Design and Operation of Farm Irrigation Systems 2nd Edition PDF

Design and
Operation of
Farm Irrigation
Systems
2nd edition
Glenn J. Hoffman
Robert G. Evans
Marvin E. Jensen
Derrel L. Martin
Ronald L. Elliott
Copyright © 2007 by the American Society of Agricultural and
Biological Engineers.
All rights reserved.
Manufactured in the United States of America.
This publication may not be reproduced

in whole or in part by any means
(with the exception of short quotes for the purpose of review)
without the permission of the publisher.
For information, contact:
ASABE, 2950 Niles Rd., St. Joseph, MI 49085-9659 USA.
Phone: 269 429 0300 Fax: 269 429 3852 E-mail: hq@asabe.org
Copy editing by Peg McCann

Cover design by Melissa Miller
Production assistance from Marcia Stults McCavit and Paticia Howard
Cover photographs courtesy of USDA Natural Resources
Conservation Service and University of Nebraska
ASABE is an educational and scientific

organization dedicated to the advancement
of engineering applicable to
agricultural, food, and biological systems.
For more information visit www.asabe.org.
The American Society of Agricultural and Biological Engineers

is not responsible for statements and opinions advanced in its meetings or
printed in its publications. They represent the views of the individual to
whom they are credited and are not binding on the Society as a whole.
International Standard Book Number (ISBN) 1-892769-64-6

Library of Congress Card Number (LCCN) 2007934701
ASABE Order Number 801P1007
PREFACE
Irrigated agriculture has played an important role in food production during the past
century and will become even more important as the global population continues to
increase. The internet has hastened the dissemination of new irrigation technology and
water management guidelines developed by specialists. Personal computers have fa-
cilitated complex calculations and control of automated irrigation systems. However,
there is still a need for irrigation system designers and system operators to have a
comprehensive book on farm irrigation systems readily available. This monograph has
been widely disseminated since first published in 1980 and reprinted in 1982. This
second edition is expected to meet an important need for several future decades.
Publication of this monograph culminates a decade of effort. Planning the second
edition first started in 1994 when a group met informally at a meeting of the American
Society of Agricultural Engineers to discuss the need for an updated version of the
first irrigation monograph. Ronald Elliott and Marvin Jensen began the organizational
and editorial processes. Leading experts in their respective areas were asked to write
various chapters. In 2000, because Elliott’s new assignment restricted his time for this
work and Jensen’s time was limited because of his involvement in several water use
studies, work on this edition was delayed. In 2002, the SW-24 Group chaired by
Robert Evans established a new editorial committee consisting of Glenn Hoffman,
Robert Evans, Gary Clark, and Derrel Martin to share the load of this monumental
task. The committee met in Denver in November, 2002, to review the status of the
revised chapters, to add new chapters, set new target dates, and to make review as-
signments. Elliott indicated that he did not wish to continue an active role and in Au-
gust, 2005, Gary Clark assumed a new position and no longer had time to continue as
an editor. The remaining work load was redistributed to Evans, Hoffman, Jensen, and
Martin.
This edition provides the latest technology in the design of surface, sprinkler, and
microirrigation systems along with basic information about soils and current informa-
tion on estimating crop water requirements. New chapters have been added on plan-
ning systems, environmental issues, efficiency and uniformity, chemigation, and use
of wastewater for irrigation.
As is the case with all ASABE monographs, the Society is indebted to many indi-
viduals who contributed significantly to the planning, writing, reviewing, editing, and
publishing of this book. The sharing of their knowledge, time, and patience in the pro-
duction of this book is greatly appreciated. All of the editors are extremely grateful for
the opportunity to work with so many dedicated and enthusiastic engineers and scien-
tists in completing this project. The assistance of the ASABE staff, especially Peg
McCann, in editing and producing this monograph is especially appreciated.
The Editors: Glenn J. Hoffman
Robert G. Evans
Marvin E. Jensen
Derrel L. Martin
August, 2007 Ronald L. Elliott
THE AUTHORS
Richard G. Allen, University of Idaho Research and Extension Center, 3793 North 3600 East,
Kimberly, ID 83341
James E. Ayars, USDA-ARS Water Management Research Laboratory, 9611 South Riverbend
Ave., Parlier, CA 93648
Evan W. Christen, CSIRO Land and Water, Griffith, New South Wales 2680, Australia
Allan W. Clark, Clark Brothers, Inc., 19772 South Elgin, Dos Palos, CA 93620
Albert J. Clemmens, USDA-ARS Arid-Land Agricultural Research Center, 21881 North
Cardon Lane, Maricopa, AZ 85238
Allen R. Dedrick, USDA-ARS National Program Staff, Beltsville, MD (retired). Current ad-
dress: 608 West Villa Rita Dr., Phoenix, AZ 85023
Harold R. Duke, USDA-ARS Water Management Research, Fort Collins, CO (retired). Current
address: 1047 Greenfield Court, Fort Collins, CO 80524
Keith O. Eggleston, Water Quality, U.S. Bureau of Reclamation, Denver Federal Center, P.O.
Box 25007 (D-5724), Denver, CO 80225 (retired)
Dean E. Eisenhauer, Department of Biological Systems Engineering, 232 Chase Hall, Univer-
sity of Nebraska, Lincoln, NE 68583
Abd El-Ghani M. El-Gindy, Agricultural Mechanization Department, Faculty of Agriculture,
Ain-Shams University, Cairo, Egypt
Ronald L. Elliott, Biosystems and Agricultural Engineering Department, 111 Ag Hall, Okla-
homa State University, Stillwater, OK 74078
Robert G. Evans, USDA-ARS Northern Plains Agricultural Research Laboratory, 1500 North
Central Avenue, Sidney, MN 59270
Robert O. Evans, Biological and Agricultural Engineering Department, P.O. Box 7625, North
Carolina State University, Raleigh, NC 27695
Delmar D. Fangmeier, University of Arizona, Tucson (retired). Current address: 848 West Sa-
fari Dr., Tucson, AZ 85704
James L. Fouss, USDA-ARS Soil and Water Research Unit, 4115 Gourrier Ave., Baton Rouge,
LA 70808
Ronald J. Gaddis, A B Consulting Co., Inc., Lincoln, NE (retired). Current address: 8100
Sanborn Dr., Lincoln, NE 68505
Leland A. Hardy, H & R Engineering, Inc., 690 Loring Dr. NW, Salem, OR 97304
Dale F. Heermann, USDA-ARS Water Management Research, Natural Resources Research
Center, 2150 Centre Ave., Building D, Suite 320, Fort Collins, CO 80526 (retired)
Glenn J. Hoffman, Department of Biological Systems Engineering, University of Nebraska,
Lincoln, NE (retired). Current address: 9203 N. Crown Ridge, Fountain Hills, AZ 85268
Sagit R. Ibatullin, Water Economy Research Institute, 12 Kolbasshy Koygeldy Str., 480022
Taraz City, Kazakhstan
Marvin E. Jensen, USDA-ARS National Program Staff, Fort Collins, CO (retired). Current ad-
dress: 1207 Springwood Dr., Fort Collins, CO 80525
Dennis C. Kincaid, USDA-ARS, Kimberly, ID (retired). Current address: 3849B North 3700
East, Hansen, ID 83334
Larry G. King, Department Agricultural and Biological Systems Engineering, Washington State
University, Pullman, WA (retired). Current address: 19855 East Silver Creek Lane, Queen
Creek, AZ 85242
vi Contents
E. Gordon Kruse, USDA-ARS Water Management Research, Fort Collins, CO (retired). Current
address: 4740 Player Dr., Fort Collins, CO 80525
Joseph M. Lord, Jr., JMLord, Inc., 267 North Fulton St., Fresno, CA 93701
William M. Lyle, Texas A&M University, Lubbock, TX (retired). Current address: Box 1679,
Hilltop Lakes, TX 77871
Mark Madison, CH2M Hill, 2020 SW 4th Ave., Portland, OR 97201
Derrel L. Martin, Department of Biological Systems Engineering, 243 Chase Hall, University of
Nebraska, Lincoln, NE 68583
Anne M. S. McFarland, Texas Institute for Applied Environmental Research, Tarleton State
University, 201 St. Felix Street, Stephenville, TX 76401
Marshall J. McFarland, Agricultural Research and Extension Center, Stephenville, TX (retired).
Current address: 1025 Darren Drive, Stephenville, TX 76401
Alan W. Moore, Cameron County Drainage District 5, 301 East Pierce St., Harlingen, TX
78550
Luis S. Pereira, Technical University of Lisbon, Lisbon, Portugal. Current address: Instituto
Superior de Agronomia, Departmento Engenharia Rural, Tapada da Ajuda, Lisboa Codex
1399, Portugal
William O. Pruitt, University of California, Davis, CA (retired). Current address: 804 West 8th
St., Davis, CA 95616
John A. Replogle, USDA-ARS Arid-Land Agricultural Research Center, 21881 North Cardon
Lane, Maricopa, AZ 85238
Matt A. Sanderson, USDA-ARS Pasture Systems and Watershed Management Research Unit,
Building 3702, Curtin Rd., University Park, PA 16802
Joseph Shalhevet, Institute of Soil, Water and Environmental Science, Agricultural Research
Organization, Bet Dagan, Israel (retired). Current address: 14 Einstein Street, Rehovot
76470, Israel
Allen G. Smajstrala, Department of Agricultural and Biological Engineering, University of Flor-
ida, Gainesville, FL (deceased)
Roger E. Smith, UDSA-ARS, Fort Collins, CO. Current address: Colorado State University,
2150 Centre Ave., Building D, Fort Collins, CO 80526
Kenneth H. Solomon, BioResource and Agricultural Engineering Department, California Poly-
technic State University, San Luis Obispo, CA (retired). Current address: 190 Kodiak St.,
Morro Bay, CA 93442
Dean D. Steele, Department of Agricultural and Biosystems Engineering, North Dakota State
University, 1221 Albrecht Blvd., P.O. Box 5626, Fargo, ND 58105
Theodor S. Strelkoff, USDA-ARS Arid-Land Agricultural Research Center, 21881 North
Cardon Lane, Maricopa, AZ 85238
Thomas J. Trout, USDA-ARS Water Management Research, Natural Resources Research Cen-
ter, 2150 Centre Ave., Building D, Suite 320, Fort Collins, CO 80526
Ted W. van der Gulik, Resource Management Branch, BC Ministry of Agriculture and Lands,
1767 Angus Campbell Rd., Abbotsford, BC V3G 2E5, Canada
Wynn R. Walker, College of Engineering, Utah State University, Logan, UT 84322
Arthur W. Warrick, Department of Soils, Water and Environmental Science, Shantz Building,
P.O. Box 210038, University of Arizona, Tucson, AZ 85721
Lyman S. Willardson, Utah State University, Logan, UT (deceased)
James L. Wright, USDA-ARS Northwest Irrigation and Soils Research Laboratory, Kimberly,
ID (retired)
I-Pai Wu, Department of Biosystems Engineering, University of Hawaii, Honolulu, HI (retired)
CONTENTS
Preface .................................................................................................iii
The Authors..........................................................................................v
Chapter 1
Introduction 1
1.1 Overview ................................................................................................................. 1
1.2 Worldwide Irrigation Development......................................................................... 2
1.3 Irrigation Development in the United States ........................................................... 7
1.4 Issues Facing Irrigated Agriculture ....................................................................... 12
1.5 Future Directions ................................................................................................... 23
References ................................................................................................................... 30
Chapter 2
Sustainable and Productive Irrigated Agriculture 33
2.1 Introduction ........................................................................................................... 33
2.2 Role of Irrigation in Food and Fiber Production ................................................... 37
2.3 Crop Production and Irrigation Water Requirements ............................................ 43
2.4 System Design and Increasing Competition for Renewable Water Supplies ........ 46
2.5 Irrigation Water Management During Droughts ................................................... 49
2.6 Other Agricultural Purposes and Benefits of Irrigation......................................... 49
2.7 System Design and Operation Challenges............................................................. 50
2.8 Summary ............................................................................................................... 51
References ................................................................................................................... 52
Chapter 3
Planning and System Selection 57
3.1 Introduction ........................................................................................................... 57
3.2 Planning for Irrigation ........................................................................................... 61
3.3 Irrigation System Selection ................................................................................... 66
References ................................................................................................................... 75
Chapter 4
Environmental Considerations 76
4.1 Introduction ........................................................................................................... 76
4.2 Water Storage, Diversion, and Consumption ........................................................ 80
4.3 Groundwater Quality ............................................................................................. 85
4.4 Surface Water Runoff............................................................................................ 93
References ................................................................................................................. 100
Chapter 5
Efficiency and Uniformity 108
5.1 Introduction ......................................................................................................... 108
viii Contents
5.2 Irrigation Scheme Physical Models ..................................................................... 109

5.3 Irrigation Performance Parameter Definitions..................................................... 111
5.4 Summary ............................................................................................................. 118
References ................................................................................................................. 118
Chapter 6
Soil Water Relationships 120
6.1 Introduction ......................................................................................................... 120
6.2 Water-Holding Characteristics of Soils ............................................................... 120
6.3 Soil Hydraulic Conductivity................................................................................ 135
6.4 Water Movement in Soil ..................................................................................... 139
6.5 Complicating Factors........................................................................................... 150
List of Symbols ......................................................................................................... 154
References ................................................................................................................. 155
Chapter 7
Controlling Salinity 160
7.1 Introduction ......................................................................................................... 160
7.2 Quantifying Salinity Hazards .............................................................................. 164
7.3. Crop Tolerance ................................................................................................... 168
7.4. Leaching ............................................................................................................. 178
7.5 Salinity Impacts on Irrigation Design.................................................................. 187
7.6 Salinity Management Practices ........................................................................... 193
7.7 Summary and Conclusions .................................................................................. 201
References ................................................................................................................. 201
Chapter 8
Water Requirements 208
8.1 Introduction ......................................................................................................... 208
8.2 Definitions ........................................................................................................... 209
8.3 Direct Measurements........................................................................................... 211
8.4 Estimation of Reference ET ................................................................................ 212
8.5 Estimating ET for Crops...................................................................................... 227
8.6 Evapotranspiration Coefficients for Landscapes ................................................. 258
8.7 Estimating Kc from Fraction of Cover ................................................................ 265
8.8 Effect of Irrigation Method on Kc ....................................................................... 266
8.9 Effects of Surface Mulching on Kc ..................................................................... 266
8.10 Precipitation Runoff .......................................................................................... 267
8.11 Other Water Requirements ................................................................................ 270
8.12 Effective Rainfall............................................................................................... 272
8.13 Design Requirements......................................................................................... 272
8.14 Annual Irrigation Water Requirements ............................................................. 277
List of Symbols ......................................................................................................... 278
References ................................................................................................................. 281
Design and Operation of Farm Irrigation Systems ix
Chapter 9
Drainage Systems 289
9.1 Introduction ......................................................................................................... 289
9.2 Environmental Considerations ............................................................................ 291
9.3 Drainage System Requirements........................................................................... 293
9.4 Approaches to Design of Subsurface Drainage Systems ..................................... 293
9.5 Determining Design Variables ............................................................................ 313
9.6 Drainage System Materials.................................................................................. 315
9.7 Construction Methods and Equipment ................................................................ 317
9.8 Drainage System Operation and Maintenance..................................................... 318
9.9 Drainage System Performance Evaluation .......................................................... 318
References ................................................................................................................. 319
Chapter 10
Land Forming for Irrigation 320
10.1 Introduction ....................................................................................................... 320
10.2 System Layout................................................................................................... 322
10.3 Soil Survey and Allowable Cuts........................................................................ 324
10.4 Topographic Survey .......................................................................................... 324
10.5 Earthwork Analyses........................................................................................... 329
10.6 Types of Equipment .......................................................................................... 334
10.7 Field Operating Procedures ............................................................................... 339
10.8 Cost and Contracting ......................................................................................... 344
10.9 Safety................................................................................................................. 344
References ................................................................................................................. 345
Chapter 11
Delivery and Distribution Systems 347
11.1 Introduction ....................................................................................................... 347
11.2 Irrigation Water Delivery .................................................................................. 348
11.3 Farm Water Distribution Systems ..................................................................... 369
References ................................................................................................................. 387
Chapter 12
Pumping Systems 392
12.1 Introduction ....................................................................................................... 392
12.2 Pump Components and Characteristics ............................................................. 394
12.3 Selecting a Pump ............................................................................................... 404
12.4 Power Units ....................................................................................................... 419
12.5 Pump Controls ................................................................................................... 425
12.6 Economic Considerations .................................................................................. 426
12.7 Impact of Pumping System Modification.......................................................... 430
12.8 Maintenance and Testing................................................................................... 432
List of Symbols ......................................................................................................... 434
References ................................................................................................................. 434
x Contents
Chapter 13
Hydraulics of Surface Systems 436
13.1 Introduction ....................................................................................................... 436
13.2 Basic Concepts of Surface Irrigation Hydraulics .............................................. 438
13.3 Variables That Influence the Surface Irrigation Process ................................... 441
13.4 Conservation Laws for Mass and Momentum................................................... 451
13.5 Hydrologic Modeling of the Surface Irrigation Process .................................... 452
13.6 Hydrodynamic Modeling of the Surface Irrigation Process .............................. 461
13.7 Estimation of Field Parameters.......................................................................... 479
References ................................................................................................................. 491
Chapter 14
Design of Surface Systems 499
14.1 Introduction ....................................................................................................... 499
14.2 Design Considerations and Approaches ............................................................ 500
14.3 Sloping-Furrow Irrigation.................................................................................. 509
14.4 Border-Strip Irrigation....................................................................................... 518
14.5 Level-Basin and Level-Furrow Irrigation.......................................................... 524
14.6 Surface Irrigation System Headworks and Control of Inflow ........................... 528
References ................................................................................................................. 530
Chapter 15
Hydraulics of Sprinkler and Microirrigation Systems 532
15.1 Introduction ....................................................................................................... 532
15.2 Hydraulics of Pipe Systems ............................................................................... 533
15.3 Irrigation System Valves ................................................................................... 547
15.4 Water Distribution to the Soil............................................................................ 554
15.5 Redistribution in the Soil................................................................................... 554
15.6 Summary ........................................................................................................... 554
List of Symbols ......................................................................................................... 555
References ................................................................................................................. 556
Chapter 16
Design and Operation of Sprinkler Systems 557
16.1 Introduction ....................................................................................................... 557
16.2 Components of Sprinkler Systems..................................................................... 558
16.3 Design Fundamentals ........................................................................................ 560
16.4 Application Uniformity ..................................................................................... 576
16.5 Solid Set Systems .............................................................................................. 580
16.6 Periodically Moved Laterals.............................................................................. 587
16.7 Center Pivots ..................................................................................................... 595
16.8 Lateral Move Systems ....................................................................................... 613
16.9 Low Energy Precision Application (LEPA) Systems........................................ 618
16.10 Travelers.......................................................................................................... 621
16.11 Auxiliary Uses of Sprinkler Systems............................................................... 626
16.12 Safety............................................................................................................... 626
Design and Operation of Farm Irrigation Systems xi
16.13 Summary ......................................................................................................... 627

List of Symbols ......................................................................................................... 627
References ................................................................................................................. 628
Chapter 17
Microirrigation Systems 632
17.1 Introduction ....................................................................................................... 632
17.2 Microirrigation Systems .................................................................................... 638
17.3 Design Factors ................................................................................................... 642
17.4 Hydraulics of Emitters and Emitter Design Variation....................................... 647
17.5 Microirrigation Design ...................................................................................... 652
17.6 Designing the System Control Head.................................................................. 656
17.7 Installation ......................................................................................................... 658
17.8 Maintenance ...................................................................................................... 658
17.9 Management ...................................................................................................... 661
17.10 Scheduling Microirrigation.............................................................................. 664
17.11 Plugging of Microirrigation Systems............................................................... 666
17.12 Subsurface Drip Irrigation ............................................................................... 668
17.13 Subirrigation .................................................................................................... 675
17.14 Microirrigation in Nurseries and Greenhouses ................................................ 675
References ................................................................................................................. 677
Chapter 18
Water Table Control Systems 684
18.1 Introduction ....................................................................................................... 685
18.2 Management of Soil Water by Water Table Control ......................................... 686
18.3 System Design and Operation in Humid Regions ............................................. 691
18.4 System Design and Operation in Arid Regions ................................................. 703
18.5 Documentation of System Design and Installation............................................ 717
18.6 Summary ........................................................................................................... 717
References ................................................................................................................. 718
Chapter 19
Chemigation 725
19.1 Introduction ....................................................................................................... 725
19.2 Backflow Prevention and Safety ....................................................................... 727
19.3 Injection Systems............................................................................................... 732
19.4 Injection System Calibration ............................................................................. 739
19.5 Irrigation System Considerations ...................................................................... 741
19.6 Calculating Injection Rates................................................................................ 750
References ................................................................................................................. 752
Chapter 20
Wastewater and Reclaimed Water Irrigation 754
20.1 Introduction ....................................................................................................... 754
20.2 Constituents and Characteristics of Wastewater and Reclaimed Water ............ 756
xii Contents
20.3 Nutrients in Wastewater and Reclaimed Water ................................................. 764

20.4 Health Concerns ................................................................................................ 765
20.5 Crops Suitable for Irrigation with Wastewater and Reclaimed Water............... 769
20.6 Design of Nitrogen Loading Rate...................................................................... 770
20.7 Irrigation Systems Using Wastewater ............................................................... 774
20.8 Design and Operation of Systems to Protect Human Health ............................. 779
20.9 Monitoring......................................................................................................... 781
References ................................................................................................................. 783
Chapter 21
Evaluating Performance 790
21.1 Introduction ....................................................................................................... 790
21.2 Management Variables ...................................................................................... 792
21.3 Measures of Performance .................................................................................. 794
21.4 Field Evaluation Methods.................................................................................. 798
21.5 Economics ......................................................................................................... 799
21.6 Analysis and Interpretation................................................................................ 800
21.7 Mobile Labs....................................................................................................... 801
21.8 Outline for Evaluations...................................................................................... 802
21.9 Conclusions ....................................................................................................... 802
References ................................................................................................................. 803
Appendix A
Glossary............................................................................................804
Appendix B
Annotated Bibliography of Irrigation Standards...........................838
Index ................................................................................................851
CHAPTER 1
INTRODUCTION
Glenn J. Hoffman (University of Nebraska,
Lincoln, Nebraska)
Robert G. Evans (USDA-ARS, Sidney,
Montana)
Abstract. This chapter introduces the second edition of “Design and Operation of
Farm Irrigation Systems.” The distribution and development of irrigation in the
United States and throughout the world is reviewed. Issues facing irrigated agricul-
ture including environmental concerns, sustainability, and changes in water policy are
discussed. The chapter concludes with a discussion of options for future directions
and some needs for irrigation research and education.
Keywords. Environmental concerns, History, Irrigation, Sustainability, Water pol-
icy, Worldwide development.
1.1 OVERVIEW
This monograph updates and complements ASAE Monograph 3, “Design and Op-
eration of Farm Irrigation Systems,” which was edited by M. E. Jensen and published
by the American Society of Agricultural Engineers in 1980. The 1980 monograph has
been the most requested and most successful of all ASAE monographs over the inter-
vening years. More than 8500 copies of the first edition have been distributed world-
wide.
Significant developments and additions to our understanding and knowledge of ir-
rigation along with major changes in the environmental, ecological, sociological, and
political scenes attest to the need for a second edition. This monograph provides cur-
rent research results and state-of-the-art knowledge on all aspects of irrigation on the
farm. It is intended to provide design procedures and scientific guidance for the man-
agement of water resources and irrigation systems.
This edition provides updated material for many of the chapters covered in the first
edition. Other chapters are significantly different or completely new. Chapters that are
built upon the subjects of the first edition are system selection, soil-water relation-
ships, salinity control, water requirements, drainage systems, land forming, delivery
systems, pumps, hydraulics, design of surface systems, design of sprinkler and micro-
irrigation systems, evaluation of performance, and irrigation management. New or
substantially different subjects include efficiency and uniformity, environmental con-
siderations, sustainability of irrigation, water table control, chemigation, wastewater
and reclaimed irrigation water, and site-specific management. By the number of new
2 Chapter 1 Introduction
subjects alone, the reader should be convinced of the many new issues in irrigation
that need to be addressed today.
This monograph should benefit practicing professionals, farmers, researchers, and
governing agencies. It provides technical data, design procedures, and management
strategies to benefit all who make decisions about the design and operation of farm
irrigation systems.
1.2 WORLDWIDE IRRIGATION DEVELOPMENT
At the beginning of a new millennium it is appropriate to review the history of irri-
gation and its impact on civilization and the global food supply. Irrigation was devel-
oped to counter both short-term and long-term drought on crop production in arid and
semiarid areas whenever a reliable supplemental water supply was nearby. The pru-
dent use of water enables the consistent production of food, fiber, and landscaping at
levels and in locations where it would not otherwise be possible. An outstanding ac-
count of the history of irrigation has been given by Postel (1999), and much of the
following historical discussion has been extracted from her writings.
It is important to note the impact irrigation has had on civilization from its earliest
beginnings to the present. Historians, through archeological evidence, credit early irri-
gation with the assurance of dependable food supplies; thereby allowing some mem-
bers of the society to pursue activities other than hunting, nomadic grazing, or farm-
ing. Among the accomplishments of these early irrigation-based societies are writing,
the wheel, water-lifting devices, yokes for draft animals, sailboats, palaces, pyramids,
temples, ceramics, fine textiles, handicrafts, the Hammurabi code of law, fired bricks,
cities, and various forms of accounting, taxation, management, and government.
1.2.1 Historical Development
Six thousand years ago, settlers in Mesopotamia (which is now Iraq, and part of
what is referred to as the Fertile Crescent) dug ditches and diverted water to their
fields from the Euphrates River and initiated the practice of irrigation. Irrigation trans-
formed both the land and society like no other previous activity by producing a de-
pendable food supply, frequently so abundant that many people were able to pursue
non-farm activities. The food surpluses had to be stored and distributed, which led to
new forms of centralized management. This need for management led to stratified
societies and centralized control. With some of the populace not needed in farming,
historians credit these early settlers, called Sumerians, with the development of writing
and the wheel as well as the creation of sailboats, water-lifting devices, and yokes for
harnessing animals. With time, as the range of activities expanded, these early socie-
ties increased in population, which led to the first true cities. The Sumerians were fol-
lowed by other irrigation-based civilizations in Mesopotamia. Among these were the
Babylonians, who are renowned for building magnificent palaces and developing
Hammurabi’s historic code of law which dealt, in part, with irrigation (Postel, 2000).
Irrigated agriculture endured in Mesopotamia through several civilizations. Archeo-
logical evidence, however, indicates that around 2400 B.C. soil salination had reached
detrimental levels. In those early civilizations, wheat was the preferred cereal for food,
and it was grown in preference to barley. Wheat, however, is less salt tolerant than
barley. Over time wheat’s share of the harvest dropped to less than two percent in
some areas and barley production dominated, and by 1700 B.C. wheat was no longer
cultivated. By then, even barley yields were declining. This archeological evidence
Design and Operation of Farm Irrigation Systems 3
has led historians to blame salination as the major contributor to the decline of civiliza-
tion in the Fertile Crescent of present-day Iraq. To this day, the southern portions of the
Fertile Crescent have never fully recovered from the demise caused by salinity.
Elsewhere in the world, early societies dependent upon irrigation arose in the Indus
River Valley in Pakistan, the Yellow River basin of China, and the Nile River Valley
of Egypt. Much later, irrigation-based cultures developed in the western hemisphere.
Central Mexico, coastal Peru, and the American Southwest each saw the rise and fall
of an advanced society built on irrigated agriculture.
The early civilization along the Indus River about 3500 B.C. also depended on irri-
gation. The complex, hierarchical society that developed by about 2300 B.C. lasted
less than 500 years. Although conquest by invaders was probably the direct cause of
the collapse of this society, instability caused by salination, siltation, flooding, and
possibly climate change also contributed to its demise.
Ancient Chinese efforts to control and use the water resources of the Yellow River
in the north China plain began about 4000 years ago. The Yellow, however, proved to
be a difficult river to tame. It emerges from the Loess Plateau carrying some 1.4 bil-
lion Mg of silt each year, making channel alterations and flooding a common occur-
rence. The Yellow has breached its dikes more than 1500 times over the last two mil-
lennia. Heavy silt deposits have elevated the river above its surrounding plain necessi-
tating the continual raising and reinforcing of dikes to prevent flooding. Ultimately a
breach of the dike occurs and a change in the course of the river results. The river has
changed course about once every century. These unpredictable shifts have been devas-
tating to irrigated agriculture and human settlement.
In sharp contrast to the other early civilizations built upon irrigation in the eastern
hemisphere, the Egyptian civilization in the Nile River Valley has endured for 5000
years without interruption. Egypt’s civilization has survived warfare and conquests
and widespread disease. Only in recent times has the sustainability of Egyptian agri-
culture come into question. In response to a 20-fold increase in population over the
last two centuries, Egypt replaced its time-tested agriculture based on the Nile’s natu-
ral flow rhythms with more intensified irrigation and flood management that requires
complete control of the river. The natural flow rhythm of the Nile was once dominated
by an annual flood that was relatively benign, predictable, and timely. With nearly
flawless predictability the river rose in southern Egypt in early July and reached flood
stage in the vicinity of Aswan by mid-August. The flood continued to surge northward
and reached the northern end of the valley by the end of September. At its peak, the
flood covered the floodplain to a depth of 1.5 meters. By late November most of the
valley was drained. Egyptian farmers then had well-watered fields that had been fertil-
ized by the rich silt carried from Ethiopia and deposited across the floodplain. Crops
were then planted as the mild winter began and harvested in the spring just in time for
the cycle to repeat. Egyptian irrigators did not experience many of the troublesome
problems of other early civilizations. Fertility was renewed each year by the silt-laden
floodwater and the inundation prior to planting pushed whatever salts had accumulated
down below the root zone.
In contrast to other ancient civilizations, early Egyptian society did not centrally
manage irrigation works. Irrigation was carried out on a local rather than a regional or
national scale. Despite the existence of many civil and criminal codes in ancient
Egypt, no evidence exists of written water law. Apparently, water management was
neither complex nor contentious, and oral traditions of common law withstood all tests
over a considerable length of time. The many political disruptions at the state level,
which included numerous conquests, did not greatly impact the system’s operation or
maintenance, probably because the politicians had no control over the flows of the Nile.
Though later in time and smaller in size, the irrigation-based civilizations that de-
veloped in the western hemisphere also shaped cultural developments. The geographic
cradle of civilization in the west was Central America. Settled villages evolved about
2000 B.C., when the productivity of domesticated corn reached a level that could sup-
port stable communities. Around 300 B.C., canals were used for irrigation throughout
the Tehuacan Valley southeast of Mexico City. Along the Peruvian coast, an irrigated
crop could be grown in four months leaving time to construct pyramids and temples
and to develop fine textiles, ceramics and handicrafts. A centralized bureaucracy was
formed to manage the large irrigation system and to control the distribution of water.
In North America, the Hohokam culture thrived for more than 1000 years along the
Gila and Salt rivers of south central Arizona but then the culture disappeared suddenly
around 1400 A.D., most likely due to prolonged severe drought. Archeologists have
documented more than 500 km of main irrigation canals, with many of them linked in
networks.
In the 16th century, Leonardo da Vinci studied the Amo watershed in northern It-
aly. His studies led to a better understanding of the relationship between a river’s
catchment area and its flow. This study helped establish the fundamentals of hydrol-
ogy and river basin management (Newson, 1992). Several hundred years passed be-
fore the principles of hydraulics and mechanized water control technologies were de-
veloped that together transformed irrigation from an art to a science.
1.2.2 Recent Developments and Trends
Over many centuries, irrigated lands have become essential to the world’s food
supply. These lands now constitute approximately 20% of the world’s total cultivated
farmland but produce about 40% of the food and fiber. Irrigated agricultural activities
also provide considerable food and foraging areas for migratory and local birds as well
as other wildlife. In short, irrigation underpins our modern world society and life-
styles.
In 1800, the worldwide irrigated area totaled about 8 million ha. The irrigated area
increased five-fold during the 19th century, mainly because during the latter half of
the century much of the scientific and technical foundation for irrigation was devel-
oped. In the 20th century, global irrigation grew from 40 to more than 270 million ha,
an almost seven-fold increase (Figure 1.1).
Currently India and China, with nearly the same amount of irrigated cropland (57
and 55 million ha, respectively), together account for about 40% of the world’s irri-
gated land. Ten countries collectively account for two-thirds of the world total (Table
1.1). It is estimated that from 36% to 47% of the world’s food is produced by irrigated
production (Gleick, 1998; Postel, 1999).
The total amount of irrigated land in a country depends upon its size, arable culti-
vated land, and climatic conditions. Thus, it is interesting to note the proportion of
each country’s arable land that is irrigated. The percent of arable land that is irrigated
within a country and by continent is given in Table 1.1. More than half of the arable
land is under irrigation in Pakistan, Iraq, Japan, Bangladesh, and Iran; nearly or all of
300
250
200
Millions of Hectares
150
100
50
0
1800 1825 1850 1875 1900 1925 1950 1975 2000
Years
Figure 1.1. Increase in irrigated area worldwide from 1800 to 2000

(adapted from Postel, 1999, and FAOSTAT, 2005).
the arable land in Uzbekistan and Egypt is irrigated. Thirty-eight percent of all the
arable land in Asia is irrigated. Brazil has seen rapid growth in recent years. In con-
trast, in Africa, a continent desperate for food, only 7% of the arable land is irrigated.
On average, irrigation is practiced on 20% of the arable land in the world (Table 1.1).
Worldwide irrigated agriculture increased significantly during the second half of
the 20th century with large government investments, major financial support from
international donors and lenders, and the spread of improved pumping technologies.
This period coincided with numerous, large-scale, government-sponsored surface wa-
ter irrigation projects in many countries and the proliferation of both private and pub-
lic groundwater wells in others.
In China, waterworks were constructed on major Huai, Hai, and Yellow River ba-
sins, mostly for rice and small grain production. By 1990, more than 4 million ha were
irrigated from the Yellow River. With improved pumping and well-drilling technolo-
gies and the availability of electricity, China turned to groundwater to irrigate the
North China Plain. The number of wells increased from 110,000 in 1961 to more than
2 million in the mid-1980s (Postel, 1999).
Likewise, irrigation by tubewells increased dramatically in India and Pakistan dur-
ing the last half of the 20th century. Irrigation from tubewells expanded from 100,000
ha in 1961 to 11.3 million ha in 1985 (World Bank, 1991). In this same time period,
canal building for surface irrigation by the Indian government doubled the surface
irrigated area. In the Indus River basin, two large storage dams and corresponding
construction of irrigation canals transformed the basin into the world’s largest con-
tiguous irrigation network, covering 14 million ha.
Table 1.1. Irrigated areas in the top 25 countries, by continents, and in

the world in 2002 (adapted from FAOSTAT, 2005). These totals are
estimates because definitions of irrigated land vary by country.
Arable Land
Irrigated Irrigated Area
Region of the World (percent) (million hectares)
Country
Egypt 100 3.4
Uzbekistan 96 4.3
Pakistan 83 17.9
Iraq 60 3.5
Japan 59 2.6
Bangladesh 58 4.6
Iran 50 7.5
Viet Nam 45 3.0
China 38 54.9
India 35 57.2
Italy 34 2.8
Romania 33 3.1
Thailand 31 5.0
Afghanistan 30 2.4
Spain 28 3.8
Mexico 25 6.3
Indonesia 23 4.8
Turkey 20 5.2
France 14 2.6
United States 13 22.5
Kazakhstan 11 2.4
Ukraine 7 2.3
Brazil 5 2.9
Australia 5 2.5
Russia 4 4.6
Continent
Asia 38 193.9
North & Central America 12 31.4
Europe 9 25.2
South America 9 10.5
Africa 7 12.9
Oceania 6 2.8
World 20 276.7
In the former Soviet Union, leaders expanded irrigation in areas with otherwise fa-
vorable climatic conditions that lacked adequate rainfall for crop production. They
concentrated on two key areas—the Central Asian republics, which accounted for
about 40% of the Soviet irrigated area before the nation’s breakup, and the southeast-
ern European region, including parts of Russia and Ukraine. With irrigation water
drawn from the rivers feeding into the Aral Sea, central Asia became a major cotton
growing region. However, the Aral Sea is a small fraction of its previous size because
of declining inflows due to greatly increased upstream diversions for irrigation. While
large scale irrigation in southeastern Europe protected this important grain-producing
region from drought there has been severe damage to the region’s natural ecosystems.
Presently, chronic scarcity of water is experienced or expected in large parts of Af-
rica and the Middle East, the northern part of China, the former Soviet Union and the
Central Asian republics, parts of India and Mexico, the western part of the U.S., and
northeast Brazil. A region is said to be water stressed when its renewable water sup-
plies drop below about 1700 cubic meters per capita (Gleick, 1998; Postel, 1999).
With supplies below this level, it becomes difficult for a country to mobilize enough
water to satisfy all the food, household, and industrial needs of its population. For ex-
ample, about 1000 Megagrams (Mg) of water are required to produce 1 Mg of grain.
With grain being the staple of the human diet, water-stressed countries import grain to
satisfy food requirements and reserve their water for other uses. Collectively, 32 of the
34 water-stressed countries in Africa and Asia import about 50 million Mg of grain
annually, about a quarter of the total traded internationally (Postel, 2000).
During the first quarter of the 21st century, the global irrigation base is expected to
grow at less than 1% a year, down from the annual growth rate of more than 3% for
the last half of the 20th century. This has largely been the result of diminishing water
supplies and increased demands from other sectors. In most areas, the best and easiest
sites are already developed. A 1995 study by the World Bank reported that irrigation
development costs on more than 190 of their funded projects averaged about $4,800
per hectare (Jones, 1995).
1.3 IRRIGATION DEVELOPMENT IN THE UNITED STATES
Irrigation has been practiced in the Southwest region of the U.S. for two millennia.
Today, irrigation is practiced in every state of the union with the total irrigated agricul-
tural cropland area exceeding 21 million ha (21.3 million ha using the USDA-NASS,
2004, estimate; 22.5 million ha using a different methodology, as listed in Table 1.1)
with a farm gate value exceeding $47 billion per year. This represents 49% of the
market value of all crops from 18% of all harvested U.S. croplands.
Nationally, per hectare sales from harvested irrigated lands average more than 4.5
times the sales from nonirrigated land (Gollehon, 2002). However, this is a somewhat
misleading figure because much higher input, labor, and equipment costs often result
in net returns similar to that for rainfed or dryland producers, especially for field
crops. This is usually not the case for tree and vine crops, vegetables, and other high-
value crops.
It is estimated that there are also approximately 16.4 (± 3.6) million ha of managed
turf in the U.S. (Milesi et al., 2005) that is not included in any of the estimates of crop-
land referenced in this chapter (or elsewhere). Turfgrass areas include lawns, airports,
institutional facilities, military bases, schools, parks, golf courses, athletic fields,
churches, cemeteries, etc. These areas are expected to continue growing, and much of
this land is irrigated to some degree, almost totally with sprinklers. The large area of
turfgrass makes it the largest irrigated crop in the U.S. Milesi et al. (2005) estimated
that irrigated turfgrass amounted to three to four times the area of irrigated corn (about
4.3 million ha), the largest irrigated agricultural crop. Applied water and other inputs
are high because turf is generally managed for appearance rather than profitability.
The amount of irrigated turfgrass is extremely significant because it indicates that ur-
ban irrigators have a large potential for water and energy savings and pollution abate-
ment that is comparable to irrigated agriculture. Furthermore, the political and eco-
nomic base clearly lies with the urban users and they will have a major voice in future
land use and water policy issues.
Thus, including turf, there is a total of about 37 million ha of irrigated land in the
US. The impact of turfgrass is quite significant because urban users are often compet-
ing for the same water supplies, especially in the 17 western states (not including Ha-
waii and Alaska) that contain about 30% of the total turf area in the U.S. In addition,
urbanization often expands on to irrigated agricultural lands, and much of the associ-
ated water is often lost to agriculture as it is then used for urban landscapes.
1.3.1 Historical Development of Irrigation in the U.S.
Irrigation by indigenous people of the Southwest is known to have existed as early
as 100 B.C. Early irrigation was practiced in the Salt River Valley, on the Colorado
Plateau, and along several watercourses elsewhere in the West. Corn, beans, squash,
milo, peaches, and other crops were grown through an intricate network of ditches and
canals.
Some of the early immigrants arriving in North America, particularly from the
Mediterranean area, brought with them a heritage of irrigation as part of farming. For
example, Spanish colonists irrigated extensively at the missions established along the
Pacific Coast beginning in the 1760s. In the mid-1800s, irrigation development in the
U.S. expanded along with the settlement of the West. Most of the early projects were
accomplished by private enterprise. In 1847, an advance party of Mormons settled in
the Salt Lake Valley of Utah and began diverting water to grow potatoes. Early Mor-
mon activities set the stage for other private irrigation ventures, and by about 1875, the
center for irrigation innovation and development had shifted to Colorado and Califor-
nia. By 1890, irrigation was practiced on more than 400,000 ha in California, 350,000
ha in Colorado, and more than 100,000 ha in Utah. About this time, irrigation became
a central theme of the federal government’s strategy to encourage settlement of the
West. The Desert Land Act of 1877 and the Carey Act of 1894 were intended to
stimulate private and state development of irrigated land. The federal government be-
came involved in irrigation development with the passage of the Reclamation Act of
1902. The Bureau of Reclamation formed by this act provided engineering expertise
and financial capital to develop large irrigation projects throughout the West.
The irrigated area in the U.S. increased from about 1 million ha in the 1880s to
about 8 million ha by the middle of the 20th century. At this time, the major irrigated
regions were the Southwest (2.2 million ha), the Mountain States (2.5 million ha), and
the Pacific Northwest (1.4 million ha) (U.S. Department of Commerce, 1983). The
drought years of the 1950s stimulated irrigation development in the southern Great
Plains. Then, with the advent of center pivot sprinkler irrigation systems and with
groundwater readily available, irrigation expanded rapidly in the central Great Plains
during the 1960s and 1970s. During the same time period, irrigation increased mark-
edly in the southeastern states. The total irrigated area declined in the 1980s because
of depressed farm commodity prices, increased energy costs, and declining water re-
sources. In the 1990s the total irrigated area recovered and data for 2003 shows the
total irrigated area to be about 21.3 million ha (USDA-NASS, 2004)
Recent growth in U.S. irrigation has been mostly in the southeastern areas of the
country, primarily in the Mississippi River delta. The 1980s and 1990s saw growth in
the lower Mississippi River areas of Missouri, Mississippi, and Louisiana; the 1960s
and 1970s had significant growth in Georgia, South Carolina, Alabama, and Arkansas.
Mississippi, Arkansas, and Missouri added about 1.2 million ha of irrigated land since
1980. This essentially doubled irrigation in these states (USDA-NASS, 2004).
1.3.2 Current Status of U.S. Irrigation
The “official” survey of agricultural irrigation in the U.S. is typically taken as part
of the U.S. Census of Agriculture. The most recent results are in the 2003 USDA Farm
and Ranch Irrigation Survey as part of the 2002 Census of Agriculture (USDA-NASS,
2004). The 2002 Census of Agriculture reported 21.3 million ha of irrigated land. The
USGS (Hutson et al., 2004) estimated about 25 million ha of total irrigated farmland
in the U.S. in 2000. The differences between the two values are due to different esti-
mating methodologies. Data reported in the figures and tables in this section are from
the 2002 Census.
The approximate agricultural cropland area irrigated in each state is indicated in
Figure 1.2. Except for Iowa and North Dakota, the states west of the Mississippi River
all have in excess of 100,000 ha of irrigated land. In comparison, only seven states
east of the Mississippi had more that 100,000 ha of irrigation.
Other interesting comparisons can be made from the U.S. agricultural census data.
The source of water for farm irrigation is summarized in the 2003 survey and indicates
that 26% of the water used for irrigation comes from surface sources off the farm,
while about 60% of the irrigation water comes from groundwater delivered from
wells, and only 14% comes from on-farm surface sources (Table 1.2).
Nationally, thermoelectric power accounts for 48%, irrigation 34%, and municipal
and industrial uses 16%, of all water withdrawn in the U.S. from both fresh and salt-
water sources (Hutson et al., 2004). However, irrigation is the largest user of freshwa-
ter supplies, especially in the arid west, accounting for about 40% of the total U.S.
freshwater withdrawals from both surface and subsurface supplies.
The methods of irrigation have changed over time. From early diversions of water
from streams by ditches dug by hand, irrigation technology has developed to include
Figure 1.2. Approximate irrigated farm land in the U.S. in 2003 by state
(adapted from USDA-NASS, 2004).
Table 1.2. Amount of water used from various sources for U.S. farm irrigation
in 2003 (adapted from USDA-NASS, 2004).
Area Volume Average
Irrigated Applied Depth Applied
Water Source (hectares) (km3) (cm)
Wells 13,094,000 53.6 41
On-farm surface water 2,946,000 14.5 49
Off-farm surface water 5,614,000 39.0 69
U.S. total 21,654,000[a] 107.1 49
[a]
Sum of land irrigated exceeds actual U.S. total (21,300,000 hectares) because some land receives water
from more than one source.
massive reservoirs and networks of canals to satisfy gravity irrigation systems, manu-
ally and mechanically moved sprinkler systems, and a variety of low-flow systems
referred to as microirrigation. Originally, irrigation was accomplished by methods
utilizing gravity to distribute and apply water. In the 20th century, sprinkler technol-
ogy along with low-cost aluminum and later PVC pipe was developed, and currently
more land is irrigated by sprinklers than by gravity. Similarly, low-flow microirriga-
tion is based on plastic technologies evolved during the last quarter of the 20th century
and is now used on about 5% of the irrigated area in the U.S. (USDA-NASS, 2004),
but on less than 1% worldwide.
Table 1.3 summarizes the amount of land irrigated by the various irrigation meth-
ods. In 2003, 50.5% of the area being irrigated was with sprinklers, 43.4% by gravity
methods, 5.6% by microirrigation, and 0.5% by subirrigation. Of the 10.9 million ha
irrigated by sprinklers, 79% of this area utilized center pivots, which amounts to about
40% of all the land irrigated in the U.S. To realize energy savings through low-
pressure applications, 90% of center pivots reported in the 2002 Census use water
pressures below 400 kPa. Furrow irrigation is practiced on 51% of the land using grav-
ity systems, while border or basin methods are utilized on 38% of the surface irrigated
lands.
The average depth of irrigation water applied per unit land area can also be calcu-
lated from the Census results. Nationwide, an average depth of 49 cm of irrigation water
is applied annually to irrigated lands in the U.S. (Table 1.2). For land irrigated by
pumped groundwater from wells, an average depth of only 41 cm of irrigation water is
applied each year. In comparison, an average depth of 69 cm of water is applied when
the supply is surface water from off the farm. The USGS reported that more than 195
billion cubic meters of water were diverted from off-farm surface sources annually
with an estimated average application depth of about 75 cm in 2000 (Hutson et al.,
2004). This is slightly higher than the 2002 Census of Agriculture estimates.
The differences in application depth estimates between the USGS and Census of
Agriculture reports are probably caused by the USGS analysis being more indicative
of gravity systems supplied from off-farm surface water sources. These tend to apply
more water than other irrigation methods. The comparisons are also biased by how
water source was considered. Center pivots are generally more efficient than gravity
methods and are the most prevalent irrigation method in the subhumid regions of the
Great Plains and humid areas in the east which use well water to supplement precipita-
tion. In contrast, gravity irrigation is dominant where the source of water is off-farm
and where the climate is generally more arid, as in the Western States.
Table 1.3 . Comparison of irrigation methods in the U.S.

in 2003 (adapted from 2002 Census of Agriculture, USDA-NASS, 2004).
Irrigated Area
Irrigation Method (ha) % of Total
Gravity systems
Furrow 4,746,000
Border/basin 3,582,000
Uncontrolled flooding 930,000
Other 104,000
U.S. total, gravity systems 9,362,000 43.4
Sprinkler systems
Center pivot, pressures above 400 kPa 785,000
Center pivot, pressures 200 to 400 kPa 3,910,000
Center pivot, pressures below 200 kPa 3,926,000
Linear move 139,000
Side roll, wheel move, or other mechanized move 739,000
Traveler or big gun 256,000
Hand move 674,000
Solid set and permanent 477,000
U.S. total, sprinkler systems 10,906,000 50.5
Microirrigation systems
Surface drip 574,000
Subsurface drip 164,000
Microsprinklers 472,000
U.S. total, microirrigation systems 1,210,000 5.6
Subirrigation 113,000 0.5
Total U.S. irrigation[a] 21,590,000
[a]
The U.S. total irrigated area is larger than the 21.3 million ha quoted previously because more than one
irrigation method may be used on some lands.
1.3.3 Trends in U.S. Irrigation

The irrigated agricultural area in the U.S. from 1939 to 2003, based on national
census data, is given in Figure 1.3. Total irrigated area increased at an average rate of
3.6% during the last six decades of the 20th century. The rate of increase is relatively
steady except for the sudden increase in the 1970s followed by a decline in the 1980s.
The surge in irrigated area in the 1970s can be attributed to the deployment of center
pivot systems and the installation of systems, particularly in the southeast, to augment
rainfall. The decline during the 1980s was created by a combination of depressed farm
commodity prices, increased energy costs, declining water resources, and some appar-
ent changes in statistical procedures in determining irrigated areas. What is interesting
is the return to a positive growth rate of irrigated land during the early 1990s, but a
slight decline again during the last 5 years of census data (1997-2002). This slight
decline is expected to continue. It is also expected that the conversion from surface
irrigation methods to self-propelled (center pivots and linear move) and microirriga-
tion technologies will increase, but will have little effect on total irrigated acres.
25
20
Millions of Hectares
15
10
0
1935 1954 1974 1992
Years
Figure 1.3. The total agricultural irrigated area in the U.S. from 1939 to 2002
(adapted from U.S. Department of Commerce, 1978, and
2002 Census of Agriculture, USDA-NASS, 2004).
1.4 ISSUES FACING IRRIGATED AGRICULTURE

Irrigation is the largest single consumer of water on the planet, accounting for
about 80% of the total freshwater consumed and about two thirds of the total diverted
for human uses, and it is responsible for more than 40% of all agricultural production.
Irrigation has permanently changed the social fabric of many regions around the
world. It has provided major economic development of many semiarid and arid areas,
stabilizing rural communities, increasing income, and providing new opportunities for
economic advancement to many. However, the world’s supply of freshwater is basi-
cally constant, and the amounts allocated to irrigated production will undoubtedly de-
crease substantially because of increased consumption by non-agricultural users.
The development of irrigation early in the 20th century in the U.S. was created by
an enormous level of governmental involvement, and despite the major contributions
to a stable and bountiful food supply, the commitment toward agriculture worldwide,
and particularly in the U.S., has diminished in recent years. Many members of society
exhibit ambivalent feelings toward irrigated agriculture.
This phenomenal development has not been without controversy and not without
problems or critics. Opposition to irrigation has two primary themes. First, irrigation is
increasingly coming into direct competition with other water uses for scarce water
resources. There are escalating demands from other water user sectors, including pota-
ble water, recreation, and tourism, that are looking to agriculture to supply the needed
water by conservation and other measures (Clemmens and Allen, 2005). Much of this
development in the Western U.S. has occurred during a long term, uncharacteristic,
wet period causing unrealistic expectations and over-allocations of water supply. Sec-
ondly, irrigated agriculture (although not alone) can have negative environmental im-
pacts on water and soil quality over broad areas. Irrigation can degrade water quality,
erode soils, reduce groundwater levels, deplete stream flow, and alter hydrologic re-
gimes leading to unhealthy trends in aquatic and riparian ecosystems. The nation’s
commitment to minimize such environmental issues continues to increase and irri-

gated agriculture will receive more and more pressure to change existing practices to
address these concerns.
It is expected that the current commodity surpluses will not last in the face of in-
creasing population coupled with the worldwide decrease in ocean fisheries. These
factors will place large worldwide demands on agriculture to increase the global pro-
duction of animal/fish protein, food, fiber, and livestock feed. Agriculture will also be
asked to help supplement the world’s ever-increasing energy needs by producing more
bio-based fuels, lubricants, and chemical plant feedstocks that are now provided by the
petrochemical industry. Irrigated agriculture, including greatly expanded intensive
greenhouse culture, is expected to provide more than 70% of this increased future food
and fiber demand worldwide in the next 25 to 50 years. Aquaculture will also expand
significantly, to offset the worldwide decline in fish catches, but the net amount of fish
protein will probably remain about the same.
However, most of the potential irrigation development has already occurred, and, in
fact, productivity of the irrigated land base around the world is declining due to soil
salination, waterlogging, and soil erosion. Despite phenomenal advancements in crop
breeding, genetic engineering, and other technologies, it is not known where or how
all of this increased productivity, needed to satisfy the increasing population, will occur.
Further stresses are being imposed on existing water resources by endangered spe-
cies regulations, international and interstate agreements on water allocations, energy
availability, and a suffering agricultural economy. Declining water tables in many
regions due to excessive pumping for irrigation are also a major concern. Temporary
and long-term water transfers between users, inter-basin diversions, and emerging
economic water markets are also confounding the issues.
Water and land resources and their many competing uses must be considered in a
regional and international framework. Withdrawals from a river deplete instream
flows that may have a large impact on downstream water users (municipalities, indus-
tries, natural areas and wildlife, agriculture, livestock operations, recreation, naviga-
tion, tourism, and hydropower production). Reservoir releases for power often conflict
with other competing demands for water, including irrigation. Withdrawals from aqui-
fers by wells can negatively affect groundwater levels over large areas and reduce
recharge to streams and other water bodies.
Global climate change may be further exacerbating the problems through changing
temperatures and probably by altering annual precipitation amounts and regional rain-
fall distribution patterns. The Intergovernmental Panel on Climate Change (2001) con-
cluded that a global warming attributable to human activity is now evident in the his-
toric record. Global mean surface temperature relative to 1990 is expected to increase
by about 2°C by 2100. If global warming occurs, it is certain to have a major impact
on water supplies, and the increased variability in precipitation will provide major
challenges for the agricultural sector. A warmer climate would accelerate the hydro-
logic cycle, increasing both the global rates of precipitation and evapotranspiration
(ET). Timing of precipitation as well as runoff from mountain snowmelt may be dif-
ferent from historical norms. There is evidence that some of these changes are already
occurring, but regional impacts are uncertain at best. The hydrologic uncertainties are
compounded because relatively small changes in precipitation and temperature can
have sizable effects on the volume and timing of runoff as well as ET, especially in
arid and semiarid areas. In short, the prospect of global warming introduces major new
uncertainties and challenges for irrigators and society as a whole.
All of these factors have created a water demand crisis that only continues to
worsen. Consequently, there is an urgent need to improve the management of irriga-
tion worldwide to conserve water, soil, and energy as well as provide food, fiber and
other critical needs. Adjustments will have to be made by all, but the major expecta-
tions will likely be focused around agriculture, especially irrigated production in arid
areas.
Equitable solutions to these issues must include irrigated agriculture as well as all
of the other users in determining future programs and directions. Managing and devel-
oping infrastructure and policies for water security to equitably satisfy the demands of
all users of this limited resource will be a difficult and lengthy task.
1.4.1 Environmental Concerns
The major environmental issues relevant to irrigation are those concerned with the
protection and management of water resources and water quality. During the past few
decades, society has become increasingly conscious of and concerned about the im-
pacts of irrigated agriculture on environmental quality. The relative significance of
environmental issues varies among regions of the country, but the types of issues con-
fronting irrigation generally are the same. However, with few exceptions, environ-
mental laws and policies have not addressed irrigation-related concerns.
Irrigated agriculture has had profound positive and negative impacts on the envi-
ronment. Irrigation has contributed to losses and changes of aquatic and riparian habi-
tats and the decline of some native species dependent on those habitats (Wilcove and
Bean, 1994). Irrigation runoff is a significant source for potential pollutants in surface
waters (National Research Council, 1989). Federal and state responses to environ-
mental concerns regarding irrigated agriculture include efforts to control soil salina-
tion and agricultural nonpoint sources of water pollution, policies to protect stream
flows and wetlands, and restrictions on the application of pesticides.
Environmental issues related to instream flow and wetland ecosystems arise wher-
ever water is withdrawn for irrigation. Dams and diversions for surface water supplies
reduce stream flows, altering the natural hydrograph and changing water temperature
and flow regimes. These changes may degrade fish spawning and rearing habitats. The
draining of wetlands for irrigated agriculture impacts waterfowl and other aquatic spe-
cies that use these habitats. As an example, 92% of the historic wetland areas in the
San Joaquin Valley of California have been converted to irrigated lands (San Joaquin
Valley Drainage Program, 1990). Large-scale water resource development projects
including irrigation, flood control, recreation, hydropower, and navigation have also
altered aquatic and riparian habitat conditions of the Platte, Colorado, Columbia, and
Snake rivers.
Irrigation development has greatly increased wetland areas in many arid areas. This
is generally perceived as a social benefit because of increased land for migrating and
non-migratory birds, wildlife, and recreational activities. Streamflows tend to be stabi-
lized due to buffering by increased subsurface return flows, thereby enhancing recrea-
tion, fisheries, and waterfowl habitat over pre-irrigation conditions. However, efforts
to improve irrigation efficiency may lead to a decrease in these artificially induced
habitats because of reduced return flows to the area, thus leading to additional contro-
versy with recreational interests, fishery and wildlife groups, and regulators.
It should also be mentioned that urbanization is creating large pollution loads on

freshwater supplies and estuaries. The amount of pollution in the world’s waterways is
increasing while total water supply is decreasing due to increasing evaporation due to
more intensive use. Urban populations are concentrated in coastal areas where treated
and untreated sewage waters are discharged into the seas where the water is no longer
available for reuse by agriculture and other uses, thereby contributing disproportiona-
bly to water scarcity. Desalination of seawater is an option in these areas but this also
raises significant environmental concerns related to salt disposal (Seckler and Amaras-
inghe, 2000).
1.4.2 Sustainability of Irrigation
Sustainability is a broad concept with many different meanings and connotations.
Some people define it solely in terms of environmental factors, such as soil and water
salinity, soil erosion, agrochemical use, and water pollution, that change the wildlife
and riparian ecologies. Others view sustainability only in terms of economics and the
continued agronomic production of food and fiber and thus include commodity prices,
infrastructure development, equipment costs, pest control, and energy. By either defi-
nition, irrigated agriculture is not in fact economically or environmentally sustainable
over the long term using existing technologies and policies.
Sustainability is an important concept, because irrigation underpins our modern
world society and lifestyles by providing at least 40% of our total food and fiber sup-
ply. In reality, society cannot afford and will not allow the loss of this tremendous
asset. Despite the current problems and negative perceptions in many sectors of soci-
ety, it is certain that irrigation will continue to be a necessary and important compo-
nent of the world’s well-being and growth.
Agricultural water security is obviously a major part of sustaining irrigated agricul-
ture. It is a term that is used to describe the need to maintain adequate water supplies
to sustain the food and fiber needs of the expanding world population. Factors that
may impact water security include competition for water, environmental concerns,
continued urbanization, government policy, and the globalization of the economy.
One of the biggest threats to the sustainability of irrigated agriculture is salinity.
Surface and groundwaters contain dissolved salts that are picked up as the water
moves through various geologic formations and soils. When plants extract water from
the soil for transpiration almost all of the salts in the soil water solution are left behind.
Evaporation from the soil surface also deposits salts. (Irrigation without salinity prob-
lems occurs in more humid regions such as the irrigated rice culture areas in Asia,
where it has been practiced for thousands of years, and also where the timing and
amount of rainfall combined with good natural drainage are sufficient to leach salts
from the system.) Excess accumulation of salts in the plant root zone can cause large
yield reductions or even total crop failure, and inability to remove these salts will
make agriculture unsustainable. Soil salination was probably the primary reason for
the failure of many ancient societies in irrigated arid areas (see Section 1.2.1 above).
Presently, about 30% of the land in the conterminous U.S. has a moderate to severe
potential for salinity problems (Tanji, 1990). Many areas in the West, such as the
Colorado River Basin, the northern Great Plains, and California’s San Joaquin and
Imperial valleys, suffer salinity problems, as do large irrigated areas in India, Pakistan,
and elsewhere around the world.
Improved irrigation efficiencies in arid areas often require additional water for
leaching salts brought in with the irrigation water, which can cause a salinity concen-
tration increase in the groundwater. Irrigation-induced soil salination can be avoided
by providing adequate drainage and appropriate water management, but drainage is
expensive and can degrade environmental water quality. In closed basins drainage
from irrigated areas can render the water terminus biologically uninhabitable, as oc-
curred in the Salton Sea in southern California. The Great Salt Lake in Utah is another
drainage catchment terminus that was naturally uninhabitable and is being further de-
graded by urban and agricultural water users.
Subsurface drainage and surface return flows from irrigation are sources for chemi-
cal and salt pollution in rivers, streams, lakes, and estuaries. Pollutants mobilized and
transported by irrigation return flows and drainage into streams and man-made wet-
lands include trace elements (e.g. selenium, boron, and molybdenum), nitrogen, and
salts, as well as pesticides, herbicides, and other chemicals. In sufficient concentra-
tions, these pollutants may be detrimental to wildlife and birds, as was the case with
Kesterson Reservoir in the San Joaquin Valley of California.
Field drainage systems are an essential part of controlling salination of agricultural
lands due to irrigation activities. Desalination or other costly types of treatment of
some drainage waters may be required to overcome these obstacles; however, disposal
of these effluents may also be a problem. Human and environmental health implica-
tions from reuse of degraded water must also be examined.
Thus, the real question facing the world today is how to make irrigation sustain-
able, both environmentally and economically. The sustainability of irrigation depends
on society’s ability to find ways to use this technology so that important benefits con-
tinue to be provided, but with less troublesome social, environmental, and economic
consequences. Society will need to improve agricultural productivity, change institu-
tional structures, modify water policies, improve delivery and on-farm systems, im-
prove management of degraded soils, enhance water reuse, improve crop water man-
agement, and address rising energy prices. Greatly increased investments in irrigation
infrastructure and economic incentives will be required throughout the world to enable
these required increases in productivity that permit sustainable irrigated production
while addressing the changes in the priorities for water use.
A recent USDA agricultural water security listening session (Dobrowolski and
O’Neill, 2005; O’Neill and Dobrowolski, 2005) developed a list of research and action
areas over the next several years for maximizing the efficiency of water use by farm-
ers, ranchers, and communities. These water issues included:
enhancing supplies with new storage facilities;
expanding existing infrastructure;
funding for water reclamation and reuse;
lowering water consumption by rural and urban users with new technologies;
developing new technologies and systems for recycling and reusing degraded
water;
providing risk assessment and management for water scarcity;
determining sociological and economic impacts of conservation technologies;
evaluating the impacts of environmental degradation and subsidies for water and
crops;
determining the role of physical and paper water banks and progressive water
rate structures and other market-based or incentive mechanisms;
applying biotechnological improvements in water use efficiency;
measuring the consequences of water conservation and reuse;
responding to public health and environmental concerns of reused water and wa-
ter treatment strategies to improve return flows from agriculture; and
expanding decision making tools for water in agricultural, rural, and urbanizing
watersheds including forecasting supply and shortage.
1.4.3 Productivity Challenges
The formidable pressures on water resources make it certain that water will be the
major natural resource issue of the 21st century (Seckler and Amarasinghe, 2000).
There will be large economic and social pressures to reduce irrigation water use. This
means that productivity per unit of water consumed must be much higher than current
values to meet the high future demands for food, fiber, livestock feed, and biological
alternatives to petroleum products. This is a major shift from the current emphasis on
maximizing yield per unit area, and it will require a significant re-thinking of how and
why irrigation is practiced.
There may be small pockets of future growth, but the world’s irrigable land base is
essentially developed. Some irrigated areas are actually declining in size and produc-
tivity due to waterlogging and soil salination, urban encroachment, soil erosion, and
declining water tables. Combining a fully exploited land base with the growing com-
petition for existing freshwater supplies along with the needs for increased agricultural
production and the concurrent need for energy conservation will require that irrigators
substantially increase efficiency and productivity per unit of water consumed. Postel
(1999) estimated that water use efficiencies across the world were only about 40%,
indicating that there are large opportunities for improvement. This applies to both ag-
ricultural and urban crop production (i.e., turfgrass). The greatest potential may be in
developing countries where improved water management strategies and practices have
the largest potential to increase production.
Productivity of a specific crop is a function of maximizing application efficiencies,
whereas improving productivity, in general, implies maximizing both efficiency and
cropland net return. Thus, to improve the productivity of crops with any irrigation
system we must consider many factors including the crop variety, plant water re-
quirements, quantity and quality of the water supply, soil characteristics, topography,
field size and shape, local climate, and a large number of economic concerns, such as
labor requirements, available capital, and other resource costs. Many of these factors
are interdependent, and it will be necessary to custom design a synergistic mix of
strategies that fit the delivery systems, farming culture, crop water use patterns, soils,
regional hydrology, and other unique environmental characteristics of an area. Clem-
mens and Allen (2005) present an excellent discussion of the environmental and eco-
nomic trade-offs involved in implementing advanced irrigation practices that improve
efficiencies.
A critical link in improved productivity is the implementation of advanced scien-
tific irrigation scheduling techniques, especially under deficit management. However,
most growers lack sufficient economic incentives to implement the techniques.
One of the greatest constraints to managing for enhanced productivity as well as
protecting water quality is the inability of agricultural producers to control inputs in
ways that account for the variability in growing conditions across a field. Infiltration
rates can vary between irrigation events as well as with location within the field. Wide
variations in soil types, soil chemical properties, subsurface conditions, topography,
drainage, insect/weed/disease problems, soil compaction, weather patterns, irrigation
system operation and maintenance, and wind distortion of sprinkler patterns, as well as
external factors such as herbicide drift, can cause yields and crop water use to vary
across a field as well as across adjacent fields. These may also be impacted by tillage
practices and crop rotations. Thus, it may be more advantageous to apply water uni-
formly over smaller portions of a field to minimize production differences and envi-
ronmental consequences due to the variability of these numerous factors rather than
managing the entire field as one management unit. The challenges and opportunities
for irrigation equipment manufacturers, designers, researchers, managers, and growers
will be immense and ultimately quite profitable.
1.4.4 Water Policy Issues
As populations increase, competition for available water supplies will intensify. In
some areas of the Western U.S., agricultural water users have some of the most senior
water rights. As cities have grown, the consumptive components of water rights have
been transferred from agriculture to the municipal sector through the willing sale and
purchase of the water rights. Depending on the area and natural rainfall amounts, the
farmer selling his water rights reduces the area irrigated or converts the non-irrigated
land to dryland farming or pasture. Irrigated land could also be fallowed in alternate
years or as part of various irrigated-dryland cropping rotations.
Worldwide, water policies involve many entities, each behaving according to its set
of rules and incentives. These entities include irrigators, landowners, irrigation dis-
tricts, water user organizations, state or provincial water agencies, national ministries,
development banks and organizations, private voluntary groups, engineering and con-
sulting firms, politicians, voters, and taxpayers. The policies and rules of these entities,
in most cases, do not encourage improved irrigation efficiency through design or op-
eration. An estimated $33 billion annually worldwide in government subsidies tend to
keep water prices artificially low and discouraging investments to conserve irrigation
water supplies (Myers, 1997). In addition, many laws and regulations have been a bar-
rier to marketing or transferring water, leading to inefficient water allocation and use.
Similarly, lack of policies to regulate groundwater use has led to over-pumping and
depletion of aquifers. There is a worldwide shortage of appropriate institutions to as-
sist growers in managing water more effectively while reducing negative environ-
mental impacts.
The following sections provide examples of current policy issues and introduce po-
tential improvements in water policies. For those desiring more information, these
issues and more are covered in great detail in numerous publications by private and
governmental groups, universities, and agencies that have conducted studies and de-
veloped plans to address complex water-related issues in the 21st century. For exam-
ple, a comprehensive series of reports was prepared for the Western Water Policy Re-
view Commission (established under the Western Water Policy Review Act of 1992
[PL 102-575, Title XXX]). These numerous reports present in-depth analyses of the
water resource issues and demographics of every major river system in the West, and
are summarized in “Water in the West: Challenge for the Next Century” (Western
Water Policy Review Commission, 1998).
1.4.4.1 Irrigation delivery management. More than 25 countries are changing the
responsibility for managing irrigation systems from the central government to local
groups (Vermillion, 1997). Most of these changes are driven as much by the need to
cut government expenditures as by the desire to improve irrigation performance. By
reducing government subsidies and oversight, however, the hope is that such man-
agement transfers will accomplish both goals.
The largest management shift of this type in recent times occurred in Mexico,
where the management of 2.8 million ha of publicly irrigated land has been turned
over to farmer organizations. With the accompanying large reduction in government
subsidies, water fees have risen to cover costs. The irrigation districts are now about
80% financially self-sufficient, up from 37% prior to the transfer. The cost of irriga-
tion water, although higher, is still about 5% of total production costs, a typical value
for irrigated agriculture. It has yet to be determined whether local management has
promoted greater equity in water distribution and more efficient water use (Johnson,
1997).
1.4.4.2 Water pricing. The inability to agree on the economic and social values of
healthy freshwater ecosystems and water quality, as well as flood control, recreation,
irrigated agriculture, and tourism, has lead to conflicts in many areas of the world.
There is a wide spectrum of options, each with its own range of consequences for
various sectors and interests.
The price of water can be difficult to alter because of the many diverse viewpoints.
In many scenarios, charging the full cost for water without changing other institutional
and economic structures would put most irrigators out of business. However, charging
little or nothing, which is the situation in many irrigation projects, is a clear invitation
to waste water and increases the potential for conflict.
One option gaining popularity is a tiered pricing structure, which provides strong
incentives to conserve water to avoid higher rates. For example, irrigators might be
charged the low rate they are accustomed to paying for up to, say, 60% of their aver-
age past water use, a significantly higher price for the next 20%, and the full (high)
cost for the last 20%. Any water used above the average use would be at full cost plus
a penalty.
The Broadview Water District in California introduced a tiered water pricing struc-
ture in the 1990s. The district determined the average amount of water used in previ-
ous years for each crop. Irrigators were charged the customary rate for up to 90% of
the average water need. Any water deliveries above 90% were 2.5 times higher in
price. Even though they were still paying much less than the real water cost, the irriga-
tors had incentive to conserve. Depending upon the crop, water savings were from 9%
to 31% while crop yields were unaffected or increased (California Department of Wa-
ter Resources, 1998). Another example is Israel, which currently has a price structure
where 65% of full evapotranspiration (ET) is provided at a reduced price, but the price
increases exponentially for any additional water.
1.4.4.3 Water marketing. Removing barriers to water marketing can also lead to a
more equitable distribution of water and improved water use efficiency. The ability to
sell some water gives irrigators an incentive to conserve water so they can profit from
the sale of the resulting saved water. Formal water markets work only where farmers
have legally enforceable rights to their water, and where those rights can be sold. Aus-
tralia, Chile, Spain, Mexico, and many states in the Western U.S. now have laws and
policies that permit water markets. One important advantage is that this mechanism is
voluntary.
As with pricing, a variety of marketing options exist. Water rights owners can sell
their water on a seasonal basis or they could enter into a multi-year contract. They
could also sell their water rights to another user, in which case the legal water entitle-
ment is transferred permanently.
1.4.4.4 Water transfers. Traditionally, there have always been limited water trans-
fers within irrigated agriculture at the farm and project level, normally from field
crops to horticultural crops of higher value. Economic decisions have also caused ag-
ricultural water to be transferred to municipal and industrial uses in many areas. How-
ever, as demands on the world’s scarce freshwater supplies increase, water transfers
from agriculture to other sectors will become more and more common and, in fact,
some may become mandated by judicial and legislative processes. Artificial ground-
water recharge as a form of water banking, as well as the treatment of degraded aqui-
fer and soil waters for later reuse, are all components in water transfer considerations
and policies.
Mutual agreements have been reached between municipal and agricultural users
that allow an irrigation district or farmers to continue irrigating, but require they make
improvements that will conserve water, i.e., reduce the consumptive use component of
water diverted. The cost of the improvements is borne by the municipal user. Site con-
ditions often will determine the effectiveness of improvements in reducing water con-
sumption. For example, lining a canal whose seepage normally returns to the river
from which the water was originally diverted may not result in much, if any, reduction
in water consumed in the river basin.
An example of a mutual agreement to conserve irrigation water that could be used
by a municipality was reached in 1988-1989 between the California Metropolitan Wa-
ter District (MWD) and the Imperial Irrigation District (IID). The IID is located in the
Imperial Valley of Southern California and diverts water by gravity from the Colorado
River. Surface and subsurface flows from irrigated lands in the IID do not return to the
Colorado River from which water was diverted, but instead flow to the Salton Sea,
which lies about 70 meters below sea level. Improvements were agreed upon that
would reduce flows to the Salton Sea, but still enable irrigated farming to continue.
The improvements were estimated to conserve about 123 million cubic meters of wa-
ter annually within IID with the saved water available to the MWD for municipal use
(MacDonnell and Rice, 1994; National Research Council, 1996). The IID-MWD con-
servation program involved lining canals within the IID, improved flow-monitoring
structures, installation of non-leak gates, prevention or recovery of canal spills, instal-
lation of regulating reservoirs and seepage recovery systems, and system automation.
Implied in this agreement is that the IID would reduce its net annual diversion of
Colorado River water at Imperial Dam by the amount conserved and the MWD would
increase its diversion upstream at Parker Dam by a like amount. The MWD benefits
because the cost of the conserved water is less than $0.10 per cubic meter, much lower
than its best option for a new supply. The IID benefits from the cash payments and an
upgraded irrigation system, and no cropland is taken out of production because the
water transferred is generated through conservation (Gomez and Steding, 1998).
In another transfer agreement with the MWD, however, farmers were required to
remove irrigated land from production. In 1992, the MWD entered into an agreement
with the Palo Verde Irrigation District located on the west side of the Colorado River
in Southern California. The agreement stipulated that farmers who laid fallow a por-
tion of their cropland for two years would be paid $3,000 for each hectare left un-
planted. More than 8000 ha were left fallow by 63 farmers. The equivalent of about
10% of the MWD’s yearly water deliveries were then transferred and stored in federal
reservoirs on the Colorado River to be used anytime before 2000 (Loh and Steding,
1996). There are also some associated environmental agreements being developed
including a three-state, 50-year agreement covering restoration of 3300 ha of native
habitat along the Colorado River from Hoover Dam to the Mexican border.
A third transfer agreement, entered into in 2002, provides mechanisms for transfer-
ring water from agricultural to urban users, and serves as the basis for California to
settle nearly seven decades of disputes among its water agencies. It involves the IID,
the MWD, the San Diego County Water Authority, and the Coachella Valley Water
District. Implementation of this 75-year landmark agreement also requires California
to meet specific benchmarks for receiving Colorado River water (surplus and other-
wise) because the state has to reduce its annual Colorado River diversions from 6.4
billion cubic meters to 5.4 billion cubic meters of water over a 15-year period. The
agreement combines temporary fallowing of irrigated lands and the transfer of about
1.2 billion cubic meters while the IID implements on-farm and system water conserva-
tion measures paid for by the urban water districts. San Diego agreed to a payment
schedule for the water transferred plus $20 million to help cover socioeconomic im-
pacts to the local Imperial Valley communities and landowners over the 15-year pe-
riod as a result of the transfers. This is in addition to the land management, crop rota-
tion, and water supply transfer agreements with the Palo Verde Irrigation District men-
tioned above.
So far, few countries have the necessary institutional and incentive structures to
guide water competition. In Chile, where government policies encourage water mar-
keting, negative impacts seem to be minimal, primarily because farmers have only
sold small portions of their water rights to cities that have funded upgrading of exist-
ing irrigation systems and operational procedures.
1.4.4.5 Economics and incentives. Both rainfed and irrigated agriculture generally
provide only a marginal rate of return to growers in today’s world markets. Costs of
production are rising while crop prices are remaining static, and many growers need
subsidy payments to remain viable despite significant improvements in farming effi-
ciencies. Thus, many producers currently cannot afford to make expensive improve-
ments to enhance productivity.
From an economic perspective, many advanced irrigation methodologies have high
initial capital costs, which add increased production costs including energy, manage-
ment, agrochemicals, and land preparation. Researchers and action agencies need to
find ways to reduce all inputs, including management time, for growers to remain
competitive in a world market. Economists and politicians must find ways to move
world crop prices upward to where they truly reflect the increased costs of production,
including higher water prices.
It is expected that a considerable amount of the cost of any new agricultural infra-
structure and field improvements would be privately, rather than publicly, financed.
Private development of an intensive, irrigated greenhouse culture will no doubt ex-
pand greatly. Individuals and farmer-owned irrigation districts will be asked to fund
upgrades to existing facilities and systems, but it should not be a one-way street.
Recent years have seen governmental investments in irrigation research and infra-
structure reduced significantly. However, it is clear that society will have to change its
current attitude of minimizing investments in irrigated agriculture, and instead make
substantial additional investments towards improving irrigation infrastructure and
management to address increased water demands. Innovative policies must reflect the
need and mechanisms for equitably funding these investments, and urban users will
need to be much more involved in this process than has generally been the case. This
is starting to become apparent in Southern California as they attempt to address com-
plex regional water issues.
Economic and social policies that include incentives need to be implemented to
motivate farmers to reduce negative externalities and consider opportunity costs when
choosing irrigation and drainage strategies. Incentives can be either negative or posi-
tive, but should be directed both at inputs in agricultural production as well as efflu-
ents such as salt, silt, nutrients, and other constituents in surface runoff and deep per-
colation. They could include collaborative water markets, payments to irrigators for
achieving higher efficiencies, higher costs for water per unit area, service charges for
each delivery time, reductions in water rights to more closely match ET, or other ways
to encourage profitable deficit irrigation strategies. However, policies and incentives
have to be economically and culturally acceptable to growers and be accompanied by
realistic programs to upgrade and support improved technologies and management.
Positive incentives could include payments to growers who meet or exceed targeted
irrigation efficiency levels. Payments could be in the form of funds to improve irriga-
tion systems and monitoring equipment, but must also include ways to offset the in-
creased costs of management and labor associated with the improved efficiency. In-
creasing the actual prices farmers receive for their products by changing governmental
policies would also provide incentives and the means for them to improve production,
management, and environmentally beneficial practices in the face of declining water
availability and rising energy costs.
Another incentive could allow landowners to move water from poorly producing
soils (whether due to salinity, erosion, or just natural causes) to more productive areas
that may be outside an irrigation district’s boundaries. Some states, such as Washing-
ton, are allowing water spreading in certain areas where water saved by converting to
low water use crops or deficit irrigation can be moved to previously nonirrigated lands
as long as total historical ET is not increased.
Some grower and societal incentives to implement improved irrigation technologies
at the irrigation district and farm scales include: reduced labor requirements, lower
costs for treating water by reducing the volumes treated, lower costs for pumping wa-
ter, reduced costs for added distribution capacity in an area of growth, less leaching of
fertilizers and chemicals and degradation of groundwater, and sustained flows in seg-
ments of streams bypassed by irrigation diversions. These savings and potential envi-
ronmental benefits accrue both to the irrigation manager and, ultimately, the general
populace. In the future, both groups will likely perceive these as necessary incentives.
Increasing crop productivity while reducing the amount of applied water depends
on the ability of a particular type of irrigation system and the managerial skills of the
operator to correctly implement the water-saving practices and techniques. Frequently,
the controlling factors are the knowledge base of the grower and the existence of in-
centives to implement the improved practices. Thus, incentives should also reward
higher management levels.
A concern is that skilled labor for irrigation is becoming more and more limited
and many are unwilling to perform the required activities and technological tasks. In
addition, the average farming population age in the U.S. is in the 50s, and this age
group is less likely than younger farmers to make major improvements or adopt new
technologies without substantial incentives.
1.5 FUTURE DIRECTIONS
The discussion above describes standard practices that can be done with current in-
stitutional and technological systems. The next step is to look into the future with all
of its uncertainty. To aid in that effort, the National Research Council (1996) devel-
oped this list of likely future directions for irrigation in the U.S.:
Irrigation will continue to play an important role in the U.S. and the world for
the foreseeable future, although there will certainly be changes in its character,
methods, and scope.
The total irrigated area will likely decline, but the value of irrigated agriculture
will remain about the same because of shifts to crops of higher value.
The amount of water dedicated to irrigated agriculture will decline as societal
values change and competition for water increases.
A major factor in the sustainability of U.S. irrigation will be determined by our
ability to compete in global markets.
Under-financed irrigation operations or those with less-skilled managers will
tend to decline in number.
Previously, irrigation meant irrigation for agriculture. During the past 25 years
irrigation has become an important part of the turf industry, and irrigation for
urban landscaping and golf courses is growing steadily as urban populations in-
crease.
With time, increasing amounts of water will be removed from agriculture to sat-
isfy environmental goals. In conjunction with this, there will be increasing pres-
sures to reduce environmental degradation associated with irrigation.
Substantial research and educational challenges must still be addressed regarding
water availability, quantity, and quality, water use, and water institutions (National
Research Council, 2001, 2004). Changes in policy and incentives will clearly become
necessary. The following sections examine some potential issues and solutions to agri-
cultural water security issues. Urban water users will have a similar set of challenges
to reduce and modify water consumption.
1.5.1 Need for Innovation
Irrigation has been practiced for more than 6000 years, but more innovation has oc-
curred in this arena in the last 100 years than in all of the preceding centuries. Almost
every aspect of irrigation has seen significant innovation: diversion works, pumping,
filtration, conveyance, distribution, application methods, drainage, power sources,
scheduling, fertigation, chemigation, erosion control, land grading, soil water meas-
urement, and water conservation.
Major future improvements in water saving will be realized through innovative de-
sign and operation of integrated irrigation systems for both agricultural and urban set-
tings. It is obvious that all these technologies will have to continue to be improved and
implemented to better manage energy, water, and soil resources. Novel irrigation
techniques and management systems will be necessary to increase the cost-
effectiveness of crop production, improve water quality, improve water reuse capabili-
ties, reduce soil erosion, and reduce energy requirements while enhancing and sustain-
ing crop production and water use efficiency. In addition, innovative water polices and
institutional structures must evolve and foster emerging irrigation technologies.
Irrigation is a valuable technology, rooted in ancient tradition, and has proven to be
dynamic and flexible. However, new and improved strategies and practices are needed
to reduce surface and groundwater contamination from agricultural lands, conserve
water and energy, and sustain food production for strategic, economic, and social
benefits. Systems must be designed and managed to minimize health hazards due to
chemical applications of fertilizers and pesticides as well as to minimize insect infesta-
tion and parasitic diseases, such as the West Nile virus and malaria. The effects of
water conservation and reuse technologies on recreation, tourism, wetlands, and
aquatic ecosystems must be assessed and balanced with other societal needs.
Future irrigators will often be operating under various managed crop water deficit
scenarios. Increasing crop productivity while reducing the amount of applied water
implies that producers will often be managing irrigations under severe to moderate soil
water deficit conditions during part or all of the growing season. Techniques such as
partial root zone drying and regulated deficit irrigation will be more and more com-
mon on tree and vine crops as well as many annual crops (Chalmers et al., 1986; Fer-
eres et al, 2003). Techniques such as fallowing of irrigated fields in alternate years to
conserve water need to be investigated as to potential water savings and reduced agro-
chemical use.
Water reuse and treatment of impaired waters will be part of agricultural water se-
curity. Innovative approaches to groundwater recharge using treated and excess sur-
face waters for later withdrawals by a multitude of users will be an essential part of
future water resource programs..
The following brief sections present more details on some of the issues that agricul-
ture will have to implement to address water security issues. These measures will in-
clude: modernizing irrigation delivery systems and on-farm systems, improved levels
of management, strategies for local water supply enhancement, and biotechnological
advances in crop breeding and selection.
1.5.2 Modernizing Delivery and On-Farm Systems
Traditional approaches to modernizing irrigation projects have focused on minimiz-
ing water loss during delivery and maximizing field application efficiencies. These are
necessary first steps, but future water delivery systems and application techniques
must be modified to enhance grower flexibility in managing rates, irrigation frequen-
cies, and durations, as well as reduce water evaporation and other losses. Small, dis-
tributed internal regulation reservoirs, closed-conduit systems to reduce evaporation
and leave unused water in the distribution system, extensive automated water-level
controls, accurate automated flow measurement, and improved ways to reduce weed
growth on canal and lateral banks to minimize non-beneficial ET are all potential
means of improving water delivery efficiencies. Some of these features are just now
beginning to be implemented in a few modernized irrigation projects.
To maximize the potential of existing and emerging technologies, irrigators must

have the flexibility to manage rate, frequency, and duration of their water supplies.
Thus, the delivery system and the farm must be considered as one integrated unit with
two parts rather than two independent systems. With the imperative need to implement
the agreements and mandates discussed previously, the Imperial Valley in California
will be a proving ground for these concepts over the next couple of decades, and there
is much to learn. In this case, the delivery system will provide irrigation water to sat-
isfy the specific field condition (i.e., rate, frequency, and duration) that is calibrated
for each crop and irrigation condition. This will require extensive canal automation
and on-farm monitoring as well as economic incentives for achieving better water use
productivity. Positive payments for achieving a certain target efficiency or tailwater
levels rather than solely raising water costs are anticipated.
The following identifies some of the potential areas where innovation is likely to
improve delivery and on-farm systems in the 21st century:
Computers and wireless control systems will play an ever-expanding role. Cell
phone and satellite communications and internet technologies will likely play an
increasingly major part in management of irrigation systems. Feedback control
technologies for automating canal operations, surface and pressurized systems,
and drainage systems must be developed, tested, and supported by incentives.
On-farm systems will benefit from advanced technologies, such as precision ir-
rigation, site-specific management, remote sensing, within-field real-time sensor
systems, and decision support systems, which collectively have great potential to
facilitate reduction of water quantity and quality problems in irrigated agriculture.
The use of real-time irrigation scheduling techniques (sensor-based) and site-
specific precision applications of water through center pivot machines and mi-
croirrigation are the next steps in the evolution of those technologies.
It will be necessary to expand modern crop production technologies to less pro-
ductive rainfed and irrigated lands characterized by poor soils, low and unstable
rainfall, steep slopes, and short growing seasons to increase food production and
stimulate economic growth. Novel approaches will be needed to address these
areas.
Microirrigation with its many variations must be made less expensive before
most growers will be able to adopt and utilize these technologies, especially in
developing countries. Some localized efforts are ongoing and one company is
manufacturing tubing at relatively low cost, but these innovative efforts and
technologies need to be extended to other areas.
For developing countries, innovative research and extension education is needed
to provide and implement simple but efficient low-cost methods of irrigation
(e.g., pitcher irrigation) to make them easy to operate, suitable for the crop, and
acceptable to growers. There is also a huge need for low-lift pumps that are in-
expensive to buy and operate in these areas. Some of this is already being done
on relatively small scales, but there is much room for innovation.
Surface irrigation methods can be made more efficient using surge flow, dead-
level basins, and other techniques for more uniform infiltration along the length
of the field. Properly designed and operated level basins eliminate runoff, can be
quite efficient and uniform, and are relatively inexpensive to construct and oper-
ate. However, considerable investment for delivery system improvements, as
well as sensor feedback controls and automation for both the delivery and appli-
cation systems, is needed to fully realize the potential water savings.
Urban and agricultural irrigators will be the primary users of degraded waters.
New approaches and techniques will be required to safely minimize detrimental
effects while maintaining production goals.
Farm- and district-level drainage systems will require improved design, evalua-
tion and simulation models defining the physical limits. Automated control sys-
tems will assist in providing a more uniform soil water environment for plant
growth to improve productivity and minimize the volumes of drainage waters
requiring treatment, especially in arid areas.
Water table elevations can be managed to permit subirrigation, if the groundwa-
ter is relatively shallow and of suitable quality, by controlling water tables or in-
ducing water tables with irrigation applications. Subirrigation has been practiced
successfully in climate regions ranging from humid to arid. Using the effluent
from deeper subsurface drainage systems as a source of irrigation water has
proven effective in many regions of the world. The biggest concern is the quality
(i.e., salinity) of the drainage system effluent.
There is still no reliable, inexpensive electronic soil water sensor that matches or
exceeds the accuracy and repeatability of neutron scattering devices. Innovative
development of such sensors is essential for water management, particularly un-
der deficit conditions.
Many of the needed and evolving technologies will require stand-alone, spatially
distributed electrical power to be feasible. Controllers, monitoring equipment,
and communications devices must be low power consumers. Photovoltaic, wind
turbine, and storage systems will need to be developed and implemented at low
cost at the farm or field level.
Economies of scale have led to large field sizes for irrigated production in many
areas, and engineers have been very successful in designing pressurized irriga-
tion systems that apply water quite uniformly over these fields. However, the
challenge over the next 50 years for the irrigation industry and designers is to
develop highly efficient systems that are also suitable for small-scale farms and
provide the necessary extension education to equip the farmers with the skills to
run them.
Although widely variable, it is estimated that 1% of the global water storage ca-
pacity in reservoirs is lost each year to sedimentation (Palimeri, 1998), decreas-
ing the ability to store water. Innovative methods to reduce erosion at the water-
shed and basin scale will be needed to increase the life of reservoirs for storage,
flood control, and recreation uses.
Innovation will be required to enable adoption of more efficient irrigation methods.
For example, high-frequency drip irrigation and other microirrigation methods have
been shown to increase the yield and quality of fruit and vegetable crops through re-
duced water and nutrient stresses. Tied to an effective soil water monitoring program,
good design, and appropriate management practices, microirrigation can be 95% effi-
cient or better. A modification of center pivot irrigation called “low energy precision
application” or LEPA has been found to be 95% efficient as well. However, microirri-
gation and LEPA irrigation are being used on less than 1% of irrigated lands world-
wide.
1.5.3 Management
Inherent in the evolution of on-farm management discussed above will be the inte-
gration of irrigation, fertilizer, and pest management strategies into systems that opti-
mize total management practices for temporal and spatial variability. This should re-
sult in substantial labor, water, and energy savings and minimize losses to the
groundwater.
Improved irrigation technologies usually result in reduced labor requirements, but
require expanded management. However, most producers who irrigate do not have
sufficient time to properly manage all their critical inputs. At certain times of the year,
a producer’s time is extremely valuable in terms of net returns from the crop, and irri-
gation water management is often not as high a priority as other concurrent cultural
factors. Thus, decision support aids must also be developed that improve the pro-
ducer’s ability to implement decisions quickly and easily because climatic variations
and pest outbreaks require precisely timed water and chemical applications on a daily
and seasonal basis. The decision support process must also provide accurate predic-
tions of application efficiencies and uniformities to improve management flexibility.
Ensuring the success of irrigated farming enterprises will require the development
of reliable and more-timely information on field and plant status to support the deci-
sion-making processes. Current plant models capable of predicting the physiological
needs of a crop over space and time tend to be complex and impractical for real-time
on-farm management. Furthermore, most of these models are point models that are not
sensitive enough to adequately predict site-specific plant needs across a field in a
timely fashion. Simpler, more appropriate models might be used but will likely need
frequent updating via automated, field-based sensor systems to readjust model vari-
ables to ensure reasonable tracking and spatial predictions of field conditions.
A more focused approach will be required for the development of spatial and tem-
poral management strategies that address site-specific crop water, nutrient, and pest
management requirements, and irrigation scheduling in real time. The ultimate goal
should be to integrate controls, sensor systems, plant and physical models, and other
techniques to provide workable solutions that reduce time requirements for busy deci-
sion makers while improving their management capacity.
Self-propelled irrigation systems, such as center pivots and linear moves, are par-
ticularly amenable to site-specific approaches because of their current level of automa-
tion and large area coverage with a single pipe lateral. These technologies hold much
promise for spatially varying water and agrochemical applications to match differ-
ences in irrigation, nutrient, and pesticide requirements throughout the field. This
could potentially increase productivity and minimize adverse water quality impacts.
By aligning irrigation water applications with variable water requirements in the field,
total water use may be reduced, decreasing deep percolation and surface runoff. This
also suggests that water-soluble fertilizers and other agrochemicals can be effectively
applied spatially through the irrigation system to match changing conditions across a
field. However, there is a need to develop more efficient methods of applying crop
amendments (e.g., nutrients, pesticides) with irrigation systems that will reduce agro-
chemical usage, improve profit margins, and reduce environmental impacts. In some
cases, it may be necessary to schedule and irrigate on a plant-by-plant basis. This
technology is available today, but is not yet economical or practical. Research is
needed to give farmers confidence that the use of these technologies is practical and
potentially valuable in improving irrigated crop production.
Much effort has gone into ET research in the past several decades, and it is one of
the most accurate estimates available to irrigation delivery and on-farm managers as
long as plants are at relatively low stress levels. However, much additional informa-
tion is still required on the yield and crop quality affect at various ET levels at differ-
ent growth stages of both major and minor crops that are in managed soil water deficit
conditions, perhaps in combination with increased soil salinity levels. The long term
effect of deficits on perennial crops also needs more research to ensure sustainability.
Critical information is lacking on actual ET under irrigation systems that are being
used for environmental modification to protect the crop from cold temperatures or
excessive heat. In addition, research on irrigation management and water use require-
ments is lacking for intercropping production systems. More research is also needed
on how to truly address spatial variability of ET across large fields. This information
will be extremely critical to future determinations of both agricultural and urban water
rights, water banking, and governmental allocations.
Remote sensing of plant and soil status using integrated satellite, aerial, and field
level plant- and soil-based sensor systems is another way of providing information, but
it also needs further development to improve spatial-temporal modeling and on-farm
management as well as irrigation district operations. Better systems and methods ca-
pable of precisely measuring specific plant parameters (e.g., nutrient status, water
status, disease, and competing weeds) on a timely basis are needed to improve crop
modeling and thus improve within-season management.
Real-time, on-the-go irrigation scheduling could be very effective in improving wa-
ter management when based on distributed networks of farm-level microclimate and
soil water sensor stations that feed into a microprocessor control system to manage
irrigations based on rules established by the producer. This effort must be supported
by expanded agricultural weather networks with a greater spatial density and grower-
friendly information delivery systems for scheduling irrigations combined with pest
management and marketing information. Input from distributed weather networks
must be integrated with other information sources to effectively contribute to on-farm
and irrigation district decision support processes.
In addition to irrigation, self-propelled irrigation systems also provide an out-
standing platform on which to mount sensors for real-time monitoring of plant and soil
conditions, which in turn provide input to the control system for optimal environ-
mental benefits. Similar sensor technologies tied to a global positioning system can
also be mounted on-farm equipment so that site-specific information is collected each
time the grower is in the field. Early detection of diseases, weeds, insects, and even
nutrient deficiencies would allow more economical spot treatments of small areas
within a field.
1.5.4 Strategies for Local Water Supply Enhancement
Techniques to capture more water in either surface or subsurface impoundments
and aquifers will only be briefly mentioned because they are beyond the scope of this
monograph. However, it is worthwhile to introduce a few practical measures to in-
crease local water availability on the farm and perhaps stimulate additional efforts in
this critical area. Unger and Howell (1999) present a much more complete review of
the pros and cons of various dryland and irrigated techniques available for local water
conservation.
Strategies to be considered as a part of these activities include irrigation system de-
sign for higher efficiency, successful treatment and reuse of degraded waters, reducing
evaporation losses, site-specific applications, managed-deficit irrigations of tree and
vine crops, and techniques to minimize leaching. However, they will not be repeated
as many of these have already been introduced and some are discussed in more detail
in other chapters.
Perhaps the most neglected local water source is the capture, diversion, and storage
of precipitation (both rainfall and snowmelt) where it falls in the field. The goal is to
capture water falling on croplands and supplement this with water from nearby non-
cropped areas. This is often referred to as “water harvesting” and is used around the
world to supplement irrigation by improving water availability. It can be an effective
water conservation tool, especially in arid and semi-arid regions (Unger and Howell,
1999).
These practices may also benefit soil fertility by capturing dissolved nutrients, and
may also provide broader environmental benefits through reduced soil erosion. Where
feasible, small reservoirs or storage tanks that minimize evaporation and seepage
losses could be constructed at both regional and farm scales to capture runoff from
storms for later use. Where groundwater is a viable water supply, techniques to re-
charge aquifers, such as recharge wells and percolation ponds, should be considered to
increase supplies.
Substituting a crop with low consumption for a crop with higher water consumption
or switching to crops with higher economic benefit or productivity per unit of water
consumed by ET can provide water savings. Reallocation of water from low-value
crops to higher-value crops can also increase the economic productivity of water.
Conservation tillage practices (e.g., strip till, ridge till) that maintain surface crop
residues or mulches to reduce evaporation can increase the share of rainfall that goes
to infiltration and ET. Contour plowing, which has been promoted as a soil-preserving
technique, can detain and infiltrate a higher share of the precipitation. Techniques such
as precision leveling can capture a higher percentage of effective rainfall and improve
soil water management.
Furrow diking in sprinkler irrigated areas (e.g., LEPA) to hold water where it falls
is beneficial. The use of mulches for weed control also reduces non-beneficial ET and
soil evaporation. Drip irrigation technologies can conserve water by greatly reducing
soil evaporation and maximizing water use efficiencies. These strategies could also
incorporate the use of alternative cropping systems including winter crops and deep-
rooted cultivars that maximize the utilization of soil water and nutrients. Weed control
to reduce non-beneficial water use is critical.
Another technique employs a series of dikes or terraces placed on the contour to
capture precipitation as it runs off and convey it to cropland. This is currently used in
certain areas of Australia where some storm water runoff is captured at the low ends of
large properties where large pumps convey the collected water into large aboveground
tanks for livestock, irrigation, and other uses.
1.5.5 Cultural/Crop/Plant Selection Factors
In combination with molecular techniques and processes, plant breeders and mo-
lecular biologists will be selecting plants that use less water with a concurrent toler-
ance to heat stress (less transpirational cooling) and greater tolerance to salinity, dis-
eases, weeds, and insect pests while producing high-quality crops. The problem is that
no single gene or series of genes has been specifically identified with increasing yields
(Mann, 1999). Comparatively small advances in disease and pest resistance as well as
drought resistance are forecasted.
Crop breeding will have greater focus on more efficient water use by selecting for
optimal growing season lengths and harvest dates that take maximum advantage of
rainfall timing at critical growth stages for each region. These activities must be ac-
companied by multidisciplinary, collaborative research into soil management, pest
control (by tillage, cultural, biological, and chemical means), green manures, and the
yield and quality responses of the new cultivars to both full and deficit irrigation
amounts, increased salinity levels, degraded reuse water from a variety of sources, and
the effects of climate change.
Intercropping, which is a cultural practice in some irrigated areas in the world, may
be advantageous to help satisfy future production expectations if expanded into rainfed
and irrigated areas that are already highly productive. However, because varieties that
can tolerate water and nutrient stress must be developed as well as production strate-
gies for interseeding and for the use of short season crops to facilitate production of
multiple crops per year in temperate to humid areas, substantial progress in these areas
is expected to be a lengthy process.
REFERENCES
California Department of Water Resources. 1998. Drainage management in the San
Joaquin Valley: A status report. Sacramento, Calif.: San Joaquin Valley Drainage
Implementation Program.
Chalmers, D. J., G. Burge, P. H. Jerie, and P. D. Mitchell. 1986. The mechanism of
regulation of Bartlett pear fruit and vegetative growth by irrigation withholding and
regulated deficit irrigation. J. American Soc. Hort. Sci. 111: 904-907.
Clemmens, A., and R. Allen. 2005. Impact of agricultural water conservation on water
availability. In Proc. of the World Water and Environmental Resources Congress.
Reston, Va.: American Society of Civil Engineers.
Dobrowolski, J. P., and M. P. O’Neill, eds. 2005. Agricultural water security listening
session final report. Washington D.C.: USDA-REE.
FAOSTAT. 2005. FAOSTAT statistical database. Rome, Italy: Food and Agriculture
Organization of the United Nations. Available at: www.faostat.fao.org.
Fereres, E., D. A. Goldhammer, and L. R. Parsons. 2003. Irrigation water management
of horticultural crops. HortSci. 38: 1036-1043.
Gleick, P. 1998. The World’s Water 1998-1999: The Biennial Report on Freshwater
Resources. Washington, D.C.: Island Press.
Gollehon, N., 2002. Irrigation and water use: Questions and answers. USDA-Economic
Research Service. Available at: www.ers.usda.gov/Briefing/WaterUse/Questions
/qa3.htm.
Gomez, S., and A. Steding. 1998. California Water Transfers: An Evaluation of the
Economic Framework and a Spatial Analysis of the Potential Impacts. Oakland,
Calif.: Pacific Institute.
Hutson, S. S., N. L. Barber, J. F. Kenny, K. S. Linsey, D. S. Lumia, and M. A.
Maupin. 2004. Estimated water use in the United States in 2000. U.S. Geological
Survey Circular 1268. Reston, Va.: U.S. Department of the Interior, Geological
Survey.
Intergovernmental Panel on Climate Change. 2001. Climate change 2001: Synthesis
report. The IPCC 3rd assessment report. Summary for policymakers. New York,
N.Y.: Cambridge Univ. Press. Available at: www.ipcc.ch/pub/pub.html.
Johnson, S. H. 1997. Irrigation management transfer: Decentralizing public irrigation
in Mexico. Water International 22: 159-167.
Jones, W. I. 1995. The World Bank and Irrigation. Washington, D.C.: World Bank.
Loh, P., and A. Steding. 1996. The Palo Verde Test Land Fallowing Program: A
Model for Future California Water Transfers? Oakland, Calif.: Pacific Institute.
MacDonnell, L., and T. Rice. 1994. Moving agricultural water to cities: The search for
smarter approaches. Hastings West-Northwest J. 2: 27-54.
Mann, C. C. 1999. Future food: Crop scientists seek a new revolution. Science
283(5400): 310-314.
Milesi, C., S. W. Running, C. D. Elvidge, J. B. Dietz, B. T. Tuttle, R. R. Nemani.
2005. Mapping and modeling the biogeochemical cycling of turf grasses in the
United States. Environmental Mgmt. 36(3): 426-438.
Myers, N. 1997. Perverse studies: Their nature, scale, and impacts. Chicago, Ill.:
Report to MacArthur Foundation.
NRC (National Research Council). 1989. Irrigation-Induced Water Quality Problems.
Washington, D.C.: National Academy Press.
NRC (National Research Council). 1996. A New Era for Irrigation. Washington, D.C.:
National Academy Press.
NRC (National Research Council). 2001. Envisioning the Agenda for Water Resources
Research in the Twenty-First Century. Washington, D.C.: National Academy Press.
NRC (National Research Council). 2004. Confronting the Nation’s Water Problems:
The Role of Research. Washington, D.C.: National Academy Press.
Newson, M. 1992. Land, Water and Development: River Basin Systems and their
Sustainable Management. London, UK: Routledge.
O’Neill, M. P., and J. P. Dobrowolski. 2005. CRREES Agricultural Water Security
White Paper. Washington, D.C.: USDA-REE CSREES.
Palmieri, A., 1998. Reservoir sedimentation and the sustainability of dams. 1998 In
Proc. of the World Bank Water Week Conference. Annapolis, Md.: World Bank.
Postel, S. 1999. Pillar of Sand: Can the Irrigation Miracle Last? New York, N.Y.:
Worldwatch Books, W. W. Norton & Co.
Postel, S. 2000. Redesigning irrigated agriculture. In Proc. 4th Nat’l Irrigation Symp.,
1-12. R. G. Evans, B. L. Benham, and T. P. Trooien, eds. St. Joseph, Mich.: ASAE.
San Joaquin Valley Drainage Program. 1990. Fish and Wildlife Resources and
Agricultural (0.1) Drainage in the San Joaquin Valley, California.
Seckler, D., and A. Amarasinghe. 2000. Chapter 3: Water supply and demand, 1995 to
2025: Water scarcity and major issues. In World Water Vision. Battaramulla, Sri
Lanka: International Water Management Institute. Available at: www.iwmi.cgiar.
org /pubs/WWVisn/WWSDOpen.htm.
Tanji, K. K. 1990. Agricultural salinity problems. In Agricultural Salinity Assessment
and Management. K. Tanji, ed. New York, N.Y.: American. Soc. Civil Engineers.
Unger, P. W., and T. A. Howell. 1999. Agricultural water conservation—A global
perspective. J. Crop Production 2(4): 1-36.
USDA-NASS (U.S. Department of Agriculture National Agricultural Statistics Service).

2004. 2002 Census of Agriculture. Vol. 3: Special studies. Part 1: Farm and ranch
irrigation survey (2003). Available at: www.nass.usda.gov/census/census02/fris/
fris03.htm.
U.S. Department of Commerce. 1978. 1974 Census of Agriculture. Chapter I:
Irrigation of Agricultural Lands, I-1 to I-63. Part 9: Irrigation and Drainage on
Farms. Washington, D.C.: Bureau of the Census.
U.S. Department of Commerce. 1983. 1982 Census of Agriculture. Vol. 1, Part 51:
United States Summary and State Data. Washington, D.C.: Bureau of the Census.
Vermillion, D. L. 1997. Impacts of Irrigation Management Transfer. Colombo, Sri
Lanka: Irrigation Water Management Institute.
Western Water Policy Review Commission. 1998. Water in the West: Challenge for
the Next Century. Springfield, Va.: NTIS.
Wilcove, D. S., and M. J. Bean. 1994. The Big Kill: Declining Biodiversity in
America’s Lakes and Rivers. New York, N.Y.: Environmental Defense Fund.
World Bank. 1991. India Irrigation Sector Review, Vol. 2. Washington, D.C.: World
Bank.
CHAPTER 2
SUSTAINABLE AND
PRODUCTIVE IRRIGATED
AGRICULTURE
Marvin E. Jensen (USDA-ARS,
Fort Collins, Colorado)
Abstract. A renewable water supply is required for sustained irrigated agriculture.
Irrigated agriculture can remain productive indefinitely with assurance of water sup-
plies and infrastructure to deliver water as needed for crop production. Emerging
issues facing designers of irrigation systems are competition for limited and declining
water resources, climate variability, effects of leasing water during drought years on
annualized water costs, environmental concerns, refined irrigation scheduling, and
drainage and salinity control issues. Designers of irrigation systems must consider the
role of irrigation in food production, and the productivity of water used for irrigation.
Keywords. Climate-based irrigation scheduling, Deficit irrigation, Design chal-
lenges, Environmental concerns, Sustainable development, Water, Water accounting.
2.1 INTRODUCTION
2.1.1 Scope of the Chapter
Since the first edition of this monograph (printed in 1980), numerous papers and
books have been published on irrigation systems and practices, trends in food produc-
tion, climate and water requirements, irrigation experiences in arid, semiarid, subhumid,
and humid areas, and various beneficial uses of irrigation water. Some of these subjects
are covered in detail in other chapters. This chapter highlights recent trends and emerg-
ing issues facing designers and operators of farm irrigation systems. Sustainable irri-
gated agriculture depends on a renewable source of water and adequate drainage for
control of soil salinity. Increasing competition for limited water supplies will require
increasing attention to the water productivity, i.e., the production of the marketable
component of a crop per unit of water consumed. Water consumed is that fraction of
water applied that is evaporated, transpired, or combined within the crop and is no
longer available for other uses. The fraction of water applied that is not consumed, i.e.,
irrigation return flows, and other environmental concerns will need to be addressed by
designers of irrigation systems. Maximizing crop yields will no longer be the main ob-
jective of irrigation; more attention will be on maximizing net benefits of irrigation.
34 Chapter 2 Sustainable and Productive Irrigated Agriculture
2.1.2 Water, Plant Growth, Irrigation, and Crop Production

Water is essential for plant growth. Seeds need water to germinate, most soils must
be moist for seedlings to emerge, and many plant growth functions require water. Wa-
ter provides the transport mechanism within plants for nutrients and products of pho-
tosynthesis. However, most of the water consumed by plants is by transpiration, a
process that is controlled by climate and leaf area when soil water is not limiting. The
combined processes of evaporation from the soil and plant surfaces and transpiration
is called evapotranspiration (ET). Procedures for estimating ET are given in Chapter
8.
Irrigation supplements precipitation by providing water essential for plant growth,
in addition to controlling soil salinity and enhancing the root environment. The impact
of irrigation on crop yields is greatest in arid and semiarid climates and in humid and
subhumid climates during drought years. Irrigated crop yields are higher and more
stable than yields on adjacent dryland areas (often referred to as rainfed areas). Na-
tional famines, once common, have been avoided because of irrigation. Irrigation was
a major factor in the success of the Green Revolution and will become even more im-
portant in the future to provide food for the ever-increasing world population. Stability
of crop production under irrigation also provides economic stability to independent
farmers and to communities. Without irrigation, increases in agricultural yields and
outputs that have fed the world’s growing population would not have been possible.
Irrigation has also stabilized food production and prices (Rosegrant et al., 2002).
Irrigation is also used for many secondary purposes. For example, it is used to con-
trol frosts in orchards and citrus groves, and to condition the soil when preparing
seedbeds and harvesting root crops such as potatoes, sugar beets, and peanuts (see
Section 2.6). More recently, irrigation has become a common and environmentally
acceptable method of making economic use of waste effluents for crop production.
Previously, most of these effluents were merely disposed of in some manner, such as
being discharged to rivers and streams, evaporated in holding ponds, or applied to
nonproductive vegetation to accelerate the evaporation process.
2.1.3 Sustainable Irrigated Agriculture
During the past decade, many authors have addressed the issue of sustainable de-
velopment and sustainable agriculture and agricultural systems (Edwards et al., 1990;
Hatfield and Karlen, 1993). Their publications address issues of soil fertility, crop and
pest management, soil erosion, soil salinity, waterlogging, economics, environment
and water quality, bioresources, ecology, policy, sociology, and socioeconomic issues.
The quantity of renewable water resources, which is the most vital component of sus-
tainable irrigated agriculture, has been addressed to a limited extent. Sustainable crop
production on irrigated lands will require a sustainable and renewable water supply.
Seckler et al. (1998) prepared a comprehensive analysis of global water demand and
supply from 1990 to 2025. A water balance approach was used to derive estimates of
water supply and demand for individual countries and regions and to identify major
regions of water scarcity. During an international conference on sustainability of irri-
gated agriculture, various views of on-farm water management, water quality, and
capacity building were discussed along with country and regional concerns in Africa,
Asia, and China (Pereira et al., 1996).
The Council for Agricultural Science and Technology (CAST, 1992) addressed the
impacts of climate change on U.S. agriculture and the limitations of renewable water
resources on adapting to climate change. A committee of the Board of Water Science

and Technology of the National Research Council (NRC) conducted a thorough study
of the future of U.S. irrigated agriculture (NRC, 1996). The Committee reported that
the availability of water has been, and is likely to remain, the principal determinant of
the status of irrigation in the western U.S. The availability of water is becoming in-
creasingly important to irrigation in eastern states as well. The cost of water and de-
mands on water resources are increasing.
In 1989, the Food and Agriculture Organization (FAO) of the United Nations de-
veloped an international action program for the 1990s on water and sustainable agri-
cultural development (FAO, 1989). It emphasized the need for long-term strategies
consistent with limited water resources and competing demands for water. The NRC
Committee on International Soil and Water Research and Development stated that one
of the challenges we face in developing agricultural strategies that is truly sustainable
is maintaining the resource base, the soil and water resources that make agriculture
possible (NRC, 1993).
Globally, the largest threat to sustainable irrigated agriculture has been, and still is,
waterlogging and soil salinity. During the three decades from 1960 to 1990, the world
witnessed a historic expansion of irrigated land. During this period, irrigated agricul-
ture played a major role in increasing food production. Currently, an estimated 36% to
47% of the world’s food is produced under irrigation (Chapter 1). During the rapid
expansion, some principles of irrigation design, operation and maintenance, and water
balance appear to have been compromised resulting in waterlogging and soil salinity.
Soils become waterlogged if water is imported in addition to precipitation in excess of
crop ET and adequate drainage is not provided to remove the excess. Increases in soil
salinity usually follow waterlogging because salts present in irrigation water become
concentrated near the soil surface as evaporation from the soil removes pure water (see
Chapter 7).
In some developing countries, the continuing availability of relatively low cost en-
ergy may be a serious restraint on converting common surface irrigation systems to
modern, more efficient systems that require more energy input. Thus, energy required
for irrigation could also become a factor affecting the sustainability of irrigated agri-
culture.
2.1.4 Productive Irrigated Agriculture
Some societies have practiced irrigation for centuries without adverse soil and envi-
ronmental effects. In general, irrigated agriculture can remain productive indefinitely
if a reasonably stable renewable water supply of suitable quality is available and the
volume of applied water does not exceed ET and the drainage capacity, so that soils
are not waterlogged. Perry (2002), referring to countries that have developed and con-
trolled their water resources for decades and centuries, listed the essential elements of
sustained water resources management as: knowledge of resource availability, govern-
ing policies, assigned priorities for developed water among users, allocation rules and
procedures, defined roles and responsibilities, and infrastructure to deliver the service
to each water user.
In regions where water is scarce, agriculture may need to temporarily relinquish
some water supplies in drought years to provide for domestic and other higher value
water uses. This has become common practice in some areas of the western U.S. and
will become a more common practice internationally. This practice may not change
the productivity of the irrigated land in a given year, but will affect long-term mean
production potential from an area because not all lands will be irrigated fully each
year. Lands not fully irrigated may be planted to a crop and some crops may not be
irrigated for near maximum yields.
In addition to water, other inputs are needed to maintain productive irrigated agri-
culture. When irrigation removes the main constraint affecting crop yields (water),
then other factors such as plant nutrition and changes in soil salinity may limit crop
production.
2.1.5 Emerging Issues Facing Designers
Designers of farm irrigation systems in the 21st century face many new challenges.
Competition for limited water supplies will require irrigators to have greater control of
the amount and uniformity at which the water is applied over the fields. Typically, this
will increase the costs of new or modernized irrigation systems. Future renewable wa-
ter supplies may not be as dependable as they have been in the past. For example, in
parts of the Great Plains farmers are faced with decreasing flow rates from wells as the
water table is lowered in the vast Ogallala aquifer. This aquifer has very limited natu-
ral recharge over much of its surface area. If a farmer already has an operating well on
the farm, the efficiency of the pump will decline as the water table drops and the flow
rate decreases. Farmers must consider the potential return on investments required to
modernize and replace the pump and existing irrigation systems with more modern
and efficient systems. The alternative is to return to dryland farming. The irrigation
system designer must take into account the effects of decreasing flow rates and limited
water supplies on irrigation system efficiencies. Particularly in regions like these,
long-term planning is necessary and designers need to consider long-term returns, not
just irrigation to meet the component of ET that is not not provided by precipitation or
the consumptive irrigation requirements.
The farmer also must consider alternative management practices to cope with the
relatively large variation in seasonal precipitation in semiarid areas. Studies in the
High Plains of Texas have shown that in most years, limited irrigation of drought-
tolerant crops can significantly increase yields and the productivity of water applied.
In years of above-average rainfall, irrigation may be of limited benefit. For example, a
three-year study in Kansas showed that when precipitation was about 30% above nor-
mal (during 1989 and 1990), there was no benefit from irrigating grain sorghum [Sor-
ghum bicolor (L.) Moench]. During the following year (1991), precipitation was be-
low normal, and sorghum and wheat yields as well as water use efficiency (WUE)
were higher in irrigated treatments than on dryland treatments (Norwood, 1995). WUE
is commonly defined as the marketable product produced per unit of water consumed
in ET. In 1991 (the year with below normal precipitation) WUE tended to be greatest
from the first increment of irrigation water applied. This example illustrates how
weather variables can complicate the design of an optimal irrigation system in semi-
arid areas.
Western U.S. water rights have long protected investments in irrigation facilities.
But water laws are changing making it easier for farmers to sell the consumptive use
(CU) component of their water right to users who are willing to pay more for the water
than the farmer’s profit potential from irrigation. If some farmers within a project or
district sell their entire CU rights rather than lease it for a season, the cost of operation
and maintenance of the district’s main distribution system will be distributed to the
remaining farmers who continue to irrigate. Where this situation occurs or is expected
to occur, designers need to give the farmer information about the alternatives of mod-
ernizing a system, rather than maintaining the present system, and also about the op-
tions of selling the water right and terminating irrigated farming operations. More
commonly, farmers are leasing their water during water-short years through various
mechanisms (NRC, 1992). Under these conditions, an irrigation system may not be
used each year, which will also affect annualized system costs.
Distributing fertilizers and pesticides in irrigation water (chemigation) has become
commonplace and can be very efficient. Chemigation requires new design considera-
tions compared with traditional systems. In addition, new environmental concerns and
associated regulations affecting return flows to streams or to the groundwater are be-
coming significant factors that the designer must consider.
Many states have installed automated weather stations to measure and disseminate
estimates of evaporative demand or crop ET. Use of such data by irrigation managers
to schedule irrigations can enhance the productivity and efficient use of water re-
sources (Ley, 1995). Designers of modern irrigation systems must consider that, in the
future, most farmers will become dependent on real-time estimates of ET and soil wa-
ter depletion when scheduling their irrigations and determining the amounts that
should be applied. The designer needs to consider simulating multiple years of opera-
tion based on estimated daily ET using procedures described in Chapter 8. For high-
value crops, many new irrigation systems will be automated with irrigations initiated
using soil water sensors, computed depletion of soil water, or remote sensing tech-
niques.
2.2 ROLE OF IRRIGATION IN FOOD AND FIBER PRODUCTION
2.2.1 Population Growth and Renewable Water Supplies
World population growth and irrigation are related, i.e., an increasing population
requires increased irrigation. World population is increasing about 1.2% per year
(Population Reference Bureau, 2005); it was about 5.3 billion in the early 1990s and
increasing at a rate of 93 million people per year. The U.N. Population Division sug-
gested that the global population will increase to 9.3 billion by 2050. With the in-
creased population comes increased food needs.
Thus, population growth and its effects on water supplies and food requirements is
a variable that must be considered by irrigation system designers and operators, par-
ticularly in many developing countries. In the U.S., population growth probably will
not be significant in the near future. However, hotspots of water scarcity occur in the
Colorado, Rio Grande, and Texas Gulf basins (Rosegrant et al., 2002). Irrigation de-
signers must consider planning at a regional scale and be aware of current regional
plans and expected new regulations in order to properly design irrigation systems.
Rosegrant et al. (2002) provided a useful table of 1995 and projected 2025 populations
by country and region using data from several sources.
The world annual per capita internal (i.e., within a country) renewable water supply
is about 7000 m3, but this varies greatly between regions. For instance, in the Middle
East and North Africa, the International Bank for Reconstruction and Development
(IBRD) estimates the per capita water supply to be only about 1000 m3 (IBRD,
1992b). Renewable water supplies are finite and can essentially be considered as a
constant within a region. The rate of population growth is often very high in regions
that already have some of the lowest per capita water supplies. For example, the
weighted average rate of increase in population in the Middle East in the late 1980s
was 2.7% per year (Jensen, 1990). In that region, the per capita water supply, which is
already low, would be expected to decrease to half its present value in just 26 years.
The estimated 2005 population in this region was 47% higher than in 1989 (Popula-
tion Reference Bureau, 2005). The weighted average annual rate of increase in 2005
had decreased slightly to 2.2%.
2.2.2 Increasing Food Demands Relative to Water Supplies
As populations increase, competition among agricultural and non-agricultural water
uses will affect irrigated agriculture in many semiarid and arid areas. These pressures
will become more severe as water supplies are more fully utilized. Conflicts will in-
crease. Solutions to these conflicts will require revisions of water law and allocations.
Australian scientists, for example, are proposing a landmark national water trading
framework to define water rights for irrigation. This water rights system would be
based on banking, share trading, and Torrens Title registration procedures which
would allow water to be traded via electronic transfers with licensed brokers and clear
trading rules. The system would have three components: an entitlement, an allocation,
and a use license (CSIRO, 2002). The authors claim that the advantage of this system
is that it would provide flexible control as climate, economic, and technical circum-
stances vary.
In the future, the price of water will be more closely related to its real economic
cost. This will affect irrigated agriculture because the current value of the marketable
product of some crops per unit volume of water consumed is low. For example, only
about 2 kg of maize grain valued at about $0.20 can be produced in arid and semiarid
areas per cubic meter of water consumed in ET. For many semiarid and arid areas, it
will not be economical to import and use a cubic meter of water (= 1 ton of water) to
produce 2 kg of grain even if a renewable source of water was available. These areas
will be forced to use local water for domestic purposes and to import more feed and
food supplies (Jensen, 1993). Importing grain rather than importing water to produce
grain locally is sometimes referred to as importing virtual water. Where agriculture
has been the primary source of foreign currency, depending on virtual water may re-
quire generating foreign currency from other sources.
Few irrigation projects are planned to supplement dryland agriculture in semiarid
areas. However, limited irrigation can significantly increase the production per unit of
precipitation and stored soil water consumed by ET. For example, in the High Plains
of Texas, dryland grain sorghum produces from 0.45 to 0.85 kg/m3 of water with
yields from 1000 to 3000 kg/ha. Limited irrigation can increase yields to about 5000
kg/ha and average productivity to about 1.2 kg/m3 (Jensen, 1984). The first 100-mm
increment of irrigation water had the most effect by increasing the yield from 2500 to
4000 kg/ha. The productivity of this increment of irrigation water would be 1.5 kg/m3.
This is an example of the future role for irrigation in semiarid areas, especially where
renewable groundwater supplies are available. Irrigation development must be inte-
grated with dryland production rather than planned, developed, and managed as a
competitive enterprise (Jensen, 1993).
2.2.3 Downstream Environmental Effects
Downstream environmental effects may include reductions in wetlands and lakes as
crops consume water that formerly flowed to lakes and wetlands. One dramatic exam-
ple is in the former Soviet Union. In 1918, the Soviet Union decided to become self-
sufficient in cotton. To irrigate cotton fields, they began diverting 55 km3 of water per
year from two rivers that feed the Aral Sea (Sun, 1988). Over the years, little water
reached the Aral Sea and it has shrunk drastically in size. Huge salt storms have car-
ried salt from the shores of the shrinking basin great distances, damaging agricultural
lands (Ellis, 1990; Micklin, 1988). The ICID organized a special session on the Aral
Sea Basin in 1994 (ICID, 1994), a NATO-sponsored workshop was held on this issue
in Wageningen, The Netherlands, in 1995, and the World Bank has established an
independent advisory panel to provide guidance to the Bank on managerial and tech-
nical issues of this problem. However, the Aral Sea problem still exists. The river sys-
tem has been encroached upon by bridges, etc., so that even if irrigation diversions
ceased, the increased flow could not reach the Sea. In hindsight, a better approach
would have been to balance the water requirements for irrigation against the inevitable
reduction in the flow to the Aral Sea and the resulting environmental consequences.
Heaven et al. (2002), using a real-time mass balance river/reservoir model for Syr
Dara (one of the rivers), concluded that water is either being used extremely ineffi-
ciently, or river offtake data or cropped area data are unreliable, or both.
Water is still being diverted for irrigation. The river system has been encroached
upon by bridges etc. so that even if diversions ceased, the increased flow could not
reach the Sea. A great misconception still exists that by increasing irrigation effi-
ciency, there would be more water for the Sea. "Irrigation efficiency" is the fraction of
water consumed.
On a much smaller scale in the U.S., the diversion of water from the Truckee River
in 1902 lowered Pyramid Lake in Nevada about 24 m during the following 80 years.
Although reallocation of water supplies appears to have stabilized the lake, the large,
deep freshwater lake, which used to overflow periodically into Winnemucca Lake, has
become slightly more saline (Jensen, 1993).
2.2.4 Water Quality Aspects
A major external effect of irrigation is the dissolved solids in return flows. ET con-
centrates the salt that is in applied water. This decrease in water quality is inevitable
with removal of pure water by any consumptive water use, not just agriculture. This is
an environmental cost that represents a trade-off between allowing water to flow in
streams to sustain instream environmental benefits versus offstream beneficial con-
sumptive uses. In the future, more regulations or compromises mutually agreed upon
by users and regulators can be expected.
Some fertilizer, mainly nitrogen, is added to the return flow downstream from irri-
gated areas by leaching. Better nutrient and irrigation management, however, can limit
the impact of this problem. For example, Ferguson et al. (1990) showed that both irri-
gation and fertilizer management influenced the amount of nitrate-nitrogen leached.
They showed that the potential reduction in fertilizer costs by using irrigation schedul-
ing, soil testing, and fertilizer management services could be profitable. Studies in
other areas have shown similar results. Application of excess plant nutrients and inef-
fective water management will continue until the real cost of water pollution is paid
for by the polluters or regulations are implemented. Pesticide residues have also been
found in return flows; these can be likewise prevented or reduced to minimal and safe
levels by improved management practices.
Irrigation in some arid areas with complex geology can affect the quality of water
flowing to downstream wildlife areas. For example, Lico (1992) reported drainage to
wildlife areas in west-central Nevada showed high concentrations of potentially toxic
arsenic, boron, lithium, molybdenum, un-ionized ammonia, uranium, and vanadium.
Several such investigations by the U.S. Geologic Survey have been completed. A
committee commissioned by the Water Science and Technology Board of the U.S.
National Research Council analyzed irrigation-induced selenium contamination and
irrigation-related drainage problems associated with the Kesterson Reservoir in Cali-
fornia (National Research Council, 1989). In that case, the reservoir captured drainage
water that originally was intended to drain to the San Francisco Bay area.
Another potential harmful effect of poorly managed irrigation is spreading human
diseases: schistosomiasis (bilharzia), dengue and dengue hemorrhagic fever, liver
fluke, filariasis, onchocerciasis (river blindness), malaria, Japanese encephalitis, and
diarrhea. Schistosomiasis is most closely linked to irrigation. Available guidelines for
policy makers, planners and managers forecast vector-borne diseases and help incor-
porate safeguards into irrigation projects (Birley, 1989, and Tiffen, 1989).
2.2.5 Irrigation Status and Trends
The area of land irrigated in the U.S. increased from the mid-1940s until the 1980s
(Figure 2.1), and is continuing to increase in some regions. In the Great Plains, deple-
tion of groundwater and some governmental programs caused the decline in irrigated
land area during the early 1980s, mainly in the High Plains of Texas. Increased irriga-
tion mainly in Nebraska and in Texas increased the total in 1997 and 2002 to about
that in 1978. (Some apparent changes in the 1980s are due to changed statistical pro-
cedures.)
Irrigation expanded rapidly in the Texas High Plains during the drought years of
the 1950s. Excellent soils, topography well-suited for surface irrigation without any
land leveling, and a good quality, easily accessible water source enabled rapid irriga-
tion expansion. The well-known Ogallala aquifer was the main source of water. How-
ever, its natural recharge rate in the area was very low and groundwater levels dropped
rapidly during the next 20 years. As a result, well flow rates decreased drastically and
pumping costs increased due to larger pumping lifts and higher fuel costs. Some lands
put under irrigation in the 1950s and 1960s have been returned to dryland farming.
Recently, most surface irrigation systems have been replaced by center pivot (CP)
sprinkler systems because center pivots can apply water uniformly with smaller flow
rates. Also, less water needs to be pumped because with CP systems there is usually
little or no runoff and less than full irrigation (deficit irrigation) can be achieved.
In Florida, farmers irrigate a high percentage of cropped land because of very sandy
soils and the need for quality control of high-value crops even though annual rainfall
is about 1270 mm. Florida growers have converted many sprinkler systems to microir-
rigation systems. In citrus groves, microsprinklers are designed to cover the citrus
trunks for freeze control (Parsons and Wheaton, 1995). Irrigated lands have also in-
creased steadily in other eastern states since 1978 (Figure 2.2). These include the
north-central states of Michigan, Minnesota, and Wisconsin where sandy soil has been
a major factor influencing expansion of irrigated lands. In these states, groundwater
has been the main source of water. Without a convenient and renewable source of wa-
ter, farmers do not irrigate, even though potential increases in crop yields would offset
8.0
7.0
Irrigated Land, hectares, millions

6.0
5.0
4.0
3.0
2.0
1.0
0.0
1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
SW NW MTN GP
Figure 2.1. Expansion of irrigated land from 1944 to 2002 in the western U.S.
SW = Southwest (Arizona, California, New Mexico, Nevada); NW = Northwest (Idaho,
Oregon, Washington); MTN = Mountain States (Colorado, Utah, Wyoming); and GP =
Great Plains (Kansas, Nebraska, Montana, North Dakota, Oklahoma, South Dakota, Texas).
Source: USDA 1997 and 2002 Censuses of Agriculture (USDA-NASS 1999, 2004).
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
FL SE L-Miss MW NC NE
Figure 2.2. Expansion of irrigated lands from 1978 to 2002 in the eastern U.S.
FL = Florida; SE = Southeast (Alabama, Georgia, Kentucky, North Carolina,
South Carolina, Tennessee, Virginia); L-Miss = Lower Mississippi (Arkansas,
Louisiana, Mississippi, Missouri); NC = North-Central (Michigan, Minnesota,
Wisconsin); and NE = 12 remaining northeast states. Source: USDA 1997 and
2002 Censuses of Agriculture (USDA-NASS 1999, 2004).
capital and operating irrigation costs, especially in drought years. Irrigated land in the
lower Mississippi states of Arkansas, Louisiana, Mississippi, and Missouri has in-
creased markedly since 1978. Over half of the irrigated land in the lower Mississippi
region is in Arkansas. Conflicts over water have developed in the southeast because of
limited groundwater supplies and lack of surface water storage facilities.
The impact of sustainable water supply on the expansion or decrease in irrigated
area is readily apparent when considering its recent history in the Great Plains. The
area of irrigated land in Texas expanded rapidly until about 1980, and then decreased
as groundwater supplies decreased. In contrast, the irrigated area in Nebraska also
increased rapidly, especially with the development of the center-pivot sprinkler system
in the 1960s, and has continued to increase to 2002 (Figure 2.3). Nebraska’s water
supply is from both surface and groundwater supplies, which have been relatively sta-
ble.
Global statistics on irrigated land area are not as robust as that in a single country
because uniform statistical methods are not always used. For example, in some coun-
tries cropping intensity is considered, so land that is double cropped within one year
may be counted twice. Also, sometimes the area commanded by a canal may be re-
ported instead of the actual land irrigated. The irrigated land reported by FAO appears
to be the most consistent and is the basis of the values reported in this section. The rate
of increase in irrigated land reached a peak of about 2.2% per year in the mid-1970s
(Jensen, 1993). The total arable land irrigated at the beginning of the 21st century was
about 275 million hectares. The annual rate of increase has decreased since the mid-
1990s to less than 0.5% per year at the beginning of the 21st century (Figure 2.4).
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
TX NE KS MT+ND+OK+SD
Figure 2.3. Expansion of irrigated land in the U.S. Great Plains regions.
TX = Texas; NE = Nebraska; KS = Kansas; and MT = Montana,
ND = North Dakota, OK = Oklahoma, SD = South Dakota.
Source: USDA 1997 Census of Agriculture (USDA-NASS, 1999).
300
y = 3.4292x - 6583.3
250

200
150
y = 2.2046x - 4232.6373
100
50
0
1970 1975 1980 1985 1990 1995 2000
Year
World Asia N&C Amer Africa S Amer Linear (World) Linear (Asia)
Figure 2.4. Expansion of irrigated land from 1972 to 2002 for the World, Asia,
North and Central America, Africa, and South America.
Source: FAO Yearbooks and FAOSTAT at www.fao.org.
2.3 CROP PRODUCTION AND

IRRIGATION WATER REQUIREMENTS
2.3.1 Crop Production-Evapotranspiration Relationships
When soil water is not adequate for plants to transpire at the climate-based evapo-
rative demand rate, plants experience water stress and plant stomates begin closing to
limit transpiration. Since plant stomates also provide for uptake of carbon dioxide, the
rate of photosynthesis decreases. Numerous studies have shown that plant growth and
crop yields decrease linearly with decreased transpiration.
Since evaporation is only a part of total ET, plant water stress distributed over the
growing season also decreases dry matter and marketable yields as ET decreases
(Howell, 1990; Howell et al., 1990). Howell (1990) reviewed the results of previous
studies and presented a detailed analysis of yield-transpiration and yield-ET relation-
ships for corn, grain sorghum, and winter wheat. The general linear statistical relation-
ship between total ET and yield of cereal grains in a given climate zone is:
Yg = b ET + a (2.1)
where Yg = the yield or marketable product, Mg/ha
ET = growing season evapotranspiration, mm
b = the slope of the yield-ET line, Mg/(ha mm)
a = a constant, Mg/ha.
Because some ET occurs before the first increment of marketable product is produced,
the value of a is negative. For example, Stewart et al. (1977) reported that for corn
yields from several non-saline treatments at Davis, California, in 1974 could be esti-
mated as Yg = 0.019 ET – 1.17 Mg/ha, i.e., the increase in corn grain per unit increase
in ET will be 1.9 kg per cubic meter of water consumed in ET. Values of b and a will
vary with crop cultivars and climate. As new cultivars are developed with higher har-
vest indices and improved cultural practices, yields may increase with little increase in
ET, thereby changing the constants. Climate also will affect the equation constants.
With frequent rains when annual plants are small, the evaporation component will be
higher than in arid areas. The slope (b) may be less in humid areas.
Similar relationships can be developed for dry matter or for forage production. For-
age yield-ET relationships from carefully controlled field experiments tend to vary
less than grain yields. For example, when using regression data from dryland studies
in North Dakota (Bauder et al., 1978), sprinkler-irrigated alfalfa in New Mexico in
1979 (Sammis, 1981), and regression data from studies in Israel (Kipness et al., 1989),
the resulting yield of alfalfa at zero moisture could be expressed as Yalf = 0.015 ET –
0.50 Mg/ha, that is, the increase in alfalfa yield per unit of ET will be 1.5 kg/m3. This
value obtained from widely differing climates agrees closely with the seven-year aver-
age of well-watered, lysimeter-based yield reported by Wright (1988). Wright reduced
the lysimeter yields 5% to agree with adjacent field yields of 1.76 kg/m3 at 12% mois-
ture, or 1.6 kg/m3 at zero moisture.
Howell et al. (1990, 2000) also presented detailed analyses of concepts of water use
efficiency (WUE), which is the the economic or marketable production per unit of wa-
ter consumed in ET (WUE = Y/ET). Bos (1980), working with an ICID group, defined
a related term of irrigation water use efficiency (WUEi), which considers the increase
in production per unit increase in ET due to irrigation, or WUEi = (Yi – Yd)/(ETi –
ETd). If there is no dryland yield or ET, the values are the same. Both of these terms
have served useful purposes as indices of the effectiveness of various agronomic prac-
tices. I have used the WUE terms since the 1950s. However, due to confusion with
irrigation efficiency that has often resulted in the literature, I suggested that these
terms be considered as indices of the productivity of water used in crop production
(Jensen, 1993). Either “total water consumed” or “irrigation water consumed” in ET is
needed to enable uniform comparisons of practices or between areas.
A comprehensive series of articles on the response of crops to carbon dioxide, tem-
perature, and water quality, and on water shortage, is presented in a publication edited
by Kirkham (1999). Chapters on water use by several crops such as cotton, olives,
rice, sorghum, sunflowers, turfgrass, and wheat are presented along with a chapter on
risk assessment of irrigation requirements of field crops. Stewart and Nielsen (1990)
compiled detailed information on irrigation of various crops in a comprehensive
monograph.
Relationships between yield and water applied are much more variable than those
between yield and ET, because the rainfall during the growing season and the amount
of water percolating below the root zone are not the same each year, or from place to
place, and will vary with the irrigation method. Even surface runoff may not be taken
into account in data reported on water applied and yield. Yield-water application rela-
tionships tend to be curvilinear, as shown by Howell et al. (1995) and Letey (1993).
Such relationships also tend to be applicable to specific conditions and are not as
transferable or as valuable in comparing production practices and systems as WUE
parameters. Unger and Howell (1999) presented a global perspective summarizing
basic principles and practices for achieving agricultural water conservation, both under
dryland (rainfed) and irrigated conditions.
2.3.2 Deficit Irrigation and WUE

A relatively new practice of not replenishing the soil water profile at each irriga-
tion, where rainfall during the growing season is significant, is called deficit irrigation.
This practice provides soil water storage space for capturing rainfall that otherwise
may percolate beyond the root zone. For example, Howell et al. (1995) have shown
that when replacing only two-thirds of the depleted soil water at irrigation, the irriga-
tion water use efficiency for irrigation water (WUEi) for corn increased from 1.8 to
2.3 kg/m3. WUEi is defined in Section 2.3.1. The net effect is that a higher proportion
of rainfall may be effective with deficit irrigation than with full irrigations.
Another form of deficit irrigation is managed plant water stress. For example,
when irrigating some small grain cultivars that are subject to lodging, mild plant water
stress prior to the boot stage can shorten the stems and increase the resistance to lodg-
ing. Terminating irrigation at critical times such as after the soft dough stage of cereals
can reduce ET and irrigation water required without a decrease in economic yield.
Similarly, on sugar beets, more leaf area may be produced than necessary. Under these
conditions, allowing increased plant water stress by limiting irrigations near the end of
the season can reduce the mass of beets produced, but increase sugar content. As a
result, the total sugar yield may not be affected. A light irrigation before harvest may
be required to harvest the beets. A side benefit is that less tonnage needs to be hauled
to the sugar factory. Studies on cotton in California showed that irrigations that are too
early not only lower final yield of cotton lint, but also result in lower production effi-
ciency (Grimes and El-Zik, 1990). These are examples of how operators of irrigation
systems can manage irrigations to reduce water requirements and costs without reduc-
ing economic yields. Local experimental data should be used as a guide to manage
deficit irrigation. FAO published a Water Report summarizing experiences of deficit
irrigation in a number of countries (FAO, 2002). The potential effects of deficit irriga-
tion practices also will need to be considered when designing farm irrigation systems.
2.3.3 Irrigation Requirements for Evapotranspiration and Salinity Control
For design purposes, in most arid areas and especially during the dry part of the
season, precipitation has little effect on estimates of irrigation water requirements for
ET and salinity control. Methodology for estimating crop ET and irrigation water re-
quirements is presented in Chapter 8. Procedures for estimates the amount of water
required for leaching are presented in Chapter 7. Leaching for control of soil salinity
does not always need to be provided at each irrigation. Pre-plant irrigation may ade-
quately leach the surface soil for new seedlings. Depending on the sensitivity of the
crops to be grown and the rainfall during the winter season, planning for leaching dur-
ing the low ET-demand rainy period may make more effective use of rainfall.
Estimating the design capacity in subhumid and humid areas can be more compli-
cated than in arid areas. The economic benefit of irrigation in subhumid and humid
areas is derived mainly during extended drought periods. If the irrigation system is not
able to replenish water during these periods, the potential economic benefits of irriga-
tion may not be achieved. These issues are discussed in detail in Chapter 8.
2.3.4 Drainage Requirements for Water and Salinity Control
In the 1990s, the World Bank (IBRD) estimated that a quarter of all irrigated land
was affected by salinity caused by poor irrigation practices (IBRD, 1992b). On newly
irrigated land, some 10 to 50 years may be required for excess water to fill the under-
lying unsaturated zone. If excess water input is ignored during this period, the rise in
the water table inevitably results in waterlogging unless the natural drainage capacity
increases with the rise in the water table. On a project basis, well-designed farm irriga-
tion systems and efficient management of excess water from the start of a project can
delay the need for increased drainage capacity for many decades, or in some cases
prevent the need to install additional drainage facilities (Jensen, 1993).
The benefits of controlled drainage are beginning to receive more attention. This
means that drainage is allowed only when the water table needs to be lowered to pre-
vent crop damage or when encouraging leaching of salts. Abbott (2003) reported a
study that showed that controlled drainage can give significant benefits to both the
farmer as increased yields and to the wider community in terms of reduced water re-
quirements. Areas of potential application for controlled drainage were identified in
North Africa (Algeria, Egypt); the Middle East (Israel, Syria, Iraq, Bahrain); India
(Punjab, Haryana, Rajasthan); and Asia (Pakistan, Northern China, Central Asian
States).
The need for increased drainage capacity is strongly linked to the types of farm ir-
rigation systems used and how they are managed. When most new irrigation projects
were dominated by surface irrigation, increased drainage capacity was nearly always
an essential requirement for sustained agricultural productivity. With improved irriga-
tion systems that apply water more uniformly and with more efficient water manage-
ment practices including deficit irrigation, the capacity of large-scale drainage above
the existing natural capacity may be decreased. However, because of nonuniform sub-
soils, some increased localized drainage capacity will usually be required in most new
irrigation projects. Seepage from unlined canals and ditches can also cause high water
table levels and increase drainage requirements. Details on salinity control and drain-
age requirements and design can be found in Chapters 7 and 9.
Reuse of waste effluents has become a common practice where renewable water
supplies are limited. Use of irrigated agriculture to dispose of waste effluents has also
become an accepted and preferred practice over adding this burden to sewage treat-
ment plants. Special management practices may be required and there usually are limi-
tations on the types of crops on which waste effluents can be used. Details on use of
waste effluents in irrigation are presented in Chapter 20.
2.4 SYSTEM DESIGN AND INCREASING COMPETITION
FOR RENEWABLE WATER SUPPLIES
Increasing competition among water uses will increase pressures on irrigated agri-
culture because agriculture consumes most of the renewable water supplies in most
semiarid and arid areas. These pressures will become more severe as water supplies
are more fully utilized. Conflicts will increase. Solutions to these conflicts may result
in revisions of water laws and allocations. In the future, the price of water will be
more closely related to its real economic cost or value. This will affect irrigated agri-
culture because the marketable product of many crops per unit volume of water con-
sumed is low.
The designer working with individual farmers or on farm irrigation systems for a
new project or a project that is being modernized must consider the available water
supplies, particularly renewable water supplies. These supplies are not only based on
hydrological aspects and the physical facilities available to store and deliver the water,
but on legal and social aspects. In the U.S., an individual farmer’s water right to water
from a common surface water source is the most common assurance of a renewable
water supply. In some cases, the water right may be granted to an quasi-public irriga-
tion district rather than to individuals. It is generally understood that in the western
U.S., when a water right is sold or transferred, only that portion of the right that is
used consumptively can be transferred.
If groundwater is the only source of water, then the expected recharge rate relative
to the expected annual withdrawal rate must be considered to determine if the water
supply can be sustained indefinitely, or at least until the investment in the irrigation
system can be recovered. In areas where groundwater is essentially being mined, as in
parts of the Ogallala aquifer of the Great Plains, state water laws regulating pumping
must be considered. In addition, if groundwater is being mined, the rates of with-
drawal and declining well-yields as the saturated layer decreases must be factored into
the planning and design process. Renewable water supply issues are not covered in
depth in this monograph. Details addressing hydrologic issues and water storage and
conveyance systems must be obtained from other sources.
The U.S. has agreements with Canada and Mexico on sharing common water sup-
plies. However, in many regions, no international agreement governs sharing waters of
rivers that flow through several countries. More than half the world's largest rivers
flow through two or more countries. The potential for conflict over shared water is
increasing greatly. The implications for food security and managing irrigation projects
are enormous (Jensen, 1993). International law for sharing water will become more
important as competition increases. Guidelines for international agreements are avail-
able (Kimball, 1992).
2.4.1 Environmental Aspects, New Dimensions
The International Conference on Water and the Environment, held in Dublin in
January, 1992, focused on sustainable development, management, and utilization of
water resources in harmony with environmental conservation. It identified environ-
ment and development priorities to be agreed upon at the June, 1992, United Nations
conference in Rio de Janeiro. Agenda 21, adopted at the Rio conference, gave high
priority to win-win policies of poverty reduction, economic efficiency, and economic
development. Also about this time, the policy of the World Bank relative to new pro-
jects changed. The Bank staff now required investment projects to be classified based
upon their potential environmental impact. Projects in Category A required a full envi-
ronmental assessment. These include dams and reservoirs and large-scale irrigation
projects (IBRD, 1992a). Policies for sustained development include “removing subsi-
dies that encourage excessive use of fossil fuels, irrigation water, and pesticides and
excessive logging” (IBRD, 1992b, p. 2).
The possible trade-offs between water for food production and water for nature be-
came one of the most contentious issues in discussions of the Second World Water
Forum in The Hague in 2000 (Rosegrant et al., 2002). They concluded that this con-
flict might be one of the most critical problems to be tackled in the early 21st century.
Gleick (2002) indicated that in the early 20th century in the U.S. West and
throughout the world, water left in the stream was considered “wasted.” The language
of water development and reclamation indicated a perception that water was a re-
source to be extracted and put to use. The meaning of beneficial use of water is begin-
ning to expand from a narrow economic interpretation to a broader view recognizing
the growing public awareness of the value of environmental resources. South Africa
has formally recognized the right to water for natural ecosystems in their new water
laws and constitution (Gleick, 2002). In the U.S., competition for water for endan-
gered species and wildlife, such as for the whooping crane in Nebraska and salmon in
northwest rivers, will affect water supplies available for irrigated agriculture
The increasing demand for domestic and other water uses may affect the order of
water reuse. Foe example, Falkenmark and Widstrand (1992) suggested that because
domestic use requires the cleanest water, the ideal order of reuse would be household
first, then industry, then agriculture. Agricultural use would be last because most irri-
gation water is consumed, i.e., it returns to the atmosphere through evapotranspiration
and is not available for reuse. This concept is becoming very real in some areas as
clean water is being provided first to municipalities. Agriculture, which used to have
first access to the water, is being required to use municipal return flow.
2.4.2 Water Supply and Consumption
In semiarid areas, the area of irrigated land alone does not indicate the volume of ir-
rigation water that is consumed because of variable rainfall and because some crops
are often not fully irrigated. Likewise, the traditional concept of irrigation efficiency is
not very meaningful when considering water supplies for irrigation and water con-
sumed by evaporation and transpiration. A better procedure for accounting for water
use, based on a water balance approach, was developed by Molden (1997). Water ac-
counting is a procedure for analyzing the uses, depletion, and productivity of water in
a water basin context. A key term is water depletion, which is the use or removal of
water from a water basin such that it is permanently unavailable for further use as a
liquid within the basin. He described process and non-process depletions. Process
depletion is where water is depleted to produce an intended good. In agriculture, proc-
ess depletion is transpiration plus that incorporated into plant tissues⎯the product.
Non-process depletion includes evaporation from soil and water surfaces and any non-
evaporated component that does not return to the freshwater resource. It would also
include depletion by weeds and other non-economic vegetation. The depleted fraction
is that part of inflow that is depleted by both process and non-process uses of water.
The productivity of water is a performance parameter that can be related to the physi-
cal mass production or economic value per unit volume of water. Molden suggested
that the productivity of water can be measured against gross or net inflow, depleted
water, process-depleted water, or available water in contrast to the production per unit
of water consumed in ET (Viets, 1962).
Water accounting can be done at various levels such as the field, irrigation service,
basin, or sub-basin levels. Molden et al. (1998) presented a detailed example of water
accounting at the basin level using data from Egypt’s Nile River, where some detailed
information on water use and productivity was available. Molden and Sakthivadival
(1999) presented another example for a district in Sri Lanka.
Numerous conferences have been held resulting in papers concerning the quantifi-
cation of real water, which is water that may be available for transfer or reallocation
to higher value uses. Irrigation efficiency concepts enter in when discussing possible
real water amounts that can be saved and made available for transfer by making
changes on irrigated lands. Roos (2001), in discussing agricultural water conservation
in relation to water available for transfer to another basin, stated: “Unless evapotran-
spiration can be reduced, such measures do not add to the regional supply. Depletion
remains essentially the same even with greater application efficiency....” His com-
ments related to the Sacramento and San Joaquin valleys, which function like closed
basins. He also stated: “But the general rule is still valid: unless depletion changes,
there is no real water savings available for transfer.”
2.5 IRRIGATION WATER MANAGEMENT
DURING DROUGHTS
Changes in water rights of renewable water supplies from agriculture to higher
value uses have been underway in the western U.S. for decades. Today, new methods
and practices are used to provide temporary exchanges in water supplies. For example,
when water supplies are plentiful, farmers who irrigate continue to use the water
available under their water rights, but when annual supplies are below normal, tempo-
rary exchanges in water uses have become more common.
NRC (1992) summarized some of the mechanisms being used to provide flexibility
in the use of renewable water supplies. The NRC Committee discussed various water
market mechanisms, third party impacts, water transfer opportunities, and the role of
water law in the transfer process. Case studies of water transfers in several areas of the
West were presented along with several examples of water banking practices.
2.6 OTHER AGRICULTURAL PURPOSES
AND BENEFITS OF IRRIGATION
Irrigation is used mainly to provide water used in ET and for control of soil salin-
ity. However, irrigation is also used for other purposes that are associated with irri-
gated agriculture. These secondary uses may be important to the success of an irri-
gated enterprise. Some of the special uses that need to be considered by the designer
are summarized below. The designer should consider local practices and special crop
needs when planning and designing an irrigation system.
2.6.1 Soil Conditioning
In arid areas, where crops are planted to meet specific marketing opportunities, irri-
gation is essential to germinate and establish plants because rainfall may not occur at
planting time or is not dependable. In some areas, solid set sprinklers are set up solely
for this purpose and then removed after the new crop has been established. After es-
tablishing the new crop, irrigations are continued using the existing irrigation system,
which is usually a surface system.
2.6.2 Crop Cooling
Crop cooling is one of several special uses of irrigation. It is a minor component of
irrigated agriculture, but is used on diverse crops for various purposes. For example,
when planting lettuce in the Imperial Valley of California about mid-September, daily
3-hour sprinkler irrigations are applied to cool the germinating lettuce seedlings until
they are established. Vineyards have been irrigated for 2 to 3 minutes at 15-minute
intervals to cool the plants down to 8° to 14°C. Other crops that have been cooled by
sprinkler irrigation are almonds, apples, beans, cherries, cotton, cranberries, cucum-
bers, flowers, potatoes, prunes, strawberries, sugar beets, tomatoes, and walnuts
(Sneed, 1972; Gray, 1970; Kidder, 1970; Carolus, 1971; Unrath 1972a,b). Air condi-
tioning of crops for higher yield and better quality has been studied in California,
Michigan, and Washington. Summaries of crop cooling principles have been presented
by Chessnes et al. (1979) and Merva and Vandenbrick (1979). Evaporative cooling of
apples in North Carolina has increased their red color and soluble solids (Unrath and
Sneed, 1974). Sprinkler irrigation has been used to delay fruit bud development in
Utah to avoid damage due to low spring temperatures (Alfaro et al., 1974; Anderson,
1974). Irrigation designers should consider recent technological developments, local
practices, and current economics in planning crop cooling by irrigation.
2.6.3 Frost and Freeze Protection
Freeze protection is another special use of irrigation. By coating plants with water
as air temperatures drop below freezing, the heat of fusion released as the water
freezes maintains the plant temperature at 0°C. The ice coating on plants must con-
tinually be in contact with unfrozen water until the ice melts. Early studies have shown
that plants could be protected against freezing against temperature as low as –9°C with
no wind and applications of 6.5 mm/h. Early studies of frost and freeze protection for
small fruit, flowers, potatoes, and grapes were described in the first edition (Braud and
Horthorne, 1965; Harrison and Gerber, 1965; Loscascio et al., 1967; Lamade, 1968;
Kidder, 1970; Sneed, 1970; Braud and Esphahani, 1971; Harrison et al., 1974). Severe
freezes in Florida in the 1980s killed about 82,000 ha of citrus. Since then, micro-
sprinklers have largely replaced heaters and wind machines in Florida (Parsons and
Wheaton, 1995).
Various techniques for frost and freeze protection were discussed in an ASAE
monograph, Modification of the Aerial Environment of Crops, by Barfield and Gerber
(1974). The designer should consider recent technological development, local prac-
tices, and current economics in planning for frost and freeze protection by irrigation.
2.6.4 Soil Conditioning
In arid areas, the surface soil may become very dry following the harvest of the
previous crop. In these areas, it is often impossible to prepare a seed bed for small-
seed crops without first moistening the surface soil. Thus, irrigation becomes a neces-
sity for mechanical purposes rather than providing water for plant growth. Similarly,
irrigation may be necessary to permit harvesting root and tuber crops such as sugar
beets, potatoes, and peanuts. Much of the water applied for this purpose may be stored
as soil water for use by the next crop, stored for the next cropping season, or used to
leach salts.
2.6.5 Applications of Chemicals in Irrigation Water
Application of fertilizers, pesticides, herbicides, desiccants, and defoliants—
chemigation—is well established and documented. Because this practice has become
so widespread a new chapter has been added on this subject (see Chapter 19).
2.7 SYSTEM DESIGN AND OPERATION CHALLENGES
Recent and emerging design issues concern environmental aspects affected when
using surface runoff from surface irrigation systems. These include any pesticides and
plant nutrients that may be included in the runoff, erosion and sediment deposition in
the collecting flow channels and receiving water bodies. Likewise, when converting
from surface irrigation systems to drip or sprinkler systems, wetlands that had devel-
oped from previous irrigation practices may gradually dry up. The impact of both sur-
face and subsurface water quality on instream and downstream water uses will become
significant factors in the future if not already a factor in the area under consideration.
Designers of irrigation must also consider the various legal, environmental and physi-
cal constraints that may affect the design and operation of an irrigation system.
Today, designers of irrigation systems must take into account many other aspects in
addition to those considered essential several decades ago. Economic aspects will be
more challenging if stable renewal water supplies are not assured and prices received
for farm produce fluctuate widely. English et al. (2002) indicate that irrigated agricul-
ture will need to adopt a management paradigm that is based on maximizing net bene-
fits rather than maximizing yields. Irrigation to maximize benefits will be more com-
plex and challenging than the conventional practice of maximizing crop yields per unit
of land. Two important concerns are sustainability and risk. The increased complexity
of analysis will require more sophisticated analytical tools.
One of the basic concepts of surface irrigation in the past is that there must be sur-
face runoff if the lower ends of the borders or furrows are to be irrigated so as to re-
place most of the available water. In many areas, because of declining water tables and
limited water supplies, surface runoff is being restricted or prohibited. The general
approach in coping with this restriction was to install recirculating or pump back sys-
tems or to physically prevent surface runoff. Recirculating systems are required in
some areas. The challenge to designers who are restructuring surface systems is to
consider variable grade systems that can increase the head to tail end uniformity of
water application with little or no surface runoff. The concept is not new (Powell et
al., 1972). However, such a design can be very complex when considering varying
infiltration rates with or without furrows, the variation in vegetative growth in border
strips and some row crops during the year, and the special requirements of some crops.
Again, there will be need for more sophisticated analytical tools to address these com-
plex relationships. Chapters 13 and 14 address these issues and provide the concepts
and tools necessary for meeting these design challenges. Chapters 15-17 address the
design of sprinkler and microirrigation systems.
In developing countries, the stability of a community organization is essential for
sustained and productive irrigated agriculture. Rapidly increasing populations and
relatively stable renewable water supplies often involve complex social, food security,
and natural resources issues. Addressing these issues will be a major challenge to de-
signers of new or improved irrigation systems to consider, especially in developing
countries.
2.8 SUMMARY
The factors to be considered in the design and operation of farm irrigation systems
have changed greatly since the first edition of this book was published in 1980. In this
chapter, I have attempted to highlight key factors that will influence the design and
operation of new irrigation systems and the upgrading of old systems in order to make
irrigated agriculture more productive and more sustainable. Some of these factors are
renewable water supplies, coping with droughts, adequacy of drainage, control of sa-
linity, the productivity of water and water use efficiency, new management concepts
such as deficit irrigation, and environmental concerns about surface runoff. I have
cited references where readers can find additional details on the problems and issues
that will need to be addressed by irrigation system designers, trainers providing irriga-
tion management guidance, and operators of irrigation systems. Sustainable and pro-
ductive irrigated agriculture will require adopting new technology and management
concepts and increased emphasis on the productivity of irrigation water.
REFERENCES
Abbott, C. 2003. Integrated irrigation and drainage to save water—“controlled
drainage.” Water, Dept. of Int’l. Devel. 16: 6.
Alfaro, J. F., E. E. Griffin, J. Keller, G. R. Hanson, J. L. Anderson, G. L. Aschroft, and
E. A. Richardson. 1974. Preventive freeze protection by preseason sprinkling to
delay bud development. Trans. ASAE 17: 1025-1028.
Anderson, J. L. M. 1974. Effects of over tree sprinkling on fruit bud development and
fruit quality. Mimo. Report. Logan, Utah: Plant Sci. Department, Utah State Univ.
Barfield, B. J., and J. F. Gerber, eds. 1979. Modification of the Aerial Environment of
Crops. St. Joseph, Mich.: ASAE.
Bauder, J. W., A. Bauer, J. M. Ramirez, and D. K. Cassel. 1978. Alfalfa water use and
production on dryland and irrigated sandy loam. Agron. J. 70: 95-99.
Birley, M. H. 1989. Guidelines for forecasting the vector-borne disease implications of
water resources development. PEEM Guidelines Series 2, VBC/89.6. Geneva,
Switzerland: World Health Organization.
Bos, M. G. 1980. Irrigation efficiencies at the crop level. ICID Bulletin, Int’l. Comm.
Irrig. and Drain. 29(2): 18-25, 60.
Braud, H. J. Jr., and P. L. Hawthorne. 1965. Cold protection for Louisiana
strawberries. Bulletin No. 591, March. Baton Rouge, La.: Louisiana State Univ.
Braud, H. J., and M. Esphahani. 1971. Direct water spray for citrus freeze protection.
ASAE Paper No. 71-234. St. Joseph, Mich.: ASAE.
Carolus, R. L. 1971. Evaporative cooling techniques for regulating plant water stress.
Hort. Sci. 6: 23-25.
CAST. 1992. Preparing U.S. Agriculture for Climatic Change. Ames, Iowa: Council
for Agricultural Science and Technology.
Chesness, J. L., L. A. Harper, and T. A. Howell. 1979. Sprinkling for heat stress
reduction. In Modification of the Aerial Environment of Crops, 388-399. B. J.
Barfield, and J. F. Gerber, eds. St. Joseph, Mich.: ASAE.
CSIRO. 2002. New framework for water. Media Release—Ref 2002/200. Clayton
South VIC, Australia: CSIRO.
Edwards, C. A., R. Lal, P. Madden, R. H. Miller, and G. House, eds. 1990.
Sustainable Agricultural Systems. Ankeny, Iowa: Soil and Water Conservation
Society.
Ellis, W. S. 1990. The Aral: A Soviet sea lies dying. National Geographic 177(2): 73-
92.
English, M. J., K. H. Solomon, and G. J. Hoffman. 2002. A paradigm shift in
irrigation management. J. Irrig. Drain. Eng. 128: 267-277.
Falkenmark, M., and C. Widstrand. 1992. Population and Water Resources: A
Delicate Balance Pop. Bulletin 47(3). Washington, D.C.: Population Reference
Bureau, Inc.
FAO (Food and Agriculture Organization). 1989. Sustainable Development and
Natural Resources. Conference, C 89/2 – Sup. 2, Nov., 54 pp. FAO.
FAO (Food and Agriculture Organization). 2002. Deficit irrigation practices. FAO
water reports No. 22. Rome, Italy: FAO.
Ferguson, R. B., D. E. Eisenhauer, T. L. Bockstadter, D. H. Krull, and G. Buttermore.
1990. Water and nitrogen management in central Platte Valley of Nebraska. J.
Irrig. Drain. Eng. 116(4): 557-565.
Gleick, P. 2002. The World’s Water, 2002-2003: Biennial Report on Freshwater

Resources. Washington, D.C.: Island Press, with contributions from the staff of the
Pacific Institute.
Gray, A. S. 1970. Environmental control using sprinkler systems. In Nat’l. Irrigation
Symp., RR1-RR10.
Grimes, D. W., and K. M. El-Zik. 1990 Cotton. In Irrigation of Agricultural Crops,
741-773. Agron. Series 30. B.A. Stewart, and D. R. Nielsen, eds. Madison, Wis.:
Am. Soc. Agron.
Harrison, D. S., and J. F. Gerber. 1965. Research with sprinkler irrigation for cold
protection. Trans. ASAE 7(4): 464-468.
Harrison, D. S., J. F. Gerber, and R. E. Choate. 1974. Sprinkler irrigation for cold
protection. Cir. 348 Tech. Gainesville, Fla.: Florida Coop. Ext. Serv., Univ. of
Florida.
Hatfield, J. L., and D. L. Karlen, eds. 1993. Sustainable Agricultural Systems. Boca
Raton, Fla.: Lewis Publishers.
Heaven, S., G. B. Koloskov, A. C. Lock, and T. W. Tanton. 2002. Water resource
management in the Aral Basin: A river basin model for the Syr Darya. Irrig. and
Drain. 51: 109-118.
Hoffman, G. J., T. A. Howell, and K. H. Solomon, eds. 1990. Management of Farm
Irrigation Systems. St. Joseph, Mich.: ASAE.
Howell, T. A. 1990. Relationships between crop production and transpiration,
evapotranspiration, and irrigation. In Irrigation of Agricultural Crops, 391-434.
B.A. Stewart and D.R. Nielsen, eds. Agron. Series 30. Madison, Wis.: American
Soc. Agron.
Howell, T. A. Irrigation’s role in enhancing water use efficiency. 2000. In Nat’l.
Irrigation Symp., 66-80. R. G. Evans, B. L. Benham, and T. P Trooien, eds. St.
Joseph, Mich.: ASAE.
Howell, T. A., R. H. Cuenca, and K. H. Solomon. 1990. Crop yield response. In
Management of Farm Irrigation Systems, 93-122. G. T. Hoffman, T. A. Howell,
and K. H. Solomon, eds. St. Joseph, Mich.: ASAE.
Howell, T. A., A. D. Schneider, and B. A. Stewart. 1995. Subsurface and surface
microirrigation of corn—Southern High Plains. In Microirigation for a Changing
World, 375-381. F. R. Lamm, ed. St. Joseph, Mich.: ASAE.
IBRD. 1992a. The World Bank and the Environment, Fiscal 1992. 3rd Annual Report,
Washington, D.C.: The World Bank.
IBRD. 1992b. World Development Report 1992: Development and the Environment.
New York, N.Y.: The World Bank, Oxford Univ. Press.
ICID. 1994. Aral Sea Basin. In Special Session, Report, Int’l. Comm. on Irrig. and
Drain. New Delhi, India: ICID.
Jensen, M. E. 1984. Water resource technology and management. In Future
Agricultural Technology and Resource Conservation, 142-146, B. C. English et al.,
eds. Ames, Iowa: Iowa State Univ. Press.
Jensen, M. E. 1990. Arid lands, impending water-population crisis. In
Hydraulics/Hydrology of Arid Lands, Proc. Int’l. Symp., 14-19. R. H. French, ed.
New York, N.Y.: American Soc. Civil Eng.
Jensen, M. E. 1993. The Impacts of Irrigation and Drainage on the Environment. 5th
Gulhati Memorial Lecture, Int’l. Comm. on Irrig. and Drain. The Hague, The
Netherlands: ICID.
Kidder, E. H. 1970. Climate modification with sprinklers. In Natl. Irrig. Symp., V1-
V6.
Kimball, L. A. 1992. Forging International Agreement: Strengthening Inter-
governmental Institutions for Environment and Development. Washington, D.C.:
World Resources Institute.
Kipness, T., I. Vaisman, and I. Granoth. 1989. Drought stress and alfalfa production in
a Mediteranean environment. Irrig. Sci. 10: 113-125.
Kirkham, M. B., ed. 1999. Water Use in Crop Production. Binghamton, N.Y.: Food
Products Press. Co-published as J. Crop Production, Vol. 2, No. 2(#4).
Lamade, W. 1968. Frost control on potatoes. Veg. Crop Mgmt. 4(4): 27-41.
Letey, J. 1993. Relationship between salinity and efficient water use. Irrig. Sci. 14: 75-
84.
Ley, T. W. 1995. Weather stations C Satellites to scientific irrigation scheduling.
Irrigation III(4): 20-23.
Lico, M. S. 1992. Detailed study of irrigation drainage in and near wildlife
management areas, West-Central Nevada, 1987-90. U.S. Geological Survey Water
Resources Investigations Report 92-4024A, Part A. Carson City, Nev.: USGS.
Locascio, S. J., D. S. Harrison, and V. P. Nettles. 1967. Sprinkler irrigation of
strawberries for freeze protection. Proc. Florida State Hort. Soc. 80: 208-211.
Merva, G. E., and C. Vandenbrick. 1979. Physical principles involved in alleviating
heat stress. In Modification of the Aerial Environment of Crops, 373-387. B. J.
Barfield, and J. F. Gerber, eds. St. Joseph, Mich.: ASAE.
Micklin, P. P. 1988. Desiccation of the Aral Sea: A water management disaster in the
Soviet Union. Science 241: 1170-1176.
Molden, D. 1997. Accounting for water use and productivity. IWMI/SWIM Paper No.
1. Colombo, Sri Lanka: Int’l. Water Mgmt. Inst.
Molden, D. J., M. el Kady, and Z. Zhu. 1998. Use and productivity of Egypt’s Nile
water. In Contemporary Challenges for Irrigation and Drainage, USCID 14th Tech.
Conf., 99-116. J. I. Burns, and S. S Anderson, eds. Denver, Colo.: USCID.
Molden, D., and R. Sakthivadivel. 1999. Water accounting to assess use and
productivity of water. Int’l. J. Water Resources Development 15(1): 55-71.
Norwood, C. A. 1995. Comparison of limited irrigated vs. dryland cropping systems
in the U.S. Great Plains. Agron. J. 87: 737-743.
NRC (National Research Council). 1989. Irrigation-Induced Water Quality Problems.
Washington, D.C.: Committee on Irrigation-Induced Water Quality Problems,
Water Science and Technology Board, National Research Council. Washingtion,
D.C.: National Academy Press.
NRC (National Research Council). 1992. Water Transfers in the West: Equity,
Efficiency, and the Environment. Washington, D.C.: Committee on Western Water
Management, Water Science and Technology Board, National Research Council.
Washingtion, D.C.: National Academy Press.
NRC (National Research Council). 1993. Toward Sustainability, Soil and Water
Research Priorities for Developing Countries. Washingtion, D.C.: National
Academy Press.
NRC (National Research Council). 1996. A New Era for Irrigation. Washington, D.C.:
Committee on the Future of Irrigation in the Face of Competing Demands, National
Research Council. Washingtion, D.C.: National Academy Press.
Parsons, L. R., and T. A. Wheaton. 1995. Freeze protection of Florida citrus with
microsprinkler irrigation. In Microirrigation for a Changing World, 25-30. F. R.
Lamm, ed. St. Joseph, Mich.: ASAE.
Pereira, L. S., J. R. Gilley, R. A. Feddes, and B. Lesaffre, eds. 1996. Sustainability of
Irrigated Agriculture. Dordrecht, The Netherlands: Kluver Academic Publishers.
Perry, C. 2002. Keynote address. Wallingford, UK. Personal communication.
Population Reference Bureau. 2005. 2005 World population data sheet-Aug. Available
at: www.prb.org.
Powell, G. M., M. E. Jensen, and L. G. King. 1972. Optimizing surface irrigation
uniformity by nonuniform slopes. ASAE Paper No. 72-721. St. Joseph, Mich.:
ASAE.
Roos, M. 2001. How do we determine the real amount of water available for transfer
from one basin to another. In Transbasin Water Transfers, Proc. USCID Water
Mgmt. Conf., 245-255. J. Schaack, ed. Denver, Colo.: USCID.
Rosegrant, M. W., X. Cai, and S. A. Cline. 2002. World Water and Food to 2025:
Dealing with Scarcity. Washington, D.C.: Int’l. Food Policy Res.
Sammis, T. W. 1981. Yield of alfalfa and cotton as influenced by irrigation. Agron. J.
73: 323-329.
Seckler, D., U. Amarasinghe, D. Molden, R. de Silva, and R. Barker. 1998. Water
demand and supply 1990 to 2025: Scenarios and issues. Colombo, Sri Lanka: Res.
Report 19. Int’l. Water Mgmt. Inst.
Sneed, R. E. 1970. Using sprinkler irrigation for crop cooling. Sprinkler Irrig. Assoc.
May: 5-7.
Sneed, R. E. 1972. Frost control. Sprinkler Irrig. Assoc. Proc., Annual Tech. Conf.,
41-53. Church Falls, Va.: The Irrigation Assoc.
Stewart, J. I., R. M. Hagan, R. E. Danielson, R. J. Hanks, and E. B. Jackson et al.
1977. Optimizing Crop Production Through Control of Water and Salinity Levels
in the Soil. Logan, Utah: Utah Water Res. Lab., Utah State Univ.
Stewart, B. A., and D. R. Nielsen, eds. 1990. Irrigation of Agricultural Crops. Agron.
Series 30. Madison, Wis.: American Soc. Agron., Madison.
Sun, M. 1988. Environmental awakening in the Soviet Union. Science 241: 1033-
1035.
Tiffen, M. 1989. Guidelines for the incorporation of health safeguards into irrigation
projects through intersectoral cooperation. PEEM Guidelines Series 1, VBC/89.5.
Geneva, Switzerland: World Health Organization.
Unger, P. W., and T. A. Howell. 1999. Agricultural water conservation: A global
perspective. In Water Use in Crop Production, 1-36. M. B. Kirkham, ed.
Binghamton, N.Y.: The Haworth Press, Inc.
Unrath, C. R. 1972a. The evaporative cooling effects of over tree sprinkler irrigation
on ‘Red Delicious’ apples. J. Am. Soc. Hort. Sci. 97(1): 55-58.
Unrath, C. R. 1972b. The quality of ‘Red Delicious’ apples as affected by over tree
sprinkler irrigation. J. Am. Soc. Hort. Sci. 97(1): 58-61.
Unrath, C. R., and R. E. Sneed. 1974. Evaporative cooling of ‘Delious Apples’: The
economic feasibility of reducing environmental heat stress. J. Am. Hort. Soc. 99(4):
372-375.
1999. 1997 Census of Agriculture. Vol. 3: Special Studies. 1998 Farm and Ranch
Irrigation Survey. Available at: www.nass.usda.gov/census/census97/fris/fris.htm.
USDA-NASS (U.S. Department of Agriculture National Agricultural Statistics Ser-
vice). 2004. 2002 Census of Agriculture. Vol. 3: Special Studies Part 1: Farm and
Ranch Irrigation Survey (2003). Available at: www.nass.usda.gov/census/census02
/fris /fris03.htm.
Viets, F. G. 1962. Fertilizer and efficient use of water. Adv. in Agron. 14: 223-264.
Wright, J. L. 1988. Daily and seasonal evapotranspiration and yield of irrigated alfalfa
in southern Idaho. Agron. J. 80: 662-669.
CHAPTER 3
PLANNING AND
SYSTEM SELECTION
Kenneth H. Solomon (California Polytechnic State
University, San Luis Obispo, California)
A. M. El-Gindy (Ain-Shams
University, Cairo, Egypt)
Sagit R. Ibatullin (Research Institute of Water
Economy,Taraz City, Kazakhstan)
Abstract. This chapter reviews the process of planning for and selecting an irriga-
tion system. Decision making involves the identification and evaluation of alternatives,
and making and implementing the decision. The initial problem situation is improved
through the cause-and-effect feedback loop, and a learning feedback loop influences
future decision making. Irrigation system selection involves both objective and subjec-
tive components of problems, goals, and values, and available augmentations to the
problem situation. An irrigation system in the broadest sense includes not only hard-
ware, such as equipment, land improvements, water conveyance, and control struc-
tures, but technology and support infrastructure as well. These may include guides for
equipment installation, operation, and maintenance, and plans for water management
and irrigation-related crop husbandry, training, and extension. Planning for an irri-
gation system must consider a wide range of physical, economic, human, and social
factors. Irrigation system selection must consider the capabilities, costs, and limita-
tions of potential irrigation methods. This chapter presents a brief review of such fac-
tors for the most common surface, sprinkler, and microirrigation systems.
Keywords. Decision making, Irrigation systems, Microirrigation, Planning, Selec-
tion, Sprinkler irrigation, Surface irrigation.
3.1 INTRODUCTION
The process of planning for, and selection of, an irrigation system should be based
upon an understanding of the goals and objectives to be met, a study of the resources
available, a creative determination of the possible alternatives, and an evaluation of
these alternatives according to economic and other criteria. This chapter presents fac-
tors to be addressed during these activities. Conceptual models for decision making
and for irrigation system selection provide a framework for the review of individual
factors. Irrigation system components beyond just hardware must be considered. Wa-
ter supply and system capacity must be sufficient to meet requirements. Economic
58 Chapter 3 Planning and System Selection
evaluation of irrigation options depends on the perspective of the evaluator and the
goals and values of all those involved with the irrigation development. Capabilities,
limitations, and evaluation factors for the most common types of irrigation systems are
briefly discussed.
3.1.1 A Conceptual Model for Decision Making
The decision-making process is illustrated in Figure 3.1. The process starts with a
problem situation to be improved or solved. Therefore, a decision must be made. The
situation at hand contains not only the actuality of the problem, but the potentiality of
a number of alternative courses of action to improve the situation or solve the prob-
lem. These action alternatives exist whether or not we are clever enough to recognize
them. Experience, judgment, and creativity increase our ability to identify the alterna-
tives that exist. Part of this process is to recognize that certain constraints are acting on
the situation with which we must deal. Notice that in the example of Figure 3.1, only
alternatives B, C and D were identified. Alternatives A and E are in fact possibilities,
but they were not thought of.
Next, value system(s) are applied to evaluate those alternatives that have been iden-
tified. Though it may be helpful to quantify these evaluations as much as possible,
value systems contain non-quantifiable elements. So, the final decision must take into
account not only the quantitative score for each alternative, but some qualitative
judgments as well. In the diagrammed example, for instance, alternative B was se-
lected over C, even though C has a higher quantitative score: apparently the intangi-
bles involved outweigh the six-point difference in score.
Like many human endeavors, decision making is a cyclical process. In this case,
two main feedback loops are identified. The primary feedback loop is the cause and
effect loop. This is the way in which the chosen action will alter the current situation.
Other aspects of the situation may be altered as well. The results of our decision will
influence the list of alternatives for further action, either with regard to this problem or
some future one.
Situation Evaluate
Action Identify Alternatives Make
Alternatives Alternatives Decision
Alt. Score
A
B B B 80 Select B
Problem C Reject
C C 86 C
D D D 62 Reject D
E
Experience Value Monitor

Judgment System(s) Results
Creativity of
Constraints Decision
Learning Process
Cause and Effect
Figure 3.1. A conceptual model of the decision process.

An additional feedback loop is the learning process. By monitoring the results of a

decision, we learn things. This learning increases our experience, judgment and per-
haps creativity. It may alter the constraints (or the way we perceive the constraints) of
the situation. Learning may allow us to identify action alternatives that we did not see
before—either for the problem just considered, or for future problems.
The learning process will also have an effect on our value systems. Learning, liv-
ing, and growing are inseparable and go hand in hand with the evolution of our value
systems. Thus the learning process will affect the way in which we evaluate alterna-
tives and make future decisions.
3.1.2 A Conceptual Model for Irrigation System Selection
Decision making within an irrigation context, based on our understanding of the
decision process in general, is illustrated in Figure 3.2. The selection process is drawn
onto a conceptual blackboard, divided into four quadrants: right to left into Objective
vs. Subjective; and top to bottom into the Existing Situation vs. things Brought to the
Situation. (Note that the items listed under each general heading are for illustration
only. The lists are not intended to be exhaustive, nor will all items listed necessarily
apply to any given situation.)
The objective part of the existing situation contains resources and problems. The
resource category is subdivided into physical and people-related resources. Problems
may be associated with resources (amount, timing, place, quality) or with constraints
of various types which prevent some otherwise desirable action. The subjective part of
the existing situation contains the goals and priorities we are trying to meet with this
irrigation decision, and the values by which the alternatives and results will be judged.
To the existing situation we bring items that we hope will solve the problems ob-
served; the existing situation is augmented. This augmentation always involves several
components: physical things such as irrigation equipment; people-related things such
Objective Subjective
Resources Problems Goals & Values

Physical People Resources: Priorities
Profit
Soil/Water Irrigators/Others Not Enough Food Health
Climate Skills Too Much Nutrition Employment
Experience Bad Timing Results Jobs Independence
Cultivars
Energy Education Poor Quality Augmented Exports Technology
Infrastructure Motivation Constraints Situation Income Tradition
Financial Markets Allows New
Matchup of
Problems &
Augmentation Capacities
for Solution
Irrigation Other
Equipment Equipment
Management Plans Management Plans New Net Benefits
Training Programs Training Programs + Advantages
New Problems New Problems – Disadvantages
Figure 3.2. A conceptual model for irrigation system selection. The items listed un-
der each general heading are for illustration only. The lists are not intended to be
exhaustive, nor will all items listed necessarily apply to any given situation.
as management plans and training; and inevitably, we bring along some new problems
to be dealt with. For example, irrigation adds a drainage requirement, or advanced
equipment adds training and maintenance requirements.
The results of our actions are formed out of the mix of the elements in the existing
situation and the items brought to the situation in the augmentation. The augmented
situation includes new capacities for solving the original problem, as well as new
problems.
However, the results box in Figure 3.2 is not the end. The results should be viewed
in light of the goals and priorities, and the values of the decision makers to determine
the net benefits. The net benefits box spans both the objective and subjective halves of
the conceptual blackboard, reminding us (as in the model for decisions) that what we
seek has both quantitative and qualitative aspects.
This is the framework in which irrigation system selection should take place. These
models do not constitute a selection process itself, but are rather a reminder of all the
things beyond the engineering details that must be considered.
3.1.3 An Irrigation System—More Than Just Hardware
An irrigation system is commonly thought of in terms of its hardware aspects: in-
field equipment, water conveyance and control structures, and land improvements
such as grading. It is important to realize, though, that a complete irrigation system
includes additional aspects of technology and support infrastructure.
A given technology includes not only relevant hardware, but the collection of
knowledge necessary to accomplish the desired function. Technology includes facts,
relationships, and rules for making decisions. In order to accomplish its function, an
irrigation system must include equipment and a variety of related “software.” Specifi-
cally, the irrigation technology package must include guides for design, installation,
operation, and maintenance of the equipment; plans for irrigation scheduling and wa-
ter management; and plans for the crop husbandry practices impacted by irrigation. It
must also include plans for irrigation-related training, research, development, and ex-
tension.
Just as equipment requires operational knowledge and plans, it requires a support
infrastructure as well. This infrastructure is partly physical and partly institutional.
Physical elements include dams, canals, offices, and facilities for research and demon-
stration. Institutional elements include water rights establishment and enforcement
bodies, irrigation districts, farmer cooperatives, repair and maintenance services, and
extension organizations. The infrastructure is required to make available the water,
other agronomic inputs, services, and technological information necessary for success-
ful irrigation.
In developed irrigation markets, the widespread existence and availability of tech-
nical irrigation information and infrastructure may lead to the misconception that irri-
gation system selection consists solely of equipment selection. Consider, for example,
the introduction of microirrigation to a developing irrigation market. In this case, it is
obvious that a choice of microirrigation equipment without a matching choice to fund
and implement a suitable package of microirrigation information, plans, and the neces-
sary support infrastructure is doomed to failure.
3.2 PLANNING FOR IRRIGATION

Planning for irrigation consists of identifying and collecting information on rele-
vant factors, followed by formulating and evaluating realistic options for irrigation
systems. “System” here is taken in the broad sense, including elements of equipment,
technical information, and infrastructure. As indicated in Figure 3.2, the initial plan-
ning phase certainly includes inventories of resources, problems, and the goals, priori-
ties, and values of the decision makers. Table 3.1 lists a number of considerations that
may have a bearing on irrigation decisions. A few of the key points listed are dis-
cussed further in the following sections.
3.2.1 Water Requirements and Water Supply
A key point to consider in any irrigation plan is whether the water available is suf-
ficient to adequately meet the water requirements for crop production. This compari-
son must be made in terms of both the total volume of water required and available
during the year, and the peak rate at which water is used and supplied. The water sup-
ply must meet or exceed the minimum requirement less any effective precipitation.
Effective precipitation is that portion of total precipitation which becomes available
to support plant growth (ASAE, 1998a). The timing and rate of precipitation can influ-
ence its availability for plant growth. Rainfall during the growing season is ineffective
if it runs off or percolates below the root zone. Precipitation falling outside the grow-
ing season is effective only to the extent that it is stored in the eventual crop root zone.
Hence, precipitation resulting in runoff, deep percolation, or evaporation prior to the
season is not effective. Precipitation and deep percolation occurring prior to the season
may be effective in meeting part of the leaching requirement (see Section 3.2.2).
The total volume of water required to produce a crop includes the water consumed
during crop growth, water used to maintain a favorable salt balance in the root zone,
and water used for certain husbandry practices such as germination, climate control,
and beneficial vegetation such as windbreaks or cover crops. If the water supply does
not meet the water volume requirement (reduced by any effective precipitation), irri-
gation options are limited. Either the supply must be augmented or the production plan
altered, for example by using shorter-season varieties, eliminating water-consuming
husbandry practices, or using a deficit irrigation schedule.
The highest rate of crop water use depends both on climatic factors and on the timing
of crop growth stages. Because weather varies from year to year, the peak-use rate will
vary as well. It is expected that the peak-use rate determined from average values of
weather variables will be exceeded one year in two. It may be prudent to study historical
weather and base plans on a more conservative figure, for example the peak-use rate
exceeded only one year in five, or one year in ten. The costs of both water supply and in-
field irrigation systems are sensitive to the peak flow rate, so it may not be economical to
design for abnormally high peak-use rates that are expected to occur only rarely.
The water supply and in-field irrigation systems should have the capacity to meet
the design peak-use rate, adjusted to compensate for water application efficiency, and
possibly adjusted again according to the strategies discussed in the following para-
graphs. Note that application efficiency may vary throughout the year. For example,
with surface irrigation systems, it may be difficult to uniformly apply the small
amounts required by crops during early growth stages, thus reducing application effi-
ciency. The application efficiency anticipated during the peak-use period should be
used in this adjustment.
Table 3.1. Factors for consideration in irrigation decision making.

Physical Factors Economic Factors
Crops, crop rotation Crop value (price)
Crop yields (especially if quality or quantity Investment capital
differs for different irrigation systems) Foreign exchange, hard currency
Cultural practices Credit
Soils Source & terms
Texture, depth, uniformity Interest rate
Intake rate, water holding capacity For long term debt
Erosion potential For operating loans
Salinity Cash flow requirements
Internal drainage Equipment life
Topography Costs & inflation
Water supply Equipment
Water rights Water, energy
Source, delivery schedule Other agricultural inputs
Quantity available, reliability Services
Current uses of water Labor, various skill levels
Water quality Supervision
Salinity & other chemical constituents Management
Suspended solids Operation & maintenance
Climate (wind, heat, frost) Repairs & replacement
Land value, availability Warranties
Flood hazard Incentives and subsidies
Catastrophic weather events Taxes
Water table Insurance
Energy availability, reliability, and form Opportunity costs for inefficiencies
Pests Markets, export/import
Infrastructure available Impact of irrigation on related field and
Water supply & conveyance factory operations
Agronomic inputs Potential for upgrades as conditions change
Equipment repair service Uncertainty
Technological support Social Factors
Weather data General values
Existing standards Specific goals and priorities
Equipment performance Legal constraints
Design, installation Political issues
Practices Local cooperation and support
Common good practice Local and governmental expectations
Human Factors Environmental issues
Labor Quality standards
Availability Wildlife habitat
Skills, experience Mitigation
Education Health issues
Potential for vandalism or theft History of irrigation experience
Level of automatic control desired Biases and taboos
Certain management strategies may allow additional adjustments in the design ca-
pacity of water supply and irrigation systems. Probably the most common of these is
an increase in the design capacity to create an allowance for downtime, that is, time
when the system is not operated so that it may be serviced, repaired, or maintained.
For example, ASAE (1998b) recommended that the design capacity for microirriga-
tion systems be sufficient to meet the peak-use rate in about 90% of the time available,
or when operating no more than 22 hours per day. In some cases the schedule of elec-
tric power rates will impact downtime decisions.
Another strategy is to plan for the use of stored soil water to meet the crop water
needs during the peak-use period. Successful use of this strategy requires that the irri-
gation system be operated so that the available soil water reservoir is full (or nearly
full) prior to entering the peak-use period. Peak-use water needs can be met by deliv-
eries from the irrigation system and by planned depletion from the root zone. The
amount that the design flow rate for the irrigation system can be decreased by this
strategy depends on the amount of available water that can be stored in the crop root
zone, and the duration of the peak-use period. This strategy carries the risk that an
abnormally long peak-use period or unexpected system failures during or just prior to
the peak-use period can result in unplanned stress to the crop due to insufficient water.
In those areas where rainfall may be expected during the peak-use period, agro-
nomic practices to increase the effectiveness of precipitation may reduce the required
flow rate for irrigation and supply systems. Tillage methods that increase temporary
surface storage of rainfall can increase effective precipitation by reducing runoff. Irri-
gation scheduling methods that don’t completely refill the crop root zone can increase
effective precipitation by reducing deep percolation.
In cases where the water supply cannot match the peak-use rate, water storage in a
pond or reservoir may solve the problem as long as the water supply volume is suffi-
cient to meet the total water volume required by the crop. During the off-peak periods,
water in excess of crop needs is diverted to storage. During the peak-use period, the
normal water supply is augmented from stored water.
Any action that increases the attainable application efficiency, such as the use of re-
turn-flow systems to capture and re-use surface irrigation runoff water, can help to
meet design peak-use needs. For a given gross application rate, increasing the applica-
tion efficiency will increase the system’s net application rate.
The advisability of these various water management strategies is frequently af-
fected by the value of the crop and the sensitivity of the crop yield to water.
3.2.2 Need for Drainage
Irrigation water inevitably contains salt, which is left behind when plants take water
from the root zone for evapotranspiration. Thus, salts tend to accumulate unless
leached away by excess water passing through the root zone. In humid regions the
excess may be provided by rainfall, but in arid regions additional irrigation water (in
excess of the crop’s consumptive use) is required for the purpose of leaching. Unless
this excess or leaching water is removed, high water tables and waterlogged soils can
result. High water tables can lead to soil salination due to upward movement of water
and salt. Drainage, natural or artificial, is necessary to carry away excess or leaching
water applied to maintain a favorable salt balance in the root zone.
Irrigation planning should include an investigation of the potential need for artifi-
cial drainage. Equipment, technology, and infrastructure for the drainage system, if
needed, should be included in the economics and evaluation of irrigation options.

3.2.3 Irrigation Economics
An important part of irrigation planning and system selection involves economic
evaluation of the alternatives. It is tempting to consider only the most obvious, initial
costs, but this is too simplistic an approach for sound decision making. Life cycle cost-
ing, which treats operation and maintenance costs as well as initial costs, is the pre-
ferred approach.
The basic price of the equipment or improvement (such as land grading) is not the
only initial cost. To this may be added sales tax, import duties or tariffs, and initial
loan fees if the purchase is to be financed with a loan. Freight and delivery charges
may also apply, and can be particularly important when considering systems utilizing
imported equipment. Freight charges are usually based on the weight of the material to
be shipped and the volume it occupies. Equipment assembly and installation may in-
volve additional initial expenses.
Important initial costs that are sometimes overlooked are expenses for a stock of
spare parts and for initial training. Particularly in remote areas or in developing mar-
kets without a fully mature maintenance and repair infrastructure, initial equipment
purchases should include an inventory of spare parts. An initial effort to train irrigators
and equipment operators may be necessary if the anticipated benefits of the chosen
irrigation system are to be achieved. The energy provider may assess an initial charge
to bring service (particularly electric power lines) to the farm site.
Annual costs to operate and maintain an irrigation system include charges for labor,
water, and energy. Allowances for preventative maintenance and repairs, repair parts,
taxes, insurance, and ongoing training should also be included in the annual operating
budget.
Different irrigation options often involve trade-offs between initial and annual op-
erating costs. For example, larger pipe sizes have higher initial costs, but reduce fric-
tion losses and hence may lower operating costs for pumping energy. Less equipment-
intensive systems may have reduced initial costs, but higher operating costs for irriga-
tion labor. In order to compare the full, life cycle cost of various irrigation options,
some method is needed to place initial and annual irrigation costs on the same basis.
A common approach is to convert the initial costs to an equivalent annual cost. This
is done by multiplying the initial cost by a factor (often called the capital recovery
factor) that depends on the interest rate and the expected economic life for the initial
cost item. The capital recovery factor may be calculated according to Equation 3.1:
i( 1 + i )n
CRF = (3.1)
( 1 + i )n − 1
where CRF = capital recovery factor, dimensionless
i = annual interest rate, decimal
n = expected economic life, years
Table 3.2 lists economic lives for different irrigation items and conditions. Increasing
the interest rate or reducing the economic life will increase the capital recovery factor
and the annualized cost. The total annualized cost for an irrigation option is the sum of
its annualized initial cost and its annual operating cost.
Table 3.2. Expected economic lives for irrigation equipment (years).

Adapted from McCulloch et al. (1967).
Item Maximum[a] Conservative[b] Difficult[c]
Irrigation systems
Side roll 20 15 12
Center pivot, linear move, LEPA 25 20 15
Surface irrigation systems 30 20 15
Individual items
Farm reservoirs (heavy silting) 25 20 15
Farm reservoirs (light silting) 50 35 25
Well 40 25 20
Irrigation pump 20 15 10
Pump power units
Diesel 15 12 8
Gas, propane 13 9 7
Gasoline 8 6 5
Electric motor 30 25 15
Water conveyance
Open ditches (unlined) 8 6 4
Concrete structures 25 20 15
Concrete pipe systems 20 15 10
Wood flumes 10 8 6
Large aluminum pipe (lightweight) 10 8 6
Coated steel pipe (underground use) 20 15 10
Coated Steel pipe (surface use) 12 10 8
Galvanized steel pipe (surface use) 25 20 15
PVC pipe (underground use) 40 20 18
Sprinkler equipment
Aluminum sprinkler laterals 25 15 10
Sprinklers (medium size) 15 12 10
Sprinklers (giant, large-volume guns) 13 10 8
Microirrigation equipment
Filters 20 15 10
Emitters, hose (permanent) 15 10 8
Thin-walled tape 3 1 1
Surface irrigation
Land grading 40 25 20
[a] [b] [c]
Assumptions regarding maintenance, operating and environmental conditions: [a] excellent mainte-
nance, ideal operating conditions, and favorable environmental circumstances; [b] typical or ordinary con-
ditions; [c] inadequate or infrequent maintenance, harsh operating or environmental conditions.
The current price for an initial or annual cost item may not always be the most ap-
propriate figure to use in an economic analysis. Subsidies, incentives, taxes, or gov-
ernmental policies may result in prices that do not reflect an item’s true cost or value.
The correct number to use in a comparative analysis can often be determined only
after considering who is paying the costs, receiving the benefits, or making the deci-
sions.
For example, consider the price and value of water. If, according to local policy, the
government pays all costs of water supply development and conveyance, the price of
water to the farmer may be zero and private irrigation decisions may be based on that
premise. For public irrigation projects, those sponsored or funded by the government
or financed by international organizations (for example, a World Bank loan), the use
of a zero water cost is questionable. In this case, a more appropriate water cost might
include allocations for at least the operation, and perhaps the facilities, of the water
delivery system. Further, perhaps credits for the recreational and environmental bene-
fits of water development, and charges for environmental degradation or the cost of
mitigation measures to avoid it, should be included as well.
Costs for other inputs to irrigated agriculture are subject to similar considerations.
While private decision making may be based on local prices, full cost, world market
cost, or even opportunity cost (net return possible by diverting that input to alternative
uses), values should be considered for public project decision making.
Note also that financial or resource constraints may prevent the selection of the
“most economical” irrigation option. Limitations on available capital, or credit limits,
may preclude some capital-intensive options, even if these would otherwise be the best
option economically. Constraints on available resources, such as land or water, may
have a similar effect on the decision process. In these instances, it may be helpful to
consider indicators such as the maximum net return per unit of constrained resource
(for example, maximum net return per unit land area, per unit water volume, or per
unit of capital invested) instead of or in addition to the usual economic indicators.
3.3 IRRIGATION SYSTEM SELECTION
To do a proper job of system selection, one must give careful consideration to the
capabilities and limitations of potential irrigation methods. The remainder of this
chapter will summarize such considerations for the most common types of irrigation
systems. Labor, management, energy, and economic factors relevant to each system
type are briefly addressed. This section is adapted from Solomon (1988), which was
based on an unpublished study by Anderson et al. (1987). U.S. costs cited are repre-
sentative of U.S. irrigation practice as of 1996, and are based on Solomon et al.
(1992), Westlands Water District (1988), as updated by the study of Jorgensen (1996).
Egyptian cost figures are based on studies conducted by the second author (El-Gindy)
in 1995, based on a field size of 20 ha.
3.3.1 Surface Irrigation
3.3.1.1 Types of surface irrigation. Basin, border strip, and furrow irrigation are
common types of surface irrigation. In basin irrigation, water is applied to a com-
pletely level (sometimes called “dead level”) area enclosed by dikes or borders. This
method of irrigation is used successfully for both field and row crops. The floor of the
basin may be flat, ridged, or shaped into beds, depending on crop and cultural prac-
tices. Basins need not be rectangular or straight sided, and the border dikes may or
may not be permanent. Basin irrigation is also called by a variety of other names:
check flooding, level borders, check irrigation, check-basin irrigation, dead-level irri-
gation, and level-basin irrigation.
Factors affecting basin geometry are available water stream size, topography, soil
properties, and degree of leveling required. Basins may be quite small or as large as
15 ha or so. Level basins simplify water management, since the irrigator need only
supply a specified volume of water to the field. With adequate, non-erosive stream
size, the water will spread quickly over the field, minimizing non-uniformities in in-
undation time. Basin irrigation is most effective on uniform soils, precisely leveled,
when large stream sizes (relative to basin area) are available. High efficiencies are
possible with low labor requirements.
Border strip irrigation uses land formed into graded strips, level across the narrow
dimension but sloping along the long dimension, and bounded by ridges or borders.
Water is turned into the upper end of the border strip and advances down the strip.
After a time, the water is turned off, and a recession front, where standing water has
soaked into the soil, moves down the strip. High irrigation efficiencies are possible
with this method of irrigation, but may be difficult to obtain in practice, because the
balancing of advance and recession phases during water application is difficult.
Border strip irrigation is one of the most complicated of all irrigation methods. The
primary design factors are border length and slope, stream size per unit width of bor-
der, planned soil water deficiency at the time of irrigation, soil intake rate, and degree
of flow retardance by the crop as the water flows down the strip. However, because of
the large variations in field conditions that occur during the season, the irrigator can
have as great an effect on irrigation efficiency as the system designer.
Furrow irrigation uses furrows, which are small, sloping channels formed in the
soil. Infiltration occurs laterally and vertically through the wetted perimeter of the
furrow. Systems may be designed with a variety of shapes and spacings. Optimal fur-
row lengths are primarily determined by intake rates and stream size. Even when soils
are uniform, the intake rates in furrows may be quite variable due to cultural practices.
The intake rate of a newly formed furrow will be greater than a furrow that has been
irrigated. Wheel-row furrows can have greatly reduced infiltration rates due to com-
paction. Because of the many design and management-controllable parameters, furrow
irrigation systems can be utilized in many situations, within the limits of soil uniform-
ity and topography. With runoff-return flow systems, furrow irrigation can be a very
uniform and efficient method of applying water. However, the uniformity and effi-
ciency are highly dependent on the timing of the irrigation and the skill of the irriga-
tor. Mismanagement can severely degrade system performance.
3.3.1.2 Capabilities and limitations. Some form of surface irrigation is adaptable
to most any crop. Basin and border strip irrigation have been successfully used on a
wide variety of crops, but furrow irrigation is less well adapted to field crops if cul-
tural practices require equipment travel across the furrows. Basin and border strip irri-
gations flood the soil surface and will cause some soils to form a crust, which may
inhibit the emergence of seedlings.
Surface irrigation systems perform better when soils are uniform, since the soil
controls the intake of water. For basin irrigation, basin size should be appropriate for
soil texture and infiltration rate. Basin lengths should be limited to 100 m on very
coarse-textured soils, but may reach 400 m on other soils. Furrow irrigation is possible
with all types of soils, but soils with extremely high or low intake rates require exces-
sive labor or capital cost adjustments that are seldom economical. Uniform, mild
slopes are best adapted to surface irrigation. Undulating topography and shallow soils
do not respond well to grading to a plane. Steep slopes and irregular topography in-
crease the cost of land leveling and reduce basin or border size. Deep cuts may expose
areas of nonproductive soils, requiring special fertility management. Erosion control
measures may be required if large stream sizes are used. In areas of high-intensity
rainfall and low intake rate soils, surface drainage should be considered with basin
irrigation, to reduce damage due to untimely inundation.
It is important that irrigation stream size be properly matched to basin or border
size and furrow lengths for uniform irrigation. Since intake rates may vary during the
season, it is helpful if the water supply rate for border and furrow systems can be var-
ied from one irrigation to the next. Border and furrow systems are not suitable for
leaching of salts for soil reclamation, since the water cannot be held on the soil for any
length of time. The basin method, however, is ideal for this purpose. Under normal
operating conditions, leaching adequate for salinity control can be maintained with
basin, border, or furrow irrigation.
High application efficiencies are possible with all surface irrigation methods, but it
is much easier to obtain these potential efficiencies with the basin method. Design
application efficiencies for basin systems should be high, perhaps 80% to 90%, for all
but very high intake rate soils. Reasonable efficiencies for border strip irrigation are
from 70% to 85%, and from 65% to 75% for furrow irrigation. With either the border
of furrow methods, runoff-return flow systems may be needed to achieve high applica-
tion efficiencies. The system designer and operator can control many of the factors
affecting irrigation efficiency, but the potential uniformity of water application with
surface irrigation is limited by the variability of soil properties (primarily infiltration
rate) throughout the field. Field studies indicate that even for relatively uniform soils,
the uniformity of infiltration rates may be only about 80%, as calculated in a manner
analogous to distribution uniformity for irrigation application amounts. Surface irriga-
tion uniformity estimates based on infiltration time differences need to be decreased
somewhat to account for soil variability.
Basin irrigation involves the least labor of the three surface methods, particularly if
the system is automated. Border and furrow systems may also be automated to some
degree to reduce labor requirements. The complicated art of border irrigation (and to a
lesser extent furrow irrigation) requires skilled irrigators to obtain high efficiencies.
The labor skill needed for setting border or furrow flows can be decreased with costly
equipment (gated pipe or automated valves/gates). The setting of siphons or slide
openings to obtain the desired flow rate is a required skill, but one that can be learned.
With surface irrigation, little or no energy is required to distribute the water
throughout the field, but some energy may be expended in bringing the water to the
field, especially when water is pumped from groundwater. In some instances these
energy costs can be substantial, particularly with low application efficiencies. Some
labor and energy will be necessary for land grading and preparation.
3.3.1.3 Economics. A major cost in surface irrigation is that of land grading or lev-
eling. The cost is directly related to the volume of earth that must be moved, the area
to be graded, and the length and size of farm ditches. Typical earth moving volumes
are on the order of 800 m3/ha, but have on occasion exceeded 2500 m3/ha. Volumes
greater than 1500 m3/ha are generally considered excessive, suggesting a design re-
view may be needed. For the U.S., typical earth moving costs are $0.50/m3. For basin
irrigation, final finishing with laser-controlled drag scrapers after primary earth mov-
ing will cost about $110/ha. Touch-up leveling (at about $50/ha) may be required
every two to three years, although some farmers choose to touch-up each year. In
Egypt, land leveling costs for surface irrigation are about $170/ha.
Ditch construction can cost from $3/m for earthen ditches to $50/m for large, con-
crete-lined ditches. Buried, low-pressure plastic or concrete pipelines for low flows
can cost about double the cost of concrete-lined ditches, and may cost 5 to 10 times as
much for higher flows. They are generally uneconomical on flat terrain where pump-
ing is not required. They may be desirable on steeper slopes (over 1%) or on undulat-
ing terrain. Lined canals for a 20-ha Egyptian farm can cost about $1300/ha.
U.S. estimates for furrow irrigation labor requirements range from 1.9 to 2.5 h/ha
per irrigation or 10 to 13 h/ha per year (based on about 5 irrigations per season). In
Egypt, labor costs for surface irrigation range from $80/ha to $170/ha per year. In the
U.S., siphon tubes may cost about $30/ha, while gated pipe may cost $370/ha to
$735/ha, depending on pipe size, material and length of furrow run. In Egypt, gated
pipe for a 20-ha field costs about $340/ha.
An operational storage reservoir (i.e., one for short-term storage of water) may be
advisable to permit use of a large stream size accumulated from a smaller steady flow.
A medium-sized, compacted-earth reservoir capable of storing 24 h water volume would
cost about $250/ha for a small (to 16 ha) farm in the U.S. For larger U.S. farms, the cost
can drop to about $125/ha. A lined reservoir may cost two to five times as much. In Cali-
fornia’s Westlands Water District (1988), a typical runoff recovery and reuse system
(3700 m3 sump capacity, serving four fields of 65 ha each) costs about $170/ha served.
3.3.2 Sprinkler Irrigation
3.3.2.1 Types of sprinkler irrigation. In sprinkler irrigation, water is delivered
through a pressurized pipe network to sprinklers, nozzles, or jets, which spray the wa-
ter into the air, to fall to the soil in an artificial “rain.” The basic components of any
sprinkler system are a water source, a pump to pressurize the water, a pipe network to
distribute the water throughout the field, sprinklers to spray the water over the land,
and valves to control the flow of water. The sprinklers, when properly spaced, give a
relatively uniform application of water over the irrigated area. Sprinkler systems are
usually designed to apply water at a lower rate than the soil infiltration rate, so that the
amount of water infiltrated at any point depends upon the application rate and time of
application, but not the soil infiltration rate.
Hand move or portable sprinkler systems employ a lateral pipeline with sprinklers
installed at regular intervals. The lateral pipe is often made of aluminum, with 6-, 9-,
or 12-m sections, and special quick-coupling connections at each pipe joint. The sprin-
kler is installed on a pipe riser so that it may operate above the crop being grown (in
orchards, the riser may be short, so that the sprinklers operate under the tree canopy).
The risers are connected to the lateral at the pipe coupling, with the length of pipe sec-
tion chosen to correspond to the desired sprinkler spacing. The sprinkler lateral is
placed in one location and operated until the desired water application has been made.
Then the lateral line is disassembled and moved to the next position to be irrigated.
This type of sprinkler system has a relatively low initial cost, but a high labor re-
quirement. It can be used on most crops, though with some, such as corn or sugar
cane, the laterals become difficult to move as the crop reaches maturity. On bare
“sticky” soils, moving the lateral lines is very difficult, and an extra line (a “dry” line)
is used to give the soil under the “wet” line some time to dry out before that particular
line is moved.
The side roll sprinkler system is a variation on the hand move lateral sprinkler line.
The lateral line is mounted on wheels, with the pipe forming the axle (specially
strengthened pipe and couplers are used). The wheel diameter is selected so that the
axle clears the crop as the lateral is moved. A drive unit, usually powered by an air-
cooled gasoline-powered engine located near the center of the lateral, is used to move
the system from one irrigation position to another by rolling the wheels.
Gun type sprinkler systems utilize a high-capacity, high-pressure sprinkler (the
“gun”) mounted on a trailer, with water being supplied through a flexible hose or from
an open ditch along which the trailer passes. The gun may be operated in a stationary
position for the desired time, and then moved to the next location. However, the most
common use is as a continuous move system, where the gun sprinkles as it moves, in a
traveling gun system. The trailer may be moved through the field by a winch and ca-
ble, or it may be pulled along as the hose is wound up on a reel at the edge of the field.
The gun used is usually a part-circle sprinkler, operating through 80% to 90% of the
circle for best uniformity, and allowing the trailer to move ahead on dry ground. These
systems can be used on most crops, though due to the large droplets and high applica-
tion rates produced, they are best suited to coarse soils having high intake rates and to
crops providing good ground cover.
The center pivot system consists of a single sprinkler lateral supported by a series
of towers. The towers are self-propelled so that the lateral rotates around a pivot point
in the center of the irrigated area. The time for the system to revolve through one
complete circle can range from a half a day to many days. To keep the lateral aligned
as it moves around the circle, the outer sections of the lateral must travel faster, and
cover larger areas per rotation, than the inner sections. Thus, the instantaneous water
application rate must increase with distance from the pivot to deliver an even applica-
tion amount. The high application rate at the outer end of the system may cause runoff
on some soils. A variety of sprinkler products have been developed specifically for use
on these machines to better match water requirements, water application rates, and soil
characteristics. Since the center pivot irrigates a circle, it leaves the corners of the field
unirrigated (unless additions of special equipment are made to the system). Center
pivots are capable of irrigating most field crops, and have on occasion been used on
tree and vine crops.
Linear move (or lateral move) systems are similar to center pivot systems in con-
struction, except that neither end of the lateral pipeline is fixed. The entire line moves
down the field in a direction perpendicular to the lateral. Water delivery to the con-
tinuously moving lateral is by flexible hose or open ditch pickup. The system is de-
signed to irrigate rectangular fields free of tall obstructions. Both the center pivot and
the linear move systems are capable of very highly controlled and efficient water ap-
plications. They require moderately high capital investments, but have low irrigation
labor requirements.
Low energy precision application (LEPA) systems are similar to center pivot and
linear move irrigation systems, but are different enough to deserve separate mention.
The lateral line is equipped with drop tubes and very low-pressure orifice emission
devices discharging water just above the ground surface into furrows. This distribution
system is often combined with micro-basin land preparation for improved runoff con-
trol (and to retain rainfall which might fall during the season). High efficiency irriga-
tion is possible, but requires either very high soil intake rates or adequate surface stor-
age in the furrow micro-basins to prevent runoff or non-uniformity along a furrow.
Solid set systems are similar in concept to the hand move lateral sprinkler system,
except that enough laterals are placed in the field that it is not necessary to move pipes
during the season. The laterals are controlled by valves that direct the water into the
laterals irrigating at any particular moment. The pipe laterals for the solid set system
are moved into the field at the beginning of the season (after planting and perhaps the
first cultivation), and are not removed until the end of the irrigation season (prior to
harvest). The solid set system utilizes labor available at the beginning and ends of the
irrigation season, but minimizes labor needs during the irrigation season. A permanent
system is a solid set system where the main supply lines and the sprinkler laterals are
buried and left in place permanently (this is usually done with PVC plastic pipe).
3.3.2.2 Capabilities and limitations. Nearly all crops can be irrigated with some
type of sprinkler system, though the characteristics of the crop, especially the height,
must be considered in system selection. Sprinklers are sometimes used to germinate
seed and establish ground cover for crops like lettuce, alfalfa, and sod. The light fre-
quent applications that are desirable for this purpose are easily achieved with some
sprinkler systems. Most soils can be irrigated with the sprinkler method, although soils
with an intake rate below 5 mm/h may require special measures. Sprinklers can be
used on soils that are too shallow to permit surface grading or too variable for efficient
surface irrigation. In general, sprinklers can be used on any topography that can be
farmed. Land leveling is not normally required.
Leaching salts from the soil for reclamation can be done with sprinklers using
much less water than is required by flooding methods, although a longer time is re-
quired to accomplish the reclamation. This can be particularly important in areas with
a high water table. A disadvantage of sprinkler irrigation is that many crops (citrus, for
example) are sensitive to foliar damage when sprinkled with saline waters. In contrast,
other crops (apples, peas) react favorably to sprinkler irrigation at harvest, actually
producing higher-quality fruit.
Attainable irrigation efficiencies for different sprinkler systems are given in Table 3.3.
Labor requirements vary depending on the degree of automation and mechanization
of the equipment used. Hand move systems require the least degree of skill, but the
greatest amount of labor. At the other extreme, center pivot, linear move, and LEPA
systems require considerable skill in operation and maintenance, but the overall
amount of labor needed is low.
Energy consumption relates to operating pressure requirements (at the inlet to the
system), which vary considerably among sprinkler systems. At the extremes, the
LEPA systems may require only 100 kPa or so, while the traveling gun system may
require 700 kPa or more. Other systems may require 200 to 400 kPa, depending on the
design of the sprinklers and nozzles chosen, spacing, crop, climate, and other factors.
Table 3.3. Attainable application efficiencies with sprinkler irrigation.
System Type Efficiency (%)
Hand move or portable 65-75
Side roll 65-75
Traveling gun 60-70
Center pivot 75-90
Linear move 75-90
LEPA 80-95
Solid set or permanent 70-80
3.3.2.3 Economics. Table 3.4 summarizes cost factors for sprinkler irrigation sys-
tems based on U.S. prices and conditions. Capital costs depend on the type of system
and size of the irrigated area. These capital costs assume that water is available at
ground level at the side of the field, and include mainline and pumping plant. Table
3.5 summarizes costs for pressurized irrigation systems in Egypt.
Energy costs are highly variable from place to place. The energy requirements
listed in Table 3.4 may be used to estimate costs by applying the locally appropriate
unit energy cost. A pumping plant efficiency of 75% has been assumed. The energy
figures cited are in terms of kWh per 1000 m3 (gross) of water applied.
Table 3.4. Sprinkler irrigation system costs, U.S. irrigation practice.

Typical Energy Labor Maintenance
Field Capital Use Required Cost
Size Cost (kWh per (hours per Factor[a]
System Type (ha) ($/ha) 1000 m3) 1000 m3) (%)
Hand move or portable 65 500-750 85-215 1.65 2
Side roll 65 1300-1500 85-215 1.17 2
Traveling gun 32 960-1200 350-490 0.68 6
Center pivot:
without corner system 55-80 800-1250 85-235 0.10 3
with corner system 60 1000-1450 100-245 0.10 6
Linear move (ditch fed) 130 1375-2500 85-235 0.19 8
Linear move (hose fed) 130 1625-2750 125-265 0.19 7
Solid set aluminum 65 3250-4000 85-215 0.97 2
Permanent 65 2500-3750 85-215 0.10 1
[a]
Annual maintenance costs are expressed as a percentage of the system capital cost.
Table 3.5. Irrigation costs, Egyptian irrigation practice, assuming a 20-ha field size.
Source Initial Energy Maintenance
of Capital & & Spare
Water Cost Labor Consumables Parts
Supply ($/ha) ($/ha) ($/ha) ($/ha)
Solid set sprinkler Canal 3570 30 160 115
Well 4860 30 170 135
Hand move sprinkler Canal 2285 60 160 85
Well 3570 60 170 130
Center pivot, Canal 2235 35 35 70
low pressure type Well 3930 35 50 117
Minisprinkler Canal 3570 60 85 100
Well 5000 60 100 160
Bubbler Canal 3570 60 85 100
Well 5000 60 100 160
Drip on orchards Canal 2235 60 85 60
Well 3570 60 100 100
Drip on vegetables Canal 3430 70 130 100
or row crops Well 4860 70 145 150
Operating labor costs vary by system type and local costs for labor. Table 3.4 gives
typical values for labor hours required per 1000 m3 (gross) of irrigation water applied.
Maintenance costs are difficult to predict, but the data in Table 3.4 may be used as an
approximate guide. The annual maintenance cost is estimated by multiplying the ini-
tial capital cost of the system by the tabulated percentage factor.
3.3.3 Microirrigation
3.3.3.1 Types of microirrigation. Microirrigation is a general term, and includes
several more specific methods (ASAE, 1998b). Drip or trickle irrigation (the terms
are often considered equivalent) applies the water through small emitters to the soil
surface, usually at or near the plant to be irrigated. Subsurface drip irrigation is the
application of water below the soil surface. Emitter discharge rates for drip and sub-
surface drip irrigation are generally less than 8 L/h for single-outlet emitters, 12 L/h
per meter for line-source emission devices. Bubbler irrigation is the application of a
small stream of water to the soil surface. The applicator discharge rate (up to 250 L/h)
exceeds the soil’s infiltration rate, so the water ponds on the soil surface. A small ba-
sin is used to control the distribution of water. Microspray irrigation applies water to
the soil surface by a small spray or mist. Discharge rates are usually less than 175 L/h..
3.3.3.2 Capabilities and limitations. Microirrigation irrigation is best suited for
tree, vine, and row crops. The main limitation is the cost of the system, which can be
quite high for closely spaced crops. Complete cover crops such as grains or pasture
cannot be economically irrigated with microirrigation systems. Microirrigation is suit-
able for most soils, with only the extremes causing any special concern. On very fine-
textured soils, localized high application rates may cause ponding, with potential run-
off, erosion, and aeration problems. On very coarse-textured soils, lateral movement of
water will be limited. This can require several emission outlets per plant to wet the
desired root volume. With proper design, using pressure-compensating emitters and
pressure regulators if required, microirrigation can be adapted to virtually any topog-
raphy. In some areas, microirrigation is successfully practiced on such steep slopes
that cultivation becomes the limiting factor.
Microirrigation uses a lower rate of water application over longer periods of time
than other irrigation methods. The most economical design would have water flowing
into the farm area throughout most of the day, every day, during peak-use periods. If
water is not available on a continuous basis, on-farm water storage may be necessary.
Microirrigation can be used successfully with waters of some salinity, although some
special cautions are needed. Salts will tend to concentrate at the perimeter of the wet-
ted soil volume. If too much time passes between irrigations, the movement of soil
water may reverse itself, bringing salts back into the root zone. Salts concentrating on
the surface around the edge of the surface wetted area can be a hazard should a light
rain occur. Such a rain can move the salts down into the root zone, without applying
enough water to leach the salts through and below the root zone. When rain falls after
a period of salt accumulation, irrigation should continue as normal until about 50 mm
of rain have fallen, to prevent salt damage. In arid regions where annual rainfall is
insufficient to leach the salts (less than 300 to 400 mm), artificial leaching may be
necessary from time to time, requiring the use of a supplemental sprinkler or surface
irrigation system.
Though a form of pressurized irrigation, microirrigation is considered a low-
pressure, low flow-rate method. These conditions require small openings in the flow
channel of the emission devices, which can be prone to plugging. The sensitivity of
emitters to plugging varies with design, but virtually all emitters will require some
degree of water treatment in agricultural situations. Cyclonic separators and screen
filters are used to remove inorganic particles from the irrigation water, and media fil-
ters are used to remove organic contaminants. Chemical treatment of the water may
also be required to control biological activity in the water, to adjust pH, or to prevent
chemical precipitation which could plug emitters. Proper design and care of the water
treatment system is vital to the successful use of microirrigation.
Properly designed and maintained microirrigation systems are capable of high effi-
ciencies. Microirrigation systems should be designed for efficiencies on the order of
90% to 95%. With reasonable care and maintenance, field efficiencies in the range of
80% to 90% may be expected. Where plugging is a problem, or emitter performance is
highly variable (high variations in emitter flow rate due to either pressure differences
in the system or inconsistent manufacturing processes), field efficiencies may be as
low as 60%. A large field study in California (Hansen et al., 1995) found that field-
measured microirrigation system efficiencies averaged 80%.
Due to their low flow characteristics, microirrigation systems usually have few
subunits and are designed for long irrigation times. The systems are easily operated
manually, but can also be fully automated. Thus, the major labor requirement is for
system maintenance and inspection. The amount of maintenance labor required is re-
lated to the sensitivity of the emitters to plugging and the quality of the irrigation wa-
ter. In a vineyard situation, one irrigator can inspect and maintain about 20 ha per day.
Another estimate for microirrigation labor is 0.4 h per 1000 m3 (gross) of irrigation
water applied.
Microirrigation systems may use less energy than other forms of pressurized irriga-
tion. The emission devices usually operate at pressures ranging from 35 to 175 kPa.
Additional pressure is required to compensate for pressure losses through the control
head (filters and control valves) and the pipe network. System pressures range from
about 135 kPa (small systems on flat terrain) to 400 kPa (larger systems on undulating
terrain).
3.3.3.3 Economics. Microirrigation system costs can vary greatly, depending on
crop (plant, and therefore emitter and hose, spacings) and type of hose employed
(semi-rigid or “disposable” thin-walled tubing called “drip tape”). Microirrigation
costs will be the lowest for widely-spaced orchard crops, and then increasingly costly
for closer spaced vines, and for very closely spaced vegetables and row crops. Mi-
croirrigation system costs are summarized in Table 3.6. These cost figures are for
high-quality systems and include pumps, filters, controls, mainlines, manifolds, and
emitters. In situations where more basic pump, filtration, and control equipment will
suffice, costs may be 20% to 25% lower than the figures given.
Typical operation and maintenance costs for microirrigation systems vary greatly
depending on local circumstances and irrigation efficiencies achieved. One approach
is to estimate operation and maintenance costs (per hectare per year) as a percentage
of the initial capital cost, as shown in Table 3.7.
Other approaches to figuring operating costs are based on estimates of energy and
labor requirements. An energy requirement of 70 to 140 kWh per 1000 m3 (gross) of
irrigation water applied may be used for microirrigation systems. A corresponding
estimate for labor required is 0.4 h per 1000 m3 (gross) of irrigation water applied.
Table 3.6. Microirrigation system costs, U.S. irrigation practice.

Crop Microirrigation System Type Capital Cost ($/ha)
Trees Drip 2250-2500
Minisprinkler 2700-3200
Microsprayer 2500-3000
Vines Drip 2000-3000
Sprinkler/drip combination[a] 5000-6000
Row crops Drip, retrievable laterals 3000-5000
Drip, disposable laterals[b] 1900-3000
[a]
Combination permanent set sprinkler and drip system for frost protection and irrigation on vines.
[b]
Disposable laterals cost $500-600/ha annually.
Table 3.7. Annual operating costs for microirrigation systems.

Annual Cost of Operation and Maintenance
Expense Category as a Percentage of Initial Capital Cost
Labor 1.5
Power[a] 3-7
Maintenance 3
Taxes and insurance 2
[a]
Depends on system efficiency.
REFERENCES
Anderson, C. L., A. W. Blair, R. D. Bliesner, A. J. Clemmens, G. L. Dobbs, M. J.
English, A. D. Halderman, D. R. Hay, J. D. Hedlund, J. L. Merriam, J. A. Replogle,
L. J. Ring, M. L. Soffran, K. H. Solomon, R. E. Walker, I. A. Walter, J. E. Welton
and M. Yitayew. 1987. Selection of Irrigation Methods for Agriculture. June 22,
1987, Draft Version. Unpublished manuscript. Reston, Va.: On-Farm Irrigation
Committee, Irrigation and Drainage Division, ASCE.
ASAE. 1998a. S526.1: Soil and water engineering terminology. St. Joseph, Mich.:
ASAE.
ASAE. 1998b. EP405.1: Design, installation and performance of trickle irrigation
systems. St. Joseph, Mich.: ASAE.
Hanson, B., B. Bowers, B. Davidoff, D. Kasapligil, A. Carvajal, and W. Bendixen.
1995. Field performance of microirrigation systems. In Proc. Fifth International
Micorirrigation Congress, Microirrigation for a Changing World: Conserving
Resources/Preserving the Environment, 769-774.
Jorgensen, G. 1996. Survey of current irrigation costs. Unpublished notes. Fresno,
Calif.: Center for Irrigation Technology, California State Univ.
McCulloch, A. W., J. Keller, R. M. Sherman, and R. C. Mueller. 1967. Ames
Irrigation Handbook for Irrigation Engineers. W. R. Ames Company, Publisher.
Solomon, K. H. 1988. Irrigation system selection. Irrigation Notes, July 1988. CATI
Publication No. 880702. Fresno, Calif.: Center for Irrigation Technology,
California State Univ.
Solomon, K. H., D. L. Nef, G. Jorgensen, and D. F. Zoldoske. 1992. Higher
agricultural electricity rates: A San Joaquin Valley perspective. Technical Report
for the California Energy Commission. Fresno, Calif.: Center for Irrigation
Technology, California State Univ.
Westlands Water District. 1988. Economic guidelines for on-farm irrigation system
modifications. Fresno, Calif.: Westlands Water District.
CHAPTER 4
ENVIRONMENTAL
CONSIDERATIONS
Thomas J. Trout (USDA-ARS,
Dean D. Steele (North Dakota State University,
Fargo, North Dakota)
Keith O. Eggleston (U.S. Bureau of Reclamation,
Denver, Colorado)
Abstact. Irrigation, through storing, diverting, spreading, consuming, and return-
ing water, and enabling more intensive agriculture, affects the environment in both
positive and negative ways. In this chapter, we provide information that will assist in
the identification and evaluation of environmental impacts of irrigation, and tools to
reduce negative impacts. Three processes that can impact the environment are dis-
cussed in detail: (1) water storage, diversion, and consumption; (2) chemical leaching
to groundwater, and (3) soil erosion and surface water runoff. Included as possible
impacts are threats to agricultural sustainability and diminished human aesthetics, as
well as effects on water quality, plant and animal habitat, and human health.
Keywords. Deep percolation, Environmental impacts, Erosion, Irrigation, Leach-
ing, Runoff.
4.1 INTRODUCTION
Each year, 3100 km3 of water—about 5 times the annual flow of the Mississippi—
is removed from the Earth’s rivers, streams, and underground aquifers to water crops.
Irrigation accounts for 70% of the global water diversions (FAO, 1996). In the U.S.
about 189 km3 or 65% of offstream freshwater withdrawals, excluding thermoelectric
power, is for irrigation. In the semi-arid western states, the portion withdrawn for irri-
gation exceeds 80% (Hutson et al., 2004). In the process of storing, diverting, trans-
porting, spreading, consuming, and returning water, and enabling more intensive agri-
cultural activities, the land, water, plant and animal, and human resources are changed.
Some changes are positive and others are negative.
The purpose and primary benefit of irrigated agriculture is the production of food
and fiber for human use and feed for livestock consumption. Irrigated agriculture pro-
duces 40% of the global harvest on 20% of the cropland (FAO, 1996). Several popu-
lous countries in Asia depend on irrigated land for a majority of their food production
(FAO, 2002). In the U.S., about 50% of crop value is produced on 18% of the har-
vested crop area that is irrigated (NASS, 2004), including most of the fresh fruits and
vegetable crops. Without irrigation, increases in agricultural yields and outputs that
have fed the world’s growing population would not have been possible (Rosegrant et
al., 2002). Without irrigation, more land would need to be brought under cultivation
and production would be more weather dependent, and thus less dependable. Irrigation
improves the quality of life for the local agricultural community by increasing and
stabilizing incomes and creating jobs. Irrigation water storage facilities may also pro-
vide recreation for humans, habitat for animals, and relief from flooding; and irriga-
tion conveyance and drainage systems often provide habitat for plants and animals.
Irrigation water often provides habitat oases in the desert.
Irrigation may also negatively impact the environment. Erosion and salination of
cropland reduce its productivity. Excessive groundwater pumping results in depletion
of this valuable resource for current and future users. Diversion of river flows may
degrade or destroy wildlife habitat and recreation opportunities. Water shortage in arid
areas may limit urban and industrial expansion. Surface and subsurface drainage water
from irrigation is usually of lower quality than the water supply and may result in pol-
lution of both groundwater and surface waters.
Proper irrigation system design and management can prevent, or at least minimize,
many of the potential negative impacts. Unavoidable negative impacts of irrigation
should be recognized and evaluated. Only through recognition and evaluation of the
various environmental effects can it be assured that the positive impacts exceed the
negative for the system owner, the community, and society.
The range of irrigation systems and environmental situations varies too widely to
discuss all possible environmental impacts. Tables 1 and 2 list many of the impacts of
irrigation on the environment. The lists are not comprehensive, but many of the proc-
esses and potential impacts are common to most systems. One common environmental
hazard, soil salination, is discussed in detail in Chapter 7 and will not be discussed
here. The authors’ experiences are primarily in semi-arid environments in the western
U.S., and the discussion will be oriented toward problems experienced in that setting.
Mock and Bolton (1993) give a more comprehensive listing of potential environmental
impacts of water projects.
In this chapter, we provide information that will assist in the identification and
evaluation of environmental impacts of irrigation, and tools to reduce negative im-
pacts. Three processes that can impact the environment are discussed in detail: (1)
water storage, diversion, and consumption; (2) chemical leaching to groundwater; and
(3) soil erosion and surface water runoff. Included as possible impacts are threats to
agricultural sustainability and diminished human aesthetics, as well as effects on water
quality, plant and animal habitat, and human health.
Irrigated agriculture requires large amounts of water. Most irrigated agriculture is
located in areas where normal water supplies during the growing season are low.
Therefore, water is collected from large areas, stored in reservoirs until required, and
diverted to irrigated lands where much of it is consumed. The surface water supply is
often augmented by pumping from groundwater aquifers that have accumulated and
reached equilibrium conditions over hundreds of years. The diverted or pumped water
that is not consumed return flows into drains, streams, lakes, and marshes or perco-
lates to groundwater aquifers. These processes disrupt the natural hydrology, and thus
change the environment associated with that hydrology. Although these processes are
often external to the on-farm irrigation process, they are driven by irrigation water
needs and are a consequence of irrigated agriculture.
78 Chapter 4 Environmental Considerations
Table 4.1. Possible negative environmental impacts of irrigation.

Impact
Agricul- Natural
tural Pro- Envi- Human Human
Process Effect Possible Consequences ductivity ronment Health Aesthetics
Degraded fish and
X X
wildlife habitat
Decreased transport
and/or dilution of sedi- X X X
Reduced or ments and contaminants
altered Increased river water
river flows temperatures and de- X
creased dissolved oxygen
Decreased water supplies
for environmental, munic- X X X
Surface ipal and industrial needs
water Increased habitat for
storage, disease vectors/hosts X X
diversion, (mosquitoes, snails, etc.)
and Physical hazards of
consumption Creation of drowning (humans and X X
artificial
animals)
surface water
Entrapment and concen-
bodies
tration of substances
(storage X X X
(nutrients, salts, metals,
reservoirs and
pesticides)
conveyance
channels) Flooding due to failure of
X X X
an irrigation structure
Submersion of land and
free flowing river sec- X X X
tions
Increased pumping costs X
Increased
Saltwater intrusion in
pumping depths X X
coastal areas
Depletion of domestic
and irrigation water sup- X X
Groundwater
plies
depletion
Aquifer consolidation X
Groundwater
Land subsidence X X
pumping
Reduced wetlands X X
and
consumption Degraded fish and
X X
wildlife habitat
Reduced
Decreased dilution of
groundwater
sediment and contami-
return flows to X X X
nant concentrations in
surface water
surface water
Increased river/lake water
X
temperatures
Table 4.1 continued.

Excess water Waterlogging of
X X X X
application agricultural land
Water table
with
rise Salination of agricultural
inadequate X X X
or native lands
drainage
Topsoil loss Loss of soil productivity X X
River/lake/wetland
X X
sedimentation
Sediment in
Irrigation facility
irrigation run- X
sedimentation
off water
Degraded fish and
X X
wildlife habitat
Nutrients in
River/lake/estuary
irrigation run-
excessive plant growth X X X
Soil off (primarily
and eutrophication
erosion phosphorus)
Crop pests in Spread of undesirable
irrigation run- pests (weeds, diseases, X X X
off (organic nematodes)
residues, weed
Degraded fish and
seeds, nema- X X
wildlife habitat
todes, diseases)
Pesticides
River/lake water
in irrigation X X
contamination
runoff
Decreased soil fertility X
Leaching of
Groundwater
nutrients X X
contamination
(primarily
nitrogen) Surface water nutrient
X X
enrichment
Increased groundwater
Water Leaching of
and surface water X X
percolation salts
salinity
through the
soil to Leaching of
groundwater potentially Contamination of
harmful groundwater and surface X X X
elements from water
soil profile
Pesticide contamination
Leaching of
of groundwater and X X X
pesticides
surface water
Table 4.2. Possible positive impacts of irrigation.

Impact
Natural
Agricultural Envi- Human Human
Process Effect Possible Consequences Productivity ronment Health Aesthetics
Increased recreation X
Creation of Increased plant and
Water X X
reservoirs animal habitat
storage
and Improved flood control X X
and
water Hydropower generation X X X
conveyance
channels More dependable
X X X
water supply
Increased quantity, quality,
and dependability of food X X X
Crop and fiber production
Irrigation
production Increased employment and
economic X X
development
4.2 WATER STORAGE, DIVERSION, AND CONSUMPTION

Irrigated agriculture requires large amounts of water. Most irrigated agriculture is
located in areas where normal water supplies during the growing season are low.
Therefore, water is collected from large areas, stored in reservoirs until required, and
diverted to irrigated lands where much of it is consumed. The surface water supply is
often augmented by pumping from groundwater aquifers that have accumulated and
reached equilibrium conditions over hundreds of years. The diverted or pumped water
that is not consumed return flows into drains, streams, lakes, and marshes or perco-
lates to groundwater aquifers. These processes disrupt the natural hydrology, and thus
change the environment associated with that hydrology. Although these processes are
often external to the on-farm irrigation process, they are driven by irrigation water
needs and are a consequence of irrigated agriculture.
4.2.1 Impacts of Water Storage on Rivers and Streams
Irrigation water supply development from surface water resources frequently re-
quires placement of structures in the stream (Figure 4.1). The purpose, size, and use of
the structure as well as the natural hydrology and biology of the stream determine the
environmental impacts that will occur both downstream and upstream of the structure.
On-stream reservoirs create large pools of water that react with the environment
differently than flowing streams. The changes can be predicted with adequate water
quality, meteorological, and other environmental data. Depending on one’s point of
view, changes can be interpreted as beneficial or adverse. One thing is certain:
changes will occur. It is important to describe the expected changes and what can be
done to prevent or minimize the effects, so the project planners and the interested pub-
lic can determine the significance of a change and possible mediation of the effects.
Under normal operation, a reservoir on a stream stores water during periods of high
or excess flows and releases it during periods of low or deficit flows. This process
changes the high and low downstream flows that existed in the past. The reduced high
Figure 4.1. Morrow Point Dam on

the Gunnison River in Colorado.
flows affect riparian habitat that depends on over-bank flooding. The flows may not be
sufficient to keep the channel gravel clean enough for spawning or to control aquatic
growth. The reduced flow variation can impact channel size and result in reduced
channel capacity downstream of the reservoir. In extreme cases (large reservoirs on
small drainages), the downstream channel may essentially disappear.
Dams trap much of the sediment that is carried under natural flow conditions. This
can change the geomorphic stability of the downstream channel and cause increased
erosion and channel instability in areas with geologic materials that are easily eroded.
Reduced sediment in the reservoir discharge will degrade the stream channel if there is
a significant amount of erodible material in the channel. This can impact natural habi-
tat, recreation, and other stream uses downstream of the dam. Likewise, reservoir wa-
ter diverted into irrigation canals can scour sediment out of canal cross-sections result-
ing in increased seepage losses. Surface irrigators with field lengths designed for infil-
tration rates reduced by sediment-laden water may find that they can no longer get the
water to the end of the row or field.
As sediment accumulates in the reservoir, nutrients and contaminants that are asso-
ciated with the sediment may become available to aquatic life in the reservoir. Con-
taminants can accumulate and concentrate in the aquatic food chain and become harm-
ful or even toxic. Actions may be needed to prevent accumulation of harmful con-
taminants.
Reservoirs also impact water temperature and dissolved oxygen. The changes de-
pend on depth of water, organic loading, and eutrophic condition. In temperate, sea-
sonal climates such as the western U.S., impounded water will stratify during summer
months if the depth is greater than 5 m. Rising spring and summer air temperatures
cause the reservoir surface water temperatures to increase faster than heat can be
transmitted to the reservoir depths or consumed in evaporation, and a temperature pro-
file is created. The stratification depends on several factors such as wind mixing, res-
ervoir detention time, solar radiation, and temperature variation. The temperature
stratification prevents the warm, less dense, surface water layer from mixing with the
deeper, colder layer. In some instances this can also cause oxygen deficiencies at depth.
Usually, oxygen deficiency in western U.S. irrigation reservoirs has not been a problem.
Reservoir outlets that draw water from near the bottom of the dam release colder
water in the summer and sometimes warmer water during the cold winter months than
would naturally flow in the stream. This temperature change may adversely impact
native biota and, in extreme cases, result in population decline or even extinction of
sensitive species in downstream reaches. Temperature change may also favor some
species, including trout. In some areas below dams where the stream conditions are
changed, excellent trout fisheries have developed; for example, the Colorado River
below Glen Canyon dam and the San Juan River below Navajo reservoir. Where water
temperatures are too cold for optimum fish production, selective withdrawal outlets
can be provided to blend water from strata of various temperatures to provide the best
downstream temperature for river management goals, such as maintaining the native
fishery or providing for maximum trout production. This was done at Flaming Gorge
reservoir on the Green River.
Dams and associated reservoirs submerge free-flowing streams and convert them to
pools of water. For recreationists who prefer rafting or fishing in free-flowing streams,
this is a significant negative impact. However, for those who prefer lake-like condi-
tions, the reservoir pool is a benefit. Some reservoirs become good fisheries and pro-
vide additional recreational benefits over the prior stream opportunities. Net recrea-
tional use in an impounded river reach has usually increased after reservoir construction.
Dams and reservoirs create fish migration barriers and adversely impact native
fisheries that are blocked from their spawning areas. Much of the decline of anadro-
mous fish (salmon) in the northwest U.S. is the result of dams blocking native spawn-
ing areas and reservoir pools slowing smolt migration to the sea. Fish ladders or other
facilities to assist fish passage can reduce impacts. Water diversions from stream or
reservoirs into canals or other conveyances may remove fish from the stream along
with the water. Fish screens or migration barriers should be constructed at points of
diversion to minimize this problem. Water diversions into large, unlined irrigation
channels sometimes create fisheries in the channels, as is the case for the Henry’s Fork
of the Snake River near St. Anthony, Idaho.
4.2.2 Impacts of Surface Water Diversions
About 110 km3 of surface water was diverted for irrigation in the U.S. in 2000
(Hutson et al., 2004). Irrigation diversions change the stream environment. These
changes usually include reduced flow and increased water temperature; they some-
times increase pollutant concentrations, and they often adversely impact the existing
aquatic habitat and stream ecosystem. The level of impact is a function of stream size
and portion of flow diverted.
Most irrigation diversions occur only during the crop growing season, usually late
April to early October in much of the western U.S. The flow is left in the channel or
stored in reservoirs during the non-growing period. The larger the percent of flow re-
moved from the stream, the greater the impact. Flow removal reduces aquatic habitat
and can make the stream less productive. The ultimate impact occurs when the diver-
sion removes all flow from the stream, either on a permanent basis or for short periods
of time, which usually ends most aquatic production in the dry reach. This can se-
verely reduce fishing and other water-based recreation. The FAO estimates that
2350 km3 of global annual runoff (6% of natural runoff) is needed for instream flow to
maintain healthy ecosystems (FAO, 1996). This is roughly equivalent to global water
currently consumed by irrigated agriculture.
Another potential impact of diversions is increase in stream temperature below the

diversion. This occurs during summer months when the remaining flow spreads in a
channel that was developed by the full flow. Usually the heat exchange opportunities
are increased by exposing more rocks and channel bottom, which absorb radiation and
then conduct the heat to the remaining flow. Frequently there is less shading of the
remaining flow, due to not filling the channel, which allows the remaining flow to
absorb proportionally more radiant energy. If stream temperatures are sufficiently ele-
vated, the stream may not support the original biota. It may change from a cold-water
fishery to a warm-water one. Increased temperatures during critical periods can also
block migration of anadromous fish in their spawning journey.
Streams dilute and have a capacity to assimilate pollutants discharged into the flow
from point and nonpoint sources. The ability to flush pollutants out of stream sections
is reduced when water is removed by diversions. This increases the adverse effects of
pollutants that enter the stream downstream of the diversion. Many of the world’s ur-
ban areas are located at the mouths of major river systems. These cities have depended
on river flows both for their water supply and to flush pollutants they generate out into
the ocean. When irrigation diversions reduce river flows, their water supply may be
degraded and their pollutants accumulate, degrading the aquatic ecosystem, increasing
health risks, and degrading the recreational opportunities and aesthetics of the river. In
some cases, the reduced flows may allow ocean saltwater to intrude further into river
estuaries. These impacts on urban populations often result in major conflicts between
urban and agricultural interests.
4.2.3 Impacts of Irrigation Pumping on Groundwater Aquifers
About 79 km3 of water was pumped from groundwater aquifers in the U.S. in 2000
for irrigation (68% of the total pumped) (Hutson et al., 2004). Irrigation pumping can
have several impacts on the aquifer sources including water table decline, groundwater
depletion, return flow reductions to surface streams, land consolidation, and saltwater
intrusion.
When groundwater pumping exceeds recharge, the water table level, or piezometric
head, declines. Although water table levels fluctuate with seasonal and annual varia-
tions in pumping and recharge, the decline will be continual if the pumping exceeds
long-term aquifer recharge. Excessive pumping is common in arid and semi-arid areas
unless the number and discharge of pumps are regulated to prevent groundwater de-
pletion. More than 4 million hectares of U.S. land are irrigated from aquifers in which
pumping exceeds recharge and water tables are declining (Postel, 1993).
Water table decline increases pumping lift and therefore pumping costs. A one me-
ter increase in the pumping depth in the U.S. requires over 1 billion MJ (300 million
kW h) of extra energy each year for pumping irrigation water. If the decline is signifi-
cant, wells may have to be deepened or replaced. With continued excess pumping,
portions of the aquifer can be completely dewatered and the groundwater resource
lost. In many agricultural areas, the aquifers supply not only irrigation but also domes-
tic and municipal water needs. Pumping amounts should be designed and regulated to
sustain the yield of the aquifer at a reasonable water table level, although sustainable
aquifer yield is often difficult to determine.
Groundwater pumping can reduce return flows from aquifers to local streams,
springs, and seeps. Water table decline can dry marsh areas and areas that receive
subirrigation from the water table. This can have serious environmental effects, espe-
cially when limited natural habitat, such as wetlands in arid areas, is destroyed. Re-
duced groundwater contributions to local streams can be critical because groundwater
sources often provide base flows during periods of low surface runoff. Groundwater
usually has a cool, constant temperature, so reduced aquifer contributions in summer
result in higher stream temperatures. Proper groundwater investigations are required to
identify expected impacts and problems and make informed decisions on the devel-
opment potential.
Aquifer depletion can cause land subsidence. Subsidence is often a greater problem
in confined than in unconfined aquifers. Non-uniform subsidence can seriously dam-
age surface structures. Subsidence of confined aquifers also damages the groundwater
resource by decreasing the potential storage capacity if the aquifer re-saturates.
Water table decline along coastal areas may allow saltwater intrusion into ground-
water aquifers. When water table levels are reduced sufficiently that the normal gradi-
ent toward the ocean reverses, saltwater moves inland making the groundwater unfit
for most uses.
4.2.4 Return Flow from Irrigated Agriculture
Irrigation is not 100% efficient, so return flow is produced. The return flow can be
surface runoff or percolation to groundwater. In some instances drains are constructed
to intercept shallow groundwater and return it to surface flows. These drainage and
surface return flows are routed through constructed drains to natural channels. Canal
operational spillage is also usually diverted to natural channels. Return flows can have
positive or negative effects on the environment.
Drainage water from irrigated lands can create marshes and wetlands along drain-
age flow paths. These wetlands can create critical habitat for wildlife in otherwise arid
environments. However, these created wetlands can also create problems for the local
residents. If wetlands are desired with the irrigation development, they should be
planned and engineered to get maximum benefits with minimum adverse effects. The
wetlands can be designed for wildlife habitat, and to remove nutrients and sediments
while minimizing water consumption.
When return or spillage flows are put in natural channels, the additional flows can
change the hydraulic equilibrium and may result in channel erosion. Sediment in sur-
face return flows can change the stream biota and have serious effects on downstream
water users. An example of return flows causing problems is the Greenfields Division
of the Sun River Project in Montana. Much of the project drainage enters Muddy
Creek, a stream that flows south along the east end of the Greenfields irrigated area,
and discharges into the Sun River. Prior to irrigation development, Muddy Creek had
minimal flows during the summer and late fall. The additional flows contributed from
the irrigation project drainage increased Muddy Creek flows and caused serious chan-
nel erosion. Sediment from Muddy Creek has changed the Sun River fishery from fair
above the confluence to poor below.
Irrigation surface return flows may increase temperatures in receiving streams,
which can cause problems for aquatic biota and water users. The combined impacts of
reduced cool groundwater return flows (see above) and warm surface return flows can
change the stream from a cold- to a warm-water environment.
A benefit of irrigation water that percolates to groundwater can be an increase of
base flows from groundwater to streams. These return flows increase stream flows
during drought and low flow periods. The cool groundwater also reduces stream water
temperatures during summer low flow periods. Both the increased flows and reduced
summer temperatures can improve the environment for fish and other biota. The South
Platte River in northeastern Colorado is a stream and riparian area that has benefited
from flow augmentation during late summer and early fall from irrigation water im-
ports, storage, and return flows.
4.2.5 Identifying and Quantifying Impacts
Prior to water diversion, storage, conveyance, or drainage project formation and de-
sign, all management goals and adverse impacts need to be identified and discussed
with the appropriate state and federal agencies and the public. Both the potential bene-
fits and negative impacts need to be carefully considered and evaluated. The National
Environmental Protection Act (NEPA) environmental impact analysis required of all
such U.S. developments formalizes that process.
Potential impacts of irrigation project development can be identified by evaluating
the impacts of projects with similar conditions. Since conditions are seldom identical,
computer models are often used to predict and evaluate expected changes. The sim-
plest approach would be to develop a spreadsheet mass balance model. More complex
models include CALSIM (CDWR, 2000), MODSIM (Labadie et al., 2000), and
CEQualW2 (Cole and Wells, 2002). The latter offers water quality options and two-
dimensional reservoir representation. Accurate model use requires thorough under-
standing of the physical and biological systems associated with the water resource to
be modeled. Proper use of models also requires sufficient input data and an under-
standing of model’s limitations. More complex models usually require more detailed
data. The models must describe both surface and groundwater resources from a quan-
tity and a quality standpoint. Results then must be related to the biological and envi-
ronmental resources to determine probable project impacts.
When an irrigation project is built, a monitoring plan should be implemented to as-
sess impacts. The monitoring plan should be designed to identify and quantify changes
that indicate potential problems or problems in areas where the project investigations
were inconclusive or did not provide a high level of confidence. This will allow pre-
ventive measures or mitigation to be implemented at the earliest possible date.
4.2.6 Strategies for Reduction or Control of Negative Impacts
Irrigation has impacts. During new project development, these impacts must be
evaluated and weighed against project benefits in the decision-making process. Ac-
tions can be implemented to reduce, partially control, or mitigate the impacts caused
by irrigation project development. Impacts that must be prevented or minimized need
to be brought into the project design process. If fish passage needs to be provided, then
instream structures must have features that will provide sufficient passage. If water
quality will be negatively impacted, then steps must be taken to correct, prevent, or
minimize the impacts. A thorough investigation of adverse impacts and ways to pre-
vent or minimize them must be included in the planning and design of irrigation pro-
jects and facilities. The National Environmental Protection Act (NEPA) process (see
above) incorporates reduction and mitigation of negative impacts.
4.3 GROUNDWATER QUALITY
4.3.1 Impacts of Irrigation on Groundwater Quality
When irrigation water application exceeds soil water storage capacity, excess infil-
trated water percolates below the root zone and eventually reaches the groundwater.
Over-application of irrigation water, non-uniform irrigation water distribution, non-

uniform soils, salt leaching requirements, and rainfall can all result in water applica-
tion in excess of the amount that can be stored in the crop root zone and thus produce
deep percolation events. Runoff from areas of a field can collect in depressions and
cause deep percolation (Derby and Knighton, 2001; Spalding et al., 2003). To support
the high yields that irrigation can provide, agricultural chemical use (nutrients and
pesticides) is often high. Some of these chemicals, and other substances that may occur
naturally in arid soils, can move with downward percolating water to the groundwater.
Among U.S. states, tribes, and territories reporting groundwater data to the U.S.
Environmental Protection Agency in 2000, fertilizer applications and irrigation prac-
tices were listed among the top ten sources of groundwater contamination by 23 and 6
of the states, respectively (USEPA, 2002). Reviews of the impacts of irrigated agricul-
ture on groundwater in the U.S. have been compiled by various authors, e.g., Bouwer
(1987), Helweg (1989), NRC (1989), and Ritter (1989). A review of the problems in
the arid western states was compiled for California (Schmidt and Sherman, 1987);
Arizona and New Mexico (Sabol et al., 1987); Oklahoma and Texas (Law, 1987); and
Idaho, Montana, Nevada, Oregon, Utah, and Washington (Sonnen et al., 1987). In the
humid regions of the U.S., the impacts of irrigation on groundwater were reviewed for
the Corn Belt and Lake States (Mossbarger and Yost, 1989); South (Shirmohammadi
and Knisel, 1989); and East (Ritter et al., 1989b; Krider, 1986).
The most common groundwater contamination problems associated with irrigation
are nitrate-nitrogen (NO3-N) and pesticides (Hunter, 1986), and these affect people in
various direct and indirect ways. Many people, especially those in rural areas, rely on
groundwater for their source of drinking water, and pollution of this resource directly
affects human health. The U.S. maximum contaminant level (MCL) or drinking water
standard for NO3-N is 10 mg L-1. Elevated levels of NO3-N can cause methemoglo-
binemia in babies (Ritter, 1989b; Dahab and Sirigina, 1994; Skipton and Hay, 1998).
Where groundwater reaches surface waters, pollutants in the groundwater may affect
the surface water’s usefulness for recreation, fishing, wildlife habitat, and associated
tourism industries. Where groundwater is used for aquaculture, pollutants may de-
crease productivity and even render the water unfit for fish production.
Nitrate-nitrogen is the crop nutrient most commonly leached from the root zone.
Although NO3-N leaching occurs widely in rainfed agriculture, some of the most se-
vere cases of NO3-N leaching to groundwater, both in the arid western states and in
more humid regions of the U.S., have been associated with irrigated agriculture (Hall-
berg, 1986; Hamilton and Helsel, 1995). One difficulty in controlling NO3-N leaching
is that plant N uptake is often inefficient compared with applied fertilizer amounts. For
example, less than half of fertilizer N applied to corn is typically removed in the grain,
and one-third to one-half of the fertilizer N may leave the profile via deep percolation
(Hallberg, 1986). On the other hand, leaching problems may be reduced in many situa-
tions with irrigation because of water control and management possible with irrigation
(Hallberg, 1989). Irrigation may reduce leaching by promoting increased N uptake by
the crop, thereby reducing the amount of N available for leaching during subsequent
off-season rainfall events (Juergens-Gschwind, 1989). Other chemicals used as fertil-
izers that may be leached to groundwater include chloride and potassium from potash,
and calcium and magnesium from lime (Hamilton and Helsel, 1995). These elements
generally do not cause water quality problems.
Pesticides include herbicides, insecticides, fungicides, soil fumigants, and nemati-

cides. Hallberg (1986) reported that generally less than 5% of applied agricultural pes-
ticides reach groundwater. Flury (1996) reported that, depending on chemical proper-
ties, as much as 5% of the applied pesticide may leach beyond the root zone in worst-
case rainfall events (i.e., heavy rainfall immediately following pesticide application).
When heavy rainfall does not immediately follow application, Flury reported that an-
nual losses are typically less than 0.1% to 1% of the applied pesticide, but may reach
as high as 4%.
Pesticides that have been found below the root zone or in groundwater underneath
irrigated areas include alachlor, atrazine and its metabolites desethylatrazine and de-
sisopropylatrazine, carbofuran, metolachlor, metribuzin (Burgard et al., 1993; Goes-
selin et al., 1997; Ma and Spalding, 1997b; Ellerbroek et al., 1998; Kraft et al., 1999;
Batista et al., 2002; Spalding et al., 2003), benomyl (Merwin et al., 1996), bromacil,
2,4-DP, diazinon, dibromochloropropane (DBCP), 1,2-dibromoethane, dicamba, 1.2-
dichloropropane, diuron, prometon, prometryn, propazine, simazine (Domagalski and
Dubrovsky, 1992; Walker and Ross, 1999; CDPR, 2000), 3,4-dichloroaniline, di-
methoate, alpha- and beta-endosulfan, lindane, molinate, prometryn (Batista et al.,
2002), cyanazine (Ritter et al., 1989a), and propalchlor (Spalding et al., 1989).
The importance of pesticide problems in groundwater may not yet be fully apparent
because some pesticides that have leached below crop root zones are still in transport
though the unsaturated zone between the root zone and the groundwater. In an irri-
gated farming area in eastern Washington, Jones and Roberts (1999) compared pesti-
cide occurrences in monitoring wells, designed to sample only the groundwater near-
est the surface, and domestic wells. They attributed the variations in groundwater
quality to well depths, groundwater age, and greater effects of sorption, dispersion,
dilution, and degradation corresponding to longer residence times, and found that over
60% of the pesticides appearing only in domestic wells were compounds no longer
used.
In arid and semiarid areas where irrigation is common, lack of rainfall over long
time periods allow soluble substances such as salts and several trace elements to ac-
cumulate and remain in the soil profile. The elements may have been present in the
original material, deposited by some past geologic event such as deposition on ancient
lake beds, or brought in with the irrigation water. In areas of high rainfall, these ele-
ments would have gradually leached out of the profile, drained to rivers, and eventu-
ally reached the ocean.
Some soils or the geologic formations underlying irrigated areas contain trace ele-
ments such as boron, selenium, cadmium, molybdenum, and arsenic that can cause
water quality problems. When subjected to water percolation from irrigation, these
elements gradually dissolve into the water and may move to shallow groundwater and
drain to lakes and marshes, where they accumulate and concentrate through evapora-
tion and can become toxic to plants or animals. Examples include the leaching of sele-
nium from shale formations underlying irrigated areas in Wyoming (Smith, 1994) and
boron and selenium from the west side of the San Joaquin valley in California (Presser
and Ohlendorf, 1987; San Joaquin Valley Drainage Program, 1990; Frankenberger and
Engberg, 1998).
Many arid and semi-arid soils and all irrigation water contains dissolved salts.
When irrigation water containing salts or water from a shallow water table or aquifer
is transpired by plants or evaporated from the soil surface, salts are left behind in the
soil. High concentrations of dissolved salts are harmful to plants and must be removed
by leaching to maintain crop production. Salt tolerance of various crops and salinity
management techniques have been extensively documented (e.g., Rhoades and Love-
day, 1990; Chapter 7 of this monograph). Leaching moves the salts and any other
soluble elements or chemicals to the groundwater. High salt concentrations in the
groundwater make it unsuitable for domestic use and limit its use for irrigation. Al-
though groundwater in some semi-arid areas was salty before agricultural develop-
ment, irrigation often compounds the problem by importing water containing salts and
by leaching salts previously stored in the soil profile into the groundwater. In many
areas, the greatest impact of irrigation is to increase deep percolation and raise water
table levels so that return flows of salty groundwater to surface waters are increased.
Where groundwater is shallow, drains are sometimes installed to lower the water table
and prevent accumulation of salts. This drainwater often has elevated salt concentra-
tions.
4.3.2 Percolation and Leaching Processes
Percolation of water through the soil has the potential to leach soluble chemicals
from surface and sub-surface soil to the groundwater. Percolation of water through and
beyond the root zone of crops occurs by flow through the porous matrix of the soil and
through preferential flow pathways. Flow through the porous matrix is described by
Darcy’s law, which states that flow is proportional to the hydraulic conductivity of the
soil and the hydraulic gradient. Water and chemical leaching potential generally in-
crease directly with soil hydraulic conductivity and inversely with water holding ca-
pacity. High hydraulic conductivity permits rapid downward movement and provides
little time for chemical uptake or transformation in the unsaturated zone. Soils with
low water holding capacity provide less room for error in irrigation scheduling and for
storage of unexpected rainfall events, and thus increase the risk of deep percolation loss.
Soils with low water holding capacity also often have high hydraulic conductivity.
Flow through preferential pathways includes water movement through soil cracks,
root channels, earthworm holes, and channels formed by other biological activities.
The resulting macropore flow of water through soil resulting from these factors pro-
vides a short-circuiting effect by which water and chemicals at the soil surface bypass
much of the soil matrix and reach groundwater faster than by Darcian flow through the
soil matrix. Saturated or near-saturated surface conditions are necessary for water to
move through macropores. Nieber (2001) presented an overview of preferential flow
mechanisms—including macropore flow, gravity-driven unstable flow, heterogeneity-
driven flow, oscillatory flow, and depression-focused recharge—and the reader is re-
ferred to his paper for a review of research on these preferential flow mechanisms and
their impact on water quality.
Surface irrigation is not the only type of irrigation that can cause preferential flow,
because saturated soil surfaces may occur with sprinkler or even drip irrigation when
water application rate exceeds infiltration. Ponding may occur in topographic depres-
sions that collect runoff from adjacent areas. This concentrated recharge may signifi-
cantly influence chemical transport to groundwater. In addition to preferential flow,
description of water flow and chemical transport in the root zone of irrigated agricul-
tural systems is complicated by nonuniformity in space and time of soil physical and
chemical properties, as well as sources and sinks within the root zone.
Chemical, physical, and biological processes affect chemical fate and transport in
porous media (Toride et al., 1993). Processes affecting the state of chemical occur-
rence in soil include adsorption, dispersion, dilution, and degradation (Jones and Rob-
erts, 1999), volatilization, amounts and timing of chemical applications, soil texture,
soil organic carbon or organic matter content, irrigation method, groundwater depth
(Hallberg, 1986; Domagalski and Dubrovsky, 1992), chemical solubility (Elliott et al.,
2000), composition of ground cover, and photochemical degradation (Wilson et al.,
1995). Allen et al. (1985) developed a numerical index (termed DRASTIC) of
groundwater vulnerability to pollution based on the following seven factors: “Depth to
water, net Recharge, Aquifer media, Soil media, Topography (slope), Impact on the
unsaturated zone, and hydraulic Conductivity of the aquifer” (Knox and Moody,
1991). Degradation rates may be quite slow for some chemicals, e.g., DBCP (Walker
and Ross, 1999), causing groundwater quality problems to persist for long time peri-
ods after chemical use has been discontinued or banned. Biological factors influencing
chemical fate and transport in the soil include crop and pest uptake, immobilization on
the organic matter in the soil, and microbiological transformations.
Models can be used to study the impacts of irrigation on groundwater (Knisel and
Leonard, 1989; Watts et al., 199). Various authors (e.g., van Genuchten and Alves,
1982; Leij et al., 1991; Leij et al., 1993) have formulated mathematical models that
include one or more of the above transformation and transport processes under differ-
ent boundary and initial conditions. Many numerical models have been developed to
describe water flow and chemical fate and transport in the unsaturated zone, including
RZWQM (Ahuja et al., 2000), UNSATCHEM (Suarez and Simunek, 1997), HY-
DRUS (Kool and van Genuchten, 1992), DRAINMOD-N (Breve et al., 1992),
TETrans (Corwin et al., 1991; Corwin and Waggoner, 1991), NLEAP (Shaffer et al.,
1991), LEACHM (Wagenet and Hutson, 1987), GLEAMS (Leonard et al., 1987),
MOUSE (Steenhuis et al., 1987), PRZM (Carsel et al., 1985), and BAM (Jury et al.,
1983).
The persistence and fate of chemicals leached to groundwater depends on the na-
ture and use of the aquifer. Chemicals are generally diluted as they reach groundwater.
Chemical and biological processes may degrade or transform the chemicals in the
groundwater. Natural and pumping-induced groundwater gradients mix and transport
chemicals within an aquifer. Subsurface formations with zero or low hydraulic con-
ductivity may effectively isolate contaminated aquifers from deeper aquifers that may
be used as drinking water sources. Groundwater carrying pesticides, salts, and metals
may return to surface waters such as streams and rivers. This return flow to surface
waters may result from natural hydraulic gradients in the groundwater system or it
may be due to discharge from manmade drainage systems installed to improve soil
aeration and to control salinity.
4.3.3 Evaluating Groundwater Problems
Chemical movement from the root zone to groundwater can be estimated from wa-
ter percolation rate and chemical concentration in the water. Quantification of down-
ward chemical movement is difficult because of chemical transformations, as well as
spatial and temporal variability in water movement and chemical concentrations.
Methods to measure or estimate water percolation to groundwater may be classi-
fied as flux methods or water balance methods (Wagenet, 1986). Flux methods use
Darcy’s law to measure or estimate deep percolation (Hillel, 1982). Water balance
methods estimate leaching by applying the principle of conservation of mass to a con-

trol volume in the unsaturated zone. The sum of inflows and outflows must equal the
change in storage in the control volume. In this case, inflows include precipitation,
irrigation, and surface and subsurface inflows (e.g., due to surface runoff reentering
the profile, tile drainage, or lateral subsurface flow under sloping topography); out-
flows include evapotranspiration (ET), surface and subsurface outflows, and leaching;
and change in storage may be determined by measuring changes in soil water content.
The percolation quantity can be calculated when other variables in the water balance
are known or can be assumed negligible.
Percolating water can be sampled for chemical constituents by extracting soil water
samples with vacuum sampling techniques (Rhodes and Oster, 1986), or by collecting
soil samples and extracting the water. Monitoring or other wells are commonly used to
sample groundwater for chemical constituents. Sampling may be done in situ using
sensors to assess parameters such as electrical conductivity, or by removing a water
sample for subsequent laboratory analysis for constituents such as nitrogen, pesticides,
and metals. Because it is difficult to estimate mixing and dilution rates in groundwater,
samples at several depths may be required. In situations where groundwater returns to
surface waters, the quality may be determined using in situ or grab sampling tech-
niques. Biological indicators in surface waters affected by irrigation return flows, such
as plant and animal populations, algae growth, or metal accumulations in fish and
wildlife, may support the overall assessment of return flow impacts. Hornsby (1990)
reviewed public health and pollution problems associated with irrigated agriculture.
For highly soluble and mobile nutrients such as nitrate-nitrogen, leaching may in
some cases be inferred from signs of visible crop deficiency such as yellowing of
leaves or from plant nutrient tests. Nitrogen leaching may also be estimated from soil
sampling and accounting for the amounts and timing of fertilizer applications and crop
uptake. A difficulty in this approach is that nitrogen mineralization rates may be diffi-
cult to determine.
The economics of irrigated agriculture can increase the potential for groundwater
pollution problems. When water or chemicals are inexpensive compared to the eco-
nomic returns they can create, overuse is more likely. Pollution potential is greatest
when high-value crops are grown and water and chemicals are relatively inexpensive.
Farmers and managers sometimes perceive that liberal application of water and
chemicals is inexpensive insurance for maximum profit.
Leaching of chemicals to groundwater can be viewed as either positive or negative,
depending on one’s perspective. Leaching of salts from the root zone is necessary to
maintain acceptable crop growth and health and provide for sustained irrigated crop
productivity. Leaching of nutrients and pesticides result in loss of valuable resources
and reduced availability for crop growth and pest control. Leaching of salts, nutrients,
and pesticides to groundwater may adversely affect groundwater resources.
4.3.4 Strategies for Reduction and Mitigation of Leaching
Leaching of chemicals to groundwater from irrigated fields can be reduced by de-
creasing water percolation amounts, by decreasing the amount of chemical available
for transport, or both. Tools available to achieve these reductions include the selection,
design, and management of irrigation systems, as well as the chemical management
and agricultural practices used in crop production. When irrigation and crop manage-
ment do not perform as desired, groundwater remediation may be necessary.
4.3.4.1 Irrigation system selection, design, and management. Irrigation system

design or management changes that increase water distribution uniformity potentially
decrease deep percolation losses and chemical leaching. Good irrigation scheduling
(Chapter 8) reduces chemical leaching by avoiding excessive water applications and
reducing deep percolation. Proper irrigation scheduling also helps ensure healthy
plants and maximum nutrient uptake, leaving smaller nutrient amounts—particularly
nitrogen—available for leaching. Where groundwater contamination is a concern, in-
tentional deficit irrigation (not refilling the soil profile to field capacity with each irri-
gation event) can reduce deep percolation, often with only slight yield reductions.
Surface, subsurface, micro-, and sprinkler irrigation systems have different charac-
teristics relative to water percolation and chemical leaching. Systems with the ability
to apply small, uniform applications have the potential to decrease deep percolation
losses. In general, systems that use pressure distribute water more precisely and with
better control than those that use gravity for distribution across the soil surface. Thus,
well-engineered drip irrigation systems have higher potential irrigation efficiencies
than sprinkler systems, and sprinkler systems are potentially better than surface sys-
tems. Generally speaking, smaller fields or irrigation zones can be irrigated more uni-
formly than larger ones.
Irrigation systems also differ in the fraction of the soil surface that is wetted and the
rate of application. Partially wetting the soil surface may allow purposely placing
chemicals to avoid downward movement with the irrigation water. Surface irrigation
always results in saturated flow at the soil surface and hence has greater potential for
preferential flow. Drip and sprinkler irrigation may not result in saturated flow condi-
tions if application rates are lower than the infiltration rate of the soil.
Subsurface irrigation systems use water table control to meet crop water require-
ments. With very high water tables, the potential for chemicals to reach shallow
groundwater is high. However, with subirrigation, since water movement should be
primarily upward through the root zone, downward movement of chemicals during
irrigation should be minimized.
4.3.4.2 Chemical management and agricultural practices. In addition to irriga-
tion management, good agronomic practices can help reduce the amount of nutrients
and chemicals leached to groundwater. These management practices can include re-
duced chemical usage, use of less mobile chemicals, chemical application timing when
leaching is least likely and when usage or transformation is most rapid, and chemical
placement where it is least likely to be leached.
Nutrient applications can be scheduled to meet crop needs through in-season soil
and plant sampling. Nitrogen content of the irrigation water should be measured and
considered to avoid over-fertilization (Smith and Cassel, 1991). Nutrient application
based on realistic potential yields rather than maximum yields may significantly re-
duce nutrient application and leaching with only small reductions in yield. Cropping
systems management may be used to mitigate N leaching. Examples of cropping sys-
tems management include accounting for the N contribution of legumes in subsequent
years; using cover crops such as rye, vetch, and wheat, planted after harvest of other
crops to immobilize residual N in the soil (Herbert et al., 1995; Wilson et al., 1995);
and planning crop sequences and timing to use nitrogen released by the breakdown of
the previous crop.
Precisely scheduled and targeted pesticide applications such as are used in inte-
grated pest management programs can result in reduced pesticide applications. Addi-
tional practices that can reduce chemical availability for leaching include using alter-
nate-furrow irrigation and chemical placement for row crops (Hefner and Tracy,
1995); placing herbicides away from anhydrous ammonia application zones (Clay et
al., 1995) to avoid the high pH regions that reduce herbicide adsorption on the soil and
increase potential for herbicide leaching (Hunter, 1986); using encapsulated, slow-
release, or controlled-release herbicides (Hickman and Vail, 1995; Williams et al.,
1995); using paraffinic oil as a herbicide carrier to increase spreading on contact and
reduce the amount of herbicide needed (Hanks, 1995); and combining banding of her-
bicides and cultivation for weed control (Baker et al., 1995).
Since the initial irrigation often produces the largest deep percolation loss with sur-
face irrigation systems, application of mobile chemicals before the first irrigation
should be avoided when possible. Chemicals are usually applied as evenly as possible
even though the needs may vary spatially. One promising solution to spatial variability
is site-specific technology and management, which divides fields into sub-field man-
agement zones and applies chemicals based on localized needs. Banding nutrients,
placement away from the irrigated furrow or dripper, and using slow-release or less-
mobile forms of nitrogen and nitrification inhibitors can reduce nitrate leaching
(Crumpton and Baker, 1993; Dahab and Sirigina, 1994).
The unpredictable nature of weather can increase the potential for leaching chemi-
cals to groundwater. For example, unexpected rainfall following irrigation can result
in deep percolation. If evaporative demand is significantly smaller than normal in a
given growing season due to cool weather, crop growth may be retarded compared to
that expected for a warmer season and N fertilizer may be underutilized. High N re-
siduals present in soil at the end of the growing season increase the potential for leach-
ing to groundwater due to off-season precipitation. Since weather cannot be predicted
with certainty, flexible chemical and water management are essential to minimize
leaching of chemicals to groundwater.
4.3.4.3 Mitigation of impacts. Groundwater remediation may be necessary when
irrigation and agronomic management practices are not sufficient to keep chemical
leaching losses below acceptable levels, or groundwater concentrations already exceed
acceptable levels. Technologies for groundwater remediation may be grouped into
four categories: containment, collection, treatment, and disposal (Hunter, 1986). Ex-
amples of the technologies are given in Table 4.3. Typical treatment processes employ
one or more of the technologies from each of the four categories in an overall treat-
ment scheme. For example, a treatment scheme could involve containment and collec-
tion of contaminated groundwater via a well, above-ground biological treatment
through irrigation of a crop, and disposal via ET and percolation back to groundwater.
In-situ treatment of groundwater can be accomplished by injecting biological or-
ganisms, chemicals, or both into the contaminated area. For example, Hunter and Fol-
let (1995) proposed injecting vegetable oil as a carbon source into the porous media
around well screens in high-nitrate groundwaters to accelerate biological denitrifica-
tion in situ. Robertson et al. (2000), used various reactive passive carbon barriers to
remediate nitrates in groundwater over a six- to seven-year period. Carbon sources for
groundwater remediation have also been studied by Obenhuber and Lowrance (1991),
Schipper and Vojvodic-Vukovic (1998, 2000), and Robins et al. (2000).
Table 4.3. Technologies for remediation of groundwater. Technologies potentially

applicable to agricultural settings are in italics (from Hunter, 1986).[a]
Category
Containment slurry wall, vibrating beam, sheet piling, concrete wall,
extraction wells, subsurface drains, ditches and trenches
Collection extraction wells only, subsurface drains, ditches and trenches,
extraction wells combined with injection wells
Treatment
In Situ microbial degradation, limestone treatment bed, activated carbon bed,
chemical treatment, neutralization
Biological activated sludge, trickling filter, rotating biological contactor, land
application, aerated lagoon, anaerobic/facultative lagoon
Physical/Chemical ion exchange, membrane separation, oxidation, reduction,
liquid/liquid extraction, carbon adsorption, air stripping, steam
stripping, spray evaporation, wet air oxidation, incineration
Disposal deep well injection, direct discharge, evaporation ponds, percolation
ponds, land application
[a]
Reprinted by permission of the National Ground Water Association. Copyright 1986. All rights reserved.
A variety of other technologies have been studied for groundwater remediation.

The use of zero-valent metals has been studied for the removal of atrazine (Singh et
al., 1998), and for sulfate reduction and denitrification (Chew and Zhang, 1998; Gu et
al., 2002). Electrokinetic methods for nitrate remediation have been studied by Chew
and Zhang (1998) and by Eid et al. (2000). Ma and Spalding (1997a), using artificial
recharge of an aquifer with river water, reported tenfold reductions in nitrate and
atrazine concentrations beneath and downgradient from the infiltration sites. Weier et
al. (1994) found that additions of ethanol to high-nitrate irrigation water may have
potential to remediate groundwater nitrates.
Biological, physical, and chemical treatment may require collecting groundwater
and bringing it above ground for treatment, and then disposing of the treated water.
The technologies for biological treatment are well established in the fields of domestic
and industrial water and wastewater treatment (Hunter, 1986; Dahab and Sirigina,
1994, and references therein). Land application of contaminated water can involve
irrigation to disperse the contaminated water for uptake or removal by plants or soil
organisms. Thus, land application can be both a treatment method and a disposal
method.
4.4 SURFACE WATER RUNOFF
4.4.1 Impacts of Irrigation-Induced Erosion and Sediment Discharge
Water running across the land surface can erode soil. Soil erosion caused by rainfall
is a commonly recognized problem, but irrigation can also cause erosion. Sprinkler
irrigation is similar to rainfall. Irrigation furrows are similar to rills with a controlled
water application. The extent of irrigation-induced erosion is not well documented,
although measurements in Idaho, Wyoming, Washington, and Utah show that it is a
serious problem in some areas of the western U.S. Koluvec et al. (1993) reviewed
Figure 4.2. Aerial views of furrow irrigated fields in southern Idaho showing white soil areas
near the inflow ends where the top soil has been eroded away. (Photo credit: Dave Carter.)
those and other irrigation-induced erosion studies carried out over the last 50 years. A
1985 USDA-Soil Conservation Service survey indicated that about 21% of irrigated
cropland in the U.S. is affected by erosion to some degree (Koluvec et al., 1993).
Soil erosion reduces soil depth and productivity. In southern Idaho, measurements
indicate that 80 years of irrigation-induced erosion of the erodible silt loam soils has
reduced crop yield potentials by 25% (Carter, 1993) (Figure 4.2). For many soils, it is
very difficult to return eroded soil to its pre-erosion production potential.
When irrigation water is applied to the land faster than it is infiltrated, a portion of
the water is redistributed within the field and may run off the field. Twenty to fifty
percent of the water applied to most furrow-irrigated fields with slopes greater than
0.5% runs off the downstream end (Trout and Mackey, 1988). This runoff water car-
ries eroded sediment as well as nutrients, pests, or chemicals that are attached to the
sediment, off the field and into surface drains, rivers, and lakes. Annual sediment run-
off between 4 and 40 Mg/ha is commonly measured from furrow-irrigated fields with
slopes greater than 1% (Koluvek et al., 1993).
Sediment and its adsorbed constituents negatively impact downstream water bodies
and users. It fills drains and downstream reservoirs and irrigation canals. Some irriga-
tion companies spend a large portion of their annual maintenance budget mechanically
removing sediment deposits from reservoirs, drains, and canals (Koluvek et al., 1993).
Often, runoff water becomes the irrigation water supply for downstream farms. Sedi-
ment-laden irrigation water increases maintenance costs of ditches, pipelines, and
ponds; increases sprinkler head and nozzle wear; and may make water treatment costs
for microirrigation prohibitively expensive. Weed seed and other soil-borne pests such
as crop viruses and nematodes can be spread from farm to farm with runoff sediment
(Sojka and Entry, 2000).
Sediment from irrigated fields has degraded many western U.S. rivers (Koluvek et
al., 1993). Irrigation return-flow sediments deposit in rivers and may cover fish
spawning beds and other natural habitats (Figure 4.3). Sediment accumulation is often
severe in low rainfall areas because river flow rates (and thus carrying capacity) are
usually low during the irrigation season in irrigated valleys, and spring flushing flows
may be reduced by upstream storage facilities. Sediments that are transported through
the rivers accumulate in downstream reservoirs, reducing reservoir storage capacity, or
at the river mouths where they may interfere with shipping or recreation facilities.
Agricultural sediments often carry sufficient adsorbed phosphorus to promote plant
growth in rivers and lakes (Westermann et al., 2001). The sediments may also carry
trace amounts of persistent agricultural chemicals such as pyrethroids and organo-
phosphates that can accumulate in river and lake beds, and reduce populations of
aquatic macroinvertebrates (Anderson et al., 2003).
Figure 4.3. Sediment deposition in the

Snake River in southern Idaho from
irrigation tailwater return flow.
(Photo credit: Dave Carter.)
4.4.2 Runoff and Erosion Processes

Soil erosion occurs when soil particles are detached, either by the force of water
droplet impact on the surface, or by the shear of water flowing over the surface, and
are then transported by flowing water to another location. Detachment depends on
both soil properties and condition (erodibility), and on the erosiveness of the water
drops or flows. Sediment transport depends on sediment particle sizes and densities,
and transport capacity of the flow. Trout and Neibling (1993) discuss irrigation-
induced erosion processes in detail.
Irrigation-induced erosion is most common with furrow irrigation, in which flows
are concentrated in small channels and directed downslope. Where closely spaced
crops are grown and flows are spread uniformly across the surface (border irrigation),
or where field slopes are very small (basin irrigation), little erosion occurs. When
sprinkler irrigation water is applied at rates high enough to produce water runoff, ero-
sion can also occur. Runoff is most common with low-pressure center pivot irrigation
where instantaneous water application rates are high at the outer end of the lateral.
Furrow irrigation is similar to concentrated flow in rills. The flow erosiveness in-
creases with the shear exerted by water on the soil, which in turn increases flow rate
and furrow slope. Doubling the flow rate more than doubles erosion. Doubling the
slope more than triples erosion (Kemper et al., 1985). Furrow flow rates decrease in
the downstream direction as water is infiltrated. Thus, the erosion rates are highest at
the inflow ends of furrows, and a portion of the eroded soil may be deposited in the
downstream half of the furrow length (assuming uniform furrow slope). The portion of
the eroded sediment that discharges from the field depends on the portion of the flow
that runs off the field. In southern Idaho furrows, most erosion is from the upstream
one-third of the furrow length, and about two-thirds of the soil is deposited in the
downstream half of the furrow length (Trout, 1996), so only about one-third of the
eroded soil leaves the field.
Sprinkler irrigation erosion is similar to rainfall-induced erosion. Water drops de-
tach soil particles and shallow overland flow transports them to rills where concen-
trated flow carries the particles down-slope. With center pivot irrigation, only a small
portion of the field is being irrigated at any one time, so runoff is often infiltrated
within the field and concentrated flow channels are often not long. Sediment discharge
from sprinkler irrigated fields is seldom a problem.
The discharge of agricultural chemicals with surface irrigation runoff depends on
the source and nature of the materials. Soluble materials in the irrigation water, such
as salts and nitrate, will discharge with runoff water. However, soluble substances that
are in the soil generally move into the profile with infiltrating water and are not carried
with surface flows or eroded surface soils. Thus nitrogen and salt concentrations in
runoff are generally not higher than in the applied water. An exception is possible with
salt accumulation from surface evaporation at the soil surface and flow through cracks
that intermingles with surface flow (Rhoades et al., 1997). Substances that are ad-
sorbed to surface soil particles or precipitated on the soil surface will be transported
with eroding soil particles (Bjorneberg et al., 2002). Total phosphorus in runoff water
with sediment is usually higher than that in the applied water (Carter et al., 1974;
Westermann et al., 2001). Recently applied or persistent pesticides may also be trans-
ported off fields with sediment (Spencer et al., 1985).
Small soil particles have large specific surface compared to large particles, and thus
have more capacity per unit mass to adsorb agricultural chemicals. Thus, a large pro-
portion of the phosphorus and other chemicals that move with sediment is associated
with the smallest sediment particles (Brown et al., 1981; Aggasi et al., 1995). Erosion
control techniques that preferentially control large sediments are less effective on agri-
cultural chemical discharge.
4.4.3. Evaluating Erosion and Sediment Discharge Problems
Erosion causes two problems: loss of soil and productivity on the field and down-
stream damage from discharged sediment. Furrow erosion can cause both problems.
Center pivot irrigation can cause erosion damage on the field but seldom has signifi-
cant sediment discharge from the field. The problem being evaluated will determine
the evaluation method.
Quantifying center pivot erosion is difficult because it varies greatly in space and
time. Visual observation of water ponding and surface flows indicates potential for
erosion. Rills and sediment deposition fans show there is a problem. Measurements of
soil depth or crop yield variation between sloped and flat field sections may show ero-
sion has occurred in the past.
Erosion in irrigation furrows is evident from furrows that down cut and evolve to
gulleys in high slope areas or near the furrow inflow end (Figure 4.4), depositional
areas where slope decreases, and very turbid water or visible bed load. As with pivot
irrigation, past erosion damage may be evident from decreased soil depth or crop yield
in eroded areas. Furrow erosion and sediment discharge can be quantified by measur-
ing the flow rate and sediment concentration in the flow (Sojka et al., 1994; Trout,
1996). Sediment concentration of periodic volumetric samples can be evaluated
Figure 4.4. Eroded irrigation furrows.

Figure 4.5. Collecting a furrow water

sample from a furrow flume and
pouring it into an Imhoff cone to de-
termine sediment concentration.
gravimetrically or with Imhoff (sedimentation) cones (Figure 4.5). Volumetric sam-

ples can likewise be collected for chemical analysis. It is critical that a representative
sample be collected, because sediment is not evenly distributed in the flow. This usu-
ally requires collecting the complete furrow flow for a short time period, without dis-
turbing the upstream flow, or by sampling a well-mixed turbulent flow in a drain
ditch. Sediment discharge is measured at the outflow point of the furrow or field. Ero-
sion or sedimentation is determined as the difference in sediment discharge between
two in-field measurement points. Because sediment often deposits on fields where
flows are low (downstream sections) or slopes are flat, field average erosion based on
field discharge measurements usually underestimates actual erosion damage.
The runoff water collection ditch at the downstream end of the field can strongly
affect sediment discharge from the field. Where water tends to pond up and flow
slowly at the field downstream end, much of the sediment in the water may be depos-
ited before leaving the field. Where field cross-slope exceeds 0.5% and farmers cut a
runoff ditch across the lower end of the field to discharge runoff, serious erosion can
occur in the ditch and at the end of the furrows, resulting in a downward sloping (con-
vex) field end and greatly increased sediment discharge.
Although much is known about erosion and sedimentation processes, they cannot
yet be accurately modeled and predicted. Soil erodibility cannot be accurately pre-
dicted. Soil erodibility varies with time and management practices, as well as with soil
texture and constituents. The Revised Universal Soil Loss Equation, RUSLE (USDA-
ARS, 1997), does not apply to irrigation-induced erosion. The Water Erosion Predic-
tion Project (WEPP) model (Nearing et al., 1989), although adapted for sprinkler and
furrow irrigation conditions, has not successfully predicted field measured erosion
(Bjorneberg et al., 1999; Kincaid, 2002).
All sediment discharging from farm fields does not usually reach downstream riv-
ers or lakes (Brown et al., 1974; Brockway and Robison, 1992). Some of the sediment
may deposit in drains or sediment basins. The runoff water may be rediverted by
downstream farmers where much of the sediment may deposit on irrigated fields. Field
measurements must be supplemented by return flow quality and quantity measure-
ments to assess sediment and chemical inflows to water bodies. Assessing damages to
rivers and lakes and their complex aquatic biological systems requires a thorough un-
derstanding of those systems as well as an accounting of the various pollutants.
4.4.4 Strategies for Reduction of Erosion and Sediment Discharge
4.4.4.1 Reducing flow erosiveness. Erosion under center pivot irrigation can be
eliminated by eliminating overland water flow. This requires reducing the water appli-
cation rate, or increasing the soil infiltration rate. The water application rate can be
reduced by increasing application width with spray booms or nozzles with larger wet-
ting patterns. Reducing total application per pass will also reduce runoff, since infiltra-
tion rates decline with application amount. Infiltration enhancement practices include
preservation of surface residues, reduced tillage, and, for sodic soils, reducing the so-
dium adsorption ratio of the soil with the addition of gypsum. Nozzles that produce
small droplets with low kinetic energy may also sustain higher infiltration by reducing
surface sealing. If the application rate cannot be reduced sufficiently, surface pitting
(diking, reservoir tillage) can store excess water in place until it can infiltrate. If runoff
cannot be avoided, surface residue and close-growing crops can reduce the exposure
of the soil to water droplet impact and reduce overland flow shear. Reduced tillage
may also reduce the erodibility of soil.
The erosiveness of furrow flows can be reduced by reducing flow rates. Reducing
flow rate usually results in longer time required to spread water across the field, and
thus lower irrigation water distribution uniformity. Thus, there is usually a trade-off
between reducing erosion and reducing irrigation uniformity. Management practices that
reduce infiltration, such as furrow compaction and surge irrigation, may counteract the
impact of reduced flow rates on uniformity. Shortening furrow lengths by subdividing
fields reduces required flow rates. However, if the number of runoff points is increased
when fields are subdivided, the amount of runoff and sediment discharge may increase.
Mid-field gated pipelines can reduce run lengths without increasing field runoff.
In many cases, average flow rates are set to insure all portions of all furrows are
adequately irrigated. Furrow-to-furrow inflow and infiltration variability creates the
need to increase average flow rates to adequately irrigate most of the field (Trout and
Mackey, 1988). Reducing flow rate to reduce erosion damage and allowing a small
portion of the field to be inadequately irrigated may be beneficial in the long term.
Furrow application systems that facilitate uniform furrow flows allow reduced average
flow rates and greatly reduce runoff rates. Reduced flow rate after stream advance is
complete (cutback) will result in reduced erosion. Irrigation scheduling may result in
smaller total application amounts and times, although erosion reduction will likely be
proportionately less than the application reduction.
Flow velocity, and thus erosiveness, is also reduced by increasing furrow rough-
ness. Furrow roughness can be increased by leaving or placing crop residue in the fur-
row (Aarstad and Miller, 1981; Carter and Berg, 1991). However, roughness also
slows water advance and may reduce irrigation uniformity. Furrow residue is a good
option for steep sections of furrows where erosion is greatest and water advance is
rapid (Brown and Kemper, 1987). Miller et al. (1987) used straw mulching in combi-
nation with surge irrigation to reduce erosion and maintain irrigation uniformity.
Carter and Berg (1991) found that no-tillage resulted in lower infiltration during early
season irrigations so irrigation uniformity was maintained while surface residue essen-
tially eliminated erosion.
Erosion can be reduced by reducing furrow slope, but substantially changing field
slopes is usually not practical. In some cases, the furrow direction can be oriented
across the slope (contour furrows) to reduce effective furrow slope. This practice can
result in severe concentrated flow erosion if water overtops rows.
On fields with a convex downstream end, if water flow velocity in the runoff ditch
can be reduced, sediment deposition may gradually fill in the depression. Carter and
Berg (1983) devised a buried pipe runoff collection system that collects sediment and
eliminates convex field ends. Eliminating runoff collection ditches and planting close-
growing crops on the convex end can slow the flow and reduce erosion and may result
in sediment deposition on convex ends. Portable canvas dam checks across eroding
runoff collection ditches can reduce ditch erosion.
4.4.4.2 Reducing soil erodibility. Our understanding of soil aggregate stability,
cohesiveness, and erodibility is poor. Few techniques are available to reduce erodibil-
ity. Erosion tends to be higher after tillage, so reducing the number and depth of tillage
operations reduces erosion (Carter and Berg, 1991). Sodium disperses clays and can
increase erosion (Lentz et al., 1996). Decreasing the sodium adsorption ratio of soil or
irrigation water can reduce erosion.
Polyacrylamide (PAM) applied in the irrigation water has been shown to dramati-
cally reduce furrow erosion. PAM has two effects: it acts as a soil stabilizer and re-
duces erodibility, and it flocculates suspended sediment particles inducing them to
deposit. Low concentrations of PAM (<10 mg/L) applied with the irrigation water,
reduces erosion by over 90% in most cases (Lentz et al., 1992; Lentz and Sojka, 1994;
Sojka et al., 1998a; Sojka et al., 1998b). Polyacrylamide applications have been shown
to reduce discharge of sediments, nutrients, pesticides, and microorganisms in surface
runoff (Lentz et al., 1998; Sojka and Entry, 2000; Entry et al., 2002). Material costs
are about $7 per ha per application, and re-application is recommended at least follow-
ing every tillage operation. Polyacrylamide was used on an estimated 400,000 ha in
1998 to control erosion (Sojka et al., 1998a).
4.4.4.3 Reducing sediment discharge. If erosion cannot be adequately controlled
on the field, off-field practices may be required to remove sediment from the runoff.
These techniques are less desirable than on-field erosion control because they do not
eliminate erosion damage to the field.
Sediment can be removed from water by slowing the flow to allow time for sus-
pended sediment particles to settle out. Sediment basins with a residence time of at
least two hours will settle out sand-sized particles, most of the silt, and a portion of the
clay (Carter, 1985). For a tailwater flow of 30 L/s, basin volume must be at least 220
m3 for two-hour residence. Sediment basin sizes vary from large ponds on major
drains to small basins at the field outflow point. Sediment basins require the accumu-
lated sediment to be periodically excavated and piled until it can be spread back onto
the fields or other areas requiring topsoil fill. Basin size must account for expected
sediment deposition rates and desired clean-out intervals.
Sediment can be collected at the low end of fields by slowing the flow in tailwater
ditches with excavated pits or earthen surface checks. These “minibasins” are more
efficient if water is directed from each basin into a ditch or buried runoff water collec-
tion system rather than allowing water to flow from basin to basin. Minibasins gener-
ally need to be rebuilt each year. Vegetative filter strips of small grains or permanent
cover crops at the downstream end of fields can also slow runoff flow and accumulate
sediment.
Sediment retention efficiency of adequately sized basins varies from 70% to 95%
(Carter et al., 1993). A weakness of sediment basins is that they least efficiently retain
small-sized sediment particles to which most of the ions are attached. Thus, their effi-
ciency in containing phosphorus and other potential pollutants is lower than their
sediment retention efficiency (Brown et al., 1981; Bjorneberg and Lentz, 2005).
4.4.4.4 Surface runoff containment and reuse. Properly designed and used irriga-
tion runoff reuse systems can contain all farm runoff and associated sediments and
contaminants on-farm. These systems must have sufficient storage and pumping ca-
pacity to allow effective use of tailwater (ASABE, 2005). With runoff reuse, a portion
of the sediment can be recycled back to fields, reducing the frequency required for
storage pond clean-out. Any substances in the runoff are also contained on the farm.
The farmer must be aware of potential problems with transporting pests or chemicals
from one field to another, but it is preferable that a farmer deal with potential prob-
lems on the originating farm.
Nutrient and other farm chemical application in irrigation water is becoming a
common practice (fertigation, chemigation). Nitrogen application in surface irrigation
water is common in some areas. For surface or sprinkler irrigation with runoff, runoff
water containment and reuse should be required when chemigating with materials that
could be harmful to downstream farmers or ecosystems.
REFERENCES
Aarstad, J. S., and D. E. Miller. 1981. Effects of small amounts of residue on furrow
erosion. Soil Sci. Soc. America J. 45(1): 116-118.
Aggasi, M., J. Letey, W. J. Farmer, and P. Clark. 1995. Soil erosion contribution to
pesticide transport by furrow irrigation. J. Environ. Qual. 24(5): 892-895.
Ahuja, L. R., K. W. Rojas, J. D. Hanson, M. J. Shaffer, and L. Ma. 2000. Root Zone
Water Quality Model: Modeling Management Effects on Water Quality and Crop
Production. Highlands Ranch, Colo.: Water Resources Pub. LLC.
Allen, L., T. Bennet, J. H. Lehr, and R. J. Petty. 1985. DRASTIC: A standardized
system for evaluating ground-water pollution potential using hydrogeologic
settings. EPA 600/2-85-018. Washington, D.C.: U.S. Govt. Print. Office.
Anderson, B. S., J. W. Hunt, B. M. Phillips, P. A. Nicely, V. de Vlaming, V. Conner,
N. Richard, and R. S. Tjeerdema. 2003. Integrated assessment of the impacts of
agricultural drainwater in the Salinas River (CA). Environ. Pollution 124: 523-532.
ASAE. 2005. EP-408.2: Irrigation runoff reuse. St. Joseph, Mich.: ASAE.
Baker, J. L., T. S. Colvin, D. C. Erbach, R. S. Kanwar, and P. A. Lawlor. 1995.
Herbicide banding to reduce inputs and environmental losses. In Proc. Clean
Water—Clean Environment—21st Century Conference, I: 13-16. St. Joseph, Mich.:
ASAE.
Batista, S., E. Silva, S. Galhardo, P. Viana, M. J. Cerejeira, I. S. Fosmgaard, and B. B.
Mogensen (eds.). 2002. Evaluation of pesticide contamination of ground water in
two agricultural areas of Portugal. In Proc. 8th Symposium on Chemistry and Fate
of Modern Pesticides, Int’l. J. Environ. Analyt. Chem. 82(8-9): 601-609.
Bjorneberg, D. L., T. J. Trout, R. E. Sojka, and J. K. Aase. 1999. Evaluating WEPP
predicted infiltration, runoff, and soil erosion for furrow irrigation. Trans. ASAE
42(6): 1733-1741.
Bjorneberg, D. L., D. T. Westermann, and J. K. Aase. 2002. Nutrient losses in surface
irrigation runoff. J. Soil Water Cons. 57: 524-529.
Bjorneberg, D. L., and R. D. Lentz. 2005. Sediment pond effectiveness for removing
phosphorus from PAM-treated irrigation furrows. Appl. Eng. Agric. 21(4): 589-593.
Bouwer, H. 1987. Effect of irrigated agriculture on groundwater. J. Irrig. Drain. Eng.
113(1): 4-15.
Breve, M. A., R. W. Skaggs, H. Kandil, J. E. Parsons, and J. W. Gilliam. 1992.
DRAINMOD-N: A nitrogen model for artificially drained soils. In Proc. 6th Int’l.
Drainage Symposium, 327-336. St. Joseph, Mich.: ASAE.
Brockway, C. E., and C. W. Robison. 1992. Middle Snake River water quality study.
Phase I: Final report. Moscow, Idaho: University of Idaho, Idaho Water Resources
Research Inst.
Brown, M. J., and W. D. Kemper. 1987. Using straw in steep furrows to reduce soil
erosion and increase dry bean yields. J. Soil Water Cons. 42(3): 187-191.
Brown, M. J., D. L. Carter, and J. A. Bondurant. 1974. Sediment in irrigation and
drainage water and sediment inputs and outputs for two large tracts in Southern
Idaho. J. Environ. Qual. 3(4): 347-351.
Brown, M. J., J. A. Bondurant, and C. E. Brockway. 1981. Ponding surface drainage
water for sediment and phosphorus removal. Trans. ASAE 24: 1478-1481.
Burgard, D. J., W. C. Koskinen, R. H. Dowdy, and H. H. Cheng. 1993. Metolachlor dis-
tribution in a sandy soil under irrigated potato production. Weed Sci. 41(4): 648-655.
CDPR (Calif. Dept. of Pesticide Regulation). 2000. Sampling for Pesticide Residues
in California Water—2000 Update of the Well Inventory Database. Available at:
www.cdpr.ca.gov/docs/empm/pubs/ehapreps/eh0015.pdf. Accessed 11 June 2003.
CDWR (Calif. Dept. of Water Resources). 2000. CALSIM Water Resources
Simulation Model. Available at: modeling.water.ca.gov/hydro/model/index.html.
Carsel, R. F., L. A. Mulkey, M. N. Lorber, and L. B. Baskin. 1985. The pesticide root
zone model (PRZM): A procedure for evaluating pesticide leaching threats to
groundwater. Ecol. Model. 30(1/2): 49-69.
Carter, D. L. 1985. Controlling erosion and sediment loss on furrow-irrigated land. In
Soil Erosion and Conservation. S. A. El-Swaify et al., eds. Ankeny, Iowa: Soil
Cons. Soc. America.
Carter, D. L. 1993. Furrow erosion lowers soil productivity. J. Irrig. Drain. Eng.
119(6): 964-974.
Carter, D. L., and R. D. Berg. 1983. A buried pipe system for controlling erosion and
sediment loss on irrigated land. Soil Sci. Soc. America J. 47(4): 749-752.
Carter, D. L., and R. D. Berg. 1991. Crop sequences and conservation tillage to
control irrigation furrow erosion and increase farmer income. J. Soil Water Cons.
46(2): 139-142.
Carter, D. L., M. J. Brown, C. W. Robbins, and J. A. Bondurant. 1974. Phosphorus
associated with sediment in irrigation and drainage waters for two large tracts in
southern Idaho. J. Environ. Qual. 3(3): 287-291.
Carter, D. L., C. E. Brockway, and K. K. Tanji. 1993. Controlling erosion and
sediment loss in irrigated agriculture. J. Irrig. Drain. Eng. 199(6): 975-988.
Chew, C. F., and T. C. Zhang. 1998. In-situ remediation of nitrate-contaminated
ground water by electrokinetics/iron wall processes. Water Sci. Tech. 38(7): 135-142.
Clay, S. A., Z. Liu, D. E. Clay, and S. S. Harper. 1995. Atrazine binding and
movement in soil as influenced by ammonia fertilizer. In Proc. Clean Water—
Clean Environment—21st Century Conference, I: 41-44. St. Joseph, Mich.: ASAE.
Cole, T. A., and S. A. Wells. 2002. CE-QUAL-W@: A two-dimensional, laterally
averaged, hydrodynamic and water quality model. Version 3.1, User Manual,
Instruction Report EL-02-1. Prepared for the U.S. Army Corps of Engineers.
Corwin, D. L., and B. L. Waggoner. 1991. TETrans: A user-friendly, functional model
of solute transport. Water Sci. Tech. 24(6): 57-65.
Corwin, D. L., B. L. Waggoner, and J. D. Rhoades. 1991. A functional model of solute

transport that accounts for bypass. J. Environ. Qual. 20(3): 647-658.
Crumpton, W. G., and J. L. Baker. 1993. Integrating wetlands into agricultural
drainage systems: Predictions of nitrate loading and loss in wetlands receiving
agricultural drainage. In Proc. Integrated Resource Management and Landscape
Modification for Environmental Protection Symposium, 118-126. J.K. Mitchell, ed.
Dahab, M. F., and S. Sirigina. 1994. Nitrate removal from water supplies using
biodenitrification and GAC-sand filter systems. Water Sci. Tech. 30(9): 133-139.
Derby, N. E., and R. E. Knighton. 2001. Field-scale preferential transport of water and
chloride tracer by depression-focused recharge. J. Environ. Qual. 30: 194-199.
Domagalski, J. L., and N. M. Dubrovsky. 1992. Pesticide residues in ground water of
the San Joaquin Valley, California. J. Hydrol. 130(1/4): 299-338.
Eid, N., D. Slack, and D. Larson. 2000. Nitrate electromigration in sandy soil: Closed
system response. J. Irrig. Drain. Eng. 126(6): 389-397.
Ellerbroek, D. A., D. S. Durnford, and J. C. Loftis. 1998. Modeling pesticide transport
in an irrigated field with variable water application and hydraulic conductivity. J.
Environ. Qual. 27(3): 495-504.
Elliott, J. A., A. J. Cessna, W. Nicholaichuk, and L. C. Tollefson. 2000. Leaching rates
and preferential flow of selected herbicides through tilled and untilled soil. J.
Environ. Qual. 29(5): 1650-1656.
Entry, J. A., R. E. Sojka, M. Watwood, and C. Ross. 2002. Polyacrylamide
preparations for protection of water quality threatened by agricultural runoff
contaminants. Environ. Poll. 120(2): 191-200.
FAO (Food and Agriculture Organization). 1996. Chapter 7: Food production: The
critical role of water. In World Food Summit Technical Background Documents,
Vol. 2. Rome, Italy: FAO. Available at: www.fao.org/docrep/003/w2612E/ w2612e
07a.htm.
FAO (Food and Agriculture Organization). 2002. The salt of the earth: Hazardous for
food production. Issue paper for World Food Summit: Five Years Later. Rome,
Italy: FAO. Available at: www.fao.org/worldfoodsummit/english/newsroom/focus
/focus1.htm.
Flury, M. 1996. Experimental evidence of transport of pesticides through field soils: A
review. J. Environ. Qual. 25: 25-45.
Frankenberger, W., and R. A. Engberg, eds. 1998. Environmental Chemistry of
Selenium. New York, N.Y.: Marcel Dekker.
Goesselin, D. C., J. Headrick, R. Tremblay, X. H. Chen, and S. Summerside. 1997.
Domestic well water quality in rural Nebraska: Focus on nitrate-nitrogen,
pesticides, and coliform bacteria. Ground Water Monit. Remediat. 17(2): 77-87.
Gu, B. H., D. B. Watson, L. Y. Wu, D. H. Phillips, D. C. White, and J. Z. Zhou. 2002.
Microbiological characteristics in a zero-valent iron reactive barrier. Environ.
Monit. Assess. 77(3): 293-309.
Hallberg, G. R. 1986. Overview of agricultural chemicals in ground water. In Proc.
Agric. Impact on Ground Water: A Conference, 1-63. Dublin, Ohio: Nat. Well
Water Assoc.
Hallberg, G. R. 1989. Nitrate in ground water in the United States. In Nitrogen
Management and Ground Water Protection, 35-74. R.F. Follett, ed. New York,
N.Y.: Elsevier.
Hamilton, P. A., and D. R. Helsel. 1995. Effects of agriculture on ground-water

quality in five regions of the United States. Ground Water 33(2): 217-226.
Hanks, J. E. 1995. Herbicide use reduction by improved application technology. In
Proc. Clean Water—Clean Environment—21st Century Conference, I: 81-84. St.
Hefner, S. G., and P. W. Tracy. 1995. Corn production using alternate furrow nitrogen
fertilization and irrigation. J. Prod. Agric. 8(1): 66-69.
Helweg, O. J. 1989. Editorial for the special section: Effect of irrigated agriculture on
groundwater quality in the humid area. J. Irrig. Drain. Eng. 115(5): 771-772.
Herbert, S. J., F. X. Mangan, G. Liu, J. Daliparthy, A. V. Barker, and L. J. Moffitt.
1995. Nitrate leaching in alternate cover crop systems. In Proc. Clean Water—
Clean Environment—21st Century Conference, II: 71-74. St. Joseph, Mich.: ASAE.
Hickman, M. V., and G. D. Vail. 1995. Role of controlled-release herbicide
formulations in reducing groundwater contamination. In Proc. Clean Water—Clean
Environment—21st Century Conference, I: 89-92. St. Joseph, Mich.: ASAE.
Hillel, D. 1982. Introduction to Soil Physics. Orlando, Fla.: Academic Press.
Hornsby, A. G. 1990. Pollution and public health problems related to irrigation. In
Irrigation of Agricultural Crops. B. A. Stewart and D. R. Nielsen, eds. Agronomy
30: 1173-1188. Madison, Wis.: ASA-CSSA-SSSA.
Hunter, Jr., W. J. 1986. A review of ground water remediation technologies and their
application to agricultural problems. In Proc. Agricultural Impact on Ground
Water: A Conference, 648-665. Dublin, Ohio: National Water Well Assoc.
Hunter, W. J., and R. F. Follett. 1995. Use of vegetable oil to bioremediate high nitrate
well water. In Proc. Clean Water—Clean Environment—21st Century Conference,
II: 79-82. St. Joseph, Mich.: ASAE.
Maupin. 2004. Estimated water use in the U.S. in 2000. U.S. Geological Survey
Circular 1268. Reston, Va.: USDI. Available at: pubs.usgs.gov/circ/2004/circ1268/.
Jones, J. L., and L. M. Roberts. 1999. The relative merits of monitoring and domestic
wells for ground water quality investigations. Ground Water Monit. Remediat.
19(3): 138-144.
Juergens-Gschwind, S. 1989. Ground water nitrates in other developed countries
(Europe): Relationships to land use patterns. In Nitrogen Management and Ground
Water Protection, 75-138. R. F. Follett, ed. New York, N.Y.: Elsevier.
Jury, W. A., W. R. Gardner, and W. H. Gardner. 1991. Soil Physics. 5th ed. New
York, N.Y.: John Wiley and Sons, Inc.
Jury, W. A., W. F. Spencer, and W. J. Farmer. 1983. Behavior assessment model for
trace organics in soil. I: Model description. J. Environ. Qual. 12: 558-564.
Kemper, W. D., T. J. Trout, M. J. Brown, and R. C. Rosenau. 1985. Furrow erosion
and water and soil management. Trans. ASAE 28(5): 1564-1572.
Kincaid, D. C. 2002. The WEPP model for runoff and erosion prediction under
sprinkler irrigation. Trans. ASAE 45(1): 67-72.
Knisel, W. G., and R. A. Leonard. 1989. Irrigation impact on groundwater: Model
study in humid region. J. Irrig. Drain. Eng. 115(5): 823-838.
Knox, E., and D. W. Moody. 1991. Influence of hydrology, soil properties, and
agricultural land use on nitrogen in groundwater. In Managing Nitrogen for
Groundwater Quality and Farm Profitability, 19-57. R. F. Follett, D. R. Keeney,
and R. M. Cruse, eds. Madison, Wis.: Soil Sci. Soc. America.

Koluvek, P. K., K. K. Tanji, and T. J. Trout. 1993. Overview of water erosion from
irrigation. J. Irrig. Drain. Eng. 119(6): 929-946.
Kool, J. B., and M. T. van Genuchten. 1992. HYDRUS: One-dimensional, variably
saturated flow and transport model with hysteresis and root water uptake. Version
3.4. Riverside, Calif.: U.S. Salinity Laboratory, USDA-ARS.
Kraft, G. J., W. Stites, and D. J. Mechenich. 1999. Impacts of irrigated vegetable
agriculture on a humid north-central U.S. sand plain aquifer. Ground Water 37(4):
572-580.
Krider, J. N. 1986. Ground water pollution potential from irrigation: Eastern U.S. In
Proc. Water Forum ’86 Conf. ASCE, 1500-1507. Reston, Va.: ASCE.
Labadie, J. W., M. L. Baldo, and R. Larson. 2000. MODSIM: Decision Support
System for River Basin Management. Available at: modsim.engr.colostate.edu/.
Law, Jr., J. P. 1987. Irrigation effects in Oklahoma and Texas. J. Irrig. Drain. Eng.
113(1): 49-56.
Leij, F. J., T. H. Skaggs, and M. T. van Genuchten. 1991. Analytical solutions for
solute transport in three-dimensional semi-infinite porous media. Water Resour.
Res. 27(10): 2719-2733.
Leij, F. J., N.Toride, and M. T. van Genuchten. 1993. Analytical solutions for non-
equilibrium solute transport in three-dimensional porous media. J. Hydrol. 151:
193-228.
Lentz, R. D., and R. E. Sojka. 1994. Field results using polyacrylamide to manage
furrow erosion and infiltration. Soil Sci. 158(4): 274-282.
Lentz, R. D., R. I. Shainberg, R. E. Sojka, and D. L. Carter. 1992. Preventing
irrigation furrow erosion with small application of polymers. Soil Sci. Soc. America
J. 56(6): 1926-1932.
Lentz, R. D., R. E. Sojka, and D. L. Carter. 1996. Furrow irrigation water-quality
effects on soil loss and infiltration. Soil Sci. Soc. America J. 60(1): 238-245.
Lentz, R. D., R. E. Sojka, and C. W. Robbins. 1998. Reducing phosphorus losses from
surface-irrigated fields: Emerging polyacrylamide technology. J. Environ. Qual.
27: 305-312.
Leonard, R. A., W. G. Knisel, and D. A. Still. 1987. GLEAMS: Groundwater loading
effects of agricultural management systems. Trans. ASAE 30(5): 1403-1418.
Ma, L., and R. F. Spalding. 1997a. Effects of artificial recharge on ground water qual-
ity and aquifer storage recovery. J. American Water Resour. Assoc. 33(3): 561-572.
Ma, L., and R. F. Spalding. 1997b. Herbicide persistence and mobility in recharge lake
watershed in York, Nebraska. J. Environ. Qual. 26(1): 115-125.
Merwin, I. A., J. A. Ray, T. S. Steenhuis, and J. Boll. 1996. Groundcover management
systems influence fungicide and nitrate-N concentrations in leachate and runoff
from a New York apple orchard. J. American Soc. Hort. Sci. 121(2): 249-257.
Miller, D. E., J. S. Aarstad, and R. G. Evans. 1987. Control of furrow erosion with
crop residues and surge flow irrigation. Soil Sci. Soc. America J. 51: 421-425.
Mock, J. F., and P. Bolton. 1993. The ICID Environmental Check-List. Wallingford,
England: H. R. Wallingford Ltd.
Mossbarger, Jr., W. A., and R. W. Yost. 1989. Effects of irrigated agriculture on
groundwater quality in Corn Belt and Lake states. J. Irrig. Drain. Eng. 115(5): 773-
790.
NASS (National Agricultural Statistics Service). 2004. 2002 Census of Agriculture.

Available at: www.nass.usda.gov/census/census02/volume1/us/index1.htm.
NASS (National Agricultural Statistics Service). 2004. 2002 Census of Agriculture:
Farm and Ranch Irrigation Survey. Vol. 3. Available at: www.nass.usda.gov/
census/census02/fris/fris03.htm.
NRC (National Research Council). 1989. Irrigation Induced Water Quality Problems.
Committee on Irrigation Induced Water Quality Problems, Water Science and
Technology Board, NRC. Washington D.C.: National Academic Press.
Nearing, M. A., G. R. Foster, L. J. Lane, and S. C. Finkner. 1989. A process-based soil
erosion model for USDA: Water erosion prediction project. Trans. ASAE 32(5):
1587-1593.
Obenhuber, D. C., and R. Lowrance. 1991. Reduction of nitrate in aquifer microcosms
by carbon additions. J. Environ. Qual. 20(1): 255-258.
Postel, S. 1993. Water and agriculture. In Water in Crisis: A Guide to the World’s Fresh
Water Resources. P. H. Gleick, ed. New York, N.Y.: Oxford Univ. Press.
Presser, T. S., and H. M. Ohlendorf. 1987. Biogeochemical cycling of selenium in the
San Joaquin Valley, California, USA. Environ. Manag. 11(6): 805-821.
Rhodes, J. D., and J. D. Oster. 1986. Chapter 42: Solute content. Part 1: Physical and
mineralogical methods. In Methods of Soil Analysis, 985-1006. 2nd ed. A. Klute,
ed. Agronomy Monograph #9, part 1. Madison, Wis.: ASA.
Rhoades, J. D., and J. Loveday. 1990. Salinity in irrigated agriculture. In Irrigation of
Agricultural Crops, 30: 1089-1142. B. A. Stewart and D. R. Nielsen, eds. Madison,
Wis.: ASA-CSSA-SSSA.
Rhoades, J. D., S. M. Lesch, S. L. Burch, J. Letey, R. D. LeMert, P. J. Shouse, J. D.
Oster, and T. O’Halloran. 1997. Salt distributions in cracking soils and salt pickup
by runoff waters. J. Irrig. Drain. Eng. 123: 323-328
Ritter, W. F. 1989. Nitrate leaching under irrigation in the United States: A review. J.
Environ. Sci. Health A24(4): 349-378.
Ritter, W. F., A. E. M. Chirnside, and R. W. Scarborough. 1989a. Pesticide movement
in a coastal plain soil under irrigation. In Toxic Substances in Agricultural Water
Supply and Drainage, 2nd Pan-American ICID Regional Conference, 389-400.
Denver, Colo.: U.S. Committee, Int’l. Commission on Irrigation and Drainage.
Ritter, W. F., F. J. Humenik, and R. W. Skaggs. 1989b. Irrigated agriculture and water
quality in East. J. Irrig. Drain. Eng. 115(5): 807-821.
Robertson, W. D., D. W. Blowes, C. J. Ptacek, and J. A. Cherry. 2000. Long-term
performance of in situ reactive barriers for nitrate remediation. Ground Water
38(5): 689-695.
Robins, J. P. J. Rock, D. F. Hayes, and F. C. Laquer. 2000. Nitrate removal for Platte
Valley, Nebraska synthetic groundwater using a constructed wetland model. Envir.
Tech. 21(6): 653-659.
Rosegrant, M. W., X. Cai, and S. A. Cline. World water and food to 2025: Dealing
with scarcity. Washington D.C.: Int’l. Food Policy Res. Inst. Available at:
www.ifpri.org/pubs/books/water2025book.htm.
Sabol, G. V., H. Bouwer, and P. J. Wierenga. 1987. Irrigation effects in Arizona and
New Mexico. J. Irrig. Drain. Eng. 113(1): 30-48.
San Joaquin Valley Drainage Program. 1990. A management plan for agricultural
subsurface drainage and related problems on the Westside San Joaquin Valley.
Final Report (Rainbow Report). Sacramento, Calif.: Calif. Dept. of Water

Resources. Available at: www.owue.water.ca.gov/docs/RainbowReportIntro.pdf.
Schipper, L., and M. Vojvodic-Vukovic. 1998. Nitrate removal from groundwater
using a denitrification wall amended with sawdust: Field trial. J. Environ. Qual.
27(3): 664-668.
Schipper, L. A., and M. Vojvodic-Vukovic. 2000. Nitrate removal from groundwater
and denitrification rates in a porous treatment wall amended with sawdust. Ecol.
Eng. 14(3): 269-278.
Schmidt, K. D., and I. Sherman. 1987. Effect of irrigation on groundwater quality in
California. J. Irrig. Drain. Eng. 113(1): 16-29.
Shaffer, M. J., A. D. Halvorson, and F. J. Pierce. 1991. Nitrate leaching and analysis
package (NLEAP): Model description and application. In Managing Nitrogen for
Groundwater Quality and Farm Profitability, 285-322. R. F. Follett, D. R. Keeney,
and R. M. Cruse, eds. Madison, Wis.: Soil Sci. Soc. America.
Shirmohammadi, A., and W. G. Knisel. 1989. Irrigated agriculture and water quality
in South. J. Irrig. Drain. Eng. 115(5): 791-806.
Singh, J., P. J. Shea, L. S. Hundal, S. D. Comfort, T. C. Zhang, and D. S. Hage. 1998.
Iron-enhanced remediation of water and soil containing atrazine. Weed Sci. 46(3):
381-388.
Skipton, S., and D. Hay. 1998. Drinking water: Nitrate and methemoglobinemia (“blue
baby” syndrome). Bull. G98-1369. Lincoln, Nebr.: Nebraska Coop. Ext.
Smith, D. M. 1994. Flow through wetlands: A way to manage salts and trace elements.
Smith, S. J., and D. K. Cassel. 1991. Estimating nitrate leaching in soil materials. In
Managing Nitrogen for Groundwater Quality and Farm Profitability, 165-188. R. F.
Follett, D. R. Keeney, and R. M. Cruse, eds. Madison, Wis.: Soil Sci. Soc. America.
Sojka, R. E., R. D. Lentz, and J. A. Foerster. 1994. Software utilizing Imhoff cone
volumes to estimate furrow-irrigation erosion. J. Soil Water Cons. 49(4): 400-406.
Sojka, R. E., R. D. Lentz, D. L. Bjorneberg, and J. K. Aase. 1998a. Polyacrylamide
effects on infiltration in irrigated agriculture. J. Soil Water Cons. 54(4): 325-331.
Sojka, R. E., R. D. Lentz, and D. T. Westermann. 1998b. Water and erosion
management with multiple application of polyacrylamide in furrow irrigation. Soil
Sci. Soc. America J. 62(6): 1672-1680.
Sojka, R. E., and J. A. Entry. 2000. Influence of polyacrylamide application to soil on
movement of microorganisms in runoff water. Environ. Pollution 108: 405-412.
Sonnen, M. B., J. L. Thomas, and J. C. Guitjens. 1987. Irrigation effects in six western
states. J. Irrig. Drain. Eng. 113(1): 57-68.
Spalding, R. F., M. E. Burbach, and M. E. Exner. 1989. Pesticides in Nebraska’s
ground water. Ground Water Monit. Rev. 9(4): 126-133.
Spalding, R. F., D. G. Watts, D. D. Snow, D. A. Cassada, M. E. Exner, and J. S.
Schepers. 2003. Herbicide loading to shallow ground water beneath Nebraska’s
Management Systems Evaluation Area. J. Environ. Qual. 32(1): 84-91.
Spencer, W. F., M. M. Cliath, J. W. Blair, and R. A. LeMert. 1985. Transport of
pesticides from irrigated fields in surface runoff and tile drain waters. USDA-ARS
Conservation Report 31. Springfield Va.: NTIS.
Steenhuis, T. S., S. Pacenka, and K. S. Porter. 1987. MOUSE: A management model
for evaluating groundwater contamination from diffuse surface sources aided by
computer graphics. Appl. Agric. Res. 2(4): 277-289.

Suarez, D. L., and J. Simunek. 1997. UNSATCHEM: Unsaturated water and solute
transport model with equilibrium and kinetic chemistry. Soil Sci. Soc. America J.
61: 1633-1646.
Toride, N., F. J. Leij, and M. T. van Genuchten. 1993. A comprehensive set of
analytical solutions for nonequilibrium solute transport with first-order decay and
zero-order production. Water Resour. Res. 29(7): 2167-2182.
Trout, T. J. 1996. Furrow irrigation erosion and sedimentation: On field distribution.
Trans. ASAE 39(5): 1717-1723.
Trout, T. J., and B. E. Mackey. 1988. Furrow inflow and infiltration variability. Trans.
ASAE 31(2): 531-537.
Trout, T. J., and W. H. Neibling. 1993. Erosion and sedimentation processes on
irrigated fields. J. Irrig. Drain. Eng. 119(6): 947-963.
USDA-ARS. 1997. Predicting soil erosion by water: A guide to planning with the
Revised Universal Soil Loss Equation. USDA Ag. Handbook #703. Also: RUSLE2
Program Users Guide. Available at: fargo.nserl.purdue.edu/rusle2_dataweb/
RUSLE2_Index.htm W. Lafayette, Ind.: USDA-ARS Natl. Soil Erosion Laboratory.
U.S. Environmental Protection Agency. 2002. National water quality inventory: 2000
Report. EPA-841-4-02-001. Available at: www.epa.gov/305b/2000report/.
van Genuchten, M. T., and W. J. Alves. 1982. Analytical Solutions of the One-
Dimensional Convective-Dispersive Solute Transport Equation. USDA Tech. Bull.
No. 1661.
Wagenet. R. J. 1986. Chapter 45: Water and solute flux. Part 1: Physical and
mineralogical methods In Methods of Soil Analysis, 1055-1088. 2nd ed. A. Klute,
ed. Agronomy Monograph #9, part 1. Madison, Wis.: ASA.
Wagenet, R. J., and J. L. Hutson. 1987. Leaching estimation and chemistry model.
Ithaca, N.Y.: Water Resources Institute, Cornell Univ.
Walker, B., and Z. Ross. 1999. Tap water to 38 central California cities tainted with
banned pesticide. Environmental Working Group Memorandum, Nov. 99.
Available at: www.ewg.org/reports/dbcp/dbcp.pdf.
Watts, D. G., N. R. Fausey, and D. A. Bucks. 1999. Background of the MSEA-
RZWQM modeling project. Agron. J. 91(2): 169-170.
Weier, K. L., J. W. Doran, A. R. Mosier, J. F. Power, and T. A. Peterson. 1994.
Potential for bioremediation of high nitrate irrigation water via denitrification. J.
Environ. Qual. 23(1): 105-110.
Westermann, D. T., D. L. Bjorneberg, J. K. Aase, and C. W. Robbins. 2001.
Phosphorus losses in furrow irrigation runoff. J. Environ. Qual. 30: 1009-1015.
Williams, C. F., S. D. Nelson, and T. J. Gish. 1995. Release and mobility of starch-
encapsulated atrazine in calcareous soils. In Proc. Clean Water—Clean
Environment—21st Century Conference, I: 173-174. St. Joseph, Mich.: ASAE.
Wilson, G. V., D. D. Tyler, J. Logan, G. W. Thomas, R. L. Blevins, M. C. Dravillas,
and W. E. Caldwell. 1995. Tillage and cover crop effects on nitrate leaching. In
Proc. Clean Water—Clean Environment—21st Century Conference, III: 251-254.
St. Joseph, Mich.: ASAE.
Wilson, P. C., T. Whitwell, and M. B. Riley. 1995. Effects of ground cover and formulation
on herbicides in runoff water from miniature nursery sites. Weed Sci. 43(4): 671-677.
CHAPTER 5
EFFICIENCY AND
UNIFORMITY
Dale F. Heermann (USDA-ARS,
Abstract. The objective of this chapter is to present the major factors that must be
considered to best meet the demands of those competing for the use of limited water
resources. It is critical for the designer to understand, define, and select the appropri-
ate efficiency and uniformity parameters when designing irrigation systems. The con-
cepts of irrigation efficiency and uniformity are often misunderstood and lead to con-
fusion. Their parameters are multidimensional in both space and time. The design
engineer must select the appropriate target performance parameters to meet the ob-
jectives of the system within the imposed constraints. Management and operation of
the system are as important as the system design in meeting the target performance
parameters.
Keywords. Basin efficiency, Beneficial use, Hydrologic basin, Irrigation design, Ir-
rigation efficiency, Microirrigation, Performance parameters, Sprinkler irrigation,
Surface irrigation, Uniformity, Water management.
5.1 INTRODUCTION
Irrigation offers tremendous benefits in increased food and fiber production. In the
U.S., irrigated fields account for only 15% of harvested cropland, but produce 38% of
the dollar value of food and fiber (Bajwa et al., 1992). Irrigators are often criticized for
growing crops that are surplus to demands for food and fiber as the public is becoming
more concerned about the efficient use of water. Worldwide, irrigation is the largest
consumer of water, more than municipal and industrial use.
Irrigation system type and design affect not only the efficiency but also the uni-
formity of water application. Uniformity refers to how uniformly water is applied; this
affects many parameters that are used to assess irrigation performance. Efficiency can
can be measured in a myriad of ways, and efficiency by one measure may not be effi-
cent by another measure. Also, the highest system efficiency may not meet economic
or environmental objectives. For example, under-irrigation may have the highest effi-
ciency in the short term but can lead to salination problems in the long term. The con-
cepts of efficiency and uniformity, as applied to irrigation, will be discussed in greater

detail below.
An engineer designing an irrigation system must, then, determine the appropriate
target performance parameters to meet the objectives of the system within the imposed
constraints. These constraints can be economic, environmental, water quality and
quantity, crops, soils, labor, service, and management skills. The target performance
parameters should be selected to meet the constraints, and the highest attainable effi-
ciency may not be appropriate. Finally, it is important to recognize that the manage-
ment and operation of the system are as important as the design of the system in meet-
ing the target performance parameters.
5.2 IRRIGATION SCHEME PHYSICAL MODELS
5.2.1 Physical and Temporal Evaluation Scale
Physical scale is important to consider when evaluating the efficiency of water use.
Efficiencies may be defined for a single field, or on larger scales up to a hydrologic
basin, as well as for the various pathways of the hydrologic cycle. Water use can be
categorized as beneficial or non-beneficial for crop production. The water diverted to
irrigation can also be divided into consumptive and nonconsumptive uses. Figure 5.1
illustrates the partitioning of water use. An understanding of these concepts is neces-
sary in formulating definitions of irrigation efficiencies. For example, beneficial uses
include more than evaporation for crop needs. The nonconsumptive use and nonbene-
ficial use may be undesirable from a single field viewpoint but may not be a loss from
a basin viewpoint.
Efficiency is not only a function of the spatial scale but also the temporal scale. Ir-
rigation application efficiencies are often evaluated for a single irrigation event, but
the assumption that this is equal to the seasonal irrigation efficiency for growing a
crop is probably incorrect. Differences in the soil conditions, stages of plant growth,
climatic conditions, and other factors can result in substantially different efficiencies.
Management and operation of the system can also change the efficiency.
Consumptive Use Nonconsumptive Use

Beneficial
Crop evapotranspiration Water for leaching

Evaporation for cooling
Evaporation for frost protection
Phreatophyte evapotranspiration Excess deep percolation

Nonbeneficial
Weed evapotranspiration Excess surface runoff

Spray evaporation Operational spill
Evaporation from soil
Reservoir and canal evaporation
Figure 5.1. Examples of the partitioning of some irrigation water uses.

110 Chapter 5 Efficiency and Uniformity
5.2.2 Irrigation System, Farm and Field Model

The objective of this monograph is to provide state-of-the-art information for the
design of irrigation systems. Economical crop production is an important considera-
tion from the users’ point of view. Conservation and environmental stewardship are
recognized as another important consideration in the selection and operation of an
irrigation system. Public concern for the environment and the competing demands for
limited water supplies increase the need to consider the various performance parame-
ters. Global efficiency is of more importance when the environment and limited water
supplies are major factors. Drought conditions increase the problem of limited water
supply. Basin efficiency and uniformity can be entirely different than that at the local
level.
Surface runoff may be of little concern to a user if sufficient water is available at
low cost and soil erosion is not a problem. This may not represent a loss in terms of
the basin hydrology. Also, costs may not be impacted, as some water supplies and
costs are based on water rights per unit of area irrigated without a limitation on the
volume. However, where water is limited, the system should be designed to limit run-
off. One option is a pumpback system, which recirculates the water back to the head
end of the field. Often a pumpback system is designed for an entire farm, where the
runoff from one field is collected and either pumped or diverted to another field.
A major concern from a basin perspective is degradation of the quality of the water,
in both runoff and deep percolation, by the pickup of salts and other pollutants. The
degradation only affects the user in the added costs of the water and chemicals that
may not be used effectively. Physical definitions of performance parameters do not
generally consider the environmental degradation of the water supply. This is a sepa-
rate issue that is growing in importance with the concern for the environment.
Water rights for either surface sources or groundwater are often limited to “reason-
able use,” which has a legal meaning that may change with technology. Generally,
reasonable uses include water for satisfying the crop evapotranspiration (ET), water
for beneficial leaching, and evaporation from canals and water emission devices.
Thus, reasonable use includes beneficial water use that can be both consumptive and
nonconsumptive (Figure 5.1). Some unavoidable, nonbeneficial consumptive uses may
also be accepted as reasonable. Even some nonbeneficial, nonconsumptive uses, such
as excess deep percolation and surface runoff, may be considered acceptable in the
design if sufficient water is available. Major concerns of excess deep percolation is
water quality degradation as the water moves through the profile into the groundwater
and the loss of soluble plant nutrients. Nonbeneficial nonconsumptive use is a loss
with respect to a field or farm but may be used to satisfy another water right within a
basin.
Uniformity of water application affects whether water use is beneficial, consump-
tive, or not. Sprinkler and microirrigation systems provide better control of water ap-
plication compared to surface irrigation systems, which have the inherent problems of
runoff and nonuniform soil infiltration that strongly affect how much water is avail-
able for each plant. Even with uniform soils, it is difficult to design and operate a sur-
face irrigation system that provides a uniform application of water, without under-
irrigation in part of the field and/or excessive deep percolation in other parts. Even
sprinkler systems do not always apply water uniformly, as they are influenced by sys-
tem hydraulics, uneven terrain, and surface translocation of applied water when appli-
cation rates exceed soil intake rates. Similarly, with microirrigation the application
may vary because of nonuniform system pressure, plugged emitters, and undetected
leaks. (Detailed characteristics of the various types of irrigation systems are given in
later chapters.)
5.2.3 Hydrologic Basin Model
A hydrologic basin is a collection of many fields or farms that are in the same hy-
drologic drainage area. The irrigation efficiency of a basin is often higher than the
typical irrigation efficiency for an individual field or farm, because inefficiencies of
water use on one farm may result in water that is used on another. Recall that irriga-
tion efficiency is typically defined as the ratio of water beneficially used to that di-
verted from the source. The surface water runoff from one farm may be rediverted and
used on another farm. Even the water that deep percolates on one farm may be
pumped for use on either the same farm or another farm downstream in the basin.
Thus, the sum of water diverted for each field in a basin often exceeds the total supply
because a given quantity of water may be diverted several times. A disadvantage of
reusing either irrigation return flow or deep percolation water is that the quality may
be degraded. In some cases, water quality degradation is such that it cannot be used
satisfactorily for crop production. Some groundwater moves in the aquifer, the quality
may be reduced, and it may become uneconomical to pump. Only surface runoff and
groundwater that move out of the basin are lost for potential use in crop production in
the basin, and even this water may be available in another basin unless it runs into the
ocean.
The current trend is to place a higher priority or value on leaving the water instream
for ecological benefits. This is another constraint that may effectively decrease the
available water supply and often has not been considered in the design of irrigation
systems.
5.3 IRRIGATION PERFORMANCE PARAMETER DEFINITIONS
The following sections define commonly used parameters that are useful in the de-
sign and performance evaluation of irrigation systems. This is not an all-inclusive list;
engineers are encouraged to clearly describe any efficiency terms that are needed to
meet the particular objectives and constraints of the system being designed. The fol-
lowing sections provide definitions published by Israelson (1950), Jensen et al. (1967),
ASCE (1978), Bos (1979), Hansen et al. (1980), Jensen et al. (1983), Walker and Sko-
gerboe (1987), and Burt et al. (1997).
5.3.1 Water Conveyance Efficiency
The conveyance of water from the source to the irrigated field can be through natu-
ral drainage ways, constructed earthen or lined channels, or closed conduits. Many
conveyance systems have transmission losses, thus, water delivered to the field is usu-
ally less than the direct diversion from a flowing stream, reservoir, or underground
aquifer. The conveyance efficiency of a conveyance system is the ratio of the volume
of water delivered to the field boundary to the volume of water diverted from the
source and can be expressed as:
Vf
ec = (5.1)
Vt
where ec = conveyance efficiency

Vf = volume of water delivered to the field

Vt = volume of water diverted from the source.
The difference between the two volumes in Equation 5.1 is the amount evaporated,
seeped from the canal, leaked from the closed conduit, spilled, and/or otherwise lost
from the conveyance and distribution system. The numerator of Equation 5.1 becomes
smaller as the losses increase, thus decreasing the conveyance efficiency. Conveyance
efficiency is often expressed for various reaches along main, secondary, and tertiary
canals. A logical subdivision is the reaches that are under common management. Each
management unit can evaluate the need, benefits, and costs of decreasing the losses
under its control. The total conveyance efficiency is the product of the efficiencies for
individual sequential reaches.
Increasing conveyance efficiency for efficiency’s sake may not be economically
justified nor increase the available water supply. The entire demand on a given water
supply must be examined to determine the necessity and value of increasing the con-
veyance efficiency. For example, operational spills may cost very little and only return
high quality water back to the stream for rediversion downstream. However, when
water demands cannot be satisfied, it may justify decreasing operational spills. Simi-
larly, actual cost and available water supply must be considered before lining canals or
installing pipelines to reduce losses, but this may be necessary to meet the demand for
water and/or to reduce water quality degradation caused by seepage.
Hydraulic grade lines can be kept more constant, which in turn provides a more
constant flow rate to each turn-out on the system. Variability in flow delivery is an
important factor that must be considered when designing an irrigation system for indi-
vidual fields on a farm. Rapid changes in the grade line can frequently lead to struc-
tural failures of canal walls and control structures, increasing the maintenance costs
and inability to make timely deliveries of irrigation water.
5.3.2 Water Application Efficiency
Water application efficiency is a measure of the fraction of applied water that is
stored in the soil profile and available for crop use. It is commonly expressed for an
individual field as field water application efficiency, expressed as:
Vs
ea = (5.2)
Vf
where ea = water application efficiency of the field

Vs = volume of irrigation water stored for evapotranspiration by the crop
Vf = volume of water delivered to the field.
Spray drift from the field, runoff, spray evaporation, and deep percolation losses all
contribute to reduce the application efficiency of irrigation systems. The volume of
water loss in spray evaporation or wind drift is difficult to estimate or measure. It may
satisfy some of the evaporative demand and reduce the potential ET for the crop, thus
indirectly satisfying part of the crop water requirements. Such water is, therefore, part
of the beneficial use and would not decrease the application efficiency. However, wa-
ter intercepted on the crop canopy is evaporated at a rate higher than the transpiration
rate of adequately watered plants and tends to decrease the application efficiency.
The water application efficiency can be unity when irrigation amounts are small,
the soil profile is not filled (minimizing deep percolation), and surface runoff is mini-
mized, as all of the water delivered is available to meet the crop water requirements
(ET plus leaching requirement). Even so, the available water may not be sufficient to
satisfy crop water requirements, causing yield reductions.
5.3.3 Soil Water Storage Efficiency
The soil water storage efficiency (Hansen et al., 1980; Walker and Skogerboe,
1987; and James, 1988) is the ratio of the volume of water stored in the soil root zone
to the volume of water required to fill the root zone to field capacity. It is expressed as:
Vs
es = (5.3)
V fc − Va
where es = soil water storage efficiency

Vs = volume of water stored in the soil root zone from an irrigation event
Vfc = volume capacity at field capacity in the soil root zone
Va = volume of water in the soil root zone prior to an irrigation event.
In other words, a high es means that irrigation brings the soil to, but not beyond,
field capacity in the root zone. To minimize deep percolation, the maximum net
amount of water that should be applied in a given irrigation event is the difference
between the field capacity and the average water content in the root zone at the time of
irrigation. We discourage the use of the soil water storage efficiency because of diffi-
culty in determining the root zone, which changes during the season and is different
for every crop, soil, and management practice. The main use of the soil water storage
efficiency is to manage surface and sprinkler irrigation systems where the objective is
to minimize labor and the number of irrigation events, and prevent overirrigation.
One problem with using storage efficiency with sprinkler and microirrigation sys-
tems is that even if it is low, frequent irrigations may still provide sufficient water for
crop production, and this management practice leaves some soil water storage room
for rainfall. Sprinkler and microirrigation systems are typically operated on a frequent
basis and can supply just the water needed without filling the profile.
5.3.4 Irrigation Efficiency
Water to satisfy crop ET requirements is not the only beneficial water that can be
supplied with an irrigation system. The ASCE On-Farm Irrigation Committee (ASCE,
1978) defines the irrigation efficiency as the ratio of the volume of water which is
beneficially used to the volume of irrigation water applied, expressed as:
Vb
ei = (5.4)
Vf
where ei = irrigation efficiency

Vb = volume of water beneficially used
Beneficial uses may include crop water use, salt leaching, frost protection, crop
cooling, and pesticide or fertilizer applications. Excessive deep percolation, surface
runoff, weed ET, wind drift (in part), and spray evaporation are not beneficial uses and
thus would tend to decrease the irrigation efficiency. Other factors that impact benefi-
cial use and thus water use efficiency include theft, misallocation, water rights, social
rules, night irrigation, and management.
5.3.5 Deep Percolation Ratio

High water tables and subsurface return flow to streams can result from deep perco-
lation. The deep percolation ratio is an important evaluation parameter when these
conditions exist and is more effectively used with another efficiency term, such as the
water application or irrigation efficiency. It is particularly significant when high water
tables need to be avoided and when the groundwater returned to the streams is of low
quality. Degradation of many streams and rivers in the arid western United States has
resulted from the return of low-quality groundwater. The deep percolation ratio is de-
fined (Walker and Skogerboe, 1987) as:
V dp
DP r = (5.5)
V f
where DPr = deep percolation ratio

Vdp = volume of water percolated below the root zone
5.3.6 Tailwater Ratio
Tailwater (surface runoff) is lost from the lower ends of fields and does not con-
tribute to crop production on that field unless it is recirculated by pumping the tailwa-
ter to the head end of the field. Tailwater may also be captured and pumped or di-
verted to another field. Tailwater is often allowed to return to a nearby stream. The
tailwater ratio is an important performance parameter if the water flows directly into
the ocean or if runoff is prohibited by law. The tailwater ratio may be quite different
from the deep percolation ratio. Disposition of tailwater relative to other components
of the total water budget is important in evaluating an irrigation system. The degree of
water quality degradation and the potential reuse in the basin affect the negative im-
pact of tailwater. The ratio is defined (Walker and Skogerboe, 1987) as:
V ro
TW r = (5.6)
Vf
where TWr = tailwater ratio
Vro = volume of surface runoff
5.3.7 Irrigation Uniformity
Irrigation efficiencies are expressed as functions of the volumes of water diverted
and the use or disposition of the water as it is applied with the irrigation system. The
nonuniformity of application within a given field is not accounted for in the efficiency
definitions. However, when or where the soil profile is not filled (perhaps in only
some areas because irrigations were not applied uniformly), the crop may exhibit
stress. An irrigation system that does not apply water uniformly must apply excess
water in some areas in order that there is enough water in other areas, such that mini-
mal plant stress occurs over the entire field. The excess water may cause surface run-
off and/or deep percolation beneath the root zone. Also, when water is applied in ex-
cess (whether by irrigation or rainfall) and the soil is saturated for several days or
more, plant oxygen stress may occur. Resulting deep percolation may even cause a
perched water table, depending on the subsoil conditions. Percolation beneath the root
zone is required to leach salts that accumulate in the root zone.
Many of the volumes used in the efficiency definitions—for irrigation, for deep
percolation, etc.—are difficult to measure in practice, because they are affected by
uniformity. Generally irrigation uniformity is based on indirect measurements. For
example, the uniformity of water that enters the soil is assumed to be related to that
caught in catch cans for sprinkler systems, to emitter discharge for microirrigation
systems, and to intake opportunity time and infiltration rates for surface systems.
The physical uniformity (size, color, wilt differences, etc.) of the growing plants is
another possible performance parameter for the design and evaluation of an irrigation
system. The purpose of an irrigation system is to enhance the growth of plants for crop
production and/or their aesthetic value. Having water application events that result in
uniform biological responses (measured yield or visual appearance) over the entire
irrigated area of production is generally desirable. This may require that the entire land
surface area receive a uniform application of water (i.e., for closely spaced seeded or
drilled crops). However, tree crops do not require uniform water applications over the
entire land surface area. Rather, application of uniform amounts of water per tree is
more important and influences the spacing requirement of microirrigation and under-
tree sprinkler irrigation systems.
Precision farming emphasizes the need to apply the production inputs in the proper
amounts and locations to maximize their effectiveness. Site-specific farming may re-
quire that irrigation systems apply water in a nonuniform pattern, but with precise con-
trol of the application to satisfy the spatial requirements. The biological and economic
models of plant growth and production are quite complex and are generally evaluated
with a physical measure of the uniformity of the inputs. The complexity of the bio-
logical and economic models result from interactions between the crop, fertility re-
quirements, soil differences, and system management. They cannot be expressed as a
single function of the irrigation uniformity. Thus, the selected design uniformity is
more subjective than objective.
Several mathematical definitions have been proposed and used to describe the uni-
formity of a system. Christiansen’s uniformity coefficient (1942) was defined to
evaluate sprinkler irrigation systems and has the strongest historical precedent in the
sprinkler irrigation industry. It is defined as:
⎡ ∑ xi − xm ⎤
CU = 100 ⎢1.0 − ⎥ (5.7)
⎣⎢ ∑ xi ⎥⎦
where CU = Christiansen’s uniformity coefficient, %
xi = measured depth (volume or mass) of water in equally spaced catch cans on
a grid
xm = mean depth (volume or mass) of water of the catch in all cans.
This requires that each catch can represents the depth applied to equal areas. This is
not true for data collected under center pivots where the catch cans are equally spaced
along a radial line from the pivot to the outer end. For center pivot systems it is neces-
sary to adjust and weight each measurement based on the area it represents (Heerman
and Hein, 1968).
Specific definitions will be given in the chapters discussing the various irrigation
systems. There are many different formulations and expressions for quantifying uni-
formity. However, most uniformities can be calculated from one another with assumed
statistical distributions of the applied depths. The standard deviation and coefficient of
variation are examples of other ways to quantify uniformity. The different formula-
tions are quite like expressing a length in either millimeters or inches. One can be led
to believe that one uniformity calculation method is more sensitive than another be-
cause of a different scale, but the various formulations will maintain the same relative
order when different irrigation systems are compared.
Since the measurement of applied depths is difficult for surface and many microir-
rigation systems, the uniformity coefficient is not generally determined directly. The
intake opportunity times or measured soil water differences from before to after an
irrigation are used to estimate application depths for surface irrigation systems.
For microirrigation systems, the emitter discharge rates are used in place of meas-
ured application depths. Thus, microirrigation uniformity is affected by the emitter
discharge rates, which are in turn affected by emitter manufacturing and irrigation
system hydraulic characteristics. The variability of emitter discharge caused by varia-
tions in orifice size (or shape) and hydraulic characteristics can result from inadequate
manufacturing quality control. Keller and Karmeli (1975) defined an empirical design
emission uniformity relationship for microirrigation systems as:
⎛ C ⎞ q
EU =100 ⎜⎜1.0 −1.27 v ⎟⎟ min (5.8)
⎜ n p ⎟ qa
⎝ ⎠
where EU = emission uniformity, %
Cv = manufacturer’s coefficient of variation
np = number of emitters per plant, at least one per plant
qmin = minimum emitter discharge rate computed from the minimum pressure
qa = average emitter discharge rate.
The above definition is based on the ratio of the discharge rate for the lowest 25% of
the emitters to the average discharge rate.
Nakayama et al. (1979) developed a coefficient of design uniformity, CUd, which is
based on the discharge rate deviations from the average rate. It is expressed as:
⎡ 0.798(C vm) ⎤
CU d = ⎢1 - ⎥100 (5.9)
⎣ n ⎦
where the terms are defined as above and the constant 0.798 results from assuming a
normal distribution of discharge rates and using the Christiansen uniformity definition
(Equation 5.7). Hart (1961) developed a similar relationship for the Christiansen dis-
tribution uniformity where the coefficient of variation of sprinkler application is de-
scribed by the ratio of the standard deviation divided by the mean.
The uniformity for surface irrigation systems is more commonly characterized by
distribution uniformity, defined as the average depth infiltrated in the low one-quarter
of the field divided by the average depth infiltrated over the entire field, expressed as:
D lq
DU = (5.10)
D av
where DU = distribution uniformity
Dlq = average depth infiltrated on the one-quarter of the field with the least in-
filtration
Dav = average depth infiltrated over the entire field.
The distribution uniformity is also often applied to microirrigation and sprinkler irriga-
tion systems including center pivots.
The literature has many definitions for evaluating the uniformity of an irrigation.
Many of them use the moments of the measured or estimated distribution of depths.
However, it has been reported (Hart and Heermann, 1976) that many measured distri-
butions can be expressed as mathematical functions of each other. Another parameter
often used assumes a normal distribution and all that are needed is the mean depth and
standard deviation. Warrick (1983) summarized the interrelationships of irrigation
uniformity terms with a number of population distributions. The Christiansen uniform-
ity and low-quarter distribution uniformity are related mathematically for normal, log
normal, uniform, specialized power, beta, and gamma distributions of water applica-
tions.
The coefficient of uniformity is typically evaluated for an individual irrigation, but
it may be more important to evaluate the uniformity of several irrigation events or
even over an entire irrigation season. The uniformity coefficient generally increases if
the depths are accumulated for multiple irrigations because of the random nature of
application and wind effects.
Redistribution in the soil can affect the variability of the water actually available to
the crop. Initial soil water content, application rates, surface distribution of applied
water, length scale of distribution on the soil surface, soil hydraulic conductivity, soil
depth, and total applied water are factors that affect the potential redistribution within
the profile. Hart (1972) estimated that CU could increase from 54% to 61% with lat-
eral redistribution for one simulation with a surface scale length of 1 m. Root distribu-
tion is another factor that can influence uniformity. It should be emphasized that crop
production or landscape response is the important consideration when evaluating an
irrigation system. If the root distribution is such that it can remove water from the ar-
eas receiving more water, the effective uniformity for crop growth may be higher than
the calculated physical parameter.
The measurement of the depths applied is important for calculating the sprinkler
CU. A number of sources of error could reduce the resulting uniformity coefficient.
Selection of the catch collector is a major consideration. The collector should have a
sharp edge so that the area of the catch is defined by the surface dimensions of the
collector and the water does not run either in or out of a broad, flat lip. The depth of
the collector must be sufficient to prevent water from splashing out of the container.
The projection of the opening of the collector must be horizontal so that the water
caught is the depth applied to the surface area and not larger or smaller. The effect of
evaporation must also be considered to prevent losses from the can before the col-
lected amounts are measured. It is recommended that a small depth of oil be added to
the catch can to limit the evaporation from the time of application until the depth is
measured. Another alternative is to measure the evaporation from an outside can. The
use of non-evaporating collectors (Clark et al., 2002) can eliminate the effects of
evaporation. Wind has also been shown to divert the water from catch cans resulting in
the measured depth being less than that actually reaching the surface of the soil for
infiltration. Another source of error for moving irrigation systems is leaks in the pipe-
line or drips off of the system trusses, support structure, or sprinkler drops that can run
directly into the catch can causing extremely large depths. Also, cans can tip over and
cause a missing or smaller reading. Caution should be taken when calculating CU with
data that may have measurement errors. It is recommended that the data obviously in
error be adjusted before use in calculation. The error would generally contribute to a
decrease in the uniformity coefficient.
Current ASABE (formerly ASAE) standards and engineering practices should be
reviewed for evaluation procedures of irrigation systems. The following are examples;
they are routinely updated.
ASAE S436.1. Test Procedure for Determining the Uniformity of Water Distri-
bution of Center Pivot and Lateral Move Irrigation Machines Equipped with
Spray or Sprinkler Devices.
ASAE EP405.1 Design and Installation of Microirrigation Systems.
ASAE EP419.1 Evaluation of Irrigation Furrows.
5.4 SUMMARY
An irrigation system is designed to enhance plant growth for crop production or
aesthetic value of turf and ornamentals. This can be accomplished with a nonuniform
and inefficient irrigation system, but that is unacceptable to the public for water quan-
tity and environmental concerns. Current designs must be uniform and efficient. This
chapter presented the various factors that need to be considered as well as performance
parameters and indices for evaluating new designs and existing irrigation systems. The
evaluation indices will provide the measure for more efficient systems to conserve
water and enhance the environment.
REFERENCES
ASCE. 1978. Describing irrigation efficiency and uniformity. J. Irrig. Drain. Div.
104(IR1): 35-41.
Bajwa, R. S., W. M. Crosswhite, J. E. Hostetler, and O. W. Wright. 1992. Agricultural
Irrigation and Water Use. ERS/USDA Agricultural Information Bulletin No. 638.
Washington, D.C.: ERS/USDA.
Bos, M. G. 1979. Standards for irrigation efficiencies of ICID. J. Irrig. Drain. Div.,
105(1): 37-43.
Burt, C. M., A. J. Clemmens, T. S. Strelkoff, K. H. Solomon, R. D. Bliesner, L. A.
Hardy, T. A. Howell, and D. E. Eisenhauer. 1997. Irrigation performance measures:
Efficiency and uniformity. J. Irrig. Drain. Eng. 123(6): 423-442.
Christiansen, J. E. 1942. Irrigation by sprinkling. Univ. Calif. Agr. Exp. Sta. Bull. 670.
Clark, G. A., D. H. Rogers, E. Dogan, and R. Krueger. 2002. The Irrigage: A non-
evaporating in-field precipitation gage. ASAE Paper No. 022068. St. Joseph,
Mich.: ASAE.
Hansen, V. E., O. W. Israelson, and G. E. Stringham. 1980. Irrigation Principles and
Practices. 4th ed. New York, N.Y.: John Wiley & Sons, Inc.
Hart, W. E. 1961. Overhead irrigation pattern parameters. Agric. Eng. 42(7): 354-355.
Hart, W. E. 1972. Subsurface distribution on nonuniformly applied surface waters.
Trans. ASAE 15(4): 656-661, 666.
Hart, W. E., and D. F. Heermann. 1976. Evaluating water distribution of sprinkler
irrigation systems. Colorado State Univ. Tech. Bulletin 128, June.
Heermann, D. F., and P. R. Hein. 1968. Performance characteristics of self-propelled

center-pivot sprinkler irrigation system. Trans. ASAE 11: 11-15.
Israelson, O. W. 1950. Irrigation Principles and Practices. 2nd ed. New York, N.Y.:
John Wiley & Sons.
James, L. G. 1988. Principles of Farm Irrigation System Design. New York, N.Y.:
John Wiley & Sons.
Jensen, M. E., L. R. Swamer, and J. T. Phelan. 1967. Improving irrigation efficiencies.
In Irrigation of Agricultural Land, 1120-1142. R. M. Hagan et al., eds. Madison,
Wis.: American Soc. Agronomy.
Jensen, M. E., D. S. Harrison, H. C. Korven, and F. E. Robinson. 1983. The role of
irrigation in food and fiber production. In Design and Operation of Farm Irrigation
Systems, 15-41. M. E. Jensen, ed. St. Joseph, Mich.: ASAE.
Keller, J., and D. Karmeli. 1975. Trickle Irrigation Design. Glendora, Calif.: Rain
Bird Sprinkler Mfg. Corp.
Nakayama, F. S., D. A. Bucks, and A. J. Clemmens. 1979. Assessing trickle emitter
application uniformity. Trans. ASAE 22(1): 816-821.
Walker, W. R., and G. V. Skogerboe. 1987. Surface Irrigation, Theory and Practice.
Englewood Cliffs, N.J.: Prentice Hall, Inc.
Warrick, A. W. 1983. Interrelationships of irrigation uniformity terms. J. Irrig. Drain.
Eng. 109(3): 317-332.
CHAPTER 6
SOIL WATER
RELATIONSHIPS
Roger E. Smith (UDSA-ARS,
Arthur W. Warrick (University of
Arizona, Tucson, Arizona)
Abstract. Basic relations of soil water and soil water flow important in irrigation
design are presented, and methods to measure soil water content, pressure head, and
conductivity are outlined. The calculation of infiltration rates and the measurement of
soil infiltration parameters are discussed, as well as many of the complexities and
challenges in applying current understanding to irrigation situations.
Keywords. Infiltration, Redistribution, Soil physics, Soil water.
6.1 INTRODUCTION
Design and operation of efficient irrigation systems require knowledge of the proc-
esses controlling movement and storage of water in soil. This chapter outlines basic
concepts of the nature of soil water and the interactive forces that affect the distribu-
tion and movement of water in the soil-water-plant system. Methods for measuring the
state of soil water, and soil properties that describe water movement and water-holding
characteristics of soils are presented and discussed. Methods to measure hydraulic
conductivity in both saturated and unsaturated soils are presented and techniques for
predicting the unsaturated hydraulic conductivity function from the soil water reten-
tion characteristic are discussed. Factors controlling infiltration rates and procedures
for measuring infiltration characteristics are also presented and discussed. References
are given to allow the reader to pursue any of these topics in greater detail, and to ob-
tain more detailed outlines of measurement methods.
6.2 WATER-HOLDING CHARACTERISTICS OF SOILS
Soil water has traditionally been of interest because of its influence on plant growth
and crop production as well as runoff processes. A growing plant must be able to bal-
ance the atmospheric demand for water with the amount it can extract from the soil.
The soil water supply is alternately depleted through evapotranspiration and replen-
ished by irrigation or precipitation. Today soil water is of increasing of concern as a
medium through which chemicals may move and potentially harm surface or ground-
water.
6.2.1 Soil Water Content

Soils hold water to the extent that they have porosity, and the water usually shares
that pore space with air. Even “saturated” soil will usually have some air trapped
within. The porosity of soil itself is quite variable, in response to both natural and ag-
ricultural practices.
Soil water content, by weight, is calculated as:
Wmoist − Wdry Wmoist
θw = = −1 (6.1)
Wdry Wdry
where θw= water content expressed on the basis of the dry weight of soil
Wmoist = moist soil weight
Wdry = oven-dry soil weight.
It is common to express soil water contents on a volumetric basis, i.e., the ratio of
the soil water volume to the total soil volume. This is done by multiplying the water
content on a dry weight basis by the ratio of the soil bulk density, ρb, and water den-
sity, ρw, as follows:
ρ
θ = θw b (6.2)
ρw
where θ = water content on a volume basis. The soil bulk density (or apparent den-
sity), ρb, is defined as the oven-dry weight of soil per unit volume, as it occurs in the
field. In the following discussion, θ will be used to mean volumetric water content.
6.2.2 Soil Water Potential
Soil water content alone is not a satisfactory criterion for describing the availability
of water to plants and attempts have been made to describe water availability in terms
of the energy state of water. Initially, empirical measurements and relationships were
developed, but these gave way to consideration of fundamental mechanisms and ex-
pressions. The soil-water-plant system is now treated as a continuous dynamic system
where water moves through the soil to plant root surfaces, into roots, through the
plant, and into the atmosphere along a path of continuously decreasing potential en-
ergy. The removal of soil water depends not only upon its amount and energy state,
but also upon the ability of the plant to absorb water and the atmospheric demand for
water from the plant. A more detailed discussion of the various potential components
can be found in a paper by Rawlins (1976).
About the beginning of the 20th century, soil water was arbitrarily classified into
different forms such as gravitational water, capillary water, hygroscopic water, etc.
(Briggs, 1897, cited by Richards and Wadleigh, 1952). These early groupings have
been replaced by a fundamental concept referred to as soil water potential. Soil water
does not occur in separable “forms” within the range of our interest, but does vary in
the energy with which it is retained in the soil. The work per unit weight to move an
infinitesimal amount of water from some reference state to a given point in the soil is
known as the total soil water potential, hT. The usual reference state, arbitrarily de-
fined as having zero potential, is an open air-water interface at some specified eleva-
tion and air pressure. Energy must be expended to remove water from an unsaturated
soil, so the soil water potential is less than the reference state and thus has a negative
sign. The potential gradient, or rate of decrease of potential energy with distance, is
the driving force causing soil water flow (Section 6.4). Thus, soil water will move
122 Chapter 6 Soil Water Relationships
from a wet area where the potential is near zero, toward a dry region where the poten-
tial is lower (a larger negative value). The soil water pressure has dimensions of
[M/LT2], and the equivalent potential hT has dimension of length. Other definitions
will follow resulting in dimensions of pressure and energy per unit mass.
The total soil water potential may be expressed as the sum of three component po-
tentials:
hT = hg + hp + ha (6.3)
where hT = total soil water potential
hg = gravitational potential
hp = matric or pressure potential
ha = pneumatic potential.
Gravitational potential, hg, is the elevation. If z is the height above a defined refer-
ence plane, hg = z. The value of hg can be positive (if above the reference) or negative
(if below the reference).
The value of hp, the pressure (and matric) potential, can be positive or negative and
is equal to the soil water pressure head (i.e., the pressure divided by the specific
weight). If the soil water pressure is greater than the adjoining gas phase pressure, then
hp will be positive. If the pressure of the soil water is less than the adjoining gas phase,
hp will be negative. This is due to the attraction of soil surfaces for water, the influence
of soil pores, and the curvature of the soil water interface. For this situation, hp is also
called the matric potential. It is convenient to consider pressure potential as a continu-
ous function of water content, which is positive in a saturated soil below the water
table and negative in unsaturated soil. Since soil water potential is generally negative,
it is often given a positive value and referred to as suction or tension.
The pneumatic potential, ha, (energy per unit mass) may be expressed as ha =
psa /(ρ g), where psa is the soil air pressure, ρ is the density of water, and g is gravita-
tional acceleration. Usually the air pressure is considered to be uniform throughout the
soil profile and the pneumatic potential is ignored in characterizing soil water flow.
Such assumptions are not always justified; see Section 6.4.
Two other ways are used to define potential. These are energy per unit volume, hT,v,
and energy per unit mass, hT,m. The dimensions of hT,v are pressure, and the relation to
hT, above, is
hT,v = ρ ghT (6.4a)
Similarly, the relationship between hT and hT,m is
hT,m= ghT (6.4b)
A useful conversion table taken from Hillel (1971) is given here as Table 6.1.
Table 6.1. Energy levels of soil water expressed in various units (from Hillel, 1971).
Soil Water Potential Soil Water Suction
Per Unit Per Unit Per Unit Per Unit
Weight Mass Volume Weight Per Unit Volume
(mm H2O) (joules/kg) (kPa) (mm H2O) (kPa) (bars)
–102.0 –1 –1 102.1 1.0 0.01
–1020. –10 –10 1020. 10. 0.1
–5100. –50 –50 5100. 50. 0.5
–10200. –100 –100 10200. 100. 1.0
–51000. –500 –500 51000. 500. 5.0
To avoid confusion among the various expressions for soil water energy status, one
must keep in mind that a low potential refers to dry soil and is a large negative num-
ber, while a high matric or pressure potential refers to a wet soil with a small negative
value of h. A high potential would be –0.10 bar while a low potential would be –15
bars. On the other hand, low suction or tension refers to wet soil with a small positive
suction value. High suction or tension means a dry soil and is a large positive number;
i.e., a low suction is +0.10 bars and a high suction is +15 bars.
The main incentive for introducing soil water potential, hT, is to describe flow rela-
tions based on spatial differences in hT. For a non-equilibrium system, flow will occur
from a higher to a lower potential. Flow is influenced by additional factors and these
factors are often included as additional components for hT (e.g., Jury et al., 1991). In
particular, osmotic effects are often included as an additional osmotic potential com-
ponent. The osmotic potential is a significant component in saline soils. For coupled
flow processes consisting of flow due to osmotic gradients, temperature gradients,
pressure gradients, and other gradients, it is not necessary to define a total potential
which includes components for each independent part. In fact, Corey and Klute (1985)
showed that the inclusion of chemical effects can lead to contradictions with the no-
tion that flow occurs from regions of high to low potentials. (This does not complicate
the formulation of coupled flows, but simply says that the same transport coefficients
cannot be used for the independent gradient terms for each component.)
6.2.3 The Soil Water Retention Characteristic
As water is removed from a soil, the matric or pressure potential of the water re-
maining is decreased (algebraically, e.g., –1 is decreased to –10). If water is added to
the soil, the matric potential is increased (such as, –10 to –1). A curve showing the
functional relationship between matric potential and soil water content is known as the
soil water characteristic or retention curve. Soil water is usually expressed as volu-
metric water content θ or as volumetric percentage of water. When the relationship is
determined by drying a wet soil, the curve is known as either the desorption curve,
water retention curve, or water release curve. When the relationship is determined as a
dry soil wets, it is called the sorption or imbibition curve. The soil water characteristic
is related in an indirect way to the pore size distribution.
Water is retained in the soil by a combination of the attraction of particle surfaces
for water and the capillary action of water in the soil pores. The matric potential is
related to the curvatures of the air-water interfaces, which in turn are affected by the
soil pore geometry, the particle aggregation, and the soil water content. At high matric
potentials (near zero), most of the soil pores are filled with water and the total porosity
and pore size distribution greatly influence the water retained. Inasmuch as soil texture
dominates the total porosity and pore size distribution, it has a marked effect on the
soil water characteristic. In general, the higher the clay content of a soil, the higher
will be the water content at any given potential. Soil aggregation, especially for fine-
textured soils, tends to increase the number of large pores. Thus, soil structure is im-
portant in the amount of water retained at high potentials. When the large pores empty,
the water remaining in the soil is held in the smaller interaggregate pores and at the
particle contact points. As the soil dries, the amount of particle surface area also af-
fects the water retained, and this is strongly influenced by soil texture. Soil compac-
tion also influences the water characteristic because compaction results in smaller
pores, reduced total porosity, and increased interparticle contact in a given soil vol-
ume. It is usually the larger pores that are reduced most by compaction, so that the
influence of compaction is greater at higher potentials.
Examples of soil water characteristics for three soils of different textures are given
in Figure 6.1. Some common functional forms for describing this relation are pre-
sented in Figure 6.2 and Table 6.2. This table uses scaled water content, Θ, defined as
(θ – θr)/(θs – θr), where θs and θr are known as the saturated and residual water con-
tents, respectively. Water potential is scaled by a parameter α with units [1/L]: h* =
αh.
Residual water content, θr, may be thought of as water which cannot be withdrawn
from a soil by suction, but in practice is often a fitting parameter. Saturated water con-
tent, θs, is a measurable quantity which is usually less than soil porosity because of
entrapped air.
Figure 6.1. Generalized water retention

relations for three different textured soils.
Figure 6.2. Examples of various algebraic forms

for describing the soil water retention relationship.
Table 6.2. Functional relationships for hydraulic characteristics.

Parameter
Function (and abbreviation) m kr = K/Ks Θ (h*) Relation
Gardner (1958) (GR) m>0 exp(h*) Θ = [exp(h*/2)(1–h*/2)]2/(m+2)
Θ p[1–
van Genuchten (1980)[a] (VG) 0<m<1 Θ = (1 + |h*|n)-m
(1 – Θ 1/m)m]2
Brooks and Corey (1964)[b] Θ = |h*|-m/(1-m) ; h* >1
m>0 Θv
(BC) = 1 ; 1< h* < 0
1 1 ⎡ ( m − 1)Θ ⎤
Broadbridge and White (1988) (m − 1)Θ 2 h* = 1 − + ln ⎢
(FBW)
m>1 Θ m ⎣ m − Θ ⎥⎦
(m − Θ )
[c]
Linear None Θ h* = ln Θ
[a]
Use p = 0.5 and commonly use n = 1/(1 – m).
[b]
Use v = 2m + 3 (sometimes v = 2m + 1 or 2m + 2).
[c]
Also may have kr(h) = dθ/dh.
The significance of the parameters α, n, and m used in Table 6.2 in connection with
the van Genuchten (1980) (abbreviated VG) and Brooks and Corey (1964) (BC) func-
tions can best be seen in Figure 6.3, which is a log-log plot of these retention relations
for specific values of m. The log slope of the asymptote is mn or m/(1 – m), often
called the pore-size distribution index, and n determines the degree of curvature in the
region near the intercept. The asymptote intercept is 1/α and is often referred to as the
air-entry head, he. As n becomes large, the shoulder curvature near he becomes
sharper, and the VG expression approaches the more simple BC relation as a limit.
It should be noted that the BC relation is a special case of the VG expression. A
generalized form of the BC relation, called the transitional Brooks-Corey relation
(TBC), has been introduced by Smith (1990). This is functionally equivalent to the VG
relation but retains the same parameters as the BC expression. The relation of m to n
often used in the VG expression (see Table 6.2) is not retained.
Figure 6.3. The parameters in the TBC or VG retention relation have

specific relationship to the shape of the curve, as illustrated here.
The term soil water capacity, C(h), refers to the slope (dθ/dh) of the soil water
characteristic at any point on the curve. This value represents the change in water con-
tent per unit change in matric potential and represents an important property for soil
water storage and release.
The soil water characteristic can be used to estimate the amount of water “released”
between any two potentials. Although the soil water potential largely determines the
ease with which a plant can obtain water, it is also important to know how much water
is in the soil at potentials above a given critical level. This, along with crop water re-
quirements, allows one to estimate the need for irrigation. The soil water potential will
decrease as a plant withdraws water. If hc is considered a critical level below which it
is not desired to deplete water, then the amount of water available at a potential h1, h1
> hc , will be θ(h1) – θ(hc). Many soils swell and shrink with wetting and drying, so
that all of the water does not come from a constant volume of soil. This is especially
important at high potentials where soil structure influences the characteristic.
6.2.3.1 Hysteresis. The soil water characteristics for sorption and desorption will
often differ because the water content in a soil at a given potential depends upon the
wetting and drying history of the soil. This history dependence in the relationship be-
tween potential and water content is called hysteresis. A schematic example of desorp-
tion and sorption curves for a soil is given in Figure 6.4. When the desorption curve is
obtained by drying an initially saturated sample and the sorption curve is obtained by
wetting an initially dry sample, the two moisture characteristics are known as the pri-
mary hysteresis loops or main branches (main drying curve, MD, and wetting curve,
MW, respectively). If the soil is not completely dry before rewetting or not completely
wet before drying, the resulting curves will fall between the two primary curves, and
they are known as scanning curves (wetting scanning curve, WS, and drying scanning
curve, DS). Wherever the starting point is within the main curves, a drying condition
will approach the MD curve, and a wetting condition will approach the MW curve.
At any given potential the water content will be greater in a drying soil (desorption)
than in a wetting soil (sorption). Field soil is rarely either completely wet before dry-
ing, or completely dry before wetting, so measured primary wetting or drying reten-
Figure 6.4. Definition diagram of the hysteresis loops

which can occur during wetting and drying of a soil.
tion curves can be used only with reservation in interpreting soil water status. The
water content or potential can only be estimated from measurement of the water con-
tent and the main curves, unless the wetting history is accurately known. However, the
amount of error involved is relatively small, compared with other errors involved such
as soil variability, climatic changes, and plant variabilities. Excellent discussions of
hysteresis are given by Jury et al. (1991), Hillel (1971), and Nielsen et al. (1972).
6.2.3.2 Methods of determining the soil water characteristic. The soil water
characteristic is usually determined in the laboratory using tension tables or pressure
plates (Figure 6.5). In all of the techniques used, a porous membrane or plate hydrauli-
cally connects the soil water with water in the lower chamber. The pores in the mem-
brane are small enough that, under the imposed pressure, water but not air can pass
through. In all cases P1 > P2, so that water is forced from the soil into the lower cham-
ber. At equilibrium, the imposed pressure (expressed in suitable terms) can be consid-
ered as the potential of the water remaining in the soil.
For high potentials the membrane may be blotter paper, fine sand, sintered glass,
porous steel, or similar materials. In this case P1 is often atmospheric and P2 is ob-
tained with a hanging water column (Figure 6.5a) or with regulated vacuum. At lower
potentials of about –1 bar or less, the pores of these materials are too large to remain
water filled and air will pass through the membrane. A fine-pored ceramic is then used
as the membrane, P2 is atmospheric, and P1 is obtained with compressed gas, usually
air or nitrogen. Ceramic membranes are available with bubbling pressures of 100 bars
and more. The air entry value of the plate should be somewhat matched to the soil
water potential of interest as the finer-pored ceramics necessary for higher pressures
tend to restrict flow. In addition to porous ceramics, porous stainless steel and plastic
materials can be appropriate for the wet range.
Regulated
Pressure
Cover to prevent
evaporation
Soil Porous
P1 P1
Membrane Soil
P2
P2 =
Hanging atmospheric
Water
Column
Figure 6.5. Two methods of determining water retention relations

for a soil sample: (left) hanging column and (right) pressure plate.
In practice, a sample of soil is placed in the pressure chamber in a retaining ring

and saturated overnight. The desired pressure is then applied until outflow ceases and
the soil water is considered to be in equilibrium with the applied pressure. The amount
of water in the soil is then determined, usually by oven drying. The process is re-
peated, with a second sample being subjected to a different pressure. The resulting soil
water contents are plotted against applied pressure or vacuum, expressed as potential
units (usually cm or millibars) to form the water characteristic. Details of apparatus
and procedure are given in Dane and Topp (2002).
Because the pore size distribution has such a large influence on water retention at
high potentials, disturbed samples (dried and sieved) often give erroneous results. Use
of so-called undisturbed soil cores is preferable, but even with these, some error is
inevitable because of the swelling and shrinking that accompanies wetting and drying
of many soils. At low potentials (approaching –15 bars) the soil-specific surface domi-
nates water retention and the error introduced by using disturbed soil samples is quite
small.
Determining the approximate local soil water characteristic in the field may be
done at sites where both soil water potential and water content are measured, using
apparatus discussed below.
6.2.4 Approximate Soil Water Parameters
Field capacity and permanent wilting point once were considered to be soil water
constants. They are now recognized as very imprecise but qualitatively useful terms.
After infiltration ceases, water within the wetted portion of the profile will redis-
tribute under the influence of potential gradients. Downward movement is relatively
rapid at first, but decreases rapidly with time. Field capacity refers to the water con-
tent in a field soil after the drainage rate has become small and it estimates the net
amount of water stored in the soil profile for plant use. While field capacity was for-
merly accepted as a physical property characterizing each soil, now it is used as only a
very rough measure of the soil water content a few days after it has been wetted. For
most soils this is a near-optimum condition for growing plants. However, soil water
will continue to move downward for many days after irrigation. Figure 6.6 (Gardner et
al., 1970) is a good illustration of the continuous nature of profile water redistribution,
contradicting the idea of a definable point associated with this common term. Indeed,
Gardner et al. (1970), referring to their experiments, stated that “the soil exhibited
nothing resembling a field capacity.”
The field capacity concept is perhaps more applicable to coarse than to fine-
textured soils because in coarse soils most of the pores empty soon after irrigation and
the capillary conductivity becomes very small at relatively high potentials. Fine soils,
however, retain more water than coarse soils as well as drain longer at significant
rates. Any interface of soil layers will inhibit water movement across the interface
(more or less, depending on the relative hydraulic properties) and thus restrict redistri-
bution and increase apparent “field capacity” (see Section 6.5.1). Other factors that
may influence soil water redistribution rate are organic matter content, depth of wet-
ting, wetting history, and plant uptake pattern. Even the cultural practices are impor-
tant: for example, the appropriate “field capacity” for dryland farming on a soil is
probably much lower than for a frequently irrigated farming system.
Figure 6.6. After irrigation, redistribution in a soil profile extends for many days,
as demonstrated here by measurements of Gardner et al. (1970).
The permanent wilting point is the soil water content below which plants growing
in the soil remain wilted even when transpiration is nearly eliminated. It represents a
condition where the rate of water supply to the plant roots is very low. The water con-
tent corresponding to the wilting point applies to the average water content of the bulk
soil and not to the soil adjacent to the root surfaces. The soil next to the root surfaces
will usually be drier than the bulk soil (Gardner, 1960), because water cannot move
toward the root surfaces fast enough to supply plant demands and a water content gra-
dient develops near the root.
Like field capacity, permanent wilting is not a soil constant nor a unique soil prop-
erty. There is no single soil water content at which plants cease to withdraw water.
However, for a given soil, the range of water contents for wilting may be quite small,
since soil water contents change little with matric potential at very low potentials.
Plant wilting is a function of demand as well as soil conditions: plants growing under
low atmospheric demand can dry soil to lower water contents than if the demand is
high, because more time is allowed for water to move through the soil to the roots.
When atmospheric demands are high, plants may temporarily wilt even though soil
water contents are considered adequate; an example is sugar beet wilting in midday
during the summer.
In the wilting range, almost all soil pores are empty of water and the water content
is determined largely by the specific surface area and the interparticle contact points.
The water content in soil subjected to a pressure potential of –15 bars is closely corre-
lated with the permanent wilting percentage for a wide range of soils (see Romano and
Santini, 2002). Because of its simplicity and the availability of reliable equipment, the
–15 bar percentage is now commonly used to estimate the permanent wilting point.
Formerly, sunflowers were the standard test plant used for determining permanent
wilting percentage (Romano and Santini, 2002).
The amount of water released by a soil between whatever is considered “field ca-
pacity” and permanent wilting is traditionally called the available water. The term
implies that the available water can be used by plants, but this is misleading. If the soil
water content approaches the wilting range, especially during periods of high atmos-
pheric demands for water or during flowering and pollination, the yield and/or quality
of most crops will decrease significantly. This concept is more important for dryland
agriculture than for irrigated conditions.
Inasmuch as the difference between field capacity and available water can be no
more meaningful than either of the terms, available water is only an estimate of the
amount of water a crop can use from a soil. Many farmers irrigate when the available
water has been depleted a certain amount, depending upon the crop. For high-water-
requiring crops such as potatoes, irrigation may be scheduled at 15% to 25% depletion
(85% to 75% available water remaining in the soil); for many other crops, the deple-
tion may go to 50% to 75% before irrigation. This bank-account type of irrigation ig-
nores any relation between depletion and water potential. For a given soil, the degree
of depletion allowed before irrigation may be roughly related to potential through the
characteristic curve. As with field capacity, available water is a useful concept, provid-
ing that its limitations are recognized, such as variations with soil depth, the influence
of climatic factors on evapotranspiration, and the effects of soil profile characteristics.
6.2.5 Methods for Characterizing Soil Water
The soil water characteristics at locations in the field may be obtained by water po-
tential measuring devices in combination with soil water measurement. Because of
natural soil variability and the variation in soil water content with wetting history (hys-
teresis), the field-determined water characteristic curve is not precise and is difficult to
duplicate. Sorption curves are more difficult and tedious to determine then desorption
curves because equilibrium is reached very slowly.
6.2.5.1 Measuring water content. Topp and Ferre (2002) and Rawlins (1976) have
discussed the various methods and associated error for measuring soil wetness. The
discussion here will be limited to those techniques considered most useful in the field.
Gravimetric. The accepted standard for soil water measurement is the gravimetric
method, which involves weighing a sample of moist soil, drying it to a constant weight
at a temperature of 105° to 110°C, and reweighing to determine the amount of water
lost on drying. The results are often expressed as the ratio of mass of water lost to
mass of dry soil. The required drying time depends upon the soil texture, soil wetness,
loading of the oven, sample size, whether the oven is a forced draft or convection type,
and other factors. Usually 24 h is sufficient but the required time is obtained by re-
peatedly weighing a sample after various periods of drying. Microwave ovens have
been used to reduce drying times (Horton et al., 1982).
The bulk density of soil may be measured by drying and weighing a known volume
of soil, or by using the clod, core, or excavation method (Grossman and Reinsch,
2002). The core method is the most commonly used. A cylindrical metal sampler of a
known volume is forced into the soil at the desired depth. The resulting soil core is
dried and the bulk density is found by dividing the mass by the volume of the cylinder.
Samples may be taken at successive depths from the surface by cleaning out the sam-
ple hole to the desired depths with an auger and then forcing the sampler into the soil
at the bottom of the hole. Alternatively, samples may be taken in a trench by forcing
the sampler into the soil horizontally at the desired depth. Obviously, this latter
method involves more labor but the sampling zones can be better observed. Excellent
core samplers are available commercially.
Neutron scattering. The neutron scattering procedure to estimate soil water content
has gained wide acceptance in recent years. A source of high energy or fast neutrons is
lowered to the desired soil depth into a previously installed access tube. Radium-
beryllium has been used, but current equipment uses americium. The fast neutrons are
emitted into the soil and gradually lose energy by collision with various atomic nuclei.
Hydrogen, present almost entirely in soil water, is the most effective element in the
soil in slowing down the neutrons. Thus, the degree of the slowing down of neutrons
is a measure of the soil water content. The slowed or thermalized neutrons form a
cloud around the source and some of these randomly return to the detector, causing
ionization of the gas within the detector and creating an electrical pulse. The number
of such pulses is measured over a given interval of time with a scalar or the rate of
pulsation can be measured with a ratemeter. The count rate is almost linearly related to
the water content.
When not in use the radiation source is housed in a shield that contains a material
high in hydrogen, such as polyethylene. This material serves as a standard by which
proper operation of the instrument can be verified. Inasmuch as instrument variations
and source decay occur, it is more satisfactory to use the count ratio method rather
than just a count. The ratio of sample count/standard count is plotted versus water con-
tent. This eliminates any systematic errors due to instrumentation that may vary from
day to day. The volume of soil measured depends upon the energy of the initial fast
neutrons and upon the wetness of the soil. With the americium-beryllium source the
volume of soil measured is a sphere of about 150 mm diameter in a wet soil and up to
500 mm or more in a dry soil (Grossman and Reinsch, 2002).
Neutron scattering has some advantages over the gravimetric method because re-
peated measurements may be made at the same location and depth, thus minimizing
the effect of soil variability on successive measurements. It also determines water con-
tent on a volume basis, with the measured soil volume influenced by the instrument
used, the soil type, and wetness. Disadvantages of neutron scattering are the initial
high investment in equipment, the time required per site because of the need to install
access tubes, and the training, licensing, and testing required for possession of a radio-
active device. Also, measurements near the soil surface are not accurate because neu-
trons are lost through the surface, and it is difficult to accurately detect any sharp de-
lineation such as a wetting front or soil layering effect.
The manufacturer usually supplies a calibration curve, but one should verify
whether it applies to a given soil. If changes in water content are desired, rather than
absolute values, a single curve is more widely applicable because the bias will be the
same in successive readings. Two calibration procedures have been used: field calibra-
tion in natural soil profiles, and laboratory calibration in large prepared soil standards.
Calibration should be done with the same type of access tubes as used in the field. For
details of the method, see Topp (2002).
Time domain reflectometry (TDR). This is a relatively new technique used to meas-
ure volumetric soil water content and soil bulk salinity based on the high-frequency elec-
trical properties of soil and water. It offers a variety of advantages including rapid, reli-
able, and repeatable measurements with a minimum of soil disturbance. In theory, TDR
requires no calibration and the soil moisture determination is independent of soil texture,
structure, salinity, density, or temperature. In reality, calibration may be necessary for
different soil types and TDR probes. Additionally, soil salinity can be evaluated with the
technique. More than 50 papers discussing the state-of-the-art developments in TDR are
published in the proceedings of a 1994 conference (U.S. Bureau of Mines, 1994).
A B
+3000
Voltage
0
TDR C
Coaxial -3000
cable t1 t2
Time
Soil surface
Lp TDR probe
Waveguide
Figure 6.7. (a) Installation diagram for time domain reflectometry (TDR) measurement of
surface soil water content. (b) The instrument trace and its interpretation, discussed in text.
TDR refers to both the overall technique and to the electronic device that generates
and measures the electrical signal used to measure the soil water content and salinity. A
sketch of a typical TDR probe is given in Figure 6.7a. It consists of a coaxial cable, han-
dle, and two or three parallel metal rods. The metal rods are also referred to as
waveguides.
A voltage pulse is sent down a cable and the returning signal monitored over time.
The velocity of the electrical pulse is proportional to the dielectric coefficient (κ) of
the soil in contact with the probes. κ is a dimensionless ratio related to the degree of
orientation of dipoles in a material when subjected to an oscillating electric field (see
Table 6.3 for some typical values of κ). For soil, κ is considered independent of den-
sity, texture, structure, temperature, salts (not necessarily true at very low water con-
tents), and others. This is mainly because changes in κ due to changes in water content
are very large compare to changes in κ for the other common constituents in soils. As
a result, it is possible to create a calibration curve relating κ to soil water content.
Topp et al. (1980) conducted a number of experiments to relate the dielectric constant
of a wide variety soil types to volumetric water content. They found a close fit for a
variety of soils and soil-like materials using a single polynomial equation:
θ = –5.3 × 10-2 + 2.9 × 10-2 κ – 5.5 × 10-4 κ2 + 4.3 × 10-6 κ3 (6.5)
This calibration curve is still used, but recent work confirms that it is best to make a
calibration curve for each soil and probe to get an optimal fit.
Table 6.3. Typical values of dielectric coefficient κ.
Material κ (dimensionless)
Perfect metal conductor ∞
Water 70 to 80
Dry soils 2 to 5
Perfect vacuum 1
The velocity, vs, of the electromagnetic wave as it travels through the soil is related
to the dielectric constant κ by
vs = c κ-0.5 (6.6)
where c is the velocity of light. vs is defined as twice the physical length of the probes
(2Lp) divided by the time (2ts) for the wave to travel the length of the probe and back,
so that
2
⎛ ct ⎞
κ =⎜ s ⎟ (6.7)
⎜ Lp ⎟
⎝ ⎠
Therefore, the dielectric constant is a function of the speed of light, the length of the
probe, and the wave travel time.
The art of finding soil water content with TDR is in knowing how to find the two
distance end points on the trace. An ideal trace is given as Figure 6.7b. Point t1 repre-
sents the time when the signal enters the soil and t2 the time when the wave has trav-
eled to the end of the probe and back to the soil surface. For locating t1 and t2, some
users choose to pick the numerical maxima and minima of the trace. For purposes of
automation and for noisy signals, special procedures are adapted for identifying points
on the trace. Also shown on Figure 6.7b is a voltage height C. The value of C is re-
lated to the effective conductivity and hence to soil salinity. For nonsaline conditions,
recovery height C is nearly to the same level as at t1; for a saline condition the height
C will be less. The analysis of the trace in this manner allows for the simultaneous
determination of water content and salinity
Other methods. The attenuation of a beam of gamma rays of known intensity
passed through a soil column is related to the mass of material through which the
beam passes. If the soil bulk density remains constant or varies in a known way, the
water content may be inferred, or if the water content is known, the soil bulk density
may be inferred. This technique has been used primarily in laboratory studies, al-
though a double-tube attenuation unit has been commercially developed for use in the
field (Reginato and Van Bavel, 1964). It is potentially accurate but only for carefully
controlled conditions.
Transient heat pulse measurements have long been used to evaluate soil thermal
properties. Water content is closely related to thermal properties. In particular, the
volumetric heat capacity, ρc, is a sum of that due to the water present and the dry soil
matrix:
ρc = ρw θ cw + ρb cm (6.8)
where ρw = density of the water
ρb = density of the bulk soil
θ = volumetric water content
cw = mass specific heat of the water
cm = mass specific heat of the dry soil.
The value of ρc is found using principles of Fourier heat flow to match temperature
rise due to an applied heat pulse. Thus, if the properties of the dry soil are known, then
θ follows. A commercially available device is available similar to that described by
Bristow et al. (1993).
6.2.5.2 Measuring soil water potential. It is often desirable to measure soil water
potential in addition to, or instead of, soil water content. The estimation of soil water
potential from water content data via the characteristic curve may not be sufficiently
accurate.
Tensiometer. Tensiometers are widely used for measuring the higher ranges of soil
water potential in the field and laboratory. The name is derived from the term “ten-
sion” that was initially applied to the energy of retention of soil water. Many commer-
cial models are available, or the necessary parts can be purchased and assembled at a
significant savings. The theory and use of tensiometers have been discussed by Young
and Sisson (2002).
Schematic diagrams of tensiometers are shown in Figure 6.8. A tensiometer con-
sists of a porous ceramic cup filled with water and connected through a water-filled
tube to a suitable vacuum measuring device. The cup, when saturated with water, must
be capable of withstanding air pressures of about 1 bar without leaking air. For normal
field applications, the vacuum is measured with a reliable vacuum gauge (Bourdon
type). For special conditions where rapid response time is needed, the vacuum meas-
urement is made with pressure transducers and almost no water flow through the cup
is required. The pressure transducers can also be used to continuously record the vac-
uum in the tensiometer, and to read many tensiometers through a switching system.
Tensiometers have been used as sensors for automating irrigation systems to maintain
a desired soil water range. Tensiometers fitted with a septum and read with a portable
pressure transducer attached to a hypodermic needle are also commercially available
(see Young and Sisson, 2002).
The major criticism of the tensiometer is that it functions reliably only in the wetter
soil range at potentials of about –0.8 bar or higher, and its range is limited by the depth
Pressure transducer
Removable air-tight cap
Vacuum guage
Barrel
Porous cup
Figure 6.8. Two types of tensiometer, using a gauge (left) or a pressure transducer (right).
With multiple tensiometers, readings may be taken by a single transducer
with a needle through septums.
of installation. At lower potentials the water inside the tensiometer vaporizes (boils)
and readings are not reliable. In drier sandy soils, when hydraulic conductivities are
very low, tensiometers may not function properly. Soil water flow away from the cup
as the soil dries may be so slow that hydraulic equilibrium with the bulk soil will not
be achieved. Under such conditions the bulk soil water potential may be much lower
than the tensiometer indicates. If it is necessary to measure water potentials at great
depths, tensiometers equipped with integral pressure transducers can be used and only
electrical leads need come to the soil surface (Watson, 1967).
Porous ceramic blocks. The water content of porous blocks in equilibrium with the
soil water may also be used as an approximate measure of soil water potential or soil
water content. Two electrodes are imbedded within a gypsum block and the resistance
between them measured with an AC ohmmeter (alternating current is used to prevent
polarization). Modern resistance blocks utilize an inert material saturated with gyp-
sum. The effect of soil solution salinity levels is masked because the electrolyte within
the block is essentially a saturated solution of calcium sulfate. Gypsum blocks are
relatively cheap and easy to use. Several companies supply them, as well as inexpen-
sive resistance meters. Even though the accuracy is not good, they do indicate soil
water conditions qualitatively and can monitor changes in θ (Topp and Ferre, 2002).
Psychrometric methods. A detailed discussion of the use of psychrometry to meas-
ure water potential has been given by Andraski and Scanlon (2002). The technique
measures the sum of the matric and osmotic potentials. The method most widely
adopted for in situ measurement of soil water potential is the measurement of wet-bulb
temperature, with a small thermocouple serving as the wet bulb. Water is condensed
on the thermocouple by passing a current through it to cool it below the dew point by
the Peltier effect. The current is then removed and the wet-bulb temperature is meas-
ured. Ambient temperature is also measured with the same thermocouple to allow cor-
rections to be made for temperature effects on the calibration curve. Units are cali-
brated against potentials of standard solutions. This technique is most useful for meas-
urement of very low potentials, since the dew point temperature is very near to ambi-
ent (i.e., high relative humidity) at high potentials.
6.3 SOIL HYDRAULIC CONDUCTIVITY
6.3.1 Conductivity and Darcy’s Law
The basic relationship for describing soil water movement was derived from ex-
periments by Darcy who found in 1856 that the flow rate in porous materials is di-
rectly proportional to the hydraulic gradient. This relation was originally set forth by
Buckingham (1907), although it is better known as Darcy’s law (Swartzendruber,
1969), and may be written as:
ΔH
q = −K (6.9)
Δs
where q is the volume of water moving through the soil in the s-direction per unit area
per unit time and ΔH/Δs is the hydraulic gradient in the same direction. The propor-
tionality factor, K, is the hydraulic conductivity, which depends on properties of both
the fluid and the porous medium. H is the hydraulic head which is the sum of the pres-
sure head, h, and the elevation head, z (Section 6.2.2). The negative sign in Equation
6.9 indicates flow is in the direction of decreasing H.
Figure 6.9. Unsaturated hydraulic conductivity as a function of water potential

(left) and water content (right). As illustrated, Ks is less than the fully saturated
conductivity since soil naturally traps a small quantity of air within its pores.
For a saturated soil K is constant, but for regions of the soil which are only partially
saturated the hydraulic conductivity varies significantly with water content, K = K(θ).
Since θ is a function of h, we may also write K = K(h). Recall that H = h + z (Section
6.2.2) where z is the vertical distance from the datum. Then for flow in the vertical
direction,
⎛ dh ⎞ dh
q = − K ( h)⎜ + 1⎟ = − K ( h ) − K (h) (6.10)
⎝ dz ⎠ dz
We noted earlier (Section 6.2.3) that soils are usually not completely saturated in
nature because of air entrapment during the wetting process. Thus, even for apparently
saturated regions below the water table the volumetric water content may not be equal
to total porosity, but to θs, the water content at residual air saturation. The value of K
corresponding to θs is Ks (Figure 6.9, right), which may still be considered constant in
regions below the water table and is sometimes referred to as the apparent saturated
conductivity. Further discussions will assume Ks is effective saturated hydraulic con-
ductivity.
6.3.2 Measuring Saturated Hydraulic Conductivity
Various methods for measuring saturated hydraulic conductivity in the field have
been described in detail by Reynolds et al. (2002). There are methods suitable for soils
with high water tables, similar to well pumping methods for obtaining aquifer trans-
missivities, and other methods suitable for unsaturated soils. The reader is referred to
Reynolds et al. (2002) or to Bouwer and Jackson (1974) for details concerning these
methods. Again all of these methods provide measurement of Ks at a point, so, due to
field variability, numerous measurements may be required to obtain a field effective
Ks value. Further, some methods measure the approximate horizontal conductivity,
which may differ significantly from the vertical conductivity.
6.3.3 Unsaturated Hydraulic Conductivity
For unsaturated soils the water moves primarily in small pores and through films
located around and between solid particles. As the water content decreases, the cross-
sectional area of the films also decreases and the flow paths become more limited. The
result is a hydraulic conductivity function that decreases very rapidly with water con-
tent as shown schematically in Figure 6.9(right). Several functions for this relation
were given above in Table 6.2.
In most cases hysteresis in the K(θ) relationship is small. However, when K = K(h)
is used as in Equation 6.10, shown in Figure 6.9(left), hysteresis may be pronounced
due to hysteresis in the h(θ) relationship (Figure 6.4).
6.3.4 Measuring Unsaturated Hydraulic Conductivity
The measurement of unsaturated hydraulic conductivity is considerably more diffi-
cult than measuring saturated hydraulic conductivity. Since the K value is dependent
on water content, both the hydraulic gradient and water content or potential must be
determined for a range of water contents to adequately define the hydraulic conductiv-
ity function. Most of the reported measurements for unsaturated soils have been con-
ducted in the laboratory where boundary conditions can be carefully controlled and
soil water content and flow rates precisely measured. Field measurements have also
been reported but are much more difficult because of the number of variables that
must be measured and the variability of soil in the field.
A major problem with both field and laboratory methods for determining K(θ) is
the time and expense required. Measurement of a single K(θ) function by a well
trained technician may require several days. Furthermore, several measurements may
be needed to adequately characterize K(θ) for a given soil type because of field vari-
ability of the soil properties.
6.3.4.1 Laboratory methods. Clothier and Scotter (2002) describe steady state
methods for measuring K(θ) based on the defining relationship given in Equation 6.10.
Essentially the method consists of setting up boundary conditions to obtain steady,
one-directional flow for adjustable pressure heads. In one method, a soil sample is
placed in an airtight cavity between two horizontal porous plates through which water
flows into and out of the sample. The average pressure head in the sample is con-
trolled by the air pressure in the cavity. The mean hydraulic gradient between two
points in the sample is determined by using tensiometers to measure the difference in
pressure head. Alternatively, the potential and the potential gradient may be controlled
by changing the elevations of the water source and the outlet. In either case, by meas-
uring the steady state flow rate, q, the conductivity may be calculated directly from
Equation 6.9. Then the potential is changed and the procedure repeated for another
value of pressure head.
Although simple in concept, this method has some disadvantages. Since soil con-
ductivities are small in general, particularly at the lower water contents, long times are
required to approach steady flow, especially for imbibition. Also the conductivity
function obtained using the above method represents a point determination, or at most
a determination for a sampled soil section. In order to incorporate some of the hetero-
geneities of natural soils in the conductivity function, it is better if conductivity deter-
minations can be made on several rather large soil samples.
Transient methods utilize a controlled boundary condition with careful measure-
ments during water movement to infer conductivity relationships by optimization with
solution of Richards’ equations. A good example of such methods is described by
Hudson et al. (1996), in which upward flow into a cylindrical soil sample is caused by
a fixed flux at the lower boundary, and water content and tension measurements are
made with TDR and tensiometer methods, respectively.
6.3.4.2 Field methods. A method originally described by Nielsen et al. (1964) (also
called the instantaneous profile method) has been frequently used (e.g., Nielsen et al.,
1973; Cassel, 1974) to determine K(θ) and h(θ) for soils. Water is applied to saturate
the surface of a field plot approximately 2 to 4 m square. After a predetermined
amount of water has infiltrated, usually 50 to 150 mm, application is ceased and the
plot covered with plastic film to prevent evaporation from the surface. Changes in soil
water pressure head during subsequent redistribution of the infiltrated water are meas-
ured using tensiometers located at 150-mm depth increments to a total depth of about
1.5 m below the surface. Changes in the soil water content with time are inferred from
tensiometer readings and the soil water characteristics. The soil water characteristics
are obtained from cores taken at each tensiometer depth. The flux at a given depth is
calculated from changes in the soil water contents above that depth. Then the hydrau-
lic gradient is determined directly from tensiometer readings and the conductivity cal-
culated from Equation 6.10.
Tension infiltrometer or disc permeameter. Figure 6.10 schematically illustrates a
disk permeameter. Like the ring infiltrometer, water is applied to a small circular area,
but the water is applied under a small tension and no guard ring is used (Hussen and
Warrick, 1994). The mariotte siphon is an integral part of the device, controlling the
entry water potential, and the disk has a fine nylon mesh surface through which water
passes to the soil. Small irregularities in the undisturbed soil surface are dealt with by
a shallow layer of fine sand which forms a contact interface. The tension may be ad-
justed from 10 or 20 up to 150 mm. An alternative arrangement, using a ring with
minimum soil penetration, allows use of the same device for application at zero to
Nylon Screen Air Inlet
Figure 6.10. Schematic of a tension infiltrometer or disk permeameter. The column

at right controls the air pressure at the entrance to the supply tube at center, and
thus controls the water pressure at the disk/soil interface.
small positive heads. The radially symmetric flow under the permeameter is analyzed
by using the method of Wooding (1968). The method can obtain local estimates of Ks
for soils that approach steady flow quickly, and can obtain points on the K(h) curve
only for soils with low values of he. Discussion of application and data analysis for
this device is given in Smith et al. (2002).
Borehole permeameters. Similar to the disk permeameter, this method employs a
constant head point source at some depth in the soil, which may be the bottom of a
bore. The Guelph permeameter is one such device (Reynolds and Elrick, 2002), and
another has been introduced by Shani et al. (1987). These methods are used in con-
junction with analytic solutions for three-dimensional flow, and can estimate K(h) for
homogeneous soils and a specific K(h) relation.
6.3.4.3 Calculation of K(θ) from the soil water characteristic. To evade the dif-
ficulty of directly measuring K(θ), numerous attempts have been made to formulate a
computational scheme so that the partial or unsaturated hydraulic conductivity may be
estimated through the knowledge of other soil properties that are easier to measure.
Such properties should be representative of the geometry of pores and their distribu-
tion in space. Since the microscopic structure of a porous medium is too complicated
to deal with in exact mathematical terms, simplifying assumptions are necessary. Ko-
sugi et al. (2002) has presented a good review of these methods. The fundamental ap-
proach, as explained by Mualem (1974), is that the soil retention curve is an analog for
the distribution of pore radii that are effective in the flow of water at any overall water
content, and that the flow in those filled pores may be described by analogy to a lami-
nar Hagen-Poisouille formula. In terms of an effective pore radius, re, and an effective
area of flowing pores, ae, the general formula can be expressed as follows:
∫ re dae
2
filled _ pores
K r (θ) = (6.11)
∫ re 2 dae
all _ pores
where Kr is relative hydraulic conductivity, K/Ks. The relations for K(h) shown in Ta-
ble 6.2 for Brooks and Corey (1964) and van Genuchten (1980) come from applica-
tions of such formulas.
6.4 WATER MOVEMENT IN SOIL
6.4.1 Mass Balance and Flow Equations
The Darcy-Buckingham flux equations (6.9 or 6.10) are adequate for steady flow in
unsaturated soils, but steady flow is not the common condition. To describe the dy-
namics of soil water, these equations must be combined with an expression for dy-
namic mass balance to obtain what is commonly called Richards’ equation (Richards,
1931):
∂θ
= −∇ ⋅ q + e (6.12)
∂t
where q is the flux vector, ∇ ⋅ represents the divergence operation (generalized spatial
differentiation) and e is a gain/loss term representing root uptake (–), for example.
For flow in the vertical z direction only, Equation 6.12 may be written as
∂θ ∂q
=− +e (6.13)
∂t ∂z
This expression is combined with Equation 6.10 to obtain the more common form of
Richards’ equation:
∂θ ∂ ⎡ ∂ h⎤ ∂ K
= ⎢ K (h) ⎥ + +e (6.14)
∂t ∂z ⎣ ∂ z⎦ ∂ z
Water content θ may be made the independent variable by substituting the soil wa-
ter diffusivity, D(θ) ≡ K(h) dh/dθ. Using this variable the equation takes the form of a
nonlinear diffusion equation:
∂θ ∂ ⎡ ∂ θ⎤ ∂ K
= ⎢ D(θ ) ⎥ + +e (6.15)
∂t ∂z⎣ ∂ z⎦ ∂ z
These equations are nonlinear because of the functional dependence of the coeffi-
cients, D(θ) or K(h). Given the soil water capacity, C(h), defined above, the variables
D(θ) and K(h) are related by D = K/C. Both parameters vary markedly with water con-
tent or pressure head as discussed in Section 6.2.3. The significant nonlinearity of the
soil parameters is the prime source of difficulty in solving the equation for agricultural
water flow conditions.
Note that use of Equation 6.14 assumes that there is no resistance to soil air move-
ment and the air pressure remains constant throughout the profile. It is also usually
assumed that the soil matrix is rigid and does not change with time, so that the soil
water characteristic and hydraulic conductivity relationships are not time variant.
These assumptions do not always hold and may cause significant errors in predicted
results, as will be discussed below.
6.4.1.1 Two- and three-dimensional flow. In several types of irrigation, water is
delivered to a point, a line, or a confined region representing only a small fraction of
the total surface area. Examples include trickle and furrow irrigation, subsurface irri-
gation, and bubbler systems with a separate outlet for individual trees or shrubs. Once
the water enters the soil, the flow is governed by gravity and capillarity, just as for the
one-dimensional cases
Solutions to water flow equations in two and three dimensions have been found for
a variety of point and line source geometries. One of the simplest, but still useful, is
the solution for a steady point source that is “buried” in a uniform soil and assuming a
hydraulic conductivity to be an exponential of the pressure head (Table 6.2). The solu-
tion is (Philip, 1968; Warrick, 2003)
ϕb ( R, z ) =
K s exp(αh)
α
=
2
q
2 0 . 5
⎛ ⎡
⎣
(
exp⎜⎜ 0.5α ⎢ z − R 2 + z 2 )
0.5 ⎤ ⎞
⎥ ⎟⎟
⎦⎠
(6.16)
8π( R + z ) ⎝
where ϕb = matric flux potential
Ks = saturated conductivity
α = m from Table 6.2
h = pressure head
3
q = source strength [L /T]
z = depth below the source
R = cylindrical radius from the source.
For any value of z and R, Equation 6.16 can be used to find the pressure head h.
The pressure head distribution based on Equation 6.16 is shown as Figure 6.11a for
Ks = 15 mm h-1 and α = 0.0050 mm-1 after Or (1995). The wettest soil (h—> 0) is near
the source, i.e., at R = z =0. The contours away from the source, –0.6, –0.7, –0.8, are
somewhat elliptical shaped and reach a deeper depth than radius due to the influence
of gravity.
In Figure 6.11b results are calculated assuming a point source at the surface rather
than buried. This solution is
K s exp(αh)
α
= 2ϕb ( x, y ) −
αq exp(αz ) ⎛
4π
⎡
(
E1 ⎜⎜ 0.5α ⎢ z − R 2 + z 2 ⎥ ⎟⎟
⎣
)
0.5 ⎤ ⎞
⎦⎠
(6.17)
⎝
where ϕb(x,y) is calculated for the buried source from Equation 6.16 and E1 is the ex-
ponential integral function (e.g., Abramowitz and Stegun, 1964, Equation 5.1.1). The
wettest region is now on the surface near the source. The contours are similar to ellip-
ses descending into the profile, but with the tops truncated. For the deeper depths, the
pressure heads are very similar to those for the buried source. For example, at a depth
0.7 m below the source, the values of h in each case is between –0.7 and –0.8 with the
surface point source showing a slightly wetter result (i.e., closer to the –0.7 contour
than for the buried source).
For more complex geometries and time-dependent cases, it is generally necessary
to go to purely numerical solutions. Computer packages to do this have generally be-
come much more available and easier to use with increasing use of microcomputers.
For example, HYDRUS-2D (Simunek and van Genuchten, 1999) is widely used.
Radial Distance (m) Radial Distance (m)

0.0 0.2 0.4 0.0 0.2 0.4
0.3 0.0
Head (m)
0.1 -0.2
0.2
0.0 0.0
Depth (m)
-0.1 -0.4 -0.2

-0.4
-0.3 -0.6 -0.6
-0.8
-1.0
-0.5 -0.8
-1.2
A B
-0.7 -1.0
Figure 6.11. Steady-state distribution of soil water pressure head for
(a) buried and (b) surface point sources (after Or, 1995; Warrick, 2003).
6.4.2 The Infiltration Process

Infiltration, usually defined as the surface entry of water into the soil profile, is a
process of great practical importance to irrigation design. Failure to adequately con-
sider the infiltration process may result in nonuniform distribution of water in the field
as well as excessive water loss due to deep percolation or runoff. For surface irriga-
tion, the most efficient furrow or border length depends in part on the infiltration ca-
pacity. Many of the soil-related factors that control infiltration also govern soil water
movement and distribution during and after the infiltration process. Hence, an under-
standing of infiltration and the factors affecting it is important to the design and opera-
tion of efficient irrigation systems.
6.4.2.1 Infiltrability. Consider a hypothetical experiment with a water depth main-
tained on the surface to create infiltration into a deep, homogeneous soil column with
a uniform initial water content. The flux or the rate water enters the soil surface is
called the infiltration rate, f. If ponded water is maintained on the surface, we will find
that f decreases with time as shown schematically in Figure 6.12. This decrease is due
to reduction in the hydraulic gradients in the soil, but in nature may also be affected by
soil changes such as surface sealing and crusting. If the experiment is continued for a
sufficiently long time the infiltration rate will approach a constant rate, fz. For homo-
geneous soil profiles fz is for practical purposes equal to Ks, the hydraulic conductivity
at residual air saturation. For layered or surface-crusted profiles, fz is different than Ks,
as discussed below.
Since water is always ponded on the surface in our hypothetical experiment, the in-
filtration rate is limited only by soil-related factors. At any time during infiltration the
maximum rate that water will infiltrate, as limited by soil properties, has often been
called the infiltration capacity of the soil, fc. Hillel (1971) noted that the term capacity
is generally used to denote an amount or volume and can be misleading when applied
to a time-rate process. He proposed the term soil infiltrability rather than infiltration
capacity, and we employ that terminology here.
Figure 6.12. Time dependency of infiltration flux under two supply rates. Although
the amounts are equal, the lower rate succeeds in putting more water in the soil,
which is the area defined by r – f.
Now consider the same soil column as described above with irrigation water or
rainfall applied at a constant rate, r, to the surface. For this case the infiltration rate
will initially be equal to r and is limited by the application rate rather than the soil
head gradient. As long as the application rate is less than the infiltrability, the infiltra-
tion rate will be controlled by the application rate; f = r. However, if r is sufficiently
greater than fz or Ks, the infiltrability may become less than r after a period of time.
Then the infiltration rate will again be controlled by the soil profile and water will
pond on the surface. The infiltration patterns for two rates r are shown in Figure 6.12.
Water supplied in excess of the infiltrability will become available for surface storage
and/or runoff. Ponding or runoff cannot occur when r < Ks, unless by saturating the
soil above a restrictive layer or a high water table.
The infiltration rate is normally expressed in units of depth of water per unit time
(or volume per unit area per unit time), e.g., mm/h. Cumulative infiltration, I = I(t), is
the total amount of water infiltrated at any time t since the start of application, and
may be expressed as:
t
∫
I (t ) = f (t ) dt (6.18)
0
where f is the infiltration rate which may or may not be equal to infiltrability, as dis-
cussed above.
The soil water distribution during ponded infiltration into a uniform unsaturated
soil profile will be approximately as shown by the dotted line in Figure 6.13. Only the
surface will be saturated, and water content will decrease with depth toward a rela-
tively sharp front known as the wetting front. Viewed through the side of a glass con-
tainer, this wetting front appears to be a sharp line, but is in fact a transition not distin-
guishable by eye. The soil often will not be completely saturated behind the wetting
front, in some places, due to air entrapment and possible local counterflow of the air
phase.
Figure 6.13. Typical shapes of water after a uniform infiltration event,

as well as the shapes after a period of distribution.
One class of formulas to describe infiltration are those derived by using Darcy’s
law (Equation 6.10), plus an expression of mass balance at the soil surface, with ap-
propriate simplifying assumptions on the hydraulic soil characteristics. These are dis-
cussed in detail by Smith et al. (2002). The most well known of this class of infiltra-
tion equations is that of Green and Ampt (1911). This equation was first derived by a
more simple conceptual approach, and originally dealt with ponded upper boundary
conditions alone. It arises from Equation 6.15 by assuming that D(θ) is a step function,
constant within the wetting zone. This allows integration of Equation 6.15 to obtain:
Δθ (G + d ) + I
fc = K s (6.19)
I
0
where G = effective capillary drive: G = ∫ k r (h)dh (6.20)
−∞
d = depth of water on the surface
kr = relative conductivity, K(h)/Ks
Δθ = (θs – θi), which is the saturation deficit, or the available water storage in
the profile.
G is an important integral measure of the capillary suction of a soil. From its defini-
tion, one can see that it can be thought of as the kr-weighted mean value of capillary
potential, plus the surface water driving head. Generalized values of G for various soil
types have been presented by Woolhiser et al. (1990).
Note that in Equation 6.19 infiltrability is expressed in terms of infiltrated depth I.
While in many cases one would find it more convenient to have fc(t), the expression of
fc in terms of I is significant for two reasons. First, it unifies the calculation of infiltra-
bility for both sprinkling (or rainfall) and ponding conditions. Second, it allows calcu-
lation of ponding and subsequent infiltrability patterns with a single function. Higher
application rates cause ponding for smaller values of I, and vice versa. The ponding
depth, Ip, can be found by substituting r for f in Equation 6.19 and solving for I = Ip.
For subsequent times, as long as rate r > f, the infiltrability is described by this equa-
tion. Over the last decades several investigators have verified that ponding times can
be accurately modeled by relating rainfall rates to infiltrated depths (Mein and Larson,
1973; Reeves and Miller, 1975; Smith and Parlange, 1978).
Infiltration relations with physical meaning have been presented by Smith and Par-
lange (1978) and Philip (1957), as well as Green and Ampt (1911) (abbreviated G-A).
These are summarized in Table 6.4 in dimensionless terms using simple scaling rela-
tions. These scaling functions (Smith et al., 2002) are as follows:
f
f* = (6.21a)
Ks
I
I* = (6.21b)
GΔθ
tK s
t* = (6.21c)
GΔθ
The step function D(θ) assumption (G-A) is most appropriate for relatively coarse
soils and relatively uniform particle size distributions, and represents one limiting type
of soil behavior. At the other end of the spectrum of soil hydraulic behavior are well
graded soils which may exhibit roughly exponential behavior of D(θ) in the region
near saturation (Parlange et al., 1982). This assumption leads from Equation 6.18 to
the infiltration relation of Smith and Parlange (1978) (abbreviated S-P) (Table 6.4).
Philip (1957) was the first to deal with solutions to the Richards’ equation, and solved
Equation 6.19 for surface ponding conditions using a series expansion. He also pro-
duced an approximate infiltration equation, by truncating the series solution. Because
of the truncation it is not as accurate an approximation as others except at very small
or very large times. The Philip expression, using scaled terms defined above, is also
given in Table 6.4. These expressions are compared graphically in Figure 6.14.
The region between the S-P and the G-A expression is where the infiltration curves
for most soils, ideally, should lie. The relative error of the truncated series of Philip in
the intermediate values of I is apparent in Figure 6.14, but the expressions are gener-
ally in agreement. Since many irrigation applications are simple in terms of time varia-
tion, an expression of fc(to) (to being opportunity time) is often useful. For that pur-
Table 6.4. Scaled forms of several analytically derived infiltration capacity functions.
Base Infiltration Relationships
Equation fc*(t*) fc*(I*) I*(t*)
t* (1 − α ) = t* (1 − α ) =
Smith- α
Parlange 1 ⎛⎜ f c* − 1 + α ⎞⎟ ⎛ f c * ⎞ f c* = 1 +
⎜ ⎟ I * ) − 1 I − ln⎜
⎛ exp(αI * ) − 1 + α ⎞
ln⎜ ⎟ − ln⎜ ⎟
exp( α * ⎜ ⎟⎟
and α ⎝ f c* − 1 ⎠ ⎝ f c* − 1 ⎠ ⎝ α ⎠
Green-
Ampt[a] 1 2 I* + 1
Philip f c* = 1 + f c* = I * = t* + 2t*
2t* 2 I* + 1 − 1
[a]
0 <α < 1: Approaches a G-A relation for α near 0, and a S-P relation when α is near 1 (Parlange et al., 1982).
Figure 6.14. Log-log plots of three analytic infiltration functions

from Table 6.4, relating scaled infiltrability, fc*, to scaled infiltrated depth, I*.
Figure 6.15. Infiltrability for a few hypothetical soils of different textures. Infiltra-
bility for a soil of any texture can vary significantly from this idealized example.
pose, a rather simple time explicit expression which mimics either of the analytic
forms and uses the same parameters can be written (Smith et al., 2002):
1
f c* = β + + (1 − β ) 2 (6.22)
2to*
The weighting parameter β is best at approximately 1/3, and is 1.0 for the Philip
expression. Another similar form is given in Smith et al. (2002)
6.4.2.2 Application rates. For sprinkler irrigation, application rates should be
selected so that surface runoff does not occur, if possible. This is usually ensured by
choosing an application rate that is less than or equal to the steady state infiltration
capacity, fz. While this procedure will work for some cases, there are many situations
in which the steady final infiltration rate is prohibitively small. For example, a soil that
has a restricting layer below the surface may have a steady state infiltration rate of
nearly zero yet be capable of large intake rates during the first stages of infiltration.
The total amount of water to be applied at one irrigation is usually fixed by the crop
root depths and the soil hydraulic properties. Thus, the maximum application intensity
can be obtained from infiltration relations using plots such as Figure 6.15.
For example, let us assume that 20 mm of water are to be applied in a single irriga-
tion on a loam soil. Using appropriate parameters for this soil and the fc(I) relationship
in Figure 6.15, we can see that the scaled application rate should not exceed 50 mm/h
in order to prevent surface ponding and runoff. It is important to remember, however,
that the soil hydraulic properties are not stable in time in a field subject to cultivation,
and these relations will change in response to soil disturbance.
6.4.2.3 Effect of soil properties on infiltration. Infiltrability fc(I) for vertical infil-
tration from a ponded surface into deep homogeneous soil profiles are plotted in Fig-
ure 6.15 for four soils. Note that these are based on sampled mean properties, and any
given soil may vary considerably from the relation shown here. As expected, infiltra-
tion rates tend to increase with coarser soil texture: values of G decrease with coarse-
ness, and values of Ks increase significantly. However, soil structure and porosity are
also important, so it is not always possible to relate infiltration capacities to texture
alone just as the hydraulic conductivity does not always change directly with soil tex-
ture. Variations in either the diffusivity or soil-water characteristic at water contents
near saturation have a very strong influence on predicted infiltration. Therefore, errors
in measuring the soil hydraulic properties are of far greater consequence for water
contents near saturation than for drier conditions, so far as infiltration is concerned.
6.4.2.4 Effects of initial water content. As the infiltration equations above indi-
cate, initial water content θi affects the soil water storage capacity, and acts mathe-
matically in tandem with the effective capillary drive: an increase in initial wetness is
like a decrease in capillary drive. If infiltration is allowed to continue indefinitely, the
infiltration rate will eventually approach Ks regardless of the initial water content. In-
filtration rates are higher at low initial contents because of higher hydraulic gradients
and more available storage volume. Because of increased storage capacity, the wetting
front advances more slowly for lower θi. For very wet initial conditions, there are ap-
propriate modifications to the f(I) relation discussed by Corradini et al. (1994).
6.4.2.5 Approximate infiltration equations. The methods outlined above provide
physically consistent means of quantifying infiltration in terms of the soil properties
governing movement of water and air. Obtaining the necessary soil property data is
usually not simple. Variation of the soil properties, both with depth and from point to
point in the field, requires numerous measurements be made to adequately describe
field conditions. Present methods of determining the hydraulic properties are difficult
and such data are only available for a limited number of soils. Thus, while the predic-
tion methods discussed in this chapter are extremely valuable in analyzing the effects
of various factors on the infiltration process, many existing engineering approaches
use empirical equations. The original purpose of such equations, in the absence of a
theoretical equation, was to fit measured data in a simple manner. Now, the analytic
forms in Table 6.4 also allow data fitting, from which come parameters with more
physical meaning.
Several of the more popular empirical algebraic equations are given in Table 6.5.
Except for a parameter representing a minimum value of f, (fo), the parameters gener-
ally have no physical meaning and cannot be measured.
Table 6.5. Algebraic infiltration relations in common use.
Infiltration Name Formula Reference
Kostiakov fc = Kk t -c Kostiakov (1932)
Modified Kostiakov fc = fo + Kk t -c
Horton fc = fo + (fi – fo) e-ct Horton (1939)
6.4.3 Infiltration Measurements

Parr and Bertrand (1960) published a thorough review of field methods for measur-
ing infiltration capacity. Basically three types of devices may be used: sprinkling infil-
trometers, flooding infiltrometers, and disk permeameters. From the application point
of view, it might be advantageous to use a sprinkling infiltrometer if sprinkler irriga-
tion is to be used, while flooding infiltrometers may be more appropriate for soils that
are to be furrow or flood irrigated. However, the flooding devices are far more fre-
quently used because they require less equipment and are easier to install and to oper-
ate than the sprinkling type. With use of physically based infiltration equations, pa-
rameters are robust, and flooding test results can be applied to sprinkling irrigation
conditions.
6.4.3.1 Ring infiltrometers. The most commonly used infiltrometer is probably the
ring or cylindrical infiltrometer, which was described in detail by Haise et al. (1956)
and is shown schematically in Figure 6.16. It consists of a metal cylinder, 200 to 450
mm in diameter, which is pressed or driven into the soil. Infiltration is measured by
ponding water inside the cylinder and measuring the rate that the free surface falls, or
by measuring the rate that water must be added to maintain a constant ponded depth.
Once the wetting front exceeds the buried cylindrical depth, lateral flow may cause the
measured infiltration rates to be higher than would occur during irrigation. Lateral
flow is especially troublesome if restrictive or subsurface compacted layers exist or if
the hydraulic conductivity decreases with depth. When restrictive layers are at a shal-
low depth, it is recommended that the infiltration cylinder be driven into or through
the layer, if possible. A common problem with ring infiltrometers is that in driving
cylinders there is always some soil disturbance near the ring wall, which can signifi-
cantly affect the measurements. A means of preventing biased measurements due to
lateral soil water flow is to use a guarded ring or buffer area around the outside of the
infiltration cylinder, as shown in Figure 6.16. Water is ponded between the two cylin-
ders at all times to prevent edge effects and to maintain vertical flow below the central
infiltration cylinder. The Wooding (1968) analytical solutions for steady state infiltra-
tion from a circular ponded surface into a deep, homogeneous soil can be used to es-
timate the guard ring diameter, D, needed for deep soils. For general purposes, a ratio
of the cylinder diameters of two or greater (D/d ≥ 2) is recommended; however, a ratio
as small as 1.25 may suffice for coarse soils. The water levels in both the cylinders
should be nearly equal, and the level can be maintained by adjusting the water supply
from separate reservoirs or by using a mariotte siphon or float valves to automatically
control the inflow. If a tall cylinder of uniform cross-section is used as the supply res-
ervoir, the infiltration volume can be determined easily and accurately by simply re-
cording the water level in the reservoir.
Buffer Supply Infiltration
Reservoir Reservoir
Infiltration Cylinder
Buffer Cylinder
Figure 6.16. Schematic of the essential features of a traditional ring infiltrometer.

6.4.3.2 Tension infiltrometer or disc permeameter. This device was introduced

in discussions above (Section 6.3.4.2). Under tension the final infiltration rate meas-
urements will be lower than for ponded infiltration, but an alternative arrangement,
using a ring with minimum soil penetration, allows use of the same device for applica-
tion at zero to small positive depths. For many soils, final infiltration rates will be ap-
proximately steady within the time taken for the cylinder to empty, due to the three-
dimensional nature of the flow. However, these values will be considerably above the
value of Ks. Approximations for sorptivity S or capillary drive G may be made if care-
ful measurements are taken very early in the experiment, but accurate rate values dur-
ing this short period are considerably more difficult to measure.
6.4.3.3 Furrow infiltrometers. Infiltration under furrow irrigation involves soil
water movement in both vertical and lateral directions. Because the infiltrability rela-
tion depends on the size and shape of the furrow, the rate water moves into the soil is
often called intake rate rather than infiltration rate. Infiltration rates determined by
sprinkler or well buffered cylindrical infiltrometers can represent primarily vertical
flow, so it is difficult to apply these results directly to furrow intake. One method fre-
quently used to determine intake rates is to make inflow and outflow measurements in
an irrigation furrow. Furrow-sized measuring flumes are sometimes used to determine
intake rates by making flow measurements at two points in an irrigation furrow lo-
cated 30 to 90 m apart, and computing spatially averaged intake rates from the differ-
ence of inflow and outflow for various times after water application begins. A detailed
discussion of the use of the inflow-outflow method is given by Merriam (1968) (also
see Chapter 17). While this method provides a good means of evaluating furrow irri-
gation systems that are already installed, it is often not convenient to use this method
to determine intake rates for the design of a new system. Bondurant (1957) developed
a furrow infiltrometer to measure intake rates in a short section of an irrigation furrow.
The furrow is blocked off by metal plates and water is applied at a rate sufficient to
maintain a constant depth. The intake rate function and/or parameters are then deter-
mined in a manner similar to that described for the cylindrical infiltrometer. Water and
any sediment entrained may be recirculated to better simulate natural conditions.
6.4.3.4 Sprinkling infiltrometers. Many types of sprinkling infiltrometers were
discussed by Parr and Bertrand (1960). Sprinkling or spray infiltrometers usually con-
sist of a plot surrounded by partially buried sheet metal barriers with facilities for
measuring the rate of surface runoff. Water is sprinkled onto the plot surface at a con-
stant intensity and the infiltration rate is determined from recorded runoff measure-
ments. In most cases the infiltration rate is determined by simply subtracting the run-
off rate from the application intensity. However, the increase of surface storage during
the initial stages of runoff should also be considered. Failure to correct for surface
storage may result in significant errors in the early part of the measured infiltration
relationships (Smith, 1976).
A relatively simple method for measuring infiltrabilities for sprinkler irrigation de-
sign was developed by Tovey and Pair (1966). The general method consists of an area
covered by sprinklers on which the application rate varies over a significant range,
along with an array of rain gauges (uniform cans) to record total depth. The Tovey and
Pair method used a rotating sprinkler which produced intermittent application rates at
each position, with rates varying with radius position. If the application rates are uni-
form in time (except for intermittency), then the local rate at any gauge is the depth
divided by the time. The gauge located at a position just at the point dividing an area
of ponding (water appearing on the surface) from an area which accepts all the applied
water will presumably record the depth of water necessary to cause ponding at that
rate. The growing area of runoff should yield a set of data which will record the pond-
ing relation such as shown in Figure 6.14, except that the soil will usually exhibit
some spatial variability in its intake relations. Gauge depths at the time of ponding
should be taken for as many gauges as possible.
6.4.4 Soil Water Redistribution
After irrigation or rainfall, the wetted region at the top of the soil profile will redis-
tribute, with the water content slowly declining, and the depth of wetting slowly increas-
ing. An analytic solution for this process for a particular mathematical form of the soil
water characteristic was presented by Warrick et al. (1990). Alternatively, many soil
water profile models simulate this redistribution with finite difference numerical solu-
tions of Equation 6.14. Parkin et al. (1995) used a form of the Warrick et al. (1990)
drainage solution to characterize drainage of a very deeply wetted soil for many soil
types. Smith et al. (1993) and Corradini et al. (1994) developed a simplified method to
estimate the redistribution rate after shallow wetting, based on the same parameters ap-
pearing in the infiltration equations. The method calculates elongation of the profile us-
ing a modified form of the capillary drive, G, which reflects the fact that with the declin-
ing water content, the upper limit in Equation 6.20 is an algebraically decreasing poten-
tial. Errors in predicting surface θo were generally less than 5% after many hours. Figure
6.13, above, illustrates the general asymptotic pattern of soil water decay with time, and
demonstrates the approximate method of Corradini et al. (1994) and Smith et al. (1993).
6.5 COMPLICATING FACTORS
The discussion thus far has not dealt with most of the complexities of the soil me-
dium: soil layers, soil surface sealing, air entrapment by infiltrating water, and random
variation in soil properties. A brief discussion of each is warranted to clarify their re-
spective effects and the state of current knowledge.
6.5.1 Soil Layers
Soil layers or horizons have important effects on water profile infiltration, holding
characteristics, and redistribution. In comparison with a uniform soil, any profile dis-
continuity that affects pore size distribution, such as a textural or structural change,
will decrease water movement. The effect of soil layers on water retention and move-
ment was reviewed several years ago by Miller (1973).
When a coarser soil overlies a finer soil layer, water will move through the upper
soil faster than through the lower, fine-textured layer. Water will accumulate above
the layer and positive hydraulic pressures may develop. Unless an appreciable slope
exists, the water will remain in the soil until it is lost by slow drainage through the
substrata, and by evapotranspiration. Inasmuch as the water would drain away if the
fine-textured layer were not present, the finer layer greatly increases the water storage
capacity of the soil profile, but only up to a point. These soils must be irrigated care-
fully to prevent adverse effects of excess water and poor aeration. A slowly permeable
layer will transmit water slowly, but for a long period of time after irrigation. Thus,
with significant evapotranspiration rates, the effective available water will be much
higher than with lower ones because the plant can extract water while the soil is drain-
ing—water that would be lost as drainage if plants were not present.
If the discontinuity consists of a fine-textured soil layer above a coarser-textured

layer, the lower layer will not conduct significant amounts of water until many of the
pores are filled with water. This will occur only at potentials much higher than those at
which the pores in the soil above are filled. As the profile drains, the layer will stop
transmitting water at relatively high potentials, and the water content in the soil above
the layer will remain higher than if the layer were not present. This situation is often
called a capillary barrier effect, and is often used to control drainage through or from
waste storage sites.
The major items to consider in evaluating the influence of coarse-textured soil lay-
ers on water retention are (1) the saturated conductivity of the layer, (2) depth and
thickness of the layer, and (3) desorption characteristics of the soil (Miller, 1973). The
soil water potential may exhibit strong gradients in either a homogeneous or a layered
soil, but the potential will be continuous across an interface, and changes in water con-
tent and in hydraulic conductivity will occur at any layer interface.
6.5.1.1 Effect of layered soils on infiltration. For a coarse- over a fine-textured
soil, infiltration proceeds exactly as for a coarse-textured soil alone until the wetting
front arrives at the boundary between the two layers. Then the progress of the wetting
front is slowed, as the lower conductivity soil has its effect; and as the wetting front
enters the lower layer, the infiltration rate approaches that predicted for the fine-
textured soil alone. The infiltration rate-time relationship for fine- over coarse-textured
soils will also decrease when the wetting front reaches the coarser layer. The value of
fz for the case of only two layers was derived approximately by Moore (1981) and
more generally by Smith (1990). In short, it depends on the relative characteristics of
the K(h) relations for the two soils and the thickness of the upper layer, but it is always
lies somewhere between the Ks values for the two layers. Thus, a thinner upper layer
of finer texture will result in fz more like the lower soil, and a deep upper, finer texture
soil will result in an fz more like the surface layer. An approximate infiltration model
for a two-layered profile, not including crust development rate, was suggested by
Smith (1990).
6.5.1.2 Surface sealing and crusting. One special form of layering in soils is a
surface seal or crust. In the proceeding discussions we have assumed that the soil ma-
trix or skeleton is rigid and does not change with time. Actually, the hydraulic proper-
ties at the soil surface may change dramatically due to tillage disturbance, freezing and
thawing, and during the application of water. Such changes may have a stronger influ-
ence on the rate of infiltration than some of the other factors discussed. Indeed, in
some of the early studies of infiltration, the “exponential” decay of infiltration rate
with time was entirely attributed to slaking of aggregates and swelling of colloids
which progressively sealed the soil (Horton, 1940).
Because of the complex nature of the sealing process and the difficulty of describ-
ing the manner in which hydraulic properties of the surface layer change with time,
there have been few attempts to use the theory of soil-water movement to analyze the
phenomena. However, the results of numerous experimental investigations indicating
the importance of surface sealing have been reported, for example, Duley (1939), Hor-
ton (1940), Mannering and Meyer (1963), Burwell and Larson (1969), Moldenhauer
and Kemper (1969), and Koon et al. (1970). Mualem and Assouline (1989) have pro-
posed and analyzed a crust or surface layer that is transitional rather than a strict layer,
and therefore more like what one would expect to develop in soil crusting. Their
analysis is limited to steady flow rather than the dynamic flux of an infiltration event.
Clearly, surface sealing and crusting has a significant, perhaps dominant, effect on
infiltration in bare or unprotected soils. This factor is of less importance when the sur-
face is protected by a crop canopy or mulch. However, it must be considered in the de-
sign of irrigation systems and in selecting methods for measuring infiltration capacities.
6.5.2 Effects of Air Entrapment and Counterflow
We noted in Section 6.4.1 that the derivation of the Richards’ equation assumes
that displaced soil air moves through the profile with negligible resistance and that the
air pressure remains constant throughout. This assumption is usually justified by the
small viscosity of air relative to that of water and by assuming that air can escape ei-
ther downward or upward through large pores that remain partially open during infil-
tration. While these assumptions may hold in many instances, there are numerous
cases where air is trapped by infiltrating water causing an air pressure buildup in ad-
vance of the wetting front and a reduction of the infiltration rate. Even for deep pro-
files, pressure buildup and subsequent counterflow of escaping air causes infiltration
rates to be less than would occur if resistance to air movement was truly negligible.
The fact that air movement may significantly affect infiltration has been recognized
for many years (e.g., Free and Palmer, 1940). Studies showing the effects of air pres-
sure buildup and the flow of air on the infiltration process have been conducted by
Wilson and Luthin (1963), Peck (1965) and Adrian and Franzini (1966). A particularly
detailed study of this phenomenon with (water-like) oil as the infiltrating liquid was
reported by McWhorter (1971). He conducted infiltration experiments for conditions
representing both semi-infinite and finite profile depths, and presented analytical
methods for predicting the effects of two-phase flow. Methods for characterizing the
effects of air movement on infiltration in terms of soil properties and boundary condi-
tions are summarized in a detailed treatment of two-phase flow in porous media by
Morel-Seytoux (1973).
An example of the effect of air movement on infiltration as measured by
McWhorter (1971) for a 2.33-m sand column is shown in Figure 6.17. When the air
phase is neglected, the infiltration rate predicted by solution of Richards’ equation
asymptotically approaches Ks, the hydraulic conductivity at residual saturation, as dis-
cussed in Section 6.4.1. However, when air is entrapped between the wetting front and
the water table or a restrictive layer, air pressure increase causes a rapid reduction in
infiltration rate. As air pressure increases, upward flow of the air phase begins, often
followed by escape of air from the surface and an increase in the infiltration rate. This
predicted infiltration rate also asymptotically approaches a constant value but this
value may be significantly less than Ks. As an example of the effect of the air phase,
McWhorter’s data for a 9.9-m long, closed column of Poudre sand shows that the
long-term infiltration rate asymptotically approaches a value that is approximately
60% of Ks. The shorter, 2.33-m column also approaches this value, but exhibits an air
escape event during the experiment, resulting in increase in f and decrease in air pres-
sure. Intermediate length closed columns (not shown here) showed higher air pressure
buildup and consequently lower asymptotic rates of f. Examples presented by Morel-
Seytoux (1973) predicted steady state infiltration rates of approximately 0.3 Ks for
soils with shallow water tables and between 0.8 Ks and 0.9 Ks for infinitely deep pro-
files.
10 300
9.9m closed column
air pressure head, mm(oil)

250
infiltration flux, mm/min.

8
200
6
air pressure head in soil
150
air-free infiltrability
4 measured infiltration rate 100
2 50
0 0
0 20 40 60 80 100
Time, min.
10 500
2.3m closed column
air pressure head, mm(oil)

8 400
infiltration flux, mm/min.
air pressure head in soil
air-free infiltrability
measured infiltration rate
6 300
4 200
2 100
0 0
0 20 40 60 80 100
Time, min.
Figure 6.17. The data of McWhorter (1971) demonstrate the effect that a
sealed bottom of a column of soil can have on infiltrability. The longer column
exhibits far less effect than the shorter one.
Morel-Seytoux and Khanji (1974) showed that the form of Equation 6.19 remains
the same when the simultaneous movement of both water and air are considered. The
resistance to air movement was accounted for by the introduction of a viscous resis-
tance correction factor, β a, (1 < β a < 2), which was defined as a function of the soil
and fluid properties, and applied as a divisor to Ks. Such a correction may be applied
to any of the equations in Table 6.3.
6.5.3 Heterogeneity
While all the above methodology is appropriate at any point on a field where soil
properties are measured and input rates are known, even the most uniform soil exhibits
variations over a field area. This is part of the reality which has led to current studies
of site-specific management and micromanagement in irrigation and chemical applica-
tion in farm practice. The challenge in dealing with large irrigated areas is to deal in a
reasonable manner with this variation. This is often called a problem in scaling up;
dealing with scales or areas larger than a sample or a plot is mostly a problem in deal-
ing with the heterogeneity encompassed by such larger scales.
Over the past few decades, there have been numerous theoretical studies of the sta-
tistics and behavior of an area composed of soils that have specific kinds of variability.
Considering the number of parameters which can and do exhibit variability, the prob-
lem is quite complex, and most studies have limited themselves to the variation of one
parameter, usually the variation of Ks (e.g., Dagan and Bresler, 1983). One type of
analysis is the treatment of a large area just as one would treat a group of small sam-
ples, to determine how the large area composed of internal variation of some assumed
or measured nature affects the large-scale behavior. This may be termed an ensemble
approach, and can be applied to soil water redistribution (Dagan and Bresler, 1983) as
well as infiltration (Smith and Hebbert, 1979; Chen et al., 1994). Other studies (Proto-
papas and Bras, 1990) have assumed parameter variation in the form of gaussian per-
turbation around a mean value, but are limited by some linearizing assumptions. For
surface infiltration, the above studies are limited because they neglect the surface run
on/runoff possibility; that is, areas with low infiltrability can contribute runoff which
adds to the input for adjacent areas of higher infiltrability. Field measurements with
sampling and dye tracing show very deep movement of surface-applied water in a very
few sites, such as old root channels and structural fractures (e.g., Kung, 1990).
In the above discussion we have considered movement of water and air in a homo-
geneous soil or in layers of homogeneous materials. However, natural soils are seldom
homogeneous and often are permeated, especially in the surface layers, by relatively
large channels formed by roots, cracking due to shrinkage, and worm holes. Obviously
such channels would have a great effect on infiltration as they would provide both
pathways for rapid inflow of water and an escape route for air as the wet front ad-
vances. Preferential flow is the name often applied to cases where water follows paths
of high conductivity existing in a soil profile. When such flow follows noticeable cracks
or tubes (such as worm-holes) these may be termed macropores. The effect of macro-
pores on infiltration has been studied by Germann and Beven (1985) and others. Large
pores, open to the soil surface, can contribute greatly to infiltration after surface ponding,
in some cases raising total infiltration by a factor of 10 or more. However, small soil air
pressures can block this contribution so large pores near to the surface, but not open,
have a much smaller effect on infiltration. Thus, infiltration can be increased by using
cultural practices designed to prevent the sealing of large pores near the surface.
Acknowledgements
This chapter is a heavily revised version of a similar chapter in the previous edition
of this monograph, written by R. W. Skaggs, D. E. Miller, and R. H. Brooks, and in
some places we have borrowed heavily, and gratefully, from their material.
LIST OF SYMBOLS
cm specific heat of soil mass
cw specific heat of soil water
C(h) specific capacity of soil, dh/dθ
d depth of surface water
f soil surface infiltration rate
fc soil surface infiltrability
g gravitational constant
G soil capillary length scale
h soil water capillary pressure head
hg soil water gravitational potential
hp soil water pressure potential
hT soil water total potential
H total soil water head, h + z

I cumulated infiltrated depth
Ip value of I when soil ponds under high value of r
kr relative hydraulic conductivity, K/Ks
K soil hydraulic conductivity
Ks saturated hydraulic conductivity, K(h) = 0
m soil retention relation parameter
n soil retention relation parameter
q soil flux
r Surface application flux value, sprinkler application
R radial distance from point source irrigation
vs velocity of soil electromagnetic wave
α soil capillary head scaling paramter
β parameter in the f(t) relation
βa air correction coefficient
θ volumetric water content of soil
θi initial value of θ, at beginning of irrigation
θr soil residual water content
θs saturated water content, value of θ at h = 0
θw water content of soil by weight
Θ scaled water content
κ soil dielectric constant
ρw mass density of water
ρb bulk density of soil
ρc volumetric heat capacity of soil
φ matric flux potential
REFERENCES
Abramowitz, M., and I. A. Stegun, eds. 1964. Handbook of Mathematical Functions.
Washington, D.C.: U.S. Government Printing Office.
Adrian, D. D., and J. B. Franzini. 1966. Impedance to infiltration by pressure build-up
ahead of the wetting front. J. Geophys. Res. 71(24): 5857-5863.
Andraski, B. J., and B. R. Scanlon. 2002. Thermocouple psychrometry. In Methods of
Soil Analysis, Part 4: Physical Methods, 609-642. J. H. Dane, and G. C. Topp, eds.
Madison, Wis.: Soil Sci. Soc. America.
Bondurant, J. A. 1957. Developing a furrow infiltrometer. Agric. Eng. 38: 602-604.
Bouwer, H., and R. D. Jackson. 1974. Determining soil properties. In Drainage for
Agriculture, 611-672. J. van Schilfgaarde, ed. Madison, Wis.: American Soc.
Agronomy
Bristow, K. L., G. S. Campbell, and K. Calissendorf. 1993. Test of a heat-pulse probe
for measuring changes in soil water content. Soil Sci. Soc. Amer. J. 57(4): 930:934.
Broadbridge, P., and I. White. 1988. Constant rate rainfall infiltration: A versatile,
nonlinear model. I. Analytic solution. Water Resources Res. 24(1): 145-154.
Brooks, R. H., and A. T. Corey. 1964. Hydraulic properties of porous media. Hydrol.
Paper No. 3. Fort Collins, Colo.: Colorado State Univ.
Buckingham, E. R. 1907. Studies on the movement of soil moisture. U.S. Dept. Agr.
Bureau of Soils Bulletin 38. Washington, D.C.: USDA.
Burwell, R. E., and W. E. Larson. 1969. Infiltration as influenced by tillage-induced
random roughness and pore space. Soil Sci. Soc. Am. Proc. 33: 449-452.
Cassel, D. K. 1974. In situ unsaturated hydraulic conductivity for selected North
Dakota soils. Bulletin No 494. Fargo, N.D.: Agr. Exp. Sta., N.D. State Univ.
Chen, Z., R. S. Govindaraju, and M. L. Kavvas. 1994. Spatial averging of unsaturated
flow equations under infiltration conditions over areally heterogeneous fields 1.
Development of models. Water Resources Res. 30(2): 523-533.
Clothier, B., and D. Scotter, 2002. Unsaturated water transmission parameters
obtained from infiltration. Section 3.5 in Methods of Soil Analysis, Part 4: Physical
Methods, 879-898. J. H. Dane, and G. C. Topp, eds. Madison, Wis.: Soil Sci. Soc.
America.
Corey, A. T., and A. Klute. 1985. Application of the potential concept to soil water
equillibrium and transport. Soil Sci. Soc. America J. 49(1): 3-11.
Corradini, C., F. Melone, and R. E. Smith. 1994. Modeling infiltration during complex
rainfall sequences. Water Resources Res. 30(10): 2777-2784.
Dagan, G., and E. Bresler. 1983. Unsaturated flow in spatially variable fields 1.
Derivation of models of infiltration and redistribution. Water Resources Res. 19(2):
413-420.
Dane, J .H., and G. C. Topp, eds. 2002. Methods of Soil Analysis, Part 4: Physical
Methods. Madison, Wis.: Soil Sci. Soc. America.
Duley, F. L. 1939. Surface factors affecting the rate of intake of water by soils. Soil
Sci. Soc. America Proc. 4: 60-64.
Free, J. R., and V. J. Palmer. 1940. Relationship of infiltration air movement and pore
size in graded silica sand. Soil Sci. Soc. America Proc. 5: 390-398.
Gardner, W. R. 1958. Some steady-state solutions of the unsaturated moisture flow
equation with application to evaporation from a water table. Soil Sci. 85: 228-232.
Gardner, W. R. 1960. Dynamic aspects of water availability to plants. Soil Sci. 89: 63-
73.
Gardner, W. H., D. Hillel, and Y. Benyamini. 1970. Post-irrigation movement of soil
water. 1: Redistribution. Water Resources Res. 6(3): 851-861.
Germann, P. F., and K. Beven. 1985. Kinematic wave approximation to infiltration
into soils with sorbing macropores. Water Resources Res. 21(7): 990-996.
Green, W. H., and G. Ampt. 1911. Studies of soil physics. Part I: The flow of air and
water through soils. J. Agr. Sci. 4: 1-24.
Grossman, R. B., and T. G. Reinsch. 2002. Bulk density and linear extensibility.
Section 2.1 in Methods of Soil Analysis, Part 4: Physical Methods, 201-228. J. H.
Dane, and G. C. Topp, eds. Madison, Wis.: Soil Sci. Soc. America.
Haise, H. R., W. W. Donnan, J. T. Phelan, L. F. Lawhon and D. G. Shockley, 1956.
The use of cylinder infiltrometers to determine the intake characteristics of irrigated
soils. USDA Publ. ARS 41-7. Washington, D.C.: USDA.
Hillel, D. 1971. Soil and Water: Physical Principles and Processes. New York, N.Y.:
Academic Press.
Horton, R. E. 1939. Analysis of runoff plot experiments with varying infiltration
capacity. Trans. American Geophys. Union Part IV: 693-694.
Horton, R. E. 1940. An approach toward a physical interpretation of infiltration-
capacity. Soil Sci. Soc. America Proc. 5: 399-417.
Horton, R., P. J. Wierenga, and D. R. Nielsen. 1982. A rapid technique for obtaining
uniform water content distributions in unsaturated soil columns. Soil Science 133:
397-399.
Hudson, D. B., P. J. Wierenga, and R. G. Hills. 1996. Unsaturated hydraulic properties
from upward flow into soil cores. Soil Sci. Soc. America J. 60: 388-396.
Hussen, A. A., and A. W. Warrick. 1994. Tension infiltrometers for measurement of
vadose zone hydraulic properties. In Handbook of Vadose Zone Characterization
and Monitoring, 189-201. L. G. Wilson, L.G. Everett, and S. J. Cullen, eds. Boca
Raton, Fla.: Lewis Publishers.
Jury, W. A., W. R. Gardner, and W. H. Gardner, 1991. Soil Physics. 5th ed. New
York, N.Y.: John Wiley & Sons.
Koon, J. L., J. G. Hendrick, and R. E. Hermanson. 1970. Some effects of surface cover
geometry on infiltration rate. Water Resources Res. 6: 246-253.
Kostiakov, A. N. 1932. On the dynamics of the coefficient of water-percolation in
soils and on the necessity for studying it from a dynamic point of view for purposes
of amelioration. Trans. 6th Comm. Internl. Soil. Sci. Soc., Russian Part A: 17-21.
Kosugi, K., J. W. Hopmans, and J. H. Dane, 2002. Parametric models. Section 3.3.4 in
Methods of Soil Analysis, Part 4: Physical Methods, 739-757. J. H. Dane, and G. C.
Topp, eds. Madison, Wis.: Soil Sci. Soc. America.
Kung, K-J.S. 1990. Preferential flow in a sandy vadose zone. 1: Field observations.
Geoderma 46(1-3): 51-58.
Mannering, J. V., and L. D. Meyer. 1963. The effects of various rates of surface mulch
on infiltration and erosion. Soil Sci. Soc. America Proc. 27: 84-86.
McWhorter, D. B. 1971. Infiltration affected by flow of air. Hydrol. Paper No. 49.
Fort Collins, Colo.: Colorado State Univ.
Mein, R. G., and C. L. Larson. 1973. Modeling infiltration during a steady rain. Water
Resources Res. 9(2): 384-394.
Merriam J. L. 1968. Irrigation System Evaluation and Improvement. San Luis Obispo,
Calif.: Blake Printing.
Miller, D. E. 1973. Water retention and flow in layered soil profiles. In Field Soil
Water Regime, 107-117. R. R. Bruce et al, eds. Madison, Wis.: Soil Sci. Soc.
America.
Moldenhauer, W. C., and W. D. Kemper, 1969. Interdependence of water drop energy
and cold size on infiltration and clod stability. Soil Sci. Soc. America Proc. 33: 297-
301.
Moore, I. D. 1981. Infiltration equations modified for surface effects. J. Irrig. Drain.
Div., ASCE 107(IR1): 71-86.
Morel-Seytoux, H. J. 1973. Two-phase flows in porous media. Advances in
Hydroscience 9: 119-202.
Morel-Seytoux, H. J., and J. Khanji. 1974. Derivation of an equation of infiltration.
Water Resources Res. 10(4): 795-800.
Mualem, Y. 1974. A conceptual model of hysteresis. Water Resources Res. 10: 514-
520.
Mualem, Y., and S. Assouline. 1989. Modeling soil seal as a nonuniform layer. Water
Resources Res. 25(10): 2101-2108.
Nielsen, D. R., J. M. Davidson, J. W. Biggar, and R. J. Miller. 1964. Water movement
through Panoche clay loam soil. Hilgardia 35: 491-506.
Nielsen, D. R., R. D. Jackson, J. W. Cary, and D. D. Evans, eds. 1972. Soil Water.
Madison, Wis.: American Soc. Agronomy.
Nielsen, D. R., J. W. Biggar, and K. T. Erh. 1973. Spatial variability of field measured
soil water properties. Hilgardia 42(7): 215-260.
Or, D. 1995. Statistical analysis of soil water monitoring for drip irrigation
management in heterogeneous soils. Soil Sci. Soc. America J. 59: 1222-1233.
Parkin, G. W., A. W. Warrick, D. E. Elrick, and R. G. Kachanoski. 1995. Analytical
solution for one-dimensional drainage: water stored in a fixed depth. Water
Resources Res. 31(5): 1267-1271.
Parlange, J-Y., I. Lisle, R. D. Braddock, and R. E. Smith. 1982. The three-parameter
infiltration equation. Soil Science 133(6): 337-341.
Parr, J. F., and A. R. Bertrand. 1960. Water infiltration into soils. Advances in Agron.
12: 311-363.
Peck, A. J. 1965. Moisture profile development and air compression during water
uptake by bounded porous bodies. 3: Vertical columns. Soil Sci. 100(1): 44-51.
Philip, J. R. 1957. The theory of infiltration. 1: The infiltration equation and its
solution. Soil Sci. 83: 435-448.
Philip, J. R. 1968. Steady infiltration from buried point sources and spherical cavities.
Water Resources Res. 4: 1039-1047.
Protopapas, A. L., and R. L. Bras. 1990. Uncertainty propagation with numerical
models for flow and transport in the unsaturated zone. Water Resources Res.
26(10): 2463-2474.
Rawlins, S. L. 1976. Measurement of water content and the state of water in soils. In
Water Deficits and Plant Growth, IV: 1-55. T. T. Koslowski, ed. New York, N.Y.:
Academic Press.
Reeves, M., and E. E. Miller. 1975. Estimating infiltration for erratic rainfall. Water
Resources Res. 11(1): 102-110.
Reginato, R. J., and C. H. M. Van Bavel. 1964. Soil water measurement with gamma
attenuation. Soil Sci. America Proc. 28: 721-724.
Reynolds, W. D., and D. E. Elrick, 2002. Constant head well permeameter (vadose
zone). Section 3.4.3.3 in Methods of Soil Analysis, Part 4: Physical Methods, 844-
837. J. H. Dane, and G. C. Topp, eds. Madison, Wis.: Soil Sci. Soc. America.
Reynolds, W. D., D. E. Elrick, E. G. Youngs, A. Amoozegar, H. W. G. Booltink, and
J. Bouma, 2002. Saturated and field-saturated water flow parameters. Section 3.4 in
Methods of Soil Analysis, Part 4: Physical Methods, 802-817. J. H. Dane, and G. C.
Topp, eds. Madison, Wis.: Soil Sci. Soc. America.
Richards, L. A. 1931. Capillary conduction of liquids in porous mediums. Physics 1:
318-333.
Richards, L. A., and C. H. Wadleigh. 1952. Soil water and plant growth. In Soil
Physical Conditions and Plant Growth, 73-251. B. T. Shaw, ed. Monog. 2.
Madison, Wis.: American Soc. Agronomy.
Romano, N, and A. Santini, 2002. Water retention and storage: Field. In Methods of
Soil Analysis, Part 4: Physical Methods, 721-738. J. H. Dane, and G. C. Topp, eds.
Sartzendruber, D. 1969. The flow of water in unsaturated soils. In Flow through
Porous Media, 215-292. R. M. DeWiest, ed. New York, N.Y.: Academic Press.
Shani, U., R. J. Hanks, E. Bresler, and C. A. S. Oliviera. 1987. Field method for
estimating hydraulic conductivity and matric potential-water content relations. Soil
Sci. Soc. America J. 51: 298-302.
Simunek, J., M. Sejna, and M. Th. van Genuchten. 1999. The HYDRUS-2D software
package for simulating the two-dimensional movement of water, heat, and multiple
solutes in variably-saturated media. Version 2.0. Riverside, Calif.: USDA-ARS
Salinity Lab.
Smith, R. E. 1976. Approximations for vertical infiltration rate patterns. Trans. ASAE
19(3): 505-509.
Smith, R. E. 1990. Analysis of infiltration through a two-layer soil profile. Soil Sci.
Soc. America J. 54(5): 1219-1227.
Smith, R. E., and R. H. B. Hebbert. 1979. A Monte Carlo analysis of the hydrologic
effects of spatial variability. Water Resources Res. 15(2): 419-429.
Smith, R. E., and J-Y. Parlange. 1978. A parameter efficient hydrologic infiltration
model. Water Resources Res. 14(3): 533-538.
Smith, R. E., C. Corradini, and F. Melone. 1993. Modeling infiltration for multistorm
runoff events. Water Resources Res. 29(1): 133-144.
Smith, R. E., K. R. J Smettem, P. Broadbridge, and D. A. Woolhiser. 2002. Infiltration
theory for hydrologic applications. Water Resources Monograph 15. Washington,
D.C.: American Geophysical Union.
Topp, G. C., J. L. Davis, and A. P. Annan. 1980. Electromagnetic determination of
soil water content: Measurements in coaxial transmission lines. Water Resources
Res. 16(2): 574-582.
Topp, G. C., and P. A. (Ty) Ferré, 2002. Water content, Section 3.1. In Methods of
Soil Analysis. Part 4: Physical Methods, 422-428. J. H. Dane, and G. C. Topp, eds.
Tovey, R., and C. H. Pair, 1966. Measurement of intake rate for sprinkler irrigation
design. Trans. ASAE 9: 359-363.
U.S. Bureau of Mines. 1994. Time domain reflectometry in environmental,
infrastructure and mining applications. In Proc. of Symposium sponsored by
Infrastructure Technology Institute at Northwestern Univ. and Los Alamos
National Lab.
van Genuchten, M. T. 1980. A closed-form equation for predicting the hydraulic
conductivity of unsaturated soils. Soil Sci. Soc. America J. 44(5): 892-898.
Warrick, A. W., D. O. Lomen, and A. Islas. 1990. An analytical solution to Richards’
equation for a draining soil profile. Water Resources Res. 26(2): 253-258.
Warrick, A. W. 2003. Soil Water Dynamics. New York, N.Y.: Oxford Univ. Press.
Watson, K. K. 1967. A recording field tensiometer with rapid response characteristics.
J. Hydrol. 5: 33-39.
Wilson, L. G., and J. N. Luthin. 1963. Effect of air flow ahead of the wetting front on
infiltration. Soil. Sci. 96(2): 136-143.
Wooding, R. 1968. Steady infiltration from a shallow circular pond. Water Resources
Res. 4: 1259-1273.
Woolhiser, D. A., R. E. Smith, and D. C. Goodrich. 1990. KINEROS, a Kinematic
Runoff and Erosion Model: Documentation and User Manual. U.S. Dept. of
Agriculture, Agri. Res. Service, ARS-77. Washington, D.C.: USDA-ARS.
Young, M. H., and J. B. Sisson. 2002. Tensiometry. In Methods of Soil Analysis. Part
4: Physical Methods, 575-608. J. H. Dane, and G. C. Topp, eds. Madison, Wis.:
Soil Sci. Soc. America.
CHAPTER 7
CONTROLLING SALINITY
Glenn J. Hoffman (University of
Nebraska, Lincoln, Nebraska)
Joseph Shalhevet (Agricultural Research
Organization, Bet Dagan, Israel)
Abstract. Strategies are presented to minimize the detrimental impacts of salinity,
sodicity, and toxicity in irrigated agriculture. The tolerances of major agricultural
crops to salinity are given along with the amount of leaching required to prevent crop
yield loss. The differences among irrigation systems and the affect of the quantity and
salt concentration of the applied water on the distribution and concentration of soil
salinity are also presented. The chapter concludes with discussions on the application
methods and the amount of applied water required to reclaim soils too high in salt
concentration to produce economical crop yield.
Keywords. Crop salt tolerance, Leaching, Reclamation, Salinity, Sodicity, Toxicity.
7.1 INTRODUCTION
A major challenge to irrigated agriculture is the hazard of salt-affected soils and
waters. Saline irrigation and drainage waters and saline and sodic soils are threats to
sustaining irrigated agriculture. Salinity is of more concern in arid than humid regions,
where precipitation is a major source of water for crop production, but it will become
more prevalent in humid regions as supplemental irrigation or drainage problems in-
crease.
Except for extremely old soils, salts from weathering of rocks and minerals seldom
accumulate in sufficient quantities to form a saline soil. Salt-affected soils usually
develop in areas that receive and accumulate salts transported from other locations by
water. Accumulations can also occur by atmospheric deposition of oceanic salts along
coastal areas; by seawater intrusion from estuaries and coastal groundwater basins,
inland saline lakes and playas; by upward movement of saline waters from groundwa-
ter; and from leaching of saline lands. Sources of salt from agriculture include irriga-
tion return flows, drainage waters, soil and irrigation amendments, animal manures
and wastes, fertilizers, and sewage sludges and effluents.
Soluble salts are present in all natural waters, and it is their concentration and com-
position that determine the suitability of soils and waters for crop production. Water
quality is normally based on three criteria: (1) salinity, (2) sodicity, and (3) toxicity.
Salinity is the osmotic stress caused by dissolved salts on crop growth. When the pro-
portion of sodium compared to other cations becomes excessive, soil structure deterio-
rates and the soil is said to be sodic. Toxicity encompasses the effects of specific sol-
utes that damage plant tissue or cause an imbalance in plant nutrition.
Soils and waters have no inherent quality independent of the specific conditions in
question. Thus, soils and waters can only be evaluated fully in the context of a speci-
fied set of conditions. A meaningful assessment of water quality for irrigation should
consider such site-specific factors as rainfall, irrigation management practices, cli-
mate, chemical and physical properties of the soil, and the chemical reactivity of con-
stituents dissolved in the water.
This chapter describes the influence of soil salinity on crop production, introduces
management techniques to cope with saline soils and waters, discusses the impact of
the type, design, and movement of an irrigation system on salinity control, and pre-
scribes methods to reclaim salt-affected soils. The chapter also addresses methods to
assess and reclaim sodic soils and briefly discusses toxic constituents that may be pre-
sent in soils or irrigation waters.
7.1.1 The Scope of the Salinity Hazard
Land occupies about 134 million km2 of the earth’s surface. Of that total, not more
than 70 million km2 are arable and only 15 million km2 are cultivated (Massoud,
1981). Of the cultivated lands, about 3.4 million km2 (23%) are saline and another 5.6
million km2 (37%) are sodic. The amount of salt-affected land is summarized in Table
7.1. Salt-affected soils cover about 13% of the total arable lands and occur in over 100
countries. Figure 7.1 shows the distribution of saline and sodic soils worldwide. Note
that lands north or south of about the 50° latitude are not salt-affected because precipi-
tation is too high and/or irrigation is not practiced.
Table 7.1. Area of salt-affected soils worldwide (from Szabolcs, 1989).
Salt-Affected
Region Land (hectares)
North America 15,800,000
Mexico and Central America 2,000,000
South America 129,200,000
Europe 50,800,000
Africa 80,500,000
South Asia 87,600,000
North and Central Asia 211,700,000
Southeast Asia 20,000,000
Australia and Pacific Islands 357,400,000
Worldwide 955,000,000
Figure 7.1. Global distribution of salt-affected soils (adapted from Szabolcs, 1989).
162 Chapter 7 Controlling Salinity
Figure 7.2. Potential water and soil salinity problems in the U.S.
(adapted from USDA, 1988).
The potential for salinity hazards in the conterminous United States is illustrated in
Figure 7.2. The map is based upon indicators of salinity in surface soils and geologic
formations on a scale of river basins or watersheds. About 30% of the land in the con-
terminous United States has a moderate to high potential for salinity. Western portions
of the United States, because of geologic formations and dry climates, are highly sus-
ceptible to salinity problems.
7.1.2 Overview of Salinity Problems
7.1.2.1 Historical perspective. Excess salinity, which is frequently the result of in-
adequate drainage, has plagued irrigated agriculture for millennia. A well-documented
case occurred in ancient Mesopotamia (now Iraq) about 4000 years ago. Cities in the
river valleys of the Tigris and Euphrates flourished in the 4th millennium B.C. The
development of these cities was based on the irrigation of wheat. Flooding, seepage,
overirrigation, and siltation resulted in a rising water table, which led to excessive soil
salinity (Gelburd, 1985). After 1000 years of successful irrigated agriculture, wheat
was largely replaced by more salt-tolerant barley. Consequently, these cities lost popu-
lation and power shifted to Babylon (Boyden, 1987). Numerous references are made
to irrigation canals in historic records of Mesopotamia, but no record of drainage ca-
nals being constructed has been found. The drainage of low-lying lands required
pumping, a technology unavailable at the time.
Historical salt and drainage problems have also been documented in North and
South America. The inhabitants of the Viru Valley on the coast of Peru developed an
irrigation system between 400 and 0 B.C. (Willey, 1953). The population peaked
about 800 A.D. before decreasing dramatically in the 13th century when people relo-
cated from the valley floor to the upper narrows of the valley. Historians attribute this
relocation to increasing soil salinity and rising water tables from inadequate drainage
(Armillas, 1961). Another example is the Hohokan Indians (Willey, 1953). They lived
in the Salt River region of what is now Arizona and practiced flood irrigation begin-
ning about 300 B.C. No evidence of the Hohokan civilization exists after 1450 A.D.
Historians surmise that waterlogging and salt accumulation caused crop failures.
Similar problems have occurred in Asia. In the Indus Plain, located in present-day
India and Pakistan, irrigation began about 2000 years ago. Since the mid-1800s, seri-
ous salinity and drainage problems have been experienced (Taylor, 1965).
7.1.2.2 Current outlook. Human-induced salination is not confined to antiquity.
One recent example is the region of the Amu Darya and Syr Darya rivers in the Com-
monwealth of Independent States. In the 1950s these river waters were diverted for the
development of irrigation before reaching the Aral Sea. These irrigated lands produced
90% of the former USSR’s cotton and 40% of its rice (Trofimenko, 1985). Due to
these major diversions, the Aral Sea began shrinking. Between 1960 and 1990 the
level of the Sea dropped 13 m in elevation and its surface area decreased 40%. This
shrinking has had devastating effects on the ecosystem and has caused health hazards
(Micklin, 1991).
Another example of prominence is the San Joaquin Valley of California. Irrigation
in the Valley began about 1850. Federal and state agencies envisioned a master drain
to serve the west side of the Valley. By 1975, however, only the southern third of the
drain, terminating at Kesterson Reservoir, had been completed. Budget restrictions and
environmental concerns prevented completion of the drain to San Francisco Bay.
Meanwhile, the drain conveyed saline drainage waters to Kesterson Reservoir, which
served as an evaporation pond. In 1982, selenium toxicity of fish and, in 1983, de-
formed and dead waterfowl were found in the Reservoir. The selenium content of the
water averaged 329 µg /L (Lee, 1990), 15 times higher than the recommended concen-
tration for irrigation water for long-term protection of plants and animals (Pratt and
Suarez, 1990). The discovery of selenium toxicosis from irrigation drainage water
emphasizes the need for concern about environmental impacts downstream from irri-
gation projects (Letey et al., 1987). This was the first documented case of selenium
toxicosis related to irrigation, but excess selenium is present in many basins in the
western U.S. (and deaths have occurred in China from excess selenium in the drinking
water). This problem has led to irrigation-drainage studies in seven western states in
the U.S. where selenium toxicity is suspected (Sylvester et al., 1988).
As these two modern-day examples illustrate, salinity can still create significant
water management problems. However, there are numerous examples of successful
irrigation enterprises under saline conditions from around the world; a few examples
are given here. In the Arkansas Valley of Colorado, irrigation water containing 1500
to 5000 mg/L of total dissolved salts has been used successfully to grow alfalfa, sor-
ghum, and wheat (Miles, 1977). Water containing 2500 mg/L of total dissolved salts
has been used for decades in the Pecos Valley of Texas (Moore and Hefner, 1977). In
regions of India receiving monsoon rains, wheat has been irrigated with waters con-
taining up to 10,000 mg/L (Dhir, 1977). Good cotton yields have been obtained in
Uzbekistan irrigating with drainage water containing 5000 to 6000 mg/L (Bressler,
1979). In the semi-arid continental-monsoon climate of China, saline water (2000 to
5000 mg/L) has been applied to wheat, corn, and cotton since the 1970s with excellent
yields (Fang et al., 1978).
7.1.2.3 Future concerns. History has shown that irrigated agriculture cannot sur-
vive in perpetuity without appropriate salt management and drainage. How long irri-
gated agriculture can survive without adequate salinity control depends on hydrogeol-
ogy and water management. The rate of salt accumulation in the soil is determined by
the salinity of the applied waters, the salinity of the soil profile, and the rate that salts
are leached out of the root zone. If layers restricting leaching are present near the soil
surface, salt accumulation and waterlogging develops within a relatively short time,
decades or less. If no restricting layers exist and the vadose region has a large water
storage capacity, irrigation may be practiced for a long time, perhaps centuries, before
salinity or waterlogging may become a threat.
7.2 QUANTIFYING SALINITY HAZARDS
Salinity, sodicity, and toxicity normally must be quantified for proper diagnosis and
management. When sampling water, 200 to 500 mL are usually sufficient for a multi-
tude of laboratory analyses. As for any sampling protocol, proper procedures must be
followed to prevent contamination and ensure accurate results. One procedure, often
ignored, is proper and complete labeling of samples. Samples should be refrigerated at
about 4°C (never frozen) and analyzed as soon as possible. Samples from wells should
be collected after pumping for at least half an hour. Typically, well-water quality will
not change significantly during the irrigation season. In a few cases, particularly when
the groundwater level changes substantially during the pumping season, periodic sam-
pling may be required because of possible changes in water quality.
Soil salinity can be analyzed in the laboratory based on water either extracted di-
rectly from the soil or taken from saturated soil samples. An alternative to laboratory
analysis is measuring salinity directly in the field.
If a field is to be sampled for laboratory analysis, a sample of about 0.5 kg of soil is
needed for each depth of interest. Samples should be air-dried, passed through a 2-mm
sieve, thoroughly mixed, and placed in durable, labeled containers. Labels should in-
clude sampling date, site, and soil depth.
Selecting a soil sampling strategy for salinity problems depends on the investiga-
tor’s objectives and the potential sources of variance in the samples. In relatively small
areas, sampling based upon judgment coupled with combining samples may be ade-
quate. Judging which locations to sample can be based on crop growth, location of
drains and irrigation appurtenances, and visual appearance of soils or plants. Fre-
quently, soil samples are composited to reduce the cost of analysis. Samples are often
composited from affected areas and compared with samples from unaffected areas.
When monitoring salinity with time, when an unbiased or accurate evaluation is
needed, or when salinity is assessed over a large area, sampling errors because of spa-
tial variability must be minimized to detect differences in salinity. The reader is en-
couraged to review Hanson and Grattan (1990) for detailed descriptions of systematic
sampling procedures.
7.2.1 Salinity
Knowing the amount of each individual solute in the soil water is ideal for diagno-
sis and management. Unfortunately, commercial field instruments are only capable of
estimating the total solute concentration, typically by measuring electrical conductiv-
ity (EC). Electrical conductivity measurements are very valuable because most plants
respond to the osmotic potential of the soil solution in the plant root zone, which can
be estimated by measuring electrical conductivity.
Laboratory analyses of soil water samples are required if the concentration of indi-
vidual solutes is desired. The standard method for laboratory analysis utilizes water
extracted from soil samples after they are brought to saturation by adding salt-free
water. This procedure enables water to be extracted easily for analysis. The ratio of the
water content of the saturated soil paste to the field water content is frequently about
2. This ratio, however, ranges from 1.8 to 3.0 among various soils. In the following
sections, measurements of electrical conductivity and concentrations of specific sol-
utes are frequently reported on the basis of field water content. To convert from con-
centrations in the soil water (Cw) to concentrations in a saturated soil extract (Ce),
knowledge of the bulk density of the soil (Db), volumetric field water content (1w), and
the gravimetric water content of the saturated soil paste on a percentage basis (satura-
tion percentage, SP) are required. The relationship is:
Ce = 100Cw1w/(Db S P) (7.1)
If the soil is relatively wet, soil water samples can be collected directly from the
soil by displacement, absorption, compaction, suction, centrifugation, or pressure
membrane extraction techniques. Of these, only suction is routinely used in the field.
Soil water can be extracted by applying vacuum into a porous cup or tube buried in the
soil. The porous sampler can be made of various ceramic or metallic materials. This
technique normally works when the soil water potential is greater than –30 J/kg.
Devices that can be installed in the field for direct measurement of soil salinity in-
clude porous-matrix salinity sensors, four-electrode units, inductive electromagnetic
meters, and time domain reflectometry probes. These methods estimate soil salinity
but not the concentrations of individual ions. Direct field measurements are particu-
larly helpful for mapping large areas or monitoring changes in salinity with time. They
also help determine the intensity and locations for detailed soil sampling.
Porous-matrix salinity sensors consist of a pair of small, corrosion-resistant elec-
trodes separated by a housing of ceramic or glass that is porous to soil water. The elec-
trodes connect to continual recording monitors or to manually operated, hand-held
meters or recorders. A temperature sensor in the probe is used to correct the electrical
conductivity readings for temperature. Response to changes in soil salinity can take a
day or longer because of the slow movement of soil solution by diffusion into the po-
rous housing. Calibration of the sensors can change, so periodic re-calibration is nec-
essary. Accuracy of commercial units is about ±0.5 dS/m (Oster and Willardson,
1971).
Four-electrode units consist of four electrodes inserted into the soil in a straight
line at a predetermined spacing. The electrical resistance to current flow between the
inner pair of electrodes is measured while a constant, alternating electric current is
passed through the soil between the outer electrodes (Rhoades and Ingvalson, 1971).
The electrical conductivity of the bulk soil depends upon the salt and water contents of
the soil, soil bulk density, and texture (Rhoades et al., 1990). A distinct advantage of
the four-electrode unit is that the spacing between the electrodes can be altered to
change the volume of soil measured. The depth of current penetration and thus the
depth of measurement is approximately equal to one-third of the distance between the
outer electrodes. Salinity within a small soil volume can be measured with portable
probes (Rhoades and van Schilfgaarde, 1976) or permanently buried probes (Rhoades,
1979). These units consist of four annular rings molded in a plastic or rubber cylinder
that is tapered slightly to maintain soil contact as it is inserted into an augured hole.
The volume of soil measured can be varied by changing the spacing between the annu-
lar rings; commercial units typically measure within a soil volume of 2 L.
An inductive electromagnetic meter can be hand-carried across the field to assess
soil salinity rapidly (Rhoades and Corwin, 1981). A transmitter coil located at one end
of the meter induces current flow in the soil which creates a secondary magnetic field.
The magnitude of the current flow in the soil is a function of the secondary field
strength and the electrical conductivity of the soil. By taking meter readings at various
heights above the soil surface, salinity by soil depth intervals can be estimated. The
meter is especially valuable for surveying salt-affected soils and mapping saline seeps.
Time domain reflectometry is a method for measuring volumetric water content and
electrical conductivity of the soil simultaneously. Soil water content is related to the
transit time and dissipation of an electromagnetic pulse launched along a set of metal-
lic parallel rods embedded in the soil. Salinity in the soil attenuates the pulse. This
measuring device is particularly beneficial because it measures soil salinity and water
content with the same probe (Dalton, 1992).
Salinity is reported using a variety of units. Various units and appropriate conver-
sion factors are presented in Table 7.2 to facilitate comparison and conversion among
units of measure. Concentration (C), the amount of substance per unit volume, is typi-
cally reported in SI metric units as moles per cubic meter of solution (mol/m3). Alter-
natively, C can be reported as grams per cubic meter (g/m3) or milligrams per liter
(mg/L). The units of g/m3 or mg/L are numerically equivalent to parts per million
(ppm). Traditionally, ionic concentration has been expressed as milliequivalents per
liter of solution (meq/L). To convert ionic concentration in units of meq/L to mol/m3,
divide by the valence of the ion. For example, divalent cations like calcium at a con-
centration of 20 meq/L would be equivalent to a concentration of 10 mol/m3. To con-
vert to g/m3, multiply the concentration in mol/m3 by the atomic weight of the ion. For
example, sodium at a concentration of 20 mol/m3 would be equivalent to a concentra-
tion of 460g/m3. Total salt concentration in units of g/m3 or mg/L is merely the sum of
the concentrations of each ion present. In units of mol/m3, it is the sum of either the
cations or the anions, but not both. The relationship between salt concentration (g/m3
or mg/L) and electrical conductivity (dS/m or mmho/cm) can be approximated by C =
640 EC.
Table 7.2. Units and conversion factors for various measurements of salinity.
Abbreviation
Measure Symbol Unit for Unit Conversion Factor
Electrical deciSiemens per meter dS/m dS/m = mmho/cm
EC
conductivity millimhos per centimeter mmho/cm
grams per cubic meter g/m3 ppm = mg/L = g/m3
milligrams per liter mg/L mol/m3 = meq/L ÷ ion valence
Concentration C parts per million ppm g/m3 = mol/m3 × atomic weight
moles per cubic meter mol/m3
milliequivalents per liter meq/L
7.2.2 Sodicity
Another water quality concern is an excess concentration of sodium, which can
cause a deterioration of soil structure. When the amount of sodium becomes excessive,
soil mineral particles tend to disperse and water penetration decreases. This becomes a
problem when the rate of infiltration is reduced to the extent that the crop is not ade-
quately supplied with water or when the hydraulic conductivity of the soil profile is
too low to provide adequate drainage. Sodium may also add to cropping difficulties
because of crusting seed beds, temporary saturation of the surface soil, and the in-
creased potential for disease, weeds, soil erosion, lack of oxygen, and inadequate nu-
trient availability. If calcium and magnesium, and not sodium, are the predominant
cations adsorbed on the soil exchange complex, the soil tends to have a granular struc-
ture that is easily tilled and readily permeable.
The sodium-adsorption ratio (SAR) of irrigation water is generally a good indicator
of the exchangeable sodium status that will occur in the soil. SAR is defined as:
C Na
SAR = (7.2)
(CCa + CMg ) 1/ 2
where all ion concentrations are in mol/m3. If the units are meq/L, the sum of CCa +
CMg must be divided by 2. Most irrigation waters from surface sources in arid areas are
supersaturated with calcite but precipitation is negligible. For such conditions, Equa-
tion 7.2 is normally a suitable indicator of SAR for soil water near the surface under
steady state conditions.
When groundwater, especially drainage water, is used for irrigation, Equation 7.2
does not accurately assess the hazard of excess sodium. A more accurate SAR (Suarez,
1981) adjusts the calcium concentration of the irrigation water to the expected equilib-
rium value in the soil and includes the effects of carbon dioxide (CO2), bicarbonate
(HCO3), and salinity upon the calcium originally present in the applied water but now
part of the soil water. This adjusted SAR (adj SAR) can be written as:
C Na
adj SAR = (7.3)
(C *Ca + CMg ) 1/ 2
where C*Ca is a modified calcium concentration. Values for C*Ca are given in Table
7.3. The adjusted SAR of the irrigation water is generally a better indicator of the ex-
changeable sodium status that will occur in the soil than SAR.
The permissible value of the SAR or adjusted SAR is a function of salinity. High
levels of salinity reduce swelling and aggregate breakdown (dispersion) and promote
water penetration, whereas high proportions of sodium produce the opposite effect
Table 7.3. The modified concentration of calcium (C*Ca) expected to remain in

soil-water near the soil surface following an irrigation with a given salinity
and ratio of bicarbonate to calcium (adapted from Suarez, 1981).
Ratio of Bicarbonate Electrical Conductivity of Applied Water (dS/m) [a]
3
to Calcium (both mol/m ) 0.1 0.5 1.0 2.0 4.0
0.1 6.1 7.2 7.6 8.2 9.0
0.2 4.2 4.5 4.8 5.2 5.7
0.5 2.3 2.5 2.6 2.8 3.1
1.0 1.4 1.6 1.6 1.8 1.9
2.0 0.90 0.98 1.0 1.1 1.2
4.0 0.57 0.62 0.66 0.70 0.77
8.0 0.36 0.39 0.41 0.44 0.48
20.0 0.20 0.21 0.22 0.24 0.26
60.0 0.09 0.10 0.10 0.12 0.12
[a]
Table values are based upon assumptions that soil has a source of lime or silicates, no precipitation of
magnesium, and a partial pressure of CO2 near the soil surface of 70 Pa.
Figure 7.3. Relative rate of water infiltration as affected by salinity and sodium-
adsorption ratio (adapted from Rhoades, 1977, and Oster and Schroer, 1979).
Figure 7.3 represents the approximate boundaries where chemical conditions se-
verely reduce infiltration of water into soil, where slight to moderate reductions occur,
and where no reduction is expected in most soils. Regardless of the sodium content,
waters with an electrical conductivity less than about 0.2 dS/m cause degradation of
the soil structure, promote soil crusting, and reduce water penetration. Rainfall and
snow melt would be prime examples of waters low in salinity that reduce water pene-
tration into soils. As Figure 7.3 illustrates, both the salinity and the sodium-adsorption
ratio of the applied water must be considered simultaneously when assessing the po-
tential effects of water quality on soil water penetration.
7.3 CROP TOLERANCE
7.3.1 Crop Salt Tolerance
Concentrations of soluble salts in the soil beyond a threshold level depress the
growth and yield of all crop plants. Growth depression is the most distinct salinity
injury symptom, usually to the exclusion of other injury signs from stress such as wilt-
ing, chlorosis, or necrosis. Crops can tolerate a certain level of salinity (the threshold)
without a measurable yield loss; the more salt tolerant the crop, the higher is this
threshold level. Beyond the threshold, yield is reduced linearly by further increases in
salinity until yield becomes zero. This relationship is depicted in Figure 7.4.
Under field conditions, the soil solution is made up of a mixture of salts. The pre-
dominant ones are chloride, sulfate, and bicarbonate salts of sodium, calcium, and
magnesium. The yield response functions, as illustrated in Figure 7.4, pertain to a
mixed salt solution as it normally occurs in nature. Some research reports in the litera-
ture relate yield to the concentration of a single salt, usually sodium chloride. The use
of single salts in experiments may create toxic effects and nutritional imbalances,
which are not necessarily related to salinity as it occurs in nature. Under most field
conditions crops respond to the total salt concentration. This response is termed the
osmotic or solute effect. Some crops, notably tree and vine crops, are specifically sen-
sitive to certain solutes, as will be discussed in Section 7.3.3. This toxicity is some-
times termed the specific solute effect.
Figure 7.4. Relative classifications of crop salt tolerance

based upon threshold and slope values.
The relation of yield to salinity is normally expressed relative to the maximum

yield that may be obtained under similar soil and climatic conditions at levels of salin-
ity much less than the threshold value. Maas and Hoffman (1977) suggested the fol-
lowing equation to express the relationship between crop yield and soil salinity:
⎧ 100 0 ≤ ECe ≤ t
⎪
Yr = ⎨ 100 − s (ECe − t ) t ≤ ECe ≤ Co (7.4)
⎪ 0 ECe ≥ Co
⎩
where Yr = relative yield, %
t = threshold salinity
s = rate of yield reduction with increasing soil salinity beyond the threshold
Co = level of soil salinity above which the yield is zero
ECe = average root zone salinity measured as the electrical conductivity of the
saturated soil extract.
Crops differ greatly in their response to salinity in both the threshold (t) and slope
(s) values. A sensitive crop such as strawberry has a threshold value of 1 dS/m and a
slope of 33% per dS/m, while a tolerant crop such as barley has a threshold of 8 dS/m
and slope of 4.5% per dS/m. The threshold and slope values for the salinity response
function are presented in Table 7.4 (Francois and Maas, 1994). The original tabulation
of these factors was reported by Maas and Hoffman (1977).
Table 7.4. Salt tolerance of agricultural crops[a] (adapted from Francois and Maas, 1994).
Crop Growth Salt Tolerance Parameters Relative
Common Botanical Measure- Threshold (t) Slope (s) Tolerance
[c]
Name Name[b] ment ECe (dS/m) (% per dS/m) Rating
Alfalfa Medicago sativa Shoot 2.0 7.3 MS
Almond Prunus duclis Shoot 1.5 19 S
Apricot Prunus armeniaca Shoot 1.6 24 S
Asparagus Asparagus officinalis Spear 4.1 2.0 T
(continued)

Barley Hordeum vulgare Grain 8.0 5.0 T
Barley (forage) Hordeum vulgare Shoot 6.0 7.1 MT
Bean Phaseolus vulgaris Seed 1.0 19 S
Bean, mung Vigna radiata Seed 1.8 20.7 S
Beet, red Beta vulgaris Storage root 4.0 9.0 MT
Bermuda grass Cynodon dactylon Shoot 6.9 6.4 T
Blackberry Rubus macropetalus Fruit 1.5 22 S
Boysenberry Rubus ursinus Fruit 1.5 22 S
Broadbean Vicia faba Shoot 1.6 9.6 MS
Broccoli Brassica oleracea Shoot 2.8 9.2 MS
(botrytis)
Cabbage Brassica oleracea Head 1.8 9.7 MS
(capitata)
Carrot Daucus carota Storage root 1.0 14 S
Celery Apium graveolens(dulce) Petiole 1.8 6.2
Clover, alsike Trifolium hybridum Shoot 1.5 12 MS
Clover, berseemTrifolium alexandrinum Shoot 1.5 5.7 MS
Clover, ladino Trifolium repens Shoot 1.5 12 MS
Clover, red Trifolium pratense Shoot 1.5 12 MS
Clover, Trifolium fragiferum Shoot 1.5 12 MS
strawberry
Corn Zea mays Ear 1.7 12 MS
Corn (forage) Zea mays Shoot 1.8 7.4 MS
Cotton Gossypium hirsutum Lint or seed 7.7 5.2 T
Cowpea Vigna unguiculata seed 4.9 12 MT
Cowpea Vigna unguiculata shoot 2.5 11 MS
(forage)
Cucumber Cucumis sativus Fruit 2.5 13 MS
Date palm Phoenix dactylifera Fruit 4.0 3.6 T
Eggplant Solanum melongena fruit 1.1 6.9 MS
Fescue, tall Festuca elatior Shoot 3.9 5.3 MT
Flax Linum usitatissimum Seed 1.7 12 MS
Foxtail, Alopecurus pratensis Shoot 1.5 9.6 MS
meadow
Garlic Allium sativum Bulb 1.7 10 MS
Grape Vitis vinifera Shoot 1.5 9.6 MS
Grapefruit Citrus × paradisi Fruit 1.2 13.5 S
Guar Cyamopsis tetragonoloba Seed 8.8 17 T
Guava Psidium guajava Shoot & 4.7 9.8 MT
root
Guayule Parthenium argentatum Rubber 7.8 10.8 T
Kenaf Hibiscus cannabinus Stem 8.1 11.6 T
Lemon Citrus limon fruit 1.5 12.8 S
Lettuce Lactuca sativa Top 1.3 13 MS
Love grass Eragrostis Shoot 2.0 8.4 MS
Muskmelon Cucumis melo Fruit 1.0 8.4 MS
Onion Allium cepa Bulb 1.2 16 S
Orange Citrus sinensis Fruit 1.3 13.1 S
Orchard grass Dactylis glomerata Shoot 1.5 6.2 MS
Pea Pisum sativum Seed 3.4 10.6 MS
(continued)

Peach Prunus persica Shoot or 1.7 21 S
fruit
Peanut Arachis hypogaea Seed 3.2 29 MS
Pepper Capsicum annuum Fruit 1.5 14 MS
Plum, prune Prunus domestica Fruit 2.6 31 MS
Potato Solanum tuberosum Tuber 1.7 12 MS
Radish Raphanus sativus Storage root 1.2 13 MS
Rice, paddy[d] Oryza sativa Grain 3.0 12 S
Rye Secale cereale Grain 11.4 10.8 T
Rye (forage) Secale cereale Shoot 7.6 4.9 T
Ryegrass, Lolium perenne Shoot 5.6 7.6 MT
perennial
Sesbania Sesbania exaltata Shoot 2.3 7.0 MS
Sorghum Sorghum bicolor Grain 6.8 16 MT
Soybean Glycine max Seed 5.0 20 MT
Spinach Spinacia oleracea Top 2.0 7.6 MS
Squash, scallop Cucurbita pepomelopepo Fruit 3.2 16 MS
Squash, Cucurbita pepomelopepo Fruit 4.7 9.4 MT
zucchini
Strawberry Fragaria × ananassa Fruit 1.0 33 S
Sudan grass Sorghum sudanense Shoot 2.8 4.3 MT
Sugar beet Beta vulgaris Storage root 7.0 5.9 T
Sugarcane Saccharum officinarum Shoot 1.7 5.9 MS
Sweet potato Ipomoea batatas Tuber 1.5 11 MS
Tomato Lycopersicon Fruit 2.5 9.9 MS
lycopersicum
Tomato, cherry Lycopersicon Fruit 1.7 9.1 MS
lycopersicum
Trefoil, big Lotus pedunculatus Shoot 2.3 19 MS
Trefoil, Lotus corniculatus Shoot 5.0 10 MT
narrowleaf
birdsfoot
Triticale × triticosecale Grain 6.1 2.5 T
Turnip Brassica rapa Storage 0.9 9.0 MS
root
Turnip (greens) Brassica rapa Top 3.3 4.3 MT
Vetch, common Vicia angustifolia Shoot 3.0 11 MS
Wheat Triticum aestivum Grain 6.0 7.1 MT
Wheat (forage) Triticum aestivum Shoot 4.5 2.6 MT
Wheat, durum Triticum turgidum durum Grain 5.9 3.8 T
Wheat, durum Triticum turgidum durum Shoot 2.1 2.5 MT
(forage)
Wheat Triticum aestivum Grain 8.6 3.0 T
(semidwarf)
Wheatgrass, tall Agropyron elongatum Shoot 7.5 4.2 T
[a]
These data serve only as a guideline to relative tolerances among crops. Absolute tolerances vary, de-
pending upon climate, soil conditions, and cultural practices. S, sensitive; MS, moderately sensitive;
MT, moderately tolerant; T, tolerant.
[b]
Botanical and common names follow the convention of Hortus Third when possible.
[c]
Ratings are defined by the boundaries in Figure 7.4.
[d]
Because paddy rice is grown under flooded conditions, values refer to the electrical conductivity of the
soil water while the plants are submerged; less tolerant during seedling stage.
7.3.2 Factors Modifying Crop Salt Tolerance

Crops are sometimes exposed to conditions that differ from those under which salt
tolerance data were obtained. A frequently asked question is whether the differences
are significant with respect to the application of salt tolerance ratings to a variety of
field conditions. Factors which may interact with salinity to cause a different yield
response include soil, crop, and environmental conditions.
7.3.2.1 Soil. Soil factors which may influence salt tolerance are texture and struc-
ture, fertility, and management schemes that alter spatial and temporal distribution of
soil salinity. Soil structure and texture impact salt tolerance through their influence on
infiltration, water holding capacity (WHC), aeration, and the ratio of saturation water
content to field water content. For the same potential evaporation rate, a sandy soil
with low WHC will lose proportionately more water than a fine-textured soil, resulting
in a more rapid increase of salt concentration in the soil solution and potentially
greater damage to the crop. This possible damage can be mitigated by irrigating a
sandy soil more frequently than a clay soil. The impact of aeration on salt tolerance is
not well understood. Some studies report yield reduction due to salinity may be less
pronounced under poor soil aeration conditions (Aubertin et al., 1968; Drew et al.,
1988); other investigations indicate about the same (John et al., 1976) or greater ef-
fects (Kriedemann and Sands, 1984; West and Taylor, 1984).
The impact of soil fertility on the yield response to soil salinity has been studied ex-
tensively. The most critical nutrient interacting with salinity is nitrogen (N), but phos-
phorus (P) and potassium (K), in rare situations, may also show interactions. Three
types of relationships between the level of fertilizer application and salinity might oc-
cur: (1) the addition of fertilizer results in the same relative yield increase at all levels
of salinity; (2) there is a larger response to fertilizer at low compared to high levels of
salinity, and (3) there is a larger increase in relative yield due to fertilizer application
at high salinity levels. Of 51 reports examined on the interaction of fertilizer applica-
tion and salinity, only five showed the third type of response, a more positive response
to fertilizer under high rather than low salinity. The remainder of the reports were di-
vided equally between the first and second types of responses. The overwhelming
conclusion is that the addition of nitrogen fertilizer can seldom overcome the deleteri-
ous effect of salinity on crop yield. In nutrient solutions, high P concentration may be
toxic at high salinity levels. Under most field conditions, the concentration of calcium
in the soil solution is high enough to result in precipitation of P, resulting in an in-
creased P fertilizer requirement. Thus, excess P is seldom found in salt-affected soils.
Most crop plants possess high selectivity for K uptake, thereby reducing the danger of
K deficiency.
The salinity distribution in the root zone must be known to calculate the average
root zone salinity. Average root zone salinity is used to report crop salt tolerance. The
bulk of the evidence indicates that the mean salinity with depth, integrated over time,
is the most representative salinity. The results of Bower et al. (1969) and Shalhevet et
al. (1969) are for soil salinity values varying widely with depth, with the maximum
values occurring at either the bottom or top of the root zone. In a field experiment on
organic soil (Hoffman et al., 1983b), sprinkler irrigation and subirrigation treatments
created widely different soil salinity distributions, but corn yield was shown to re-
spond to the average soil salinity. An exception to this conclusion is the result of
Bernstein and Francois (1973) who suggested that the representative soil salinity is the
weighted salinity in proportion to the relative depth distribution of water uptake. For
widely spaced plantings, such as orchards, the spatial salt distribution in three dimen-
sions within the root zone of a single tree must be considered.
7.3.2.2 Crop. Crop factors that may modify the response to salinity are variety,
rootstock, and stage of growth. The normal breeding and selection process of crop
varieties typically emphasizes high productivity rather than tolerance to salinity. Con-
sequently, varietal differences in salt tolerance are not common among field and gar-
den crops. Nevertheless, some varietal differences do exist and may cause confusion
with regard to the application of salt tolerance data. Some documented examples are:
barley (Epstein et al., 1980), wheat (Torres and Bingham, 1973), tomato, soybean,
lettuce (Shannon, 1980), and melon (Shannon and Francois, 1978). Breeding programs
to improve crop salt tolerance have resulted in a limited number of tolerant varieties.
Shannon and Noble (1990) list eleven such varieties. Large differences in salt toler-
ance are known among rootstocks of subtropical crops and deciduous fruit trees. These
crops are frequently salt sensitive and are known to be especially sensitive to chloride
and sodium salts.
Determination of the relative sensitivity of crops to salinity at different stages of
growth is difficult because the effect on growth during an early stage may influence
the response during later growth stages. Some studies claiming to show stage of
growth sensitivity, in fact, demonstrate a sensitivity to the duration the roots were ex-
posed to salinity. Mass and coworkers (1986; 1989 a,b) applied salinity treatments for
30-day periods during the vegetative, reproductive, and maturity stages of growth for
several crops. The major conclusion from their results is that crops are specifically
sensitive during the early seedling period. Re-analysis of their results on a time-
weighted basis, taking the number of days from treatment initiation to harvest as the
weighing factor, shows very similar results for the vegetative and reproductive stages.
Once salinity is applied for a significant period of time and then removed, there does
not appear to be significant recovery—the damage is sustained to the end of the sea-
son. Naturally, the later salinity is applied the less the damage will be. An important
management option resulting from this analysis is that saline water may be applied late
in the season with very little damage to crop production (Francois et al., 1994).
Salinity may delay seed germination and seedling emergence, but most crops are
capable of germinating at higher salinity levels than they can tolerate at later stages of
growth. For example, a salinity level of 4.7 dS/m at planting was shown to reduce pea-
nut yield by 50%, while seedling growth after emergence was not reduced by half until
ECe reached 7.5 d/Sm and 50% germination was achieved at an ECe of 13 dS/m (Shal-
hevet et al., 1969). Similar results were obtained with corn (Maas et al., 1983) where
germination was satisfactory at an electrical conductivity of the soil water, ECsw, of 10
dS/m while the threshold for grain yield was at an ECsw of 5.5 dS/m. Threshold salin-
ity, ECsw, for seedling growth, however, was only 1.0 dS/m.
7.3.2.3 Environment. Six elements of the aerial environment have been shown to
impact crop salt tolerance: temperature, humidity, rainfall, radiation, carbon dioxide,
and air pollution. Temperature is a critical environmental factor for crop salt tolerance.
High temperature increases the stress level to which a crop is exposed, either because of
increased transpirational demand or because of the effect of temperature on the bio-
chemical transformations in the leaf. The increase in the level of stress due to high tem-
perature increases the sensitivity of the crop to salinity (Magistad et al., 1943, for al-
falfa, bean, beet, carrot, cotton, onion, and tomato; Lunt et al., 1960, for cherry, chrysan-
themum, and kidney bean; Hoffman et al., 1978, for bean; and Ehlig, 1960, for grape).
High atmospheric humidity tends to decrease the crop stress level, thus reducing sa-
linity damage. High humidity generally increases the salt tolerance of salt-sensitive
more than salt-tolerant crops (Hoffman et al., 1971; Hoffman and Jobes, 1978).
Rainfall has no direct effect on the salinity response function, but it may increase
leaching and replace potentially saline sources for satisfying evapotranspiration. Rain-
fall prior to and during the irrigation season makes it possible to use more saline irri-
gation water because of the dilution of the soil solution. This impact is quantified in
Section 7.4.1.
For some crops, high atmospheric carbon dioxide may reduce salinity damage.
Schwartz and Gale (1984) found only an 18% yield reduction for beans at high carbon
dioxide levels (2.5 mL CO2/L of air) compared with 64% reduction for the same in-
crease in salinity at the ambient carbon dioxide concentration (0.3 mL CO2/L of air).
No interaction with carbon dioxide was demonstrated for corn.
The apparent tolerance to salinity improves with an increase in the concentration of
ozone, an important air pollutant. For example, at low levels of salinity, high levels of
ozone caused a 95% yield reduction for bean while only a 44% reduction occurred for
the same ozone level at high salinity levels (Hoffman et al., 1973). As with other fac-
tors (e.g., soil fertility), when yield potential is suppressed by one limiting factor, the
relative effect of another limiting factor is reduced.
Contrary to expectation, plants may be more sensitive to salinity as solar radiation
is reduced. Despite half the yield potential in the shade as in the sun, salinity caused a
37% reduction in dry matter for Maranta leuconeura, while in the sun the reduction
was only 15% (Nolan et al., 1982). Reduced radiation significantly decreased the salt
tolerance threshold of cantaloupe grown in a greenhouse (Meiri et al., 1982).
7.3.3 Tolerance to Specific Solutes
Specific solutes, such as boron, chloride, and sodium, are potentially toxic to crops
that take up these solutes through the root system and accumulate them in the leaves.
Many trace elements are also toxic to plants at very low concentrations. Suggested
maximum concentrations for a number of trace elements are given by Pratt (1973).
Fortunately, most irrigation supplies contain insignificant concentrations of these po-
tentially toxic trace elements and toxicity is generally not a problem.
7.3.3.1 Boron. Boron is a trace element found occasionally in saline soils. At con-
centrations only slightly greater than the 0.2 to 0.5 g/m3 needed for optimum growth,
boron may be phytotoxic. Boron toxicity symptoms typically appear at the tip and
along the edges of older leaves as yellowing, spotting, and/or drying of leaf tissue. The
damage gradually progresses between the veins toward midleaf. A gummosis or exu-
date on limbs or trunks is sometimes noticeable on boron-affected trees, such as al-
mond. Many sensitive crops show toxicity symptoms when boron concentrations in
leaf blades exceed 250 mg/kg (dry-mass basis), but not all sensitive crops accumulate
boron in their leaves. Stone fruits (e.g., peach, plum, almond) and pome fruits (pear,
apple, and others) are examples of crops that do not accumulate boron in leaf tissue.
Many crops have been tested for foliar damage caused by boron (i.e., Eaton, 1944). A
much smaller selection of crops has been evaluated for yield loss.
The Maas-Hoffman model for salt tolerance has also been applied to boron toxicity.
Both relative yield and relative shoot growth have been found to be described ade-
quately by this model (Francois, 1984, 1986; Bingham et al., 1985). The Maas-
Hoffman model for boron tolerance is:
Yr =100 – sb (Cb – Cbt) (7.5)
where the terms are the concentrations for boron (b) rather than electrical conductivity
for salinity as in Equation 7.4. The model coefficients for threshold and slope (when
measured) and qualitative tolerance ratings for crops studied to date are given in Table
7.5. Although the threshold tolerance for boron varies widely among crops, the slopes
Table 7.5. Coefficients of threshold (Cbt) and slope (sb) for boron tolerance using the Maas-
Hoffman model. Boron concentrations are for soil water (adapted from Maas, 1990).
Boron Concentration of Soil Water
Growth Threshold, Cbt Slope, sb Qualitative
Crop Measurement (g/m3) (% per g/m3) Rating[a]
Alfalfa Shoot 4-6 T
Barley Grain 3.4 4.4 MT
Bean Shoot 0.7 - 1 S
Beet, red Shoot 4-6 T
Broccoli Head 1.0 1.8 MS
Cabbage Shoot 2-4 MT
Carrot Shoot 1-2 MS
Cauliflower Curd 4.0 1.9 T
Corn Shoot 2-4 MT
Cotton Shoot 6 - 10 T
Cowpea Shoot 2.5 12 MT
Cucumber Shoot 1-2 MS
Grape Shoot 0.5 - 0.7 S
Grapefruit Shoot 0.5 - 0.7 S
Lemon Shoot < 0.5 S
Lettuce Shoot 1.3 1.7 MS
Muskmelon Shoot 2-4 MT
Onion Shoot 0.5 - 0.7 S
Orange Shoot 0.5 - 0.7 S
Peach Shoot 0.5 - 0.7 S
Peanut Seed 0.7 - 1 S
Pecan Shoot 0.5 - 0.7 S
Pepper Shoot 1-2 MS
Plum Shoot 0.5 - 0.7 S
Potato Tuber 1-2 MS
Radish Storage root 1.0 1.4 MS
Sorghum Grain 7.4 4.7 T
Squash Shoot 2-4 MT
Strawberry Shoot 0.7 - 1 S
Sugar beet Storage root 4.9 4.1 T
Sunflower Seed 0.7 - 1 S
Sweet potato Tuber 0.7 - 1 S
Tomato Fruit 5.7 3.4 T
Walnut Shoot 0.5 - 0.7 S
Wheat Grain 0.7 - 1 3.3 S
[a]
Rating is based upon the threshold value. Separations between sensitive (S), moderately sensitive (MS),
moderately tolerant (MT), and tolerant (T) are concentrations of 1, 2, and 4 g of boron per m3 of soil so-
lution.
Table 7.6. Chloride tolerance limits of some fruit-crop

cultivars and rootstocks (adapted from Maas, 1990).
Maximum Permissible
Cl- in Soil Water without
3
Crop Rootstock or Cultivar Leaf Injury (mol/m )
Avocado West Indian 15
Guatemalan 12
Mexican 10
Citrus Sunki mandarin, grapefruit, Cleopatra mandarin, 50
Rangpur lime
Sampson tangelo, rough lemon, sour orange, 30
Ponkan mandarin
Citrumelo 4475, trifoliate orange, Cuban shaddock, 20
Calamondin, sweet orange, Savage citrange,
Rusk citrange, Troyer citrange
Grape Salt Creek, 1613-3 80
Dog ridge 60
Stone fruit Marianna 50
Lovell, Shalil 20
Yunnan 15
Berries Boysenberry, Olallie blackberry 20
Indian Summer raspberry 10
Grape Thompson seedless, Perlette 40
Cardinal, Black Rose 20
Strawberry Lassen 15
Shasta 10
of the yield reductions only vary from 1.4% to 12% (Table 7.5). This indicates that
although crop yields are reduced by relatively low concentrations of boron, yields do
not decline rapidly as boron levels increase in excess of the threshold value.
7.3.3.2 Chloride. Most herbaceous crops are not particularly sensitive to chloride.
Soybeans are an exception; salt-sensitive soybean cultivars accumulate excessive
amounts of chloride. This problem has been avoided by breeding cultivars that restrict
the transport of chloride to the shoots. Woody plant species, on the other hand, are all
generally susceptible to chloride toxicity. Tolerances vary among species and even
among varieties or rootstocks within a species. These differences usually reflect the
prevention or retardation of chloride accumulation in the crop canopy. Table 7.6 lists
chloride tolerance limits for some rootstocks and cultivars of fruit crops for which data
are available. The data in Table 7.6 indicate the maximum chloride concentrations
permissible in the soil water that do not cause leaf injury.
7.3.3.3 Sodium. Plant growth and yield may be reduced by the accumulation of
toxic levels of sodium. Sodium injury is generally limited to woody species like avo-
cado, citrus, and stone-fruit trees. Sodium concentrations as low as 5 mol/m3 in soil
water or 0.25% to 0.5% on a dry weight basis in leaves can cause injury in some tree
crops. Sodium is normally retained in the roots and the lower trunk of stone-fruit trees,
but after several years, the conversion of sapwood to heartwood apparently releases
the accumulated sodium, which is transported to the leaves causing tip and margin
necrosis (Maas, 1990).
In sodic, nonsaline soils, total soluble salt concentrations are low and, conse-
quently, calcium and/or magnesium concentrations may be nutritionally inadequate.
These deficiencies rather than sodium toxicity are usually the primary cause of poor
plant growth among non-woody species. Sodium uptake by plants is strongly regu-
lated by calcium in the soil solution. Thus, the presence of sufficient calcium is essen-
tial to prevent the accumulation of toxic levels of sodium. As a general guide, calcium
and magnesium concentrations in the soil solution above 1 to 2 mol/m3 each, with the
concentration of calcium being at least as great as magnesium, are nutritionally ade-
quate in sodic, nonsaline soils.
7.3.4 Crop Water Production Functions
The function relating yield (Y) to transpiration (T) is the water production function.
The slope of this function is the water use efficiency. The production function is lin-
ear, crop-specific, and independent of ambient conditions, as long as T is normalized
with respect to potential evaporation (Ep) (de Wit, 1958). In the field, it is seldom pos-
sible to measure transpiration directly. The more commonly used relationship is yield
as a function of applied water (AW). Water quantities applied in the field (irrigation
plus rainfall) are equivalent to the sum of transpiration, evaporation, changes in soil
water storage, leaching, and runoff. An example of the relationship between yield and
water application is given in Figure 7.5 for tomato. In figures such as 7.5, the amount
of water applied at zero yield is typically taken to be the amount of water applied that
evaporates or may be transpired without yield production. This component may shift
the intersect on the water quantity axis (see Figure 7.5), but will not change the slope.
The leaching component may change the slope and the shape of the curve from linear
to exponential.
Figure 7.5. Yield of marketable tomato fruit as a function of water application

depths for three levels of salinity of the irrigation water (Vinten et al., 1986).
The relation of yield to water application is independent of salinity, as long as there

is no leaching (Childs and Hanks, 1975). Shalhevet (1984) summarized experiments in
which both salinity and water quantity were varied to show production functions for
nine crops. Salinity may affect the production function in three ways. First, the maxi-
mum yield possible (the yield plateau) will be lower as soil salinity increases (see Fig-
ure 7.5). At the same time, transpiration will be reduced commensurate with the slope
of the particular crop water production function. Second, the intercept on the water
quantity axis may shift to the right because the evaporation component may be greater
than under nonsaline conditions. For example, Shalhevet et al. (1969) found half of the
total evapotranspiration to be lost in evaporation because the peanut canopy was slow
to develop and did not reach full soil cover under saline conditions. Third, the need to
satisfy the leaching requirement may result in a curvilinear function as demonstrated
in Figure 7.6. A larger quantity of saline water must be applied to maintain a given
yield level compared to nonsaline conditions. In Figure 7.6 the deviation from linear-
ity reflects the leaching fraction, (AW2–AW1)/AW2). The distance Ymax – Y4 is the yield
loss due to an irrigation salinity of 4 dS/m at maximum leaching.
Figure 7.6. Relative yield of tall fescue as related to the ratio of applied water
to potential evaporation (adapted from Letey et al., 1985).
7.4 LEACHING
Water must drain through the crop root zone to prevent solutes from increasing to
concentrations detrimental to crop production. All soils have an inherent ability to
transmit soil water provided a hydraulic gradient exists and the hydraulic conductivity
is reasonable. This is usually denoted as natural drainage capacity. Drainage occurs
when the hydraulic gradient is positive in the downward direction. Natural drainage in
many instances is sufficient to leach salts from the root zone. Most soils of fine tex-
ture, soils with compacted layers, and soils with layers of low hydraulic conductivity
may be so restrictive to downward water movement that the natural drainage capacity
is insufficient to provide adequate leaching. In addition, the hydrogeology may be
such that the hydraulic gradient is predominately upward, which leads to waterlogging
and salination.
If natural drainage is not adequate, a man-made or artificial drainage system must
be installed to sustain productivity. Before designing an artificial drainage system, the
natural drainage capacity should be determined. If, after applying excess water, the
hydraulic gradient and the soil’s hydraulic conductivity permit soil water to drain out
of the crop root zone, the intensity of the artificial system can be reduced. Alterna-
tively, upward flow into the root zone can intensify the need for artificial drainage.
With either natural or artificial drainage, water leaching below the crop root zone
must go elsewhere. It may take decades or just a season for productivity to be reduced,
depending on the hydrogeology of the area, but without drainage to an appropriate
outlet agricultural productivity can not be sustained.
7.4.1 The Leaching Requirement
The amount of irrigation water needed to satisfy the crop’s water requirement can
be estimated from water and salt balances within the crop root zone. The major flows
of water into the root zone are irrigation, rainfall, and upward flow from the ground-
water. Water flows out by evaporation, transpiration, and drainage. Under steady-state
conditions, the change in depth of soil water and salt storage are essentially zero. If the
total water inflow is less than evaporation plus transpiration, water is extracted from
soil storage and drainage is reduced, with time, the difference between inflows and
outflows becomes zero. In the absence of net downward flow, salt will accumulate,
crop growth will be suppressed, and transpiration will be reduced by a value consistent
with the crop water production function.
In the presence of a shallow water table, deficiencies in the irrigation and rainfall
amounts may be offset by upward flow from the groundwater. Upward flow will carry
salts into the root zone. If upward flow continues and sufficient leaching does not oc-
cur, soil salinity will ultimately reduce crop growth and water consumption. Over the
long term, a net downward flow of water is required to control salination and sustain
crop productivity.
Rarely do conditions controlling the water that flows into and out of the root zone
prevail long enough for a true steady state to exist. However, it is instructive to con-
sider a simple form of the steady-state equation to understand the relationship between
drainage and salinity. If it is assumed that the upward movement of salt is negligible,
the quantities of salt dissolved from soil minerals plus salt added as fertilizer or
amendments is essentially equal to the sum of precipitated salts plus salt removed in
the harvested crop, and the change in salt storage is zero under steady-state conditions,
the leaching fraction (L) can be written as:
Dd C a EC a
L= = = (7.6)
Da C d EC d
where EC is electrical conductivity and the subscripts d and a designate drainage and
applied water (irrigation plus rainfall). Equation 7.6 applies only to salt constituents
that remain dissolved. The importance of Equation 7.6 is greater than the accuracy of
predicting the drainage amount or salt concentration because the relationship between
the leaching fraction and root-zone salinity is extremely important in managing soil
salinity.
The minimum leaching fraction that a crop can endure without yield reduction is
termed the leaching requirement, Lr, which can be expressed as follows:
Dd* C a EC a
Lr = = = (7.7)
Da C d* EC d*
The notation in Equation 7.7 is the same as in Equation 7.6 except the superscript (*)
distinguishes required from actual values. Several models have been proposed to relate
Cd* in Equation 7.7 to some readily available value of soil salinity that indicates the
crop’s leaching requirement. Hoffman and van Genuchten (1983) determined the line-
arly averaged, mean root-zone salinity by solving the continuity equation for one-
dimensional vertical flow of water through soil, assuming an exponential soil water
uptake function (Raats, 1974). The linearly averaged salt concentration of the root
zone given as the concentration of the saturated extract (C ) is given by:
C 1 δ (7.8)
= + ln[ L + (1 − L) e − Z / δ ]
C a L ZL
where Ca = salt concentration of the applied water

L = leaching fraction
Z = depth of the root zone
δ = an empirical constant assumed equal to 0.2 Z.
Figure 7.7 illustrates the relationship given in Equation 7.8, taking into account the
salinity of rainfall and irrigation water. Note salinity in Figure 7.7 is expressed in
terms of electrical conductivity of the soil saturation extract (EC e or C e) assuming C is
twice Ce.
Plants adjust osmotically as soil salinity increases (Maas and Nieman, 1978). Be-
cause salt-tolerance trials are usually designed to maintain leaching fractions of about
0.5, the osmotic adjustment consistent with no loss of yield can be estimated by the
mean soil salinity at 50% leaching, as given in Figure 7.7. As a first approximation,
Figure 7.7. The mean root-zone soil salinity as a function of the salinity of the applied
water and leaching fraction (L) (adapted from Hoffman and van Genuchten, 1983).
Figure 7.8. Leaching requirement (Lr) as a function of the

salinity of the applied water and the salt-tolerant threshold value
for the crop (adapted from Hoffman and van Genuchten, 1983).
the leaching requirement can be expressed as a function of the crop salt tolerance
threshold value by reducing C at any given L by C at 50% leaching. Figure 7.8 illus-
trates this relationship among the salinity of the applied water, the salt-tolerance
threshold of the crop, and the leaching requirement. Hoffman (1985) compared calcu-
lated leaching requirements from this and other models with experimental results. Of
the four models tested, the one presented here agrees well with the measured values
throughout the range of Lr of agricultural interest.
The following illustrates the procedure for estimating the leaching requirement. As-
sume tomatoes are to be grown in an arid region and irrigated by water with an ECi of
3 dS/m. The salt-tolerance threshold, t, of tomatoes is 2.5 dS/m (Table 7.4). If rainfall
is insignificant compared to irrigation and there is no upflow from a shallow water
table, the leaching requirement can be determined directly from Figure 7.8. In this
case, the leaching requirement equals 0.2. The drainage water would have an EC of 15
dS/m from Equation 7.7. To calculate the salinity of the applied water when rainfall is
significant, the equation used is:
CaDa = CrDr + CiDi (7.9)
The variable C is salt content and D is depth. The subscripts a, r, and i, indicate ap-
plied, rain, and irrigation water, respectively. An iterative process must be used to
determine Lr since Di is unknown until Lr is known. As an illustration, assume
evapotranspiration, De + Dt, is 750 mm and rainfall, Dr, is 150 mm. Begin calculations
by assuming that Di is 900 mm and Cr = 0. Because Da = Dr + Di, then Ca = CiDi/(Dr
+ Di) = 3(900)/(150 + 900) = 2.6 dS/m. With Ca = 2.6, Lr for tomato is 0.18 and Di =
De + Dt + Dd = 750 + 0.18 Di. Thus Di is 915 mm. This is sufficiently close to the
assumed value of 900 mm that further iterations are unnecessary.
Accounting for nonuniformity of irrigations in estimating Lr has not been addressed
to date. If the leaching requirement is not met everywhere in the field, saline soil will
develop wherever evapotranspiration plus the leaching requirement is not met. One
must evaluate whether to apply copious amounts of water to assure that the leaching
requirement is met throughout the field or to accept some reduction in yield in parts of
the field rather than overirrigate most of the field.
Extensive advancements in irrigation technology and management are needed to
approach the goal of matching the leaching requirement. Present irrigation practices in
many areas are inefficient and inadvertently provide excessive leaching. This is costly.
It leads to a loss of water, energy, and nutrients, and increases the need for drainage
facilities. Consequently, knowing the leaching requirements of crops and striving to
attain them with properly designed and managed irrigation systems is vital. Neverthe-
less, under the best conditions at present, some excess irrigation water is generally
applied to achieve maximum yield.
7.4.2 Soil Salinity with Restricted Leaching
An example of the effect of virtually no drainage on soil salinity is given by Peck et
al. (1981) for southwest Australia. Figure 7.9 illustrates the chloride concentration and
downward velocity of the soil solution in a native eucalyptus forest with an annual
rainfall of 800 mm. Chloride concentration was highest at a soil depth of 7 m, where
the downward velocity of the soil solution equaled 0.04% of the annual rainfall, or 0.3
mm/yr. Below 7 m, the chloride concentration decreased linearly to less than 2000
mg/L just above the water table at a depth of 17 m.
The deeper the soil, the greater the capacity to store salt with minimal yield reduc-
tion. High levels of salinity in the lower portion of a crop root zone have a minor in-
fluence on yield if the upper portion is maintained at a relatively low level of salinity
(Bingham and Garber, 1970; Bernstein and Francois, 1973). Plants compensate for
reduced uptake of water from a zone of highly salinized soil by increasing uptake from
a zone low in salinity (Wadleigh et al., 1947; Lunin and Gallantin, 1965; Shalhevet
and Bernstein, 1968;). Although this minimizes yield loss, there is general concern
about how much salt can be stored in the crop root zone before leaching is needed.
Chlor ide Concent r at ion (m g/L) Soil Solut ion Velocit y (m m /yr)
Figure 7.9. Soil chloride concentration and the downward rate of soil-water
movement as a function of soil depth in poorly drained, non-irrigated soils of
southwest Australia (Peck et al., 1981).
Table 7.7. Change in the area and yield of cotton and tomato in Broadview
Water District compared to Fresno County (from Wichelns et al., 1988).
1968 to 1972 1978 to 1982 Change (%)
Broad- Fresno Broad- Fresno Broad- Fresno
view County view County view County
Cotton
Area (ha) 715 76,000 2,130 167,000 +200 +140
Lint yield (kg/ha) 1,190 1,190 1,305 1,190 +10 0
Tomato
Area (ha) 605 8,150 170 16,300 –0 +100
Fruit yield (Mg/ha) 72 50 59 64 –8 +28
One of the first studies to address the question of how much salt can be stored in
the root zone without significant yield reduction involved alfalfa grown in a green-
house and irrigated with water having an electrical conductivity (ECi) of 1 dS/m
(Francois, 1981). The plants were grown without leaching in a sandy loam soil with
water table depths of 0.6 m, 1.2 m, or 1.8 m for periods of 9, 14, and 20 months, re-
spectively. In all cases, yield was reduced less than 25%, yet 14, 30, and 45 Mg/ha of
salt were stored in the lower portions of the three different soil profile depths. Drastic
reductions in yield occurred only when the salt began to accumulate in the upper por-
tion of the root zone. This study demonstrates that regardless of soil depth, alfalfa can
be grown for a considerable period of time without leaching, provided the upper part of
the root zone, where most roots are concentrated, is maintained at a low level of salinity.
On a larger scale, the Broadview Irrigation District on the west side of California’s
San Joaquin Valley is a documented example of the effect of accumulating soil salin-
ity because of limited drainage (Wichelns et al., 1988). The district, consisting of 4000
ha of field crops, has had adequate imported surface water for irrigation since 1957.
This imported water contained approximately 300 mg/L of salt (ECi of 0.5 dS/m). An-
nual rainfall averages 150 mm. To facilitate leaching, subsurface drains were installed
on more than 80% of the irrigated land. The district, however, had no drainage outlet
until 1983, so it blended its surface runoff and subsurface drainage effluent with the
irrigation water supply. The proportion of drainwater mixed with the low-salinity irri-
gation water increased from near zero in the early 1960s to about half in the early
1980s, when the mean salt content of the drainage water was about 2800 mg/L. Thus,
although the fields were leached, the salts were reapplied to the field and no salt dis-
posal occurred. As illustrated in Table 7.7, crop selection switched to salt-tolerant
crops, such as cotton, to maintain economic income. The amount grown and the yield
of more salt-sensitive crops, such as tomato, dropped drastically as soil salinity in-
creased over time. Irrigators in nearby Fresno County, in contrast, did not blend drain-
age effluent with their irrigation water and they were able to maintain both tomato and
cotton yields throughout the same period of time.
7.4.3 Impact of Shallow Groundwater
The upward movement of shallow groundwater and its subsequent evaporation at
the soil surface leads to salination. To minimize the rate of salt accumulation, artificial
drainage is normally installed to lower the water table. The depth of the water table,
soil properties, and the rate of upward movement must be known to establish the ap-
propriate depth to maintain the water table. This information is also needed to estimate
the amount of groundwater available for consumption by plants.
7.4.3.1. Capillary flow. Starting from saturation, the drying rate at the soil surface
is first limited by the atmospheric evaporative conditions. As the soil surface dries, the
evaporation rate becomes limited by the rate of water movement to the soil surface in
the liquid phase. As the soil dries further, liquid movement ceases but vapor move-
ment upward continues, but is relatively unimportant. In arid and semiarid regions,
evaporative demands are high so water-transmitting properties of the soil normally
limit upward flow.
Gardner and Fireman (1958) studies the relationship between the rate of upward
flow and the depth of the water table. Their study verified a relation, proposed by
Gardner (1957), between hydraulic conductivity, K, and soil suction, S, of the form:
a
K= n (7.10)
S +b
where a, n, and b are constants. For many soils, values of n equal to 2 or 3 fit experi-
mental data well. Figure 7.10 gives the theoretical maximum rate of upward flow from
a stationary water table for two soils as a function of the depth to the water table.
The discussion above is valid when no drainage is discharged. When a net down-
ward flux is maintained by an appropriately designed drainage system, leaching oc-
curs. In this case groundwater salinity loses its significance as a criterion determining
the desired depth of drainage. The main criterion becomes the need to maintain a suf-
ficiently deep aerated root zone.
The water depth that will restrict upward flow to a minimum for these two soil
types can be determined from Figure 7.10. Lowering the water table from the surface
to a depth of about 1 m will be of little benefit in most soils. Upward flow at these
shallow depths could exceed 2.5 mm per day for clay soils and be even greater for
Figure 7.10. Maximum theoretical rate of upward water flow for Chino clay and Pachap-
pa sandy loam as a function of the depth of the water table (Gardner and Fireman, 1958).
coarser-textured soils. Lowering the water table from 1.2 m to 3 m in Pachappa sandy
loam decreases upward flow by a factor of 10. When the water table is at 2.5 m, fur-
ther lowering of the water table reduces upward flow only slightly. Keep in mind that
upward movement and evaporation of water from the surface of the soil are possible
even with a water table at a depth of 10 m. Harmful amounts of soluble salts could
slowly accumulate if the groundwater is sufficiently saline and rainfall and irrigations
amounts are sufficiently low. These results, verified by field observations, have led to
the installation of most subsurface drainage systems at depths of at least 1.5 m wher-
ever salinity poses a hazard.
7.4.3.2 Crop water use from groundwater. Water supplied to a crop by capillary
rise from shallow groundwater can be an important resource. Benefits of using this
water include reduced irrigation, lower production costs, the movement of less
groundwater to deeper aquifers, and a decrease in the amount of groundwater that re-
quires disposal through subsurface drainage systems. Although researchers have made
progress in determining the individual effects of salinity and water table depth on crop
water use, the combined effects of these two factors are not well understood. Some
experiments that have been conducted are presented, but generalizations are difficult
to make.
In a lysimeter study in Texas on Willacy fine sandy loam, Namken et al. (1969)
studied the impact of two irrigation regimes and three depths to the water table on
saline soil profiles (Figure 7.11). For the first year of the study, the salinity level of the
groundwater, ECG, was 6 to 8 dS/m. During the last three years, the ECG ranged from
Figure 7.11. Soil salinity distribution for different salt contents and
depths of groundwater (Namken et al., 1969).
0.9 to 1.6 dS/m. The groundwater supplied 57%, 38%, and 28% of the total water used
by a cotton crop when the water table was at depths of 0.9 m, 1.8 m, and 2.8 m, re-
spectively. When the water table was 1.8 m deep or deeper, the upper part of the pro-
file remained nonsaline, while the lower part became salinized. When the depth was
0.9 m, the salinity of the groundwater influenced the entire profile.
In a field study in the San Joaquin valley of California, cotton grown on a loam soil
received at least 60% of its evapotranspiration from a water table located 2.0 m to 2.5
m below the surface with an ECG of 6 dS/m (Wallender et al., 1979). The fewer the
number of irrigations, the more groundwater contributed to evapotranspiration. How-
ever, the lint yield was reduced.
Some of the most consistent data showing the relationship between evapotranspira-
tion provided from groundwater and water table depth have been obtained with cotton
(Figure 7.12). The relationship between cotton’s use of water from the groundwater
and the depth of the water table for soils ranging from clay to clay loam is based on
results from three different field experiments on the west side of California’s San Joa-
quin Valley (Grimes et al., 1984; Hanson and Kite, 1984; Ayars and Schoneman,
1986). The relationship for sandy loam is based on a lysimeter study in Texas (Nam-
ken et al., 1969). The impact of salinity is not included in Figure 7.12 but results from
California indicate that cotton’s uptake of groundwater is not reduced measurably until
ECG exceeds 12 dS/m.
Use of groundwater by alfalfa and corn varies from 15% to 60% of the total sea-
sonal use, but the data are too inconsistent to establish a relationship. In the Grand
Valley of Colorado, alfalfa’s use of groundwater from a water table depth of 0.6 m
varied from 46% to 94% of the total seasonal use in two different years, when ECG
equaled 0.7 dS/m (Kruse et al., 1986). Groundwater use varied from 23% to 91% of
Figure 7.12. Contribution of shallow, saline groundwater to evapotranspiration

(ET) of cotton as a function of soil type and depth to water table
(adapted from Hoffman et al., 1990).
the total seasonal use in other years, when ECG equaled 6 dS/m. In a joint study, Kruse
et al. (1985) reported that corn obtained 52% to 68% of its seasonal water requirement
when the water table was 0.6 m deep and obtained 25% to 32% of its seasonal water
requirement when the water table was 1 m deep. The proportion of use remained unaf-
fected when ECG varied from 0.7 dS/m to 6 dS/m.
7.5 SALINITY IMPACTS ON IRRIGATION DESIGN
7.5.1 Influence of Irrigation Method
The pattern of salt distribution within a given field depends on the variability in soil
properties, differences in water management, and the design of the irrigation system.
The soil salinity profile that develops as water is transpired or evaporated depends, in
part, on the water distribution pattern inherent with the irrigation method. Distinctly
different salinity profiles develop for different irrigation methods. Each irrigation
method has specific advantages and disadvantages for salinity management.
The major methods of flood irrigation are borders and basins. Border methods
commonly have excessive water penetration near the levees, at the upper end, and at
the lower end of the strips if surface drainage is prevented, and inadequate water pene-
tration midway down the strip, which may result in detrimental salt accumulation. If
insufficient amounts of water are supplied to one end of the basins, the far ends may
have excessive salt accumulations. The basin method of flooding has the potential for
more uniform water applications than other flooding methods provided the basins are
leveled, sized properly, and have uniform soils.
With furrow irrigation, salts tend to accumulate in the seed beds because leaching
occurs primarily below the furrows. If the surface soil is mixed between crops and the
irrigation water is not too saline, the increase in salt over several growing seasons may
not be serious. If excess salt does accumulate, leaching by another irrigation method
may be required or the application of special agronomic techniques (such as planting
on the side slopes of the beds, planting in the furrows, etc.) can be helpful. In the fur-
row and flood methods, the length of run, irrigation application rate, soil characteris-
tics, slope of the land, and time of application are the factors that govern the depth and
uniformity of application. Proper balance among these factors is beneficial in control-
ling salinity.
Sprinkler irrigation is unavoidably accompanied by wetting of crop foliage unless
below-canopy sprinkling is possible. Salts can be absorbed directly into the leaves,
thus some crops experience foliar injury and yield reductions that may not occur when
they are surface irrigated with the same water (Maas et al., 1982). Table 7.8 gives ap-
proximate concentrations of chloride and sodium in sprinkled waters that cause foliar
injury for some crops. Intermittent sprinkling with saline water frequently causes more
foliar injury than sprinkling continuously. This is attributed to the increased salt con-
Table 7.8. Relative susceptibility of crops to foliar injury from

saline sprinkling waters (adapted from Maas, 1990).
Sodium or Chloride Concentration (mol/m3) Causing Foliar Injury
<5 5 to 10 10 to 20 > 20
almond grape alfalfa cauliflower
apricot pepper barley cotton
citrus potato corn sugar beet
plum tomato sorghum sunflower
centration each time water evaporates from the leaves. Foliar injury can be particularly
acute if sprinkling is done during the day when high evaporative conditions exist.
Flooding and sprinkler irrigation systems that wet the entire soil surface create a
profile that at steady state increases in salinity with soil depth to the bottom of the root
zone, provided that moderate leaching is applied, application is uniform, and no shal-
low, saline groundwater is present. If irrigations are infrequent, the salt concentration
in the soil solution increases with time between irrigations, particularly near the soil
surface. In the presence of a shallow, saline water table and, particularly, with inap-
propriate irrigation management techniques, salts will accumulate near the soil sur-
face. Frequently, this situation leads to a severe salinity problem.
Trickle or drip irrigation systems where water is applied from point sources have
the advantage that high leaching is provided near the emitters and high soil water con-
tents can be maintained in the root zone by frequent but small water applications. Plant
roots tend to proliferate in the leached zone of high soil water content near the water
sources. This allows water of relatively high salt content to be used successfully in
many cases. Possible emitter clogging, the redistribution of water required to germi-
nate seeds, and the accumulation of the salts at the soil surface between emitters are
management concerns.
The salinity profile under line water sources, such as furrows and either porous or
multi-emitter drip irrigation systems, has lateral and downward components. The typi-
cal cross-sectional profile has an isolated pocket of accumulated salts at the soil sur-
face midway between the line sources of water and a second, deep zone of accumula-
tion, with the concentration depending on the amount of leaching. A leached zone
occurs directly beneath the line source. Its size depends on the irrigation rate, the
amount and frequency of irrigations, and the crop’s water extraction pattern.
Whereas the salt distribution from line sources increases laterally and downward,
the distribution from point irrigation sources, such as micro-basins and drip systems
with widely spaced emitters, increases radially from the water source in all directions
below the soil surface. As the rate of water application increases, the shape of the sa-
linity distribution changes. The mathematical model of Bresler (1975) predicts that the
salinity distribution in uniform, isotropic sand changes from elliptical (with the maxi-
mum movement vertical) to more circular as the rate of water movement decreases. In
isotropic and layered soils, the horizontal rate of water movement is greater than the
vertical, resulting in relatively shallow salt accumulations. For tree crops irrigated with
several drip emitters per tree, the wetting patterns may overlap, thereby reducing the
level of salt accumulation midway between the emitters under a tree.
The continuous upward water movement from a subsurface irrigation system re-
sults in salt accumulation near the soil surface as water is lost by evapotranspiration.
Subsurface systems provide no means of leaching these shallow accumulations unless
the soil is leached periodically by rainfall or surface irrigations.
The salt distribution profiles shown in Figure 7.13 are typical of those just de-
scribed. The lateral distribution of salts under sprinkler irrigation is relatively uniform,
with evaporation accounting for the salt accumulation near the soil surface. The salin-
ity distributions for both line source systems (furrow and drip) are similar, with salin-
ity levels relatively low beneath the water sources and relatively high midway between
the sources. Observed differences in salt accumulation below the side slopes of the
Figure 7.13. Influence of the irrigation system on the soil salinity pattern of bell pepper
at two levels of irrigation water quality (adapted from Bernstein and Francois, 1973).
planting bed are caused by additional leaching during furrow irrigation. Of course, the
more saline irrigation water accounts for higher salt concentrations for each case illus-
trated. Because the salt distributions were determined after only one irrigation season
for soil that had previously been well leached, salt had not yet accumulated at greater
soil depths.
7.5.2 Conjunctive Use of Waters
Historically, conservative salinity standards have been applied to assess the suit-
ability of water for irrigation. Consequently, saline waters, regardless of their origin,
have normally been avoided even as supplemental irrigation supplies. Using rational
water assessment procedures, however, saline waters, typically exported from irriga-
tion districts, still have potential value for irrigation (Rhoades, 1977, 1987). One situa-
tion for using saline irrigation water is in conjunction with nonsaline water. Where
high-quality water costs are prohibitive, crops of moderate to high salt tolerance can
be irrigated with saline water, especially at later growth stages, with economical ad-
vantages. This strategy may be economical even with some reduction in crop yields.
Irrigating with saline drainage water would permit the expansion of irrigated agricul-
ture while reducing drainage disposal and pollution problems (Rhoades, 1987).
Two strategies for conjunctive use of saline irrigation water are blending and sea-
sonal cycling with a source of low-salinity water. Each strategy has it advantages and
limitations.
7.5.2.1 Blending. The mixing of saline and nonsaline waters to obtain a composite
water suitable for irrigation is usually referred to as blending. The goals of blending
are to improve the suitability of the saline water and to increase the total water supply.
However, the entire volume of any saline water cannot be consumed in crop produc-
tion; the higher its salinity, the less that can be consumed. Blending is beneficial only
when the resulting mixed water has a salinity lower than the threshold salinity of the
crop and leaching is provided by rain or by irrigation. For example, the threshold sa-
linity of zucchini squash is 4.7 dS/m (see Table 7.4). Blending saline water of 5 dS/m
in equal volume with low-salinity water of 1 dS/m will yield water of 3 dS/m. The
resultant soil salinity (ECe) will be about 4.5 dS/m (ECe =1.5 × ECi) which is lower
than the threshold salinity. Blending doubles the supply of water for irrigation without
any loss in yield. Using the saline water alone would result in a 26% yield reduction
assuming the resultant soil salinity is 7.5 dS/m.
If this same blended water is used to irrigate strawberries, with a threshold salinity
value of 1.0 dS/m, the resultant yield would be nil. Consequently, the portion of low-
salinity water used in the blend, which could have resulted in only a 5% yield loss on
half the area, is made useless. The low-salinity water can be used to irrigate corn with-
out yield loss because its threshold salinity is 1.7 dS/m. Irrigating with saline water
alone will result in 70% yield reduction while the reduction with the blended water
will be 35%. The same 35% yield reduction would result from irrigating half of the
area with nonsaline water and half with saline water without the need of blending.
Blending in this case offers no advantage. Additionally, for some crops the blended
water will require leaching, thereby expending more nonsaline water.
If the blending strategy is adopted, two mixing processes, network dilution or soil
dilution, are possible (Shalhevet, 1984). With network dilution, water supplies are
blended in the irrigation conveyance system. With soil dilution, the soil acts as the
mixing medium. Meiri et al. (1986), conducting a three-year study in Israel on a rota-
tion of potato and peanut under drip irrigation, concluded that crops responded to the
weighted mean water salinity, regardless of the blending method. Other examples of
blending techniques are given by Rains et al. (1987) and Rolston et al. (1988).
7.5.2.2 Cycling. The cycling strategy utilizes low-salinity water for pre-plant and
early irrigation of salt-tolerant crops in the rotation and for all irrigations of salt-
sensitive crops. The saline water is used to irrigate salt-tolerant crops at later stages of
growth. The timing and amount of substitution will vary with salt tolerance of the
crops grown, the quality of the two waters, the climate, and the irrigation system.
Whatever salt buildup occurs in the soil from irrigating with saline water is alleviated
in a subsequent cropping period when nonsaline water is applied. Furthermore, the
yield of the sensitive crop is not reduced if proper pre-plant irrigations and appropriate
water management are used during germination and seedling establishment to leach
salts out of the seed area and shallow soil depths. Subsequent irrigations with non-
saline water leach these salts farther down in the profile ahead of the advancing root
system and reclaim the soil. The cyclic strategy prevents the soil from becoming ex-
cessively saline while permitting, over the long term, substitution of saline water for a
significant fraction of the irrigation water requirements.
This strategy has been tested for two cropping patterns on a farm in the Imperial
Valley of California (Rhoades et al., 1988). One test consisted of two successive rota-
tions of wheat, sugar beet and cantaloupe. Wheat and sugar beet are relatively salt
tolerant and cantaloupe is relatively salt sensitive. Results for this successive rotation
are given in Table 7.9. In this rotation, Colorado River water (1.2 dS/m, relatively
nonsaline water) was used throughout for the control treatment (C) and for the pre-
plant and early irrigations of wheat and sugar beet in the other two treatments (Ca and
Ac) and for all irrigations during the cantaloupe crop. The saline water (A) was from
the Alamo River which had an EC of 4 dS/m. During the two rotations, 35% of the
irrigation water was saline for treatment Ca while 53% was saline for treatment cA.
No significant yield losses of wheat or sugar beet occurred in either cycle of the rota-
tion from substituting saline water for Colorado River water after seedling establish-
ment. Furthermore, the commercial yield of fresh-market cantaloupe in the fields that
had previously been irrigated with saline water was not reduced significantly.
The second cropping pattern tested was a four-year block rotation consisting of two
years of cotton followed by wheat (both relatively salt-tolerant crops) and then by al-
falfa (a relatively salt-sensitive crop). For the two cotton crops, the three treatments
were Colorado River water only (C), nonsaline water for seedling establishment fol-
lowed by saline water (cA), and saline water from the Alamo River only (A). Colo-
rado River water was used exclusively for wheat and alfalfa in all three treatments.
During this four-year rotation, 25% of the irrigation water was saline for treatment cA
and 45% was saline for treatment A. Lint cotton yield in 1982 was not reduced from
use of saline water (see Table 7.9) even when it was applied during pre-plant and
seedling establishment periods. In the second cotton crop (1983) there was no yield
loss from use of saline water following seedling establishment. A significant loss of
Table 7.9. Yields of crops in two rotations cycling saline and nonsaline irrigation
waters in the Imperial Valley of California (adapted from Rhoades et al., 1988).
Successive Rotation
Wheat Sugar beet Cantaloupe Wheat Sugar beet Cantaloupe
1982 1982-83 1983 1984 1984-85 1985
(Mg grain/ha) (Mg sugar/ha) (kg seed/ha) (Mg grain/ha) (Mg sugar/ha) (cartons/ha)
Irrigation Water Quality
C 8.1 9.7 440 7.9 9.2 435
Ca 8.1 9.7 430 7.8 9.2 535
cA 8.3 9.2 400 8.0 8.8 525
Block Rotation
Cotton Cotton Wheat Alfalfa
1982 1983 1984 1985
(Mg lint/ha) (Mg lint/ha) (Mg grain/ha) (Mg forage/ha)
Irrigation Water Quality
C 1.4 1.1 7.7 17.5
cA[a] 1.4 1.1 7.7 15.7
A[a] 1.5 0.7 7.7 16.6
[a]
All treatments of wheat and alfalfa irrigated with Colorado River water only.
lint yield did occur in 1983 from using saline water for all irrigations (treatment A).
This yield reduction was caused by excess salinity in the seed bed during the estab-
lishment period, which reduced plant population. No loss in yield of wheat grain or
alfalfa forage occurred from the previous use of saline water on cotton when irrigated
with Colorado River water. Studies by Ayars et al. (1986) and Shennon et al. (1987)
showed similar results.
There may be difficulty in adopting the cyclic strategy if the availability of saline
water does not coincide with peak crop water use. In the San Joaquin Valley of Cali-
fornia, drain effluent occurs primarily from January to June when most crops require
high-quality water. Using drainage water late in the season may not be feasible if the
flow rate needed to irrigate a field effectively exceeds the flow rate from the drains.
To avoid this lack of drainage water, a storage reservoir could be constructed. Another
option is to close the drains and allow the soil to act as the reservoir and then pump as
necessary. The latter option is more desirable since it does not take land out of produc-
tion. A third option is to develop a reuse system on a regional basis with separate
dedicated areas of drainage water collection and use. However, regardless of where
the drainage water is stored, a drainage water collection and distribution system must
be constructed to implement this strategy (Grattan and Rhoades, 1990).
7.5.3 Environmental Consequences
Irrigating with saline water to alleviate an environmental concern is possible, par-
ticularly where saline water disposal is impractical due to physical, environmental, or
social constraints. Because many crops are sensitive to shallow or fluctuating water
tables, a means of lowering the shallow water table is essential to sustain crop produc-
tion. Drainage water, collected to lower the water table, can frequently be used for
irrigation. This reduces the amount of nonsaline water required for salt-tolerant crops,
and decreases the volume of drainage water requiring disposal or treatment. Many
growers in the San Joaquin Valley of California, for example, are considering this
option as a solution to reducing drainage volume in response to an environmental con-
cern.
Another potential use of saline water is the irrigation of specific crops that have the
ability to accumulate large quantities of undesirable constituents (e.g., selenium, ni-
trate, boron, molybdenum) before the subsequent discharge of drainage effluent. A
number of plants can be used for biofiltration, the term used to describe this process
(Cervinka et al., 1987; Wu et al., 1988). For example, they found that certain grasses
and native species in California are effective in accumulating substantial amounts of
selenium in their shoots. Potentially toxic constituents could therefore be removed by
harvesting the shoots. This alternative management practice is most attractive where
drainage disposal is an environmental concern, the bioaccumulator has economic
value, or other treatment processes are unavailable or too expensive.
Saline waters may contain specific solutes, such as boron and chloride, that can ac-
cumulate in plants to levels that cause foliar injury and a yield reduction. In portions
of the San Joaquin Valley of California, shallow groundwaters often contain boron
concentrations in excess of 5 mg/L. The removal of boron from the soil requires far
more water for leaching than does removal of salts. Thus, the environmental conse-
quences of excess boron can be more detrimental than salinity.
7.6 SALINITY MANAGEMENT PRACTICES

7.6.1 Irrigation Scheduling
Irrigation scheduling involves timing as well as the quantity of water to be applied
for each irrigation and for the entire season. Scheduling of irrigations may require
modification for saline waters.
7.6.1.1 Irrigation water requirement. As discussed in Section 7.3.4, the slope of
the crop-water production function for saline water is the same as for nonsaline water,
provided leaching water is not applied during the irrigation season. Leaching by irriga-
tion is frequently postponed and accomplished following the crop growing season or
from winter rainfall.
If the irrigation water is moderately saline and maximum yields can be maintained
by leaching, the seasonal water requirement can be estimated from the crop-water pro-
duction function and the leaching requirement. When salinity is expected to reduce
crop productivity, the yield decrement can be estimated from the salinity response
function (see Equation 7.4). The water requirement can be estimated from the water
production function. A simplified approach combines the water production function as
defined by Stewart et al. (1976) with the salinity response function to obtain:
ETr = 1– (s/c) (ECe – t) (7.11)
where ETr = relative evapotranspiration
t = threshold of the salinity response function
s = slope of the salinity response function
c = slope of the water production function.
7.6.1.2 Irrigation interval. As a soil dries between irrigations, both the osmotic
and matric potentials decrease. The rate at which these processes occur depends on ET
and on the water desorption properties of the soil. Shortening the interval between
irrigations causes a higher pre-irrigation soil water content and a lower pre-irrigation
salt concentration; both features enhance plant growth.
Intuitively, one expects this favorable effect to be greater for saline than for non-
saline irrigation water. However, several processes occur which result in a smaller
response to irrigation interval under saline than under nonsaline conditions. More fre-
quent irrigations result in an upward shift in the depth of the peak salt concentration
into regions of greater root concentration (Bernstein and Francois, 1973). In addition
to salt loading from irrigating to satisfy crop water requirements, the salt load in the
soil also increases because of the need to apply additional water to compensate for
leaching. Under sprinkler irrigation, reducing the irrigation interval may result in more
severe leaf damage because of the more frequent wetting of the foliage. More signifi-
cant, however, is the fact that salinity reduces ET, resulting in a slower rate of soil
drying. Thus, for the same irrigation interval the total pre-irrigation soil water poten-
tial (matric plus osmotic) may be lower, causing a greater yield reduction under non-
saline than under saline conditions (Figure 7.14). In this case, reducing the irrigation
interval will be more beneficial under irrigation with nonsaline than saline water.
Field experiments in three different soils in two ecological regions of Israel tested
the impact of irrigation interval and irrigation water salinities on eggplant (Shalhevet
et al., 1983) and corn (Shalhevet et al., 1986). A unified function of relative yield ver-
sus mean root zone salinity could be used for both crops for all irrigation intervals
tested. The results for the corn experiment are presented in Figure 7.15 for irrigation
intervals from 3.5 to 21 days.
Figure 7.14. Soil matric potential as a function of time following an irrigation for
waters having an electrical conductivity (adapted from Shalhevet et al., 1986).
Figure 7.15. Field experiment showing the lack of impact from irrigation interval on
corn yields as a function of soil salinity created by different irrigation water qualities
(adapted from Shalhevet et al., 1986).
Most experimental results show either no difference in relative yield response or a

lower response to more frequent irrigations with saline than with nonsaline water. No
differences in response were shown by Ayers et al. (1943) with kidney beans, Bern-
stein and Francois (1975) with beans, Greenway (1965) with citrus, and Hoffman et al.
(1983a) with tall fescue. Lower response was shown by Ayoub (1977) with senna,
Bernstein and Francois (1975) and Goldberg and Shimueli (1971) with pepper, and
Wagenet et al. (1980) with barley. Only guayule (Wadleigh et al., 1946) showed an
advantage of increasing irrigation frequency with saline compared with nonsaline wa-
ter. The conclusion from this analysis is that recommendations given regarding irriga-
tion frequency with nonsaline water are suitable for saline water irrigation as well. Most
studies show no benefit of decreasing the irrigation interval under saline conditions.
7.6.2 Crop Considerations
A few crop management options are available to counteract a possible reduction in
yield from salinity. These options include crop selection, seed placement, and planting
density.
7.6.2.1 Crop selection. As described in Section 7.3.1, crops vary widely in their
tolerance to saline conditions. Crops can be selected from Table 7.4 to withstand the
restrictions imposed by the expected soil salinity. Some leaching is frequently pro-
vided by irrigation inefficiency or excess expectations of crop water use. When sig-
nificant leaching is required, especially when drainage or infiltration is a problem, it
might be preferable to select more tolerant crops than to depend on leaching to control
soil salinity. When the irrigation water is high in a potentially toxic constituent, such
as chloride or boron, crops selected must be tolerant to these specific solutes.
When nonsaline water is available, along with saline water, more tolerant crops,
which are generally of lower economic value, may be selected to be irrigated with the
saline source while the nonsaline water is applied to more valuable salt-sensitive
crops. Leaching of accumulated salts can be provided during the periods of nonsaline
irrigation provided the leaching requirement is met (see Section 7.4.1).
7.6.2.2 Seed placement. The pattern of salt accumulation within the seed bed for
furrow irrigation is critical to seed germination (Bernstein et al., 1955) Placing a single
or double seed row on the shoulder of the bed close to the water line, rather than in the
center of the bed, places the seed in an area of the lowest salt concentration (see Figure
7.16). When alternate-furrow irrigation is practiced, seeds should be placed in the bed
center or on the side of the bed away from the dry furrow. A sloping bed configuration
with the seeds planted on the sloping side is an alternative solution to the salt accumu-
lation problem in furrow irrigation (Bernstein and Fireman, 1957).
Under drip irrigation, care must be exercised not to place seeds between last sea-
son’s points of irrigation where salt accumulation may prevent germination. When
precision seeding is not feasible, sprinkler irrigation should be used to leach salts out
of the surface soil if off-season rainfall has been inadequate to provide the required
leaching.
7.6.2.3 Planting density. As pointed out previously, the most distinct symptom of
salinity damage to crops is stunted growth. This reduction in growth may be exploited
to overcome yield reduction in some crops by increasing the number of plants per unit
area (Francois, 1982). Increasing planting density should be done by decreasing inter-
row spacing rather than increasing intra-row spacing of plants (Keren et al., 1983). To
date, this idea has only been tested on cotton.
Figure 7.16. Typical salt accumulation patterns in ridge and bed cross-sections
under furrow irrigation (Bernstein et al., 1955; Bernstein and Fireman, 1957).
7.6.3 Infiltration
Infiltration refers to the entry of water into the soil. An infiltration rate of 3 mm/h
or less is considered low for many situations while a rate of more than 12 mm/h is
relatively high. Infiltration can be affected by many factors other than water quality,
including soil texture, the type of clay minerals, and exchangeable cations. Only infil-
tration problems caused by chemical constituents are discussed here. Figure 7.3 illus-
trates that both salinity and the sodium-adsorption ratio of the applied water influence
infiltration.
Water with an electrical conductivity less than 0.5 dS/m and especially below 0.2
dS/m tends to leach surface soils free of soluble minerals and salts, especially calcium
and magnesium. This process reduces the stabilizing influence of salts on soil aggre-
gates and soil structure. Without salts, and in particular without calcium and magne-
sium, the soil disperses, and the displaced finer soil particles fill the smaller soil pore
spaces, sealing the surface and reducing infiltration. Soil crusting and seedling emer-
gence problems often result. Rainfall is a very low-salinity water and irrigated areas
frequently experience exceptionally low rates of infiltration from rainwater and sig-
nificant amounts of surface ponding, runoff, and erosion can occur.
Management options to improve infiltration can be based on either chemical or
physical changes. Chemical practices involve altering the soil or water chemistry that
influences infiltration. This is normally accomplished by adding a chemical amend-
ment, such as gypsum, or blending sources of water with different chemical composi-
tions to reduce the potential hazard. Physical practices include cultivation, deep till-
age, and preserving crop residues.
Chemical amendments added to soil or water to improve infiltration increase the
soluble calcium content and/or increase the salinity of the applied water significantly.
Amendments will not help if the cause of inadequate infiltration is soil texture, soil
compaction, restrictive soil layers, or a high water table. Most amendments supply
calcium directly (i.e., gypsum) or indirectly by applying an acid or acid-forming sub-
stance (i.e., sulphuric acid or sulphur) that reacts with lime (CaCO3) in the soil to re-
lease calcium into the soil solution. If lime is not present, acids or acid-forming mate-
rials will not be beneficial. Comparative data for the more common amendments are
given in Table 7.10. Amendments are expensive, and they are only justified if their
use, substantiated by field trials, results in a significant improvement. Water amend-
ments are most effective if the infiltration problem is caused by a low-salinity water
(ECi < 0.2 dS/m) or by high SAR in a water of low to moderate salinity (ECi < 1
dS/m). If water salinity is moderate to high (ECi > 1 dS/m) and SAR is high, soil-
applied amendments are often more effective.
Gypsum is the most commonly used and widely available amendment. In practice it
is unusual to get more than 0.5 to 2 mol/m3 of calcium dissolved from gypsum into an
irrigation water supply. These relatively small amounts of calcium in a low-salinity
water, however, may increase infiltration by as much as 100% to 300%. If the water is
relatively saline, these small amounts of calcium are far less effective and increase
infiltration to a much smaller degree. The rate at which gypsum goes into solution is
dependent upon its particle size. Finely ground gypsum (less than 0.25 mm in diame-
ter) dissolves much more rapidly than coarse material. The coarsely ground, low-
purity forms of gypsum are more satisfactory for application to the soil; given time for
dissolution, low grades of gypsum have been applied successfully in the irrigation water.
Tillage to create a rough, yet thoroughly disturbed, soil surface is frequently an ef-
fective practice for improving infiltration. Cultivation prior to an irrigation roughens
the soil and opens cracks and air spaces that increase the surface area exposed for in-
filtration. Deep tillage (chiseling, subsoiling) for many soils, particularly fine-textured
soils, improves deep water penetration for only a few irrigations because the soil sur-
Table 7.10. Chemical properties of various amendments for reclaiming sodic soil.
Amount
Solubility in Equivalent to
Chemical Physical Cold Water 1 kg of 100%
Amendment Composition Description (kg/m3) Gypsum (kg)
Gypsum .
CaSO4 2H2O white mineral 2.4 1.0
Sulfur S8 yellow element 0 0.2
Sulfuric acid H2SO4 corrosive liquid Very High 0.6
Lime sulfur 9% Ca + 24% S yellow-brown Very high 0.8
solution
Calcium carbonate CaCO3 white mineral 0.014 0.6
Calcium chloride CaCl2.2H2O white salt 977 0.9
Ferrous sulfate FeSO4.7H2O blue-green salt 156 1.6
Pyrite FeS2 yellow-black mineral 0.005 0.5
Ferric sulfate Fe2(SO4).9H2O yellow-brown salt 4400 0.6
Aluminum sulfate Al2(SO4)3.18H2O corrosive granules 869 1.3
face quickly returns to its original condition. Although improvement is not permanent,
deep tillage may temporarily allow sufficient water to enter to make an appreciable
improvement in stored soil water and in crop yield. Deep tillage, done only when the
soil is relatively dry to prevent compaction, physically tears, shatters, and rips the soil.
It is done prior to planting or during dormancy of permanent crops.
The preservation or application of crop residue, livestock manure, and other or-
ganic matter on the soil surface will improve infiltration and is a widely accepted prac-
tice. The fibrous and less easily decomposed crop residues, such as from barley, rice,
wheat, corn, and sorghum, have improved water penetration markedly, whereas resi-
dues from legumes and vegetable crops have been less effective. Relatively large
quantities of residues are needed; for instance, manure has been added at rates of 40 to
400 Mg/ha. An application of organic residues in the range of 10% to 30% by soil
volume in the upper soil layer may be needed to be effective.
7.6.4 Reclamation of Salt-Affected Soils
Soils may be naturally saline or suffer from salination because of irrigation mis-
management or inadequate drainage. Such soils require reclamation before irrigated
agriculture can be profitable. The only proven method of reclaiming salt-affected soils
is by leaching. Sodic soils normally require the addition of an amendment or tillage to
promote the leaching process. Soils high in boron are particularly difficult to reclaim
because of the tenacity by which boron is held in the soil. Leaching can be enhanced
by growing very salt-tolerant plants during the reclamation process. Reclaiming saline
soils by harvesting plants with the salts they have taken up is not practical. Likewise,
flushing water over the soil surface to remove visible crusts of salt is not effective.
Adequate drainage and suitable disposal of the leaching water are absolute prerequi-
sites for reclamation.
7.6.4.1 Saline soils. The amount of water required for the removal of salts from a
saline soil depends on the initial level of salinity, soil physical characteristics, tech-
nique of applying water, and soil water content. Leaching by continuous flooding is
the fastest method but it requires larger quantities of water than leaching by sprinkler
or by intermittent flooding. Intermittent flooding or sprinkling, though slower, is more
efficient and will require less water, particularly on finer-textured soils, than continu-
ous ponding.
The relationship between the fraction of the initial salt concentration remaining in
the soil profile, c/co, and the amount of water leaching though the profile per unit
depth of soil, dL/dS, when water is ponded continuously on the soil surface can be de-
scribed by
(c/co) (dL/dS) = K (7.12)
where K is a constant that differs with soil type. Equation 7.12 defines the curves in
Figure 7.17 for organic (peat) soils when K = 0.45, for fine-textured (clay loam) soils
when K = 0.3, and for coarse-textured (sandy loam) soils when K = 0.1. Equation 7.12
is valid when dL/dS exceeds K. Figure 7.17 summarizes the results of nine field ex-
periments (Hoffman, 1986).
The amount of water required for leaching soluble salts, particularly for fine-
textured soils, can be reduced by intermittent applications of ponded water or by sprin-
kling. The results of three leaching trials by intermittent ponding where no water table
was present are summarized in Figure 7.18. Agreement among experiments is excel-
Figure 7.17. Depth of leaching water per unit depth of soil required to reclaim
a saline soil by continuous ponding (adapted from Hoffman, 1986).
Figure 7.18. Depth of leaching water per unit depth of soil required to reclaim a
saline soil by ponding water intermittently (adapted from Hoffman, 1986).
lent, considering the variety of soil textures and the depth of water applied each cycle
(50 to 150 mm) with corresponding ponding intervals varying from weekly to
monthly. The relationship between c/co and dL/dS for intermittent ponding can be ap-
proximated by
(c/ co) (dL/dS) = 0.1 (7.13)
For fine-textured soils, intermittent ponding requires only about one-third as much
water to remove about 70% of the soluble salts initially present, compared to continu-
ous ponding.
Reclamation by flooding may be relatively inefficient when tile or open drains are
present as most of the water flow will take place through the soil near the drains, while
midway between the drains there will be far less leaching. The regions between drains
should be leached by basins separated from the regions above the drains, or leaching
should be done by sprinklers to improve efficiency (Luthin et al., 1969).
When a soil is both saline and sodic, the initial removal of salts may create sodic
conditions and reduce permeability. The equivalent dilution method, which is based on
the principle that divalent cations tend to replace monovalent cations on the exchange
complex as a result of the dilutions of the soil solution, was proposed to overcome this
problem (Reeve and Doering, 1966). By successive dilution of initially applied high-
salt water, sodic soils can be reclaimed without a reduction in infiltration capacity.
7.6.4.2 Sodic soils. The reclamation of sodic soils is very difficult, if not impossi-
ble, without the use of chemical amendments to replace exchangeable sodium by cal-
cium. In some cases, the equivalent dilution method can be used to reclaim sodic soils.
Properties of typical amendments are given in Table 7.10. In a calcarious soil, the ad-
dition of acids or acid-forming materials may dissolve sufficient lime to provide suffi-
cient exchangeable calcium. Gypsum, however, is the most common additive because
of its low cost, good solubility, and availability.
The gypsum requirement to reclaim a sodic soil depends on the amount of ex-
changeable sodium (Na) to be replaced by calcium. It can be calculated from (Keren
and Miyamoto, 1990)
GR = 0.86 dsDb (CEC) (ENai – ENaf) (7.14)
where GR = gypsum requirement
dS = depth of soil to be reclaimed, m
Db = soil bulk density, Mg/m3
CEC = cation exchange capacity of the soil, mol/kg
ENai and ENaf = initial and desired final exchangeable sodium fractions.
According to this equation about 10 Mg/ha of gypsum will replace 30 mol/kg of Na to
a depth of 0.3 m for a soil having a bulk density of 1.32 Mg/m3. Typically, a 1-cm
depth of water per hectare will dissolve about 0.25 Mg of gypsum. Thus, a 40-cm
depth of water will be required to reclaim the soil in the above example to a depth of
30 cm. It is difficult to reclaim a deep sodic soil profile in a single leaching operation.
The usual procedure is to partially reclaim the soil during the first year, then plant a
shallow-rooted crop and continue the reclamation process in the following years until
the entire profile is reclaimed (Keren and Miyamoto, 1990).
7.6.4.3 Boron leaching. Like other salts, excess boron must be removed. More wa-
ter is needed to leach boron than other salts because it is tightly adsorbed on soil parti-
cles. Similar to leaching soluble salts, the relationship between c/co and dL/dS can be
approximated by
(c/co) (dL/dS) = 0.6 (7.15)

In addition, periodic leachings may be required to remove additional boron released
from the soil particles over time (Oster et al., 1984).
7.7 SUMMARY AND CONCLUSIONS
The design and management of irrigation systems must be adequate to prevent
harmful accumulations of salt in the crop root zone. Timely irrigations must be of suf-
ficient quantity and uniformity to meet the crop’s needs and leach salts adequately,
without excessive surface runoff or deep percolation. All irrigation water contains salt;
as evapotranspiration proceeds, the salt concentration increases in the remaining soil
water. Without proper irrigation management, the land can become waterlogged and
salinized. Regardless of management, drainage water from irrigated lands carries salt
that requires appropriate disposal.
The response of crops to salinity, sodicity, and toxicity varies widely among plant
species. The relationship between crop yield and soil salinity has been quantified for
many crops under typical growing conditions. This relationship, however, depends on
a number of soil, crop, and environmental factors. Sodicity typically reduces infiltra-
tion, which leads to reduced crop yields. Crops can also be sensitive to specific sol-
utes, such as chloride, sodium, and boron. With proper crop selection and appropriate
irrigation management, economic yields are usually possible under low to moderate
saline conditions.
Salt-affected soils can be reclaimed through leaching. Sodic soils normally require
a chemical amendment that supplies calcium or magnesium to replace sodium. Copi-
ous amounts of water are frequently required to reclaim a salt-affected soil.
REFERENCES
Armillas, P. 1961. Land use in pre-columbian America. In A History of Land Use in
Arid Regions, 225-276. L. D. Stamp, ed. Paris, France: UNESCO Arid Zone
Research, 17.
Aubertin, G. M., W. R. Rickman, and J. Letey. 1968. Differential salt-oxygen levels
influence plant growth. Agron. J. 60: 345-353.
Ayars, J. E. and R. A. Schoneman. 1986. Use of saline water from a shallow water
table by cotton. Trans. ASAE 29: 1674-1678.
Ayars, J. E., R. B. Hutmacher, R. A. Schoneman, S. S. Vail, and D. Felleke. 1986.
Drip irrigation of cotton with saline drainage water. Trans. ASAE 29(6): 1668-1673.
Ayers, A. D., C. H. Wadleigh, and O. C. Magitad. 1943. The interrelationship of salt
concentration and soil moisture content with the growth of beans. American Soc.
Agron. J. 35: 796-810.
Ayoub, A. T. 1977. Some primary features of salt tolerance in senna (Cassia
acutifolia). J. Exp. Bot. 28: 484-492.
Bernstein, L., and M. Fireman. 1957. Laboratory studies on salt distribution of furrow
irrigated soil with special reference to pre-emergence period. Soil Sci. 83: 249-263.
Bernstein, L., and L. E. Francois. 1973. Leaching requirement studies: Sensitivity of
alfalfa to salinity of irrigation and drainage waters. Soil Sci. Soc. America Proc. 37:
931-943.
Bernstein, L., and L. E. Francois. 1975. Effect of frequency of sprinkling with saline
waters compared with daily drip irrigation. Agron. J. 67: 185-190.
Bernstein, L., M. Fireman, and R. C. Reeve. 1955. Control of salinity in the Imperial
Valley, California. USDA, ARS 41(4).
Bingham, F. T., and M. J. Garber. 1970. Zonal salinization of the root system with
NaCl and boron in relation to growth and water uptake of corn plants. Soil Sci. Soc.
America Proc. 34: 122-126.
Bingham, F. T., J. E. Strong, J. D. Rhoades, and R. Keren. 1985. An application of the
Maas-Hoffman salinity response model for boron toxicity. Soil Sci. Soc. America J.
69: 672-674.
Bower, C. A., G. Ogata, and J. M. Tucker. 1969. Root zone salt profiles and alfalfa
growth as influenced by irrigation water salinity and leaching fraction. Agron. J.
61: 783-785.
Boyden, S. 1987. Western Civilization in Biological Perspective: Patterns in
Biohistory. Oxford, U.K.: Oxford Univ. Press.
Bresler, E. 1975. Two-dimensional transport of solutes during non-steady infiltration
from a trickle source. Soil Sci. Soc. America Proc. 39: 604-613.
Bressler, M. B. 1979. The use of saline water for irrigation in the U.S.S.R. Joint
Commission on Scientific and Technical Cooperation. Water Resources.
Cervinka, V., C. Finch, J. Beyer, F. Menezens, and R. Ramirez. 1987. The agro-forestry
demonstration program in the San Joaquin Valley. Progress Report. Sacramento,
Calif.: Calif. Dept. of Food and Agriculture, Agricultural Resources Branch.
Childs, S. W., and R. J. Hanks. 1975. Model for soil salinity effects on crop growth.
Soil Sci. Soc. America Proc. 39: 617-622.
Dalton, F. N. 1992. Development of time-domain reflectometry for measuring soil
water content and bulk soil electrical conductivity. In Advances in Measurement of
Soil Physical Properties Bringing Theory into Practice, 143-167. C. Topp, and R.
Green, eds. ASA Special Publication No. 30. Madison, Wis.: Soil Sci. Soc. America.
de Wit, C. T. 1958. Transpiration and Crop Yields. Verslag Landbouck Onderzoek
No. 64.6.Wageningen, The Netherlands.
Dhir, R. D. 1977. Investigations into use of highly saline waters in an arid
environment. I: Salinity and alkali hazard conditions in soil under a cyclic
management system. In Proc. Intl. Salinity Conf. on Managing Saline Water for
Irrigation, 608-609. Lubbock, Tex.: Texas Tech. Univ.
Drew, M. C., J. Guenther, and A. Lauchli. 1988. The combined effect of salinity and
root anoxia on growth and net Na+ and K+- accumulation in Zea mays growth in
solution culture. Ann. Bot. 62: 41-53.
Eaton, F. M. 1944. Deficiency, toxicity, and accumulation of boron in plants. J. Agric.
Res. 69: 237-277.
Ehlig, C. F. 1960. Effect of salinity on four varieties of table grapes grown in sand
culture. American Soc. Hort. Sci. 76: 323-331.
Epstein, E., J. D. Norlyn, D. W. Rush, R. W. Kingsbury, R. W. Kelley, G. A.
Cunningham, and A. F. Wrona. 1980. Saline culture of crops: A genetic approach.
Science 210: 399-404.
Fang, S., Y. Tian, and D. Xin. 1978. Comprehensive control of drought, waterlogging
salinization, and saline groundwater. In Selected Works of Symposium on the
Reclamation of Salt-Affected Soil in China. Shandong Publ. House of Scientific
Technology.
Francois, L. E. 1981. Alfalfa management under saline conditions with zero leaching.
Agron. J. 73: 1042-1046.

Francois, L. E. 1982. Narrow row cotton (Gossypium hirsutum L.) under saline
conditions. Irrig. Sci. 3: 149-156.
Francois, L. E. 1984. Effect of excess boron on tomato yield, fruit size, and vegetative
growth. J. American Soc. Hort. Sci. 109: 322-324.
Francois, L. E. 1986. Effect of excess boron on broccoli, cauliflower, and radish. J.
American Soc. Hort. Sci. 111: 494-498.
Francois, L. E., and E. V. Maas. 1994. Crop response and management on salt-
effected soils. In Handbook of Plant and Crop Stress, 149-181. M. Passaraklie, ed.
New York, N.Y.: Marcel Dekker.
Francois, L. E., C. M. Grieve, E. V. Maas, and S. M. Lesch. 1994. Time and salt stress
affect growth and yield components of irrigated wheat. Agro. J. 86: 100-107.
Gardner, W. R. 1957. Some steady-state solutions of the unsaturated moisture flow
equation with application to evaporation from a water table. Soil Sci. 85(4): 228-232.
Gardner, W. R., and M. Fireman. 1958. Laboratory studies of evaporation from soil
columns in the presence of a water table. Soil Sci. 85(5): 244-249.
Gelburd, D. E. 1985. Managing salinity lessons from the past. J. Soil Water Cons. 40:
329-331.
Goldberg, D., and M. Shimueli. 1971. Sprinkler and trickle irrigation of green pepper
in an arid zone. Hort. Sci. 6: 559-564.
Grattan, S. R., and J. D. Rhoades. 1990. Irrigation with saline ground water and
drainage water. In Agricultural Salinity Assessment and Management, 432-449. K.
K. Tanji, ed. New York, N.Y.: American Soc. Civil Engineers.
Greenway, H. 1965. Plant response to saline substrates. VII Aust. Biol. Sci. 18: 763-779.
Grimes, D. W., R. L. Sharma, and D. W. Henderson. 1984. Developing the Resource
Potential of a Shallow Water Table. Contribution No. 188. Davis, Calif.: Calif.
Water Resource Center, Univ. of California.
Hanson, B. R., and S. R. Grattan. 1990. Field sampling of soil, water, and plants. In
Agricultural Salinity Assessment and Management, 186-200. K. K. Tanji, ed.
ASCE Manuals and Reports on Engineering Practices. No. 71. New York, N.Y.:
American Soc. Civil Engineers.
Hanson, B. R., and S. W. Kite. 1984. Irrigation scheduling under saline high water
tables. Trans. ASAE 27: 1430-1434.
Hoffman, G. J. 1985. Drainage required to manage salinity. J. Irrig. Drain. Eng. 111:
199-206.
Hoffman, G. J. 1986. Guidelines for reclamation of salt-affected soils. Applied Agric.
Res. 1(2): 65-72.
Hoffman, G. J., and J. A. Jobes. 1978. Growth and water relations of cereal crops as
influenced by salinity and relative humidity. Agron. J. 70: 765-769.
Hoffman, G. J., and M. T. van Genuchten. 1983. Soil properties and efficient water
use: Water management for salinity control. In Limitations to Efficient Water Use
in Crop Production, 73-85. H. M. Taylor, W. Jordan, and T. Sinclair, eds. Madison,
Wis.: American Soc. Agronomy.
Hoffman, G. J., J. A. Jobes, and W. J. Alves. 1983a. Response of tall fescue to
irrigation water salinity, leaching, fraction and irrigation frequency. Agric. Water
Mgmt. 7: 439-456.
Hoffman, G. J., E. V. Maas, and S. L. Rawlins. 1973. Salinity-ozone interactive effect
on yield and water relation of pinto beans. J. Environ. Quality 2: 148-152.

Hoffman, G. J., E. V. Maas, T. L. Prichard, and J. L. Meyer. 1983b. Salt tolerance of
corn in the Sacramento-San Joaquin Delta of California. Irrig. Sci. 4: 31-44.
Hoffman, G. J., J. A. Jobes, Z. Hanscom, and E. V. Maas. 1978. Timing of
environmental stress affects growth, water relation and salt tolerance of Pinto
beans. Trans. ASAE 21: 713-718.
Hoffman, G. J., S. L. Rawlins, M. J. Garber, and E. M. Cullen. 1971. Water relations
and growth of cotton as influenced by salinity and relative humidity. Agron. J. 63:
822-826.
Hoffman, G. J., J. D. Rhoades, J. Letey, and F. Sheng. 1990. Chapt. 18: Salinity
management. In Management of Farm Irrigation Systems, 667-715. G. J. Hoffman,
T. A. Howell, and K. H. Solomon, eds. St. Joseph, Mich.: ASAE.
John, C. D., V. Limpinuntana, and H. Greenway. 1976. Interaction of salinity and
anaerobiosis in barley and rice. J. Exp. Bot. 28: 133-141.
Keren, R., and S. Miyamoto. 1990. Reclamation of saline, sodic, and boron-affected
soils. In Agricultural Salinity Assessment and Management, 410-431. K. K. Tanji,
ed. ASCE Manuals and Reports on Engineering Practices No. 71. New York, N.Y.:
Keren, R., A. Meiri, and Y. Kalo. 1983. Plant spacing effect on yield of cotton
irrigated with saline water. Plant Soil 74: 461-465.
Kriedemann, P. E., and R. Sands. 1984. Salt resistance and adaptation to root zone
hypoxia in sunflower. Aust. J. Plant Physiol. 11: 287-301.
Kruse, E. G., D. A. Young, and D. F. Champion. 1985. Effects of saline water tables
on corn irrigation. Proc. Specialty Conference, 444-453. New York, N.Y.:
Kruse, E. G., R. E. Yoder, D. L. Cuevas, and D. F. Champion. 1986. Alfalfa water use
from high, saline water tables. ASAE Paper No. 86-2597. St. Joseph, Mich.:
ASAE.
Lee, E. W. 1990. Drainage water treatment and disposal solutions. In Agricultural
Salinity Assessment and Management, 450-468. K. K. Tanji, ed. New York, N.Y.:
Letey, J., A. Dinar, and K. C. Knapp. 1985. Crop water production function model for
saline irrigation water. Soil Sci. Soc. America J. 49: 1005-1009.
Letey, J., C. Roberts, M. Penberth, and C. Vasek. 1987. An agricultural dilemma:
Drainage water and toxic disposal in the San Joaquin Valley. Special Publication
3319. Div. of Agric. and Natural Resource, Univ. of Calif.
Lunin, J., and M. H. Gallatin. 1965. Zonal salinization of the root system in relation to
plant growth. Soil Sci. Soc. America Proc. 29: 608-612.
Lunt, O. R., J. J. Oertli, and K. C. Kohl. 1960. Influence of environmental conditions on
the salinity tolerance of several plant species. 7th Int. Congr. Soil Sci. 1: 560-570.
Luthin, J. N., P. Fernandez, J. Woerner, and F. Robinson. 1969. Displacement front
under ponded leaching. J. Irrig. Drain. Div., ASCE 95(IR1): 117-125.
Mass, E. V. 1990. Crop salt tolerance. In Agricultural Salinity Assessment and
Management, 262-304. K.K. Tanji, ed. ASCE Manuals and Reports on Engineering
Practices, No. 71. New York, N.Y.: American Soc. Civil Engineers.
Maas, E. V., and G. J. Hoffman. 1977. Crop salt tolerance: Current assessment. J.
Irrig. Drain. Div., ASCE 103: 115-134.
Maas, E. V., and R. H. Nieman. 1978. Physiology of plant tolerance to salinity. In Crop
Tolerance to Suboptimal Land Conditions, 277-299. G. A. Jung, ed. Spec. Publ. 32.
Maas, E. V., and J. A. Poss. 1989a. Salt sensitivity of wheat at various growth stages.
Irrig. Sci.10: 313-320.
Maas, E. V., and J. A. Poss. 1989b. Sensitivity of cowpea to salt stress at three growth
stages. Irrig. Sci. 10: 29-40.
Maas, E. V., S. R. Grattan, and G. Ogata. 1982. Foliar salt accumulation and injury in
crops sprinkled with saline water. Irrig. Sci. 3: 157-168.
Maas, E. V., J. A. Poss, and G. J. Hoffman. 1986. Salinity sensitivity of sorghum at
three growth stages. Irrig. Sci. 7: 1-11.
Maas, E. V., G. J. Hoffman, G. D. Chaba, J. A. Poss, and M. C. Shannon. 1983. Salt
sensitivity of corn at various growth stages. Irrig. Sci. 4: 45-57.
Magistad, O. C., A. D. Ayers, C. H. Wadleigh, and H. G. Gauch. 1943. Effect of salt
concentration, kind of salt and climate on plant growth in sand cultures. Plant
Physiol. 18: 151-166.
Massoud, F. I. 1981. Salt affected soils at a global scale and concepts for control.
Technical Paper. Rome, Italy: Land and Water Development Div., FAO.
Meiri, A., G. J. Hoffman, M. C. Shannon, and J. A. Poss. 1982. Salt tolerance of two
muskmelon cultivars under two radiation levels. J. American Soc. Hort. Sci. 170:
1168-1172.
Meiri, A., J. Shalhevet, L. H. Stoley, G. Sinai, and R. Steinhardt. 1986. Managing
multi-source irrigation water of different salinities for optimum crop production.
BARD technical report #1-402-81. Bet Sagan, Israel: Volcani Center.
Micklin, P. P. 1991. The Water Management Crisis in Soviet Central Asia. Pittsburgh,
Pa.: Univ. Pittsburgh, Center for Russian and East European Studies.
Miles, D. L. 1977. Salinity in the Arkansas Valley of Colorado. Interagency
Agreement Report EPA-AIC-D4-0544. Denver, Colo.: EPA.
Moore, J., and J. V. Hefner. 1977. Irrigation with saline water in the Pecos Valley of
West Texas. In Proc. Intl. Salinity Conf. on Managing Saline Water for Irrigation,
339-344. Lubbock, Tex.: Texas Tech. Univ.
Namken, L. N., C. L. Wiegand, and R. B. Brown. 1969. Water use by cotton from low
and moderately saline static water tables. Agron. J. 61: 305-310.
Nolan, S. L., T. H. Ashe, R. S. Lindstrom, and D. C. Martens. 1982. Effect of sodium
levels on four foliage plants grown at two light levels. HortSci. 17: 815-817.
Oster, J. D., and F. W. Schroer. 1979. Infiltration as influenced by irrigation water
quality. Soil Sci. Soc. America J. 43: 444-447.
Oster, J. D., and L. S. Willardson. 1971. Reliability of salinity sensors for the
management of soil salinity. Agron. J. 63: 695-698.
Oster, J. D., G. J. Hoffman, and F. E. Robinson. 1984. Management alternatives:
Crop, water, and soil. California Agric. 38: 29-32.
Peck, A. J., C. D. Johnston, and D. R. Williamson. 1981. Analyses of solute
distributions in deeply weathered soils. Agric. Water Mgmt. 4: 83-102.
Pratt, P. F. 1973. Quality criteria for trace elements in irrigation waters. Calif. Agric.
Exp. Station Bulletin.
Pratt, P. F., and D. L. Suarez. 1990. Irrigation water quality assessments. In
Agricultural Salinity Assessment and Management, 220-236. K. K. Tanji, ed. New
York, N.Y.: American Soc. Civil Engineers.
Raats, P. A. C. 1974. Steady flows of water and salt in uniform soil profiles with plant
roots. Soil Sci. Soc. America Proc. 38: 717-722.
Rains, D. W., S. Goyal, R. Weyranch, and A. Lauchli. 1987. Saline drainage water
reuse in a cotton rotation system. Calif. Agric. 41(9 and 10): 24-26.
Reeve, R. C., and E. J. Doering. 1966. The high salt water dilution method for
reclaiming sodic soils. Soil Sci. Soc. America Proc. 30: 498-504.
Rhoades, J. D. 1977. Potential for using saline agricultural drainage waters for
irrigation. In Proc. Water Mgmt. for Irrig. & Drain., 85-116. Reston, Va.:
Rhoades, J. D. 1979. Inexpensive four-electrode probe for monitoring soil salinity.
Soil Sci. Soc. America J. 43: 817-818.
Rhoades, J. D. 1987. Use of saline water for irrigation. Water Quality Bull. 12: 14-20.
Natl. Water Res. Inst., Ontario, Canada special bulletin, Water Quality, Burlington.
Rhoades, J. D., and D. L Corwin. 1981. Determining soil electrical conductivity depth
relations using an inductive electromagnetic soil conductivity meter. Soil Sci. Soc.
America J. 45: 225-260.
Rhoades, J. D., and R. D. Ingvalson. 1971. Determining salinity in field soils with soil
resistance measurements. Soil Sci. Soc. America Proc. 35: 54-60.
Rhoades, J. D., and J. van Schilfgaarde. 1976. An electrical conductivity probe for
determining soil salinity. Soil Sci. Soc. America J. 40: 647-651.
Rhoades, J. D., P. J. Shouse, W. J. Aloes, N. A. Monteghi, and S. M. Lesch. 1990.
Determining soil salinity from soil electrical conductivity using different models
and estimates. Soil Sci. Soc. America J. 54: 46-54.
Rhoades, J. D., F. T. Bingham, J. Letey, A. R. Dedrick, M. Bean, G. J. Hoffman, W. J.
Alves, R. V. Swain, P. G. Pacheco, and R. D. LeMert. 1988. Reuse of drainage
water for irrigation: Results of Imperial Valley study. I: Hypothesis, experimental
procedures and cropping results. Hilgardia 56: 1-16.
Rolston, D. E., D. W. Rains, J. W. Biggar, and A. Lauchli. 1988. Reuse of saline drain
water for irrigation. Presented at UCD/INIFAP Conference. Guadalajara, Mexico.
Schwartz, M., and J. Gale. 1984. Growth response to salinity at high levels of carbon
dioxide. J. Exp. Bot. 35: 193-196.
Shalhevet, J. 1984. Management of irrigation with brackish water. In Soil Salinity
under Irrigation, Processes and Management, Ecological Studies, 298-318. I.
Shainberg, and J. Shalhevet, eds. New York, N.Y.: Springer-Verlag.
Shalhevet, J., and L. Bernstein. 1968. Effects of vertically heterogeneous soil salinity
on plant growth and water uptake. Soil Sci. 106: 85-93.
Shalhevet, J., B. Heuer, and A. Meiri. 1983. Irrigation interval as a factor in salt
tolerance of eggplant. Irrig. Sci. 4: 83-93.
Shalhevet, J., P. Reiniger, and D. Shimshi. 1969. Peanut response to uniform and non-
uniform soil salinity. Agron. J. 1: 384-387.
Shalhevet, J., A. Vinten, and A. Meiri. 1986. Irrigation interval as a factor in sweet
corn response to salinity. Agron. J. 78: 539-545.
Shannon, M. C. 1980. Differences in salt tolerance within ‘Empire’ lettuce. J.
Shannon, M. C., and L. E. Francois. 1978. Salt tolerance of three muskmelon
cultivars. J. American Soc. Hort. Sci. 103: 127-130.
Shannon, M. C., and C. L. Noble. 1990. Genetic approaches for developing economic
salt tolerant crops. In Agricultural Salinity Assessment and Management, 161-185.

K. K. Tanji, ed. ASCE Manuals and Reports on Engineering Practices, No. 71.
New York, N.Y.: American Soc. Civil Engineers.
Shennan, C., S. Grattan, D. May, R. Burau, and B. Hanson. 1987. Potential for the
long-term cyclic use of saline drainage water for the production of vegetable crops,
142-146. Technical Progress Report. U.C. Salinity/Drainage Task Force, Div. of
Ag. and Natural Resources, Univ. of California.
Stewart, J. I., R. M. Hagan, and W. O. Pruitt. 1976. Salinity effects on corn yield,
evapotranspiration, leaching fraction and irrigation efficiency in managing saline
water for irrigation. In Proc. Intl. Salinity Conf., 316-332.
Suarez, D. L. 1981. Relationship between Ph and SAR and an alternative method of
estimating SAR of soil or drainage water. Soil Sci. Soc. America J. 45: 469-475.
Sylvester, M. A., J. P. Deason, H. R. Feltz, and R. A. Engberg. 1988. Preliminary
results of the Department of Interior’s irrigation drainage studies. In Planning Now
for Irrigation and Drainage in the 21st Century, Proc. of a conference sponsored
by the Irrigation and Drainage Division of the American Society of Civil
Engineers, 665-677. D. L. Hay, ed. New York, N.Y.: American Soc. Civil Engineers.
Szabolcs, I. 1989. Salt Affected Soils. Boca Raton, Fla.: CRC Press, Inc.
Taylor, A. C. 1965. Water history and the Indus Plain. Natural History 24(5): 40-49.
Torres, C. B., and F. T. Bingham. 1973. Salt tolerance of Mexican wheat. I: Effects of
NO3 and NaC1 on mineral nutrition, growth and grain production of four wheat
varieties. Soil Sci. Soc. America Proc. 37: 711-715.
Trofimenko, S. 1985. The state and density of the Aral Sea. Ambio. 14(3): 181-182.
USDA. 1988. Water quality education and technical assistance plan. USDA–Soil Con-
servation Service and USDA–Extension Service Report. Washington, D.C.: USDA.
Vinten, A., J. Shalhevet, A. Meiri, and J. Peretz. 1986. Water and leaching
requirements of industrial tomatoes with brackish water. Irrig. Sci. 7: 13-25.
Wadleigh, C. H., H. G. Gauch, and O. C. Magistad. 1946. Growth and rubber
accumulation in guayule. USDA Tech. Bulletin No. 925. Washington, D.C.: USDA.
Wadleigh, C. H., H. G. Gauch, and D. G. Strong. 1947. Root penetration and moisture
extraction in saline soil by crop plants. Soil Sci. 63: 341-349.
Wagenet, R. J., W. P. Campbell, A. M. Bamatraff, and D. L. Turner. 1980. Salinity, irri-
gation frequency, and fertilization effect on barley growth. Agron. J. 72: 969-974.
Wallender, W. W., D. W. Grimes, D. W. Henderson, and L. K. Stromberg. 1979.
Estimating the contribution of a perched water table to the seasonal
evapotranspiration of cotton. Agron. J. 71: 1056-1060.
West, D. W., and J. A. Taylor. 1984. Response of six grape cultivars to the combined
effects of salinity and root zone waterlogging. J. American Soc. Hort. Sci. 109:
844-851.
Wichelns, D., D. Nelson, and T. Weaver. 1988. Farm-level analyses of irrigated crop
production in areas with salinity and drainage problems. San Joaquin Valley
Drainage Program. Sacramento, Calif.: U.S. Bureau of Reclamation.
Willey, G. R. 1953. Prehistoric settlement patterns in the Viru Valley, Peru.
Smithsonian Inst., Bureau of American Ethnology, Bull. 155. Wash. D.C.:
Smithsonian Inst.
Wu, L., Z. Huang, and R. G. Burau. 1988. Selenium accumulation and selenium-salt
co-tolerance in five grass species. Crop Sci. 28: 517-522.
CHAPTER 8
WATER REQUIREMENTS
Richard G. Allen (University of Idaho,
Kimberly, Idaho)
James L. Wright (USDA-ARS,
Kimberly, Idaho)
William O. Pruitt (University of California,
Davis, California)
Luis S. Pereira (University of Lisbon,
Lisbon, Portugal)
Marvin E. Jensen (USDA-ARS,
Abstract. Evapotranspiration (ET) calculation guidelines are based on the crop co-
efficient-reference evapotranspiration method (Kc ETref). Equations for the ASCE-
EWRI standardized Penman-Monteith method are provided for grass and alfalfa ref-
erences, where the grass reference standardization follows the FAO Penman-Monteith
procedure. Linearized FAO-style crop coefficients from FAO-56 and curvilinear coef-
ficients from Wright are presented as both mean and as dual (basal) crop coefficients.
ET coefficients for landscape utilize a decoupled procedure similar to that summa-
rized by the Irrigation Association Water Management Committee. Guidelines for
calculating irrigation water requirements and peak system design rates are described.
Keywords. Evapotranspiration, Evaporation, Penman, Penman-Monteith, Irriga-
tion requirement, Effective precipitation, Crop coefficient, Landscape coefficient.
8.1 INTRODUCTION
Evapotranspiration (ET) from vegetation is the primary water requirement for agri-
cultural crops. The quantification of ET is necessary for design and sizing of irrigation
system components, for operating irrigation and water resources systems, for conduct-
ing water balances, and for conducting hydrologic analyses. Many types of systems
are available for measuring ET, although direct measurement of ET is generally diffi-
cult or expensive. The primary focus of this chapter is estimation of ET using histori-
cal or real-time weather data, because design and operation of irrigation systems gen-
erally requires long and often continuous records of irrigation water requirements.
Most standard, operational procedures for determining ET utilize the crop coeffi-
cient (Kc)-reference evapotranspiration (ETref) approach due to its simplicity, repro-
ducibility, relatively good accuracy, and transportability among locations and cli-
mates. Two families of Kc curves are presented: the linear procedure of the FAO-56
Irrigation and Drainage Paper (Allen et al., 1998) and the curvilinear procedure of
Wright (1982). In addition, within each family, two types of Kc are defined: mean or
time-averaged Kc used for irrigation systems design and general planning, and dual Kc
where basal transpiration and evaporation from soil are separated to increase accuracy.
Two reference ET types are defined and described: clipped cool-season grass and full-
cover alfalfa, since these are both in common usage within the U.S.
Numerous ETref equations have been introduced during the past 50 years. Many of
these have been described and evaluated in prior literature (Jensen 1974; Burman et al.
1980; Hatfield and Fuchs, 1990; Jensen et al. 1990). The three reference ET equations
described in this chapter are the Penman-Monteith (PM) equation, the classical Pen-
man equation and the 1985 Hargreaves equation. These equations are considered to be
the most commonly used methods today and are appropriate and applicable to irriga-
tion systems design and operation under a wide range of application situations and
climates. The PM equation has been standardized to both clipped grass and alfalfa
references by ASCE-EWRI (2005), and is considered to be the best method when a
full complement of weather data is available (solar radiation, air temperature, humid-
ity, and wind speed data). The 1985 Hargreaves equation and the PM equation with
estimated weather data are recommended when only air temperature data are avail-
able. The FAO-24 pan evaporation method is also useful for situations where good-
quality pan evaporation data are measured in conjunction with some weather data.
That method is described in Doorenbos and Pruitt (1977), Jensen et al. (1990), and
Allen et al. (1998).
Other methods for estimating ET include crop simulation computer models that
“grow” and simulate the evaporation and transpiration components of ET separately.
These may also simulate the photosynthesis and respiration of plants (Ritchie and Ot-
ter, 1985, Jones et al., 1987). Complex ET models include “multi-layer” resistance
equations that consider the effects of stomatal, leaf boundary, and aerodynamic char-
acteristics for various portions of the plant canopy (Shuttleworth and Wallace, 1985;
Dolman, 1993; Huntingford et al., 1995). However, data and time requirements to
prepare model parameters for new or localized applications often place these models
outside the realm of irrigation systems design and operation. The Kc-ETref method is a
tried and true method that empirically incorporates many of the physiological and
aerodynamic variables governing crop evapotranspiration. The Kc-ETref method, when
applied carefully, can produce estimates of ET that are sufficiently accurate for irriga-
tion systems design and operation (Allen et al., 2005c).
8.2 DEFINITIONS
Several important quantities are defined in this section. Most of these definitions
are common to agricultural literature.
8.2.1 Evapotranspiration
The term evapotranspiration, abbreviated ET, is defined as the combined process
by which water is converted from liquid or solid forms via evaporation from soil and
wet plant surfaces and via evaporation of water from within plant tissue. The latter
process is known as transpiration. ET can be expressed as the energy consumed as
latent heat energy per unit area (and denoted as LE) or as the equivalent depth of
evaporated water. Units for ET are typically mm t-1 where t denotes a time unit (hour,
day, month, growing season, or year) and units for LE are typically W m-2 or MJ m-2 t-1.
The term reference evapotranspiration has been used as a standardized and repro-
ducible index approximating the climatic demand for water vapor and is generally
210 Chapter 8 Water Requirements
abbreviated ETref. Reference ET is the ET rate from an extensive surface of reference

vegetation having a standardized uniform height and that is actively growing, com-
pletely shading the ground, has a dry but healthy and dense leaf surface, and is not
short of water. This definition is applied to the standardized reference crops of grass
(ETo) and alfalfa (ETr).
The advantage of using the reference concept is that it enables the measurement
and validation of estimated reference ET using living, standardized crops. In addition,
because stomatal control of the alfalfa reference crop approximates that of most agri-
cultural crops at full cover, ET from the reference crop is more like crop ET than is
potential ET (Pereira et al., 1999).
Crop evapotranspiration, abbreviated ETc, is defined as the rate of ET from an ex-
tensive surface of a specific crop. The ETc rate is influenced by crop growth stages,
amount and frequency of wetting of the soil surface, environmental conditions, and
crop management. Crop ET is usually less than about 1.3 ETo or 1.0 ETr when crop
foliage does not completely shade the ground or when the crop has begun to mature
and senesce.1 Crop ET may exceed ETo when the crop has substantially more leaf
area, is taller, or has less stomatal control as compared to the clipped grass reference,
or has a wet leaf or soil surface. Crop ET is normally expressed in units of mm h-1,
mm d-1, mm month-1, or mm season-1, and is synonymous with the term consumptive
use. ETc representing ET under any condition, ideal or nonideal, is termed “actual
ETc” and is denoted as ETc act.
The “extensive surface” in the definition of ETc and calculation methodologies im-
plies that the crop covers a large enough area that the energy exchange at the crop sur-
face, and the wind speed, temperature and humidity profiles above the crop, are in
equilibrium. Only when this equilibrium exists do energy-based flux-profile equations
such as the Penman and Penman-Monteith equations attain the highest accuracy. The
extensive surface condition applies to field sizes having dimensions greater than about
200 m (equivalent to 4 ha or 10 acres).
Landscape evapotranspiration, abbreviated ETL, is ET from residential and urban
landscapes. Procedures for estimating ETL are similar to those for ETc, with two dis-
tinctions: (1) landscape systems are nearly always comprised of a mixture of multiple
types and species of vegetation, thereby complicating the estimation of the landscape
coefficient and ET; and (2) typically, the objective of landscape irrigation is to pro-
mote appearance rather than biomass production as is the case in agriculture. There-
fore, target ET for landscape may include an intentional “stress” factor in the baseline
value for ETc act. This adjustment can result in significant water conservation. A multi-
component approach for estimating ETL is described that covers a wide range of land-
scape vegetation and environmental conditions. This approach is similar to that
adopted by the Irrigation Association (2003).
8.2.2 Effective Precipitation
Effective precipitation, abbreviated Pe, was defined by Dastane (1974) in the con-
text of irrigation water management as precipitation that “is useful or usable in any
phase of crop production.” This definition implies that precipitation must infiltrate and
remain resident in the effective root zone of a cropped field long enough to be ex-
1
Senescence describes the natural aging process of leaves whereby leaves begin to yellow and die, and
stomatal function and exchange of carbon dioxide and water vapor reduce. Senescence may be accelerated
by environmental stresses such as disease and water shortage.
tracted by crop roots during the ET process or where evaporation from wet soil re-
duces some transpiration demand. Effective precipitation is discussed further in Sec-
tions 8.10 and 8.12, and in the context here includes evaporation from soil incidental
to production agriculture and irrigation.
8.2.3 Irrigation Water Requirements
The designer or operator of an irrigation system most frequently estimates irriga-
tion water requirements (IR) for short periods and on a seasonal basis. Short-term es-
timates are needed for systems sizing and for day to day operations. Seasonal esti-
mates are needed for allocation of water supplies and in water rights administration.
The units for IR are often expressed as volume per unit area per unit time (for exam-
ple, m3 ha-1 d-1) or as depth per unit time (for example, mm d-1). The “net” irrigation
water requirement, abbreviated IRn, is defined as the depth of water needed to fulfill
the ET requirement in excess of any effective precipitation for a disease-free crop
growing in large fields under non-restricting soil and soil water conditions and under
adequate fertility. In addition, IRn considers contributions of shallow ground water
(GW) and change in stored soil water during the period of interest. In equation form:
IRn = ETc − Pe − GW − Δ θ z s (8.1)
where Pe = effective precipitation during the period of calculation
Δθ = change in soil water content in the root zone during the period of
calculation
zs = depth of soil experiencing the change in water content.
Units for all terms are the same. In essence, IRn is the beneficially consumed portion of
an irrigation application.
The “gross” irrigation water requirement, abbreviated IR, includes the water re-
quired for the IRn in addition to water for leaching salts and to satisfy delivery and
field system losses. Delivery system losses include spillage and seepage. Field system
losses include surface runoff and deep percolation. In equation form:
IRn ETc − Pe − GW − Δ θ z s
IR = = (8.2)
(1 − LR ) CF (1 − LR) CF
where LR is the leaching requirement (a fraction) and CF is the consumed fraction of
applied water, generally equivalent to the so-called “irrigation efficiency.” Usually, LR
is considered only when irrigation uniformity is high. Under low to moderate uniform-
ity, deep percolation incidental to irrigation is often sufficient to fulfill LR. Calculation
of the leaching requirement is described in Chapter 7.
8.3 DIRECT MEASUREMENTS
System designers and operators obtain ET data from either direct field measure-
ments or from estimates that are based on weather and crop data. Use of direct, real-
time field measurements for ET in design and operation is rare due to the expense and
operation and maintenance requirements of equipment. Direct measurements are pri-
marily used to provide data for calibrating climatic and weather-based ET methods
and for monitoring soil water conditions.
Although direct measurements are essential for ET equation development or validation,
the user must exercise caution when using these systems, data collected by these systems,
and also coefficients calibrated from such data. Errors involved can be subtle, yet large.
For example, eddy covariance data are prone to under-measurement by up to 30% (Twine
et al., 2000; Wilson et al., 2002). Bowen ratio (BR)-based estimates are prone to error and
bias in net radiation and soil heat flux measurements (Payero et al., 2003). Both LE
and BR estimates are sensitive to instrument siting relative to the roughness sublayer
and evaporative footprint. Users should familiarize themselves with the problems as-
sociated with design, installation, maintenance, and operation of direct ET measuring
devices that may influence the values and integrity of direct ET measurement data.
One caution on measurement of ET is that measurements must represent conditions
for which the data are applied. For irrigation systems design, measurements should be
from or representative of ET from large expanses of vegetation where the ET process
is essentially one-dimensional (upward). ET measurements should not be made from
small groups of plants that are spatially isolated. Spatial isolation can increase transfer
of heat and radiation energy from outside the measured group of plants to the plants,
thereby increasing measured ET and making it unrepresentative of ET from large
fields (Pruitt and Lourence, 1985; Meyer and Mateos, 1990; Pruitt, 1991; Allen et al.,
1991b; Grebet and Cuenca, 1991).
The reader is referred to other literature sources for background and description of
various measurement methods, including lysimeters: WMO (1966), Tanner (1967),
Aboukhaled et al. (1982), Pruitt and Lourence (1985), Allen et al. (1991a), and Howell
et al. (1991); eddy covariance: Swinback (1951), Businger et al. (1967), Brutsaert
(1982), Campbell and Tanner (1985), Tanner (1988), Twine et al. (2000), Wilson et al.
(2002), and Munger and Loescher (2004); and Bowen ratio: Sellers (1965), Tanner
(1967), Blad and Rosenberg (1974), Dyer (1974), Sinclair et al. (1975), Brutsaert
(1982), Pruitt et al. (1987), and Payero et al. (2003).
8.4 ESTIMATION OF REFERENCE ET
Evaporation of water requires relatively large amounts of energy obtained either
from transfer of sensible heat from the air stream or from radiant energy. Therefore,
the ET process is largely governed by energy exchange at the vegetation surface and is
limited by the amount of energy available. Because of this limitation, it is possible to
estimate the rate of ET based on a net balance of energy fluxes. This is the basis for
the Penman and Penman-Monteith equations.
8.4.1 Reference Evapotranspiration
Two types of reference crops have been widely applied. These are clipped, cool-
season grass, such as fescue and perennial ryegrass, and full-cover alfalfa. Reference
ET for clipped grass is commonly denoted as ETo and reference ET for alfalfa is com-
monly denoted as ETr.
8.4.2 Standardized Definitions and the Penman-Monteith Method
Grass reference ETo was defined in FAO-24 (Doorenbos and Pruitt, 1977) as “the
rate of evapotranspiration from an extensive surface of 8 to 15 cm tall, green grass
cover of uniform height, actively growing, completely shading the ground and not
short of water.” It is generally accepted that the grass reference crop is a “cool-
season,” C-3 type of grass with roughness, density, leaf area, and bulk surface resis-
tance characteristics similar to perennial ryegrass (Lolium perenne L.) or Alta fescue
(Festuca arundinacea Schreb. ‘Alta’).
Alfalfa reference ETr was defined by Wright and Jensen (1972) as: “... ET from
well watered, actively growing alfalfa with 8 in. (20 cm) or more of growth ...” and by
Wright (1982) as “... when the alfalfa crop was well watered, actively growing, and at
least 30 cm tall; so that measured ET was essentially at the maximum expected level
for the existing climatic conditions.” The height of alfalfa (Medicago sativa L., v.
Ranger) in the data set used to develop the 1982 Kimberly Penman (Wright, 1982) and
used in developing resistance algorithms for the Penman-Monteith method (Allen et
al., 1989) ranged from about 0.15 to 0.80 m in height and averaged 0.47 m (unpub-
lished data from Wright, 1985).
Generally, ETr is about 1.1 to 1.4 times that of ETo due to the increased roughness
and leaf area of alfalfa. The higher value (1.4) represents the ratio of ETr to ETo under
extremely arid and windy conditions (minimum daytime relative humidity [RH] <
20% and wind speed > 5 m s-1 [11 mph]) and the lower value (1.1) represents the ratio
of ETr to ETo under humid, calm conditions. Wright (referenced in Jensen et al.
(1990), Table 6.9) has measured an average ratio of ETr to ETo at Kimberly, Idaho,
equal to about 1.25. This value represents conditions common to the semiarid moun-
tain states in the U.S.
Because of the challenges in growing and maintaining a living reference crop, the
PM equation has been used by the Food and Agriculture Organization (FAO) (Smith
et al., 1991, 1996; Allen et al., 1998) and ASCE-EWRI (2005) to represent a fixed
definition for ETo. The FAO definition for ETo in terms of the PM equation is “the rate
of evapotranspiration from a hypothetical reference crop with an assumed crop height
of 0.12 m, a fixed surface resistance of 70 s m-1 and an albedo of 0.23, closely resem-
bling the evapotranspiration from an extensive surface of green grass of uniform
height, actively growing, completely shading the ground and with adequate water”
(Allen et al. 1998). ASCE-EWRI (2005) has adopted this same definition for stan-
dardization of ETo, with the provision for lower surface resistance (50 s m-1), when
calculating on hourly or shorter timesteps. This lower resistance for hourly calcula-
tions has subsequently been adopted by FAO (Allen et al., 2006). ASCE-EWRI addi-
tionally defined a standardized ETr for the alfalfa reference, also based on the PM,
where the standardized height is 0.5 m.
The ASCE-EWRI (2005) standardized PM method has the form:
Cn
0.408 Δ ( Rn − G ) + γ u2 (es − ea )
T + 273
ETref = (8.3)
Δ + γ (1 + Cd u 2 )
where ETref = standardized reference ET for short (ETos) or tall (ETrs) surfaces, mm d-1
for daily timesteps or mm h-1 for hourly timesteps
Rn = calculated net radiation at the crop surface, MJ m-2 d-1for daily timesteps
or MJ m-2 h-1 for hourly timesteps
G = soil heat flux density at the soil surface, MJ m-2 d-1 for daily timesteps or
MJ m-2 h-1 for hourly timesteps
T = mean daily or hourly air temperature at 1.5 to 2.5-m height, °C
u2 = mean daily or hourly wind speed at 2-m height, m s-1
es = saturation vapor pressure at 1.5- to 2.5-m height, kPa, calculated for daily
timesteps as the average of saturation vapor pressure at maximum and
minimum air temperature
ea = mean actual vapor pressure at 1.5 to 2.5-m height, kPa
Δ = slope of the saturation vapor pressure-temperature curve, kPa °C-1
γ = the psychrometric constant, kPa °C-1
Table 8.1. Values for Cn and Cd in Equation 8.3 (from ASCE-EWRI, 2005).
Short Reference, Tall Reference,
ETos ETrs Units for Units for
(clipped grass) (alfalfa) ETos, R n,
Calculation Timestep Cn Cd Cn Cd ETrs G
-1
Daily 900 0.34 1600 0.38 mm d MJ m-2 d-1
-1
Hourly during daytime 37 0.24 66 0.25 mm h MJ m-2 h-1
-1
Hourly during nighttime 37 0.96 66 1.7 mm h MJ m-2 h-1
Cn = a numerator constant that changes with reference type and calculation time
step, K mm s3 Mg-1 d-1 or K mm s3 Mg-1 h-1
Cd = a denominator constant that changes with reference type and calculation
timestep, s m-1.
Table 8.1 provides values for Cn and Cd. The values for Cn consider the timestep
and aerodynamic roughness of the surface (i.e., reference type). The constant in the
denominator, Cd, considers the timestep, bulk surface resistance, and aerodynamic
roughness of the surface. Cn and Cd were derived by simplifying several terms within
the ASCE PM equation of ASCE Manual 70 (Allen et al., 1989; Jensen et al., 1990)
and rounding the result. Units for the 0.408 coefficient are m2 mm MJ-1. Daytime is
defined as occurring when Rn during an hourly period is positive. The ASCE-EWRI
definition uses smaller values for surface resistance for hourly or shorter calculation
timesteps (during daytime) than for daily calculation timesteps. The FAO-PM (Allen
et al., 1998) is equivalent to Equation 8.3, where Cn = 900 and Cd = 0.34. ASCE-
EWRI (2005) and Allen et al. (2006) recommended applying the FAO-PM equation
for ETo using the Cn and Cd coefficients for hourly and shorter calculation timesteps in
Table 1. Daily ETref are obtained by summing ETref values from hourly or shorter peri-
ods or by applying the ETref equations on a 24 h-timestep. Generally, summing hourly
calculations is considered to be more accurate. ASCE-EWRI recommended applica-
tion of ETos and ETrs for calculating reference evapotranspiration and development of
new crop coefficients, and for facilitating transfer of crop coefficients.
8.4.3 The Classical Penman Equation

Besides the standardized Penman-Monteith equation that is now recommended by
FAO (Allen et al., 1998) and by ASCE-EWRI (2005), some forms of the original
Penman equation (Penman, 1948) are still in use that perform similarly to the ASCE
PM. The classical form of the Penman equation is:
Δ
ET ref = 0.408 (Rn − G ) + K w γ (a w + bw u 2 ) (es − ea ) (8.4)
Δ +γ Δ +γ
where terms and definitions are the same as those used for the PM equation in Equa-
tion 8.3. Parameter Kw is a units parameter, with Kw = 2.62 for ETref in mm d-1 and Kw
= 0.109 for ETref in mm hour-1. The aw and bw terms are empirical wind coefficients
that have often been locally or regionally calibrated. The values for aw and bw were 1.0
and 0.537, respectively, in the original Penman equation (Penman, 1948, 1963) for
grass ETo, for wind speed in m s-1, (es – ea) in kPa and ETo in mm d-1. Generally, local
calibration of the Penman method is neither required nor recommended and was often
an artifact of earlier studies required to overcome biases or faulty measurements in
weather and ET data. Currently the commonly used Penman values for aw and bw are
those for the hourly CIMIS Penman method for ETo and daily 1982 Kimberly Penman
method for ETr.
8.4.3.1 CIMIS Penman equation for grass reference. Pruitt (Pruitt and Dooren-
bos 1977a; Pruitt et al., 1984; Snyder and Pruitt, 1992) developed aw and bw for esti-
mating grass ETo for hourly periods. These coefficients have been used since the
1980s for standard ETo estimation in the California Irrigation Management Informa-
tion Service (CIMIS). The CIMIS Penman ETo equation uses aw = 0.29 and bw = 0.53
for Rn > 0, and aw = 1.14 and bw = 0.40 for Rn < 0, and is applied hourly, where in
Equation 8.4, ETo = mm h-1, Rn = MJ m-2 h-1, and Kw = 0.109. Standard CIMIS calcula-
tions assume G = 0. When applied hourly, ETo from the CIMIS Penman equation is
summed over 24-h periods to obtain daily totals, similar to what is done using the PM
method.
8.4.3.2 1982 Kimberly-Penman alfalfa reference. The 1982 Kimberly-Penman
equation was developed from intensive studies of ET using precision weighing lysime-
ters at Kimberly, Idaho (Wright and Jensen 1972; Wright, 1981, 1982, 1988) and is
intended for application with 24-hour timesteps (Kw = 2.62 in Equation 8.4). The Kim-
berly-Penman wind function is calibrated to estimate ET from full-cover alfalfa, where
the aw and bw wind function coefficients vary with time of year (Jensen et al. 1990):
⎧⎪ ⎡ J − 173 2⎤ ⎫⎪
⎛ ⎞
a w = 0.4 + 1.4 exp ⎨− ⎢ ⎜ ⎟ ⎥ ⎬ (8.5)
⎪⎩ ⎢⎣ ⎝ 58 ⎠ ⎥⎦ ⎪⎭
⎧ ⎡⎛ J − 243 ⎞ 2⎤ ⎫⎪
⎪
b w = 0.605 + 0.345 exp⎨− ⎢⎜ ⎟ ⎥⎬ (8.6)
⎪⎩ ⎢⎣⎝ 80 ⎠ ⎥⎦ ⎪
⎭
where J is the day of the year. For latitudes south of the equator, Equation 8.5 and 8.6
are applied using J' in place of J, where J' = (J – 182) for J ≥ 182 and J' = (J + 182)
for J < 182. The (es – ea) term in the 1982 Kimberly Penman is computed the same as
for the standardized PM equation (as the average of es computed at daily maximum
and minimum temperatures; Equation 8.7).
8.4.4 Computation of Parameters for the
Penman-Monteith and Penman Equations
It is recommended that standardized procedures and equations be used to calculate
parameters in ETref equations. This insures agreement among independent calculations
and simplifies calculation verification. Equations presented in this section follow pro-
cedures standardized by FAO-56 and ASCE-EWRI (2005).
8.4.4.1 Saturation vapor pressure of the air. For 24-hour or longer calculation
timesteps, es, the saturation vapor pressure of the air, is computed as:
eo (T max) + eo (T min)
es = (8.7)
2
where Tmax and Tmin are daily maximum and minimum air temperature, °C, at the
measurement height (1.5 to 2 m), and e° is the saturation vapor pressure function. For
hourly applications, es is calculated as e°(T) where T is average hourly air temperature.
The saturation vapor pressure function is:
⎛ 17.27 T ⎞
e o (T ) = 0.6108 exp ⎜ ⎟ (8.8)
⎝ T+ 237.3 ⎠
where e°(T) is in kPa and T is in °C (Tetens, 1930).
8.4.4.2 Actual vapor pressure of the air. Actual vapor pressure of the air, ea, is
equivalent to saturation vapor pressure at the dew point, Td. For 24-h or longer
timesteps, Td is taken as mean daily or early morning dewpoint temperature, °C. Hu-
midity of the air can be measured using many different methods, including relative
humidity sensors, dewpoint sensors, and wet bulb/dry bulb psychrometers, so that ea
can be calculated using several ways. The recommended procedures, in order of what
are considered to be the most reliable to the least reliable, are (ASCE-EWRI, 2005):
1. For 24-h periods, averaging ea measured or computed hourly over the 24-h pe-
riod.
2. For 24-h periods, calculating ea from dewpoint, Td, that is measured or computed
hourly over the 24-h period:
⎛ 17.27 T d ⎞
e a = eo ( T d ) = 0.6108 exp ⎜⎜ ⎟⎟ (8.9)
⎝ T d + 237.3 ⎠
where ea is in kPa and Td is in °C.
3. Psychrometer measurements using dry and wet bulb thermometers, where
(Bosen, 1958):
(
ea = e o (Twet ) − γ psy Tdry − Twet ) (8.10)
where e°(Twet) is saturation vapor pressure at the wet bulb temperature in kPa
(Equation 8.8), γpsy is the psychrometric constant for the psychrometer, in kPa
°C-1, and Tdry – Twet is the wet bulb depression, where Tdry is dry bulb tempera-
ture and Twet is the wet bulb temperature, both in °C. For 24-h periods, Tdry and
Twet are preferably averaged over the 24-h period. Alternatively, once-per-day
readings of Tdry and Twet can be used, for example, readings taken at 7 or 8 a.m.
Tdry and Twet must be measured simultaneously.
The psychrometric constant for the psychrometer is:
γ psy = a psy P (8.11)
where apsy is a coefficient depending on the type of ventilation of the wet bulb,
in °C-1, and P is the mean atmospheric pressure, in kPa. The coefficient apsy de-
pends primarily on the design of the psychrometer and on the rate of ventilation
around the wet bulb. The following values are often used:
apsy = 0.000662 for ventilated (Asmann type) psychrometers having air move-
ment between 2 and 10 m s-1 for Twet > 0 and 0.000594 for Twet < 0
= 0.000800 for naturally ventilated psychrometers having air movement
of about 1 m s-1
= 0.001200 for non-ventilated psychrometers installed in glass or plastic
greenhouses.
4. For 24-hour or longer timesteps, relative humidity (RH) measurements taken
twice daily (early morning, corresponding to Tmin and early afternoon, corre-
sponding to Tmax) can be combined to yield an approximation for 24-hour aver-
age ea:
RHmax RH
e o (Tmin ) + e o (Tmax ) min
ea = 100 100
2 (8.12)
where RHmax is daily maximum relative humidity (%) (during early morning),
and RHmin is daily minimum relative humidity (%) (during early afternoon,
around 1400 hours).
For hourly calculations, ea is calculated from relative humidity as:
RH o
ea = e (T ) (8.13)
100
where RH is mean relative humidity for the hourly or shorter period, %, and T is
mean air temperature for the hourly or shorter period, °C.
5. From daily RHmax and Tmin:
RHmax
ea = e o (Tmin ) (8.14)
100
6. From daily RHmin and Tmax:
RHmin
ea = e o (Tmax ) (8.15)
100
7. If daily humidity data are missing or are of questionable quality, ea can be ap-
proximated for the reference environment assuming that Td is near Tmin:
Td = Tmin − K o (8.16)
where Ko is approximately 2° to 4°C in dry seasons in semiarid and arid climates
and Ko is approximately 0°C in humid to subhumid climates and the rainy sea-
son in semi-arid climates (ASCE-EWRI, 2005).
8. In the absence of RHmax and RHmin data, but where daily mean RH data are avail-
able, ea can be estimated as:
RH mean o
ea = e (Tm ean ) (8.17)
100
where RHmean is mean daily relative humidity, generally defined as the average
between RHmax and RHmin, and Tmean is mean daily air temperature. For daily or
longer periods, Equation 8.17 is generally less desirable than methods 1-7 for
calculating ea because the e°(T) relationship is nonlinear.
8.4.4.3 Psychrometric constant. The psychrometric constant (γ) in the Penman
and PM equations is calculated as (Brunt, 1952):
γ = 0.000665 P (8.18)
-1
where P has units of kPa and γ has units of kPa °C .
8.4.4.4 Atmospheric pressure. For purposes of ET estimation, mean atmospheric
pressure, P, can be based on elevation using an equation simplified from the universal
gas law relationship (Burman et al., 1987, ASCE-EWRI, 2005):
P = (2.406 - 0.0000534 z ) 5.26 (8.19)
where P has units of kPa and z is the weather site elevation (m) above mean sea level.
8.4.4.5 Slope of the saturation vapor pressure-temperature curve. The slope of

the saturation vapor pressure-temperature curve, Δ, is computed as:
⎛ 17.27 T ⎞
2503 exp ⎜⎜ ⎟⎟
Δ= ⎝ T + 237.3 ⎠ (8.20)
(T + 237.3)2
where Δ has units of kPa °C-1 and T is daily or hourly mean air temperature in °C.
8.4.4.6 Wind speed at 2 meters. Wind speed varies with height above the ground
surface. For the calculation of standardized ETref, the wind speed measurement must
be made equivalent to that 2 m above the surface. Therefore, wind measured at other
heights is adjusted by:
4.87
u2 = u z (8.21)
ln(67.8 z w − 5.42)
where u2 = wind speed at 2 m above ground surface, m s-1
uz = measured wind speed at zw m above ground surface, m s-1
zw = height of wind measurement above the ground surface, m.
Equation 8.21 is used for measurements taken above a short grass (or similar) surface,
based on the logarithmic wind speed profile equation. For wind measurements made
above surfaces other than clipped grass, the user should apply a full logarithmic equa-
tion that considers the influence of vegetation height and roughness on the shape of
the wind profile. These alternative adjustments are described in Allen and Wright
(1997) and ASCE-EWRI (2005). Wind speed data collected at heights above 2 m are
acceptable to use in the standardized equations following adjustment to 2 m, and are
preferred if vegetation adjacent to the weather station commonly exceeds 0.5 m.
Measurement at a greater height, for example 3 m, reduces the influence of the taller
vegetation surrounding the weather measurement site.
8.4.4.7 Net radiation. Net radiation, Rn, in the context of ET, is the net amount of
radiant energy available at a vegetation or soil surface for evaporating water, heating
the air, or heating the surface. Rn includes both short and long wave radiation compo-
nents. Net radiation flux density can be measured directly using hemispherical net
radiometers. However, measurement is difficult because net radiometers are problem-
atic to maintain and calibrate, creating the likelihood of systematic biases. Therefore,
Rn is often estimated from observed short wave (solar) radiation, vapor pressure, and
air temperature. This estimation is routine and generally highly accurate for well de-
fined vegetated surfaces. If Rn is measured, then care and attention must be given to
the calibration of the radiometer, the surface over which it is located, maintenance of
the sensor domes, and level of the instrument. The condition of the vegetation surface
is as important as the sensor. For purposes of calculating reference ET, the measure-
ment surface for Rn is generally assumed to be clipped grass or alfalfa at full cover.
When calculated, Rn is estimated from the short wave and long wave components:
Rn = Rns – Rnl (8.22)
-2 -1 -2 -1
where Rns is net short-wave radiation (in MJ m d or MJ m h ), defined as being
positive downwards and negative upwards, and Rnl is net outgoing long-wave radiation
(in MJ m-2 d-1 or MJ m-2 h-1), defined as being positive upwards and negative down-
wards. Rns and Rnl are generally positive or zero in value.
Net short-wave radiation resulting from the balance between incoming and re-
flected solar radiation is given by:
Rns = Rs − α Rs = (1 − α ) Rs (8.23)
where α is albedo, fixed in the ASCE-EWRI standardization at 0.23 for both daily and
hourly timesteps (dimensionless), and Rs is incoming solar radiation (MJ m-2 d-1 or MJ
m-2 h-1). Users may elect to use a different estimation for albedo; however, it is essen-
tial to ascertain the validity and accuracy of an alternative method using good meas-
urements of incoming and reflected solar radiation.
The standardized procedure of FAO-56 and ASCE-EWRI for estimating Rnl is
based on the Brunt (1932, 1952) approach for estimating net emissivity. Users may
choose to utilize a different approach for calculating Rnl; however, as with albedo it is
essential to ascertain the validity and accuracy of the Rn method using net radiometers
in excellent condition and that are calibrated to some dependable and recognized stan-
dard. Rnl, net long-wave radiation, is the difference between upward long-wave radia-
tion from the standardized surface (Rlu) and downward long-wave radiation from the
sky (Rld), so that Rnl = Rlu – Rld. The ASCE-EWRI procedure for daily Rnl is:
⎡ TK4 max + TK4 min ⎤
(
daily Rnl = σ f cd 0.34 − 0.14 ea ⎢
⎢
) 2
⎥
⎥
(8.24)
⎣ ⎦
and for hourly Rnl is:
( )
hourly Rnl = σ f cd 0.34 − 0.14 ea TK4 hr (8.25)
where Rnl = net outgoing long-wave radiation, MJ m-2 d-1or MJ m-2 h-1
σ = Stefan-Boltzmann constant, 4.901 × 10-9 MJ K-4 m-2 d-1 or 2.042 ×
10-10 MJ K-4 m-2 h-1
fcd = a cloudiness function (dimensionless) and limited to 0.05 < fcd < 1.0 (see
below)
ea = actual vapor pressure , kPa
TK max = maximum absolute temperature during the 24-hour period, K (recall
that K = °C + 273.15)
TK min = minimum absolute temperature during the 24-hour period, K
TK hr = mean absolute temperature during the hourly period, K.
The superscripts “4” in Equations 8.24 and 8.25 indicate the need to raise the air tem-
perature, expressed in Kelvin units, to the power of 4.
For daily and monthly calculation timesteps, fcd is calculated as:
Rs
f cd = 1.35 − 0.35 (8.26)
Rso
Rs is measured or calculated solar radiation and Rso is calculated clear-sky radiation,
both in MJ m-2 d-1. The relative solar radiation, Rs/Rso, represents relative cloudiness
and is limited to 0.3 < Rs /Rso < 1.0 so that fcd has limits of 0.05 < fcd < 1.0.
For hourly periods during daytime when the sun is more than about 15° above the
horizon, fcd is calculated using Equation 8.26 with the same limits applied. For hourly
periods during nighttime, Rso, by definition, equals 0, and Equation 8.26 is undefined.
Therefore, fcd during periods of low sun angle and during nighttime is defined using fcd
2
from prior periods. When sun angle (β) above the horizon at the midpoint of the
hourly or shorter time period is less than 0.3 radians (~17°), then:
fcd = fcd β >0.3 (8.27)
where fcd β >0.3 is the cloudiness function (dimensionless) for the time period prior to
when sun angle β (in the afternoon or evening) falls below 0.3 radians.
If the calculation timestep is shorter than one hour, then fcd from several periods can
be averaged into fcd β >0.3 to obtain a representative average value. In mountain valleys
where the sun may set near or above 0.3 radians (~17°), the user should increase the
sun angle at which fcd β >0.3 is computed and imposed. For example, for a location
where mountain peaks are 20° above the horizon, a period should be selected for com-
puting fcd β >0.3 where the sun angle at the end of the time period is 25° to 30° above the
horizon. The same adjustment is necessary when deciding when to resume computa-
tion of fcd during morning hours when mountains lie to the east.
Only one value for fcd β >0.3 is calculated per day for use during dusk, nighttime, and
dawn periods. That value for fcd β >0.3 is then applied to the time period when β at the
midpoint of the period first falls below 0.3 radians (~17°) and to all subsequent peri-
ods until after sunrise when β again exceeds 0.3 radians.
Equations 8.26 and 8.27 will not apply at latitudes and times of the year when there
are no hourly (or shorter) periods having sun angle of 0.3 radians or greater. These
situations occur at latitudes of 50° for about one month per year (in winter), at lati-
tudes of 60° for about five months per year, and at latitudes of 70° for about seven
months per year (ASCE-EWRI, 2005). Under these conditions, the application can
average fcd β> 0.3 from fewer time periods or, in the absence of any daylight, can assume
a ratio of Rs/Rso ranging from 0.3 for complete cloud cover to 1.0 for no cloud cover.
Under these extreme conditions, the estimation of Rn is only approximate.
The application of Equation 8.27 presumes that cloudiness during periods of low
sun angle and nighttime is similar to that during late afternoon or early evening. This
is generally a reasonable assumption and is commensurate with the relative simplicity
and moderate accuracy of the procedure.
8.4.4.8 Clear-sky solar radiation (Rso). Clear-sky solar radiation (Rso) is used in
the calculation of net radiation (Rn). Rso is defined as the amount of short-wave radia-
tion that would be received at the weather measurement site under conditions of clear-
sky (i.e., cloud-free). Daily Rso is a function of the time of year and latitude and is im-
pacted by station elevation (affecting atmospheric thickness and transmissivity), the
amount of precipitable water in the atmosphere (affecting the absorption of some
short-wave radiation), and the amount of dust or aerosols in the air. Rso, for purposes
of calculating Rn, can be computed as:
Rso = (0.75 + 2 × 10-5 z) Ra (8.28)
where z is station elevation above sea level in m. More complex estimates for Rso,
which include impacts of turbidity and water vapor on radiation absorption, can be
used for assessing integrity of solar radiation data and are described in Appendix D of
ASCE-EWRI (2005).
2
The sun angle β is defined as the angle of a line from the measurement site to the center of the sun’s disk
relative to a line from the measurement site to directly below the sun and tangent to the earth’s surface. This
definition assumes a flat surface.
8.4.4.9 Exoatmospheric radiation. Exoatmospheric radiation, Ra, (also known as

extraterrestrial radiation) is defined as the short-wave solar radiation in the absence of
an atmosphere, and is a well-behaved function of the day of the year, time of day, and
latitude. Ra is needed for calculating Rso, which is in turn used in calculating Rn. For
daily (24-hour) periods, Ra is estimated from the solar constant, the solar declination,
and the day of the year:
24
Ra = Gsc d r [ω s sin(ϕ ) sin(δ ) + cos(ϕ ) cos(δ ) sin(ω s )] (8.29)
π
where Ra = exoatmospheric radiation, MJ m-2 d-1
Gsc = solar constant, 4.92 MJ m-2 h-1
dr = inverse relative distance factor (squared) for the earth-sun, unitless
ωs = sunset hour angle, radians
ϕ = station latitude, radians, positive for the northern hemisphere and negative
for the southern hemisphere
δ = solar declination, radians.
Parameters dr and δ are calculated as:
⎛ 2π ⎞
d r = 1 + 0.033 cos ⎜ J⎟ (8.30)
⎝ 365 ⎠
⎛2π ⎞
δ = 0.409 sin ⎜ J − 1.39 ⎟ (8.31)
⎝ 365 ⎠
where J is the number of the day in the year between 1 (1 January) and 365 or 366 (31
December). The constant 365 in Equations 8.23 and 8.24 is held at 365 even during a
leap year. J can be calculated as:
⎛ M ⎞ ⎛ 3 ⎞ ⎛ M Mod(Y , 4) ⎞
J = DM - 32+ Int ⎜ 275 ⎟ + 2 Int ⎜ ⎟ + Int ⎜ - + 0.975 ⎟ (8.32a)
⎝ 9 ⎠ ⎝ M + 1⎠ ⎝ 100 4 ⎠
where DM is the day of the month (1-31), M is the number of the month (1-12), and Y
is the number of the year (for example 1996 or 96). The “Int” function in Equation
8.32 finds the integer number of the argument in parentheses by rounding downward.
The “Mod(Y,4)” function finds the modulus (remainder) of the quotient Y/4.
For monthly periods, the day of the year at the middle of the month (Jmonth) is ap-
proximately:
J month = Int(30.4 M − 15 ) (8.32b)
The sunset hour angle, ωs, is given by:
ω s = arccos [ − tan (ϕ ) tan (δ )] (8.33)
The “arccos” function is the arc-cosine function and represents the inverse of the co-
sine. This function is not available in all computer languages, so that ωs can alterna-
tively be computed using the arc-tangent (inverse tangent) function:
π ⎡ − tan(ϕ ) tan(δ ) ⎤
ωs = − arctan ⎢ ⎥ (8.34)
2 ⎣ X 0.5 ⎦
where X = 1 − [tan(ϕ )] 2 [tan(δ )] 2 (8.35)
and X = 0.00001 if X ≤ 0
For hourly time periods, the solar time angle at the beginning and end of the period
serve as integration endpoints for calculating Ra:
12
Ra = Gsc d r [(ω2 −ω1 )sin(ϕ)sin(δ )+cos(ϕ)cos(δ ) (sin(ω2 )−sin(ω1) )] (8.36)
π
where Ra = exoatmospheric radiation, MJ m-2 hour-1
Gsc = solar constant, 4.92 MJ m-2 h-1
ω1 = solar time angle at beginning of period, radians
ω2 = solar time angle at end of period, radians.
πt
ω1 and ω2 are given by ω1 = ω − 1 (8.37)
24
π t1
ω2 = ω + (8.38)
24
where ω is solar time angle at the midpoint of the period (radians), and tl is the length
of the calculation period (hour): i.e., 1 for hourly periods or 0.5 for 30-minute periods.
The solar time angle at the midpoint of the hourly or shorter period is:
π
ω = {[t + 0.06667( Lz − Lm ) + Sc ] − 12} (8.39)
12
where t = standard clock time at the midpoint of the period (hour) after correcting time
for any daylight savings shift. For example, for a period between 1400 and
1500, t = 14.5 hours.
Lz = longitude of the center of the local time zone, expressed as positive
degrees west of Greenwich, England. In the U.S., Lz = 75°, 90°, 105°, and
120° for the Eastern, Central, Rocky Mountain, and Pacific time zones,
respectively, and Lz = 0° for Greenwich, 345° for Paris, France, and
255° for Bangkok, Thailand.
Lm = longitude of the solar radiation measurement site, expressed as positive
degrees west of Greenwich, England
Sc = a seasonal correction for solar time, hours.
Because ωs is the sunset hour angle and –ωs is the sunrise hour angle (noon has ω =
0), values of ω < –ωs or ω > ωs from Equation 8.39 indicate that the sun is below the
horizon, so that, by definition, Ra and Rso are zero and their calculation has no mean-
ing. When the values for ω1 and ω2 span the value for –ωs or for ωs, this indicates that
sunrise or sunset occurs within the hourly (or shorter) period. In this case, the integration
limits for applying Equation 8.36 should be correctly set using the following conditionals:
If ω1 < –ωs then ω1 = –ωs
If ω2 < –ωs then ω2 = –ωs (8.40)
If ω1 > ωs then ω1 = ωs
If ω2 > ωs then ω2 = ωs
If ω1 > ω2 then ω1 = ω2
The above conditionals can be applied for all timesteps to insure numerical stability
of the application of Equation 8.36 as well as correctly computing the theoretical
quantity of solar radiation early and late in the day. Where there are hills or mountains,
the hour angle when the sun first appears or disappears may increase for sunrise or
decrease for sunset; however, generally, no corrections are necessary. The seasonal
correction for solar time is:
S c = 0.1645 sin(2 b) − 0.1255 cos(b) − 0.025 sin(b) (8.41)
2π(J −81)
b= (8.42)
364
where J is the number of the day in the year and b has units of radians.
The user should confirm accurate setting of the datalogger clock. If clock times are
in error by more than 5–10 minutes, estimates of exoatmospheric and clear sky radia-
tion may be significantly impacted. This can lead to errors in estimating Rn on an
hourly or shorter basis, especially early and late in the day. A shift in “phase” between
measured Rs and Rso estimated from Ra according to the data logger clock can indicate
error in the reported time. Discussion is given in Appendix D of ASCE-EWRI (2005).
The angle of the sun above the horizon, β, at the midpoint of the hourly or shorter
time period is computed as:
β = arcsin[sin(ϕ )sin(δ )+cos(ϕ )cos(δ )cos(ω) ] (8.43)
where β = angle of the sun above the horizon, radians
ϕ = station latitude, radians
δ = solar declination, radians
ω = solar time angle at the midpoint of the period, radians.
The “arcsin” function is the arc-sine function and represents the inverse of the sine.
This function is not available in all computer languages, but β can alternatively be
computed using the arc tangent (inverse tangent) function:
⎡ Y ⎤
β = arctan ⎢ ⎥ (8.44)
(
⎢ 1−Y
⎣ )
2 0.5 ⎥
⎦
where Y = sin(ϕ )sin(δ ) + cos(ϕ )cos(δ )cos(ω ) (8.45)
and all other parameters are as defined previously.
8.4.4.10 Soil heat flux. The magnitude of soil heat storage or release can be sig-
nificant in the surface energy balance over a period of a few hours, but is usually small
day to day because heat stored early in the day as the soil warms is lost late in the day
or at night when the soil cools. Since the magnitude of 24-hour average soil heat flux
under a crop canopy over 10- to 30-day periods is relatively small, G normally can be
neglected for daily and longer timesteps. The total G over a complete growing period,
however, may be significant, especially for 30 days or longer. As defined here and by
ASCE-EWRI (2005), G is positive when the soil is warming and negative when the
soil is cooling.
For daily periods, the magnitude of G averaged over 24 hours beneath a fully vege-
tated grass or alfalfa reference surface is relatively small in comparison with Rn.
Therefore, it is ignored in the standardized ET calculations so that
Gday = 0 (8.46)
where Gday is the daily soil heat flux density, MJ m-2 d-1.
Over a monthly period, G for the soil profile can be significant, especially during
spring and fall. Assuming a constant soil heat capacity of 2.0 MJ m-3 °C-1 and an ef-
fectively warmed soil depth of 2 m, G for monthly periods in MJ m-2 d-1 is estimated
from the change in mean monthly air temperature as:
Gmonth,i = 0.07 (Tmonth,i +1 − Tmonth,i−1 ) (8.47)
or, if Tmonth,i+1 is unknown,
Gmonth,i = 0.14 (Tmonth,i − Tmonth,i−1 ) (8.48)
where Tmonth,i is mean air temperature of month i, Tmonth,i-1 is mean air temperature of
the previous month, and Tmonth,i+1 is the mean air temperature of the next month (all
°C).
For application to short-periods, e.g., when hourly data for G are required, other
approaches must be used. One method common to research uses heat flux plates in-
stalled near the soil surface, usually at a depth of about 0.1 to 0.15 m. The total heat
flux density is determined by performing a calorimetric balance of the soil layer above
the plate. Because soil heat flux plates are not a common measurement at weather sta-
tions, the reader is referred to other references (Tanner, 1960; Brutsaert, 1982; and
Allen et al., 1996) for application descriptions.
For hourly or shorter time periods, G, in the standardized calculation is expressed
as a function of net radiation for the two reference types. For the standardized short
reference ETos :
Ghr daytime = 0.1 Rn (8.49a)
Ghr nightime = 0.5 Rn (8.49b)
where G and Rn have the same measurement units (MJ m-2 h-1 for hourly or shorter
time periods). For the standardized tall reference ETrs :
Ghr daytime = 0.04 Rn (8.50a)
Ghr nightime = 0.2 Rn (8.50b)
For standardization, nighttime is defined as when measured or calculated hourly net
radiation Rn is < 0 (i.e., negative). The amount of energy consumed by G is subtracted
from Rn when estimating ETos or ETrs. The coefficient 0.1 in Equation 8.49a represents
the condition of only a small amount of dead thatch underneath the leaf canopy of the
short (clipped grass) reference. Large amounts of thatch insulate the soil surface, re-
ducing the daytime coefficient for grass to about 0.05. However, the 0.1 coefficient is
part of the EWRI and FAO standardizations.
8.4.5 1985 Hargreaves Grass Reference Equation
The Hargreaves equation (Hargreaves and Samani, 1982, 1985; Hargreaves et al.,
1985) is suggested as a means for estimating ETo in situations where weather data are
limited and only maximum and minimum air temperature data are available. The form
of the 1985 Hargreaves equation is:
ETo = 0.0023 (Tmax – Tmin)0.5 (Tmean + 17.8) (Ra) (8.51)
where Tmax and Tmin = maximum and minimum daily air temperature, °C,
Tmean = mean daily air temperature, (Tmax + Tmin)/2

Ra = average daily exoatmospheric radiation (Equation 8.29).
ETo in Equation 8.51 has the same units as Ra and can be converted from MJ m-2 d-1 to
mm d-1 by dividing by λ = 2.45 MJ kg-1.
The Hargreaves equation was the highest-ranked temperature-based method in the
ASCE Manual 70 analysis (Jensen et al., 1990). Droogers and Allen (2002) and Har-
greaves and Allen (2003) reported Equation 8.51 to estimate well over a wide range of
latitudes and climates for periods of 5 days or longer without significant error, pro-
vided mean wind speeds averaged between 1 and 3 m s-1. Itenfisu et al. (2003) com-
pared daily ETo by Equation 8.51 and that by the standardized ASCE PM method for
49 sites in 16 states in U.S. and found mean ratios of Hargreaves ETo to ASCE PM
ETos to range from 1.43 to 0.79, with a mean of 1.06 and a standard deviation of 0.13.
The Hargreaves equation tended to estimate higher than the ASCE PM method when
mean daily ETo was low, and vice versa.
One advantage of an equation such as the Equation 8.51 relative to more complex
equations such as the Penman or PM equation, which is often overlooked, is the re-
duced data requirement and therefore reduced chance for data error. This is advanta-
geous in regions where solar radiation, humidity, and wind data are lacking or are of
low or questionable quality (Droogers and Allen, 2002). Generally, air temperature
can be measured with less error, with less sophisticated equipment, and by less trained
individuals than can the other three parameters required by combination equations.
Equation 8.51 can be calibrated against the PM equation (Equation 8.3) when data are
available to produce a regionally calibrated temperature equation. Examples of this
type of calibration in Spain include Martinez-Cob and Tejero-Juste (2004), Vanderlin-
den et al. (2004), and Gavilan et al. (2005). An alternative to using Equation 8.51 when
data are lacking is to employ the PM equation using estimates for missing variables.
8.4.6 Effect of Timestep Size on Calculations
The Penman and Penman-Monteith equations can be applied to hourly and 24-h
timesteps. The 24-h timesteps can use daily, weekly, 10-d, and monthly averages for
weather data. Under many climatic conditions, calculating ETo or ETr using hourly
timesteps and then summing over 24 hours provides estimates that closely equal ETo
or ETr calculated using 24-h average data with 24-h calculation timesteps, especially
when applying the standardized ASCE-EWRI PM method (Itenfisu et al., 2003;
ASCE-EWRI, 2005). Generally, 24-h ETo and ETr have potential for higher accuracy
when computed using hourly or shorter timesteps and then summed to 24-hour totals.
Hourly calculation is better able to consider impacts of abrupt and gradual changes in
weather parameters during the course of a day on ET (Irmak et al., 2005; Allen et al.,
2006). Examples of this are high wind conditions during afternoons with low humid-
ity, overpass of cloud fronts and rain events, wintertime, and nighttime calm.
8.4.7 Limited Data Availability
Many historical weather data sets include only measurements of daily air tempera-
ture. When ETo estimates are desired, one of three approaches is recommended:
1. When approximate calculations of ETo are suitable, apply Equation 8.51.
2. When more accuracy is required, calibrate Equation 8.51 at a regional station or
stations that have Rs and u2 data. If Td or other humidity data are not available,
Td can be estimated from Tmin as described in Equation 8.16. The desired stan-
dard reference method can be used as the calibration basis (e.g., Equation 8.3) at
the regional station. The calibration can be of the form:

ET ref ET ref - a
b= or b = (8.52)
ET equation ET equation
where ETref is the reference ET against which to calibrate and ETequation is ET es-
timated by the equation being calibrated. Coefficients b and a are calibration
factors that may vary by month. These coefficients can be determined by linear
regression.
Reference ET is then calculated at locations with limited data as:
ETref = b ETequation or ETref = a + b ETequation (8.53)
As an example of regional calibration, Allen et al. (1983) determined values for
b by month for the state of Idaho when calibrating the FAO-Blaney-Criddle
equation against ETr. The user is cautioned, however, that use of faulty regional
weather data for calibration can create more error in the ET estimate than using
an original equation.
3. The user may elect to use the Penman or PM equation, estimating Rs, Td, and/or
u2. This is the approach recommended by FAO-56 for most situations. The bene-
fit is that a physically correct equation for ETo is used, so that local or regional
calibration should be unnecessary. The accuracy of the ETo estimate depends on
how well missing data parameters are estimated. FAO-56 and ASCE-EWRI
(2005) describe procedures for estimating Rs, Td and u2 , where Rs is estimated
from Tmax, Tmin, and Ra, Td is estimated from Tmin, and u2 is estimated from re-
gional averages or from some regional weather station. In some parts of the
U.S., gridded estimates of weather data are available from weather forecast sys-
tems such as by NCEP (McQueen et al., 2004). Allen and Robison (2007) ap-
plied the ASCE PM method at 107 National Weather Service stations in Idaho
where only daily maximum and minimum air temperature data were available
for long historic periods. They estimated Rs using a procedure by Thornton and
Running (1999) following regional calibration.
8.4.8 Weather Data Integrity
ETo and ETr estimates are only as good as the weather data used in their estimation.
Weather data should be screened to insure integrity and representativeness. This is
especially important with electronically collected data, since human oversight and
maintenance may be limited. Solar radiation can be screened by plotting measure-
ments against a clear sky Rso envelope provided by Equation 8.28, or a more accurate
and complicated method in ASCE-EWRI (2005) Appendix D, illustrated in Allen
(1996a). Humidity data (Td, RH, ea) can be evaluated by examining daily maximum
RH (RHmax) or by comparing Td with Tmin. Under reference conditions, RHmax generally
exceeds 80% during early morning and Td approaches Tmin (Allen 1996a; ASCE-
EWRI, 2005).
Weather data should be representative of the reference condition. Data collected at
or near airports can be negatively influenced by the local aridity, especially in arid and
semiarid climates. Data from dry or urban settings may cause overestimation of ETo or
ETr due to air temperature measurements that are too high and humidity measurements
that are too low, relative to the reference condition. Allen et al. (1998) and ASCE-
EWRI (2005) suggest simple adjustments for “nonreference” weather data to provide
data more reflective of well-watered settings. Allen and Gichuki (1989) and Ley et al.
(1996) have suggested more sophisticated approaches.
Often, substituting Td = Tmin – Ko for measured Td, as suggested in Equation 8.16,
can improve ETo estimates made with the combination equation when data are from a
nonreference setting. Using the nonreference data in Equation 8.16 will overestimate
the true Td and ea that would occur under reference conditions, since Tmin will be
higher in the dry setting and, consequently, so will estimated Td. However, because es
and ea in the nonreference setting are both inflated when calculated using Tmax, Tmin,
and Td estimated from Equation 8.16, the es – ea difference in the combination equa-
tion is often brought more in line with that expected for the reference condition and a
more accurate estimate for ETref results.
8.5 ESTIMATING ET FOR CROPS
The crop coefficient, Kc, has been developed over the past half-century to simplify
and standardize the calculation and estimation of crop water use. The Kc is defined as
the ratio of ET from a specific surface to ETref. The specific surface can be comprised
of bare soil, soil with partial vegetation cover, or full vegetation cover. The Kc repre-
sents an integration of effects of crop height, crop-soil resistance, and surface albedo
that distinguish the surface from the ETref definition. The value for Kc often changes
during the growing season as plants grow and develop, as the fraction of ground cov-
ered by vegetation changes, as the wetness of the underlying soil surface changes, and
as plants age and mature.
The potential crop ET is calculated by multiplying ETref by the crop coefficient:
ETc = Kc ETref (8.54)
The reference crop corresponds to a living, agricultural crop (clipped grass or full-
cover alfalfa) and it incorporates the majority of the effects of variable weather into
the ETref estimate. Because ETref represents an index of climatic evaporative demand,
the Kc varies predominately with specific crop characteristics and a small amount with
climate in the case of ETo. This enables the transfer of standard values for Kc between
locations and between climates. The transfer has led to the widespread acceptance and
usefulness of the Kc-ETref approach. Values for Kc based on ETo tend to average 1.2 to
1.4 times Kc based on ETr due to differences in magnitudes between the two ETref
types. For this reason, ASCE-EWRI (2005) has recommended differentiation between
the two families of Kc by referring to ETo-based Kc as Kco and to ETr-based Kc as Kcr.
The Kc and ETc in Equation 8.54 represent ET under potential growing conditions
with no stresses caused by shortage of soil water or salinity. These are the general
conditions for agricultural production. Both water and salinity stress reduce transpira-
tion and thus ET by causing plant canopies to reduce stomatal opening and water loss.
When water or salinity-induced reductions are considered, ET is calculated as:
ETc act = Kc act ETref (8.55)
where ETc act and Kc act represent ET and associated Kc under actual field conditions
that may include effects of environmental stresses.
8.5.1 Mean Kc and Dual Kc Methods
Two approaches to Kc are described in this section. The first approach uses a mean
or “single” Kc where time-averaged effects of evaporation from the soil surface are
averaged into the Kc value. The mean Kc represents, on any particular day, average
evaporation fluxes from the soil and plant surfaces. The second Kc approach is the
dual Kc method, where the Kc value is divided into a basal crop coefficient, Kcb, and a
separate component, Ke, representing evaporation from the soil surface. The basal crop
coefficient represents ET conditions when the soil surface is dry, but sufficient root
zone moisture is present to support full transpiration. Generally, a daily calculation
timestep is required to apply the dual Kc method, whereas the mean Kc method can be
applied on daily, weekly, or monthly timesteps.
The mean (single) Kc approach is used for planning studies and irrigation systems
design where averaged effects of soil wetting are appropriate. For typical irrigation
management, this use is valid. The dual Kc approach, which requires more numerical
calculations, is suited for irrigation scheduling, for soil water balance computations,
and for research studies where specific effects of day-to-day variation in soil surface
wetness and the resulting impacts on daily ETc, the soil water profile, and deep perco-
lation fluxes are important.
The form of the equation for Kc act in the dual Kc approach is:
Kc act = Ks Kcb + Ke (8.56)
where Kact = actual crop coefficient that considers effects of moisture stressa stress
reduction coefficient, 0 to 1
Ks = a stress reduction coefficient, 0 to 1
Kcb = basal crop coefficient, 0 to −1.4 for Kco and 0 to ~1.0 for Kcr
Ke = soil water evaporation coefficient, 0 to −1.4 for Kco and 0 to ~1.0 for Kc r.
All three terms are dimensionless. Kcb is defined as the ratio of ETc to ETref when the
soil surface layer is dry, but where the average soil water content of the root zone is
adequate to sustain full plant transpiration. Ks reduces the value of Kcb when the aver-
age soil water content of the root zone is not adequate to sustain full plant transpiration
and is described later. Ke quantifies the evaporation component from wet soil in addi-
tion to the evapotranspiration represented in Kcb.
The mean (single) Kc is a simplified, single coefficient that includes effects of time-
averaged Ke:
K c = K cb + K e (8.57)
and K c act = K s K c (8.58)
where Ke represents the time-averaged (i.e., multi-day) Ke.
8.5.2 The Crop Coefficient Curve
The crop coefficient curve represents the changes in Kc over the course of the
growing season, depending on changes in vegetation cover and maturation. During the
initial period, shortly after planting of annuals or shortly after the initiation of new
leaves for perennials, the value of Kc is small, often less than 0.4. Figure 8.1 illustrates
two general shapes of Kc curves as used by FAO (Doorenbos and Pruitt, 1977; Allen et
al., 1998) and as used by Wright (1982). The curves by Wright exhibit a smoothed
change in Kc with time, whereas Kc curves by FAO are constructed using linear line
segments.
The simple, linear FAO Kc curve is widely used and generally provides sufficiently
accurate description of linear crop growth and development and subsequently Kc, and
is recommended for most applications. Definitions for three benchmark Kc values re-
quired to construct the curve and associated definitions for growth stage periods and
relative ground cover are illustrated in Figure 8.2.
1.4
K c Wright
1.2 K c FAO
1.0
0.8
K c 0.6
0.4
0.2
0.0
May June July August
Figure 8.1. Typical crop coefficient curves from

Wright (1982) and from FAO-24 and FAO-56.
Kc 1.4 K c mid
1.2
0.8
0.6
Kc ini
0.4
K c end
0.2
0
time (days)
initial crop development mid-season late season
Figure 8.2. FAO Kc curve with four crop stages and three
Kc values relative to typical ground cover (from FAO-56).
Kc= K cb+ K e
Kc 1.4
1.2 Ke
1
0.8
0.6
0.4
0.2
Kcb
0
time (days)
initial crop development mid-season late season
Figure 8.3. Generalized FAO Kc curve definitions showing the basal Kcb,
soil evaporation, Ke, and time-averaged mean Kc values.
Typical shapes for the Kcb, Ke, and mean Kc curves are shown in Figure 8.3. The Kcb
curve represents the minimum Kc for conditions of adequate soil moisture and dry soil
surface. The Ke spikes represent increased evaporation following wetting of the soil
surface by precipitation or irrigation. The spikes reach a maximum of about 1.2 for Kco
and about 1.0 for Kcr. When summed, the values for Kcb and for Ke represent the total
crop coefficient, Kc. The mean Kc curve shown as the dashed line illustrates the effect
of averaging Ke over time and is displayed as a smoothed curve.
The mean Kc lies above the Kcb curve, with potentially large differences during the
initial and development stages, depending on the frequency of soil wetting. During the
midseason period when the crop is likely to be near full cover, the effects of soil wet-
ness are less pronounced.
8.5.3 Construction of the FAO Kc Curve
The linear FAO style Kc curve is constructed by the following steps.
1. Divide the growing period into four general growth stages describing crop
phenology and development (Figure 8.2). The four stages are: (1) initial period
from planting or green-up until about 10% ground cover; (2) crop development
period; (3) mid-season period from effective full cover to the beginning of the
late season period; and (4) late-season period from the beginning of senescence
until harvest, crop death, or full senescence.
2. Specify the three Kc values corresponding to Kc ini, Kc mid, and Kc end, where Kc ini
represents the average Kc during the initial period, Kc mid represents the average
Kc during the midseason period, and Kc end represents the Kc at the end of the
late-season period.
3. Connect straight line segments through each of the four growth stage periods,
with horizontal lines drawn through Kc ini during the initial period and through
Kc mid during the midseason period. Diagonal lines are drawn from Kc ini to Kc mid
within the domain of the development period and from Kc mid to Kc end within the
domain of the late-season period.
In the case of basal Kcb, the same procedure for curve construction is followed. Kc
mid represents the mean maximum Kc expected during the midseason period, and does
not necessarily represent the absolute peak Kc reached by the crop. This is illustrated
in Figure 8.1, where the smoothed Kc curve, in this case for dry, edible beans devel-
oped by Wright (1982; from Jensen et al. 1990), rises above the linearized curve dur-
ing part of the midseason. As illustrated, the linearized FAO curve approximates the
trends in Kc, and accurately estimates seasonal ET when the lengths of the growth
stages and the value for Kc mid are properly selected.
Values for the grass-based Kco ini, Kco mid, and Kco end, and for Kcbo ini, Kcbo mid, and
Kcbo end are listed in Table 8.2 for various agricultural crops (the o subscript indicates
an ETo basis). The three Kco columns represent typical irrigation management and pre-
cipitation frequencies. The Kc values are taken from FAO-56 (Allen et al., 1998) and
are largely based on Doorenbos and Pruitt (1977) and Doorenbos and Kassam (1979).
A primary departure made here is for trees and grapes where, rather than listing values
for Kc and Kcb, values are instead given for Kcb full, Kc min and Kcb cover to be used in
Equation 8.85 (with a density coefficient from Equation 8.80) introduced later. This
modification provides flexibility in adjusting the value for Kc and Kcb according to the
fraction of ground shaded by canopy, which is highly variable among orchards and
cultures. Parameters are listed in footnotes for applying Equation 8.80 that reproduce
various Kc values reported in the literature. The Kco in Table 8.2 are applicable with
grass reference ETo as defined by the standardized FAO/ASCE PM Equation (8.3) and
are generally valid for use with other grass reference equations, provided these agree
relatively closely with the standardized PM definition (ASCE-EWRI, 2005).
For several group types, only one value for Kc ini is listed for the group, since tabu-
larized Kc ini are only approximate. Figures 8.5a-8.5c (in Section 8.5.3.4), from FAO-56,
improve estimates for Kc ini by accounting for frequency of soil wetting and soil type.
8.5.3.1 The climatic basis of Table 8.2. The Kc mid values in Table 8.2 are typical
values expected under a standard climatic condition defined in FAO-56 (Allen et al.,
1998) as a subhumid climate having average daytime minimum relative humidity
(RHmin) of 45 % and having calm to moderate wind speeds averaging 2 m s-1. More
arid climates and conditions of greater wind speed have higher values for Kc and Kcb,
especially for tall crops and more humid climates, and conditions of lower wind speed
have lower values according to the relationship:
0.3
⎛h⎞
K c = K c table + [0.04 (u 2 − 2) − 0.004 (RH min − 45)] ⎜ ⎟ (8.59)
⎝3⎠
where Kc table is the Kc value (or Kcb value) from Table 8.2 for Kc mid (and for Kc end
when Kc end > 0.45) and h is mean crop height in m. The Kc for crops 2 to 3 m in height
can increase by as much as 40% when going from a calm, humid climate (for example,
u2 = 1 m s-1 and RHmin = 70%) to an extremely windy, arid climate (for example, u2 =
5 m s-1 and RHmin = 15%). The increase in Kc is due to the influence of the larger aero-
dynamic roughness of tall crops relative to grass on the transport of water vapor from
the surface. The adjustments to Kc for climate are generally made using mean values
for u2 and RHmin for the entire midseason period.
8.5.3.2 Lengths of crop growth stages. The four crop growth stages for the FAO
Kco curves are characterized in terms of crop growth benchmarks as illustrated in Fig-
ure 8.2. The crop development period is presumed to begin when approximately 10%
of the ground is covered by vegetation and ends at attainment of effective full cover.
Effective cover can be defined for row crops such as beans, sugar beets, potatoes, and
corn as the time when some leaves of plants in adjacent rows begin to intermingle so
that soil shading becomes nearly complete, or when plants reach nearly full size, if no
intermingling occurs. For crops taller than 0.5 m, the average fraction of the ground
surface covered by vegetation at the time of effective full cover is about 0.7 to 0.8
(Neale et al., 1989; Grattan et al., 1998). Effective full cover for many crops begins at
the initiation of flowering. It is understood that the crop or plant can continue to grow
in both height and leaf area after the attainment of effective full cover. Another way to
estimate the occurrence of full cover is when the leaf area index (LAI) reaches about 3
(Ritchie, 1972; Wright, 1982; Ritchie and NeSmith, 1991).
The lengths of the initial and development periods are relatively short for decidu-
ous trees and shrubs, depending on the amount of pruning, which develop new leaves
in the spring at relatively fast rates. These two periods may thus be only a few days in
length for trees and may include the flowering period. The Kc ini selected for trees and
shrubs should reflect the ground condition prior to leaf emergence or initiation, since
Kc ini is affected by the amount of grass or weed cover, soil wetness, tree density, and
mulch density.
Table 8.2. Mean crop coefficients, Kco, and basal crop coefficients, Kcbo, for
well-managed crops in a subhumid climate, for use with ETos (following FAO-56*).
Crop Kc ini[1] Kc mid Kc end Kcb ini Kcb mid Kcb end
a. Small Vegetables 0.7 1.05 0.95 0.15 0.95 0.85
Broccoli 1.05 0.95 0.95 0.85
Brussels sprouts 1.05 0.95 0.95 0.85
Cabbage 1.05 0.95 0.95 0.85
Carrots 1.05 0.95 0.95 0.85
Cauliflower 1.05 0.95 0.95 0.85
Celery 1.05 1.00 0.95 0.90
Garlic 1.00 0.70 0.90 0.60
Lettuce 1.00 0.95 0.90 0.90
Onions, dry 1.05 0.75 0.95 0.65
green 1.00 1.00 0.90 0.90
seed 1.05 0.80 1.05 0.70
Spinach 1.00 0.95 0.90 0.85
Radish 0.90 0.85 0.85 0.75
b. Vegetables, Solanum
Family (Solanaceae) 0.6 1.15 0.80 0.15 1.10 0.70
Eggplant 1.05 0.90 1.00 0.80
Sweet peppers (bell) 1.05[2] 0.90 1.00[2] 0.80
Tomato 1.15[2] 0.70-0.90 1.10[2] 0.60-0.80
c. Vegetables, Cucumber
Family (Cucurbitaceae) 0.5 1.00 0.80 0.15 0.95 0.70
Cantaloupe 0.5 0.85 0.60 0.75 0.50
Cucumber, fresh market 0.6 1.00[2] 0.75 0.95[2] 0.70
machine harvest 0.5 1.00 0.90 0.95 0.80
Pumpkin, winter squash 1.00 0.80 0.95 0.70
Squash, zucchini 0.95 0.75 0.90 0.70
Sweet melons 1.05 0.75 1.00 0.70
Watermelon 0.4 1.00 0.75 0.95 0.70
d. Roots and Tubers 0.5 1.10 0.95 0.15 1.00 0.85
Beets, table 1.05 0.95 0.95 0.85
Cassava, year 1 0.3 0.80[3] 0.30 0.70[3] 0.20
year 2 0.3 1.10 0.50 1.00 0.45
Parsnip 0.5 1.05 0.95 0.95 0.85
Potato 1.15 0.75[4] 1.10 0.65[4]
Sweet potato 1.15 0.65 1.10 0.55
Turnip (and rutabaga) 1.10 0.95 1.00 0.85
Sugar beet 0.35 1.20 0.70[5] 1.15 0.50[5]
* Primary sources of Table 8.2: FAO-56 (Allen et al., 1998), with Kc ini traceable to Doorenbos and Kassam (1979) and
Kc mid and Kc end traceable to Doorenbos and Pruitt (1977), Pruitt (1986), Wright (1981, 1982), and Snyder et al. (1989a,b).
[1]
These are general values for Kc ini under typical irrigation management and soil wetting. For frequent wettings such as
with high frequency sprinkle irrigation or daily rainfall, these values may increase substantially and may approach 1.0
to 1.2. Kc ini is a function of wetting interval and potential evaporation rate during the initial and development periods
and is more accurately estimated using Figure 8.5 or using the dual Kcb ini + Ke.
[2]
Beans, peas, legumes, tomatoes, peppers and cucumbers are sometimes grown on stalks reaching 1.5 to 2 meters in
height. In such cases, increased Kc values need to be taken. For green beans, peppers and cucumbers, 1.15 can be taken,
and for tomatoes, dry beans and peas, 1.20. Under these conditions h should be increased also.
[3]
The midseason values for cassava assume non-stressed conditions during or following the rainy season. The Kc end and
Kcb end values account for dormancy during the dry season.
[4]
The Kc end and Kcb end values for potatoes are about 0.40 and 0.35 for long season potatoes with vine kill.
[5]
This Kc end and Kcb end values are for no irrigation during the last month of the growing season. The Kc end and Kcb end
values for sugar beets are higher, up to 1.0 and 0.9, when irrigation or significant rain occurs during the last month.
(continued)
e. Legumes (Leguminosae) 0.4 1.15 0.55 0.15 1.10 0.50
Beans, green 0.5 1.05[2] 0.90 1.00[2] 0.80
Beans, dry, and pulses 0.4 1.15[2] 0.35 1.10[2] 0.25
Chick pea 1.00 0.35 0.95 0.25
Faba bean (broad bean), fresh 0.5 1.15[2] 1.10 1.10[2] 1.05
dry/seed 0.5 1.15[2] 0.30 1.10[2] 0.20
Garbanzo 0.4 1.15 0.35 1.05 0.25
Green gram and cowpeas 1.05 0.60-0.35[6] 1.00 0.55-0.25[6]
Groundnut (peanut) 1.15 0.60 1.10 0.50
Lentil 1.10 0.30 1.05 0.20
Peas, fresh 0.5 1.15[2] 1.10 1.10[2] 1.05
dry/seed 1.15 0.30 1.10 0.20
Soybeans 1.15 0.50 1.10 0.30
f. Perennial Vegetables (with winter dormancy and
initially bare or mulched) 0.5 1.00 0.80
Artichokes 0.5 1.00 0.95 0.15 0.95 0.90
Asparagus 0.5 0.95[7] 0.30 0.15 0.90[7] 0.20
Mint 0.60 1.15 1.10 0.40 1.10 1.05
Strawberries 0.40 0.85 0.75 0.30 0.80 0.70
g. Fiber Crops 0.35 0.15
Cotton 1.15-1.20 0.70-0.50 1.10-1.15 0.50-0.40
Flax 1.10 0.25 1.05 0.20
Sisal[8] 0.4-0.7 0.4-0.7 0.4-0.7 0.4-0.7
h. Oil Crops 0.35 1.15 0.35 0.15 1.10 0.25
Castor bean (Ricinus) 1.15 0.55 1.10 0.45
Rapeseed, canola 1.0-1.15[9] 0.35 0.95-1.10[9] 0.25
Safflower 1.0-1.15[9] 0.25 0.95-1.10[9] 0.20
Sesame 1.10 0.25 1.05 0.20
Sunflower 1.0-1.15[9] 0.35 0.95-1.10[9] 0.25
i. Cereals 0.3 1.15 0.4 0.15 1.10 0.25
Barley, oats 1.15 0.25 1.10 0.15
Spring wheat 1.15 0.25-0.4[10] 1.10 0.15-0.3[10]
[10] [11]
Winter wheat, frozen soils 0.4 1.15 0.25-0.4 0.15-0.5 1.10 0.15-0.3[10]
with non-frozen soils 0.7 1.15 0.25-0.4[10]
Maize, field (grain corn) [12] 1.20 0.60,0.35 0.15 1.15 0.50,0.15
Maize, sweet (sweet corn) [12] 1.15 1.05[13] 1.10 1.00[13]
Millet 1.00 0.30 0.95 0.20
Sorghum, grain 1.00-1.10 0.55 0.95-1.05 0.35
sweet 1.20 1.05 1.15 1.00
Rice 1.05 1.20 0.90-0.60 1.00 1.15 0.70-0.45
[6]
The first Kc end is for harvested fresh. The second value is for harvested dry.
[7]
The Kc for asparagus usually remains at Kc ini during harvest of the spears, due to sparse ground cover. The Kc mid value
is for following regrowth of plant vegetation following termination of harvest of spears.
[8]
Kc for sisal depends on the planting density and water management (e.g., intentional moisture stress).
[9]
The lower values are for rainfed crops having less dense plant populations.
[10]
The higher value is for hand-harvested crops.
[11]
The two Kcb ini values for winter wheat are for less than 10% ground cover and for during the dormant, winter period, if
the vegetation fully covers the ground, but conditions are nonfrozen.
[12]
The Kc mid and Kcb mid values are for populations > 50,000 plants ha-1. For less dense populations or less uniform growth,
Kc mid and Kcb mid can be reduced by 0.10 to 0.2. The first Kc end value is for harvest at high grain moisture. The second Kc
end value is for harvest after complete field drying of the grain (to about 18% moisture, wet mass basis).
[13]
If harvested fresh for human consumption. Use Kc end for field maize if the sweet maize is allowed to mature and dry in the field.
(continued)
j. Forages
Alfalfa hay, ave.cutting effects 0.40 0.95[13] 0.90
individual cutting periods 0.40[14] 1.20[14] 1.15[14] 0.30[14] 1.15[14] 1.10[14]
for seed 0.40 0.50 0.50 0.30 0.45 0.45
Bermuda hay, ave.cutting effects 0.55 1.00[13] 0.85 0.50 0.95[15] 0.80
spring crop for seed 0.35 0.90 0.65 0.15 0.85 0.60
Clover hay, berseem,
averaged cutting effects 0.40 0.90[13] 0.85
individual cutting periods 0.40[14] 1.15[14] 1.10[14] 0.30[14] 1.10[14] 1.05[14]
Ryegrass hay, ave. cuttings 0.95 1.05 1.00 0.85 1.00[15] 0.95
Sudan grass hay (annual),
averaged cutting effects 0.50 0.90[13] 0.85
[14]
individual cutting periods 0.50 1.15[14] 1.10[14] 0.30[14] 1.10[14] 1.05[14]
Grazing pasture, rotated grazing 0.40 0.85-1.05 0.85 0.30 0.80-1.00 0.80
extensive grazing 0.30 0.75 0.75 0.30 0.70 0.70
Turf grass, cool season[16] 0.90 0.90 0.90 0.80 0.85 0.85
warm season[16] 0.85 0.90 0.90 0.75 0.80 0.80
k. Sugar Cane 0.40 1.25 0.75 0.15 1.20 0.70
l. Tropical Fruits and Trees
Banana, 1st year 0.50 1.10 1.00 0.15 1.05 0.90
2nd year 1.00 1.20 1.10 0.60 1.10 1.05
Cacao 1.00 1.05 1.05 0.90 1.00 1.00
Coffee, bare ground cover 0.90 0.95 0.95 0.80 0.90 0.90
with weeds 1.05 1.10 1.10 1.00 1.05 1.05
Date palms 0.90 0.95 0.95 0.80 0.85 0.85
Palm trees 0.95 1.00 1.00 0.85 0.90 0.90
Pineapple,[17] bare soil 0.50 0.30 0.30 0.15 0.25 0.25
with grass cover 0.50 0.50 0.50 0.30 0.45 0.45
Rubber trees 0.95 1.00 1.00 0.85 0.90 0.90
Tea, non-shaded 0.95 1.00 1.00 0.90 0.95 0.90
shaded[18] 1.10 1.15 1.15 1.00 1.10 1.05
m. Berries and Hops
Berries (bushes) 0.30 1.05 0.50 0.20 1.00 0.40
Hops 0.30 1.05 0.85 0.15 1.00 0.80
[13]
This Kc mid for hay crops represents an averaged Kc for before and following cuttings. It is applied to the period follow-
ing the first development period until the beginning of the last late season period of the growing season.
[14]
These Kc coefficients for hay crops represent immediately following cutting; at full cover; and immediately before
cutting, respectively. The growing season is described as a series of individual cutting periods (Figure 8.4).
[15]
This Kcb mid for bermuda and ryegrass hay crops represents an averaged Kcb mid for before and following cuttings. It is
applied to the period following the first development period until the beginning of the last late season period.
[16]
Cool season grass varieties include dense stands of bluegrass, ryegrass, and fescue. Warm season varieties include
bermuda and St. Augustine. Values given are for potential conditions representing a 0.06- to 0.08-m mowing height.
Turf, especially warm season varieties, can be stressed at moderate levels and still maintain appearance (see Section 8.6
and Table 8.13). Generally a value for the stress coefficient Ks of 0.9 for cool-season and 0.7 for warm-season varieties
can be employed where careful water management is practiced and rapid growth is not required. Incorporating these Ks
values into an “actual Kc” will yield Kc act values of about 0.8 for cool-season and 0.65 for warm-season turf.
[17]
The pineapple plant has very low transpiration because it closes its stomates during the day and opens them during the
night. Therefore, the majority of ETc from pineapple is evaporation from the soil. The Kc mid < Kc ini since Kc mid occurs during full
ground cover so that soil evaporation is less. Values assume that 50% of the ground surface is covered by black plastic mulch
and that irrigation is by sprinkler. For drip irrigation beneath the plastic mulch, Kc values given can be reduced by 0.10.
[18]
Includes the water requirements of the shade trees.
(continued)
n. Fruit Trees and Grapes
Conifer trees[19] 1.00 1.00 1.00 0.95 0.95 0.95
Kiwi 0.40 1.05 1.05 0.20 1.00 1.00
The following, shaded, section for trees and grapes gives values for Kcb full used in Eq. 8.85.
Kcb full[20] Kcb full[20] Kcb full[20] Kc min[20] Kcb cover[20] Kcb cover[20]
initial mid end initial mid, end
Almonds, no ground cover[21] 0.20 1.00 0.70[22] 0.15
ground cover 0.20 1.00 0.70[22] 0.15 0.75 0.80
Apples, cherries, pears;
killing frost[23] 0.30 1.15 0.80[22] 0.15 0.40 0.80
no killing frost[23] 0.30 1.15 0.80[22] 0.15 0.75 0.80
Apricots, peaches, pears,
plums, pecans;
killing frost[24] 0.30 1.20 0.80[22] 0.15 0.40 0.80
[25]
no killing frost 0.30 1.20 0.80[22] 0.15 0.70 0.80
Avocado, no grnd cover[26] 0.30 1.00 0.90 0.15
ground cover 0.30 1.00 0.90 0.15 0.75 0.80
Citrus[27] 0.80 0.80 0.80 0.15 0.75 0.80
Mango, no ground cover[28] 0.25 0.85 0.70 0.15
Olives[29] 0.60 0.70 0.60 0.15 0.70 0.70
Pistachios 0.30 1.00 0.70 0.15 0.70 0.70
Walnut[30] 0.40 1.10 0.65 0.15 0.75 0.80
Grapes, table or raisin[31] 0.20 1.15 0.90 0.15 0.70 0.70
wine[32] 0.20 0.80 0.60 0.15 0.70 0.70
[19]
Conifers exhibit substantial stomatal control due to soil water deficit. The Kc can easily reduce below the values pre-
sented, which represent well-watered conditions for large forests.
[20]
The first three columns for the orchard crops in the following rows are values for Kcb full for initial, mid- and end-season
periods to be used in Eq. 8.85 along with the Kc min of column 4 to calculate Kcb for the initial, midseason and late-
season periods, where the density coefficient, Kd, is calculated from Eq. 8.80 using effective fraction of cover fc eff and
plant height, h, as noted in the following footnotes. The last two columns are Kcb cover for initial and for mid- and end-
season periods to be used in Eq. 8.85 for when there is active ground cover. Generally the value for Kc ini is estimated as
0.10 + Kcb ini from Eq. 8.85 and Kc mid and Kc end are estimated as 0.05 + Kcb mid or Kcb end from Eq. 8.85.
[21]
Apply Eq. 8.80 with fc eff = 0.4, ML = 1.5 and h = 4 m for Kd in Eq. 8.85 to derive Kcb values similar to FAO-56.
[22]
These Kc end values represent Kc prior to leaf drop. After leaf drop, Kc end ≈ 0.20 for bare, dry soil or dead ground cover
and Kc end ≈ 0.50 to 0.80 for actively growing ground cover.
[23]
Apply Eq. 8.80 with fc eff = 0.5, ML = 2 and h = 3 m for Kd in Eq. 8.85 to derive Kcb values similar to FAO-56.
[24]
[25]
Apply Eq. 8.80 with fc eff = 0.8, ML = 1.5 and h = 3 m for Kd in Eq. 8.85 to derive Kc values similar to Johnson et al.
(2005), with fc eff = 0.6, ML = 1.5, h = 3 m for Kc values similar to those derived from Girona et al. (2005), with fc eff =
0.45, ML = 1.5, h = 3 m for Kcb values similar to FAO-56, with fc eff = 0.29, ML = 1.5, h = 2.5 m for Kc values similar to
Paço et al. (2006), and with any fc eff ≤ 0.7, ML = 1.5, h = 3 m to approximate Kc estimates by Ayars et al. (2003).
[26]
Apply Eq. 8.80 with fc eff = 0.4, ML = 2 and h = 4 m for Kd in Eq. 8.85 to derive Kcb values similar to FAO-56.
[27]
Apply Eq. 8.80 with fc eff = 0.2, 0.5 and 0.7, ML = 1.5 and h = 2, 2.5 and 3 m for Kd in Eq. 8.85 to derive recommended
Kcb values that are about 15% higher than the values entered in FAO-56 for these same three levels of fc eff .
[28]
Apply Eq. 8.80 with fc eff = 0.7 to 0.85, ML = 1.5 and h = 5 m for Kd in Eq. 8.85 for Kcb values similar to those derived
from de Azevedo et al. (2003).
[29]
Apply Eq. 8.80 with fc eff = 0.7, ML = 1.5 and h = 4 m for Kd in Eq. 8.85 to derive Kcb values similar to FAO-56. Apply
with fc eff = 0.6, ML = 1.5, h = 4 m to derive Kc values similar to Pastor and Orgaz (1994), but with Kcb mid = 0.45. Apply
with fc eff = 0.3 to 0.4, ML = 1.5, h = 4 m to derive Kcb values similar to Villalobos et al. (2000). Apply with fc eff = 0.05
and 0.25, ML = 1.5, h = 2 and 3 m to derive Kc values similar to Testi et al. (2004).
[30]
[31]
Apply Eq. 8.80 with fc eff = 0.65, ML = 1.5 and h = 2 m for Kd in Eq. 8.85 to derive Kc values similar to Johnson.et al.
(2005). Apply with fc eff = 0.45, ML = 1.5, h = 2 m for Kcb values similar to FAO-56.
[32]
(continued)

Crop Kc ini[1] Kc mid Kc end
o. Wetlands, Temperate Climate
Cattails, bulrushes,
killing frost 0.30 1.20 0.30
Cattails, bulrushes, no frost 0.60 1.20 0.60
Short veg., no frost 1.05 1.10 1.10
Reed swamp, standing water 1.00 1.20 1.00
Reed swamp, moist soil 0.90 1.20 0.70
p. Special
Open water, < 2 m depth or
in subhumid climates or
tropics 1.05 1.05
Open water, > 5 m depth,
clear of turbidity, temperate
climate 0.65[33] 1.25[33]
[33]
These Kc values are for deep water in temperate latitudes where large temperature changes in the water body occur
during the year, and initial and peak period evaporation is low as radiation energy is absorbed into the deep water body.
During fall and winter periods (Kc end), heat is released from the water body that increases the evaporation above that for
grass. Therefore, Kc mid corresponds to the period when the water body is gaining thermal energy and Kc end when releas-
ing thermal energy. These Kc values should be used with caution.
The start of maturity and beginning of decline in Kc is often signaled by the begin-
ning of aging, yellowing or senescence of leaves, leaf drop, or the browning of fruit so
that ETc is reduced relative to ETo. Calculations for Kc and ETc are sometimes pre-
sumed to end when the crop is harvested, dries out naturally, reaches full senescence,
or experiences leaf drop. For some perennial vegetation in frost-free climates, crops
may grow year round so that the date of termination is the same as the date of “plant-
ing.” The length of the late season period may be relatively short (less than 10 days)
for vegetation killed by frost (for example, maize at high elevations in latitudes
> 40°N) or for agricultural crops that are harvested fresh (for example, table beets and
small vegetables). The value for Kc end should reflect the condition of the soil surface
(average water content and any mulch cover) and the condition of the vegetation fol-
lowing harvest or after full senescence. Kc during nongrowing periods having little or
no green ground cover can be estimated using the equation for Kc ini as described later.
FAO-56 provides general lengths of growth (development) stages for a wide vari-
ety of crops under different climates and locations. This information is reproduced in
Table 8.3. The lengths in Table 8.3 serve only to indicate typical proportions of grow-
ing season lengths under a variety of climates. In all applications, local observations of
the specific plant stage development should be made to account for local effects of
plant variety, climate, and cultural practices. Local information can be obtained by
interviewing farmers, ranchers, agricultural extension agents, and local researchers, by
conducting local surveys, or by remote sensing (Neale et al., 1989; Tasumi et al.,
2005).
Table 8.3. Lengths of crop development stages* for various

planting periods and climatic regions (days) (from FAO-56).
Init. Dev. Mid Late
Crop (Lini) (Ldev) (Lmid) (Llate) Total Plant Date Region
a. Small Vegetables
Broccoli 35 45 40 15 135 Sept Calif. desert, U.S.
Cabbage 40 60 50 15 165 Sept Calif. desert, U.S.
20 30 50/30 20 100 Oct/Jan arid climate
Carrots 30 40 60 20 150 Feb/Mar Mediterranean
30 50 90 30 200 Oct Calif. desert, U.S.
Cauliflower 35 50 40 15 140 Sept Calif. desert, U.S.
25 40 95 20 180 Oct (semi)arid
Celery 25 40 45 15 125 April Mediterranean
30 55 105 20 210 Jan (semi)arid
20 30 20 10 80 April Mediterranean
Crucifers[1] 25 35 25 10 95 February Mediterranean
30 35 90 40 195 Oct/Nov Mediterranean
30 40 25 10 105 Nov/Jan Mediterranean
Lettuce
25 35 30 10 100 Oct/Nov arid region
35 50 45 10 140 Feb Mediterranean
Onion, dry
20 35 110 45 210 Oct; Jan. arid region; Calif.
25 30 10 5 70 April/May Mediterranean
Onion, green 20 45 20 10 95 October arid region
30 55 55 40 180 March Calif., U.S.
Onion, seed 20 45 165 45 275 Sept Calif. desert, U.S.
20 20 15/25 5 60/70 Apr; Sep/Oct Mediterranean
Spinach
20 30 40 10 100 November arid region
5 10 15 5 35 Mar/Apr Medit.; Europe
Radish
10 10 15 5 40 Winter arid region
b. Vegetables, Solanum Family (Solanaceae)
30 40 40 20 130 October arid region
Egg plant
30 45 40 25 140 May/June Mediterranean
Sweet peppers 25/30 35 40 20 125 April/June Europe and Medit.
(bell) 30 40 110 30 210 October arid region
30 40 40 25 135 January arid region
35 40 50 30 155 Apr/May Calif., U.S.
Tomato 25 40 60 30 155 Jan Calif. desert, U.S.
30 40 45 30 145 April/May Mediterranean
c. Vegetables, Cucumber Family (Cucurbitaceae)
30 45 35 10 120 Jan Calif., U.S.
Cantaloupe
10 60 25 25 120 Aug Calif., U.S.
20 30 40 15 105 June/Aug arid region
Cucumber
25 35 50 20 130 Nov; Feb arid region
* Lengths of crop development stages provided in this table are indicative of general conditions, but may
vary substantially from region to region, with climate and cropping conditions, and with crop variety.
The user is strongly encouraged to obtain appropriate local information.
[1]
Crucifers include cabbage, cauliflower, broccoli, and Brussels sprouts.
(continued)
Init. Dev. Mid Late
Pumpkin, 20 30 30 20 100 Mar, Aug Mediterranean
winter squash 25 35 35 25 120 June Europe
Squash, 25 35 25 15 100 Apr; Dec. Medit.; arid reg.
zucchini 20 30 25 15 90 May/June Medit.; Europe
25 35 40 20 120 May Mediterranean
30 30 50 30 140 March Calif., U.S.
Sweet melons
15 40 65 15 135 Aug Calif. desert, U.S.
30 45 65 20 160 Dec/Jan arid region
20 30 30 30 110 April Italy
Watermelons
10 20 20 30 80 Mat/Aug Near East (desert)
d. Roots and Tubers
15 25 20 10 70 Apr/May Mediterranean
Beets, table
25 30 25 10 90 Feb/Mar Mediterranean, arid
Cassava, yr 1 20 40 90 60 210 Rainy season tropical regions
yr 2 150 40 110 60 360
25 30 30/45 30 115/130 Jan/Nov (semi)arid climate
25 30 45 30 130 May continental climate
Potato 30 35 50 30 145 April Europe
45 30 70 20 165 Apr/May Idaho, U.S.
30 35 50 25 140 Dec Calif. Desert, U.S.
Sweet potato
15 30 50 30 125 Rainy seas. tropical regions
30 45 90 15 180 March Calif., U.S.
25 30 90 10 155 June Calif., U.S.
25 65 100 65 255 Sept Calif. desert, U.S.
Sugar beet 50 40 50 40 180 April Idaho, U.S.
25 35 50 50 160 May Mediterranean
45 75 80 30 230 November Mediterranean
35 60 70 40 205 November arid regions
e. Legumes (Leguminosae)
20 30 30 10 90 Feb/Mar Calif.,Mediterranean
Beans, green
15 25 25 10 75 Aug/Sep Calif.,Egypt,Lebanon
20 30 40 20 110 May/June continental climate
Beans, dry 15 25 35 20 95 June Pakistan, Calif.
25 25 30 20 100 June Idaho, U.S.
Faba bean 15 25 35 15 90 May Europe
Broad bean 20 30 35 15 100 Mar/Apr Mediterranean
Dry bean 90 45 40 60 235 Nov Europe
Green bean 90 45 40 0 175 Nov Europe
Green gram,
cowpeas 20 30 30 20 110 March Mediterranean
25 35 45 25 130 Dry season West Africa
Groundnut 35 35 35 35 140 May high latitudes
20 30 60 40 150 April Europe
Lentil
15 25 35 15 90 May Europe
Peas 20 30 35 15 100 Mar/Apr Mediterranean
35 25 30 20 110 April Idaho, U.S.
(continued)

Init. Dev. Mid Late
15 15 40 15 85 Dec tropics
Soybeans 20 30/35 60 25 140 May central U.S.
20 25 75 30 150 June Japan
f. Perennial Vegetables (with winter dormancy and initially bare or mulched soil)
40 40 250 30 360 Apr (1st yr) California
Artichoke
20 25 250 30 325 May (2nd yr) (cut in May)
50 30 100 50 230 Feb warm winter
Asparagus
90 30 200 45 365 Feb Mediterranean
g. Fiber Crops
30 50 60 55 195 Mar-May Egypt,Pakistan,Cal.
45 90 45 45 225 Mar Calif. desert,U.S.
Cotton
30 50 60 55 195 Sept Yemen
30 50 55 45 180 April Texas
25 35 50 40 150 April Europe
Flax
30 40 100 50 220 October Arizona
h. Oil Crops
25 40 65 50 180 March (semi)arid climate
Castorbeans
20 40 50 25 135 Nov. Indonesia
20 35 45 25 125 April California, U.S.
Safflower 25 35 55 30 145 Mar high latitudes
Sesame 20 30 40 20 100 June China
Sunflower 25 35 45 25 130 April/May Medit.; Calif.
i. Cereals
15 25 50 30 120 November central India
20 25 60 30 135 March/Apr 35-45°L
Barley, July
15 30 65 40 150 East Africa
oats,
40 30 40 20 130 Apr
wheat
40 60 60 40 200 Nov
20 50 60 30 160 Dec Calif. desert,U.S.
20[2] 60[2] 70 30 180 December California, U.S
Winter
30 140 40 30 240 November Mediterranean
wheat
160 75 75 25 335 October Idaho, U.S.
Grains 20 30 60 40 150 April Mediterranean
(small) 25 35 65 40 165 Oct/Nov Pakistan; arid reg.
30 50 60 40 180 April East Africa (alt.)
25 40 45 30 140 Dec/Jan arid climate
Maize 20 35 40 30 125 June Nigeria (humid)
(grain) 20 35 40 30 125 October India (dry, cool)
30 40 50 30 150 April Spain(spr,sum);Cal.
30 40 50 50 170 April Idaho, U.S.
[2]
These periods for winter wheat will lengthen in frozen climates according to days having zero
growth potential and wheat dormancy. Under general conditions and in the absence of local data, fall
planting of winter wheat can be presumed to occur in northern temperate climates when the 10-day
running average of mean daily air temperature decreases to 17°C or December 1, whichever comes
first. Planting of spring wheat can be presumed to occur when the 10-day running average of mean
daily air temperature increases to 5°C. Spring planting of maize-grain can be presumed to occur
when the 10-day running average of mean daily air temperature increases to 13°C.
(continued)
Init. Dev. Mid Late
20 20 30 10 80 March Philippines
Maize
20 30 50/30 10 90 Oct/Dec arid climate
(sweet)
30 30 30 10[3] 110 April Idaho, U.S.
20 40 70 10 140 Jan Calif. desert, U.S.
15 25 40 25 105 June Pakistan
Millet
20 30 55 35 140 April central U.S.
20 35 40 30 130 May/June U.S., Pakis., Med.
Sorghum
20 35 45 30 140 Mar/April arid region
30 30 60 30 150 Dec; May tropics; Mediterr.
Rice
30 30 80 40 180 May tropics
j. Forages
Alfalfa, total last –4°C in spring
10 30 var. var. var.
season[4] until first -4°C in fall
Alfalfa, 1st 10 20 20 10 60 Jan Calif., U.S.
cutting 10 30 25 10 75 Apr Idaho, U.S.
cycle[4] (last –4°C)
Alfalfa, other
5 10 10 5 30 Mar Calif., U.S.
cutting
[4] 5 20 10 10 45 Jun Idaho, U.S.
cycles
Bermuda
10 25 35 35 105 March Calif. desert, U.S.
for seed
Bermuda
for hay
10 15 75 35 135 – Calif. desert, U.S.
(several
cuttings)
7 days before last
Grass –4°C in spring
10 20 – – –
pasture[4] until 7 days after
first –4°C in fall
Sudan, 1st
25 25 15 10 75 Apr Calif. desert, U.S.
cutting cycle
Sudan, other
3 15 12 7 37 June Calif. desert, U.S.
cutting cycles
k. Sugar Cane
35 60 190 120 405 low latitudes
Sugar cane,
50 70 220 140 480 tropics
virgin
75 105 330 210 720 Hawaii, U.S.
25 70 135 50 280 low latitudes
Sugar cane,
30 50 180 60 320 tropics
ratoon
35 105 210 70 420 Hawaii, U.S.
Banana, 1st yr 120 90 120 60 390 Mar Mediterranean
Banana, 2nd yr 120 60 180 5 365 Feb Mediterranean
Pineapple 60 120 600 10 790 Hawaii, U.S.
[3]
The late season for sweet maize will be about 35 days if the grain is allowed to mature and dry.
[4]
In climates having killing frosts, growing seasons can be estimated for alfalfa and grass as:
alfalfa: last –4°C in spring until first –4°C in fall (Everson et al., 1978).
grass: 7 days before last –4°C in spring and 7 days after last –4°C in fall (Kruse and Haise, 1974).
(continued)

Init. Dev. Mid Late
m. Grapes and Berries
20 40 120 60 240 April low latitudes
20 50 75 60 205 Mar Calif., U.S.
Grapes
20 50 90 20 180 May high latitudes
30 60 40 80 210 April mid-latitudes (wine)
Hops 25 40 80 10 155 April Idaho, U.S.
n. Fruit Trees
Citrus 60 90 120 95 365 Jan Mediterranean
Deciduous 20 70 90 30 210 March high latitudes
orchard, 20 70 120 60 270 March low latitudes
little pruning 30 50 130 30 240 March Calif., U.S.
Olives 30 90 60 90 270[5] March Mediterranean
Pistachios 20 60 30 40 150 Feb Mediterranean
Walnuts 20 10 130 30 190 April Utah, U.S.
o. Wetlands, Temperate Climate
Wetlands 10 30 80 20 140 May Utah, U.S.(killing frost)
(cattails, 180 60 90 35 365 November Florida, U.S.
bulrush)
Wetlands
180 60 90 35 365 November frost-free climate
(short veg.)
[5]
Olive trees gain new leaves in March and often have transpiration during winter, where the Kc continues
outside of the “growing period” and total season length may be set to 365 days.
8.5.3.3 Kc curves for forage crops. Many crops grown for forage or hay receive
multiple harvests during the growing season. Each harvest essentially terminates a
sub-growing season and associated Kc curve and initiates a new sub-growing season
and associated Kc curve. The resulting Kc curve for the entire growing season is the
aggregation of a series of Kc curves associated with each sub-cycle. A Kc curve con-
structed for alfalfa grown for hay in southern Idaho is illustrated in Figure 8.4. Cut-
tings may create a ground surface with less than 10% vegetation cover. Cutting cycle 1
may have longer duration than cycles 2, 3, and 4 if low air and soil temperatures or
short daylength during this period moderate the crop growth rate. Frosts terminate the
growing season in southern Idaho sometime in the fall, usually in early to mid-October
(day of year 280 to 300, see footnote 4 of Table 8.3). Magnitudes of Kc values during
midseason periods for each cutting cycle change from cycle to cycle as a result of ad-
justing Kc mid and Kc end for each period using Equation 8.59. Basal Kcb curves for for-
age or hay crops can be constructed similar to Figure 8.4.
8.5.3.4 Mean Kc for the initial stage (annual crops). ET during the initial stage for
annual crops is predominantly in the form of evaporation from soil. Accurate estimates
for the time-averaged Kc ini for the mean Kc must consider the frequency that the soil
surface is wetted. Kc mid and Kc end are less affected by wetting frequency since vegetation
during these periods is generally near full ground cover so that effects of surface evapo-
ration are smaller. Figures 8.5a-8.5c from FAO-56 estimate Kc ini as a function of ETo,
soil type, and wetting frequency. Equations for these curves are given in Allen et al.,
(1998, 2005b). Figure 8.5a is used for all soil types when wetting events (precipitation
and irrigation) are light (i.e., infiltrated depths average about 10 mm per wetting event),
Figure 8.5b is used for “heavy” wetting events, where infiltrated depths are greater than
1.5
Alfalfa Hay Kimberly, Idaho
First Cutting
Cycle
1.0
Kc Second Third Fourth
Cutting Cutting Cutting
Cycle Cycle Cycle
0.5
Frost
L ini L dev L mid L late Cuttings
0.0
90 120 150 180 210 240 270 300
Day of the Year
Figure 8.4. Crop coefficient curve for alfalfa hay crop in southern Idaho having four cuttings.
30 to 40 mm, on coarse-textured soils3 and Figure 8.5c is used for heavy wetting events
on fine and medium-textured soils. In general, the mean time interval is estimated by
counting all rainfall and irrigation events occurring during the initial period that are
greater than a few mm. Wetting events occurring on adjacent days are typically counted
as one event. When average infiltration depths are between 10 and 40 mm, the value for
Kc ini can be interpolated between Figure 8.5a and Figure 8.5b or Figure 8.5c.
8.5.4 The Dual Kc Method: Incorporating Specific Wet Soil Effects
The dual Kc method introduced in this section, based on FAO-56, is applicable to
both Kc based on ETo (Kco) and Kc based on ETr (Kcr), with differences only in the
value for Kc max. The Ke component describes the evaporation component of ETc. Be-
cause the dual Kc method incorporates the effects of specific wetting patterns and fre-
quencies that may be unique to a single field, this method can provide more accurate
estimates of evaporation components and total ET on an individual field basis.
When the soil surface layer is wet, following rain or irrigation, Ke is at a maximum,
and when the soil surface layer is dry, Ke is small, even zero. When the soil is wet,
evaporation occurs at some maximum rate and Kcb + Ke is limited by a maximum
value Kc max:
K e = K r ( K c max − K cb ) ≤ f ew K c max (8.60)
where Kc max is the maximum value of Kc following rain or irrigation, Kr is a dimen-
sionless evaporation reduction coefficient (defined later) and is dependent on the cu-
mulative depth of water depleted (evaporated), and few is the fraction of the soil that is
both exposed to solar radiation and that is wetted. The evaporation rate is restricted by
the estimated amount of energy available at the exposed soil fraction, i.e., Ke cannot
exceed few Kc max. Kc max for the ETo basis (Kc max o) ranges from about 1.05 to 1.30:
⎛⎧ ⎞
⎛ h ⎞ ⎫⎪
0.3
⎪
K c max o = max⎜ ⎨1.2 + [0.04(u2 − 2) − 0.004 ( RH min − 45) ]⎜ ⎟ ⎬, K cb o + 0.05 { }⎟⎟ (8.61a)
⎜ ⎝ 3 ⎠ ⎪⎭
⎝ ⎪⎩ ⎠
3
Coarse-textured soils include sands and loamy sand textured soils. Medium-textured soils include sandy loam,
loam, silt loam, and silt textured soils. Fine-textured soils include silty clay loam, silty clay, and clay textured
soils.
Table 8.4. Typical crop height, ranges of maximum effective rooting depth (Zr), and
soil water depletion fraction for no stress (p), for common crops (from FAO-56).
Depletion
Maximum Crop Maximum Root Fraction[2] (p) for
Crop Height (h) (m) Depth[1] (m) ETc = 5 mm d-1
a. Small Vegetables
Broccoli 0.3 0.4-0.6 0.45
Brussels sprouts 0.4 0.4-0.6 0.45
Cabbage 0.4 0.5-0.8 0.45
Carrots 0.3 0.5-1.0 0.35
Cauliflower 0.4 0.4-0.7 0.45
Celery 0.6 0.3-0.5 0.20
Garlic 0.3 0.3-0.5 0.30
Lettuce 0.3 0.3-0.5 0.30
Onions, dry 0.4 0.3-0.6 0.30
green 0.3 0.3-0.6 0.30
seed 0.5 0.3-0.6 0.35
Spinach 0.3 0.3-0.5 0.20
Radishes 0.3 0.3-0.5 0.30
b. Vegetables, Solanum Family (Solanaceae)
Eggplant 0.8 0.7-1.2 0.45
Sweet peppers (bell) 0.7 0.5-1.0 0.30
Tomato 0.6 0.7-1.5 0.40
c. Vegetables, Cucumber Family (Cucurbitaceae)
Cantaloupe 0.3 0.9-1.5 0.45
Cucumber, fresh market 0.3 0.7-1.2 0.50
machine harvest 0.3 0.7-1.2 0.50
Pumpkin, winter squash 0.4 1.0-1.5 0.35
Squash, zucchini 0.3 0.6-1.0 0.50
Sweet melons 0.4 0.8-1.5 0.40
Watermelon 0.4 0.8-1.5 0.40
d. Roots and Tubers
Beets, table 0.4 0.6-1.0 0.50
Cassava, year 1 1.0 0.5-0.8 0.35
year 2 1.5 0.7-1.0 0.40
Parsnip 0.4 0.5-1.0 0.40
Potato 0.6 0.4-0.6 0.35
Sweet potato 0.4 1.0-1.5 0.65
Turnip and rutabaga 0.6 0.5-1.0 0.50
Sugar beet 0.5 0.7-1.2 0.55[3]
[1]
The larger values for Zr are for soils having no significant layering or other characteristics that can re-
strict rooting depth. The smaller values for Zr may be used for irrigation scheduling and the larger values
for modeling soil water stress or for rainfed conditions.
[2]
The tabled values for p apply for ETc ≈ 5 mm/day. The value for p can be adjusted for different ETc
according to p = pTable 8.4 + 0.04 (5 – ETc) where p is expressed as a fraction and ETc as mm/day.
[3]
Sugar beets often experience late afternoon wilting in arid climates even at p < 0.55, with usually only
minor impact on sugar yield.
(continued)

Depletion
Maximum Crop Maximum Root Fraction[2] (p)
Crop Height (h) (m) Depth[1] (m) for ETc = 5 mm d-1
e. Legumes (Leguminosae)
Beans, green 0.4 0.5-0.7 0.45
Beans, dry and pulses 0.4 0.6-0.9 0.45
Beans, lima, large vines 0.4 0.8-1.2 0.45
Chick pea 0.4 0.6-1.0 0.50
Faba bean (broad bean), fresh 0.8 0.5-0.7 0.45
dry/seed 0.8 0.5-0.7 0.45
Garbanzo 0.8 0.6-1.0 0.45
Green gram and cowpeas 0.4 0.6-1.0 0.45
Groundnut (peanut) 0.4 0.5-1.0 0.50
Lentil 0.5 0.6-0.8 0.50
Peas, fresh 0.5 0.6-1.0 0.35
dry/seed 0.5 0.6-1.0 0.40
Soybeans 0.5-1.0 0.6-1.3 0.50
f. Perennial Vegetables (with winter dormancy and initially bare or mulched soil)
Artichokes 0.7 0.6-0.9 0.45
Asparagus 0.2-0.8 1.2-1.8 0.45
Mint 0.6-0.8 0.4-0.8 0.40
Strawberries 0.2 0.2-0.3 0.20
g. Fiber Crops
Cotton 1.2-1.5 1.0-1.7 0.65
Flax 1.2 1.0-1.5 0.50
Sisal 1.5 0.5-1.0 0.80
h. Oil Crops
Castor bean (Ricinus) 0.3 1.0-2.0 0.50
Rapeseed, canola 0.6 1.0-1.5 0.60
Safflower 0.8 1.0-2.0 0.60
Sesame 1.0 1.0-1.5 0.60
Sunflower 2.0 0.8-1.5 0.45
i. Cereals
Barley 1 1.0-1.5 0.55
Oats 1 1.0-1.5 0.55
Spring wheat 1 1.0-1.5 0.55
Winter wheat 1 1.5-1.8 0.55
Maize, field (grain) (field corn) 2 1.0-1.7 0.55
Maize, sweet (sweet corn) 1.5 0.8-1.2 0.50
Millet 1.5 1.0-2.0 0.55
Sorghum, grain 1-2 1.0-2.0 0.55
sweet 2-4 1.0-2.0 0.50
Rice 1 0.5-1.0 0.20[4]
[4]
The value for p for rice is 0.20 of saturation.
(continued)

Depletion
Maximum Crop Maximum Root Fraction[2] (p)
Crop Height (h) (m) Depth[1] (m) for ETc = 5 mm d-1
j. Forages
Alfalfa, for hay 0.7 1.0-2.0 0.55
for seed 0.7 1.0-3.0 0.60
Bermuda, for hay 0.35 1.0-1.5 0.55
spring crop for seed 0.4 1.0-1.5 0.60
Clover hay, berseem 0.6 0.6-0.9 0.50
Ryegrass hay 0.3 0.6-1.0 0.60
Sudan grass hay (annual) 1.2 1.0-1.5 0.55
Grazing pasture - rotated grazing 0.15-0.30 0.5-1.5 0.60
extensive grazing 0.10 0.5-1.5 0.60
Turf grass, cool season[5] 0.10 0.5-1.0 0.40
warm season[5] 0.10 0.5-1.0 0.50
k. Sugar Cane 3 1.2-2.0 0.65
Banana, 1st year 3 0.5-0.9 0.35
2nd year 4 0.5-0.9 0.35
Cacao 3 0.7-1.0 0.30
Coffee 2-3 0.9-1.5 0.40
Date palms 8 1.5-2.5 0.50
Palm trees 8 0.7-1.1 0.65
Pineapple 0.6-1.2 0.3-0.6 0.50
Rubber trees 10 1.0-1.5 0.40
Tea, non-shaded 1.5 0.9-1.5 0.40
shaded 2 0.9-1.5 0.45
m. Grapes and Berries
Berries (bushes) 1.5 0.6-1.2 0.50
Grapes, table or raisin 2 1.0-2.0 0.35
wine 1.5-2 1.0-2.0 0.45
Hops 5 1.0-1.2 0.50
n. Fruit Trees
Almonds 5 1.0-2.0 0.40
Apples, cherries, pears 4 1.0-2.0 0.50
Apricots, peaches, stone fruit 3 1.0-2.0 0.50
Avocado 3 0.5-1.0 0.70
Citrus
70% canopy 4 1.2-1.5 0.50
50% canopy 3 1.1-1.5 0.50
20% canopy 2 0.8-1.1 0.50
Conifer trees 10 1.0-1.5 0.70
Kiwi 3 0.7-1.3 0.35
Olives (40% to 60% ground 3-5 1.2-1.7 0.65
coverage by canopy)
Pistachios 3-5 1.0-1.5 0.40
Walnut orchard 4-5 1.7-2.4 0.50
[5]
Cool season grass varieties include bluegrass, ryegrass, and fescue. Warm season varieties include ber-
muda grass, buffalo grass, and St. Augustine grass. Grasses are variable in rooting depth. Some root be-
low 1.2 m while others have shallow rooting depths. The deeper rooting depths for grasses represent
conditions where careful water management is practiced with higher depletion between irrigations to
encourage the deeper root exploration.
1.2
small infiltration depths
~ 10 mm
1.0
0.8 1 da
y
K c ini
0.6
2 day
0.4
10
day 4 day
0.2 20 da 7 day
y
(a) 0.0
0 1 2 3 4 5 6 7 8 9 10 11 12
ETo mm/day
low moderate high very high
1.2
large infiltration depths
> 40 mm
1.0
1 day
0.8
2 day
K c ini
0.6
0.4 4 day
10 da 7 day
y
0.2 20 day
a. coarse textures
(b) 0.0
0 1 2 3 4 5 6 7 8 9 10 11 12
ETo mm/day
1.2
1 day
large infiltration depths
1.0 > 40 mm
2 day
0.8
K c ini
4 day
0.6
7 day
0.4
10 day
0.2 20 day
b. fine and medium textures
0.0
(c) 0 1 2 3 4 5 6 7 8 9 10 11 12
ETo mm/day
Figure 8.5. Average Kc ini for the initial crop development stage as related to the level of
ETo and the interval between irrigations and/or significant rain during the initial period
for (a) all soil types when wetting events are light (about 10 mm per event); (b) coarse-
textured soils when wetting events are greater than about 40 mm; and (c) medium and
fine-textured soils when wetting events are greater than about 40 mm (from FAO-56);
Equations for these curves are given in Allen et al. (2005b).
Table 8.5. Typical soil water characteristics for different soil types (from FAO-56).
Depth of Water That Can
Soil Type Be Depleted by Evaporation
(USDA Soil Stages 1 and 2 Stages 1 and 2
Texture Soil Water Characteristics Stage 1 TEW[1] TEW[1]
Classifica- θFC θWP (θFC – θWP) REW (Ze = 0.10 m) (Ze = 0.15 m)
tion) m3 m-3 m3 m-3 m3 m-3 mm mm mm
Sand 0.07-0.17 0.02-0.07 0.05-0.11 2-7 612 9-13
Loamy sand 0.11-0.19 0.03-0.10 0.06-0.12 4-8 9-14 13-21
Sandy loam 0.18-0.28 0.06-0.16 0.11-0.15 6-10 15-20 22-30
Loam 0.20-0.30 0.07-0.17 0.13-0.18 8-10 16-22 24-33
Silt loam 0.22-0.36 0.09-0.21 0.13-0.19 8-11 18-25 27-37
Silt 0.28-0.36 0.12-0.22 0.16-0.20 8-11 22-26 33-39
Silt clay loam 0.30-0.37 0.17-0.24 0.13-0.18 8-11 22-27 33-40
Silty clay 0.30-0.42 0.17-0.29 0.13-0.19 8-12 22-28 33-42
Clay 0.32-0.40 0.20-0.24 0.12-0.20 8-12 22-29 33-43
[1]
TEW = (θFC – 0.5 θWP) Ze
where h is the mean plant height (in m) during the period of calculation (initial, devel-
opment, midseason, or late-season), and the max ( ) function indicates the selection of
the maximum of values separated by the comma. Kc max for the tall reference ETr
(Kc max r) does not require adjustment for climate, due to the greater roughness of the
reference basis:
Kc max r = max (1.0, {Kcbr + 0.05}) (8.61b)
Kcbo and Kcbr are the basal Kc for the ETo and ETr bases. Equation 8.61 ensures that
Kc max is always greater than or equal to the sum Kcb + 0.05, suggesting that wet soil
increases the Kc value above Kcb by 0.05 following complete wetting of the soil sur-
face, even during periods of full ground cover. Equation 8.56 and Equations 8.60 to
8.69 can be applied with both the straight-line Kcb curve style of FAO and with curvi-
linear Kcb curves such as by Wright (1982), as illustrated later in this section.
The surface soil layer is presumed to dry to an air-dry water content approximated
as halfway between wilting point, θ WP, and oven dry. The amount of water that can be
removed by evaporation during a complete drying cycle is estimated as:
TEW = 1000 ( θ FC − 0.5 θWP ) Z e (8.62)
where TEW (total evaporable water) is the maximum depth of water that can be evapo-
rated from the surface soil layer when the layer has been initially completely wetted
(in mm). Field capacity, θ FC, and θ WP are expressed in m3 m-3 and Ze is the effective
depth (m) of the surface soil subject to drying to 0.5 θ WP by way of evaporation. Typi-
cal values for θ FC, θ WP, and TEW are given in Table 8.5 for a range in soil types. Ze is
an empirical value based on observation. Some evaporation or soil drying will be ob-
served to occur below the Ze depth. FAO-56 recommended values for Ze of 0.10 to
0.15 m, with 0.1 m recommended for coarse soils and 0.15 m recommended for fine-
textured soils.
The Kr coefficient of Equation (8.60) is calculated, assuming a 2-stage drying process:
K r = 1.0 for De, j -1 ≤ REW (8.63a)
TEW − De, j -1
Kr = for De, j -1 > REW (8.63b)
TEW − REW
where De,j-1 is cumulative depletion from the soil surface layer at the end of day j-1
(the previous day) (in mm), REW is the readily evaporable water that is evaporated
during stage 1, and TEW and REW are in mm (REW < TEW). Evaporation from the
soil beneath the crop canopy, occurring at a slower rate, is assumed included in the
basal Kcb coefficient.
In the FAO-56 model, the term fw is defined as the fraction of the surface wetted by
irrigation and/or precipitation. This term defines the potential spatial extent of evapo-
ration. Common values for fw are listed in Table 8.6 and illustrated in Figure 8.6.
When the soil surface is completely wetted, as by precipitation or sprinkler, few of
Equation 8.60 is set equal to (1 – fc), where fc is the fraction of soil surface effectively
covered by vegetation. For irrigation systems where only a fraction of the ground sur-
face is wetted, few is limited to fw:
f ew = min(1 − f c , f w ) (8.64)
Both 1 – fc and fw, for numerical stability, have limits of 0.01 to 1. In the case of
drip irrigation, Allen et al. (1998) suggest that where the majority of soil wetted by
irrigation is beneath the crop canopy and is shaded, fw be reduced to about one-half to
one-third of that given in Table 8.6. Their general recommendation for drip irrigation
is to multiply fw by [1 – (2/3) fc]. Pruitt et al. (1984) and Bonachela et al. (2001) have
described evaporation patterns and extent under drip irrigation.
The value for fc is limited to < 0.99 for numerical stability and is generally deter-
mined by visual observation. For estimating few, fc can be estimated from Kcb as:
(1 + 0.5 h )
⎛ K cb − K c min ⎞
fc = ⎜ ⎟ (8.65)
⎜ K c max − K c min ⎟
⎝ ⎠
where fc is limited to 0 to 0.99 and Kc min is the minimum Kc for dry bare soil with no
ground cover and h is crop height in meters. The difference Kcb – Kc min is limited to >
0.01 for numerical stability. The value for fc will change daily as Kcb changes. Kc min
ordinarily has the same value as Kcb ini used for annual crops under nearly bare soil
conditions (i.e., Kc min ~ 0.15). However, Kc min is set to 0 or nearly zero under condi-
tions with large time periods between wetting events, for example in applications with
natural vegetation in deserts. The value for fc decreases during the late-season period
in proportion to Kcb to account for local transport of sensible heat from senescing
leaves to the soil surface.
Table 8.6. Common values for the fraction of soil surface

wetted by irrigation or precipitation (after FAO-56).
Wetting Event fw
Precipitation 1.0
Sprinkler irrigation, field crops 1.0
Sprinkler irrigation, orchards 0.7 to 1.0
Basin irrigation 1.0
Border irrigation 1.0
Furrow irrigation (every furrow), narrow bed 0.6 to 1.0
Furrow irrigation (every furrow), wide bed 0.4 to 0.6
Furrow irrigation (alternated furrows) 0.3 to 0.5
Microspray irrigation, orchards 0.5 to 0.8
Trickle (drip) irrigation 0.3 to 0.4
initial mid and late

crop development season
1 - fc fc 1 - fc fc
Rain
Basin
Border
Sprinkler
fw = 1 fw = 1
Irrigation
1 - fc fc ewf = (1 - f c1)- fc fc
Furrow
Irrigation
fw fw
few = f w few = (1 - f c)
1 - fc fc 1 - fc fc
Drip
Irrigation
fw fw
few = 1.0 ... 0.3 f w
Figure 8.6. Crop Determination of few (grayed areas) as a function of the fraction of
ground surface coverage (fc) and the fraction of the surface wetted (fw) (from FAO-56).
8.5.4.1 Water balance of the soil surface layer. Estimation of Ke requires a daily
water balance for the few fraction of the surface soil layer. The daily soil water balance
equation is:
Ij Ej
De, j = De, j -1 − ( Pj − RO j ) − + + Te, j + DPe, j (8.66)
fw f ew
where De,j-1 and De,j = cumulative depletion depth at the ends of days j–1 and j, mm
Pj = precipitation on day j, mm
ROj = precipitation runoff from the soil surface on day j, mm
Ij = irrigation depth on day j that infiltrates the soil, mm
Ej = evaporation on day j (i.e., Ej = Ke ETref), mm
Te, j = depth of transpiration from the exposed and wetted fraction of the soil
surface layer on day j, mm
DPe j = deep percolation from the few fraction of the soil surface layer on day j
if soil water content exceeds field capacity, mm.
Assuming that the surface layer is at field capacity following heavy rain or irrigation,
the minimum value for De,j is zero. The limits imposed on De,j are consequently 0 <
De,j < TEW. It is recognized that water content of the soil surface layer can exceed
TEW for short periods of time while drainage is occurring. However, because the
length of time that this occurs varies with soil texture, wetting depth, and tillage, De,j >
0 is assumed. Additionally, it is recognized that some drainage in soil occurs at very
small rates at water contents below field capacity. To some extent, impacts of these
simple assumptions can be compensated for, if needed, in setting the value for Ze or
TEW. The irrigation depth Ij is divided by fw to approximate the infiltration depth to
the fw portion of the soil surface. Similarly, Ej is divided by few because it is assumed
that all Ej (other than residual evaporation implicit to the Kcb coefficient) is taken from
the few fraction of the surface layer.
8.5.4.2 Transpiration from the surface layer. The amount of transpiration ex-
tracted from the few fraction of the evaporating soil layer is generally a small fraction
of total transpiration. However, for shallow-rooted annual crops where the depth of the
maximum rooting is less than about 0.5 m, Te may have significant effect on the water
balance of the surface layer and therefore on estimation of the evaporation component,
especially for the period midway through the development period. The following ex-
tension to FAO-56 by Allen et al. (2005a) estimates Te from the few fraction of the
evaporation layer in proportion to the water content of that layer:
Te = K t K s K cb ETref (8.67)
where Kt, having a range of 0 to 1, is the proportion of basal ET (= Kcb ETref) extracted
as transpiration from the few fraction of the surface soil layer. Ks is the soil water stress
factor computed for the root zone (range of 0 to 1). Kt is determined by comparing
relative water availability in the Ze and Zr layers along with the presumed rooting dis-
tribution. For the few fraction:
⎛ D ⎞
⎜ 1 − e ⎟ ⎛ Z ⎞ 0.6
Kt = ⎜ TEW ⎟ ⎜ e ⎟ (8.68)
⎜ 1 − Dr ⎟ ⎜⎝ Z r ⎟⎠
⎜ ⎟
⎝ TAW ⎠
where the numerator and denominator of the first expression are limited to > 0.001
and the value for Kt is limited to < 1.0. TAW is the total available water (mm) in the
root zone and Dr is the current depletion from the effective root zone.
In the simple water balance procedure of FAO-56, it is assumed that the soil water
content is limited to < θ FC on the day of a complete wetting event. This is a reason-
able assumption considering the shallowness of the surface layer. Downward drainage
(percolation) of water from the surface layer is calculated as:
Ij
DPe, j = ( Pj − RO j ) + − De, j -1 ≥0 (8.69)
fw
8.5.4.3 Initialization of the water balance and order of calculation. To initiate
the water balance for the evaporating layer, the user can assume that the soil surface
layer is near θ FC following a heavy rain or irrigation so that De,j-1 = 0. Where a long
period of time has elapsed since the last wetting, the user can assume that all evapor-
able water has been depleted from the evaporation layer at the beginning of calcula-
tions so that De, j-1 = TEW = 1000 (θ FC – 0.5 θWP) Ze. Calculations for the dual Kcb +
Ke procedure, for example when using a spreadsheet, proceed in the following order:
Kcb, h, Kc max, fc, fw, few, Kr, Ke, E, DPe, De, I, Kc, and ETc.
8.5.4.4 Impacts of water stress. The Ks component in Equation 8.56 and 8.58 is
the water stress coefficient used to reduce Kc or Kcb under conditions of water stress or
salinity stress. FAO-56 describes the application of the salinity stress function for re-
ducing ET. The water stress function is described here. Mean water content of the root
zone is expressed as root zone depletion, Dr, defined as water shortage relative to field
capacity. At field capacity, Dr = 0. The degree of stress is presumed to progressively
increase as Dr increases past RAW, the depth of readily available water in the root
zone. For Dr > RAW, Ks is:
TAW − Dr TAW − Dr
Ks = = (8.70)
TAW − RAW (1 − p ) TAW
where TAW is total available soil water in the root zone (mm), and p is the fraction of
TAW that a crop can extract from the root zone without suffering water stress. When
Dr < RAW, Ks = 1. Values for p are listed in Table 8.4. The total available water in the
root zone is estimated as the difference between the water content at field capacity and
wilting point:
TAW = 1000 (θ FC – θ WP) Zr (8.71)
where Zr is the effective rooting depth (m) and Zr contains Ze. RAW is estimated as:
RAW = p TAW (8.72)
where RAW has units of TAW (mm). Table 8.4 contains typical maximum values for
Zr. FAO-56 describes several means to estimate the development (increase) in Zr with
time for annual crops including in proportion to development of Kcb and in proportion
to time. Other methods for Zr development include a sine function of time (Borg and
Grimes, 1986), an exponential function of time dampened by soil temperature and soil
moisture (Danuso et al., 1995) and a full root-growth simulation model by Jones et al.
(1991).
8.5.5 Conditions for Maximum Transpiration
The Kc values in Table 8.2 represent potential water consumption by healthy, rela-
tively disease-free, and densely planted stands of vegetation having adequate levels of
soil water. When stand density, height, or leaf area are significantly less than that at-
tained under ideal or normal (pristine) conditions, the value for Kc may be reduced.
Low stand density, height, and leaf area are caused by disease, low soil fertility, high
soil salinity, waterlogging or water shortage (moisture stress), or by poor stand estab-
lishment. The reduction in the value for Kc during the midseason for poor crop stands
can be as much as 0.3 to 0.5 for extremely poor crop stands and can be approximated
according to the amount of effective (green) leaf area relative to that for healthy vegeta-
tion having normal planting densities. Procedures for reducing Kc according to the re-
duction in leaf area and the fraction of ground cover are given in Chapter 9 of FAO-56.
8.5.6 Application of the Basal Kcb Procedure over a Growing Season
The first step in applying the basal Kcb approach is to construct the Kcb curve using
Kcb ini, Kcb mid, and Kcb end similar to constructing the Kc curve. Equations for computing
Ke (and Ks if necessary) are applied on a daily calculation timestep where daily Kcb is
interpolated from the constructed Kcb curve. Figure 8.7 is an illustration of applying
the Kcb + Ke procedure for a snap bean crop harvested for dry seed. The measured ETc
1.6 300
Snap Beans
Precipitation, net irrigation, mm

FAO-56 PM ETo
FAO-56 Kcbo
1.4 FAO-56 Ke
250
1.2
200
1
Kcb, Kc
0.8 150
0.6
100
0.4 I I
I I I I
50
0.2
P I
PP P P PP
0 0
135 150 165 180 195 210 225 240 255
Day of Year, 1974
Basal Kcb Ks Kcb + Ke Kc from Lys. P = precipitation event I = irrigation
Figure 8.7. Measured (circles) and estimated (thin line) daily crop coefficients
for a snap bean crop at Kimberly, Idaho. The basal crop curve (Kcb)
was derived from Kcb values given in Table 8.2. (Data from Wright, 1990.)
data are from a precision lysimeter system at Kimberly, Idaho (Wright, 1990; Vander-
kimpen, 1991). The soil at Kimberly had a silt loam texture. Soil evaporation parame-
ters were Ze = 0.15 m, TEW = 34 mm, and REW = 8 mm. Nearly all wetting events
were from alternate-row furrow irrigation so that the value for fw was set to 0.5. Irriga-
tion events occurred at about midday or during early afternoon. The agreement be-
tween the estimated values for daily Kc from Equation 8.56 (thin, continuous line) and
actual 24-hour measurements (symbols) is relatively good.
8.5.7 Alfalfa-Based Crop Coefficients
Wright (1982, 1995) defined crop coefficients for crops common to higher eleva-
tions and latitudes of the western U.S. These coefficients, derived at Kimberly, Idaho,
were initially based on the alfalfa reference ETr represented by the 1982 Kimberly
Penman Equation (Equations 8.4–8.6). Wright developed both mean and basal Kc,
denoted here as Kcr and Kcbr to distinguish them from the grass based Kc described
earlier. These coefficients, if converted to an ETo basis by multiplying by the ratio of ETr
to ETo and simplified to linear segment forms, agree closely with the Kco from Table 8.2
for similar crops. The Kimberly coefficients were converted from the 1982 Kimberly
Penman ETr basis to the ASCE standardized Penman-Monteith ETrs basis by Allen and
Wright (2006) and are listed in Tables 8.7 and 8.8. The Kc curves of Wright (1981, 1982)
were changed little by the conversion due to the similarity in the two ETr references.
Coefficient curves by Wright are based on the relative time from planting to effec-
tive full cover and on days after effective full cover. Effective full cover was defined
by Wright (1982) as the time when leaves between row crops begin to interlock, at
heading of grain, and flowering of peas. The Kc curves are constructed from the data
in Tables 8.7 and 8.8 by using linear or curvilinear interpolation between adjacent
columns or by fitting regression equations to the first half (planting to effective full
cover) and to the second half (days after effective full cover) for each curve. Table 8.9
lists strategic dates of crop development recorded by Wright (1982).
Table 8.7. Alfalfa-based mean (single) crop coefficients (Kcr) by Wright (1981, 1995)
converted for use with the ASCE standardized PM equation (Allen and Wright, 2006).
Percent Time from Planting to Effective Full Cover Date
Crop 0 10 20 30 40 50 60 70 80 90 100
Spring grain[1] 0.20[2] 0.20 0.20 0.25 0.37 0.50 0.63 0.76 1.00 1.03 1.03
Peas 0.15 0.17 0.19 0.21 0.32 0.42 0.52 0.63 0.73 0.83 0.93
Sugar beets 0.26 0.26 0.26 0.26 0.26 0.28 0.30 0.38 0.55 0.74 1.03
Potatoes 0.20 0.20 0.20 0.22 0.30 0.41 0.53 0.67 0.73 0.77 0.8
Corn 0.20 0.20 0.20 0.20 0.24 0.34 0.44 0.58 0.72 0.90 1.00
Beans 0.20 0.20 0.22 0.26 0.35 0.45 0.55 0.68 0.83 0.95 0.97
Winter wheat 0.25 0.25 0.27 0.38 0.60 0.80 0.90 0.96 1.00 1.03 1.03
Days after Effective Full Cover Date
Crop 0 10 20 30 40 50 60 70 80 90 100
Spring grain[1] 1.03 1.03 1.03 1.03 0.94 0.50 0.30 0.15 0.10
Peas 0.93 0.93 0.70 0.54 0.38 0.22 0.12 0.10
Sugar beets 1.03 1.03 1.03 1.00 0.97 0.92 0.82 0.74 0.65 0.61 0.56
Potatoes 0.80 0.80 0.76 0.72 0.68 0.63 0.58 0.50 0.38 0.20 0.15
Field corn 1.00 0.99 0.98 0.95 0.88 0.80 0.72 0.63 0.35 0.18
Sweet corn 1.00 0.97 0.94 0.90 0.84 0.70 0.55 0.35 0.20 0.10
Beans 0.97 0.97 0.94 0.64 0.32 0.15 0.10 0.05
Winter wheat 1.03 1.03 1.03 1.03 1.00 0.55 0.25 0.15 0.10
Alfalfa Hay: Percent Time from New Growth to Harvest and
from Harvest to Harvest
Cycle 0 10 20 30 40 50 60 70 80 90 100
First[3] 0.50 0.62 0.73 0.83 0.88 0.94 1.00 1.00 1.00 0.98 0.96
Intermediate 0.30 0.40 0.55 0.80 0.94 0.97 1.00 1.00 1.00 0.97 0.94
Last 0.30 0.35 0.45 0.53 0.58 0.58 0.54 0.48 0.46 0.44 0.44
Alfalfa Hay and Ryegrass: Total Season (including all cutting effects) (days)
Crop 0 20 40 60 80 100 120 140 160 180 200
Alfalfa[4] 0.45 0.69 0.87 0.88 0.70 0.75 0.88 0.81 0.88 0.71 0.65
Alfalfa[5] 0.44 0.77 0.82 0.86 0.90 0.88 0.85 0.82 0.78 0.66 0.50
Alfalfa, overall
seasonal mean 0.85
Perennial ryegrass 0.55 0.66 0.77 0.8 0.8 0.8 0.78 0.76 0.72 0.68 0.55
[1]
Spring grain includes wheat and barley.
[2]
The initial values 0.15 to 0.26 list for all crops are appropriate for relatively dry surface soil conditions
from planting until significant crop development. For moderately wet surface soil, as with pre-
emergence irrigation(s) or some precipitation, use 0.35, and for very wet conditions use 0.50.
[3]
First denotes first growth cycle. Intermediate growth cycles (following first cutting) may number one or
more, depending on length of season. The last harvest terminates when crop becomes dormant in freez-
ing weather. Alfalfa cultivar was Ranger.
[4]
Each value represents the average Kc over the 20-day period at Kimberly, Idaho. The first value repre-
sents from 0 to 10 days and the last value from 190 to 200 days.
[5]
This is a seasonal curve for alfalfa cut periodically for hay where the values are from a smoothed, run-
ning average of Kc.
Table 8.8. Alfalfa-based basal crop coefficients (Kcbr) by Wright (1982, 1995), converted
for application with the ASCE standardized PM equation (Allen and Wright, 2006).
Percent Time from Planting to Effective Full Cover Date
Crop 0 10 20 30 40 50 60 70 80 90 100
Spring grain[1] 0.15 0.15 0.15 0.19 0.24 0.36 0.48 0.62 0.92 0.98 1.03
Peas 0.12 0.13 0.14 0.15 0.18 0.27 0.36 0.50 0.65 0.78 0.92
Sugar beets 0.15 0.15 0.15 0.15 0.15 0.16 0.17 0.21 0.40 0.66 1.03
Potatoes 0.15 0.15 0.15 0.15 0.15 0.20 0.34 0.49 0.64 0.72 0.77
Corn 0.15 0.15 0.15 0.16 0.17 0.20 0.27 0.41 0.55 0.80 0.96
Beans 0.15 0.15 0.17 0.19 0.23 0.35 0.46 0.60 0.78 0.93 0.95
Winter wheat 0.12 0.12 0.14 0.22 0.45 0.70 0.84 0.96 1.00 1.03 1.03
Days after Effective Full Cover Date
Crop 0 10 20 30 40 50 60 70 80 90 100
Spring grain[1] 1.03 1.03 1.03 1.03 0.94 0.40 0.15 0.07 0.05
Peas 0.92 0.92 0.72 0.52 0.32 0.16 0.07 0.05
Sugar beets 1.03 1.03 1.02 0.98 0.93 0.86 0.78 0.72 0.66 0.60 0.54
Potatoes 0.77 0.77 0.73 0.68 0.64 0.59 0.54 0.47 0.20 0.08 0.08
Field corn 0.96 0.96 0.96 0.92 0.85 0.79 0.72 0.62 0.28 0.16 0.12
Sweet corn 0.96 0.95 0.93 0.88 0.80 0.65 0.47 0.23 0.12
Beans 0.95 0.95 0.88 0.64 0.30 0.09 0.05
Winter wheat 1.03 1.03 1.03 1.03 1.00 0.50 0.20 0.10 0.05
Alfalfa Hay: Percent Time from New Growth to Harvest and
from Harvest to Harvest
Cycle 0 10 20 30 40 50 60 70 80 90 100
First[2] 0.35 0.45 0.56 0.72 0.82 0.90 1.00 1.00 1.00 0.98 0.96
Intermediate 0.25 0.30 0.42 0.72 0.90 0.95 1.00 1.00 0.98 0.96 0.94
Last 0.25 0.27 0.36 0.42 0.50 0.45 0.35 0.30 0.25 0.22 0.22
[1]
Spring grain includes wheat and barley.
[2]
First denotes first growth cycle. Intermediate growth cycles (following first cutting) may number one or
more, depending on length of season. The last harvest terminates when crop becomes dormant in freez-
ing weather. Alfalfa cultivar was Ranger.
Table 8.9. Dates (month/day) for crop growth stages at Kimberly, Idaho used
in development of crop coefficient curves (after Wright, 1982).
Rapid Full Head/
Crop Planting Emergence Growth Cover Bloom Ripening Harvest
Spring grain 4/01 4/15 5/10 6/10 6/10 7/20 8/10
Peas 4/05 4/25 5/10 6/05 6/15 7/05 7/25
Sugar beets 4/15 5/10 6/01 7/10 10/15
Potatoes 4/25 5/25 6/10 7/10 7/01 9/20 10/10
Field corn 5/05 5/25 6/10 7/15 7/30 9/10 9/20
Sweet corn 5/05 5/25 6/10 7/15 7/20 8/15
Beans 5/22 6/05 6/15 7/15 7/05 8/15 8/30
Winter wheat (2/15)[1] (3/01) [1] 3/20 6/05 6/05 7/15 8/10
Alfalfa
First cutting 4/01 4/20 6/15
Second cutting 6/15 6/25 7/31
Third cutting 7/31 8/10 9/15
Fourth cutting 9/15 10/01 10/30
[1]
For winter wheat, the first two dates are “effective dates” for green-up in the spring. Actual dates were
10/10 for planting and 10/25 for emergence, the previous year.
1.4 300
Snap Beans
Precipitation, net irrigation, mm

ASCE-PM ETr
Wright (1982) Kcbr
1.2 FAO-56 Ke
250
1
200
0.8
Kcb, Kc
150
0.6
100
0.4
I I
I I I I
0.2 50
I
PP PP P PP
0 0
135 150 165 180 195 210 225 240
Day of Year, 1974
Basal Kcb Ks Kcb + Ke Kc from Lys. P = precipitation event I = irrigation
Figure 8.8. Measured (circles) and estimated (thin line) daily crop coefficients
for a snap bean crop at Kimberly, Idaho. The basal crop curve (Kcb)
was derived from Kcbr values given in Table 8.8. (Data from Wright, 1990.)
8.5.8 Comparisons between FAO and Wright Kc Methods.

Estimates of daily ETc by the linearized FAO-56 basal Kcb + Ke method and ETos
basis for a snap bean crop at Kimberly, Idaho, were shown in Figure 8.7 along with
measured daily Kc. The same measurement data are plotted in Figure 8.8, but using the
Wright (1982) curvilinear basal Kcb data from Table 8.8, instead, for snap beans, and
ETrs by the ASCE PM ETr equation instead of ETos. Ke was calculated in both applica-
tions using the FAO-56 method (Equations 8.60 to 8.68), with the only difference be-
ing in the calculation for Kc max (Equation 8.61). The measured Kc values were calcu-
lated for each day by dividing ET measured by the precision lysimeter system by ETo
or ETr calculations. Stage lengths for the 1974 bean crop were 25/25/30/20 days for
the FAO Kcb application and stage dates for constructing the curvilinear Kcb curve of
Wright were taken from Table 8.9.
Both methods for constructing the daily Kcb + Ke curves fit observed Kc relatively
well (Figures 8.7 and 8.8), where wet soil evaporation following precipitation and
irrigation temporarily increased values for Kc act. The water-balance based FAO-56 Ke
method estimated Kc act to remain above the Kcb curve throughout the midseason period
for the bean crop due to the frequent irrigation and low, sustained rate of evaporation
caused by the nearly full ground cover. The estimated Kc act followed measured Kc
relatively closely during the initial period when reduction in Kcb and Es was simulated
due to dryness of the surface soil layer, and during the late season period when soil
water content of the root zone was estimated to be insufficient to sustain ET at poten-
tial levels. These estimates were borne out by the lysimeter record.
Lysimeter-computed Kc act on about five days around day 190 (beginning of mid-
season period) lay beneath both the FAO and Wright (1982) Kcb curves. The Wright
Kcb curve (Figure 8.8) was constructed by Wright based on two years of data (only one
is shown). The FAO-56 Kcb curve (Figure 8.7) was constructed to follow the same
general shape as the Wright curve via the selection of stage lengths to provide for con-
sistent comparisons between methods. The maximum estimated and measured Kc act for
the FAO-56 ETo-based method averaged about 1.35 during the midseason period (Figure
8.7), whereas the estimated and measured Kc act for the Wright method averaged about
1.0 during the midseason period (Figure 8.8) due to the use of the alfalfa reference basis.
8.5.9 Soil Water Balance
The calculation of Ks requires a daily water balance computation for the root zone.
Often, for purposes of estimating Kc, the water content of the root zone is expressed in
terms of net depletion. This makes the adding and subtracting of losses and gains
straightforward and various components of the soil water budget can be expressed in
terms of water depth. Rainfall, irrigation, and capillary rise of groundwater add water
to the root zone and decrease the root zone depletion. Soil evaporation, crop transpira-
tion, and percolation losses remove water from the root zone and increase the deple-
tion. A daily water balance, expressed in terms of depletion at the end of the day, is:
Dr , i = Dr , i -1 − ( P − RO) i − I i − CRi + ETc act , i + DPi (8.73)
where Dr,i = root zone depletion at the end of day i, mm
Dr,i-1 = water content in the root zone at the end of the previous day, i-1, mm
Pi = precipitation on day i, mm
ROi = runoff from the soil surface on day i, mm
Ii = net irrigation depth on day i that infiltrates the soil, mm
CRi = capillary rise from the groundwater table on day i, mm
ETc act,i = actual crop evapotranspiration on day i, mm
DPi = water flux out of the root zone by deep percolation on day i, mm.
Although following heavy rain or irrigation, soil water content might temporally
exceed field capacity, in the above equation, the total amount of water exceeding field
capacity is assumed to be lost the same day via deep percolation, following any ET for
that day. This does permit the extraction of one day’s ET from this excess before per-
colation. The root zone depletion will gradually increase as a result of ET. In the ab-
sence of a wetting event, the root zone depletion will reach the value TAW defined
from rooting depth, θ FC and θWP from Equation 8.71. At that moment no water is left
for ET, and Ks becomes zero, from Equation 8.70. Limits imposed on Dr,i are:
0 ≤ Dr , i ≤ TAW (8.74)
8.5.9.1 Initial depletion. To initiate the water balance for the root zone, the initial
depletion Dr,i-1 can be derived from measured soil water content by:
Dr , i-1 = 1000(θ FC − θ i -1 ) Z r (8.75)
where θ i-1 is the average soil water content at the end of day i – 1 for the effective root
zone. Following heavy rain or irrigation, the user can assume that the root zone is near
field capacity, i.e., Dr,i-1 ≈ 0.
Pi is equivalent to daily precipitation. Daily precipitation in amounts less than about
0.2 ETref is normally entirely evaporated and can generally be ignored in depletion
calculations. Ii is equivalent to the mean infiltrated irrigation depth expressed for the
entire field surface. Runoff from the surface during precipitation can be estimated us-
ing standard procedures from hydrological texts.
8.5.9.2 Capillary rise (CR). The amount of water transported upwards by capillary
rise from the water table to the root zone depends on the soil type, the depth of the
water table and the wetness of the root zone. Capillary rise can normally be assumed
to be zero when the water table is more than about 1 m below the bottom of the root
zone. CR can be computed from parametric equations as a function of the water table
depth, water content of the root zone, root density and soil characteristics (Liu et al.,
2006). Some information on capillary rise is presented in Chapter 9.
8.5.9.3 Deep percolation from the root zone (DP). Following heavy rain or irriga-
tion, the soil water content in the root zone may exceed field capacity. In applications
of Equation 8.73, DP is assumed to occur within the same day of a wetting event, so
that the depletion Dr, i in Equation 8.73 becomes zero. Therefore,
DPi = ( Pi − ROi ) + I i − ETc, i − Dr , i -1 ≥0 (8.76)
As long as the soil water content in the root zone is below field capacity (i.e., Dr, i > 0),
the soil will not drain and DPi = 0. The DPi term in Equations 8.73 and 8.76 is not to be
confused with the DPe,i term used in Equations 8.66 and 8.69 for the evaporation layer.
Both terms can be calculated at the same time, but are independent of one another.
8.5.9.4 Depth of root zone. The depth of the effective root zone can be estimated as:
DoY i - DoY ini
z r i = z r min + ( z r max - z r min ) (8.77)
Lroot growth
for DoYi ≤ DoYini + Lroot growth
where zr i = effective depth of the root zone on day i, mm
zr min = initial effective depth of the root zone (generally at DoY = DoYini)
zr max = maximum effective depth of the root zone, m, reached Lroot growth days
following DoYini
DoYi = day of the year (1–365) corresponding to day i
DoYini = day of the year corresponding to the date of planting or initiation of
growth (or January 1 if a perennial is growing all of the year).
When DoYi > DoYini + Lroot growth, zr = zr max. Other nonlinear schemes for root
growth such as that by Borg and Grimes (1986) can be applied, although these are
often not very different from linear estimates.
8.5.10 ET During the Nongrowing Season
During nongrowing periods, ET is dominated by evaporation, rather than transpira-
tion, especially if the nongrowing season is caused by killing frosts. Nongrowing sea-
son ET is therefore generally best estimated using techniques that accurately estimate
evaporation from the soil surface. Evaporation is a strong function of wetting fre-
quency and reference ET rate and therefore, the Kc ini estimated from Figure 8.4 can be
used as an estimate of Kc during the nongrowing season (Martin and Gilley 1993)
with some adjustment for impacts of surface cover by dead vegetation, as described in
Section 8.9.2. Alternatively, Kc during the nongrowing season can be estimated using
the dual Kcb + Ke method with some adjustment to TEW and REW during cold periods
(Allen et al., 2005a,b). The dual procedure was applied by Allen and Robison (2007)
for complete calendar years including winter, with adjustments made during periods of
snow cover. Snyder and Eching (2004, 2005) suggested a procedure for combining the
Kc during the nongrowing season, estimated using a procedure similar to Figure 8.4,
with the Kc curve for the growing season to create a continuous Kc for the entire calen-
dar year. This was done by taking the maximum, for each day, of Kc ini as estimated from
Figure 8.4 or similar curve and the Kc obtained from the growing season Kc ini curve.
8.6 EVAPOTRANSPIRATION COEFFICIENTS

FOR LANDSCAPES
Over the past several decades, the water requirements and water consumption by
residential and urban landscapes have become increasingly important in terms of
quantity and value of water consumed. Procedures similar to those from agriculture
have been adapted to estimate ET from landscapes. However, two distinctions be-
tween agriculture and landscapes exist regarding ET quantification: landscape systems
are nearly always comprised of a mixture of multiple types and species of vegetation,
thereby complicating the estimation of ET, and, typically, the objective of landscape
irrigation is to promote appearance rather than biomass production, as is the case in
agriculture. Therefore, target ET for landscapes may include an intentional “stress”
factor in the baseline value for ETc act, where landscape plants are watered less than
they would be if they were irrigated like a crop. They are watered enough to look good
and to survive, but the plants are stressed and will not be at maximum productivity.
This adjustment can result in significant water conservation. The magnitude of the
stress factor depends on physiological and morphological requirements of the plants;
the goal is to sustain health and appearance with minimal irrigation. For example, wa-
ter conservation studies on turfgrasses have demonstrated that water savings of 30%
for cool-season turfgrasses and 40% for warm-season turfgrasses may be attainable
without significant loss of quality (Pittenger and Shaw, 2001). Many shrubs and
groundcovers can be managed for even more stress-induced reduction in ET. Addi-
tionally, few landscape sites meet the “extensive surface” requirement.
Because of the frequent inclusion of water stress in target ET values for landscape
design and management, distinction must be made between these target ET values and
actual ET values. Actual ET values may exceed target ET values if the landscape re-
ceives more water than required by the target that includes intentional stress. Under
these conditions, landscape vegetation may exploit the additional available water, sub-
ject to some limit constrained by environmental energy for evaporation and leaf area.
This limit, which follows behavior and principles used for agricultural crops, may ex-
ceed the targeted ET rate for the particular landscape cover. Conversely, actual ET
may be less than target ET values if actual stress levels to the landscape are greater
than targeted. Therefore, two ET values for landscape are distinguished here. The first
is the target landscape ET, referred to as ETL, that is based on minimum ET levels,
relative to climate, necessary to sustain a healthy, attractive landscape. The second ET
value is the actual landscape ET, ETL act, that is based on landscape type and on actual
water availability.
The target ET for a landscape is calculated as
ETL = KL ETo (8.78)
-1 -1 -1
where ETL is the target landscape ET (in mm d , mm month , or mm year ), and ETo
is the grass reference ET in the same units. KL is the target landscape coefficient, simi-
lar to the crop coefficient used in agricultural applications.
There has been relatively limited experimental research on quantifying water needs
of the diverse array of landscape plant types (Pittenger and Henry 2005). Much of the
existing information is based largely on observation rather than on scientifically ob-
tained data. Some of the leading work on landscape ET has been done in California,
where water applied to landscapes in southern California is estimated to be 25% to
30% of all water used in the region (Pittenger and Shaw 2001). Pittenger and Shaw
(2007, unpublished work from University of California Cooperative Extension) have
produced a table of KL values for 35 landscape groundcovers and shrubs that provide
acceptable landscape performance after establishment, but that cause a managed
amount of moisture stress via limited water application. Costello et al. (2000) and IA
(2005) described a recent procedure termed WUCOLS (water use classification of
landscape species), where the KL has been decoupled into reproducible and visually
apparent components representing the effects of three or fourfour factors that control
the value for KL. The decoupling was done to provide for application to the wide di-
versity of vegetation types and environments of landscape systems. Snyder and Eching
(2004, 2005) have proposed a similar decoupled procedure for estimating a formulated
KL, but which uses different ranges for the compoents. The Snyder and Eching (2004)
procedure is described here, where
KL = Kv Kd Kmc Ksm (8.79)
where Kv is a vegetation species factor, Kd is a vegetation density factor, Kmc is a mi-
croclimate factor, and Ksm is a managed stress factor. Kv can be considered to be the
ratio of ETL to ETo for a specific single or mixture of plant species under full or nearly
full ground cover and full soil water supply. Factors Kd, Kmc and Ksm modify Kv for
less than effective full ground-cover, for impacts of shading or exposure to advective
or reflective sources, and for intentional water stress. Each of these factors can be es-
timated separate from the other, based on visual observation of the landscape (Kd and
Kmc) and based on visual observation and grower experience (Ksm). Following the es-
timation of the individual factors, KL is calculated using Eq. 8.79 and represents a rela-
tively accurate and reproducible estimate of relative landscape ET. The procedure of
Snyder and Eching (2004, 2005), used in the University of California-Davis LIMP
software, differs from that of WUCOLS (Costello et al. 2000) in the ranges used to
define Kd. In addition, the procedure of WUCOLS combines the values for Kv and Ksm
into a “species” coefficient which can make the combined product difficult to esti-
mate. The procedure and factor ranges of LIMP and Pittenger et al. (2001) may be
more likely to produce more accurate and reproducible estimates of landscape ET.
8.6.1 Vegetation Coefficient
The value for Kv for landscape vegetation represents the ratio of ETL to ETo for full
or nearly full cover (shading) of the ground and with full soil water supply and is used
to estimate the maximum, potential rate of ratio of KL for the vegetation under ideal
conditions. Costello and Jones (2000) have suggested values for a “species coeffi-
cient,” Ksp, for hundreds of landscape species in California, however their values in-
clude the Ksm factor and therefore should not be used with the LIMP procedure. Based
on the definition for Kv used here, where Kv is fraction of ETo when the foliage is at
near maximum density (Kd = 1) and full water availability (Ksm = 1), large numbers of
landscape vegetation tend to have similar values for Kv due to similarity in total leaf
area and stomatal response. Therefore, a condensed table of typical values for general
species types can be used to provide general estimates for Kv for used in Equation
8.79, where Kv ranges from about 0.8 to 1.2. Historically the grass reference ETo has
been used to estimate ET from landscapes, thus the upper limit for Kv can exceed 1.0
for some tall, leafy vegetation. Table 8.10 contains general values for Kv for general
types of landscape vegetation.
Table 8.10. Vegetation (species) factors (Kv) for general plant types.
Vegetation Category[1] Kv
Trees 1.15
Shrubs, desert species 0.7
nondesert 0.8
Groundcover 1.0
Annuals (flowers) 0.9
Mixture of trees, shrubs,
and groundcover[2] 1.20
Cool season turfgrass[3] 0.9
Warm season turfgrass[4] 0.9
[1]
The tree, shrub, and groundcover categories listed are for landscapes that are
composed solely or predominantly of one of these vegetation types.
[2]
Mixed plantings are composed of two or three vegetation types (i.e., where a
single vegetation type does not predominate).
[3]
Cool season grasses include Kentucky Blue grass, fescues, perennial ryegrass
[4]
Warm season grasses include Bermuda grass, St. Augustine grass, buffalo
grass, and blue grama.
The typical Kv values in Table 8.10 represent nearly full effective ground cover (fc
> ~0.7) (see Section 8.5.3), and no water stress. The values in Table 8.10 are general
and under most conditions, these values are not met due to the density factor being
less than 1.0 and managed, intentional moisture stress. The Kv value for warm season
grass is equal to that for cool season grass in Table 8.10 since both of these values
represent ETL/ETo for nonstress conditions. Typically, warm season grasses can toler-
ate more moisture stress than cool season grasses so that a lower managed stress factor
can be applied to warm season grasses with less visual effect, as is illustrated later in
Table 8.13. The Kv values for both cool season and warm season grasses are less than
1.0 in Table 8.10 due to the tendency for their mean height to be less than that of the
standardized 0.12 m grass reference.
8.6.2 Density Factor
Landscapes can vary considerably in vegetation density, due to wide variations in
plant spacing and maturity. Plant density factors, Kd, are listed in Table 8.11 for the pri-
mary landscape vegetation types. Vegetation density refers to the collective leaf area of
all plants in the unit landscape area. More densely growing vegetation has a higher Kd
and will transpire and require more water. Immature and sparsely planted landscapes
typically have less total leaf area per unit landscape area than mature landscapes and
are assigned a Kd value in the low category. Often, landscapes have two and three tiers
of vegetation including turf or groundcover, shrubs, and trees. Overlapping tiers are
capable of more radiative and other energy exchange and tend to increase ET.
Allen et al. (1998) introduced a general equation for Kd that is based on the fraction
of ground covered (or shaded at noon) by vegetation and mean plant height. This rela-
tionship has the form:
⎛ ⎛ 1 ⎞⎞
⎜ ⎟
⎜ 1+h ⎠ ⎟ (8.80)
K d = min⎜1, M L f c eff , f c⎝eff ⎟
⎜ ⎟
⎝ ⎠
where fc eff is the effective fraction of ground covered or shaded by vegetation (0 to
1.0) near solar noon, h is the mean height of the vegetation in m, and ML is a multiplier
on fc eff to impose an upper limit on relative transpiration per unit ground area as
Table 8.11. Density factor (Kd) for different vegetation

heights, h, over a range of effective fraction of ground covered
or shaded by vegetation, fc eff, from Equation 8.80, where ML = 1.5.
fc eff h = 0.1 m h = 0.4 m h=1m h=4m
0.0 0.00 0.00 0.00 0.00
0.1 0.12 0.15 0.15 0.15
0.2 0.23 0.30 0.30 0.30
0.3 0.33 0.42 0.45 0.45
0.4 0.43 0.52 0.60 0.60
0.5 0.53 0.61 0.71 0.87
0.6 0.63 0.69 0.77 0.90
0.7 0.72 0.78 0.84 0.93
0.8 0.82 0.85 0.89 0.96
0.9 0.91 0.93 0.95 0.98
1.0 1.00 1.00 1.00 1.00
represented by fc eff. Equation 8.80 estimates larger values for Kd as vegetation height
increases. This accounts for the impact of larger aerodynamic roughness and generally
more leaf area with taller vegetation, given the same fraction of ground covered or
shaded. Allen et al. (1998) provided means to estimate values for fc eff as a function of
time of year, latitude, vegetation height, canopy shape and, in the case of row vegeta-
tion, row orientation. These relationships are provided in Equations 8.81 and 8.82 for
randomly placed vegetation. For practical purposes, and due to the uncertainties asso-
ciated with estimating fc for landscape vegetation, the value for fc eff can often be as-
sumed to be the same as fc. The “min” function takes the smallest of the three values
separated by commas. Parameter ML is expected to range from 1.5 to 2.0, depending
on the canopy density and thickness and the value can be modified to fit the specific
vegetation. The ML fc eff limit usually becomes invoked only for h greater than about 1
to 2 m.
For canopies such as trees or randomly planted vegetation, fc eff can be estimated as
fc
f c eff = ≤1 (8.81)
sin(β )
where β is the mean angle of the sun above the horizon during the period of maximum
ET (generally between 1100 and 1500 hours). β is calculated from Equation 8.43.
Generally, fc eff can be calculated at solar noon when ω = 0 and Equation 8.43 becomes
β = arcsin [sin(ϕ )sin(δ ) + cos(ϕ )cos(δ )] (8.82)
Figure 8.9 shows Kd estimated by Equation 8.80 over a range of fraction of (effec-
tive) ground cover and various mean heights for vegetation and compared to estimates
for Kd from other sources for specific vegetation types. Hernandez-Suarez (1998) pro-
duced estimates for Kd for low-growing vegetation and Fereres (1991) produced esti-
mates for tree orchards. Snyder and Eching (2005) introduced an equation fitted to the
Fereres (1991) data (Sammis et al., 2004). The Kd estimates by Hernandez-Suarez are
similar to those by Equation 8.80 for h of 0.1 to 0.3 m and those by Fereres (1991) are
simlar to those by Equation 8.80 for h = 5 m. These comparisons suggest that Equation
8.80 works well as a general estimate for Kd over a range of fc eff and h.
1.2
Fereres and h=5m
1.0 Snyder and Eching
(2005) (Orchards)
0.8 h=3m
h=1m
0.6
K
d h = 0.4 m
0.4 Hernandez-Suarez
(Vegetables)
0.2 Eq. 8.82 , h = 0.1 m
0.0
0 0.2 0.4 0.6 0.8 1
Fraction of Ground Cover
Figure 8.9. Density coefficient, Kd, estimated from Equation 8.80 with ML = 2 over a range
of fraction of ground cover and various plant heights, and compared with estimates
by Fereres (1981) for orchards and Hernandez-Suarez (1988) for vegetables.
When two substantial tiers of vegetation are present, for example trees shading
grass or flowers, the value for h can be approximated as a geometric mean of the
heights of the vegetation or in proportion to the fc for each tier. The value for fc should
be the total fc for the two vegetation tiers combined, accounting for any overlapped
shading, i.e. using a weighted average.
It is important to note that the Kd factor estimated in this section from fc presumes
that the fraction of ground surface not covered by vegetation (i.e., bare soil), is rela-
tively dry and therefore does not contribute substantially to the ETL. In situations
where the exposed soil surface is wet a majority of the time, the value for Kd should be
increased by 10 to 20%, subject to Kd ≤ 1, to account for soil surface evaporation, es-
pecially for trees and shrubs. Alternatively, a daily dual Kcb + Ke procedure, described
previously, can be applied where Kcb is set equal to KL from Eq. 8.79 and Ke is esti-
mated from wetting frequency.
8.6.3 Microclimate Factor
Structures and paved areas typical of urban landscapes have pronounced effect on
the local energy balance and thus ET demand of adjacent vegetated areas due to the
transfer of additional energy for evaporation. The environmental conditions of a land-
scape may vary significantly across the landscape, for example, areas on the south side
of a building vs. on the north side. The microclimate factor, Kmc, accounts for impacts
of sun, shading, protected areas, hot or cool areas, reflected and emitted radiation from
structures, wind, and transfer of heat energy from low ET surroundings on ET. Plant-
ings adjacent to paved, open areas may have 50% higher ET demand than similar
plantings among other vegetation due to the transfer of energy to the vegetation. Con-
versely, plantings in shaded areas from sun and wind may have ET rates only one-half
as high as those for open settings.
Values for Kmc are listed in Table 8.12 for the general classes of vegetation. In gen-
eral, the “high” category (Kmc > 1) in Table 8.12 reflects harsh microclimate condi-
tions such as planting in direct sunlight near a paved or other nonvegetated surface,
planting near a reflecting window or heat-absorbing surfaces, or in exposed, windy
conditions. The “low” category (Kmc < 1) reflects an environment where the planting is
Table 8.12. Microclimate factor, Kmc,

for different plant types (from Irrigation Association, 2003).
Average
(reference
Vegetation High condition) Low
Trees 1.4 1.0 0.5
Shrubs 1.3 1.0 0.5
Groundcover, flowers 1.2 1.0 0.5
Mixture of trees, shrubs, and
groundcover 1.4 1.0 0.5
Turfgrass 1.2 1.0 0.8
located in shade, shielded from wind, and away from dry, hot surfaces. The average or
medium category (Kmc = 1) represents a reference condition similar to an open park
setting, where conditions caused by buildings, pavement, shade, and reflection do not
influence the ET by the landscape. The values given for Kmc are only approximate and
local measurements are recommended to confirm or to derive local values.
The water manager should select the appropriate microclimate adjustment factor for
each sector of a landscape or for each irrigation zone. For example, a turfgrass zone
that thrives in sun may have an average Kmc of 1.0. The same zone in full shade of a
building during midday may be assigned a Kmc of 0.8 or lower to better reflect the ac-
tual plant water needs.
8.6.4 Managed Stress Factor
As stated previously, the typical objective of landscape irrigation is to promote ap-
pearance rather than biomass production, as is the case in agriculture. Therefore, target
ET for landscapes may include an intentional and managed “stress” factor in the base-
line value for ETc act, where landscape plants are watered less than they would be if
they were irrigated like a crop. This management is done by adjusting irrigation water
schedules to apply less water than the vegetation will potentially transpire. The magni-
tude of the stress factor depends on physiological and morphological requirements of
the plants. For example, water conservation studies on turfgrasses have demonstrated
that water savings of 30% for cool-season turfgrasses and 40% for warm-season
turfgrasses may be attainable without significant loss of quality (Pittenger and Shaw,
2001). Many woody shrubs and groundcovers can be managed for even more stress-
induced reduction in ET (Kjelgren et al., 2000).
Pittenger et al. (2001), Shaw and Pittenger (2004), and Pittenger and Shaw (2007)
defined water needs of non-turf landscape plants as a percentage of ETo needed to
maintain their appearance and intended function (e.g. shade, green foliage, screening
element). In Equation 8.79, the landscape coefficient KL (equivalent to the crop coeffi-
cient) is decoupled into components that describe the impacts of vegetation type, den-
sity of vegetation, microclimatic effects, and managed stress factor. The managed
stress factor, Ksm, represents the fraction of full ET rate targeted to obtain the func-
tional and visual characteristics of the landscape vegetation. Parameter Ksm has a range
of 0 to 1.0 where 1.0 represents conditions of no moisture stress (and no real water
conservation) and 0 represents complete lapse of plant transpiration (i.e., probable
plant death). High values for Ksm will sustain relatively more lush, high leaf-area vege-
tation stands that tend to maximize ET and that may be necessary to sustain long-term
plant health or appearance. Low values for Ksm represent substantial managed plant
water stress and reduction in ET, generally at the cost of biomass accumulation and
potentially visual effects (Richie and Pittenger, 2000).
Many landscape species can exercise significant stomatal control and can be
“forced” to support relatively lower levels of ET. For instance, the low range for
groundcover is 0.2, which is appropriate for a selected group of drought-tolerant
groundcover species. This value may not be appropriate for some ornamental ground-
covers, however, that require higher amounts of water (and less water stress) to main-
tain health and appearance. It is recommended that one consult with local or regional
sources to determine the appropriate managed stress factor (Ksm) to apply. Pettinger
and Shaw (2007) suggested KL for more than 30 groundcovers and shrubs grown in
southern California that contain low Ksm components and thus provide good water
conservation. Many of the vegetation types listed by Pettinger and Shaw are native
desert vegetation types that tolerate water stress. Other sources of KL information for
specific species where the KL includes an implied Ksm < 1 include the WUCOLS pub-
lication by Costello and Jones (1999).
8.6.5 Actual ET from Landscapes
As discussed in the previous section, the vegetation coefficient KV represents a suf-
ficient water supply to support full ET from relatively dense vegetation having near
maximum ground cover and open environmental exposure. However, the KL coeffi-
cient may contain an implicit amount of managed stress for purposes of water conser-
vation.. The degree of implied managed stress is quantified in Eq. 8.79 by the Ksm
term. Therefore, the KL derived from Equation 8.79 using a recommended Ksm term
may not represent actual conditions where actual stress deviates from the managed or
target stress. Under these conditions, for purposes of water balance and determination
of consumptive use from a landscaped or larger area, the “managed stress” coefficient
in Equation 8.79 must be replaced by an “actual” stress coefficient, Ks, where Ks is
computed from Equation 8.70 based on soil water depletion determined from a daily
balance of root-zone soil water, for example usng Equation 8.73. Equation 8.79 there-
fore assumes the form:
KL act = Kv Kd Kmc Ks (8.83)
where Ksm has been replaced by an actual stress coefficient Ks and where KL act is the
actual ET anticipated from the landscape under actual water availability. Following
Table 8.13. Managed stress factors (Ksm) for general plant types
and the general values of depletion fraction for no stress.
Average Depletion
managed fraction, p,
Vegetation Category High stress stress Low stress for no stress
Trees 0.4 0.6 0.8 0.6
Shrubs, desert species 0.3 0.4 0.6 0.6
nondesert 0.4 0.6 0.8 0.6
Groundcover 0.3 0.5 0.8 0.5
Annuals (flowers) 0.5 0.7 0.8 0.4
Mixture of trees, shrubs,
and groundcover[1] 0.4 0.6 0.8 0.6
Cool season turfgrass 0.7 0.8 0.9 0.4
Warm season turfgrass 0.6 0.7 0.8 0.5
[1]
Mixed plantings are composed of two or three vegetation types and where a single vegetation type
does not predominate.
calculation of KL act, actual ET from the landscape under actual watering conditions is
ETL act = K L act ETo (8.84)
When Ks is estimated from Equation 8.70, the depletion fraction parameter p used
to estimate RAW can be set to values listed in Table 8.13. However, it is recom-
mended that one use specific values determined for specific species if these are avail-
able. The effective depth of the root zone used to estimate TAW can be species or vari-
ety specific and therefore information specific to the variety can be important.
8.7 ESTIMATING Kc FROM FRACTION OF COVER
When a Kc value is needed for vegetation that is not similar to that listed in Tables
8.2, 8.3, or 8.5, or where a more specific value for KL (or the product KLKd ) is desired
than provided by Table 8.10, the function of FAO-56 used to create the Kd density
function and factor of Equation 8.80 can be applied.
The estimation of Kc for the initial growth stage of annuals, where the soil surface
is mostly bare, can be determined from following Section 8.5.3.4 for the mean Kc (Kc
ini) where the crop coefficient in this stage is primarily determined by the frequency
with which the soil is wetted. In the dual Kc approach, the Kcb for the initial period can
be estimated as 0.1 to 0.15 for bare soil.
The Kc during the midseason and late season periods, if for a low fraction of ground
cover, will be affected to a large extent by the frequency of precipitation and/or irriga-
tion. Therefore, the basal Kcb + Ke approach is recommended, with Kcb estimated using
the following Equation 8.85 based on the fraction of ground covered by vegetation
during the particular period. For landscape crops, the impact of less than full ground
shading on the ETL is incorporated by the Kd parameter in Equation 8.79 or 8.83 for
KL. In the case of agricultural crops, Kd is used in a similar way:
( )
K cb = K c min + K d K cb full − K c min (8.85a)
where Kcb = estimated basal Kcb (for example, during the midseason) when plant
density and/or leaf area are at or lower than full cover
Kc min = the minimum Kcb representing bare soil
Kcb full = the basal Kcb anticipated for the vegetation under full cover conditions
and corrected for climate
Kd = the density factor from Equation 8.80.
Kc min ≈ 0.0 during long periods of no rain or irrigation and Kc min ≈ 0.15 to 0.20 during
periods of rain or irrigation. For tree crops having grass or other ground cover, Equa-
tion 8.85a can be re-expressed as:
[ ]
K cb = K cb cover + K d (max K cb full − K cb cover , 0 ) (8.85b)
where Kcb cover is the Kcb of the ground cover in the absence of tree foliage.
Kcb full for use with ETo can be approximated as a function of mean plant height and
adjusted for climate similar to the Kcb mid:
0 .3
⎛h⎞
K cb full = min (1.0 + 0.1 h, 1.20 ) + [0.04 (u 2 − 2) − 0.004 ( RH min − 45) ]⎜ ⎟ (8.86a)
⎝3⎠
For use with alfalfa reference ETr, the climatic correction is not required for Kcb full
because of the characteristics of the alfalfa reference crop. The equation is then:
K cb full = min (0.8 + 0.1 h, 1.0 ) (8.86b)
8.8 EFFECT OF IRRIGATION METHOD ON KC

The type of irrigation system can affect the magnitude of ETc, principally in the
fraction of the field surface that is wetted during irrigation and in the irrigation inter-
val. These two characteristics affect the average rate of evaporation from the soil sur-
face and the frequency with which the soil surface is wetted. Other factors influencing
ETc are the amounts of evaporation losses from droplets during sprinkle irrigation and
evaporation of intercepted water from wet canopies following sprinkle irrigation.
All water evaporated from sprinkle droplets and from wet canopies consumes en-
ergy, as liquid is changed to vapor. This energy is supplied from sensible heat con-
tained in the lower atmosphere and from solar radiation. The evaporation process
cools and humidifies the air. As a result, the ET demand placed on the vegetation can-
opy itself (inside the leaf stomates) is generally moderated during the free water
evaporation of water droplets. This effectively reduces the ETc demand from the soil
profile estimated to occur in the absence of the free water evaporation. Therefore, the
evaporation losses computed from free water surfaces during sprinkle irrigation cannot
be directly added to the potential ETc demand as computed with Equation 8.56. As
implied by Equation 8.61, an upper limit on the sum of free water evaporation from
sprinkler irrigation and ETc exists. This sum approximately equals Kc max ETref. There-
fore, Kc max ETref can be used to represent the maximum amount of evaporation and
transpiration losses to be expected from an entire field during and immediately follow-
ing irrigation for most types of irrigation systems.
8.9 EFFECTS OF SURFACE MULCHING ON KC
Mulches are frequently used in vegetable production to reduce evaporation losses
from the soil surface, to accelerate crop development in cool climates by increasing soil
temperature, to reduce erosion, or to assist in weed control. Mulches may be composed
of organic plant materials or they may be synthetic mulches comprised of plastic sheets.
Plastic mulches are the most common type of mulch used in vegetable production.
8.9.1 Plastic Mulches
Plastic mulches (or covers) are generally comprised of thin sheets of polyethylene or
similar material placed over the ground surface and generally along plant rows. Holes
are cut through the film at plant spacings to allow emergence of vegetation. Polyethylene
covers are usually either clear or black. Effects on ETc by the two colors are generally
similar (Haddadin and Ghawi, 1983; Battikhi and Hill, 1988a,b; and Safadi, 1991). Plas-
tic mulches substantially reduce the evaporation of water from the soil surface, espe-
cially under trickle irrigation systems. Associated with the reduction in evaporation is a
general increase in transpiration from vegetation caused by transfer of both sensible and
radiative heat from the surface of the plastic cover to adjacent vegetation. Usually, the
ETc from mulched vegetables is about 5% to 30% lower than for vegetable production
without a plastic mulch. A summary of observed reductions in Kc, evaporation, and in-
creases in transpiration over growing seasons is given in Table 8.14 for five vegetable
crops. Even though the transpiration rates under mulch may increase by an average of
10% to 30% over the season as compared to using no mulch, the Kc’s decrease by an aver-
age of 10% to 30% due to the 50% to 80% reduction in evaporation from wet soil. Gener-
ally, crop growth rates and vegetable yields are increased with the use of plastic mulches.
To consider the effects of plastic mulch on ETc, the values for mean Kc mid and Kc end
for vegetables listed in tables can be reduced by 10% to 30%, depending on the fre-
Table 8.14. Approximate reductions in Kc, surface evaporation, and increases in

transpiration for various vegetable crops under plastic mulch as compared
with no mulch using trickle irrigation (from Allen et al., 1998).
Reduction Reduction in Increase in
in Kc[1] Evaporation[1] Transpiration[1]
Crop (%) (%) (%)
Squash 5-15 40-70 10-30
Cucumber 15-20 40-60 15-30
Cantaloupe 5-10 80 35
Watermelon 25-30 90 –10[2]
Tomato 35 n/a n/a
Average 10-30 50-80 10-30
[1]
Relative to using no mulch.
[2]
A negative increase, i.e., a decrease in transpiration.
quency of irrigation; use the higher value for frequent trickle irrigation. The value for
Kc ini is often as low as 0.10. When adjusting the basal Kcb for mulched production, less
adjustment is needed, as compared to the mean Kc curve, being on the order of perhaps
5% to 15% reduction in Kcb, since the basal evaporation of water from the soil surface
is less with a plastic mulch, but the transpiration is relatively more. Local calibration
of Kcb (and Ke) is encouraged. When applying a basal approach with plastic mulch, fw
should represent the relative equivalent fraction of the ground surface that contributes
to evaporation through the vent holes in the plastic cover. This fraction can be substan-
tially (at least 2 to 3 times) larger than the area of the vent holes to account for vapor
transfer from under the sheet.
8.9.2 Organic Mulches
Organic mulches are sometimes used with orchard production and row crops under
reduced-tillage operations. Organic mulches may be comprised of unincorporated
plant residue or foreign material imported to the field. The depth of the organic mulch
and the fraction of the soil surface covered can vary widely. These two parameters
affect the amount of reduction in evaporation from the soil surface.
A general rule of thumb with a mulched surface is to reduce the amount of soil wa-
ter evaporation by about 5% for each 10% of soil surface that is covered by the or-
ganic mulch. For example, if 50% of the soil surface were covered by an organic crop-
residue mulch, then the soil evaporation would be reduced by about 25%. To apply
this to the Kc values in tables, one would reduce Kc ini values by about 25% and would
reduce Kc mid values by 25% of the difference between Kc mid and Kcb mid.
When applying the basal approach with separate water balance of the surface soil
layer, the magnitude of Esp can be reduced by about 5% for each 10% of soil surface
covered by the organic mulch. These recommendations are approximate and attempt
to account for the effects of partial reflection of solar radiation from residue, microad-
vection of heat from residue into the soil, lateral movement of soil water from below
residue to exposed soil, and the insulating effect of the organic cover. These parame-
ters can vary widely, so that local research and measurement are encouraged.
8.10 PRECIPITATION RUNOFF
Runoff during precipitation events is strongly influenced by land cover, soil tex-
ture, soil structure, sealing and crusting of the soil surface, land slope, local land form-
ing (tillage and furrowing), antecedent moisture, and precipitation intensity and dura-
tion. Therefore, estimation of runoff during precipitation is generally fraught with un-
certainty. For general purposes, runoff can be estimated using the USDA-NRCS curve
number approach. The NRCS curve number is simple to apply and is widely used
within the hydrologic, soils, and water resources communities. Required data are daily
precipitation depth and computation of a daily soil water balance by which to select
the antecedent soil water condition.
8.10.1 The NRCS Curve Number
The curve number (CN) represents the relative imperviousness of a soil-vegetation
complex and ranges from 0 for infinite perviousness and infiltration to 100 for com-
plete imperviousness and total runoff (beyond abstraction). Generally the value for CN
is selected from standard tables based on general crop and soil type and is adjusted for
the soil water content prior to the wetting event. Examples of values for CN for vari-
ous crop and soil combinations are given in Table 8.15. Parameter S in the CN proce-
dure is the maximum depth of water that can be retained as infiltration and canopy
interception during a single precipitation event (in mm). S is calculated as
⎛ 100 ⎞
S = 250 ⎜ − 1⎟ (8.87)
⎝ CN ⎠
and surface runoff is then calculated for P > 0.2 S as
RO =
(P − 0.2S )2 (8.88)
P + 0 .8 S
where RO is the depth of surface runoff during the event (mm), and P is the depth of
rainfall during the event (mm). The 0.2S term is abstracted precipitation, i.e., that in-
tercepted by canopy and soil surface before any runoff occurs. If P < 0.2 S, then RO =
0.0. In addition, RO ≤ P applies.
Table 8.15. Typical curve numbers for general crops for antecedent
soil water condition (AWC) II from USDA-SCS (1972) and Allen (1988).
Soil Texture
Crop Coarse Medium Fine
Spring wheat 63 75 85
Winter wheat 65 75 85
Field corn 67 75 85
Potatoes 70 76 88
Sugar beets 67 74 86
Peas 63 70 82
Dry edible beans 67 75 85
Sorghum 67 73 82
Cotton 67 75 83
Paddy rice 50 60 70
Sugar cane, virgin 60 69 75
Sugar cane, ratoon 60 68 76
Fruit trees, bare soil 65 72 82
Fruit trees, ground cover 60 68 70
Small garden vegetables 72 80 88
Tomatoes 65 72 82
Alfalfa hay 60 68 77
Suggested defaults 65 72 82
The curve number is affected by the soil water content prior to the rainfall event,
since soil water content affects the soil infiltration rate. Therefore, the CN is adjusted
according to estimated soil water content prior to the rainfall event. This soil water
content is termed the antecedent soil water condition or AWC. Adjustment ranges for
CN were defined by SCS (1972) for dry (AWC I) and wet (AWC III) conditions. SCS
defined AWC I as occurring when “watershed soils are dry enough for satisfactory
plowing or cultivation to take place” and AWC III as when the “watershed is practi-
cally saturated from antecedent rains” (USDA-SCS, 1972, p. 4.10). AWC II is defined
as the “average condition” and represents values in Table 8.14.
Hawkins et al. (1985) expressed tabular relationships in USDA-SCS (1972) in the
form of equations relating CN for AWC I and AWC III to CN for AWC II:
CN II
CN I = (8.89)
2.281 − 0.01281 CN II
CN II
CN III = (8.90)
0.427 + 0.00573 CN II
where CNI is the curve number associated with AWC I (dry) [0 - 100], CNII is the
curve number associated with AWC II (average condition) [0 - 100], and CNIII is the
curve number associated with AWC III (wet) [0 - 100].
The soil surface layer water balance associated with the dual Kc procedure (Equa-
tion 8.66) can be used to estimate the AWC condition. An approximation for the de-
pletion of the soil surface layer at AWC III (wet) is when De = 0.5 REW, i.e., when the
evaporation process is halfway through stage 1 drying. This point will normally be
when approximately 5 mm or less have evaporated from the top 150 mm of soil since
the time it was last completely wetted. Thus, the relationship:
De- AWC III = 0 .5 REW (8.91)
where De-AWC III is the depletion of the evaporative layer at AWC III. AWC I can be
estimated to occur when 15 to 20 mm of water have evaporated from the top 150 mm
of soil from the time it was last completely wetted. This is equivalent to when the
evaporation layer has dried to the point at which De exceeds 30% of the total evapor-
able water in the surface layer beyond REW. This depletion amount is expressed as De
= REW + 0.3(TEW – REW), where TEW is the total evaporable water in the surface
layer. Therefore,
De - AWC I = 0 .7 REW + 0.3 TEW (8.92)
where TEW is the cumulative evaporation from the surface soil layer at the end of
stage 2 drying. When De is in between these two extremes, i.e., 0.5 REW < De < 0.7
REW + 0.3 TEW, then the AWC is in the AWC II condition and the CN value is line-
arly interpolated between CNI and CNIII. In equation form:
CN = CN III for De ≤ 0.5 REW (8.93)
CN = CN I for De ≥ 0.7 REW + 0.3 TEW (8.94)
and, for 0.5 REW < De < REW + 0.3 (TEW – REW):
CN =
(De − 0.5 REW )CN I + (0.7 REW + 0.3 TEW − De )CN III (8.95)
0.2 REW + 0.3 TEW
Equation 8.95 produces CNII when De is halfway between the endpoints of CNI and
CNIII due to the symmetry of CNI and CNIII relative to CNII.
8.10.2 Infiltrated Precipitation
Once the surface runoff depth is estimated using the curve number procedure, the
depth of rainfall infiltrated is calculated as
Pinf = P – RO (8.96)
where Pinf = depth of infiltrated precipitation, mm
P = measured precipitation depth, mm
RO = depth of surface runoff, mm.
If the soil will not hold the amount infiltrated, the remainder goes to deep percolation.
8.11 OTHER WATER REQUIREMENTS
8.11.1 Germination of Seeds
Seeds for many crops are planted within only a few cm of the soil surface where the
soil dries quickly, as estimated in Section 8.5.4. Therefore, frequent irrigation during
the week following planting may be required in dry environments to increase plant
viability and health. Sprinkler irrigation is especially suited to seed germination be-
cause small depths of water can be applied. Soil wetting by furrow irrigation is prac-
ticed in many areas, but more water is required than with sprinklers since water re-
quires more time to move horizontally from the furrow to the ridge where the seed
may be planted. Furthermore, salt tends to concentrate in the ridge by evaporation
when saline water is used.
Sprinkler systems for germination are normally solid set or center pivot systems
that allow for daily or even more frequent wetting. Evapotranspiration requirements
under conditions of high frequency irrigation are estimated using the Kc = Kcb + Ke
procedure, where Kc typically approaches 1.2 to 1.3 for grass reference ETo and 1.0 for
alfalfa reference ETr during the germination period. Evaporation from soil following
irrigation reduces the warming of the soil, which may benefit seedling growth.
8.11.2 Climate Modification
Irrigation can be used to both warm and cool the crop environment to increase
yields. Evaporation during irrigation and from wet soil following irrigation cools the
air boundary layer and improves growing conditions during periods of high air tem-
perature for sensitive crops of fresh vegetables and fruits including peas, tomatoes,
cucumbers, muskmelons, strawberries, apples, and grapes (Burman et al., 1980). Cool-
ing is accomplished by conversion of sensible heat of the air and soil into latent heat
of vaporization.
All crops generally have foliar temperature thresholds above which photosynthesis
is reduced due to stomatal closure or where a component of the carbon fixation proc-
ess moves outside its kinetic window (Hatfield et al., 1987). For cool-season crops this
occurs at temperatures in the neighborhood of 23° to 28° C and for warm-season crops
at about 32° C (Hatfield et al., 1987; Keller and Bliesner, 1990). Foliar cooling re-
quires high-intensity irrigation, where from two to six short applications are needed
each hour. These intensities can only be practiced with automated fixed (solid set)
sprinkler systems. Application rates are light, typically averaging 1 mm h-1 (Keller and
Bliesner, 1990). Foliar cooling is of course more effective in arid climates where the
low vapor pressure of the air and corresponding large vapor pressure deficit (es – ea)
facilitates conversion of sensible heat into heat of vaporization. Misting to improve
greenhouse environments is a common practice.

Wolfe et al. (1976) and Griffin (1976) explored the use of sprinkler irrigation systems
to cool orchard crops during early season warm periods in order to delay fruit bloom
until later periods so as to reduce the risk of bloom freezing. Griffin (1976) and Griffin
and Richardson (1979) developed mathematical models to estimate when irrigation
bloom delay is effective. Their procedure was reviewed by Keller and Bliesner (1990).
8.11.3 Freeze Protection
Both overhead and under-tree sprinkler systems are used for freeze protection. The
goal of irrigation for freeze protection is to convert enough liquid irrigation water to
solid (ice) that the latent heat of fusion (approximately 335 kJ kg-1) released during the
freezing of irrigation water is sufficient to maintain air temperature at 0°C. Normally
fruit and vegetables will not freeze at 0°C due to the presence of sugars and other
molecules in the biomass. Because irrigation for freeze protection is practiced when
air temperatures and wind speeds are low and generally during nighttime, evaporation
rates are low. Therefore, most irrigation water added infiltrates the soil or runs off
following melting. Further information is provided by Snyder et al. (2005).
Design of sprinkler systems for frost control is covered in Keller and Bliesner (1990),
Martin and Gilley (1993), and in Chapter 16. Minimum application rates are typically
2.5 to 3 mm h-1 (Keller and Bliesner, 1990). Larger applications are needed as the atmos-
pheric dewpoint decreases below 0°C (Blanc et al., 1963; Keller and Bliesner, 1990).
8.11.4 Fertilizer Application
Application of fertilizer by irrigation is often more economical than by other
means. Generally, irrigation for fertilizer, herbicide, or pesticide application is only
initiated when there is sufficient water deficit in the root zone to retain the applied
water. Otherwise, some of the applied chemical will leach below the root zone. There-
fore, the evapotranspiration requirement for fertilizer application is considered to be
essentially zero, unless it requires more frequent irrigation than required for soil water
replacement only. This topic is described in detail in Chapter 19.
8.11.5 Soil Temperature
Soil temperatures can be markedly affected by irrigation water, through cooling by
the water itself and by cooling during evaporation of water from wet soil following
irrigation. Low irrigation water temperatures may depress soil temperatures and im-
pede plant development. Soil cooling may be desirable under certain conditions, such
as establishing seedling stands for head lettuce and during germination of grasses.
8.11.6 Dust Suppression
Dust suppression, though not limited to agricultural fields, can be achieved using
sprinkler systems. Feedlot dust can be generated in hot, dry climates when cattle be-
come active in early evening and the air boundary layer is stable, impeding mixing.
Dust is common around construction sites and on unpaved roads. In all cases, dust can
be suppressed by sprinkling. The evaporation component can be estimated using the
dual Kc procedure in Section 8.5.4. A reasonable approximation can be made using
Kc ini from Figure 8.4a (light wetting events) where the wetting frequency and ETo rate
are used to estimate the average Kc during each wetting period.
8.11.7 Leaching Requirements
Leaching of salts from the crop root zone is an essential component of irrigation
when the irrigation water carries salts. Generally, leaching of salts is accomplished
during irrigation to replace ET losses so that the ET requirement is not increased by

leaching. However, in some arid climates having heavy soils, special leaching irriga-
tions are necessary. In these situations, the annual consumptive irrigation water re-
quirement is increased by the evaporation from wet soil during and following each
leaching irrigation. Chapter 7 describes the estimation of leaching requirements.
8.12 EFFECTIVE RAINFALL
Estimation of effective rainfall is necessary in calculating net and gross irrigation
water requirements (Equations 8.1 and 8.2). In very arid climates, the majority of infil-
trating rainfall is retained in the root zone and is consumed at some point as evapora-
tion or transpiration. In more humid climates where rainfall totals are often greater
than ET, portions of rainfall percolate below the root zone or runoff, thereby reducing
the fraction that is effective in supplying crop ET. Evaporation of rainfall from wet
soil is estimated using the Kcb + Ke procedure described in Section 8.5.4.
The depth of effective rainfall is estimated by subtracting runoff estimated using
the curve number (Section 8.10.1) or other infiltration estimation approach and sub-
tracting deep percolation from the root zone computed by Equation 8.76. The soil water
balance for the root zone is updated each day using Equation 8.73 or similar equation.
Patwardhan et al. (1990) found that using a daily soil water balance equation
(Equation 8.73) to estimate effectiveness of precipitation to be significantly more ac-
curate than more simple and vague procedures such as the SCS monthly effective pre-
cipitation method (NRCS, 1993; Dastane, 1974), especially for poorly drained soils.
The effectiveness of rainfall is influenced by irrigation, since irrigation prior to a rain-
fall event reduces the capacity of the soil to both infiltrate and retain the precipitation
in the profile. The impact of irrigation on rainfall effectiveness is best determined by
daily soil water balance. Application of the Kcb + Ke procedure provides an estimate of
fraction of rainfall that evaporates with and without irrigation. The difference in DP
and RO over the season with and without irrigation provides an indication of the rela-
tive impact of irrigation on retained precipitation.
The daily soil water balance is the best method for estimating the carryover of win-
ter precipitation into the growing period. However, the application is complicated by
the impact of frozen soils on infiltration, by the estimation of crop coefficients and ETc
during the dormant season, and by the challenge of estimating the energy and evapora-
tion balance from any snow cover (Allen, 1996b; Wright, 1996). Chapter 11 of FAO-
56 provides guidance on estimating ET during winter and dormant seasons.
There is some question on whether the evaporating compoenent of P (E from the
soil surface) should be considered to be “effective.” This is especially true if estimat-
ing accumulation of nongrowing season P in the root zone. In this case, only the por-
tion of P that infiltrates and does not evaporate from the surface evaporation layer will
accrue over time. Similar arguments can be made in estimating effective Pduring the
growing season. If the dual Kcb + Ke method is ued to estimate ET, then the evapora-
tive component of P is accounted for in the daily soil water balance and the estimation
for effective precipitation is more accurate.
8.13 DESIGN REQUIREMENTS
Design and operation of irrigation systems is affected by both the peak irrigation
water requirement and by the seasonal irrigation requirement. The seasonal irrigation
water requirement dictates the annual operating time for the system and corresponding
costs for labor, water, and energy. The peak water requirement dictates the minimum
capacity for supply pipes, pumps, and canals to sustain potential crop growth. It also
governs the maximum land area that can be adequately supplied by a sprinkler irriga-
tion lateral. The peak water requirement decreases as the allowable interval between
irrigations increases as discussed in the next section.
8.13.1 Estimating Peak ET Using Climatic Records
Because evapotranspiration is driven by weather parameters, ET can vary signifi-
cantly and randomly from day to day and from year to year. This is illustrated in Fig-
ure 8.10 where daily alfalfa reference ETr estimated using the ASCE-EWRI standard-
ized Penman-Monteith equation (Equation 8.3) is shown for a 20-year period at Kim-
berly, Idaho. The variation in daily ETr is large. Overlain on the daily ETr are prob-
ability lines for various levels of nonexceedence, based on a normal probability distri-
bution. Nonexceedence is defined as the value of ET that is not expected to be ex-
ceeded p% of the time, where p is the probability level.
A normal probability distribution presumes that the coefficient of skewness (CS) =
0, where CS is the ratio of skew to the population mean. Generally, with evapotranspi-
ration, CS approaches 0, so that frequency estimates based on the normal distribution
are generally valid (Allen and Wright, 1983; Allen et al., 1983). In the case of the
normal distribution, an estimate for ET for a specific probability of nonexceedance is
estimated as:
ET Pn = ET mean + K Pn s (8.97)
where ETPn is the ET rate expected to be exceeded only 100% – p% of the time, KPn is
a probability factor, and ETmean and s are the estimates for the mean and standard de-
viation of the underlying ET population. Generally, ETmean, s, and ETPn are computed
for a specific time period during the growing season, for example, for the peak 30-day
period. The ETmean is calculated as:
∑ ET
ET mean = (8.98)
n
14
Daily ETr, mm/day
12
10 98% 90% 95%

70%
8
50%
6 5%
2%
30%
4 10%
2
0
0 30 60 90 120 150 180 210 240 270 300 330 360
Day of year
Figure 8.10. Distribution of daily calculated alfalfa reference ETr at Kimberly, Idaho,
over a 20-year period and overlay of probability lines of nonexceedence.
Table 8.16. Probability factors (KPn) for a normal distribution.

Probability of non-
exceedance, p (%) 2 5 10 25 50 75 80 90 95 98
Probability of
exceedance (%) 98 95 90 75 50 25 20 10 5 2
Return period 1.0 1.0 1.1 1.3 2 4 5 10 20 50
(years)
KPn –2.054 –1.645 –1.28 –0.675 0.0 0.675 0.841 1.28 1.645 2.054
where ET is the ET for each day of the period, over all years of record available for
analysis, and n is the number of observations. Usually, a minimum of five years of
records are required to obtain viable estimations for ETmean, s, and ETPn. ET is gener-
ally estimated from weather records. The unbiased estimate for standard deviation is
calculated as:
0.5 0.5
⎡ ∑(ET − ET 2⎤ ⎡ n ∑( ET 2 ) - (∑ ET )2 ⎤
mean )
s=⎢ ⎥ =⎢ ⎥ (8.99)
⎢⎣ (n - 1 ) ⎥⎦ ⎣⎢ n (n - 1 ) ⎦⎥
Table 8.16 lists KPn values for probabilities of interest in irrigation systems design and
operation based on the normal distribution.
When the population of ET can not be considered to follow a normal distribution,
for example, when the coefficient of skewness is less than about –0.5 or greater than
about 0.5, then a Pearson Type III distribution can be used to estimate ETPn:
ETPn = ETmean + KPPIII s (8.100)
where KPPIII is the probability factor for the Pearson Type III distribution. For cases
where –1.0 < CS < 1.0, an expression from Wilson and Hilferty (1931) and USGS
(1981) can be used for KPPIII :
2 ⎡⎢ ⎡⎛ ⎤
3
CS ⎞ CS ⎤
K PPIII = ⎢⎜ K Pn - ⎟ + 1⎥ - 1⎥ (8.101)
CS ⎢ ⎣⎝ 6 ⎠ 6 ⎦ ⎥
⎣ ⎦
where KPn is from Table 8.16 for the normal distribution and CS is the coefficient of
skewness:
n ∑( ET - ET mean )3
CS =
(n - 1) (n - 2) s3
(8.102)
n 2 ∑( ET 3) - 3 n ∑ ET ∑( ET 2) + 2 (∑ ET )3
=
n (n - 1) (n - 2) s3
When CS exceeds the limits for Equation 8.100, values for KPPIII can be taken from
tables given in USGS (1981) or other texts on frequency analysis. Martin and Gilley
(1993) described the application of the Weibull probability distribution.
The length of the averaging period has substantial effect on the values for ETPn es-
timated for a specific probability of nonexceedence. The length of averaging period in
irrigation systems design and operation should normally be based on the allowable
number of days between irrigation events. This time length, known as the irrigation
interval, Iint, is estimated as:
RAW
I int = (8.103)
ET c
where RAW is the depth of water that can be depleted with no stress to the plant, from
Equation 8.72 and ET c is the average ETc during the period:
i = d2
∑ ET ci
i = d1
ET c = (8.104)
I int
where ETci is ETc on day i and d1 and d2 are beginning and ending days for the Iint av-
eraging period, estimated as:
⎛ I int ⎞
d 1 = INT⎜ J - + 0.6 ⎟ (8.105)
⎝ 2 ⎠
⎛ I int ⎞
d 2 = INT⎜ J + - 0.4 ⎟ (8.106)
⎝ 2 ⎠
where INT( ) is the integer expression of the argument within the parentheses and J is
the day of year number at the center of the period.
Figure 8.11 shows values for ETPn for the peak June 21–July 20 period at Kim-
berly, Idaho between 1966 and 1985 as a function of the averaging period (i.e., irriga-
tion interval). Alfalfa reference ETr is shown. ETPn is shown for both the normal and
Pearson Type III frequency distributions, where differences between distributions are
small. Coefficients of skewness for the data ranged from CS = –0.5 for Iint = 1 day to CS
= –0.1 for Iint = 30 days. The comparison of ETPn computed by the two distributions in-
dicates that the normal distribution is valid and representative for the Kimberly climate.
Average ETr during period, mm/day
12
10 98%
95%
75% 90%
8 50%
25%
10%
5%
6 2%
4
2
0
0 10 20 30
Number of days in averaging period
Figure 8.11. Expected alfalfa reference ET rate at specified levels of nonexceedence as a
function of number of days in the averaging period for peak June 21–July 20 period,
1966–1985, at Kimberly, Idaho. Solid lines represent normal distributions
and grey lines represent log Pearson type III distributions.
Figure 8.11 illustrates how ETPn decreases as the length of the averaging period (or
irrigation interval) increases. For example, if an irrigation system design were based
on probability of nonexceedance = 90% (10-year return period), the ETr used in design
calculations at Kimberly would be 9.4 mm d-1 for crops having an irrigation interval of
only 3 days. Design ETr would be 9.1 mm d-1 for crops having an irrigation interval of
7 days, and 8.6 mm d-1 for crops having an irrigation interval of 30 days. In other
words, irrigation pipe, pump, and canal systems for the 30-day irrigation interval at
Kimberly could be sized about 9% smaller than systems required for the 3-day irriga-
tion interval (assuming that crop coefficients were the same). A 3-day interval would
represent shallow rooted crops (Zr ~ 0.6 m) on coarse soils. The 30-day interval would
represent deep-rooted crops (Zr ~ 2.5 m) on medium-textured soils.
8.13.2 General Design Curves
Figure 8.11 illustrates the variation in ETPn with length of averaging period for the
southern Idaho climate, which is characterized as semiarid with relatively cloudless
days during the peak monthly period. When climates have more variable cloudiness,
the variation in day-to-day ET increases since solar radiation is a primary driver of
evapotranspiration. Under variable cloudiness conditions, the likelihood of experienc-
ing a period with cloud-free weather and corresponding high ET increases as the
length of averaging period decreases. Therefore, the ratio of peak ET during short av-
eraging periods relative to mean monthly ET increases as mean cloudiness of a loca-
tion increases. This effect is illustrated in Figure 8.12 (adapted from Doorenbos and
Pruitt, 1977), where the ratio of peak ET relative to mean monthly ET is shown to
increase with increasing humidity and cloudiness. The peak ET estimated by Figure
8.12 is for the 75% probability level of nonexceedence (i.e., the ET level that is ex-
ceeded in 2.5 years out of every 10). The curves in Figure 8.12 are expressed as a
function of net application depth. Net application depth (Dn) is related to time length
between irrigations, in days, as Dn = Iint ET c and where Dn is equivalent to the nu-
merator in Equation 8.103.
1.4
4 4 Mid-continental subhumid to
Peak ET (75% Prob.)
humid climates with highly

variable cloudiness during
Mean Monthly ET
1.3 3 peak month.

3 Mid-continental climates with
variable cloudiness and peak
monthly ET = 8 mm/day.
2 2 Mid-continental climates with
variable cloudiness and peak
1.2 monthly ET = 5 mm/day.
1 Arid and semiarid climates
with predominantly clear
weather during peak month.
1.1 1
1.0
0 40 80 120 160 200
Net Application Depth, mm
Figure 8.12. Expected ratio of peak design ET to mean monthly ET of the peak month for
a 75% level of nonexceedence for four types of climates (after Doorenbos and Pruitt, 1977).
Figure 8.11 follows Curve 1 in Figure 8.12 relatively closely, where Curve 1 repre-
sents arid and semiarid climates with predominantly clear weather. The four curves in
Figure 8.12 can be reproduced mathematically by:
ET peak
= (1.18 + 0.2 N cl ) D n( - 0.037 - 0.02 N cl ) (8.107)
ET month
where Ncl is the number of the curve in Figure 8.12 [1 - 4] and Dn is the net application
depth, mm. The standard error of estimate for the ratio estimated by Equation 8.107 is
0.012 relative to the original curves of Doorenbos and Pruitt (1977). Equation 8.107
should be applied with limits of Dn > 20 mm.
8.13.3 Population Statistics from ET Estimation Methods
ET estimation equations, such as the Penman-Monteith, Penman, and Hargreaves
equations described in this chapter, may not reproduce the same variation in ET as is
found in the actual, underlying ET population. In general, the more weather parame-
ters contained in the ET equation, the larger will be the variation in ET estimates.
Therefore, in applications where weather data are represented by long-term averages,
for example wind speed, variation among ET estimates is reduced. Deviation in statis-
tics from the underlying population, as represented by standard deviation, s, and coef-
ficient of skewness, CS, will impact the estimation of ETPn or ETPPIII and subsequent
irrigation design and operation. This behavior was discussed in more detail by Allen et
al. (1983) and Allen and Wright (1983).
8.14 ANNUAL IRRIGATION WATER REQUIREMENTS
Annual evapotranspiration requirements can be calculated by computing ETref on
daily, weekly, 10-day, or monthly timesteps and applying a crop coefficient over each
period. When calculations are performed on a daily basis, the Kcb + Ke procedure can
be utilized to improve accuracy. When weekly or longer timesteps are used, a time-
averaged Kc is applied to ETref. Annual irrigation water requirements are estimated by
summing the calculated ETc over the year and subtracting effective rainfall. Gross
annual water requirements require division by a uniformity term (consumed fraction of
applied water) and potentially a leaching factor (see Equation 8.2).
Ordinarily, ET and irrigation water requirements are needed for the growing season
only. The soil water stored in the soil profile at the start of the season is estimated by
operating the soil water balance over part or all of the winter period. In winter climates
with frozen soils, this can be challenging. Often, in irrigated agriculture, the soil pro-
file is recharged to near field capacity by the start of the growing season. In climates
where the growing season is year-round, continuity between ending soil water content
from one crop to the beginning soil water content of a subsequent crop should be con-
sidered to adequately account for the total water requirement.
Acknowledgements
The ASCE-EWRI, with support by the Irrigation Association, is commended for
the work on standardization of the calculation of reference evapotranspiration that
provided the means for beginning the standardization of crop coefficients. Work by
the Irrigation Association Water Management committee has helped shape the struc-
ture for the landscape ET coefficients. The authors appreciate and acknowledge input
and advice on agricultural crop coefficients and landscape coefficients by Dr. R. L.
Snyder of the University of California, Davis.
LIST OF SYMBOLS
BR Bowen ratio
Cd denominator coefficient for the ASCE standardized PM equation, s m-1
Cn numerator coefficient for the ASCE standardized PM equation;
K mm s3 Mg-1 d-1, K mm s3 Mg-1 h-1
CF consumed fraction of applied water
CN curve number
CRi capillary rise from the groundwater table on day i, mm
CS coefficient of skewness
De,j-1 cumulative depletion from the soil surface layer at the end of day j – 1, mm
DM day of month (1–31)
Dr cumulative depletion from the root zone, including De, mm
DoY day of year
DPei,j deep percolation from the few fraction of the soil surface layer on day j, mm
Ej evaporation on day j, mm
ET evapotranspiration rate; mm d-1, mm h-1
ETc ET from a particular crop; mm d-1, mm h-1
ETc act actual evapotranspiration rate; mm d-1, mm h-1
ETL target ET from a particular landscape vegetation; mm d-1, mm h-1
ETmean mean ET of the underlying population; mm d-1
ETmonth mean ET during a month, mm d-1
ETo ET from a well-watered grass reference crop; mm d-1, mm h-1
ETpeak mean ET during peak ET period, mm d-1
ETr ET from a well-watered alfalfa reference crop; mm d-1, mm h-1
ETref reference evapotranspiration, general; mm d-1, mm h-1
G heat flux density to the ground; MJ m-2 d-1, MJ m-2 h-1, W m-2
Gsc solar constant, MJ m-2 h-1
GW contributions of shallow ground water; mm d-1, mm h-1
Ij irrigation depth on day j that infiltrates the soil, mm
Iint irrigation interval, d
J day of the year
Jmonth day of the year for the middle of the month
IR irrigation water requirement; mm d-1, mm season-1
Kc crop coefficient general
Kc act crop coefficient under actual field conditions
Kcb crop coefficient (basal), soil water not limiting transpiration, but the soil
surface is visually dry
Kcb adj basal Kcb (or Ksp pot) during the midseason when plant density and/or leaf
area are at or lower than full cover
Kcb full basal Kcb anticipated for vegetation under full cover conditions
Kd vegetation density factor
Ke evaporation coefficient
Kc end Kc at the end of the late season period
Kc ini average Kc during the initial period
Kc m mean crop coefficient
Kc max maximum value of Kc following rain or irrigation
Kc mid average Kc during the midseason period
Kc min minimum Kc for dry bare soil with no ground cover

Kco crop coefficient based on grass (ETo) reference
Kcr crop coefficient based on alfalfa (ETr) reference
Kd vegetation density coefficient
KL target landscape coefficient
KL act actual KL anticipated from the landscape under water availability
Kmc microclimate factor
Ko offset between mean daily Td and daily Tmin, °C
KPn probability factor
KPPIII probability factor for the Pearson Type III distribution
Kr evaporation reduction coefficient
Ks adjustment coefficient for water stress
Ksm coefficient for managed water stress
Kv vegetation species factor
Ksp pot potential vegetation species factor under fully watered conditions
Kt proportion of basal ET extracted as transpiration from surface soil layer
Kw units parameter in the Penman equation; mm d-1kPa-1, mm h-1kPa-1
Ldev length of development period of crop growing season, d
Lini length of initial period of crop growing season, d
Lmid length of midseason period of crop growing season, d
Llate length of late season period of crop growing season, d
Lroot growth length of period of root growth, d
Lm longitude of the solar radiation measurement site, degrees west of Greenwich
Lz longitude of the center of the local time zone, degrees west of Greenwich
LAI leaf area index
LE latent heat energy (ET) per unit area; MJ m-2 d-1, MJ m-2 h-1, W m-2
LR leaching requirement
M month number (1–12)
P atmospheric pressure, kPa
P precipitation, mm
Pe effective precipitation (entering and remaining in the root zone), mm
Pinf depth of infiltrated precipitation, mm
Ra exoatmospheric solar radiation; MJ m-2 d-1, MJ m-2 h-1, W m-2
Rn net radiation; MJ m-2 d-1, MJ m-2 h-1, W m-2
Rnl net long-wave radiation; MJ m-2 d-1, MJ m-2 h-1, W m-2
Rns net solar radiation; MJ m-2 d-1, MJ m-2 h-1, W m-2
Rs solar radiation at the surface, MJ m-2 d-1, MJ m-2 h-1, W m-2
Rso solar radiation for cloudless sky; MJ m-2 d-1, MJ m-2 h-1, W m-2
RAW readily available water in the root zone, mm
REW readily evaporable water from the surface soil layer, mm
RH relative humidity, %
RHmax maximum daily relative humidity, %
RHmean mean daily or hourly relative humidity, %
RHmin minimum daily relative humidity, %
RO surface runoff, mm
S maximum depth of water in CN procedure that can be retained as infiltra-
tion, mm
Sc seasonal correction for solar time, h

T mean daily or hourly air temperature, °C
Td dewpoint temperature of the air, °C
Tdry dry bulb temperature of the air, °C
Tei,j depth of transpiration from the few fraction of the soil surface on day j, mm
TKmax maximum daily air temperature, K
TKmin minimum daily air temperature, K
Tmax maximum daily air temperature, °C
Tmean mean daily or hourly temperature, °C
Tmin minimum daily temperature, °C
Twet wet bulb temperature of the air, °C
TAW total available water in the root zone, mm
TEW total evaporable water that can be evaporated from the surface soil layer, mm
Y year (2007, for example)
Ze effective depth of the surface soil subject to drying by way of evaporation, m
Zr maximum effective rooting depth, m
a, b constants; see usage
apsy ventilation coefficient for psychrometric constant
aw empirical coefficient in Penman wind function
bw empirical coefficient in Penman wind function, s m-1
dr inverse relative distance factor (squared) for the earth-sun
ea mean actual vapor pressure, kPa
es saturation vapor pressure, kPa
eo(T) saturation vapor pressure function at temperature T, kPa
fc fraction of the soil surface covered by vegetation
fcd cloudiness function
fc eff effective fraction of the soil surface effectively covered by vegetation
few fraction of the soil that is both exposed to solar radiation and that is wetted
fw fraction of the soil surface wetted by irrigation and/or precipitation
h height of vegetation, m
n number of observations
p soil water depletion fraction for no stress
s standard deviation of the underlying ET population, mm d-1
t time, s, h, d
tl time length of the calculation period, h
uz horizontal wind speed at height z, m s-1
u2 horizontal wind speed at 2 m above the ground surface, m s-1
z elevation, m
zs depth of soil experiencing the change in water content, mm, m
zw height of anemometer above ground surface, m
α shortwave reflectance coefficient or albedo
β angle of the sun above the horizon, radians
γ γ = 0.000665 P, Pa °C-1
γpsy psychrometric constant for psychrometers, kPa °C-1
Δ slope of the saturation vapor pressure-temperature curve, de/dT, kPa °C-1
Δθ change in soil water content in the root zone during period of calculation
δ solar declination, radians
θ volumetric soil water content

θFC volumetric soil water content at field capacity
θWP volumetric soil water content at wilting point
ϕ station latitude, radians
Φ Stefan-Boltzmann constant, kJ m-2 h-1 K-4, MJ m-2 d-1 K-4
ω solar time angle at midpoint of hourly or shorter period, radians
ωs sunset hour angle, radians
ω1 solar hour angle at beginning of hourly or shorter period, radians
ω2 solar hour angle at end of hourly or shorter period, radians
REFERENCES
Aboukhaled, A., A. Alfaro, and M. Smith. 1982. Lysimeters. Irrigation and Drainage
Paper 39. Rome, Italy: FAO.
Allen, R. G. 1988. IRRiSKED: Irrigation scheduling program for demand and rotation
scheduling: Logan, Utah: Dept. Biological and Irrigation Engineering, Utah State Univ.
Allen, R. G. 1996a. Assessing integrity of weather data for use in reference evapotran-
spiration estimation. J. Irrig. Drain. Eng. 122(2):97-106.
Allen, R.G. 1996b. Nongrowing season evaporation in Northern Utah. In Proc. North
American Water and Environment Congress, ASCE. Reston, Va.: American Soc.
Civil Engineers.
Allen, R. G., C. E. Brockway, and J. L. Wright. 1983. Weather station siting and consump-
tive use estimates. J. Water Resources Plan. and Man. Div., ASCE 109(2): 134-146.
Allen, R. G., A. J. Clemmens, C. M. Burt, K. Solomon, and T. O’Halloran. 2005. Pre-
diction accuracy for project-wide evapotranspiration using crop coefficients and
reference evapotranspiration. J. Irrig. Drain. Eng. 131(1): 24-36.
Allen, R. G., and F. N. Gichuki. 1989. Effects of projected CO2-induced climate
changes on irrigation water requirements in the Great Plains states (Texas, Okla-
homa, Kansas and Nebraska). In The Potential Effects of Global Climate Change
on the United States: Appendix C-Agriculture. EPA-230-05-89-053. Washington,
D.C.: U.S. Environmental Protection Agency, Office of Policy, Planning and Eval.
Allen, R. G., T. A. Howell, W. O. Pruitt, I. A. Walter, and M. E. Jensen, eds. 1991a.
Lysimeters for Evapotranspiration and Environmental Measurements. New York,
N.Y.: American Soc. Civil Engineers.
Allen, R. G., M. E. Jensen, J. L. Wright and R. D. Burman. 1989. Operational esti-
mates of reference evapotranspiration. Agronomy J. 81: 650-662.
Allen, R. G., L. S. Pereira, D. Raes, and M. Smith. 1998.Crop Evapotranspiration:
Guidelines for Computing Crop Water Requirements. United Nations FAO, Irriga-
tion and Drainage Paper 56. Rome, Italy: FAO.
Allen, R. G., L. S. Pereira, M. Smith, D. Raes, and J. L. Wright. 2005a. FAO-56 dual
crop coefficient method for estimating evaporation from soil and application exten-
sions. J. Irrig. Drain. Eng. 131(1): 2-13.
Allen, R. G., W. O. Pruitt, J. A. Businger, L. J. Fritschen, M. E. Jensen, and F. H.
Quinn 1996. Chapt. 4: Evaporation and transpiration. In ASCE Handbook of Hy-
drology. Wootton et al., eds. New York, N.Y.: American Soc. Civil Engineers.
Allen, R. G., W. O. Pruitt, and M. E. Jensen. 1991b. Environmental requirements of
lysimeters. In Lysimeters for Evapotranspiration and Environmental Measure-
ments, 170-181. R. G. Allen, T. A. Howell, W. O. Pruitt, I. A. Walter, and M. E.

Jensen, eds. New York, N.Y.: American Soc. Civil Engineers.
Allen, R. G., W. O. Pruitt, D. Raes, M. Smith, and L. S. Pereira. 2005b. Estimating
evaporation from bare soil and the crop coefficient for the initial period using
common soils information. J. Irrig. Drain. Eng. 131(1): 14-23.
Allen, R. G., W. O. Pruitt, J. L. Wright, T. A. Howell, F. Ventura, R. Snyder, D. Iten-
fisu, P. Steduto, J. Berengena, J. Baselga Yrisarry, M. Smith, L. S. Pereira, D.
Raes, A. Perrier, I. Alves, I. Walter, and R. Elliott. 2006. A recommendation on
standardized surface resistance for hourly calculation of reference ETo by the
FAO56 Penman-Monteith method. Agric. Water Mgmt.
Allen, R.G. and C.W. Robison. 2007 Evapotranspiration and Irrigation Water Re-
quirements for Crops in Idaho. Research Completion Report, Univ. Idaho submit-
ted to Idaho Dept. Water Resources. 277 pp.
Allen, R. G., and J. L. Wright. 1983. Variation within the measured and estimated con-
sumptive use requirements. In Proc. of the 1983 Specialty Conference, Irrigation and
Drainage Division, 1-12. Jackson Hole, Wyo.: American Soc. Civil Engineers.
Allen, R. G., and J. L. Wright. 1997. Translating wind measurements from weather
stations to agricultural crops. J. Hydrol. Eng. ASCE 2(1): 26-35.
Allen, R. G., and J. L. Wright. 2007. Conversion of Kimberly, Idaho crop coefficients
to the ASCE standardized Penman-Monteith reference evapotranspiration basis. In-
house report, Univ. Idaho, Kimberly, Idaho.
ASCE-EWRI. 2005. The ASCE Standardized Reference Evapotranspiration Equation.
Report 0-7844-0805-X, ASCE Task Committee on Standardization of Reference
Evapotranspiration. Reston, Va.: American Soc. Civil Engineers.
Ayars, J. E., R. S. Johnson, C. J. Phene, T. J. Trout, D. A. Clark, and R. M. Mead.
2003. Water use by drip-irrigated late-season peaches. Irrig. Sci. 22: 187-194.
Battikhi, A. M., and R. W. Hill. 1986a. Irrigation scheduling and watermelon yield
model for the Jordan Valley. J. Agronomy Crop Sci. 157: 145-155.
Battikhi, A. M., and R. W. Hill. 1986b. Irrigation scheduling and cantaloupe yield
model for the Jordan Valley. Agric. Water Mgmt. 15: 177-187.
Blad, B. L., and N. J. Rosenberg. 1974. Lysimetric calibration of the Bowen-ratio en-
ergy balance method for evapotranspiration estimation in the Central Great plains.
J. App. Meteorol. 13(2): 227-236.
Blanc, M. L., H. Geslin, I. A. Holzberg, and B. Mason. 1963. Protection against frost
damage. Tech. Note No. 51. Geneva, Switzerland: World Meteorol. Org.
Bonachela, S., F. Orgaz, F. J. Villalobos, and E. Fereres. 2001. Soil evaporation from
drip-irrigated olive orchards. Irrig. Sci. 20(2): 65-71.
Borg, H., and D. W. Grimes. 1986. Depth development of roots with time: An empiri-
cal description. Trans. ASAE 29: 194-197.
Bosen, J. F. 1958. An approximation formula to compute relative humidity from dry
bulb and dew point temperatures. Monthly Weather Rev. 86(12): 486.
Brutsaert, W. H. 1982. Evaporation into the Atmosphere. Dordrecht, Holland: R. Dei-
del Publ. Co.
Brunt, D. 1932. Notes on radiation In the atmosphere: I. Quart. J. Roy. Meteorol. Soc.
58: 389-420.
Brunt, D. 1952. Physical and Dynamical Meteorology. 2nd ed. Cambridge, UK: Cam-
bridge Univ. Press.
Burman, R. D., P. R. Nixon, J. L. Wright, and W. O. Pruitt. 1980. Water requirements.

In Design and Operation of Farm Irrigation Systems, 189-232. M. E. Jensen, ed.
St. Joseph, Mich.: ASAE.
Burman, R. D., M. E. Jensen, and R. G. Allen. 1987. Thermodynamic factors in
evapotranspiration. In Proc. Irrig. and Drain. Spec. Conf., ASCE, 28-30. L. G.
James, and M. J. English, eds. Reston, Va.: American Soc. Civil Engineers.
Businger, J. A., M. Miyake, A. J. Dyer, and E. F. Bradley. 1967. On direct determina-
tion of turbulent heat flux near the ground. J. Appl. Meteorol. 6(6): 1025-1032.
Campbell, G. S., and B. D. Tanner. 1985. A krypton hygrometer for measurement of
atmospheric water vapor concentration. In Moisture and Humidity, 609-612. Re-
search Triangle Park, N.C.: ISA.
Costello, L. R., N. P. Matheny, and J. R. Clark. 2000. Estimating the irrigation water
needs of landscape plantings in California: Part 1: The landscape coefficient
method. Univ. Calif. Coop. Ext. and Calif. Dept. Water Resources.
Costello, L. R., and K. S. Jones. 1999. Estimating the irrigation water needs of land-
scape plantings in California. Part 2:WUCOLS III-Water use classification of land-
scape species.
Danuso F., M. Gani, and R. Giovanardi. 1995. Field water balance: BIdriCo 2. In
Crop Water models in Practice, 49-73. L. S. Pereira, B. J. Van den Broek, P. Ka-
bat, and R. G. Allen, eds. Wageningen, The Netherlands: Wageningen Press.
Dastane, N. G. 1974. Effective rainfall in irrigated agriculture. FAO Irrig. and Drain-
age Paper No. 25. Rome, Italy: United Nations Food and Agr. Organization.
de Azevedo, P. V., B. B. da Silva, and V. P. R. da Silva. 2003. Water requirements of
irrigated mango orchards in northeast Brazil. Agric. Water Man. 58: 241-254.
Dolman, A. J. 1993. A multiple-source land surface energy balance model for use in
general circulation models. Agric. For. Meteorol. 65: 21-45.
Doorenbos, J., and A. H. Kassam. 1979. Yield response to water. Irrig. and Drainage
Paper No. 33, (rev.) Rome, Italy: FAO.
Doorenbos, J., and W. O. Pruitt. 1977. Crop water requirements. Irrig. and Drainage
Paper No. 24, (rev.) Rome, Italy: FAO.
Droogers, P., and R. G. Allen. 2002. Estimating reference evapotranspiration under
inaccurate data conditions. Irrig. Drain. Systems (16): 33-45
Dyer, A. J. 1974. A review of flux-profile relationships. Boundary Layer Meterol. 7:
363-372.
Everson, D. O., M. Faubion, and D. E. Amos. 1978. Freezing temperatures and grow-
ing seasons in Idaho. Univ. Idaho Agric. Exp. Sta. Bull. 494.
Fereres, E. 1981. Drip irrigation management. Leaflet No. 21259. Berkeley, Calif.:
Cooperative Ext., Univ. California.
Gavilan, P., I. J. Lorite, S. Tornero, and J. Berengena. 2005. Regional calibration of
Hargreaves equation for estimating reference ET in a semiarid environment. Agric.
Water Mgmt. 81(3): 257–281.
Girona, J., M. Gelly, M. Mata, A. Arbones, J. Rufat, and J. Marsal. 2005. Peach tree
response to single and combined deficit irrigation regimes in deep soils. Agric.
Water Man. 72:97-108.
Grattan, S. R., W. Bowers, A. Dong, R. L. Snyder, J. J. Carroll, and W. George. 1998.
New crop coefficients estimate water use of vegetables, row crops. Calif. Agric.
52(1): 16-21.
Grebet, P., and R. H. Cuenca. 1991. History of lysimeter design and effects of envi-
ronmental disturbances. Lysimeters for Evapotranspiration and Environmental
Measurements, 10-18. R. G. Allen, T. A. Howell, W. O. Pruitt, I. A. Walter, and M.
E. Jensen, eds. New York, N.Y.: American Soc. Civil Engineers.
Griffin, R. E. 1976. Micro-climate control of deciduous fruit production with over-
head sprinklers. In Reducing Fruit Losses Caused by Low Spring Temperatures,
Appendix F. Document No. 10550101. Final report of Utah Agric. Exp. Sta. to
Four Corners Regional Commission. Logan, Utah: Utah State Univ.
Griffin, R. E., and E. A. Richardson. 1979. Sprinklers for micro climate cooling of bud
development. In Modfication of the Aerial Environment of Plants, 441-455. B. J.
Barfield, and J. Gerber, eds. St. Joseph, Mich.: ASAE.
Haddadin, S. H., and I. Ghawi. 1983. Effect of plastic mulches on soil water conservation
and soil temperature in field grown tomato in the Jordan Valley. Dirasat 13(8): 25-34.
Hargreaves, G. L., G. H. Hargreaves, and J. P. Riley. 1985. Agricultural benefits for
Senegal River Basin. J. Irrig. Drain. Eng. 111: 113-124.
Hargreaves, G. H., and Z. A. Samani. 1982. Estimating potential evapotranspiration.
Tech. Note. J. Irrig. Drain. Eng. 108(3): 225-230.
Hargreaves, G. H., and Z. A. Samani. 1985. Reference crop evapotranspiration from
temperature. Appl. Eng. Agric. 1(2): 96-99.
Hargreaves, G. H., and R. G. Allen. 2003. History and evaluation of the Hargreaves
evapotranspiration equation. J. Irrig. Drain. Eng. 129(1): 53-63.
Hatfield, J. L., J. J. Burke, J. R. Mahan, and D. F. Wanjura. 1987. Foliage temperature
measurements: A link between the biological and physical environment. In Proc.
International Conf. on Measurement of Soil and Plant Water Status, 2: 99-102.
Logan, Utah: Utah State Univ.
Hatfield, J. L., and M. Fuchs. 1990. Chapt. 3: Evapotranspiration models. In Manage-
ment of Farm Irrigation Systems. ASAE Monograph 33-59.St. Joseph, Mich.: ASAE.
Hawkins, R. H., A. T. Hjelmfelt, and A. W. Zevenbergen. 1985. Runoff probability,
storm depth, and curve numbers. J. Irrig. Drain. Eng. 111(4): 330-340.
Hernandez-Suarez, M. 1988. Modeling irrigation scheduling and its components and
optimization of water delivery scheduling with dynamic programming and stochas-
tic ETo data. Ph.D. diss. Davis, Calif.: Univ. California, Davis.
Howell, T. A., A. D. Schneider, and M. E. Jensen. 1991. History of lysimeter design
and use for evapotranspiration measurements. In Lysimeters for Evapotranspiration
and Environmental Measurements, 1-9. R. G. Allen, T. A. Howell, W. O. Pruitt, I.
A. Walter, and M. E. Jensen, eds. New York, N.Y: American Soc. Civil Engineers.
Huntingford, C., S. J. Allen, and R. J. Harding 1995, An intercomparsion of single and
dual-source vegetation-atmosphere transfer models applied to transpiration from
Sahelian savannah. Boundary-Layer Meteorology 74: 397-418.
Irmak, S., T. A. Howell, R. G. Allen, J. O. Payero, and D. L. Martin. 2005. Standard-
ized ASCE Penman-Monteith: Impact of sum-of-hourly vs. 24-hour timestep com-
putations at reference weather station sites. Trans. ASAE 48(3): 1063-1077.
Irrigation Association. 2003. Landscape Irrigation Scheduling and Water Manage-
ment: Practices Guidelines. Report by Water Management Committee. J. McCabe,
J. Ossa, R. G. Allen, B. Carleton, B. Carruthers, C. Corcos, T. A. Howell, R. Mar-
low, B. Mecham, and T. L. Spofford, eds.
Itenfisu, D. R. L. Elliott, R. G. Allen, and I. A. Walter. 2003. Comparison of reference
evapotranspiration calculations as a part of the ASCE standardization effort. J. Ir-

rig. Drain. Eng. 129(6): 440-448.
Jensen, M. E., ed. 1974. Consumptive use of water and irrigation water requirements.
Irrig. Drain. Div. Report, American Soc. Civil Engineers.
Jensen, M. E., R. D. Burman, R. G. Allen, eds. 1990. Evapotranspiration and Irrigation
Water Requirements. ASCE Manual 70. Reston, Va.: American Soc. Civil Engineers.
Johnson, R. S., L. E. Williams, J. E. Ayars, and T. J. Trout. 2005. Weighing lysimeters
aid study of water relations in tree and vine crops. Calif. Agric. 59(2):133-136.
Jones, J. W., K. J. Boote, S. S. Jagtap, G. G. Wilderson, G. Hoogenboom, and J. W.
Mishoe. 1987. SOYGRO V5.4 Technical Documentation. Gainesville, Fla.: Agric.
Eng. Dept. Res. Report, Univ. of Florida.
Jones, C. A., W. L. Bland, J. T. Ritchie, and J. R. Williams. 1991. Simulation of root
growth. In Modeling Plant and Soil Systems, 91-123. Monograph 31. J. Hanks and
J. T. Ritchie, eds. Madison, Wis.: American Soc. Agronomy.
Keller, J., and R. D. Bliesner. 1990. Sprinkle and Trickle Irrigation. New York, N.Y.:
van Nostrand Reinhold.
Kruse, E. G., and R. H. Haise. 1974. Water use by native grasses in high altitude Colorado
meadows. Western Region report ARS-W-6-1974. Washington, D.C.: USDA-ARS.
Ley, T. W., R. G. Allen, and R. W. Hill. 1996. Weather station siting effects on refer-
ence evapotranspiration. In Proc. ASAE International Conference on Evapotranspi-
ration and Irrigation Scheduling, 727-734. St. Joseph, Mich.: ASAE.
Liu Y., L. S. Pereira, and R. M. Fernando. 2006. Fluxes through the bottom boundary
of the root zone in silty soils: Parametric approaches to estimate groundwater con-
tribution and percolation. Agric. Water Man. 84: 27-40.
Martin, D., and J. Gilley. 1993. Chapt. 2: Irrigation water requirements. In National
Engineering Handbook, Part 623. Natural Resources Conservation Service.
Martınez-Cob, A., and M. Tejero-Juste. 2004. A wind-based qualitative calibration of
the Hargreaves ETo estimation equation in semiarid regions. Agric. Water Manage.
64: 251-264.
McQueen, J. 2004. Air Quality Forecast Model Evaluation Study during
ICARTT/NEAQS-2004, NOAA/NCEP: CMAQ-ETA: 301-763-8000 x7226. At:
www.al.noaa.gov/ICARTT/StudyCoordination/ICARTTdocs/Plan2k4_AQFM.pdf.
Meyer, W. S., and L. Mateos. 1990. Effects of soil type on soybean crop water use in
weighing lysimeters. II: Effect of lysimeter canopy height discontinuity on evapo-
ration. Irrig. Sci. 11: 233-237.
Munger, J. W., and H. W. Loescher. 2004. Guidelines for making eddy covariance
flux measurements. At: public.ornl.gov/ameriflux/measurement_standards_4.doc.
National Resouce Conservation Service. 1993. Part 623.0207: Effective precipitation.
In National Engineering Handbook, 2.142-2.154.
Neale, C. M. U., W. C. Bausch, and D. E. Heermann. 1989. Development of reflec-
tance based crop coefficients for corn. Trans. ASAE 32(6): 1891-1899.
Pastor, M., and F. Orgaz. 1994. Los programas de recorte de riego en olivar. Agricul-
tura 746: 768-776.
Patwardhan, A. S., J. L. Nieber, and E. L. Johns. 1990. Effective rainfall estimation
methods. J. Irrig. Drain. Eng. 116(2): 182-193.
Paço, T. A., M.I. Ferreira, and N. Conceição. 2006. Peach orchard evapotranspiration
in a sandy soil: Comparison between eddy covariance measurements and estimates
by the FAO 56 approach. Agric. Water Man. 85(3):305-313.

Payero, J. O., C. M. U. Neale, J. L. Wright, and R. G. Allen. 2003. Guidelines for
validating Bowen ratio data. Trans. ASAE 46(4): 1051-1060.
Penman, H. L. 1948. Natural evaporation from open water, bare soil and grass. Proc.
Roy.Soc. London A193: 120-146.
Penman, H. L. 1963. Vegetation and Hydrology. Tech. Comm. No. 53. Harpenden,
England: Commonwealth Bureau of Soils.
Pereira L.S., A. Perrier, R. G. Allen, and I. Alves. 1999. Evapotranspiration: Review
of concepts and future trends. J. Irrig. Drain. Eng. 125(2): 45-51.
Pruitt, W. O. 1986. Traditional methods: Evapotranspiration research priorities for the
next decade. ASAE Paper No. 86-2629. St. Joseph, Mich.: ASAE.
Pruitt, W. O. 1991. Development of crop coefficients using lysimeters. Lysimeters for
Evapotranspiration and Environmental Measurements, Proc. of the ASCE Int.
Symp. on Lysimetry, 182-190. R. G. Allen, T. A. Howell, W. O. Pruitt, I. A. Walter,
and M. E. Jensen, eds., New York, N.Y.: American Soc. Civil Engineers.
Pruitt, W. O., and F. J. Lourence. 1985. Experiences in lysimetry for ET and surface drag
measurements. In Advances in Evapotranspiration, 51-69. St. Joseph, Mich.: ASAE.
Pruitt, W. O., E. Fereres, P. E. Martin, H. Singh, D. W. Henderson, R. M. Hagan, E.
Tarantino, and B. Chandio. 1984. Microclimate, evapotranspiration, and water-use
efficiency for drip- and furrow-irrigated tomatoes. In Proc. 12th Congress, Interna-
tional Commission on Irrigation and Drainage, 367-394.
Pruitt, W. O., B. D. Swann, A. Held, B. Sutton, A. Matista, and T. C. Hsiao. 1987.
Bowen ratio and Penman: Australian-California tests. In Irrigation Systems for the
21st Century, Proc. of the ASCE Nat’l Conf. Irrig. and Drainage Engineering, 149-
158. L. G. James and M. J. English, eds. American Soc. Civil Engineers
Ritchie, J. T. 1972. Model for predicting evaporation from a row crop with incomplete
cover. Water Resources Res. 8: 1204-1213.
Ritchie, J. T., and D. S. NeSmith. 1991. Chapter 2: Temperature and crop develop-
ment. In Modeling Plant and Soil Systems, 5-29. R. J. Hanks and J. T. Ritchie, eds.
Agronomy Series No. 31. Madison, Wis.: American Soc. Agron.
Ritchie, J. T., and S. Otter. 1985. Description and performance of CERES-Wheat: A
user-oriented wheat yield model. In ARS Wheat Yield Project, 159-175. ARS-38.
Springfield, Va.: National Tech. Info. Serv.
Richie, W. E., and D. R. Pittenger. 2000. Mixed landscape irrigation research findings.
Proc. UCR Turfgrass and Landscape Management Research Conf., 12-13.
Rogers, J. S., L. H. Allen, and D. J. Calvert. 1983. Evapotranspiration for humid re-
gions: Developing citrus grove, grass cover. Trans. ASAE 26(6): 1778-83, 92.
Safadi, A. S. 1991. Squash and cucumber yield and water use models. Unpublished
Ph.D. diss. Logan, Utah: Dept. Biological and Irrigation Engineering, Utah State Univ.
Sammis, T. W., Andales, A., and L. Simmons. 2004. Adjustment of closed canopy
crop coefficients of pecans for open canopy orchards. 38th Western Irrigation Pe-
can Conference Proceedings, 117-119.
Sellers, W. D. 1965. Physical Climatology. Chigago, Ill.: Univ. Chicago Press.
Shaw, D. A. and D. R. Pittenger. 2004. Performance of landscape ornamentals given
irrigation treatments based on reference evapotranspiration. Acta Hort. 664: 607-613.
Shuttleworth, W. J., and J. S. Wallace. 1985. Evaporation from sparse crops: An en-
ergy combination theory. Quart. J. Roy. Meteorol. Soc. 111: 839-853.
Sinclair, T. R., L. H. Allen, and E. R. Lemon. 1975. An analysis of errors in the calcu-
lation of energy flux densities above vegetation by a Bowen ratio profile method.
Boundary-Layer Meteorol. 8: 129-139.
Smith, M., R. G. Allen, J. L. Monteith, A. Perrier, L. S. Pereira, and A. Segeren. 1991.
Report of the expert consultation on procedures for revision of FAO guidelines for
prediction of crop water requirements. Rome, Italy: UN-FAO.
Smith, M., R. G. Allen, and L. S. Pereira. 1996. Revised FAO methodology for crop
water requirements. In Proc. International Conf. on Evapotranspiration and Irriga-
tion Scheduling, 116-123. St. Joseph, Mich.: ASAE.
Snyder, R. L., J. P. De Melo-Abreu, and S. Matulich. 2005. Frost Protection: Funda-
mentals, Practice and Economics. Vol. 2, FAO Environment and Natural Re-
sources Service Series 10. Rome: FAO. 240 pp and 64 pp +CD-ROM.
Snyder, R. L,. and S. Eching. 2004. Landscape Irrigation Management Program—
IS005 Quick Answer. Univ. California, Davis. biomet.ucdavis.edu/irrigation_
scheduling/LIMP/limp.pdf.
Snyder, R. L., and S. Eching. 2005. Urban Landscape Evapotranspiration. California
State Water Plan, vol. 4, 691-693. www.waterplan.water.ca.gov/reference/index.
cfm#infrastructure
Snyder, R. L., and W. O. Pruitt. 1992. Evapotranspiration data management in Cali-
fornia. In Proc. of the Irrigation and Drainage sessions at ASCE Water Forum ’92,
128-133. T. Engman, ed. New York, N.Y.: American Soc. Civil Engineers.
Snyder, R. L., B. J. Lanini, D. A. Shaw, and W. O. Pruitt. 1989a. Using reference
evapotranspiration (ETo) and crop coefficients to estimate crop evapotranspiration
(ETc) for agronomic crops, grasses, and vegetable crops. Leaflet No. 21427. Berke-
ley, Calif.: Cooperative Extension, Univ. California.
Snyder, R. L., B. J. Lanini, D. A. Shaw, and W. O. Pruitt. 1989b. Using reference
evapotranspiration (ETo) and crop coefficients to estimate crop evapotranspiration
(ETc) for trees and vines. Leaflet No. 21428. Berkeley, Calif.: Cooperative Exten-
sion, Univ. California.
Soil Conservation Service. 1982. National Engineering Handbook. U.S. Govt. Print-
ing Office.
Sun, J., S. K. Esbensen, and L. Mahrt. 1995. Estimation of surface heat flux. J. Atmos.
Sci. 52: 3162-3171.
Swinbank, W. C. 1951. The measurement of vertical transfer of heat and water vapor
by eddies in the lower atmosphere. J. Meteorol. 8: 135-145.
Tanner, C. B. 1960. Energy balance approach to evapotranspiration from crops. Soil
Sci. Soc. America Proc. 24: 1-9.
Tanner, C. B. 1967. Measurement of evapotranspiration. In Irrigation of Agricultural
Lands, 534-574. R. M. Hagan, H. R. Haise, and T. W. Edminster, eds. Madison,
Wis.: American Soc. of Agron.
Tanner, B. D. 1988. Use requirements for Bowen ratio and eddy correlation determi-
nation of evapotranspiration. In Management of Irrigation and Drainage Systems:
An Integrated Approach, Proc. National Conference on Irrigation and Drainage
Engineering, 605-616 R. G. Allen, and C. M. Neale, eds. New York, N.Y.: Ameri-
can Soc. Civil Engineers.
Tasumi, M., R. G. Allen, R. Trezza, and J. L. Wright. 2005. Satellite-based energy
balance to assess within-population variance of crop coefficient curves. J. Irrig.
Drain. Eng. 131(1): 94-109.

Testi, L., F. J. Villalobos, and F. Orgaza. 2004. Evapotranspiration of a young irri-
gated olive orchard in southern Spain. Agric. and For. Meteor. 121(1,2):1-18.
Tetens, O. 1930. Uber einige meteorologische Begriffe. Z. Geophys. 6: 297-309.
Thornton, P. E., and S. W. Running. 1999. An improved algorithm for estimating
incident daily solar radiation from measurements of temperature, humidity, and
precipitation. Agric. and For. Meteor. 93: 211-228.
Twine, T. E., W.P. Kustas, J.M. Norman, D.R. Cook, P.R. Houser, T. P. Meyers, J. H.
Prueger, P. J. Starks, and M.L. Wesely. 2000. Correcting eddy-covariance flux un-
derestimates over a grassland. Agric. For. Meteor. 103: 279-300.
USDA-SCS. 1972. National Engineering Handbook, Section 4, Table 10.1.
USGS (U.S. Geological Survey). 1981. Guidelines for determining flood flow fre-
quency. Bulletin 17B. Reston, Va.: Office of Water Data Coordination, Hydrology
Subcommittee, Interagency Advisory Committee on Water Data.
Vanderkimpen, P. J. 1991. Estimation of crop evapotranspiration by means of the
Penman-Monteith equation. Unpublished Ph.D. diss. Logan, Utah: Dept. of Biol.
and Irrig. Engineering, Utah State Univ.
Vanderlinden, K., J. V. Giraldez, and M. Van Mervenne. 2004. Assessing reference
evapotranspiration by the Hargreaves method in Southern Spain. J. Irrig. Drain.
Eng. 129(1): 53-63.
Villalobos, F. J., F. Orgaz, L. Testi, and E. Fereres. 2000. Measurement and modeling
of evapotranspiration of olive orchards. European J. Agron. 13:155-163.
Wilson, E. B., and M. M. Hilferty. 1931. The distribution of Chi-square. Proc. Na-
tional Academy of Science 17(12): 684-688.
Wilson, K. et al. 2002. Energy balance closure at FLUXNET sites. Agric. For. Meteol.
113: 223-243.
Wolfe, J. W., P. B. Lombard, and M. Tabor. 1976. The effectiveness of a mist versus a
low pressure sprinkler system for bloom delay. Trans. ASAE 19(3): 510-513.
World Meteorological Organization. 1966. Measurement and estimation of evapora-
tion and evapotranspiration. Tech. Note No. 83. Geneva, Switzerland: WMO.
Wright, J. L. 1981. Crop coefficients for estimates of daily crop evapotranspiration. Ir-
rig. Scheduling for Water and Energy Conserv. in the 80’s. St. Joseph, Mich.: ASAE.
Wright, J. L. 1982. New evapotranspiration crop coefficients. J. Irrig. Drain. Div.,
ASCE 108: 57-74.
Wright, J. L. 1985. Lysimeter measurements of alfalfa evapotranspiration. Unpub-
lished data. Kimberly, Idaho: USDA-ARS.
Wright, J. L. 1988. Daily and seasonal evapotranspiration and yield of irrigated alfalfa
in southern Idaho. Agron. J. 80: 662-669.
Wright, J. L. 1990. Evapotranspiration data for dry, edible beans, sugar beets, and
sweet corn at Kimberly, Idaho. Unpublished data. Kimberly, Idaho: USDA-ARS.
Wright, J. L. 1995. Calibrating an ET procedure and deriving ET crop coefficients. In
Proc. of the 1995 Specialty Seminar. Sponsored by the Water Res. Comm., Ameri-
can Consulting Engin. Council of Colo. and the Colo. State Engineers Office.
Wright, J. L. 1996. Dormant Season Evaporation in Southern Idaho. Presentation at
the North American Water and Environment Congress, ASCE.
Wright, J. L., and M. E. Jensen. 1972. Peak water requirements of crops in southern
Idaho. J. Irrig. Drain. Div., ASCE 96(IR1): 193-201.
CHAPTER 9
DRAINAGE SYSTEMS
Larry G. King (Washington State University,
Pullman, Washington)
Lyman S. Willardson (Utah State University,
Logan, Utah)
Abstract. This chapter describes the basics of design, operation, and maintenance
of subsurface drainage systems as needed for sustained irrigated agriculture. Empha-
sis is on the control and management of the water table as needed for crop production
in harmony with environmental concerns. Two approaches to design of subsurface
drainage systems are based on steady state and transient flows. Steady state design of
drains is based on Hooghoudt’s equations as modified by Moody. Two examples are
presented in the form of spreadsheets. Transient flow design is based on procedures
developed by Dumm and Glover and others in the U.S. Bureau of Reclamation
(USBR). Spreadsheets are presented using the USBR transient approach for parallel
relief drains. Spreadsheets are also presented for calculating drain spacing using the
Hooghoudt equation and the USBR transient flow approach. Procedures are de-
scribed for determining design variables. Drainage materials, construction methods,
and equipment are described. The chapter concludes with brief summaries of system
operation, maintenance, and performance evaluation.
Keywords. Drainage systems, Drainage system design, Drainage system materials,
Environmental considerations, Hooghoudt equation, Leaching and salinity control,
Moody equations, Operation and maintenance, Steady state flow, Subsurface drainage
systems, Transient flow, Water table control.
9.1 INTRODUCTION
This chapter addresses agricultural drainage with emphasis on drainage under irri-
gated agriculture, in contrast to drainage of building foundations, dams, airport run-
ways, highway embankments, etc. Of course the principles can be applied to these
latter cases of drainage, but the focus will be on drainage of agricultural lands for crop
production under irrigation. Thus, more attention will be given to control and man-
agement of the water table in the soil profile than to removal of excess water from the
land surface.
Proper application of principles to drainage of irrigated land should not be in direct
conflict with the maintenance and/or restoration of wetlands, since naturally occurring
wetlands are not removed. In some cases, as within the Columbia Basin Project in
central Washington, large wetlands areas have been created as a result of an irrigation
project. Drainage of irrigated fields has provided water to supply wetlands within the
290 Chapter 9 Drainage Systems
project. However, engineers must be very careful with uses of drainage effluent due to
the quality of this water, as has been so dramatically demonstrated by the problems in
California’s Central Valley and Kesterson Wildlife Reserve.
It may appear paradoxical that after large sums are spent on infrastructure to deliver
irrigation water to farms, even larger sums must be spent to remove water through
drainage systems. It is true that improved irrigation systems and water management
may in many situations reduce the costs of drainage systems. However, even after
making practical improvements in irrigation water management at various scales,
drainage system construction will be necessary.
The goal of this chapter is to develop and discuss the basics of design, operation,
and maintenance of subsurface drainage systems to serve irrigated farms. The material
presented in this chapter provides scientifically based engineering methods for design
of these systems. Because the principals are the same for both humid and arid regions,
and because design is more critical in arid regions, the chapter will treat drainage de-
sign in arid regions in detail. Any important differences in design related to humid
areas will be pointed out as appropriate.
9.1.1 Definition of Drainage Systems
Agricultural drainage systems related to crop production are classified as surface
drainage systems or subsurface drainage systems depending on where the water caus-
ing the drainage problem exists. That is, the classification is based on the function of
the system not on the location of system components. Surface drainage systems re-
move excess water resulting from rainfall, snowmelt, and surplus irrigation from the
land surface. Subsurface drainage systems remove soil water to keep the water table
level below the bottom of the active root zone, to avoid waterlogging. Buried pipes
may be part of a surface or subsurface drainage system depending upon the source of
the water they are conveying. Likewise, an open ditch may function to control the wa-
ter table by removing soil water and be part of a subsurface drainage system, or it may
carry excess water from the soil surface as part of a surface drainage system. In arid
regions, subsurface drains are often deeper than in humid regions because of the need
for salinity control; deep subsurface drains make salinity control easier.
Surface drains may be shallow or deep and are normally unlined open ditches. Sub-
surface drains are deeper, and may be open ditches, but are usually buried perforated
pipes. Before the advent of corrugated plastic drain tubing, buried subsurface drains
were constructed of rigid clay tiles or concrete pipes with open joints between short
lengths of these pipes. The corrugated plastic drain tubing has perforations to admit
water and usually is surrounded by a graded gravel envelope. Many buried pipe drain-
age systems discharge by gravity directly into waterways or constructed open chan-
nels. Others discharge to a sump from which the water is pumped into a suitable open
channel or waterway.
Typically, drainage systems consist of laterals, collectors, and outlets. The water to
be removed by the system first enters the lateral. The water from several laterals enters
a collector. If the system is large we may also have sublaterals and subcollectors
where the water first enters the sublateral and sequentially goes to the lateral, subcol-
lector, and collector. Depending on system size, the collector or collectors convey the
water to the system outlet. The outlet is the discharge point or points of the system
where the water enters existing channels or waterways.
9.2 ENVIRONMENTAL CONSIDERATIONS

The time has long since past (if it ever existed) when an engineer can ignore the
environmental consequences of designs. This is as true in agricultural drainage as in
any other engineering design effort. This section briefly discusses some issues to be
considered, including quality of irrigation water, soil water, and drainage system ef-
fluent water; the quantity of drainage system effluent water; drainage in relation to
wetlands; and the reclamation of salinized soils. It is hoped that through discussion of
these environmental factors, the designers of agricultural drainage systems will con-
sider a holistic approach to the designs so that the systems work in harmony with the
environment rather than in conflict with other uses of the environment.
9.2.1 Irrigation Water Quality
Irrigation water quality has little to do with the actual physical design of either sur-
face or subsurface drainage systems. If irrigation water is of poor quality from a salin-
ity standpoint, the amount of water that must pass through the soil to control salinity
may be larger than when good quality water is used. The subsurface drainage water is
naturally of lower quality than the applied irrigation water; and, if irrigation water
quality is poor, the subsurface drainage water may need to be kept separated from the
water supply to other users. However, it may be possible to reuse most of the drainage
water for irrigation. Surface drainage waters can be reused without restriction unless
they happen to contain high concentrations of herbicides, pesticides, or pathogens that
may damage crops where the waters are used.
9.2.2 Soil Water Quality
For plants to grow unaffected by the quality of the water in the soil pores, some
proportion of the applied irrigation water must pass downward through the soil profile.
Table 7.4 in Chapter 7 lists the salt tolerance of agricultural crops. If the salinity of the
irrigation water is high, a leaching fraction must be added to the water applied in order
to keep the soil water salinity below the threshold values shown in the table and avoid
crop yield losses. See Chapter 7 for additional details.
9.2.3 Drainage System Effluent Quality
The quality of drainage system effluent is a function of (1) the quality of the water
entering the soil surface; (2) the timing and amounts of fertilizers and pesticides ap-
plied to the crop; (3) the natural materials in the soil; and (4) the proportion of the wa-
ter extracted by the plants for evapotranspiration. If the plants extract a high propor-
tion of the water from the soil profile, the salt concentration in the remaining water
will be increased significantly because most agronomic plants extract nearly pure wa-
ter and leave most of the salts in the soil. Highly water-soluble fertilizers and pesti-
cides that are not used or degraded by the soil environment will appear in the drainage
water. Some pesticides or their daughter products are molecularly bound to the soil
particles and do not remain in the percolating water. Nitrate nitrogen from fertilizer
applications may appear in drainwater if it is leached through the profile by excess
irrigation or rainfall before the fertilizer is consumed by the plants, and is often used
as a general indicator of pollution by other agents as well. If there is doubt about the
quality of drainage water, samples should be taken over time and analyzed to deter-
mine the level of dissolved constituents in the water. Some soils naturally contain
toxic elements (e.g., selenium) that may be mobilized if water is leached through the
soil.
9.2.4 Drainage System Effluent Quantity

Water discharged from subsurface agricultural drains must come from water that
infiltrated through the soil surface and water seeping into the drains from the water
table. In humid areas, the largest volume of drainage water comes from excess rainfall
that seeps into the soil and must be removed to prevent a detrimental rise in the water
table. At the time of peak discharge, drains in humid areas may discharge as much as
2.3 L/s per ha, or a depth of 20 mm/day, but are normally designed to carry about 0.8
L/s per ha (7 mm/day). In arid areas, drain discharge rates usually do not exceed 0.3
L/s per ha (3 mm/day). Drain discharge rates may be higher if the drains are discharg-
ing groundwater that moves into the area from higher lands or canal seepage.
9.2.5 Drainage and Wetlands
The basic purpose of agricultural drainage is to remove the excess water from the
soil surface and the soil profile to levels such that crops can be grown successfully.
Drainage of a land area will therefore change it from having a wet soil to having a soil
dry enough to be cultivated. This can create some interesting economic, institutional,
and social issues.
If an area has been wet for long periods and cannot be cultivated, it will develop an
ecology that is typical of a “wetland” and may have a microenvironment that serves as
habitat for various species of plants and animals. If biological studies show that the
wet area serves a public good, it may be mandated that it cannot be drained. A nearby
wetland of similar size can sometimes be substituted, allowing artificial drainage of a
troublesome wet area in a field to improve the efficiency of the farming operation.
In the U.S., any drainage of an area that could potentially be declared a wetland
must be approved according to the procedures and regulations in Section 404 of the
U.S. Clean Water Act. To avoid federal penalties, USDA Natural Resource Conserva-
tion Service specialists or members of the U.S. Army Corps of Engineers should be
contacted for clearance before any construction takes place.
9.2.6 Reclamation of Salinized Soils
Soils become salinized in arid areas because the water table is too close to the sur-
face. Water from a shallow water table can move easily to the soil surface where it
evaporates. The salt carried by the water is therefore deposited on or near the soil sur-
face. When the salt content of the soil becomes too high, crop seeds will not germinate
and the soil itself may be physically damaged by the salt.
The first step in reclaiming a salinized soil is to take samples for analysis to deter-
mine whether drainage alone will remedy the problem or whether it may be necessary
to add some source of soluble calcium, such as gypsum, to aid in the reclamation
process. If the soil does not need extra calcium, drains can be installed to lower the
water table to an appropriate depth. For arid regions, drains should be installed at a
depth of at least 2 m to prevent resalination of the soil following reclamation. When
the drains are installed and functioning, the soil can be irrigated a few times, in the
absence of a crop, to leach the salts out of the surface soils. Once the salt content of
the surface soil has been lowered sufficiently that crop seeds can be germinated, salt-
tolerant crops can be grown economically. Until the salt content of the surface soil is
lowered to a satisfactory amount, overirrigation should be practiced. In some cases, it
may be desirable to install the drainage system and then pond the soil surface to
quickly leach the soil profile. Refer to Section 7.6.4 to estimate the amount of water
that will be needed for reclamation. Once crops are growing, uniform irrigations will
eventually remove the excess salt from the profile.
9.3 DRAINAGE SYSTEM REQUIREMENTS
In Pakistan, the people appropriately refer to the “double menace” of waterlogging
and salinity when describing drainage problems. Thus, drainage systems in irrigated
areas should provide control of the water table and allow for leaching water to control
soil salinity in the crop root zone.
In addition to control of the water table and soil salinity, successful agriculture re-
quires appropriate timing of certain critical field operations. Where natural drainage is
inadequate, artificial drains are needed to remove excess soil water at times during the
year when the soil is generally too wet for these operations. This may apply to entire
farms or to only a small area of a large field. In the latter case, localized drains may be
installed in a portion of the field to overcome the lack of natural drainage or to inter-
cept seepage from a canal or other sources.
Before proceeding to the details of drainage system design approaches in Section
9.4, it is helpful to consider two important aspects of the need for constructed drainage
systems in irrigated agriculture.
9.3.1 Water Table Control
Plants grow best when the water table is kept at or below the bottom of the crop
root zone. For control of excess water in the root zone, subsurface drains are generally
designed to keep the water table midway between adjacent drains from getting closer
to the surface than 1 m. (In organic soils, the water table should be maintained at ap-
proximately 0.5 m deep to prevent excess oxidation of the organic soil materials.) The
depth and spacing of the subsurface drains determines the midpoint water table depth
at the design discharge rate. In humid areas the drains are commonly installed at
depths of 1.0 to 1.5 m, while in arid areas drains are typically installed at depths of 2
m or more. Drain spacing in arid areas may be 4 to 5 times greater than in humid areas
because there is less excess water to be removed.
9.3.2 Leaching and Salinity Control
If drains are installed at a 2-m depth or deeper in an arid area and if adequate
amounts of irrigation water are applied in the course of normal cultivation of a crop,
salinity will be controlled and adequate leaching will occur to prevent soil salination.
Drains do not need to be made larger for reclamation of saline soils because all
movement of water is downward in the soil profile. If the water temporarily exceeds
the carrying capacity of the drains, the flow of water through the soil will be slower
and the reclamation may be more effective since the salt in the soil has more time to
dissolve in the leaching water.
9.4 APPROACHES TO DESIGN OF
SUBSURFACE DRAINAGE SYSTEMS
There are two general approaches for calculating drain spacing for a subsurface
drainage system, the steady state and the transient approaches. In a steady state de-
sign, the water table is assumed to be maintained at a constant level by a continuous
slow uniform recharge of water through the soil surface. A steady state design theory
was developed by Hooghoudt (1940) in the Netherlands, where rainfall is slow and
occurs over a long period of time. When recharge stops, the water level between
drains goes down until it reaches the level of the drains. The transient design theory
was developed by the U.S. Bureau of Reclamation (USBR, 1993) to more closely rep-
resent the periodic recharge to the groundwater that occurs under irrigated conditions.
In a transient design, the water table is assumed to rise on the date of irrigation and to
fall continuously until the next irrigation or recharge event. In the real world, water
tables fluctuate between drains regardless of whether a steady state or a transient
method was used to determine spacing.
Constructed drainage systems can be classified according to the nature of the water
source causing the drainage problem and the protection offered by the drainage sys-
tem. If the source of water causing the drainage problem originates within the area
protected by the drainage system, the system is called a relief drainage system. The
drains in this case are referred to as relief drains. The spacing of parallel relief drains
may be determined by either the steady state or transient design approach (See Section
9.4.1). If the water originates outside the area being protected by the drains, then the
drains intercept this water and are called interceptor drains. Interceptor drains are a
special steady state design case and are treated in Section 9.4.2.
The plan view geometry (or layout) of drainage systems can be modified to meet
special physical conditions. NRCS (2001) shows various layouts for field drains that
describe their appearance on a map: random, herringbone, or parallel (see also USBR,
1993).
9.4.1 Relief Drains
Drains installed in a field to cause a general lowering of the water table are called
relief drains. They effectively “relieve” the high water table. As an alternative to relief
drains, networks of small wells can be installed in an area with a high water table for
the purpose of lowering the water table. They are called relief wells and have the same
effect as relief drains installed horizontally below the soil surface. If the groundwater
is of acceptable quality, it can be used directly for irrigation or even municipal use. If
the groundwater is salty, horizontal relief drains at a relatively close spacing are pre-
ferred over wells or deep drains at wide spacing to minimize the amount of deep saline
groundwater removed.
9.4.1.1 Steady flow. Drainage systems in both humid and arid areas can be de-
signed using steady state theory. When the appropriate design parameters are selected,
the proper design will result. The amount of water to be removed by the drainage sys-
tem must be determined. This amount is described as a volume per unit area per unit
of time and is called the drainage coefficient. The units of the drainage coefficient are
usually expressed in mm/day or m/day, and must be the same as the units of hydraulic
conductivity of the soil that are used in the design calculations. In humid areas, the
drainage coefficient is based on local experience and is approximately 7 mm/day. The
assumption made is that there will be 7 mm/day of excess water from rainfall that will
enter the soil profile and that must be removed by the drainage system to prevent the
water table from rising closer than 1 m to the soil surface. In arid areas, a steady state
drainage coefficient is computed from information about the average peak rate of
evapotranspiration (ET) and the fraction of the applied irrigation water that is not con-
sumed by the crop. The fraction of the irrigation water that infiltrates into the soil, but
is not consumed by the crop, is related to the irrigation efficiency or water application
efficiency. Any surface runoff water from the irrigation should not be included in the
drainage coefficient or the drain spacing will be smaller than necessary. If salinity
control is required, a leaching fraction must be added to the drainage coefficient.
Equation 9.1 can be used to determine the drainage coefficient for an arid area:
q = ET [1 – SF(1 – LF)]/SF (9.1)
where q is the drainage coefficient expressed in the same units as ET, ET is expressed
as a rate of use per day or as a depth of water consumed over a specific period of time,
LF is the leaching fraction (see Chapter 7), and SF is the fraction of the infiltrated irri-
gation water that is stored in the root zone. The stored fraction (SF) of the irrigation
water can be calculated from data for a typical irrigation as
SF = ET/[IW(1 – RF) – ET(LF)] (9.2)
where IW is the gross amount of water applied, ET is the amount of water consumed
by the plants since the last recharge event, and RF is the fraction of the irrigation water
that runs off. The stored fraction SF has also been called irrigation efficiency (of the
infiltrated water) or water application efficiency (not including runoff).
A typical value for the drainage coefficient for an arid area is from 1 to 3 mm/day.
If the equation gives a higher value, the wrong efficiency is being used. The leaching
fraction used in the drainage coefficient equation will vary from 0.0 to 0.2 depending
on the salt concentration of the irrigation water (see Chapter 7) but will seldom exceed
0.05. Improper irrigation practices that result in poor uniformity (low efficiency)
should not be assumed to satisfy the annual leaching requirement. Calculation or se-
lection of a steady flow drainage coefficient provides part of the necessary input to
steady state drainage system design.
The steady state design equation used here was developed for an idealized flow
system for two parallel buried pipe drains taken from the middle of a larger set of par-
allel buried pipe drains. The drains are at some distance D above a horizontal imper-
meable barrier. This idealized flow system is shown in Figure 9.1. It is assumed in
Figure 9.1 that the ground surface and the impermeable barrier are horizontal and the
two drains are at the same depth. It is customary to assume that the drains are half full
(i.e., the design depth for the drains is to the drain pipe centerline.)
Figure 9.1. Idealized flow system for steady state buried pipe drainage to two drains
taken from a set of parallel drains at a distance D above an impermeable barrier.
Using the Dupuit-Forchheimer (D-F) assumptions, Hooghoudt (1940) developed an

equation of the form:
S = (4h / q )( K1h + 2 K 2 De ) (9.3)
where S = drain spacing (in the horizontal direction)
q = drainage coefficient
h = height of the water table above the drain pipe centerline midway between
the two drains
K1, = hydraulic conductivity values for the soil above the drains
K2 = hydraulic conductivity values for the soil below the drains
De is the Hooghoudt equivalent depth coefficient, which is explained below.
Note from Figure 9.1 and Equation 9.3 that the Hooghoudt equation applies di-
rectly to the special layered soil case where the drains lie at the interface between the
two soil layers having different values of hydraulic conductivity. Also note that for K1
= K2, the soil is homogeneous and the equation also applies with K1 = K2 = K. For the
case when K2 << K1 and the soil below the centerline of the drains may be considered
the impermeable barrier, Equation 9.3 still applies with K2 = 0 and K1 = K (also D =
0). Thus, for this case of drains on the barrier, Equation 9.3 reduces to S = (4Kh2/q)1/2
from which S may be calculated directly from value of K, h, and q. Steady state is as-
sumed wherein a constant rate recharge, q (volume per unit area per unit time), is re-
moved by the drains. The recharge is uniformly distributed (uniform rainfall) as it en-
ters the saturated groundwater zone. Also, in the derivation of Equation 9.3, it has
been assumed that the depth of water in the drains is small relative to h.
Hooghoudt recognized that the assumption of horizontal flow could be a serious
limitation in the use of Equation 9.3, especially for larger values of the ratio D/S.
Thus, he developed the concept of equivalent depth, De. For all calculations, De should
be used instead of the actual depth from the drains to the impermeable barrier, D.
The Hooghoudt equivalent depth, De, is defined as a fictitious depth to a fictitious
impermeable barrier such that if the spacing S is computed with Equation 9.3, i.e.,
using De instead of D, the result would be the same as using the actual D in a more
accurate theory. Perhaps this concept is more easily understood by using the graphics
of Figure 9.2. The real situation with buried pipe drains at a distance D above the bar-
rier is replaced with the theoretical model of fully penetrating ditch drains but with
depth of water in the ditch drain of De, the Hooghoudt equivalent depth.
Figure 9.2. A graphical explanation of the concept of Hooghoudt equivalent depth, De,
as depicted by the difference between the real situation and the theoretical model.
Note that De always is either less than or equal to D. The extra resistance of radial
flow near the drains is modeled by a shallower total system of the same hydraulic con-
ductivity. In Figure 9.2, note that the hydraulic conductivity is the same in both the
real situation and the theoretical model. Of course there may be different values for
K1, above the drains and K2 below the drains or for homogeneous soils, K1 = K2 = K.
Also, in Figure 9.2 the values of S and h in the real situation are exactly the same as S
and h in the theoretical model.
The concept of Hooghoudt equivalent depth is widely accepted as a reasonable way
of using the D-F theory to calculate spacing of parallel drains regardless of whether
steady state or transient drain spacing theories are applied. Note that the concept is
based upon equivalent overall system performance and not upon whether the water
flow is steady or transient.
Moody (1966) presented equations to relate the Hooghoudt equivalent depth to the
actual depth of the system, the drain radius, and the drain spacing. In the nomenclature
used above, his equations are:
De = D /{1+ (D/S) [(8/π )ln(D/r) – α ]} for 0 ≤ D/S ≤ 0.31 (9.4)
where α = 3.55 − 1.6( D / S ) + 2( D / S ) 2 (9.5)

and r is the drain radius. Moody noted that for practical purposes, α = 3.4 can be used
with little error. When computers are used to perform the calculations the use of Equa-
tion 9.5 to determine α is recommended. For the case where D/S is larger than 0.31,
the Moody equation is
De = S / {(8 / π )[ln(S / r ) − 1.15] } for D/S > 0.31 (9.6)
Charts to determine Hooghoudt equivalent depth have been prepared and presented
in the literature by various authors. We prefer the Moody equations especially for ease
of use with computers.
By inspection of the Hooghoudt and Moody equations, it is easily seen that the cal-
culation of the proper drain spacing for relief drains is an iterative (or trial and error)
procedure because S depends on De and De depends upon S. In fact, the whole design
procedure is iterative. Also note that the depth of the drains below ground surface is
not expressly included in the Hooghoudt equation (Figure 9.1, Equation 9.3). The it-
erative design procedure starts with a depth of root zone appropriate for the crop to be
grown. If the designer is inexperienced, calculations for all crops that may be grown
should be made and then the most critical one should be chosen as the design crop.
Once the depth of the root zone, Drz, has been determined, a trial drain depth is cho-
sen. See Section 9.5.2 for discussion of proper drain depths. Since the drain depth is
the sum of Drz and h, the choice of the trial drain depth determines the design value for
h. Once the depth from ground surface to the impermeable barrier (see Section 9.5.3)
is known, D has also been determined. Next, the designer must choose a trial diameter
of drain pipe from which the drain radius r is determined. The effective drain radius is
half the drain diameter plus the thickness of gravel envelope surrounding the drain
pipe. After the drain spacing has been calculated, the drain discharge must be calcu-
lated and compared to the pipe capacity to see if this trial drain diameter is satisfac-
tory.
A B C D E F G H
1 THIS SPREADSHEET USES HOOGHOUDT'S EQUATION FILE NAME:
2 FOR STEADY FLOW WHEN 0 < D/S < 0.31 or D/S = 0.31 HODESNL
3 FOR LAYERED SOILS
4 INSTRUCTIONS for use of this spreadsheet to calculate drain spacing:
5 1. Enter the proper values for the input data into CELLS A14 through A19
6 2. If you wish to see the effect of a new trial drain spacing, change the first value of S
7 (in CELL A21)
8 3. Accept the last value of S in the first column as the correct drain spacing ONLY if:
9 (a) this last value is equal to the immediately preceding value of S ,
10 AND
11 (b) D/S lies in the range 0 < D/S < 0.31 or D/S = 0.31
12 4. If D/S > 0.31, then use the spreadsheet with file name HODSN31L
13 INPUT DATA:
14 0.75 m/day = K1 (hydraulic conductivity above drains)
15 0.75 m/day = K2 (hydraulic conductivity below drains)
16 0.0025 m/day = q (drainage coefficient)
17 0.75 m = h (water table height above drains at mid-point)
18 0.15 m = r (effective drain radius)
19 2.5 m = D (depth of barrier below drains)
20 S (m) D/S D e (m) S 2 (m2)
21 100.00 0.03 2.29 4798
22 69.27 0.04 2.21 4649
23 68.18 0.04 2.20 4641
24 68.13 0.04 2.20 4641
25 68.12 0.04 2.20 4641
26 68.12 0.04 2.20 4641
Figure 9.3. Spreadsheet using the Hooghoudt and Moody equations
to calculate steady state relief drain spacing for 0 < D/S ≤ 0.31.
With the foregoing variables determined and knowing the soil hydraulic conductiv-
ity K (or K1 and K2) and the drainage requirement q, the designer is ready to use the
Hooghoudt and Moody equations to calculate the drain spacing S. While the iterative
solution of these equations can be done with a hand calculator, a computer spreadsheet
is more useful, especially for the inexperienced designer. Two such spreadsheets are
given as Figures 9.3 and 9.4. Figure 9.3 is the one used most often and is for 0 < D/S <
0.31. Figure 9.4 applies when D/S > 0.31.
Instructions for use of the spreadsheets are included within the spreadsheets. Use of
the spreadsheets will be discussed first and then details of how they are constructed
will be given. The known or already determined values of the input variables are
placed into cells A14 through A19. Next a trial value for drain spacing S is entered
into cell A21. The spreadsheet then calculates the values row by row as shown. The
correct drain spacing is given as the last value in column A if the last value of S is
identical to the next-to-last value of S immediately above and the values of D/S lie in
the range 0 < D/S ≤ 0.31. If D/S > 0.31, then the spreadsheet of Figure 9.4 must be
used.
The spreadsheets are constructed as follows. In Figure 9.3, the first trial value of
drain spacing is entered into cell A21. Then the value of D from cell A19 is divided by
A B C D E F G H
2 FOR STEADY FLOW WHEN D/S > 0.31 HODSN31L
3 FOR LAYERED SOILS
7 (in CELL A22)
8 3. Accept the last value of S in the first column as the correct drain spacing ONLY if::
10 AND
11 (b) D/S lies in the range D/S > 0.31
12 4. If 0 < D/S < 0.31 or D/S = 0.31, then use the spreadsheet with file name HODESNL
13 INPUT DATA:
19 50 m = D (depth of barrier below drains)
20 S (m) D/S D e (m) S 2 (m2)
21 200.00 0.25 12.99 24060
22 155.11 0.32 10.52 19607
23 140.03 0.36 9.67 18073
24 134.44 0.37 9.35 17499
25 132.29 0.38 9.22 17278
26 131.44 0.38 9.18 17191
27 131.11 0.38 9.16 17157
28 130.98 0.38 9.15 17143
29 130.93 0.38 9.15 17138
30 130.91 0.38 9.14 17136
31 130.90 0.38 9.14 17135
32 130.90 0.38 9.14 17134
Figure 9.4. Spreadsheet using the Hooghoudt and Moody equations
to calculate steady state relief drain spacing for D/S > 0.31.
this trial S to give D/S in cell B21. In cell C21, α is calculated from Equation 9.5, and
then De is calculated from Equation 9.4. Next S2 is calculated from Equation 9.3 in cell
D21. The square root of this last value gives the new trial value of S in cell A22 and
the process is repeated for row 22, etc. The construction and use of the spreadsheet of
Figure 9.4 is entirely parallel to that of Figure 9.3. Equation 9.6 is used to calculate De
in Figure 9.4, whereas Equations 9.4 and 9.5 are used for this calculation in Figure 9.3.
9.4.1.2 Steady state relief drain design examples. The spreadsheets of Figure 9.3
and 9.4 also show examples of the drain design process. In Figure 9.3, suppose that the
root zone depth to be protected by the drainage system is 1.25 m and the drains are to
be placed 2 m deep. Then h is 0.75 m. Suppose the drain pipe is to be 10 cm diameter
and the gravel envelope surrounding the drain pipe is 10 cm thick, thus the effective
drain radius r is 0.15 m. Also, the drainage coefficient is 2.5 mm/day, the soil is ho-
mogeneous with hydraulic conductivity of 0.75 m/day, and the impermeable barrier is
4.5 m below ground surface. For these conditions, the input variables are as given in
cells A14 through A19 of Figure 9.3. If a trial drain spacing of 100 m is chosen, Fig-
ure 9.3 shows the calculations of this example resulting in a drain spacing of 68 m
(rounded to the nearest meter). After constructing the spreadsheet, a designer can eas-
ily show that the procedure converges rapidly to the same drain spacing regardless of
the value chosen for the first trial drain spacing. A first trial value of 10 m will pro-
duce the same result for this example.
Assuming that the drain depth and the root zone depth of this example are satisfac-
tory, all that remains is to determine if the drain pipe diameter of 10 cm is adequate.
Since the flow to the drains is steady, the drain discharge per meter length is equal to
the product of the drainage coefficient q and the drain spacing S. Thus for this exam-
ple, the drain discharge is calculated as Qm = qS = 0.0025(68) = 0.170 m3/day per m of
drain. Now if the drain laterals are 300 m long, the discharge from the lateral QL is
51.0 m3/day or 5.9×10-4 m3/s. Using a Manning n value of 0.017 for corrugated plastic
drain tubing, the capacity of the drain on a grade of 0.001 is calculated to be 1.25×10-3
m3/s from the Manning equation:
Q = VA = (1/n) AR2/3 Sf 1/2 (9.7)
where V = velocity
A = area
R = hydraulic radius
Sf = friction slope
n = the Manning roughness coefficient.
Since the capacity of the lateral (1.25×10-3 m3/s) exceeds the calculated discharge
(5.9×10-4 m3/s), the drain pipe diameter of 10 cm is adequate. If the discharge com-
puted with the Manning equation is less than the discharge required from the lateral,
choose a larger drain diameter and repeat the spreadsheet calculations with the larger
trial drain diameter.
The example shown in Figure 9.4 is for a very deep impermeable barrier. All other
values are as given in the previous example. The drains must handle a discharge of
1.14×10-3 m3/s and the 10 cm drain diameter is still adequate for the 300-m drains on a
grade of 0.001 m/m.
9.4.1.3 Transient flow. Transient drain spacing designs require the translation of
an amount of recharge water entering the soil into a rise of the water table. When an
irrigation occurs, no water will reach the water table and cause it to rise until the com-
plete root zone of the plants has been restored to field capacity. Therefore, the amount
of water required to refill the root zone is equal to the amount of water that has been
consumed by the crop since the last recharge event. If the water added to the soil does
not refill the soil profile to replace the water consumed by the plants, the water table
will not rise in response to an irrigation. The amount of the irrigation water entering
the soil, as well as the storage space in the soil is expressed as a volume of water per
unit area of soil, which is equivalent to a depth of water. When the ET since the last
recharge event, expressed as a depth of water, is subtracted from the average depth of
irrigation water infiltrating the soil (volume per unit area excluding any runoff), the
difference will be a depth of water that will cause the water table to rise. The actual
water table rise that will occur is the depth of recharge water divided by the specific
yield of the soil. Specific yield is explained in Section 9.5.6, where it is called drain-
able porosity.
The assumption made for transient flow designs is that any recharge causes an in-
stantaneous rise of the water table on the day of recharge. Periods of one day are used
in the computations. The day of recharge is the last day of the discharge from the pre-
vious recharge event and the zero day for computation for the rate of fall of the water
table until the next recharge event.
The transient design procedure presented herein follows closely that developed by
Dumm (1968) and others of the U.S. Bureau of Reclamation (USBR, 1993). This de-
sign procedure is based upon a concept of dynamic equilibrium wherein the annual
cycle of water table fluctuations is relatively consistent from year to year when ade-
quate drainage is achieved for proper water table levels for the crops grown. In some
cases this dynamic equilibrium is reached without construction of buried pipe drains,
but in many situations natural drainage is inadequate to provide dynamic equilibrium
and subsurface drains are required.
Glover and Dumm (USBR, 1993), using heat flow analogies which neglect conver-
gence of flow toward the drains, presented an equation with the initial water table
given by a fourth degree parabola:
y(x,0) = (8/S4) [y(S/2,0)] ( S3x – 3S2x2 +4 Sx3 – 2x4) for 0 < x < S (9.8)
where the coordinates y(x,z) have their origin at the centerline of one of the drains
such that y = 0 when z = 0, y is the water table height above the centerlines of two par-
allel relief drains taken from the middle of a larger set of parallel relief drains as was
done for steady state relief drains (see Section 9.4.1), and S is the drain spacing. By
truncating the resulting infinite series solution and using only the first term of the se-
ries, an approximate equation results:
h/h0 = (36.37/π3) exp(–π K D t/[fS2]) (9.9)
where D = the time averaged flow depth for any drainout period assumed to be given
as D = D + y0/2 at any location x (at the midpoint between the two drains,
D = D + h0/2)
h0 = height of the water table above the drains midway between the two drains
at the start of a water table recession or drainout period
h = height of the water table above the drains midway between the two drains at
any time t after the start of a drainout period
K = soil hydraulic conductivity
f = specific yield
S = spacing between the two parallel drains.
Solving Equation 9.9 for the drain spacing gives:
1/ 2
{ }
S = π KDt / ⎡⎣ f ln(36.37h0 /(π 3 h)) ⎤⎦ (9.10)
Van Schilfgaarde (1965) stated that correction for the convergence effect can be
made by applying Hooghoudt’s equivalent depth concept. There is nothing inherent in
the concept of equivalent depth that restricts its application to steady flow. Thus, it can
be expected to apply to spacing calculations based upon various transient theories that
neglect convergence of flow toward the drains. The Moody (1966) equations for De
(Equations 9.4, 9.5, and 9.6) will be used here. Thus, replacing D with De to correct
for convergence of flow toward the drains, Equation 9.10 becomes
1/ 2
{ ⎣ (
S = π KDe t / ⎡⎢ f ln 36.37h0 / π 3 / h ⎤⎥
( ⎦ )) } (9.11)
The equation for drain spacing if the drains sit on an impermeable barrier is
(Dumm, 1968):
1/ 2
S = {9 Kh0t / [ 2 f (h0 / h) − 1]} (9.12)
Dumm (1968) presented a summary of the USBR approach for designing pipe
drain systems using a transient equation. He reported that Australian and Canadian
research workers and a USBR project provided data for checking the validity of the
mathematical developments of the transient flow theory under various field conditions.
The reader should refer to Tables 1 and 2 and Figures 2, 3, and 4 of the paper (Dumm,
1968) for these comparisons.
The concept of dynamic equilibrium as it relates to drainage is illustrated by Figure
9.5 of Dumm (1968) and by Figure 5-3 of the USBR Drainage Manual (USBR, 1993).
In addition to the earlier discussion of the dynamic equilibrium concept, the main ele-
ments of this concept are explained as follows. When recharge and discharge become
equal, the highest level and the range in cyclic annual fluctuation of the water table
become reasonably constant from year to year. The water table rises during the irriga-
tion season and reaches its highest elevation after the last irrigation. When year-around
cropping is practiced, this usually occurs at the end of the peak portion of the irriga-
tion season.
Dumm (1968), gave equations for the transient drain discharge. Replacing D with
De in the equation for drains above the barrier gives:
Q = 2 πK D e / S (9.13)
and for drains on the barrier:
Q = 4 Kh 2 / S
(9.14)
3
in which Q is the drain discharge in m /day per lineal meter of drain, K is in m/day,
and De , S, and h are in m. Water balance calculations were shown in Tables 5 and 6
of Dumm (1968) by averaging the beginning and end of each drainout period. Note
that Q is a discharge rate and is a function of time because h is a function of time. See
comparisons in Figures 9, 10, 11, and 12 of Dumm (1968).
Information needed for the USBR approach include estimates of deep percolation
losses for each irrigation, which is given in Table 9.1, and methods for estimating spe-
cific yield, f. In estimating the specific yield, hysteresis is ignored in the water content
as a function of capillary pressure. The USBR approach is to estimate specific yield
from the hydraulic conductivity determined by the auger hole method. Figure 9.5
shows a curve adapted from USBR (1993) for obtaining the specific yield.
9.4.1.4 Transient relief drain design example. Perhaps the best way to explain
the design process for relief drains using transient flow theories is by a detailed exam-
ple. Some of the steps involved are the same as for steady state design of relief drains.
First, choose a trial drain depth. Next, from the root zone depth to be protected and the
trial drain depth, determine the maximum allowable height of the water table above
the drains midway between the drains (maximum h or h0). Then, from the trial drain
Table 9.1. Approximate deep percolation loss below the root zone for use
in drain spacing calculations (adapted from Table 3 of Dumm, 1968).[a]
On Basis of Texture
Deep Percolation Deep Percolation
Texture Percentage[b] Texture Percentage[b]
Loamy sand 30 Sandy clay loam 14
Sandy loam 26 Clay loam 10
Loam 22 Silty clay loam 6
Silt loam 18 Sandy clay, and clay 6
On Basis of Infiltration Rate
Infiltration Rate Deep Percolation Infiltration Rate Deep Percolation
(mm/hr) Percentage[b] (mm/hr) Percentage[b]
1.27 3 25.40 20
2.54 5 31.75 22
5.08 8 38.10 24
7.62 10 50.80 28
10.16 12 63.50 31
12.70 14 76.20 33
15.24 16 101.60 37
20.32 18
[a]
These deep percolation percentages are usually adequate to provide for effective leaching to maintain
salt balance in the root zone. However, if the leaching requirement is larger than the percentage given in
the table, the larger value should be used in drain spacing calculations.
[b]
Percentage is based on the application or net input.
Figure 9.5. Curve used by USBR to estimate specific yield from hydraulic conductivity
(adapted from Figure 2-4, USBR, 1993).
Table 9.2. Seasonal ET of sugar beets used for the design example.
Period ET (mm/day) No. of Days
151 May – 31 May 3.0 17
1 June – 10 June 5.1 10
11 June – 27 June 5.6 17
28 June – 5 Sept 6.4 70
6 Sept – 25 Sept 5.6 20
26 Sept – 15 Oct 5.1 20
depth and the depth of the impermeable barrier below ground surface, determine D.
Also, choose a trial drain pipe diameter and envelope thickness to select the effective
drain radius r. From this point on, additional information must be obtained for tran-
sient design beyond that required for steady state design.
For transient design, an irrigation schedule must be developed as well as a schedule
for all water additions to the soil surface, such as rainfall and/or snowmelt, that will be
significant in raising the water table. An ET schedule for water use from the soil root
zone must be developed. A criterion must be developed to estimate how much water is
used before irrigation is complete. Deep percolation of water below the root zone for
each water addition must also be determined. The specific yield of the soil must be
determined so that it can be used to estimate the amount of water table buildup (rise)
that will be caused by each water addition to the water table. Once all these calcula-
tions have been done, the actual calculations of drain spacing can be undertaken.
Again, computer spreadsheets will be used for this purpose.
Now an example will be given. The details and assumptions given herein comprise
just one possible scenario. The scenario used for any specific design must match as
close as possible the common irrigation and farming practices for the area. If this ex-
ample does not match practices for the area where the designer is working, modifica-
tions will be required to obtain the best schedule for the specific area. For this exam-
ple, the silt loam soil has an average hydraulic conductivity of 20 mm/hr as measured
by the auger hole method and an available water content of 183 mm/m. The depth
from ground surface to the barrier is 12.2 m and the root zone depth for the crop of
sugar beets is 1.2 m. Choose a trial drain depth of 2.4 m. To develop an irrigation
schedule to be used for design, it is decided that the first irrigation will occur on 15
May and ET prior to 15 May is negligible. The design distribution of ET for the grow-
ing season is given in Table 9.2.
For the distribution of the sugar beet roots, it is assumed that 40% of the total ET
comes from the top quarter of the root zone (top 0.3 m). The irrigation schedule will
be based upon use of 75% of the available water in the top quarter of the root zone
(top 0.3 m) between irrigations (0.75 × 183 mm/m × 0.3 m = 41 mm). It is assumed
that each irrigation application more than fills the root zone to field capacity. Since
40% of the total ET comes from the top quarter of the root zone, then the total root
zone ET per irrigation equals 41/0.40 or 103 mm. Assume that the ET on an irrigation
day is extracted as part of the drainout period following that irrigation. Now, 17 days
at 3.0 mm/day = 51 mm of water and 103 – 51 = 52 mm can still be used before the
next irrigation is needed. The next 10 days at 5.1 mm/day uses 51 mm of water and the
total use since the irrigation on 15 May has been 51 + 51 = 102 mm for the first 27
days. One more day (11 June) will use 6 mm, bringing the total to 108 mm since 15
May. This exceeds the allowable use of 103 mm, hence the next (2nd) irrigation will
be on 11 June. These calculations are continued (see Table 9.3) to produce the irriga-
tion schedule. Thus, there are nine irrigations scheduled for 15 May, 11 June, 29 June,
15 July, 31 July, 16 August, 1 September, 18 September, and 7 October covering a
total span of 145 days (27 + 18 + 16 + 16 + 16 + 16 + 17 + 19). The nonirrigation sea-
son then lasts for 365 – 145 = 220 days.
Weather records of the area show that the average year will produce about 25 mm
of water from snowmelt, which will recharge to the groundwater below the water table
on 20 February. The elapsed time from this snowmelt event until the first irrigation on
15 May is 84 days. The rest of the nonirrigation season amounts to 136 days (220 –
84). Because of possible errors using excessively long drainout time periods, the
USBR (1993) recommends splitting this part of the nonirrigation season into two
nearly equal drainout time periods. Thus, the 136 days is split into two drainout peri-
ods of 68 days each.
The next steps involve calculation of the deep percolation losses from the irriga-
tions and the water table buildup (rise) from each of the water additions. It is assumed
that the silt loam soil has an infiltration rate equal to its hydraulic conductivity. From
Table 9.1, for the silt loam soil with an infiltration rate of 20 mm/hr, the deep percolation
from each irrigation is estimated at 18%. The net water depth added to the groundwa-
ter is then obtained from the 103 mm of water added at each irrigation as follows:
Net input = 103/ (1 – 0.18) = 103/0.82 = 126 mm per irrigation.
Deep percolation = 126 –103 = 23 mm per irrigation (or 18% of 126).
From Figure 9.5 with K = 0.02 m/h, the specific yield is 9% (f = 0.09).
Thus for each irrigation, the water table buildup is 23/0.09 = 256 mm or 0.256 m. For
the snowmelt the buildup is 25/0.09 = 278 mm or 0.278 m.
Table 9.3. Details of irrigation schedule calculations.

First irrigation on 15 May Second irrigation on 11 June
(17)(3.0) = 51 mm (17)(5.6) = 95 mm
(10)(5.1) = 51 mm 103 – 95 = 8 mm, 8/6.4 = 1.3 days
27 days 102 mm (1)(6.4) = 6 mm
18 days 101 mm
Third irrigation on 29 June Fourth irrigation on 15 July
103/6.4 = 16.1 days 103/6.4 = 16.1 days
(16 days)(6.4) = 102 mm (16 days)(6.4) = 102 mm
Fifth irrigation on 31 July Sixth irrigation on 16 August
103/6.4 = 16.1 days 103/6.4 = 16.1 days
(16 days)(6.4) = 102 mm (16 days)(6.4) = 102 mm
Seventh irrigation on 1 September Eighth irrigation on 18 September
(5)(6.4) = 32 mm (8)(5.6) = 45 mm
103 – 32 = 71 mm, 71/5.6 = 12.7 days 103 – 45 = 58 mm, 58/5.1 = 11.4 days
(12)(5.6) = 67 mm (11)(5.1) = 56 mm
17 days 99 mm 19 days 101 mm
Ninth irrigation on 7 October
(9)(5.1) = 46 mm, so a tenth irrigation
is unnecessary and ET beyond
15 October is neglected due to begin-
ning of harvest on 16 October.
As discussed in Section 9.4.1, the USBR approach is based upon a concept of dy-
namic equilibrium in which the water table fluctuates during the irrigation season, but
generally rises and reaches its highest elevation immediately after the last irrigation.
Spreadsheets are now used to calculate the proper drain spacing. This example uses
the spreadsheet with filename USBR. This spreadsheet is designed for the USBR tran-
sient flow approach using the USBR equation for drains above the barrier. The spread-
sheet uses the Moody equations for Hooghoudt equivalent depth.
Figure 9.6 shows the first spreadsheet calculation using the input values for this ex-
ample and a trial drain spacing of 150 m. The instructions for spreadsheet use are in-
cluded as part of the spreadsheet. An effective drain radius r of 0.15 m (a 10-cm drain
diameter with 10-cm thick gravel envelope) is used. Also the trial drain depth is 2.4 m
(below ground surface) so that D = 12.2 – 2.4 = 9.8 m. The choice of this first trial
drain spacing is not critical since we can use linear interpolation or extrapolation after
two trials to get values much closer to the correct drain spacing. Note that since the
value of h0 (1.473) at the bottom of Figure 9.6 is greater than the desired value
(1.200), the assumed drain spacing of 150 m is too large. Choose a smaller value, say,
120 m, and recalculate using the spreadsheet. Figure 9.7 shows the resulting calculated
spreadsheet. Since the last value of h0 (1.016) is too small, the assumed drain spacing
(120 m) is too small. The designer may interpolate between the two calculated drain
spacings (150 m and 120 m) to obtain a new trial value of drain spacing (132 m). In-
stead, once the spreadsheet has been constructed, the designer may just wish to enter
new trials of S into cell A18 until the last value of h0 matches the desired value
(1.200). This latter approach was used to produce Figure 9.8 showing the correct drain
spacing to be 132.6 m, rounded to 133 m.
9.4.1.5 Discussion of relief drain design examples. Because of all the data re-
quirements and many assumptions upon which the transient method rests, it would be
interesting to know how close results of using a simple approach such as Hooghoudt’s
steady state equation would be to those of the transient method. The spreadsheet HO-
DESNL (Figure 9.3) can be used to calculate drain spacing for comparison. The ques-
tion arises as to how to choose the value of the drainage coefficient, q. Suppose q is
calculated as the average recharge rate for the peak irrigation period, i.e., when the
interval between irrigations is the shortest. Then for the transient example given in
Section 9.4.1,
q = (0.023 m)/(16 days) = 0.00144 m/day
The calculation of drain spacing with this value of q in Hooghoudt’s equation as
shown in Figure 9.9 gives a value of 152.9 m for drain spacing S. This spacing is 15%
larger than given by the transient method of calculation (Figure 9.8).
Now assume that the transient method gives a good description of water table posi-
tion throughout the irrigation season. The spreadsheet USBR (Figure 9.6) can be used
to find a starting value of h0, to satisfy dynamic equilibrium conditions so that the ef-
fect on water table position throughout the year from drains placed at the spacing cal-
culated from the Hooghoudt equation can be predicted. This is done by trial and error
until the starting and ending values of h0 in column D match for a drain spacing of
153 m as given by the Hooghoudt equation (Figure 9.9). This result is shown in Figure
9.10 where the dynamic equilibrium value of h0, is given as 1.527 m for a drain spac-
ing of 153 m.
A B C D E F G H
1 DRAIN SPACING CALCULATIONS - TRANSIENT FLOW FILE NAME:
2 USBR APPROACH, EQUATION 9.11 USBR
4 1. This spreadsheet is designed for a specific irrigation and recharge schedule. You MUST
5 modify the first three columns of the table for your particular schedule!
6 2. Enter data into CELLS A14 through A18.
7 3. Enter desired height of water table above drain centers as the first h 0 into CELL D28.
8 4. Print the spreadsheet.
9 5. Enter a new estimate of drain spacing into CELL A18 and print the new spreadsheet.
10 6. Use interpolation between spreadsheets to get an improved estimate of drain spacing.
11 7. Repeat process until h 0 at bottom of column 4 agrees with the value at top of column 4.
12 8. For D = 0, do not use this spreadsheet! Use file name USBRDOB.
13 INPUT DATA:
14 0.48 m/day = K (hydraulic conductivity)
15 0.09 = f (specific yield)
18 150 m = S (estimate of drain spacing)
19 CALCULATIONS:
20 0.065333 = D/S
21 3.454004 = α Equation 9.5
22 6.668109 = D e Hooghoudt's equivalent depth for 0 < D/S < 0.31, or D/S = 0.31
23 10.23053 = D e Hooghoudt's equivalent depth for D/S > 0.31
24 6.668109 = D e Hooghoudt's equivalent depth for conditions of this design
25 IRRIG. t Buildup h0 K(D e +h 0/2)t h/h 0 h
26 NO. (days) (m) (m) f S2 (m)
27 9
28 68 1.200 0.1172 0.369 0.443
29 68 0.000 0.443 0.1110 0.392 0.174
30 SM 0.278
31 84 0.452 0.1373 0.303 0.137
32 1 0.256
33 27 0.393 0.0439 0.760 0.299
34 2 0.256
35 18 0.555 0.0296 0.876 0.486
36 3 0.256
37 16 0.742 0.0267 0.901 0.668
38 4 0.256
39 16 0.924 0.0270 0.898 0.830
40 5 0.256
41 16 1.086 0.0273 0.896 0.973
42 6 0.256
43 16 1.229 0.0276 0.893 1.097
44 7 0.256
45 17 1.353 0.0296 0.876 1.185
46 8 0.256
47 19 1.441 0.0333 0.845 1.217
48 9 0.256
49 1.473
Figure 9.6. Spreadsheet using the USBR transient approach to calculate parallel
relief drain spacing showing first trial drain spacing that is too large.
A B C D E F G H
13 INPUT DATA:
19 CALCULATIONS:
20 0.081667 = D/S
21 3.432672 = α Equation 9.5
26 NO. (days) (m) (m) f S2 (m)
27 9
28 68 1.200 0.1705 0.218 0.262
29 68 0.000 0.262 0.1586 0.245 0.064
30 SM 0.278
31 84 0.342 0.1972 0.167 0.057
32 1 0.256
33 27 0.313 0.0632 0.628 0.197
34 2 0.256
35 18 0.453 0.0426 0.770 0.349
36 3 0.256
37 16 0.605 0.0383 0.803 0.486
38 4 0.256
39 16 0.742 0.0387 0.800 0.594
40 5 0.256
41 16 0.850 0.0391 0.798 0.678
42 6 0.256
43 16 0.934 0.0393 0.796 0.743
44 7 0.256
45 17 0.999 0.0420 0.775 0.774
46 8 0.256
47 19 1.030 0.0470 0.737 0.760
48 9 0.256
49 1.016
relief drain spacing showing second trial drain spacing that is too small.
A B C D E F G H
13 INPUT DATA:
18 132.6 m = S (estimate of drain spacing)
19 CALCULATIONS:
20 0.073906 = D/S
21 3.442674 = α Equation 9.5
26 NO. (days) (m) (m) f S2 (m)
27 9
28 68 1.200 0.1443 0.282 0.339
29 68 0.000 0.339 0.1354 0.308 0.104
30 SM 0.278
31 84 0.382 0.1678 0.224 0.086
32 1 0.256
33 27 0.342 0.0538 0.690 0.236
34 2 0.256
35 18 0.492 0.0363 0.820 0.403
36 3 0.256
37 16 0.659 0.0326 0.850 0.560
38 4 0.256
39 16 0.816 0.0330 0.847 0.691
40 5 0.256
41 16 0.947 0.0333 0.844 0.799
42 6 0.256
43 16 1.055 0.0336 0.842 0.889
44 7 0.256
45 17 1.145 0.0359 0.823 0.942
46 8 0.256
47 19 1.198 0.0403 0.788 0.944
48 9 0.256
49 1.200
relief drain spacing showing correct drain spacing obtained by successive trials of S.
A B C D E F G H
2 FOR STEADY FLOW WHEN 0 < D/S < 0.31 or D/S = 0.31 HODESNL
3 FOR LAYERED SOILS
7 (in CELL A21)
8 3. Accept the last value of S in the first column as the correct drain spacing ONLY if:
10 AND
11 (b) D/S lies in the range 0 < D/S < 0.31 or D/S = 0.31
12 4. If D/S > 0.31, then use the spreadsheet with file name HODSN31L
13 INPUT DATA:
20 S (m) D/S D e (m) S 2 (m2)
21 100.00 0.10 5.74 20274
22 142.39 0.07 6.55 22895
23 151.31 0.06 6.69 23318
24 152.70 0.06 6.71 23381
25 152.91 0.06 6.71 23390
26 152.94 0.06 6.71 23391
Figure 9.9. Drain spacing estimated from Hooghoutd’s equation with q as
the average recharge rate for the peak irrigation period of this example.
Careful study of Figure 9.10 shows that after irrigations 6, 7, 8, and 9, h0 exceeds
the desired value of 1.200 m and h exceeds this value at the ends of the drainout peri-
ods following irrigations 7 and 8. At all other times of the year, the water table is be-
low the 1.200 m depth as desired for the crop of this example. A designer must evalu-
ate the economic damages of such water table positions before deciding whether one is
justified in using the simpler steady state approach instead of the much more data-
demanding transient approach to design the drainage system. Of course, many more
sets of conditions and situations than considered here would have to be evaluated be-
fore one could reach any general conclusions on this question. Further work is left to
the individual designer.
Of course, as was mentioned in Section 9.4.1 on steady state design, the designer
must complete the work by ensuring that the trial drain pipe diameter is sufficient to
carry the expected drain discharge (refer to Equation 9.7). One could also develop a
spreadsheet for the special transient design case where the drains rest upon the imper-
meable barrier using Equation 9.1.
The designer is cautioned to use the design tools presented here with judgment
tempered by experience. Especially with the transient approach, an appropriate irriga-
tion schedule must be built into the spreadsheet to handle the specific situation for the
design conditions at hand.
A B C D E F G H
13 INPUT DATA:
19 CALCULATIONS:
20 0.064052 = D/S
21 3.455722 = α Equation 9.5
26 NO. (days) (m) (m) f S2 (m)
27 9
28 68 1.527 0.1158 0.374 0.571
29 68 0.000 0.571 0.1084 0.402 0.230
30 SM 0.278
31 84 0.508 0.1333 0.315 0.160
32 1 0.256
33 27 0.416 0.0426 0.771 0.320
34 2 0.256
35 18 0.576 0.0287 0.884 0.509
36 3 0.256
37 16 0.765 0.0259 0.909 0.696
38 4 0.256
39 16 0.952 0.0262 0.906 0.862
40 5 0.256
41 16 1.118 0.0265 0.903 1.009
42 6 0.256
43 16 1.265 0.0268 0.901 1.140
44 7 0.256
45 17 1.396 0.0287 0.884 1.233
46 8 0.256
47 19 1.489 0.0323 0.853 1.271
48 9 0.256
49 1.527
Figure 9.10. Dynamic equilibrium at a drain spacing of 153 m for this example.
9.4.2 Interceptor Drains

Interceptor drains are single drains installed nearly along a ground surface contour
at a depth to intercept subsurface water moving downslope from higher lands. In con-
trast, relief drainage assumes that all the water removed by the drains originates from
the soil surface above the system of drains and that there is no lateral movement of
water into the area drained.
Interceptor drains can be open drains or buried pipes. Both kinds are equally effec-
tive if functioning as designed. We prefer the buried pipe interceptors. The main ad-
vantages of buried pipe interceptors are no loss of farmed area, no restriction to
equipment movements, no weeds, and low maintenance. Open ditch interceptors in-
troduce problems with weeds, instability of banks, restrictions of farming operations,
loss of significant farm area, and difficutly of maintaining hydraulic grade. The depth
of the drains and the proportion of the subsurface flow cross-section intercepted by the
drain determine the amount of water that will be collected by the drain.
The principal design factor for interceptor drains is the quantity of water entering
the drains, which determines the best pipe size. The depth and location of the drain is
controlled by the local site conditions, particularly the hydraulic conductivity of the
soil, the slope of the groundwater table in the vicinity of the drain, and the depth from
the water table to any low-permeability (restrictive) layer in the soil profile.
Darcy’s law is used for the design of interceptor drains. Consider a vertical section
aligned with the direction of flow of water to be intercepted. The original flow rate
under normal conditions is calculated for a unit width of soil (length of drain) in this
vertical section from Equation 9.15.
Q = Kd(dh/dl) (9.15)
where Q = flow rate, m3/day per m width of soil (length of drain)
K = hydraulic conductivity, m/day
d = depth from the water table to the restrictive layer, in m
dh/dl = hydraulic gradient.
The depth of the drain is selected based on soil profile characteristics and the kind
of excavating equipment available for construction. If the low-permeability layer can
be reached, the drain should be placed on top of the layer. The drain should not be
installed in or under a restrictive layer. The distance of the interceptor drain below the
original groundwater surface divided by the original total depth of the flowing water
above the restrictive layer multiplied by the calculated original flow will be the flow
per unit length of interceptor. That is, the interceptor can be considered to skim all
water arriving between its depth and the original (pre-construction) groundwater sur-
face. The flow per unit length of drain can then be used to choose a pipe size.
The post-construction groundwater surface downslope of the interceptor will be
parallel to the original groundwater surface and at the depth of the interceptor. The
groundwater surface in the soil upslope from the drain will be at the original level be-
yond a distance of 10 times the depth of the drain below the original groundwater sur-
face. The upslope drawdown curve to the drain has an approximate parabolic shape.
The designer is referred to USBR (1993) for additional details on interceptor drains,
including how to place additional drains downslope in an irrigated area requiring a
combination of interceptor and relief drains.
Keller and Robinson (1959) presented results of an extensive sand-tank flume study
of interceptor drains that can be used as a guide for design. Their results include ways
of estimating the increased seepage loss from a canal when an interceptor drain is in-
stalled near the canal. The interceptor drain actually induces more seepage from the
canal than existed before the drain was constructed. As long as the proper size of drain
pipe is used, the interceptor drain will protect the land downslope as intended.
9.5 DETERMINING DESIGN VARIABLES
Drainage design is not an exact science because of the great variability that exists in
natural soils. Frequently, the design variables needed can only be approximated, but
every possible attempt should be made to use accurate and precise information. Rain-
fall is not constant and irrigations are not perfectly uniform, so drainage coefficients are
not precise. Likewise, neither the soil hydraulic conductivity nor the depth to the im-
permeable barrier can be determined with a high degree of accuracy under any circum-
stances, so drainage design also requires the use of reasonable engineering judgment.
9.5.1 Water Table Depth
As given in Section 9.3.1, the design depth to water table at the midpoint between a
system of parallel relief drains is normally taken to be 1.0 m. The USBR (1993) speci-
fies a design depth to water table of 1.2 m. The deeper depth gives a factor of safety in
areas where salinity control is important. If a steady state drain spacing is calculated
using a water table depth of 1 m, a drain spacing calculated by the transient method
will be less, even though the physical variables used in the calculation are identical.
The preferred method is to calculate the steady state drain spacing with a water table
depth of 1 m and then use the calculated steady state drain spacing to calculate the
maximum height of the water table at the time of recharge using a transient design
equation. The fluctuations of the water table will be controlled by the drain depth and
spacing and the recharge events, and not by the design method, steady state or tran-
sient, used to determine the drain spacing.
9.5.2 Drain Depth
Drains should be installed on or above a low-permeability layer and in highly per-
meable layers that are at or near the proper depth. Beyond those guidelines, the depth
of drains is determined primarily by the type of excavating equipment available. Plows
and trenchers used to install subsurface drains have a limited depth capability, depend-
ing on their design. Open drains can be as deep as necessary, but to maintain stable
side slopes, deep open drains take large areas of land out of service and create serious
maintenance problems. Special trenchers have been manufactured to install buried
subsurface drains to depths of 4 to 5 m, but they are uncommon. The USBR (1993)
made an economic analysis of drain depth. Deeper drains can be installed at wider
spacing, but excavation costs are higher. The balance between excavation costs, speed
of construction, and cost of materials resulted in an optimum drain depth of approxi-
mately 2 m. Shallower drains will cost more because more materials will be needed for
the closer spacing. Deeper drains will cost more because labor and excavation costs are
higher. Local conditions may cause the true economic drain depth to be different.
9.5.3 Depth to Barrier
In the design of subsurface drainage systems, one of the necessary variables is the
distance from the drain to some layer in the soil that would be a barrier to vertical wa-
ter movement. Since permeability of the soil is a relative value, any horizontal soil
layer with a hydraulic conductivity less than a tenth to a fifth of the weighted average
hydraulic conductivity of all the overlying layers that are below the water table is con-
sidered to be a barrier. Work of Toksoz and Kirkham (1971a,b) can be used to evalu-
ate this assumption of barrier depth. The designer may also use the appropriate spread-
sheet in this chapter to evaluate the effects of different barrier depths on calculated
drain spacing. In most drain spacing equations, the depth to barrier is replaced by an
“equivalent depth” (see Section 9.4.1). Since the equivalent depth equation contains
the drain spacing and the drain spacing equation contains the equivalent depth, the
solution to the equations must be by trial and error. If the depth to the barrier is un-
known from soil borings or well logs or some other source, the equivalent depth can
be assumed to be 25% of the drain spacing.
9.5.4 Hydraulic Conductivity
In subsurface drain spacing calculations one of the most important factors is the per-
meability or hydraulic conductivity of the soil. Hydraulic conductivity can be deter-
mined in the field by means of an auger hole test (USBR, 1993; Matsuno, 1991) or a
shallow well pump-in test (USBR, 1993). In the absence of any other information, the
basic intake rate of the soil can be used as an estimate of hydraulic conductivity. It
should be remembered, however, that the basic intake rate applies to the highly structured
surface soil, but the water flowing to the drains is moving through the deeper soil layers.
9.5.5 Drainage Requirements
The drainage requirements and goals in humid and arid areas are different (see Sec-
tion 9.1.1). In humid regions, excess water removal is the objective of subsurface
drainage. The hydraulic conductivity of soils in humid areas tends to decrease with
depth so that drains must be closely spaced at a shallow depth. The normal installation
depth for humid area drains is 1.2 m and the selected design steady state water table
height above the drains is 0.2 to 0.3 m. The design depth to the midpoint water table is
therefore about 1 m. Local experience is used to decide on the drainage coefficient,
which is in the range of 7 to 8 mm per day. Humid area soils should not be over-
drained, i.e., drained to such depths that the plants become short of water between
precipitation events or irrigations. The depths below the water table can be used as a
soil water storage reservoir. Organic soils should not be drained below a depth of 0.5
m to protect the soils against accelerated oxidation.
In arid areas, the primary purpose of subsurface drainage is to keep the water table
well below the soil surface so that excessive movement of salts from the water table to
the root zone or the soil surface is avoided. In fine-textured soils, water can move up-
ward by capillarity a distance of 2 m or more, so to avoid salinity problems drains
must be installed at a depth of at least 2 m. In undrained arid areas, natural water ta-
bles should be below 3 m. In arid areas that are irrigated throughout the year or
throughout the full growing season, drains can be installed at depths shallower than 2
m, but the land should not be allowed to stand fallow for any significant period of time
or the surface will become salinized from capillary rise of water. With relatively fre-
quent irrigations, salts will be leached downward and will not accumulate on the soil
surface. The drainage requirement can be calculated using Equation 9.1 to determine
the proper drainage coefficient for use in drain spacing equations.
9.5.6 Drainable Porosity
A saturated soil, in the absence of restrictive drainage conditions, will drain to field
capacity. The difference in water content of soil between saturation and field capacity
is the drainable porosity. If the drainable porosity of the soil is 10%, adding 10 mm of
water to a soil at an equilibrium field capacity water content will cause the water table
to rise 100 mm. A better term to use in the design of drainage systems is specific yield.
The definition of specific yield is the volume of water released by a unit area of soil
for a unit drop in the water table. Specific yield changes with depth to the water table
until the water table is deeper than about 3 m. Specific yield is correlated with hydrau-
lic conductivity for normal drainage depths and the USBR (1993) has published a
curve that can be used to estimate specific yield when the hydraulic conductivity is
known. Field measurements of drain discharge as the depth to water table changes can
also be used to estimate specific yield. The most common laboratory method for de-
termining the specific yield requires interpretation of a soil water retention curve for
the soil in question.
9.6 DRAINAGE SYSTEM MATERIALS
Drainage design is not finished nor can discussion of the design procedure be com-
plete without some consideration of the materials from which the system is to be con-
structed. This section considers the tubing and pipe, envelope, outlet, and manholes
for buried pipe drainage systems.
9.6.1 Tubing and Pipe
The first pipes used for agricultural subsurface drains were short lengths (0.3 m) of
unglazed fired clay pipe with an internal diameter of 0.1 m. Since the pipes were made
in the same kilns as other types of clay tiles, they were called drain tile and drains
constructed of these rigid pipes were called tile drains. The reader is referred to
Weaver (1964) for a more complete history of tile drains in the U.S. The pipes were
merely butted together in the bottom of a trench and depended on the irregularity of
the ends of the pipes to admit the drainage water. Standards were developed for
strength and the pipes were used successfully for many years.
Concrete pipes also have been used for subsurface drains. They were first made in
0.3-m lengths with square ends and were called concrete tiles. Later, special concrete
drain pipes were made in approximately 1-m lengths with tongue and groove or bell
and spigot joints that fit loosely to admit water.
In the 1960s, corrugated and perforated plastic tubing, made primarily of high den-
sity polyethylene in the U.S. and polyvinyl chloride in Canada and Europe, was devel-
oped and was rapidly accepted as a substitute for clay tile and concrete pipe as subsur-
face drains. Today, concrete pipes are normally used for buried subsurface collector
drains larger than 0.3 m and corrugated plastic tubing is the dominant material for
smaller-diameter collector drains and laterals. Clay and concrete pipe are rigid pipes
and have strength and loading characteristics that are very different than the flexible
corrugated plastic tubing. Corrugated plastic tubing (called simply plastic drain in the
following) is classified as flexible even though it feels rigid. Plastic drains must be
carefully bedded to prevent being crushed by the weight of the backfill in the drain
trench. If a gravel envelope is not used, plastic drains should be laid in a specially
formed 90° or semicircular groove (with the same diameter as the outside of the drain
tubing) made in the bottom of the trench. Otherwise, the tubing may be deformed or
crushed. The full resistive strength of plastic drains can only be attained with the
proper bedding described above. Placing a plastic drain in a flat-bottomed trench with
loose material along the sides as backfill can result in flattening or failure of the drain.
Surrounding plastic drains with a fine gravel or coarse sand envelope will provide the
drain with adequate structural support in a flat-bottomed trench.
There are ASTM standards for both rigid pipes and flexible drain tubing. The stan-
dards provide specifications for strength and durability and protection of plastic drains
from ultraviolet light damage. The USBR (1993) and the NRCS (2001) have drainage
handbooks containing additional information about pipe and tubing strength and in-
stallation requirements.
Studies in Europe (Dierickx, 1992) have shown that the open area for entry of wa-
ter into drain pipes or tubing should be 20 cm2 per meter length as a minimum and that
open area up to 50 cm2 per meter improves drain performance. Plastic drains should
have at least one perforation in every corrugation.
The advantages of plastic drains over clay or concrete pipe are light weight, ease of
handling, lack of loose joints to become misaligned, low cost, and durability. Disad-
vantages are its low structural strength without proper bedding, susceptibility to crushing
from point loads during backfilling, and greater hydraulic roughness than rigid pipes.
Any pipe that will not deteriorate while buried in the soil can be used as a drain
pipe if it has adequate open area for the entry of water. Pipe perforations must be small
enough to prevent the entry of sediment if the pipe is not protected by synthetic or
natural envelope materials.
9.6.2 Envelope
Porous materials such as coarse sand or fine gravel are placed around drains in un-
stable soils to prevent movement of soil particles into the drains. These materials are
called drain envelopes. Early in the history of buried pipe drainage these envelopes
were called filters. A filter is exactly what we do not want to place around our drains.
Since a filter is designed to trap particles, it will clog over time and prevent water
movement into the drain. Conversly, a well-designed envelope of properly graded
granular material will naturally “bridge” to prevent most soil particles from passing
through but will allow very fine soil particles into the drain. These very fine particles
will be transported on out through the system with the normal water flow. The NRCS
(2001) and USBR (1993) have published criteria for the design of granular material
envelopes for drains.
Soils (particularly fine sands and coarse silts) in arid regions are frequently struc-
turally unstable and thus most subsurface drains require the use of envelopes. Drain
envelopes will always improve the hydraulic functioning of a drain. Soils in humid
areas tend to be structurally stable so that drain envelopes are not normally necessary.
In recent years, various synthetic fabrics called geotextiles have been used for drain
envelopes. While they may be cheaper and easier to install, they are not as effective as
granular materials. Where success with geotextiles has been reported, probably no enve-
lope was needed. We recommend envelopes of properly graded granular materials.
9.6.3 Outlet
As stated in Section 9.1.1, the outlet(s) is the discharge point(s) of the drainage sys-
tem where the water enters an existing channel or waterway. Small drainage systems
may have only one outlet but larger, more complex systems usually have multiple outlets.
The importance of the outlet should not be underestimated. The adequacy of the entire
drainage system depends on a properly designed, constructed, and maintained outlet.
Gravity flow of all water from the drains to the receiving bodies of water is usually
most desirable. When this is not possible or is difficult to do, sumps equipped with
water-level controlled pumps may be used to lift the drainage effluent to the proper
elevation.
Regardless of whether the drainage system uses sump pumps or discharges by

gravity flow, the last section of buried pipe should be solid rigid pipe that extends over
the receiving body of water. This pipe is necessary to prevent damage to the system
from livestock and/or wildlife. Usually the last section of pipe is specified as corru-
gated metal culvert pipe of suitable diameter. To protect the outlet pipe from en-
croachment by muskrats or other animals, flap valves, grates, or other devices may be
installed at the exit end of the pipe (NRCS, 2001). We recommend that the invert of
the outlet pipe be at least 0.3 m above the high-water stage of the waterway.
9.6.4 Receiving Bodies of Water
Although formally not a part of the drainage system, the existing waterway or other
receiving body of water must be carefully considered. The stage-frequency relation-
ship of the waterway must be known, as well as its capacity to carry the drain dis-
charge together with its normal flow. The quality of water in the receiving water body
must be such that the drainage water can be safely discharged into it. Subsurface
drainage outlets have been classified as point source pollution if they damage the wa-
ter quality of the receiving stream.
In determining the effective elevation of a proposed drainage outlet, the elevation
of the water surface throughout the year should be determined to eliminate the possi-
bility of submerging the drain system outlet at a time when drainage is most needed.
9.6.5 Manholes
Manholes in subsurface drainage systems serve the same purpose as manholes in a
sewer system; they provide access for inspection and cleaning. As a general rule,
manholes should not be used in a subsurface drainage system unless absolutely neces-
sary. Manholes interfere with cultivation if they are placed in a field. If they are left
open they can be dangerous to children and animals, and are entry points for trash that
may clog the drainlines. Manholes are useful at points where collector systems come
together so that performance of individual systems can be observed. Rather than in-
stalling manholes at regular intervals, it is more economical to dig down to a drain
with a suspected problem.
9.7 CONSTRUCTION METHODS AND EQUIPMENT
Subsurface drains are installed in trenches that may be hand dug or opened by a va-
riety of digging machines. Most drains are installed by mechanical trenchers designed
and equipped especially for drain installation. Fouss (1974) shows photographs of
drainage equipment which has not changed much in recent decades except for more
widespread use of laser grade control. Fouss shows a modern chain-digger trencher
equipped with a shield to prevent collapse of the trench before the pipe can be in-
stalled in an unstable soil, and a drainplow that can install drains to depths of 1.5 m
without making an excavation. (Drainplows are uneconomical for installations deeper
than 1.5 m.) Trenching machines are classified according to the digging method used:
a wheel, flexible ladder, or digging chain. The chain-type trencher is used for narrow
trenches common for small diameter lateral drains. Wheel and ladder trenchers are
used for the wide trenches needed for large diameter drains. Drainplows are used for
small-diameter drains installed with or without geotextile envelope materials, but not
with gravel/sand envelopes.
Trenches can be opened with a backhoe or other type of excavating machine, but
extra effort is then needed to bring the bottom of the trench to the necessary firm uni-
form grade. Drain pipes should be laid straight or on gently curving uniform grades
without vertical deviations greater than 15% of the pipe diameter. Vertical deviations
(humps or valleys) in a drainline reduce the flow gradient and make a deadwater area
upslope of the hump where sediment can collect and reduce the flow capacity of the
pipe. Depressions in a drainline also collect sediment and reduce flow capacity. Over-
excavation of a trench and bringing the trench to grade by throwing in loose material
may result in vertical grade deviations, pipe misalignment, and structural failure of the
pipe. Thus, bringing an over-excavated trench back onto grade should be done only
with gravel material—an expensive process. The foregoing discussion points to the
need for good grade control during the construction of drains. Controlling grade with a
laser is now commonly used with drain plows and drain trenchers to assure adequate
control of grade and alignment.
9.8 DRAINAGE SYSTEM OPERATION AND MAINTENANCE
Subsurface drainage systems with gravity outlets operate continuously and auto-
matically with minimum maintenance. If the outlet is a sump with a pump, then con-
tinuous maintenance of that part of the system is necessary to assure adequate drain-
age. The operation of the system is fixed at the time of the design and consists of the
design rate of water removal or the rate of lowering of the water table at the midpoint
between drains. If the drain flows respond to irrigations or rainfalls in excess of the
water storage capacity of the soil, the drains are assumed to be operating properly. The
appearance of new wet spots near a drainline in a field is an indication that the drain is
clogged at that point. The drain can then be excavated, examined, and repaired. The
crop can be watched for indications of inadequate drainage, such as salt accumulations
on the soil surface or differences in crop growth (poor growth in wet periods or good
crop growth in dry periods). In periods of water shortage, plant roots can invade and
clog the drains. Sugar beet, cotton, willows, poplars, and deep-rooted desert plants can
clog drains in a relatively short period during times of water shortage. Use of rigid
drain pipes is recommended if such conditions are expected.
Maintenance of buried subsurface rigid-pipe drains is primarily accomplished by
high-pressure water jetting machines similar to sewer cleaners used in the city (Grass
et al., 1975). Sections of drains up to 200 m long can be easily cleared of roots, sedi-
ment, and iron ochre deposits by jet cleaning. Pump pressures are approximately
4 MPa. The drains are cleaned in sections from excavated access holes. After an ap-
propriate plug is placed into the downstream section of drain at the access hole, clean-
ing is done in an upstream direction from this plug and the water and debris are
pumped out of the access hole. When cleaning of this section is finished, the drain
pipe and envelope are repaired and the access hole is filled. Cleaning costs are usually
20% or less than the cost of installing a new drain.
9.9 DRAINAGE SYSTEM PERFORMANCE EVALUATION
Drainage system performance criteria are fixed at the time of design. Drain depths
are set by the capacity of the available digging machines, the requirements of the area,
and the stratification of the soil. Drains should not be installed in or below soil layers
that are slowly permeable. The minimum allowable depth to the water table at the
midpoint between drains is fixed by local experience and is usually approximately
1 m. The drainage requirement or design drainage coefficient determines the rate of
water removal necessary and then the soil hydraulic conductivity is used in a suitable
equation to determine the appropriate drain spacing. If the soil drains acceptably, the
performance criteria are assumed to be met. If the soil does not drain properly, further
investigation will be needed to determine whether the drain is faulty or whether addi-
tional drains are necessary.
REFERENCES
Dierickx, W. 1992. Use of drainage materials. In Proc. of 5th Drainage Workshop,
Vol. I. New Delhi, India: ICID-CIID.
Dumm, L. D. 1968. Subsurface drainage by transient-flow theory. J. Irrig. Drain.
Div., ASCE 94(IR4): 505-519.
Fouss, J. L. 1974. Drain tube materials and installation. In Drainage for Agriculture,
147-177. Monograph No. 17. J. van Schilfgaarde, ed. Madison, Wis.: American
Soc. Agronomy
Grass, L. B., A. J. Mac Kenzie, and L. S. Willardson. 1975. Inspecting and cleaning
subsurface drain systems. Farmers Bulletin No. 2258. Washington, D.C.: USDA-
ARS.
Hooghoudt, S. M. 1940. Bijdragen tot de kennis van eenige natuurkundige grootheden
van den grond, 7, Algemeene beschouwing van het probleem van de detail
ontwatering en de infiltratie door middel van parallel loopende drains, greppels,
slooten, en kanalen. Versl. Landbouwk Ond. 46: 5 15-707.
Keller, J., and A. R. Robinson. 1959. Laboratory research on interceptor drains. J.
Irrig. Drain. Div., ASCE 85(IR3): 25-40.
NRCS. 2001. Chapter 14: Water management drainage. In Engineer Field Handbook,
Part 650, 14-56. Washington, D.C.: USDA, Natural Resources Conservation
Service.
Matsuno, Y. 1991. Improved measurement of hydraulic conductivity in an auger hole.
MS thesis. Pullman, Wash.: Washington State Univ.
Moody, W. T. 1966. Nonlinear differential equation of drain spacing. J. Irrig. Drain.
Div., ASCE 92(IR2): 1-9.
Toksoz, S., and D. Kirkham. 1971a. Steady drainage of layered soils. I: Theory. J.
Toksoz, S., and D. Kirkham. 1971b. Steady drainage of layered soils. II: Nomographs.
J. Irrig. Drain. Div., ASCE 97(IR1): 19-37.
USDI-BR. 1993. Drainage Manual. 3rd ed. Denver, Colo.: U.S. Department of
Interior, Bureau of Reclamation.
van Schilfgaarde, J. 1965. Transient design of drainage systems. J. Irrig. Drain. Div.,
ASCE 91(IR3): 9-22.
Weaver, M. M. 1964. History of Tile Drainage (in America Prior to 1900). Waterloo,
N.Y.: M. M. Weaver.
CHAPTER 10
LAND FORMING
FOR IRRIGATION
Allen R. Dedrick (USDA-ARS,
Beltsville, Maryland)
Ronald J. Gaddis (A B Consulting,
Lincoln, Nebraska)
Allan W. Clark (Clark Brothers,
Dos Palos, California)
Alan W. Moore (Cameron County Drainage
District 5,Harlingen, Texas)
Abstract. Land forming for irrigation is a key part of the land improvement process
in all regions where crop production is dependent upon irrigation. The state of the art
of land forming as it relates to irrigation development, particularly surface irrigation,
is presented. Topics addressed include system layout, soil survey and allowable cuts,
topographic surveys, earthwork analyses, types of equipment, field operating proce-
dures, cost and contracting, and safety. Computer-aided design, including various
computer software packages, and extensive changes in equipment for land forming
and its control emerging over the past 25 to 30 years are included.
Keywords. Irrigation, GPS, Land forming, Land improvement, Land levelers, Land
leveling, Land shaping, Land smoothing, Lasers, Scrapers, Surface drainage, Surface
irrigation, Water conservation, Water management.
10.1 INTRODUCTION
Land forming for irrigation is a key part of the land improvement process with the
goal of sustaining high crop productivity. It provides a land surface ideal for surface
irrigation and drainage. Both irrigation and drainage systems require land surfaces that
transport water sufficiently slowly to allow adequate infiltration and minimize erosion,
while at the same time transporting water rapidly and uniformly to prevent injury to
plants. The benefits of land forming are extensive (Springer, 1966; Lyles, 1967;
Quackenbush, 1967; Buras and Manor, 1970; Sewell, 1970; Harris, 1971; Kriz et al.,
1973; Thomas and Cassel, 1979; Shih et al., 1981; and Shih, 1992) and include:
improving surface drainage by eliminating low areas and adding grade to areas
of no slope;
permitting more efficient use of irrigation water and making possible the irriga-
tion of a larger area with a limited water supply;
improving efficient control of water by providing for uniform distribution;

reducing the number of surface field ditches required for good drainage since
each furrow carries its own water;
reducing erosion by preventing water from concentrating in natural drainage
ways;
reducing herbicide and fertility losses because of better control of water;
reducing ditch maintenance by minimizing the number of deep ditches;
saving machinery-operating time in turning around at the end of large fields as
compared to small fields;
increasing the cropping area by removing ditches and fencerows;
increasing machinery efficiency because point rows are eliminated and machin-
ery can be operated at higher speeds;
increasing the precision of inter-row tillage;
increasing the accuracy of height or depth of the cutting edge of harvesting ma-
chines;
reducing wear and damage to equipment working in the field;
providing better crop and soil management because the uniform ground surface
allows precision planting and fertilizer application, which results in more uni-
form germination and crop maturation;
increasing crop yield because of better soil, water, and crop management;
reducing organic soil subsidence by maintaining a more even water table depth
within a field;
reducing excessive organic soil mineralization and nitrification by eliminating
the high spots; and
saving energy in paddy rice production by reducing amount of water pumped for
flooding the fields.
The benefits touch on all aspects of productivity. Further, the list addresses issues
pertaining to all climatic regions, i.e., irrigated areas where production is dependent
upon irrigation (arid, semi-arid regions) as well as areas where irrigation is supple-
mental to precipitation. In irrigated areas with higher precipitation, land forming is
extremely important to control surface drainage and improves the uniformity of water
application by surface irrigation. In some instances, the ideal land forming for surface
drainage and surface irrigation systems have conflicting requirements: flatter slopes
for surface irrigation systems generally result in higher irrigation efficiencies, but
drainage may be inadequate.
Irrigation systems (whether surface, sprinkle, or micro) are selected and designed
considering various factors applied to specific site conditions. These factors include:
water supply—source location, delivery flow rate, quality, delivery schedule,
delivery point;
crop characteristics—effective root zone, growth characteristics, cultural prac-
tices;
soil features—water-holding capacity, intake rate, depth;
topographic features—surface irregularities, steepness, direction of slope;
geographic features—field shape, natural drains, buildings, utilities or obstruc-
tions;
labor—amount required, quality, quantity available, cost;
energy—amount required and available, associated cost;
322 Chapter 10 Land Forming for Irrigation
system cost—hardware and other ancillary equipment, installation, operation,

maintenance;
farm equipment—type commonly used and/or available; and
farmer’s preference.
All irrigation methods may require some land forming to attain maximum produc-
tion, depending mostly on the climatic region and irrigation method used. However,
the performance of surface irrigation methods is impacted most by the field surface
topography. To guide the process of land forming, topographic features influenced by
the soil depth and type are paramount. Depending on the irrigation system selected,
the field may be graded with a single or a series of slopes in the direction of irrigation,
and either with or without side fall, i.e., field slope other than that in the direction of
irrigation (discussed further in Section 10.2).
As used herein, land forming (or shaping) refers to both land grading and land
smoothing. Land grading refers to the movement of soil associated with meeting the
general slope requirements of the irrigation system design, while land smoothing (or
finishing) generally refers to the removal of nonuniformities in that soil surface. The
average finished slope is a statistical measure of how accurately the field was graded
to meet the design slope. Land smoothing can be statistically described by the preci-
sion of the field elevations around the design slope (i.e., the standard deviation of the
field elevations from the design slope).
Land grading and smoothing improve the distribution uniformity of irrigation wa-
ter. This improved uniformity provides an opportunity to improve the irrigation appli-
cation efficiency of the system. Generally, uniform irrigations improve the resultant
crop yield, because of the potential negative impacts of low and high areas. Low areas
are generally areas of over-application, resulting in poor soil aeration, associated oxy-
gen deficiencies, and potential excess leaching of nutrients. High areas are generally
areas of under-application, thus resulting in insufficient water to meet crop needs and
leading to potential build-up of salts. Many times irrigators will apply additional irriga-
tion water to sufficiently irrigate the high areas, which potentially exacerbates the nega-
tive impacts associated with the low areas and lowering the overall irrigation efficiency.
10.1.1 Scope
The objective of this chapter is to review the state of the art of land forming as it re-
lates to irrigation development, particularly surface irrigation. Land forming associ-
ated with nonirrigated land is not addressed directly, although much of the information
presented would apply. Topics addressed include system layout, soil survey and al-
lowable cuts, topographic surveys, earthwork analyses, types of equipment, field oper-
ating procedures, cost and contracting, and safety. Computer-aided design, including
various computer software packages, and extensive changes in equipment for land
forming and its control have emerged in the past 25 to 30 years.
10.2 SYSTEM LAYOUT
Before the land forming for a field can be designed, the farm irrigation system,
drainage system, main field boundaries, roads, and other physical features must be
planned. From soil information, topography, method of water application, crops to be
grown, and the farmer’s own personal likes and dislikes, alternative field arrangements
are developed. Field arrangement and design are important parts of the process of land
forming. It is essential that planning be done with imagination. All of the practical
potential layouts should be considered and the one best suited to the site selected.
The planner should consider and include, when appropriate, the following items
when developing the irrigation system layout:
location and elevation of the water supply;
amount and arrangement of supply ditches and/or pipelines, as well as water
control requirements;
location, number, size, and elevation of drainage ditches;
location, size and method of operation of the return-flow system;
type of farm equipment to be used and ease of access to and operation within all
fields for that equipment;
field location and arrangement, to minimize soil variations within each field;
combining small fields into larger fields, especially if larger farming equipment
is to be used;
minimizing non-cropped area;
ease of management of irrigation water and cultural operations; and
current and future crops to be grown.
Design criteria must be established before the field can be subdivided and the de-
sign for land forming prepared. Both irrigation and drainage affect the allowable
length of fields. Excessively long fields may create erosion problems during irrigation
and/or rainfall. Short fields require more delivery and drainage facilities for a given
area. Efficiency of machinery operation depends on the type of machinery used, but
generally increases with length of field.
Distribution uniformity (DU), when used as the design criterion for surface irriga-
tion systems, can be described as
DU = f (I, n, so, L, Qin, Zn) (10.1)
where I = infiltration characteristic, generally defined by a relationship between
cumulative infiltrated depth and infiltration opportunity time
n = flow resistance (Manning n)
so = slope (or series of slopes) in the direction of irrigation of the area to be
irrigated
L = length of the furrow or border
qin = unit flow rate, i.e., flow rate per unit width (Qin/W) or flow rate per furrow
(Qin is the inflow rate to the field, and W is the width of the area being
irrigated in a particular set)
Zn = net depth to apply.
Infiltration and flow resistance are characteristics inherent to the irrigation site. At
the time of design, the other four factors (slope, length, unit flow rate, and net depth to
apply) are adjusted within various constraints to provide an irrigation system that will
meet the targeted design performance criteria. The design slope in Equation 10.1 re-
flects land grading requirements. In the U.S., irrigation and drainage guidance for the
particular type of surface irrigation system(s) being considered, i.e., how the last four
factors of Equation 10.1 are combined, can be obtained from various sources such as
the USDA-Natural Resources Conservation Service (formerly USDA-Soil Conserva-
tion Service), Cooperative Extension Service, universities, USDA-Agricultural Re-
search Service, or private consultants. These design guidelines, when combined with
soils, depth of water to apply, and flow information, provide optimum field slopes and
lengths to achieve the targeted design criteria. Additional information on surface hy-
draulics is given in Chapter 13 and design procedures are given in Chapter 14.
10.3 SOIL SURVEY AND ALLOWABLE CUTS

Land forming should only be done after first knowing the soil profile conditions
and the maximum cut that can be made without permanently affecting agricultural
production by exposing infertile and/or rapidly draining subsoils. General soil survey
maps are available for many areas and, if not available, the services of a soil scientist
should be obtained. In the U.S., soils information may be obtained from the local
Natural Resources Conservation Service (NRCS) office, local Cooperative Extension
Service, or a state university. The standard soil survey generally delineates areas larger
than 2 ha. The soil profiles are described to a depth of 1.5 m. In alluvial soils, sand and
gravel pockets may occur within a soil-mapping unit of some other soil type. A de-
tailed site investigation should be made to determine the extent of such material before
the land grading begins so that adequate fill material can be made available. Sand and
gravel pockets that occur near the water inlet end of a surface irrigation system may
greatly limit the irrigation efficiency because of high intake rates. It is often advisable
to remove these sand and gravel materials and replace them with soil similar to the rest
of the field. If there are doubts as to the soil depths, investigation with an auger in sev-
eral representative locations within the land area is mandatory.
Soils with deep, well-drained subsoil generally have few limitations on depth of
cut. Some shallow soils may be suitable for irrigation, but may not have sufficient
depth (e.g., shallow soils over bedrock, gravels, caliche, heavy claypan, or fragipans)
to permit the necessary land forming needed for surface irrigation, while they could be
successfully irrigated with sprinkle or microirrigation systems. Cross-slope benching
is sometimes used on sloping land to reduce the depth of cut needed to obtain suitable
field grade and can reduce the cost of land forming (Dedrick, 1989).
10.4 TOPOGRAPHIC SURVEY
It is extremely important that the entire farm be studied before any grading work is
attempted, even though land grading is usually done on a field-by-field basis, because
the most economical design for an individual field may be undesirable when the entire
farm system is considered. The factors noted in Section 10.2 should be considered by
the designer in planning the overall farm development. A general topographic map of
the entire farm showing the location of all physical features can provide much of this
information. The designer must select the irrigation layout; the most desirable location
for the various elements of the farm irrigation, water reuse, and drainage systems; and
location of field boundaries and field roads. By presenting this type of information,
several alternative designs can be developed for consideration and selection by the
farmer or farm manager. The selected plan should permit land grading to be carried
out on individual fields, even over a period of years, if necessary.
Planning is aided by development of a detailed topographic map. When collecting
data for the map, consider:
What is the size of the area to be planned?
What are the availability and capability of personnel?
What are the time constraints for getting the job done?
How will the topographic data be analyzed, i.e., how will field slopes, earthwork
quantities, and haul distances be calculated?
Does the type of grading and smoothing equipment and the procedure used dur-
ing the grading and smoothing operations dictate a certain topographic layout?
Will grid stakes be required for the grading operation?
What type of surveying equipment will be used to collect the data?

How will the topographic map be generated, by hand or by computer?
In simplest terms, the topographic data collected includes the elevation, z, of vari-
ously located x and y coordinates, including boundaries to the area. The survey data
collected in the field will be structured either according to a uniformly spaced, grid-
based pattern or irregular spacing based on documenting the extremes and changes of
the surface. The choice of data collection pattern will depend on the general smooth-
ness of the area, the method to be used to analyze the data, type of grading equipment
to be used and associated design required, and how the field surfaces will be mapped.
Highly accurate global positioning systems (GPS) have been introduced to the land
forming industry during the past 10 to 15 years. In one system, the elevation informa-
tion is obtained from a laser transmitter/receiver system and the x-y coordinates from
the GPS. A second system gets both the position and elevation from the GPS. Both sys-
tems can be used with either grid-based data collection or irregularly spaced patterns.
10.4.1 Grid-Based Pattern
10.4.1.1 General description. The grid-based pattern is best suited to areas that are
relatively smooth. If, during construction, the method of grading is to be based on
lanes and controlled by periodic cut/fill stakes, a grid layout has the advantage of de-
fining the lanes. If the method of grading depends on manual control, grid locations
must be preserved until construction when these same locations will be used for the
cut/fill stakes. If laser-controlled scrapers are used for grading, only a few points of
known elevation are required and systematically located cut/fill stakes are not needed.
Information needed regarding the finished slope will depend on the type of laser
hardware used. The main disadvantage of the grid pattern is that critical high and low
points and changes in slope do not often occur at grid points, and thus they are missed
in data collection and analyses leading to inaccurate surface portrayal.
For ease in computing volume and area, especially when not computer aided, the
grids should be made square; however, stakes may be set at any spacing desired by the
designer. The most commonly used spacing is 30 m. Larger spacing makes it more
difficult for the operator of manually controlled grading equipment to maintain the
grade from one stake to the next. Closer spacing greatly increases the work of staking,
surveying, and computing the grading design. Occasionally, it may be necessary to
locate additional points in the grid pattern to record significant topographic changes in
the land surface or to show irregularly shaped boundaries. If closer spacing or numer-
ous additional points are needed to describe the topography, irregularly spaced data
collection may be more appropriate.
Grid-based elevation readings are either taken in the center of each square or at the
corners. Each data set type requires a different method of analysis. An elevation read-
ing taken in the center of each square is assumed to represent the mean elevation of
that entire square. The field surface analysis resembles a checkerboard with each
square at a different elevation. Exact elevations around the border of the surface are
unknown and either must be estimated or can be defined by adding readings at a few
locations. The field layout is simple, usually adequate for flat, uniform surfaces, and
lends itself readily to simplified approximations with regard to slope design and
earthwork calculations.
When the elevation readings are taken at the edges of the squares, a more accurate
surface analysis model (based on triangulation) is possible since each square actually
defines two triangles. This type of grid data can be reduced to the first type of grid
representation if one of the simple approximation methods is to be used for design of
land forming.
10.4.1.2 Staking the field. Wood laths (10 mm × 45 mm × 1.2 m) are excellent
stakes for setting the grid points. They are easily set by the technician from a standing
position, and cut or fill markings near the top of the lath are easily seen by the equip-
ment operator. The laths should be sharpened and driven into the ground far enough so
that they will remain standing until the grading work is completed.
A procedure is outlined by Anderson et al. (1980) to guide the complete staking of
a field. The procedure entails the establishment of key rows of stakes across the ends
of the field from which the remaining grid stakes can be set by eye. For larger fields it
may be desirable to establish the guide rows near the center of the field to eliminate
some of the errors that occur in setting long lines by visual sighting.
10.4.1.3 Survey equipment and techniques. Since the x-y coordinates are defined
by the grid layout, the elevation is the only remaining unknown. The elevations are
usually determined with an engineering level and rod or a rod with a laser receiver. To
facilitate field readings, the rodman may use an all-terrain vehicle or pickup truck to
move quickly from reading site to reading site. In some cases, the laser-control equip-
ment attached to the tractor-scraper unit is used.
10.4.2 Irregularly Spaced Pattern
10.4.2.1 General description. Irregularly spaced patterns can be used to document
the field surface. Irregularly spaced data points are useful in mapping the area to be
irrigated as well as to map important planning features outside the irrigated area. No
prior layout is required. The location of data points is selected in the field and should
reflect various controlling features or break lines. Break lines are topographic features
that have more or less uniform slope, such as natural and man-made surface drains,
streams, lakeshores, roads, railroads, ditches, and ridgelines. Numerous data points
can be used to adequately represent any feature. On uniform portions of the ground
surface, the distance between readings can be increased without compromising the
integrity of the data analysis model.
10.4.2.2 Surveying equipment and techniques. Analyses of irregularly spaced
data can result from any surveying technique that provides both the location (x-y co-
ordinates) and the associated elevation readings. Typical equipment used includes the
transit, plane table and alidade, automated total station, aerial photographs with stereo
mapping, and laser-GPS and GPS surveying systems. If accurate and detailed contour
maps for the area already exist, electronic digitization can be used to develop a data-
base for reproducing the topographical map and for calculation of land grading .
Transit. The transit provides both location and elevation. The surveyor uses the sta-
dia readings to determine distance from the instrument and, with the horizontal and
vertical angles, to calculate the coordinates of a point. Electronic distance measuring
equipment (EDM) can be used instead of the stadia. EDM is rapid and eliminates
some hand calculations and associated errors, especially when combined with hand-
held, battery-powered electronic notebooks or data loggers to register and store data,
thereby reducing the need for hand recording all survey records.
Plane table and alidade. The plane table and associated alidade are best suited for
situations where the contour map itself is the end product. The method is of limited
use where data are needed for land grading design. Often, an engineering level is used
with the plane table and alidade to supplement the data collection by determining the
data point elevations. The alidade is used to sight the rod. Stadia crosshair readings are
used to determine the distance to the sighted point. The rectangular base (also a ruler
or scale) of the alidade rests directly on the map as it is drawn. The line of sight and
the scaled horizontal distance to the rod are plotted by hand to locate the x-y coordi-
nates of each data point as the field is being surveyed. Location of the point replaces
the recording of the horizontal angle.
The plane table and alidade can be used to develop contour lines directly by fixing
a target on the rod at a height that will result in the elevation of a desired contour line.
A series of irregularly spaced points, found to be at the desired elevation using the
level, are located and drawn on the map with the alidade. Connecting the points draws
the contour lines.
Total station. A popular, current surveying technique involves the use of the auto-
matic total station. This apparatus is a battery-powered, computerized transit using
EDM technology that also performs various data calculation, recording, and transfer
functions. The total station electronically registers the x-y distances and elevations of
points selected by a rodman carrying a mirror-like prism. The instrument operator
transmits a beam of light to the prism enabling the instrument to measure horizontal
and vertical angles (like a transit) and the distance from the instrument to the target.
The 3D coordinates of each selected ground location are electronically calculated and
stored, later to be transferred for data analysis. It is often possible to survey a very
large area from a single location since such equipment functions accurately over long
distances, with actual distance depending on atmospheric conditions.
Aerial mapping. Aerial photography with stereo plotting is best used when a large
area (several thousand hectares) is to be planned as a unit or where work is to take
place on several large adjacent farms at the same time. The photography should be
done during the season of low vegetative growth and the maps plotted, as they are
needed. This procedure requires that vertical and horizontal control points be estab-
lished before the photos are taken. These points, which must be visible on the photos,
usually are made with white plastic strips in the form of a cross. Low-level photo-
graphs are taken from an airplane. The height of the plane above ground level will
depend on the camera used and the detail required for the topographic map. For exam-
ple, a 150-mm focal length lens in the camera at about 900 m to 1200 m above the
ground will provide the resolution needed to plot 0.6-m contour lines. The topographic
map is made with a stereoscopic plotter using stereo pairs of photos. While excellent
for general planning purposes, the individual areas to be graded will require additional
surveying to provide accurate data for grading design calculations.
Laser-GPS system. This system combines the use of laser and GPS technology to
provide survey information similar to that provided by the total station. In addition,
such surveying systems are coupled with computer-aided land leveling software that
produces various design and display information that is used directly with land form-
ing equipment and/or as support material for the equipment operator.
GPS surveying system. This recently introduced system performs all of the func-
tions of a total station utilizing the GPS to obtain both the field coordinates and the
elevation.
10.4.2.3 Digitization. Digitizing provides a means of transforming a contour map
from any source (including plane table and alidade data) into a database for analysis.
This can be done either manually or using a digitizer. If done manually, either a grid is
established prior to the survey so that the x-y coordinates are known, or the contour
map of the area to be irrigated is drawn using irregular survey points as needed. A
clear plastic overlay, marked with grid locations, is superimposed on the map enabling
the elevations to be interpolated. Digitizing technology registers the location of each
x-y coordinate of interest from maps or data points. The elevation is entered either by
defining a series of x-y coordinates on a contour line or by entering individual eleva-
tions for the selected points. When using digitized contour lines in surface computa-
tions, the user needs to insure that points on one contour line triangulate with those of
an adjacent contour. Computer software exists that allows such user intervention. Fur-
ther, the user must prevent three points on the same contour from forming a triangle,
which would then erroneously form a level triangle.
10.4.3 Digital Elevation Models (DEMs)
The availability of computers with the capability to rapidly process large quantities
of data has made hand or calculator analyses almost obsolete. The same skills once
used to carefully draw maps and make various earthwork calculations are now applied
toward managing inexpensive computer software to rapidly produce, revise, and refine
contour maps and system design options. The digital representation of the topographic
survey data, i.e., an array of points whose horizontal positions are given by their x and
y coordinates and whose elevations are given as z coordinates, are known as digital
elevation models (DEMs) or digital terrain models (DTMs). Generally, when an ir-
regularly spaced DEM is utilized, a triangulated irregular network (TIN) surface
model of the area is formed. The TIN model is constructed by connecting points in the
array to create a network of adjoining triangles. (The same procedure can be applied to
grid-based data.) Each of the triangles is formed by lines joining together three of the
actual survey points, one at each corner of the triangle. Creation of a network of “near-
equilateral” triangles is a criteria commonly used in the development of TIN models.
Further, the TIN depends on proper identification of break lines. Break lines must be
identified to the computer in the input array. Once the break lines are appropriately
identified, the break lines become sides of triangles in the network of triangles created
when developing the TIN model. Information from this network of triangles can then
be used in all subsequent data analysis and display.
10.4.4 Automated Contouring Systems
Available contouring software allows a topographic map to be drawn by techni-
cians with no experience with hand-drawn contours. Automated contouring systems
generally make two assumptions concerning TIN models: (1) each triangle side has a
constant slope, and (2) the surface area of any triangle is a plane. Based on these as-
sumptions, elevations of contour crossings are interpolated along triangle edges. After
review of the initial contour map, the user may modify any portion of the resulting
triangulation as needed, and thereby insure that the surface model accurately simulates
the actual ground surface. Although the contouring program forms the surface model
automatically, most programs incorporate some means by which the user designates
the boundary of the surface to be mapped. In some cases, the boundary is assumed by
the program and can be altered if it is incorrect.
Various mathematical techniques are used to approximate field contours. Data in-
terpolation is a common technique that involves interpolation between adjacent grid
points to determine the coordinates of any contour line that may pass between the
points. Without smoothing, however, this method may produce erratic values and con-
tours that are difficult to interpret. Polynomial regression has been used (Portier and
Shih, 1982) to estimate parameters for a mathematical model that produces a smooth,
continuous surface with well-behaved contours. A disadvantage is that none of the
original observations are likely to be found on the fitted surface, however, this may
not be a problem if the goal of the analysis is to obtain a general impression of the
field for making decisions.
The software utilized for developing topographic maps generally consists of a data
editing program, a contouring program, and a computer-aided design (CAD) program.
Within the data-editing program, the data are screened to exclude points such as
benchmarks and other references that are not on the field surface. A 3D model of the
ground surface is formed within the contouring program. From this model the contour
map is then drawn according to the interval selected by the user. Usually, topographic
maps with contour lines using 0.3-m intervals are adequate for description and design.
On very flat or irregular topography, 0.15-m increments may be desirable. When the
slope exceeds 1%, the planner may desire a contour interval of 0.6 m or greater to
have a more readable map. Contouring software is available that enables designers to
produce 3D views of the present (existing) terrain and as it will appear after the pro-
posed modification. These views can be extremely helpful in assisting landowners
and/or project officials to visualize, and assess, the various development alternatives.
The contour map can be output to a computer printer, a plotter, and/or a CAD pro-
gram. It is important that the topographic map illustrates other physical features of the
farm. Usually within a separate CAD program, the designer combines the contour map
with other data that show the location and elevation of benchmarks, location and
source of irrigation water, existing field boundaries, drainage patterns and outlets,
farmstead, farm roads, location of both buried and aboveground utilities, and any other
physical features that may affect the planning of the system and the design of the land
grading for each field.
10.5 EARTHWORK ANALYSES
Design and analysis for land forming include processes for defining the finished
field surface, adjusting for cut and fills to compensate for shrinkage of the soil, and
making the necessary earthwork calculations. At the present time, computers are used
nearly exclusively to optimize cuts and fills. Powerful, easy to use, and relatively in-
expensive software for land forming design is commercially available from vendors of
surveying software. Many consulting engineering firms, federal agencies, and state
universities involved in irrigation development have such programs. Most commercial
software is able to handle irregular field shapes, irregularly spaced data, and other
complex design situations. Cut-fill maps and an analysis of the work and haul patterns
required to achieve the design are important options now available. The following
discussion is intended to provide an overview of design procedures commonly used in
design and analysis of land forming, supplemented with an ample number of refer-
ences. A limited number of computational equations are given to provide examples of
how certain issues are addressed. In most instances, numerous procedures are avail-
able for solving various aspects of the earthwork analysis process; it is left to the inter-
ested readers to decide which procedures best accommodate their needs.
10.5.1 Defining the Finished Field Surface

As noted earlier, the survey data collected can be according to either a grid-based
or irregular pattern. Generally, a mathematical description of the design surface is de-
veloped that meets the functional requirements of the land forming operation. An
equation for a linear plane is used to describe the linear land-grading plane where the
elevation, H(x,y), at any point (x,y) is given by
H ( x, y ) = a + bx + cy (10.2)
where a is the elevation of the plane at the origin of the x and y coordinate system, and
b and c are the two edge slopes of the plane in the x and y directions, respectively.
The best-fit linear plane can be determined by linear regression analysis techniques.
Best-fit approaches are useful, especially, for drainage purposes and when the slopes, b
and c in Equation 10.2, are within tolerances for the irrigation system. In many surface
irrigation design situations, the design plane will differ from the best-fit plane (e.g., the
design plane generally has two fixed design specifications which leaves only one dimen-
sion in which the plane can be changed to balance cut and fill volumes). For example, in
border irrigation there normally is no cross slope in the border strips and the field eleva-
tion at the inlet may be fixed. This means that the plane has only one factor, the longitu-
dinal slope, which can be varied in the design. Level basins, where the basins are de-
signed to be level in both directions (Erie and Dedrick, 1979; also see Chapter 14), are
another example. In this case, elevation is the only factor that can be varied for land
grading. To minimize earthwork volumes, iterative adjustment of the slope or elevation of
the design plane within prescribed limits is done by many earthwork software packages.
Numerous procedures for defining the finished field surface have been based on
uniformly spaced, grid-based patterns. In this work, much of the land grading design
has involved determining a linear plane that balances cut and fill volumes to minimize
earthwork. Further, design procedures for nonlinear surfaces have been presented
(Hamad and Ali, 1990; Ebne-Jalal, 2004). The following paragraphs provide an over-
view of some of the linear plane procedures developed for analyzing grid-based data.
Note that the focus is on minimizing the earthwork requirement.
Givan (1940) introduced the first systematic procedure to perform land-grading de-
signs for rectangular fields. Givan used least squares procedures to define a plane sur-
face that fits the natural ground surface with a minimum of earthwork. Chugg (1947)
developed the first land grading design method for irregularly shaped fields. He ap-
plied the least squares method to calculate the best-fit slope, using transparent and
coordinate papers to simplify and organize the computational process. The procedure
requires the graphical determination of rectangular distances from the origin to the
centroid and a separate determination of the centroid elevation.
Raju (1960) developed a method to calculate the slopes of the graded plane, which
he called the fixed volume center method. His method is based on the criterion that the
total volume of earth and the center of volume will be the same before and after grad-
ing. This assumption is required to ensure a balance between cut and fill and to obtain
the least amount of grading and movement of earth. The fixed volume center method
has proven to be an accurate method of land grading, and although time consuming
when done by hand, can readily be used with computers.
Harris et al. (1966) introduced what they called the best-fit warped surface, in
which they recognized the need for other types of design by suggesting that a field-
surface shape that allows row and cross-row grades to vary within prescribed limits
would be useful, especially when surface drainage was needed. Such a surface has
variable slope in the irrigation direction, while the cross slope follows the natural
ground profile with minor modifications. Design limits are maximum and minimum
row grades and cross slopes, maximum change in row grade and cross slope at a sta-
tion, and the cut-fill ratio.
Shih and Kriz (1971) presented the symmetrical residuals method of land grading
design for both rectangular and irregularly shaped fields. The method is based on re-
sidual properties, Newton’s divided-difference interpolation procedure, and statistical
properties of the best design with an unbiased estimate and minimum variance. Sowell
et al. (1973) applied linear programming techniques to the land-grading designs pre-
sented by Shih and Kriz (1971) and found that the linear programming technique gave
a smaller total sum of depth of cut.
Paul (1973) developed two methods to calculate the slopes of the best-fit plane, the
double-centroid and the computer-minimized cost methods. Hamad (1981) proposed a
method for quick estimation of the volumes of earthwork by using probability theory.
Manela (1983) applied the least squares method to find the best-fit plane. Scalloppi
and Willardson (1986) further developed a practical procedure to calculate the slopes
of a graded plane using the least squares method.
All of the above design methods involved trial and error procedures to determine
the plane that balances cut and fill volumes, considering the shrinkage of the soil, to
minimize earthwork. Easa (1989) presented formulas for direct land-grading design
that eliminates the need for trial and error procedures to minimize earthwork. The
method explicitly considers the required design specifications, which may include the
two edge slopes of the plane, one edge slope and a control point, or two control points.
Such an approach is important in many practical situations in the field where many
times the finished plane must conform to certain boundary conditions.
Procedures also have been developed for analyzing irregularly spaced data. The best-
fit design plane generally can be generated by a modified least-squares technique in
which each data point is weighted in the analysis in proportion to the area it represents.
10.5.2 Computation of Cut and Fill Adjustments
In land shaping it is essential that the volume of material excavated be adequate to make
the fills. If the cuts equal the fills without borrowing or wasting material, the earthwork is
in balance. Experience has shown that the volume of cuts computed from the design must
exceed the fills within a certain percent. The cut-fill volume ratio, R, is defined as
C′
R= (10.3)
F′
where C′ and F′ are the cut and fill volumes, respectively, after the parameters of the
plane have been adjusted.
Cut-fill ratios generally range from 1.2 to 1.6, but can be as low as 1.1 and as high
as 2.0 (Michael, 1978). The cut-fill ratio depends on factors such as soil type and its
moisture content, both of which generally vary with depth of cut; the earthmoving
equipment weight, its tire size, and number of passes made over the soil; and the depth
of the layer (commonly referenced as “lift”) of soil deposited with each pass of the
equipment. All of these factors influence the extent of compaction. The correct estima-
tion of the cut-fill ratio is an important factor in the earthwork design. Shih (1992)
provided a generalized guide for selecting cut-fill ratios dependent upon soil texture,
depth of cut and fill, organic matter content of soil, and roughness of the soil surface
(Table 10.1). Generally, it is best to base this estimate upon the judgment of those hav-
ing local experience with the soils in question. For large projects, it may be appropri-
ate to use a portion of the area as a test of the cut-fill ratio to be able to adjust the
earthwork design for the remainder.
The best-fit linear plane (Equation 10.2) assumes that the cut and fill volumes are
equal. This formula for the best-fit plane can be used directly if the shrinkage of the
soil during leveling is negligible. However, under most conditions, to obtain the de-
sired cut-fill ratio (Equation 10.3) the parameters, a, b, and c, of the design land grad-
ing plane (Equation 10.2) must be adjusted. Clearly, the adjustments of the parameters
of the land-grading plane should have the effect of lowering the plane. Most adjust-
ment procedures have focused on lowering the plane, i.e., reducing a in Equation 10.2
by some amount, Δa, where Δa = a – a′ and a′ is the elevation of the adjusted plane:
1+ R
Δa = (R ∑ F - ∑ C) (10.4)
2 RN t
where Δa = increment the grade line is lowered, m
R = cut-fill ratio
Nt = number of both cut and fills coordinates
ΣC = summation of cuts before lowering the plane, m
ΣF = summation of fills before lowering the plane, m.
A trial and error procedure generally is used to adjust a, b, and c to satisfy R in Equa-
tion 10.3. As noted earlier, Easa (1989) developed approximate adjustment formulas for
all design parameters a, b, and c that can be applied directly to determine the position of
the plane that satisfies the cut-fill ratio. Approximation formulas like those presented by
Easa (1989) are essential when calculations are being done manually and can be a useful
tool within computer programs. Since R is simply an estimate based on experience, the
adequacy of the cut-fill ratio should be checked as the land shaping progresses.
Sometimes it is necessary to use material from the field to construct farm roads or
elevated ditches, or material excavated from a drain may need to be spread on the
field. The adjustment in the plane to allow for this material can be estimated by the
following equation:
V
Δa ′ = K (10.5)
A
where Δa′ = increment the grade line is lowered, m
V = volume of compacted fill in a road or elevated ditch to be taken (“borrowed”)
from the graded surface, or the volume of soil to be distributed (“wasted”)
over the graded surface when in its compacted state, m3
A = area of the field, m2
K = dimensional constant equal to 1.0 for Δa′ in meters.
Table 10.1. General guidelines for selecting cut-fill ratios for land grading design (Shih, 1992).
Soil Depth of Cut Organic Roughness of
Cut-Fill Ratio Texture and Fill Matter Field Surface
Low, 1.10–1.25 coarse deep low smooth
Medium, 1.25–1.40 medium moderate medium moderately smooth
High, 1.40–1.70 fine shallow high rough
10.5.3 Earthwork Calculations

The designer of the land grading program must be concerned with minimizing the
land grading costs while carefully considering the layout of the system, the irrigation
efficiency attainable, labor requirements, cost of operation and maintenance, and
farmability. Thus, the optimum or minimum cost of land forming will be the minimum
cost associated with meeting the above criteria. The volume of excavation, or cut, is
generally used as a basis for contracting and for estimating equipment requirements
and job schedule. Various procedures have been developed to minimize cuts and fills.
In most instances, minimizing cuts and fills does not minimize costs (flat, uniform
fields being an exception). The goal is to minimize the amount of work that must be
done which includes minimizing the volume of cut and haul distance. Type of equip-
ment also can affect the costs of land forming. In some instances, a distance criterion
such as average haul distance is used to supplement the purely volumetric method
generally used. The following paragraphs discuss some of the commonly used meth-
ods for computing volume of cut or fill. They all represent ways to calculate the volu-
metric difference between two surface models, one being the existing or original sur-
face and the second representing the proposed design surface.
The summation method, the least accurate of the methods presented, is used to pro-
vide quick estimates of volumes to be excavated (USDA-SCS, 1961). The method
assumes that a given cut or fill at a grid point represents an area midway to the next
grid point. The grid point is considered as being in the center of the grid rather than at
the corner. The product of the cut in meters and the area of the grid in square meters
provide the required excavation volume in cubic meters.
Where excavation involves the entire grid, the volume of earthwork could be ob-
tained by multiplying the area of the grid by the average cut at the four corners of the
grid. Since cut and fill both often occur within a grid, a procedure known as the four-
point method is used. This method is based on the equations
2
L2 ( ∑ C )
V c= (10.6)
K ∑C + ∑ F
2
L2 ( ∑ F )
Vf= (10.7)
K ∑C + ∑ F
where Vc = volume of cut, m3
Vf = volume of fill, m3
ΣC =sum of cuts on the four corners on a grid square, m
ΣF = sum of fills on the four corners of a grid square, m
L = length of the grid square, m
K = dimensional constant equal to 4 when C, F, and L in m and V in m3.
An accurate method of computing the volume of cut or fill in land forming makes
use of the prismoidal formula,
L
V = ( A 1 + 4 Am + A2 ) (10.8)
6
where V = volume, m3
L = distance between end planes, m
A1 and A2 = end plane areas, m2
Am = middle section plane area, m2.
This formula is easily adapted to computer applications, but it is laborious by hand or

calculator.
Many currently used computer-aided land forming programs determine the soil
volume through the use of a triangulated irregular network (TIN) of elevation differ-
ences between the original and design surfaces (digital terrain model, DTM) to gener-
ate tetrahedrons from which volume can be calculated. The tetrahedron is used since it
is the simplest enclosed figure that can be made from the least number of planes. By
representing the original surface by its survey locations, it preserves the exact eleva-
tions as surveyed. The procedure includes the following steps:
1. Calculate the difference in elevation between the existing ground DTM and the
proposed design surface DTM for all key points. Key points might include all data
points on the existing ground model, defining data points for the design surface (e.g.,
corners of the area to be graded), and various intermediate points that may be formed
by the intersection of triangle sides between the two DTMs. The calculation is done by
projecting each point onto the other surface, interpolating the corresponding elevation,
and calculating the difference.
2. A TIN is generated to represent the DTM created by the elevation differences.
The triangles for the elevation difference TIN are generated honoring all break lines
(e.g., elevations along a ridge and/or along the bottom of a drainage way) and points
common to both DTM surfaces.
3. The TIN is used to determine volumes by breaking the data for each triangle into
the appropriate number of tetrahedrons. The volume of each tetrahedron is calculated
and the volumes are added together to compute the cut and fill volumes between the
original and the design DTMs.
4. A form of the prismoidal formula (Equation 10.8) is used to calculate the volume
of the tetrahedrons. Various solution procedures are used within the earthwork soft-
ware packages to apply the formula. Most procedures are proprietary.
10.6 TYPES OF EQUIPMENT
There is a great variety of equipment that can be used for land forming. In some in-
stances, machinery has been made more efficient, more reliable, more compact, and
can move more soil than formerly. Each type of machine has its own capabilities and
limitations. Over the past 25 to 30 years, laser-control technology has been adapted to
Figure 10.1. GPS-based leveling systems

use signals from satellites to control
a scraper or scrapers as they move
through a field, shown graphically on the
left. The system is anchored in the field by
a base station reference (right).
A display, control unit, and GPS antenna
are mounted on the tractor.
most tools used in land forming; and a class of scrapers that can both carry, and in
some instances, precisely smooth the design surface has been developed. As noted in
Section 10.4, highly accurate global positioning systems have been introduced in the
industry of land forming, and more recently have been adapted to its tools (Figure
10.1). These technologies have significantly impacted the choice of equipment and
how it is used. The examples of equipment described in this section were selected to
illustrate some of the types that are being used (Gattis et al., 1959; Haynes, 1966;
Fisher et al., 1973; Shih, 1992; Wiedemann et al., 1977; Dedrick et al., 1982). All
equipment for land forming must be kept in good repair. The best type and size de-
pends on the particular job.
10.6.1 Power Units
Prior to the 1960s, crawler tractors were used to pull most scrapers. Crawler trac-
tors are still used in special circumstances since they have low soil compaction charac-
teristics, provide excellent traction on a variety of soils, and can move large volumes
of earth when used with 6- to 15-m3 carrier-type scrapers. They are well suited for
short hauls but their slow speed (about 8 km/h) limits their suitability for broad use.
Most of the land forming is now being done with either self-propelled rubber tired
scrapers or scrapers pulled with rubber-tired wheel tractors. Their high field speeds
and maneuverability makes them well suited for this type of work. Two-wheel units
are more maneuverable than four-wheel units. Either can be used with scrapers to
great advantage on land-grading jobs requiring long-distance hauling. They need to be
appropriately matched to the earth-moving equipment as certain combinations are bet-
ter than others.
Bulldozers are suitable for making heavy cuts and moving earth a short distance.
Earth moving distances by bulldozer should not exceed 90 m, but less than 60 m is
preferred. Bulldozers are well-suited for certain rough grading work and are essential
as pusher tractors when extra power is required for loading large scrapers.
10.6.2 Equipment for Land Grading
Most land grading where large quantities of earth are to be moved over appreciable
distances is done with earth-moving scrapers ranging in capacity from about 1.5 m3 to
20 m3. The advent of the fixed-blade scraper has made it much easier to control depths
of cut and fill allowing much larger-volume scrapers to be pulled by smaller power
units (e.g., farm tractors).
Motor graders and blade-type graders pulled by tractors may be used for land form-
ing on small fields, especially for narrow benches or where only minor grading work
is needed. They are used more extensively for shaping ridges on the downhill side of
benches and the slope from one bench to another; excavating relatively small chan-
nels, such as shallow drainage ditches, where the excavated material can be spread or
shaped along one or both sides of the channels; constructing divisions between irri-
gated borders or basins and crowning land; and maintaining shallow drainage ditches
or field laterals.
Many types of scrapers can be, and generally are, equipped with laser and/or GPS
control systems. The laser technology was originally applied in agriculture for auto-
matic grade control for subsurface drainage equipment (Fouss et al., 1964) and was
later adapted to equipment used in land forming. Development of highly precise GPS
technology over the past 10 years has facilitated its use with such equipment. Laser
and GPS precision-guidance systems have revolutionized the process of land forming.
The carrier-type scraper is a tractor-drawn implement that excavates, loads, hauls,

and spreads earth. It consists of a bucket mounted on rubber-tired wheels with a blade
and apron across its front end for cutting, scooping and retaining a load of earth. Car-
rier-type scrapers can be used efficiently on many land-grading jobs and are most effi-
cient where the depth of cut is sufficient for rapid loading. The scraper must be pulled
by the right size and type of tractor. Often, on large earth-moving jobs, these scrapers
are operated in groups with a pusher tractor. Carrier-type scraper sizes range from 2 to
20 m3 capacity with the smaller capacity being designed for use with farm tractors.
Elevating scrapers are used for land grading, especially where intermediate depths
of cut are required. Farm tractors commonly used with four-row equipment can handle
the 4-m3 size. Large farm tractors or contractor-type tractors are used to pull the larger
scrapers. A very desirable feature is the ability of the elevating scraper to load under
most depths of cut and soil textures. In addition, cutting and spreading can be done
evenly.
The ejector scraper, a relatively recent addition to the land grading and smoothing
industry, is a fixed-blade scraper that can be used in place of the elevating scraper in
some instances and is exceptionally useful where relatively small cuts are required
(Figure 10.2). Ejector scrapers range in capacity from a few cubic meters to more than
10 m3 and can be used singly or in tandem. The back wall of the bucket is moveable.
Soil is forced or ejected out of the scraper when hydraulic cylinders push the back wall
forward. The soil flows out of the scraper at an even rate, simplifying the grading op-
eration. Further, the cutting edge can be precisely positioned with laser-control equip-
ment resulting in precisely finished grades. The moveable back wall of the ejector
scraper can be fixed in a forward position to create a land-smoothing device similar to
the bottomless drag scraper discussed below. Thus, the scraper is a tool that can be
used to cut, haul, and fill over relatively large distances while also being used to pro-
vide the final smoothing operation.
10.6.3 Equipment for Land Smoothing
Normally, it is impractical and too expensive to finish land surfaces to exact grade
with heavy earth-moving machines. The heavy scraper work should provide a field
surface such that two or three passes over the field with land smoothing equipment, de-
signed to remove small grade irregularities, will produce the desired uniform surface.
10.6.3.1 Bottomless scrapers. The bucket or bowl of a bottomless scraper has little
or no bottom and earth moving is accomplished by scraping soil from areas that are
above grade, dragging it a short distance, and depositing it in areas that are below
grade. Bottomless scrapers work best in a loose soil that has adequate moisture so that
a smooth, relatively firm surface remains after the smoothing is completed. Dry, pow-
dery soil conditions should be avoided. The field should be watered when such condi-
tions exist, then when the field has sufficiently dried, the surface is worked with a disk
or shallow chisel before the final smoothing is done.
There are two basic types of bottomless scrapers, the land plane and the drag
scraper, which differ in the way in which grade is controlled.
The land plane is a bottomless scraper with a long frame extending both in front of
and in back of the scraper, with wheels mounted at the ends of the frame. The blade on
the land plane is mounted midway on the frame and is adjustable vertically so that the
depth of cut and the amount of earth carried by the bucket can be regulated. Some
Figure 10.2. Ejector scrapers are a type of fixed-blade scraper used for moderate
land grading and smoothing, operated singly (above) and in tandem (below).
planes are equipped with hydraulic controls so that the tractor operator can control the
blade level from the tractor. Other planes are manually adjusted. When in use, the
blade is set at a level that will maintain about one-third to one-half load in the bucket.
If the blade is set too low, the soil in the bucket will become excessive, spilling around
the ends and over the top of the bucket. If it is set too high, there will be insufficient
soil to adequately fill the low areas. When the adjustment is properly made, the plane
will automatically remove the high spots and fill in the depressions of a diameter equal
to nearly half the machine length. Cuts of 6 to 9 cm or less can be made within the
length of the machine.
Land planes are best suited for relatively large fields. They are manufactured in
lengths up to about 27 m and with blade widths up to 4.5 m. From a smoothing stand-
point, the longer the plane the better the smoothing job. From a maneuverability
standpoint, the longer machines require a wider turning area, and thus will leave a
wider strip around the boundaries of the field. Smaller fields will have greater propor-
tions that cannot be properly smoothed. On fields of 8 ha or less and on narrow
benches, smaller and more maneuverable planes are more desirable.
The two-wheel bottomless scraper, commonly known as a drag scraper, can be op-
erated either manually by the tractor operator or by using laser-guided controls. Laser-
controlled drag scrapers (Figure 10.3) and ejector scrapers (Section 10.6.2, Figure
10.2) have become the precision, land smoothing tools of choice. These machines are
manufactured in a number of widths, up to 5.5 m, to serve various purposes. The drag
scraper is well-suited for handling large volumes of earth over a short haul, while the
ejector works for both short and long hauls. Manual smoothing can be done with the
scrapers through visual control by the operator observing the grade to control the qual-
ity of finishing. Through the use of the hydraulic controls, the tractor operator can cut,
drag, and unload as desired. When the drag scraper is used with operator control, it
may be necessary to land plane the area to attain the required results. The laser-
controlled drag scraper will do a more accurate job than a skilled operator using a
manually operated drag scraper and/or land plane. Further, it relieves the operators of
some of their machine control responsibilities, and if combined with available field
Figure 10.3. Laser-controlled drag scraper used for precision surface finishing.
The battery-powered laser transmitter, with solar charger, is shown in the
foreground and the receiving unit is shown mounted on the scraper.
surveying equipment can reduce or eliminate the need for traditional survey crews.
Often manually operated drag scrapers are used before land planing to remove surface
irregularities too minor to be taken care of economically with a carrier-type scraper.
10.6.3.2 Other levelers and floats. Farmers in their farm shops as well as various
commercial companies have made excellent levelers and floats in many variations.
Generally, the earth-moving principles are the same as for land planes, and the effec-
tiveness is dependent on the length of the frame. Levelers and floats that can be pulled
by the average farm tractor are more important in restoring the smoothness of a field
after tillage than in removing the irregularities left by heavy earth-moving equipment.
10.7 FIELD OPERATING PROCEDURES
The land grading work must be planned and scheduled to assure satisfactory field
conditions. Some of the factors that must be considered are timing, weather, and field
condition.
Timing—The field operations must fit into the cropping sequence. When appro-
priate, the grading should be scheduled so that the work can be carried out be-
tween the harvest of one crop and the planting of the next crop. The time avail-
able between crops will help determine the type of equipment to be used.
Weather—While weather conditions cannot always be predicted, there is gener-
ally a time in each region when conditions are most favorable for grading work.
If possible, the work should be planned for the most favorable season to avoid
long shut-down periods. In particular, fieldwork should not be scheduled for pe-
riods of high rainfall, freezing temperatures, or other weather conditions that
may cause excessive delays in earth moving or cause wind or water erosion.
Field conditions—Some field conditions during land grading can have serious
adverse effects on the quality of the work and crop production after it is done.
Grading should not be done when the fields are excessively wet. Operating
scrapers in wet soils is difficult and can cause compaction and other serious
damage to the soil structure, which results in poor soil conditions for crop pro-
duction. Grading of very dry soils should also be avoided. Cuts are difficult to
make, fills become loose and powdery, and the result is poor quality work,
higher cost of operating the equipment, and a dust hazard. Grading with exces-
sive crop residues also should be avoided since it can result in poor compaction
and excessive settling in deeper fill areas.
The quality of the land forming operation depends on the equipment used, condi-
tion of the field, and the skill of the operator. There are probably as many different
ways of approaching the actual earth moving process as there are equipment operators.
The earth-moving work should be carried out so that the grid stakes are not disturbed
until the job is ready for the final smoothing operation. The farmer or contractor is
normally given a map showing the work to be done. The information shown on the
map depends on local custom and contractor preferences. In some cases it may be a
map showing just the amount of cut or fill. A typical cut-and-fill map is shown in Fig-
ure 10.4. Numbers indicate the cut or fill in tenths of a meter. For example, C02 indi-
cates that the cut is to be 0.2 m. If the grid point were marked F01, the fill would be
0.1 m. Some engineers make a detailed study of the contractor’s map and prepare
earthwork balance areas suggesting the direction and location that earth is to be
moved. Most contractors or operators study the map and make their own decision on
Figure 10.4. A typical cut-fill map used by the equipment operator.

Cuts and fills are in tenths of a meter.
Figure 10.5. Field sketch of suggested haul lanes to be used by the land
forming equipment operator to optimize the earthmoving process.
matching cut and fill areas. Figure 10.5 is an example of how the earth movement pat-
terns might be visualized by the operator. On large, more complicated, jobs the haul
patterns may be sketched out on the contractor’s map. In some instances, the contrac-
tor may want a complete map showing original ground elevation, planned finished
grade elevations, and the cut and fill figures. Other procedures include such things as
indicating the cut with red figures and the fill with blue, or by cross-hatching the cut
areas in red and the fill areas in blue.
10.7.1 Land Grading
Cutting and filling is normally done parallel to the grid stake rows. The procedure
varies with operator preference and equipment. In some instances, when the scrapers
are equipped with laser control equipment, field stakes are not used. The operator
merely uses a cut-fill map of the area, some idea as to the most efficient method of
going about the grading job, and uses the laser controls to provide the grade guidance
needed. If the grade stakes are used, some operators cut and fill a strip the width of the
scraper adjacent to the stake rows. The same procedure is followed for the opposite
side of the lane along the next row of stakes. When these are brought approximately to
grade, work is carried on in the intermediate area. When done without laser control,
the grade between stake rows is carried visually across the lane. This process is con-
tinued until the entire field has been shaped. Another procedure is to make a cut along
the stake row as deep as can be made on the first pass. The second strip is made just
far enough over so that the tractor can straddle the strip that is left. The strip that was
left is then removed as a third strip. This procedure is repeated across the cut area.
Regardless of the procedure followed, the cuts and fills should be made in uniform
layers. This reduces uneven settling of the field, facilitates carrying the proper grade
from one stake row to the other, permits faster travel, and is easier on equipment. Nor-
mally the small area around each stake is left until the shaping is completed. If deep
cuts are made leaving a large, high mound at the stake, it may be desirable to remove
the stake, cut away the mound and replace the stake. Cut areas should be disked as
soon as possible after the cuts are made to keep the cut surface from drying and be-
coming hard. This will make the preparations for the final smoothing easier.
Normally grades are checked with a permissible tolerance of ±60 mm. On level or
flat grades it may be necessary to check within ±30 mm. In no case should reverse
grades in the direction of irrigation be permitted. When the field construction has been
approved, the stakes should be removed, the mounds and depressions at the grid stakes
graded to the level of the adjacent surface, and the field disked or chiseled to prepare it
for the final smoothing operation.
10.7.2 Land Smoothing
10.7.2.1 Land plane. Since the cut-and-fill operation by the heavy earth-moving
equipment is normally carried out parallel to the rows of the grid stakes, it is generally
desirable to carry out the first planing operations diagonally to the grid pattern. It is cus-
tomary to land plane the field in three directions. The first two operations are in diagonal
directions, perpendicular to each other, and the third is in the direction of irrigation. Dur-
ing the smoothing operations, considerable time is lost in making the turns at the field
boundaries. Figure 10.6 shows how the first two diagonal planing operations can be car-
ried out to cover the field with two passes of the plane in the minimum amount of time.
This can reduce the time required by about 25% over planing the entire field in one di-
agonal direction and then repeating the operation at right angles to the first operation.
Figure 10.6. A method of performing two diagonal

land planings in which distance traveled is minimized.
10.7.2.2 Drag and ejector scrapers. In recent years, the drag scraper or the ejector
scraper with the moveable back of the scraper in the forward position, has become an
excellent final smoothing machine, with laser control becoming state of the art after
first being introduced in the mid-1970s as a tool to assist in land forming. Figure 10.3
shows a scraper and laser equipment in the field. The laser transmitter is mounted on a
tripod, special stand, or trailer. It generates a thin laser beam that is rotated rapidly to
form a virtual plane of light over the working area. The plane either can be level (for
level basins) or sloping. The maximum range of the laser is about 300 m, but dust,
wind, and heat can be limiting. However, these do not affect GPS-controlled systems,
which have been adapted to land forming tools like the drag and ejector scrapers.
There are many personal preferences as to how the smoothing operation will be
completed. If there are significant high and low areas, a typical approach is to use a
field map as a guide to move soil from high to low areas. Once the field is close to on
grade, or if there are no significant high and low areas to start with, the operator nor-
mally runs the scraper over the field in two or three directions with the last pass in the
direction of irrigation. Fields smoothed in this way normally do not require further
smoothing (e.g., land planing not required).
As noted earlier, land smoothing can be described statistically by the variability of
the field elevations around the design slope. Field elevations are normally distributed;
thus, the standard deviation of the field elevations from the design slope is an appro-
priate criterion for assessing smoothing adequacy. With well-trained operators and
carefully adjusted equipment, standard deviations of the finished field elevations from
the targeted field plane finished with laser-controlled scrapers range from about 12 to
15 mm on 4 to 16 ha fields (Dedrick, 1979; Dedrick et al., 1982). On level basins in
the southwest, NRCS standards require 80% of a basin to be within +15 mm and
100% within ±30 mm. To meet the 80% criteria the standard deviation must be
slightly less than 12 mm. Typically, standard deviations on “level” basins prior to la-
ser-controlled smoothing ranged from 25 to 30 mm.
10.7.2.3 Water leveling. Water leveling for rice land is a method of field benching
that was used extensively in the past, but in the United States it essentially has been
replaced with laser-controlled scrapers. Rice is normally grown on soil having a re-
strictive layer with a permeability rate of about 0.5 mm or less per hour. If this restric-
tive layer is thick enough to support a tractor and scraper under a water-saturated con-
dition, water leveling can be used. The equipment commonly used for water leveling
is a bottomless scraper or a blade mounted on the back of a farm tractor. The general
procedure is as follows:
Smooth the field with a land plane to remove as many of the surface irregulari-
ties as possible.
Locate and construct the permanent levees. Levees should be straightened by
filling across the low areas.
Flood as much area as can be leveled in one day to a depth that will cover the
high point.
Carry out the leveling work using the tractor-mounted blade or the bottomless
scraper. Tractor movement produces water waves that assist the leveling opera-
tion. More than one tractor is often used to speed up the work and to create more
wave action.
Soil movement can be determined by water depth measurements.
After the suspended soil material has settled and the water has become clear, the
remaining water should be drained off.
When the soil has dried sufficiently, the area between levees should be worked
and smoothed with a land plane.
In some rice production areas, land forming is done to reduce the number of levees
and to construct parallel levees. Final smoothing of the areas could be done with either
water-leveling techniques or laser-controlled scrapers. Faulkner (1964) provides more
detailed information on water leveling.
10.7.3 Maintenance
Land forming often requires a large investment of capital. Annual maintenance is
an important part of the farmer’s operation to obtain the proper return on this invest-
ment. Special treatment and annual maintenance is needed to bring land back to full
production and keep the surface uniform. To improve soil condition after smoothing,
tillage (e.g., plowing, disking, or chiseling) mixes the disturbed soil and loosens areas
that have been compacted. In cold climates, leaving the soil surface in a roughened
condition over winter will allow the frost action to break down clods and improve in-
filtration. If land forming has exposed infertile subsoils, soil testing is recommended.
Reclamation practices might include a green manure crop the first year to add organic
matter and/or applications of barnyard manure. These reclamation procedures may
need to be supplemented with the application of fertilizer to meet crop requirements.
Trace minerals, such as iron or zinc, are sometimes deficient in cut areas.
Following land shaping, irrigation, and farming operations, fill areas tend to settle
and cut areas “fluff up” leaving the surface uneven. In some instances, final smoothing
after major land forming may not be justified until the land has been irrigated. To pro-
vide an opportunity to refinish the area, farmers generally plant an annual crop the first
year after land forming. The extent of the continuing maintenance procedures will
depend on several factors such as the type of tillage equipment that is being used, time
since major land forming processes were completed, type of crop to be grown, value
of the crop to be grown, and the susceptibility of the crop to damage caused by soil
surface irregularities and associated nonuniform water application.
Cultivation should be done carefully to help maintain precisely smoothed surfaces.

Plowing should be done with a two-way plow. In some cases, sweeps or chisels are
used to cut through the soil, raising and dropping the soil to provide a loosening effect.
10.8 COST AND CONTRACTING
The cost of earth work in land forming will vary with a number of factors: topogra-
phy, field conditions, length of haul, total volume and volume of earth movement per
unit area, type of equipment used, and the operator’s skill. Competition among con-
tractors and the need for off-season work often affects the cost.
In most areas the major earthwork is contracted based upon a volumetric basis. Fi-
nal smoothing generally is contracted on a per unit area basis. In some instances,
where the contractor does both the major land moving and finishing, the quoted price
per unit volume includes final smoothing.
Most contracting is done on a verbal agreement between the contractor and the
farmer or landowner. This sometimes leads to misunderstandings. The following items
are suggested as some of the things that should be covered by a written agreement:
field location and description of site conditions;
basis for the construction such as the contractor’s work map, special construc-
tion not shown on the map, allowance for settlement in heavy fill areas, cut-to-
fill ratio, etc.;
facilities, materials, and assistance which the farmer will provide to the contractor;
who will do the smoothing and when it is to be done, if the final smoothing is to
be done as a separate job;
completion date with provisions for extension if factors beyond the contractor’s
control delay the progress of the work;
basis for payment (if based on volume, the method to be used for determining
the volume and who is going to do the computation should be stated; agreement
should be made on how the field will be checked for completion);
liability insurance;
responsibility for locating and protecting buried utilities;
final clean-up of rubbish, excess materials and equipment, broken equipment,
etc; and
pay schedule.
10.9 SAFETY
All equipment must have the required safety devices and should be operated in a
reasonable and safe manner. Safety shields must be in place and securely fastened.
Special precautions should be taken when operating equipment around electric power
lines, buried utilities, and communication lines. Buried utilities and communication
lines have been damaged and lives lost during land grading operations. The general
location of the buried utility should be indicated on the contractor’s map. Most utility
companies will, on request, stake the exact location of their utility line and indicate the
depth of the cover over the line. While some governmental agencies serve written no-
tice to the farmer that he is responsible for notifying the utility company of the work to
be done on his farm, all parties involved in the land forming must be safety conscious.
Dust associated with land forming can be a serious problem, especially during and
after the final smoothing operation. Dust affects nearby residents, creates hazardous
conditions on nearby roads, and can damage crops on adjacent farmland. As noted
earlier, land forming on dry soils should be avoided, both from a quality of job stand-
point (i.e., loose soil is difficult to smooth) and from a dust (health and safety) stand-
point. In some instances, temporary irrigation facilities, such as portable sprinklers,
may need to be used to provide better working conditions. If soil moisture is adequate,
working the field with small chisels or a spring-tooth harrow may provide sufficient
surface roughening to control blowing dust. The finished field should be irrigated and
planted as soon as possible. If the planting of the next crop is to be delayed for any
length of time, it may be advisable to plant a cover crop to provide adequate protection
from wind erosion.
REFERENCES
Anderson, C. L., A. D. Halderman, H. A. Paul, and E. Rapp. 1980. Land shaping
requirements. In Design and Operation of Farm Irrigation Systems, 281-314. M. E.
Jensen, ed. St. Joseph, Mich.: ASAE.
Buras, N., and G. Manor. 1970. Field comparison of land smoothers. Trans. ASAE 13:
639-640, 643.
Chugg, G. E. 1947. Calculations for land gradation. Agric. Eng. 28(11): 461-463.
Dedrick, A. R. 1979. Land leveling-precision attained with laser-controlled drag
scrapers. ASAE Paper No. 79-2565. St. Joseph, Mich.: ASAE.
Dedrick, A. R., L. J. Erie, and A. J. Clemmens. 1982. Level basin irrigation. In
Advances in Irrigation, 105-145. D. I. Hillel, ed. Academic Press.
Dedrick, A. R. 1989. Surface-drained level basin systems. In Proc. Int. Conf. on
Irrigation: Theory & Practice, 566-575. Southampton, U.K.
Easa, S. M. 1989. Direct land grading design of irrigation plane surfaces. J. Irrig.
Drain. Eng. 115(2): 285-301.
Ebne-Jalal, R. 2004. Weighted average method for land grading design. J. Irrig.
Drain. Eng. 130(3): 239-247.
Erie, L. J., and A. R. Dedrick. 1979. Level-basin irrigation: A method of conserving
water and labor. USDA Farmers’ Bulletin No. 2261.Washington, D.C.: USDA.
Faulkner, M. D. 1964. Leveling Rice Land in Water. Crowley, La.: Louisiana Sta.
Univ., Rice Exp. Sta.
Fisher, C. E., H. T. Wiedemann, C. H. Meadors, and J. H. Brock. 1973. Chapt. 7:
Mechanical control of mesquite. In Mesquite-Growth and Development,
Management Economics, Control, Uses. Res. Mono. 1, Texas Agricultural
Energetics. Westport, Conn.: AVI.
Fouss, J. L., R. G. Holmes, and G. O. Schwab. 1964. Automatic grade control for
subsurface drainage equipment. Trans. ASAE 7(2): 111-113.
Gattis, J. L., K. A. Koch, and J. L. McVey. 1959. Land grading for surface irrigation.
Agric. Ext. Serv. Circ. 491. Little Rock, Ark.: Univ. Arkansas.
Givan, C. V. 1940. Land grading calculations. Agric. Eng. 21(1): 11-12.
Hamad, S. N. 1981. Using probability theory in land grading earthwork. Al-Muhandis
80: 23-28.
Hamad, S. N., and A. M. Ali. 1990. Land-grading design by using nonlinear
programming. J. Irrig. Drain. Eng. 116(2): 219-226.
Harris, W. S. 1971. Crop yields as affected by row grades. Trans. ASAE 14: 860-863.
Harris, W. S., J. C. Wait, and R. H. Benedict. 1966. Warped-surface method of land
grading. Trans. ASAE 9: 65-65.
Haynes, H. D. 1966. Machinery and methods for constructing and maintaining surface
drainage on farm lands in humid areas. Trans. ASAE 9: 185-188, 193.
Kriz, G. J., L. D. Hunning, R. E. Sneed, and S. F. Shih. 1973. Land-forming designs
for water management. J. Soil Water Cons. 28: 134-136.
Lyles, L. 1967. Effect of dryland leveling on soil moisture storage and grain sorghum
yield. Trans. ASAE 10: 523-525.
Manela, M. 1983. Construcao de planos geometricos em sistematizacao de solos pelo
metodo dos minimos quadralos generalizados. Irrigacao e Tecnologia Moderna 10:
32-34, 36-37.
Michael, A. 1978. Irrigation: Theory and Practice. Vikas, New Delhi, India: Vikas
Pvt Ltd.
Paul, H. A. 1973. Theoretical derivation and economic evaluation of the double
centroid and computer-minimized cost methods of calculating slopes for land
grading. Thesis. Logan, Utah: Utah State Univ.
Portier, K. M., and S. F. Shih. 1982. Spline fitting in land-forming design. Soil Crop-
Sci. Soc. Fla. Proc. 41: 39-43.
Quackenbush, T. H. 1967. Land modification for efficient use of water. J. Irrig. Drain.
Eng. 93: 7014.
Raju, V. S. 1960. Land grading for irrigation. Trans. ASAE 3(1): 38-41, 54.
Scalloppi, E. J., and L. S. Willardson. 1986. Practical land grading based upon least
squares. J. Irrig. Drain. Eng.112(2): 98-109.
Sewell, J. I. 1970. Land grading for improved surface drainage. Trans. ASAE 13: 817-
819.
Shih, S. F. 1992 Energy requirements for land clearing, forming, and ditching. Energy
in World Agric. 6: 101-116.
Shih, S. F., and G. J. Kriz. 1971. Symmetrical residuals method for land forming
design. Trans. ASAE 14(6): 1195-1200.
Shih, S. F., D. E. Vandergrift, D. L. Myhre, G. S. Rahi, and D.S. Harrison. 1981. The
effect of land forming on subsidence in the Florida Everglades’ organic soil. Soil
Sci. Soc. America J. 45: 1206-1209.
Sowell, R. S., S. F. Shih, and G. J. Kriz. 1973. Land-forming design by linear
programming. Trans. ASAE 16: 296-301.
Springer, G. 1966. Land reshaping for cherry orchards. Trans. ASAE 9: 292-294.
Thomas, D. J., and D. K. Cassel. 1979. Land-forming Atlantic Coastal Plain soils:
Crop yield relationships to soil physical and chemical properties. J. Soil Water
Cons. 34: 20-24.
USDA-SCS. 1961. Chapt. 12: Land leveling. In National Engineering Handbook,
Sect. 15. Washington, D.C.: USDA Soil Conservation Service.
Wiedemann, H. T., B. T. Cross, and C. E. Fisher. 1977. Low-energy grubber for
controlling brush. Trans. ASAE 20: 210-213.
CHAPTER 11
DELIVERY AND
DISTRIBUTION SYSTEMS
John A. Replogle (ARS-USDA,
Maricopa, Arizona)
E. Gordon Kruse (ARS-USDA,
Abstract. The supply of irrigation water may be on-farm or delivered to the farm
through a water-user association or irrigation district from more distant, often more
reliable, and larger water supplies. We summarize farm water delivery policies and
their impacts on farm labor, crops, and on-farm investment needed to achieve effective
water management. Water delivery usually implies that several farms share access to
the water. The irrigation schedule, including flow rate, duration of access, and the
time to access the flow again, can be rigid or flexible. Hardware items, including
automatic gates, methods of flow metering, and systems used to convey irrigation wa-
ter from the farm supply to the fields are described. Canal and pipeline water meas-
urement is presented. Pipelines or lined ditches offer greater seepage control, ease of
water diversion, and reduced maintenance, but have a higher initial cost than earthen
ditches. There is a tendency for mechanizing and automating supply, distribution, and
application systems in order to deliver water with a flexible schedule. Requirements of
system management and equipment, including reservoir use and management, struc-
tures and water conveyance design, system operation and maintenance, and mosquito
control, are presented.
Keywords. Canals, Ditches, Flow control, Gates, Irrigation district, Irrigation
management, Irrigation scheduling, Low-pressure pipe systems, Mosquito control,
Pipelines, Reservior, Water management.
11.1 INTRODUCTION
When the supply of irrigation water exists on the farm, either from a well (as are an
estimated 42% of the water supplies for irrigated farms) or directly diverted from a
stream (as are an estimated 16%), the farm operator can usually control the timing and
quantity of the delivery. Most other farm water supplies are surface waters delivered
to the farms through a system operated by a water-user association or irrigation dis-
trict, whose purpose is to facilitate water delivery, application, and removal. Such or-
ganizations give the farmer access to more distant, often more reliable, and larger water
supplies by sharing the construction and operating costs with many other users, but there
are usually more restrictions and external control of amounts delivered and flow rates.
348 Chapter 11 Delivery and Distribution Systems
The first portion of this chapter summarizes the various forms of farm water deliv-
ery policies and their impacts on farm labor, crops, and on-farm investment needed to
achieve effective water management. Delivery policies and methods are classified in
terms of distinguishing physical parameters. Physical features of the systems and
hardware items, like automatic gates and methods of flow metering, that can signifi-
cantly contribute to their improved operation are described. The delivery system is
purposely viewed as a subsystem of the farm production scheme, as part of the total
water handling cost of the farm unit, because the farmer ultimately pays essentially all
costs. Improvements and new developments are stressed.
The remainder of the chapter describes systems used to convey irrigation water
from the farm supply—a canal, reservoir, or well—to the fields. Such systems gener-
ally have a capacity less than about 0.3 m3/s. Ditches or pipelines used for such con-
veyance should be well designed and maintained to minimize water loss by seepage
and to allow efficient irrigation. Generally, pipelines or lined ditches offer greater
seepage control, ease of water diversion, and reduced maintenance, but have a higher
initial cost than earthen ditches.
11.2 IRRIGATION WATER DELIVERY
Water delivery through canals or pipelines usually implies that several farms must
somehow share access to the water in terms of flow rate, duration of access, and the
return time to access the flow again. The canal or pipeline system must accommodate
these three factors. The implied goal is to provide water at the right time in the right
amount and at the right place. Canals and pipelines within the farm must then be oper-
ated to accommodate the delivery restrictions.
11.2.1 Schedules
The delivery policy used by the water supply organization to meet this three-part
implied goal is called a scheduling system or schedule. This scheduling system may
have been established by law or based on economic and engineering factors that are
important, or were important when the organization was established.
Schedules consist of the three factors: delivery flow rate to the farm, delivery fre-
quency to the farm, and delivery duration. The relations between these three factors
control the capital cost and operating expenses of the delivery system. The same fac-
tors bear heavily on the effective and economical use of water, labor, energy, and capi-
tal investment on the farm. They may even limit the crops that can be considered.
An appreciation of the operational range available for different types of schedule
can be obtained by examining the three scheduling components in more detail.
Flow rate—For the delivery flow rate, policy may specify a constant flow at
each delivery cycle throughout the season. One limit is a continuous, unchang-
ing, uninterrupted flow. Variations would include continuous but seasonally
modified flow rates. The more common cycled flow is usually in some kind of
rotation system among users. The canal delivery equipment associated with con-
stant deliveries is frequently a fixed orifice or fixed pipe size. Seasonal changes
in flow rate to users can be implemented within narrow limits by adjusting the
canal water surface elevation in some instances.
Frequency—Delivery frequency, like flow rate, can vary widely. Water can be
delivered periodically throughout the season, and this period can be changed
seasonally. For example, the cycle may repeat at two-week intervals in the hot
season and be changed to three- or four-week intervals in the cooler seasons.
Duration—Delivery duration, the length of time that flow is allowed through the
farm gate or valve, can vary from continuous throughout the season to a few
minutes or hours per irrigation cycle. In some projects, all farms on a canal take
small flows for the entire period the canal is operating. The canal may be filled
on alternate weeks. Other projects deliver in 24-hour units or in some other
standard but fixed time block.
An additional consideration is who controls the rate, frequency, and duration. This
can be local farm-operator control, central-authority control or some combination.
11.2.1.1 Schedule types. Schedule types may be broadly classified as either rigid
(predetermined) schedules or flexible (modifiable) schedules. Variations are summa-
rized in Table 11.1, with decreasing flexibility from top to bottom. Descriptions of the
three components, how they are handled, and who controls each component are useful
for understanding the advantages and limitations. In general, the farm water delivery
policy must be compatible with the distribution system on the farm. In many cases this
has required on-farm designs to be adapted to an existing delivery system with the net
result that the total production system is severely limited. Often, economical im-
provements in the design and operation of on-farm systems are made possible by ap-
propriate modifications in the water delivery system. Thus, improvements in the dis-
tribution system, although they may increase costs, may be justified by more effective
water use and reduced on-farm labor, capital, or energy costs.
A system of schedule naming was proposed by Replogle and Merriam (1980). The
names for various schedules differed slightly from names commonly used, but not
always with the same meanings among users. The proposed names were subsequently
revised by Replogle (1984, 1987). Clemmens (1987) grouped the schedules according
to the degree of flexibility provided to the farmers in terms of rate, frequency, and
duration, and the level within the delivery system at which irrigation delivery deci-
sions are made. He otherwise retained most of the naming system described above.
Table 11.1 summarizes some of the more prominent schedule combinations and the
source of control. The constraints on each of the schedules are indicated to help pro-
vide an understanding of the possible implications of comparative schedules.
11.2.1.2 Rigid-schedule irrigation systems. Rotation (rigid) schedules can some-
times improve irrigation efficiencies in areas where farm operators are new to irriga-
tion technology, compared to the free access of a flexible system, which could be mis-
used. Rotation schedules require relatively low capital investment in canals or distri-
bution pipelines and involve the least water agency management. The rotation sched-
ules have one of the lowest staffing requirements in relation to the capital investments
in the system. Field tasks may be simply to open or close a valve or gate as instructed,
with no precision regulation required. Without on-farm water storage, this schedule
forces the farm operators to operate at low water use efficiency. The irrigator may
have to forego crops that must be watered often, require precise timing of applications
(some vegetable and hay crops), or require careful water management to discourage
pests and diseases.
The only communication required between farm operator and water agency is a post-
ing (or other general unilateral notice) of the schedule, a desirable simplicity in areas
with numerous small farms and limited telephone service. However, demand systems
(see Section 11.2.1.3 below) also have limited communication requirements because
they allow the farm operator to access the water supply without notice to anyone.
Table 11.1. Definitions of delivery scheduling methods, after Clemmons (1987).

System Constraints
Schedule Categories Rate Frequency Duration
Local control
Demand schedules
Demand U U U
Limited rate demand L U U
Arranged frequency demand L A U
Intermediate Control
Arranged schedules
Arranged A A A
Limited rate arranged L A A
Restricted arranged C A C
Fixed duration arranged C A F
Fixed rate/restricted arranged F A C
Central control
Central system schedules
Central system V V V
Fixed amount V F F
Rotation schedules
Rotation F F F
Varied amount rotation F F(V) F(V)
Varied frequency rotation F F(V) F(V)
Continuous flow F(V)
Terminology:
U: Unlimited, no restriction, under user control.
L: Limited to maximum flow rate, but still arranged.
A: Arranged between user and water authority.
C: Constant during irrigation as arranged.
F: Fixed by central policy.
V: Varied by central authority, at authority’s discretion.
(V): Varied by central authority, seasonally, by policy.
The use of rigid rotation-scheduling policies should at best be considered as an in-

terim delivery policy and efforts should be made to direct rehabilitation, redesign, and
education towards providing better-informed users and more flexible delivery sched-
ules that can respond to soil- and crop-water needs and conserve water and energy.
Rigid-schedule systems can sometimes be made more effective if some flexibility
in water delivery can be introduced. Farm operators, for instance, might make their
needs known through a preseason meeting with the water delivery authorities. Thus,
these listed schedules could be further categorized as central-controlled schedules or
schedules for which the farm operator has some input, but not enough day-to-day op-
erational input to be classified among the arranged-schedule groups.
Typical flow rates of rigid-schedule systems are from 15 to 30 L/s; typical frequen-
cies are 7 to 14 days; and typical durations are 8, 12, or 24 hours. The larger flows are
usually in canals and the smaller flows are frequently in low-pressure pipelines. Limit-
ing cases, as mentioned above, are the continuous flow schedule, and durations that
last as long as flow is in the canal.
11.2.1.3 Flexible-schedule irrigation systems. Demand (flexible) schedules, in
which the irrigator may have water as desired—flexible in frequency, rate, and dura-
tion—are often too expensive, especially if large rates or complete time flexibility are
offered. The capacity of such a system would be too large for reasonable capitalization
and operating costs, because it must be large enough to meet the combined probable
demand of all the users at any one time.
Thus, the actual schedule is a limited-rate demand schedule, which has often been
called a modified-demand schedule. It allows flexible rates, up to some maximum, as
well as flexible durations and frequencies.
In some irrigation districts, requests must be made from 1 to 3 or 4 days in advance
of the desired delivery date. The maximum rate is set fairly high to economically use
labor on the farm, but it is limited by the economics of the total farm and distribution
systems. With this limited-rate demand schedule the rate may be varied by the irriga-
tor during irrigation, and the irrigator controls the duration within the requested 24-h
period.
Obviously, this schedule requires operational spillage, storage, or automation of the
supply because the irrigator may vary the flows. Regulating reservoirs established at
several locations along the mains or at heads of laterals are suitable. They can be filled
at night when on-farm demand drops off. This permits mains and sub-mains to operate
near peak capacity most of the time. When sub-mains and laterals are automated by
use of level-top, float-controlled canals, or closed or semi-closed pipelines, operations
become very simple (see Section 11.2.3). Main canals need to be automated so that
gates at all checks can be simultaneously set to pass any flow demanded any place in
the system. Peak flow demands are met from the strategically placed reservoirs so
main canal flows are relatively stable. Non-automated, sloping main canals can also
function satisfactorily if they have reservoirs to accumulate water that would other-
wise be spilled. They must have adequate flow to satisfy the day-to-day peak demand
in the upper reach, and an adequate automated service area below the reservoirs to
utilize the operational variations. On-line system reservoirs may be located slightly
above or below the canal and the necessary flows pumped into or out of them. The
lifts and quantities are often small. The grade may be reduced on a steep canal to pro-
vide the elevation differences needed for the reservoir either by reconstruction or by
adding check structures.
The limited-rate demand schedule is highly practical for the water user. It enables
the farm operator to irrigate each crop when needed, and to use a stream size that is
economical and efficient for the particular situation. Thus, differences in soils and
crop requirements can normally be accommodated. Limitations on crop types that can
be successfully grown are reduced. A ditch rider, who is an agency employee, may
need to be on call 24 hours a day to reduce operational spillage as the farm operator
changes his flow rate or finishes irrigating, unless the system has automated canals or
closed or semi-closed pipelines, and suitable reservoirs.
Nearly all large projects in the northwestern U.S. operate on a type of arranged
schedule. In some cases it is possible to deliver water on a straight demand schedule
during part of the season, and change to one of the arranged systems or even to a rota-
tion schedule during peak crop-growth periods. The demand schedules are practical
when the discharge from the supply system has adequate capacity, can be varied, and
has automated canals and/or closed or semi-closed pipelines. A totalizing meter meas-
ures the volume of water, but the irrigator controls the flow rate and duration (Mer-
riam, 1973, 1977). With a fully adequate capacity, it is not necessary to schedule the
delivery day, thus greatly reducing operation and maintenance costs.
11.2.2 Flow Control and Measurement to Farms

Good irrigation water management requires knowledge of the quantity of water
available. Water at rest (i.e., stored water) is measured in units of volume, such as cu-
bic meters (m3). Water in motion is measured in units of flow rate, as volume per unit
time, such as cubic meters per second (m3/s) or liters per second (L/s). Many methods
are available for measuring water deliveries to farms and for measuring rates and vol-
umes of water on-farm, but accuracy varies widely. Generally, flow meters can be
classified into two major groups: those for measuring the total quantity of flow (total-
izing meters) and those for measuring flow rate (rate meters). With appropriate auxil-
iary equipment, meters from either group can be converted to the other group.
11.2.2.1 Totalizing meters. The most common totalizing meters are the propeller
meters that measure flow in pipelines. Direct totalizing meters for open channels are
rare. One example is the Dethridge meter, which is used mainly in Australia to meas-
ure accumulated delivered volume flows in canals. More recently, open channel me-
ters based on use of ultrasonics can provide flow rate and accumulated volume and are
thus totalizing meters.
Propeller meters. Propeller meters are often used to measure discharges from wells
and for pipeline deliveries to farms. In such service, they are usually fitted to show
both totalized flow quantity and flow rate. These meters use a multi-bladed propeller
with two to six blades, made of metal, plastic, or rubber, that drive the indicating de-
vice to give any desired volumetric and rate units. Modern meters may use magnetic
coupling between the propeller and the indicator, and the indicator may be electronic
rather than mechanical. The meters are designed and calibrated for operation in closed
conduits, usually round pipes, that are flowing full. The propeller diameter usually
ranges from 50% to 80% of the pipe diameter. Meters are commercially available for
pipe sizes ranging from 0.05 to 1.8 m in diameter. The effective low-to-high flow
range for an individual meter is about 1:10. Ordinarily, these meters will operate
within flow velocities ranging from 0.15 to 5.0 m/s.
The principle involved is simply counting the revolutions of the propeller as the
water passes, the rate of revolution being proportional to rate of flow. Changes in the
pattern of flow approaching the meter or changes in the bearing friction or the resis-
tance of the propeller surfaces affect the accuracy. Routine accuracy of a well-
maintained propeller meter is usually about ± 5%. Higher accuracy may sometimes be
achieved, particularly if calibrated in place. Replogle (1999) and Replogle and Wahlin
(2000) describe one method of checking propeller meters in systems with a free out-
flow.
Propeller meters are sometimes used to measure flow in irrigation canals by direct-
ing the flow to be measured through a length of pipe that is flowing full. Often a sub-
merged culvert or pipe turnout serves this purpose. Because of debris problems, the
meter may be placed in the flow just long enough to get a valid measurement of flow
rate and then removed. Flow rate is then assumed to be constant throughout the deliv-
ery period to produce totalized flow. However, it may be left in place for the entire
time of a delivery and actually totalize the flow, as may be needed for the limited-rate
demand schedule.
Dethridge meters. Of primarily historical interest, Dethridge meters were designed
by J. S. Dethridge of Australia in 1910 (Dethridge, 1913) and have since been used
extensively on that continent (Bos, 1989; Replogle, 1970). A Dethridge meter is basi-
cally an undershot waterwheel turned by the canal water passing through its emplace-
ment, a short concrete section specially shaped to provide only minimum practical
clearance of the lower portion of the wheel. The Dethridge meter registers volume of
delivered water. It is simple and sturdy in construction and operates with a small head
loss. These are being phased out in favor of rate meters with totalizing and rate indica-
tion in the form of overshot weirs, on the argument that the Dethridge meter is a safety
hazard. It has not been used, to the authors’ knowledge, in the U.S., probably because
of its bulkiness and safety concerns. This is a direct totalizing meter for open channels,
but flow rate can be obtained by using a stopwatch or other time records.
11.2.2.2 Rate meters for canal-system deliveries. Rate meters that are most
commonly used in canal-system deliveries to farms are orifices, weirs, and flumes.
Orifice, Venturi, ultrasonic, electromagnetic, and vortex meters are available for pipe-
line deliveries. Vortex and electromagnetic meters are not commonly used in irrigation
practices. For a discussion of these, refer to USBR (1997).
Orifices. Orifices are used for both canal flow measurements and in pipelines. Ori-
fices are openings, of well-defined area, in a plate or bulkhead, the tops of which are
well below the upstream water level in the case of canal flows. If the orifice dis-
charges into the air, it is a free-discharging orifice. If it discharges under water, it is a
submerged orifice. Free-discharging orifices require considerable head and are usually
limited to special locations where head loss is not critical. Submerged orifices require
less head and are more commonly used. Orifice shapes vary widely. Circular, sharp-
edged orifices are easily machined. Segmental orifices, which can be made by inserting a
plate part way into the side of a pipe to form semicircular openings, are used in some
pump discharge pipes because they will pass both air and sand if properly oriented.
In practice, circular, sharp-edged orifices placed into canal service are fully con-
tracted so that the bed and sides of the approach channel and the upstream water sur-
face are remote enough to avoid influencing the discharging jet. Usually orifices
formed by partly opening a gate are suppressed from one or more sides and may require
special rating procedures to be sufficiently accurate. The term “suppressed” means that
the contraction of the flow does not occur, that is, the contraction is suppressed.
A general disadvantage of orifices is debris clogging that prevents accurate meas-
urements. It is frequently difficult to tell if partial clogging has occurred, especially in
sediment-laden water. To prevent canal overtopping, the elevation of the top of the
orifice wall or other emergency overflow should be at or slightly below the maximum
expected upstream water level so that flows can be safely bypassed if the orifice plugs.
The head-discharge relation for circular, sharp-edged, submerged orifices is:
Q = C d A 2 g ( h1 − h2 ) (11.1)
3
where Q = discharge rate, m /s
Cd = discharge coefficient, ranging between 0.57 and 0.61
A = area of the orifice opening, m2
g = gravitational constant, 9.8 m/s2
(h1– h2) = head differential across the orifice, m.
For freely discharging orifices, h2 = zero and hl is measured from the center of the
orifice. Orifices should be installed and maintained so that the velocity of approach is
negligible. Circular orifices can produce accuracies within about ± 1% if properly ma-
chined, installed, and maintained.
Other orifice devices. Orifices are much less sensitive to head variations than over-
fall weirs and flumes because they respond only to the square root of the head change.
A variation that is even less sensitive is the NEYRTEC module gate, which delivers
water at a nearly constant flow rate, even though the head may vary considerably
(Goussard, 1987; Bos, 1989). Fixed guide vanes cause flow that modifies the vena
contracta in the module, reducing or increasing the area of the exit flow enough to
approximately compensate for changing head within a limited range. These modules
are equipped with a separate slide panel that acts as a simple on-off gate. This gate is
not intended for flow regulation. Flow regulation is accomplished by changing the
number of parallel mounted modules.
The constant-head orifice (CHO) is a combination regulating and measuring struc-
ture used on turnouts to farms. It usually consists of two sluice-type gates in tandem,
installed about 1 m apart. The CHO was developed by the U.S. Bureau of Reclamation
(USBR, 1997) and is so named because it is operated by adjusting the downstream
gate to maintain a constant head differential (h1 – h2), across the orifice formed by the
upstream gate. The differential head is usually 0.06 m. The discharge will then be at
the required value to an accuracy of about ± 7%. Because of the small head differ-
ences, extreme care is needed for setting the required head readings.
Sharp-crested weirs. Sharp-crested weirs (Brater and King, 1976; USBR, 1997) are
overflow structures made from thin plates whose thickness at the overfall edge is less
than 2 mm. The weir opening is usually rectangular, triangular, or trapezoidal, al-
though other shapes have been used. These weirs are accurate, i.e., will repeat their
calibration to within 1% or 2%, especially the rectangular and triangular types. The
trapezoidal versions are considered to be reliable to ± 5%. The most popular trapezoi-
dal style is the Cipoletti weir with 0.25:1 side slopes (horizontal:vertical). To obtain
high accuracy, the nappe (which is the falling sheet of wather springing from the weir
plate) must be ventilated on the downstream side of the weir plate so that the pressure
there will be atmospheric. If the weir is not ventilated, the resulting vacuum will cause
variable and undetermined over-discharging. The downstream water level should be
kept at least 5 cm below the crest of the weir. This means that the required head loss
always exceeds by 5 cm the upstream head over the weir crest. This required high
head loss is one of the major limitations to using sharp-crested weirs for irrigation
flows, especially in relatively flat irrigated areas.
The discharge over rectangular sharp-crested weirs can be approximated with a
power function of the form
Q = CLH 3 / 2 (11.2)
where C incorporates the gravitational constant, the discharge coefficient, and other
effects; L is the weir crest length; and H is the total energy head on the weir (the flow
depth plus the approach velocity energy). Methods for determining C are given by
Brater and King (1976). For design purposes, C = 1.66 for SI units or 3.0 for foot-
pound (customary) units. Methods to more precisely define K for rectangular sharp-
crested weirs are also given by Kindsvater and Carter (1959) and Bos (1989), and in-
clude other shapes of opening and the effects of weir suppression.
Critical-flow flumes. Critical-flow flumes are rate meters particularly suited for
measuring irrigation water in open channels. They consist of a specified constriction
built into the canal floor and/or sidewalls that raises the upstream flow level above that
which would have existed without the constriction. This rise must create enough drop
to cause critical flow in the contracted section, or throat, of the flume, and is fre-
quently less than 10% or 15% of the original flow depth. This makes them attractive
for measuring canal flows in many relatively flat irrigated areas of the world.
Of historical interest, because of their wide applications for more than a half-
century, are Parshall flumes (Parshall, 1950; USBR, 1997), followed by H-flumes
(Bos, 1989; USDA, 1979), cutthroat flumes (Skogerboe et al., 1972), and special
trapezoidal and triangular flumes (Robinson, 1968). These styles, however, permit
curvilinear flow to occur in the control section, making them hydraulically complex
with their calibrations depending on laboratory ratings. Additionally, most of the
trapezoidal and triangular flumes require depth measurement in a converging flow
section, which is not favored by modern hydraulics practice. Because of the complex
hydraulic behavior and method of head sensing, these flumes must be carefully con-
structed throughout their length, and can become relatively expensive devices.
Another general type of flume is the long-throated flume, which may be contracted
from the sides and bottom, and which includes some broad-crested weirs that are con-
tracted from the bottom with rounded upstream edges. These have been used for many
years in Europe (Ackers and Harrison, 1963; Ackers et al., 1978; Bos, 1989). These
are called long-throated flumes because the contraction length is about twice the
maximum flow depth for styles resembling broad-crested weirs, causing nearly paral-
lel flow as it passes through the flume throat. For side contractions only, the throat
length may need to be based on a length-to-width ratio. This is because the shock
waves from the contraction at the edges must cross to the center of the flow for the
central region to “know” that the side contraction exists. A study of this effect is pre-
sented in Wang et al. (2005). This shock wave propagates at about 45° across the
channel, which would suggest a minimum throat length exceeding one throat width.
The cross-sectional shape can be almost any regular geometric pattern—
rectangular, triangular, trapezoidal, circular, parabolic, or complex. The flow depth is
sensed in an upstream channel section, again, in an area of parallel flow. The stage-
discharge ratings are based partly on calculated relations, combined with laboratory-
determined discharge coefficients, and usually meet the desired ± 5% accuracy re-
quirement demanded of the flumes. More recent calibration procedures, involving
mathematical modeling, velocity distribution, and considering boundary layer devel-
opment caused by frictional effects, improve the reliability of these calibrations to
within ± 2% (Bos et al., 1991; Clemmens et al., 2001). Simplified field installation
techniques have greatly decreased the skill requirements of field construction. Individ-
ual discharge ratings can be restored to the original intended accuracy in spite of most
field construction anomalies.
An important aspect of flumes is the high submergence limit that can be tolerated
as compared with sharp-crested weirs. (Submergence limit is defined here as that point
of submergence caused by downstream water level at which the real discharge devi-
ates by 1% from that indicated by the observed flow stage.) A high submergence limit
minimizes required head loss for the measuring device in the canal system. Because
sharp-crested weirs can withstand no submergence, the minimum required head loss is
equal to the flow depth over the weir plus about 5 cm, to assure free flow. Flumes, on
the other hand, have submergence limits that range from about 50% to as high as 95%.
This corresponds to absolute head losses through flumes of about 5% to 50% of the
canal flow depth.
The submergence limit of critical-flow devices, like flumes and weirs, is a function
of the streamline curvature at the control section and the losses in the downstream
expansion. Broad-crested weirs and long-throated flumes, which have straight and
parallel streamlines at the control section, can approach a submergence limit of 95%,
if part of the kinetic energy is recovered by a proper exit expansion (Bos, 1989). The
submergence limits for truncated long-throated flumes, i.e., flumes that have no ex-
pansion section, are about 70% to 85%, depending on whether the throat area ap-
proximates a rectangular or a triangular shape. These submergence limits, even for
truncated versions, are higher than those routinely listed for H-flumes, cutthroat
flumes, and Parshall flumes, which are 25%, 65%, and 70%, respectively (Chow,
1959; Clemmens et al., 2001) for the usual sizes of flumes on farms.
The older flumes, however, are quite usable, and many are in continual service and
most certainly need not be replaced if they were properly installed and subsequently
maintained. Abt et al. (1989) and Abt and Staker (1990) present ratings corrections for
flumes that have settled out of level, either longitudinally or laterally. Accurate and
much less expensive designs are now available for new installations. These have the
added capability of being tailored to most existing channels. Many open channels can
now be retrofitted with a low head-loss measuring device that can operate at high
submergence with a single depth reading.
When installing any of the critical-flow flumes, care must be taken to assure ade-
quate free flow discharge at both high and low ranges. Checking only at high flow is
risky. For example, a rectangular flume placed in a trapezoidal channel, where the
channel, rather than a check gate, controls the flow level, may be free flowing at high
stages but submerged at low flows (Replogle, 1968). Usually, combinations of these
shapes need to be checked for submergence at low flows. On the other hand, a trape-
zoidal or triangular flume placed into a rectangular channel will require checking the
calculations at high flows and will tend to become less submerged at lower flows.
Rectangular flumes, including the Parshall and the cutthroat, are sometimes installed
with a vertical offset to avoid submergence at low flows (Replogle, 1968; Humphreys
and Bondurant, 1977), at the expense of higher head loss at maximum flow.
The trapezoidal, broad-crested weir with upstream approach ramps is a variation on
the long-throated flume. It is particularly suited for the common, small concrete ca-
nals, constructed in a continuous manner with sliding forms, which are called slip-
formed canals. Such weirs can inexpensively and accurately measure irrigation flows.
They are usually truncated at the downstream end of the weir sill, although an exit
ramp can improve the submergence limit from about 85% to 95% in some configura-
tions. Clemmens et al. (2001) present design information for broad-crested weirs in
trapezoidal and other channel shapes.
Analysis shows that the error in discharge is almost proportional to the error in
constructed sill width (Replogle, 1977; Replogle et al., 1997; Clemmens and Replogle,
1980), and corrections to the rating tables due to these types of construction anomalies
can usually be accurately adjusted.
A rectangular broad-crested weir discharges nearly equal quantities of water over
equal widths. The flow is nearly two-dimensional over the weir, so that rating tables
can provide the flow rate, q, in m3/s per meter width of sill for each value of h1. De-
sign and calibration information for these weirs is given in Clemmens et al. (2001) and
Wahl et al. (2005).
Portable flumes for flow rates up to about 50 L/s are described in Bos et al. (1991)
for trapezoidal and rectangular cross-sections. Adjustable sill-height flumes are now
available that allow easy installation at an estimated elevation and applying post-
installation adjustment to flow conditions. They can be used in both lined and unlined
small canals. Commercial offerings are available in sizes up to 1000 L/s, although
only the smaller sizes of less than about 300 L/s maximum capacity are considered
portable.
Simple constructions for small trapezoidal canals. The structural requirements of
broad-crested weirs of the long-throated type described above allows simple construc-
tion methods in the common-sized trapezoidal canals with 30- and 60-cm bottom
widths, 1:1 side slopes and capacities on the order of 500 L/s. Construction sugges-
tions are shown in Figure 11.1. The concrete requirements are quantities that can be
hand mixed.
Structures for circular channels. The broad-crested style of long-throated flume has
been adapted for flow measurements in circular channels and pipes with a free water
surface because of its relatively simple construction. Other shapes, such as side con-
tractions and center line piers, have been used. The reader is referred to the computer
program and design procedure for most calibrations (Clemmens et al., 1993a, and the
update to Windows XP provided in Wahl et al., 2005). There are two situations of
particular interest for designing a weir in a pipe. The weir can be placed somewhere in
the middle of a straight pipe, or at the end of a pipe, such as the entrance to a deep
manhole, which is probably more common because the site is more accessible.
When the weir is placed in the interior of a straight pipe, the presence of the weir
must cause a rise in the upstream water surface to develop the required head loss. This
increase in flow area upstream causes a proportional decrease in velocity and may
cause subsequent sediment deposition. Where this is a problem, a downstream ramp
should be considered to reduce the required head loss and thus accommodate a smaller
sill height. Sediment problems are further aggravated at low flows because the free-
flowing (normal depth) water surface drops proportionally faster than the water sur-
face upstream from the weir. Thus, the velocity difference becomes greater and greater
with decreasing discharge. For situations where wide fluctuations in flow rates exist
and sedimentation is a problem, an alternative weir shape or location should be con-
sidered.
An alternative for the measuring location is the end of a pipe, particularly where
there is a large drop in water surface. A weir can often be designed so that the water
level upstream from the weir either matches or is below the normal depth of water in
the pipe. This may avoid aggravating sedimentation problems.
Portable flumes using plastic pipe. Circular pipes can also be used as the basic for-
mat for portable and permanent structures (Figure 11.2). For portable use, the translo-
cated stilling well is convenient because it renders the device less sensitive to longitu-
dinal and transverse leveling. The installation and use of the translocated stilling well
is described in detail in USBR (1997), and is similar to that shown in Figure 11.2. A
broad-crested weir, usually 0.2 to 0.5 times the pipe diameter in height, is placed in the
pipe. The free-flow limiting depth in the pipe is about 0.9 times the pipe diameter. The
most convenient approach to the general calibration computations is application
Drain tube
allows canal to
empty without
pumping. Usually
needs cleaning
with a rod
when drainage
is needed.
1. Select a concrete lined 2. Fill with field soil to within 5 to 10 cm 3. Put steel angle-iron forms in place.
canal that is free of major (2 to 4") of finished broad-crested weir. These must be carefully shimmed to
cracks. Place a drain tube Temporarily form a downstream level in all directions. See below for
(about 3 cm (1") plastic) in canal. ramp with the soil even if a concrete example construction of these forms,
which are left in the concrete.
Concrete thickness
Reinforcing
bars may
be added
Soil
to ramp. 3
should
1 wash away
from under ramp
when canal flow starts.
4. Assure that the forms stay 5. Pour the ramp on about a 3:1 6. Mount sidewall gauge at proper location
level while filling with concrete. slope. This slope is not critical. and elevation. Usually set elevation of most
used flow rate. Check zero and evaluate if
gage and canal wall slope are well enough
matched (±3 mm ( ±1/8")) is usually acceptable)
et
ra ck Wall-gage mounting
gB bracket made from
tin
un
e Mo galvanized sheet metal.
Wall gauge can be made using
ug
Ga Attach to canal wall with
screws. a chisel and punch to make the
marks and numbers. A paper pattern
useful as a chisel guide.
Use 4 cm (1-1/2 ")

angle or
larger.
Cut ends
to match
canal side slope.
After sliding gauge into

Form for sill bracket to proper vertical
Need be only
concrete
approximate
to lock it tightly.
Make as accurate
as practical
Figure 11.1. Simple construction method for long-throated flumes in small farm canals.
of the WinFlume calibration software for long-throated flumes of many cross-

sectional shapes presented in Clemmens et al. (2001) and updated in Wahl et al.
(2005). The following design and calibration information was produced using the
WinFlume computer model and its predecessors.
Qcfs = ( D ft )2.5 K1, ft ( h1, ft / D ft ) + K 2 , ft )U (11.3)
Stilling Well Location
To
h1
D y1 Stilling y2
Well p1
La Lb L
bc
Figure 11.2. Broad-crested weir for flow measurement in a circular culvert.
where Qcf s = discharge, ft3/s (for SI units, multiply Qcfs by 28.32 to obtain L/s)
Dft = diameter of pipe, ft (multiply SI units in m by 3.28 to obtain ft)
K1,ft = constant from Table 11.2
K2,ft = constant from Table 11.2
h1,ft = head measured from sill top (bottom of contracted section), ft (multiply SI
units in m by 3.28 to obtain ft)
U = exponent.
As mentioned, the flumes represented in Table 11.2 are Froude models of each
other. These and all the long-throated flume shapes can be similarly scaled without
using the computer model, as long as all dimensions remain proportional. Small dif-
ferences from direct computer results are to be expected because the roughness of con-
struction materials is not usually scaled. Smooth concrete roughness was used to de-
velop the values in Table 11.2. The calibration equations from Table 11.2 will usually
perform satisfactorily to within ± 3%, not counting the error of the readout detection
method, for scale expansions or contractions of about 5. Scale expansions of 10 tend
to over-emphasize roughness and will under-predict discharge by 5% to 10%. Accu-
racy at the ± 2% range requires individual computation of the constructed device using
the computer model of Wahl et al. (2005).
As with the other broad-crested weirs, and flumes, the width of the sill surface, w,
is one of the two most important dimensions in the flume. The other is the accurate
referencing of the zero elevation of the head-measuring device.
For portable measuring weirs, the translocated stilling well, which references the
upstream head to the sill floor without the necessity of accurately leveling the flume, is
recommended. This minimizes errors in head determination even if the weir is not
exactly level (Bos et al., 1991; Wahl et al., 2005). The upstream gage should be used
only if the flume is accurately leveled or is permanently installed.
Table 11.2. Equation parameters for sills in circular flumes. These values represent
minimum lengths in direction of flow and may be increased 30% with only slight change
in calibration, and should be suitable for most applications.[a]
Pre-Gage Dist., Lpg ≥ hmax Sill Height = p1/D
Approach, Lap ≥ hmax hmin = 0.07 D
Converging, Lcv = 3 p1 hmax = [0.85 D – p1]
Control, Lcn ≥ 1.5 D – p1
K1,ft K2,ft Qrange/D5/2
p1/D La/D Lb/D L/D (ft) (ft) U hrange/D (ft3/s) bc/D
0.20 0.65 0.60 1.200 4.130 0.004 1.736 0.080-0.65 0.056–1.975 0.800
0.25 0.60 0.75 1.125 3.970 0.004 1.689 0.070-0.60 0.048–1.689 0.866
0.30 0.55 0.90 1.050 3.780 0.0 1.625 0.070-0.55 0.050–1.434 0.917
0.35 0.50 1.05 0.975 3.641 0.0 1.597 0.065-0.50 0.046–1.202 0.954
0.40 0.45 1.20 0.900 3.507 0.0 1.573 0.060-0.45 0.042–0.991 0.980
0.45 0.40 1.35 0.825 3.378 0.0 1.554 0.055-0.40 0.037–0.807 0.995
0.50 0.35 1.50 0.750 3.251 0.0 1.540 0.050-0.35 0.032–0.640 1.000
[a]
The values in the table are only valid for foot-pound (customary) units. To use SI units, first convert D
and h1 from m to ft by multiplying by 3.28 ft/m for entry into table, then convert the resulting Qcfs to m3/s
or L/s. Alternately, use the computer program of Wahl et al. (2005).
In Figure 11.2, the stilling well is placed at the usual location. However, it can be
placed at another location, such as near the outlet, provided the sensing of depth at the
gage location is accurately transmitted to the alternate location. It can even be placed
in the channel if it does not significantly obstruct flow or divert flow to one side and
cause erosion. Placing it in the upstream channel often causes detrimental flow pat-
terns that can affect the function of the flow-measuring device, unless it is dug deep
into the bank or placed at substantial distance upstream.
Alternately, a static head tube, consisting of several holes about 3 mm in diameter
drilled into a length of 2- or 3-cm diameter PVC pipe, can be used to monitor the pres-
sure at the gage location as shown in Figure 11.3. These holes are about 30 cm from
the end of the capped pipe so that flow separation around the end of the pipe is negli-
gible by the time the flow passes the pressure sensing holes. The water level sensed
here is transmitted to the stilling well where the depth can be observed by any of the
several flow depth methods discussed elsewhere. The system shown can be used as the
primary readout method or it can be used to properly set the zero for another depth-
sensing device. If a static pressure tube is a permanent part of, for example, a stilling
well placed downstream, then the tube should be clamped tightly to the wall of the
flume so that debris trapping is minimized, and well above the floor level to reduce
sediment plugging. The area obstruction of the pipe crossing the sill control area is
small and can usually be ignored, except in small flumes where the pipe represents
more than about 1% of the flow area.
Head reading, h1detected
Inlet holes for in suspended cup
static head tube
h1 h1
SIDE VIEW END VIEW

Figure 11.3. Using a static head tube for a portable flow-measuring system in a half pipe.
11.2.2.3 Rate meters for pipe system deliveries. Rate meters commonly used in
irrigation pipes are the propeller meters mentioned in 11.2.2.1, which can also totalize
flow. Differential-head meters, such as the orifice meter for pipe flows, operate on the
principles described in section 11.2.2.2.1. Other selected meters described herein in-
clude Venturi meters and special ultrasonic meter applications.
Venturi meters. Venturi meters are applicable to irrigation pipeline flows but are
not widely used for routine farm turnout use because their length and the straight-pipe
requirements usually make them impractical. Despite this, they represent one of the
older and more reliable of the differential head meters. The head loss is relatively low,
and they can pass slurries readily. In irrigation works, small Venturi meters can be
used to create a vacuum for chemical injection applications. In sizes compatible with
irrigation wells, they are usually considered too costly. The flow range is narrow but
similar to that of the orifice meters (about 3 to 1, high- to low-flow rate). These de-
vices are well covered in the literature. The reader is referred to ASME (1971), and to
Brater and King (1976), for a more complete treatment.
Field-constructed Venturi meters. Special Venturi meters constructed from plastic
pipe fittings address the economic limitations of commercially available meters, foul-
ing problems, and water management requirements. Individual calibration of each
constructed meter is not required for effective applications in irrigation practice. Vari-
ability between constructions is usually within ± 2% and the head loss is between 18%
and 25% of head reading (compared to 10% to 14% for the “standard” Herschel-type
Venturi tube) (ASME, 1971). The discharge coefficient is also slightly less than the
Herschel-type tube. A constant value of Cd = 0.95 is suitable for most irrigation appli-
cations. More complete details are given in Replogle and Wahlin (1994). General con-
struction details are shown in Figure 11.4.
App ly suc tion or

pressure to ad just Clamp
the differential
Pre-m arked flow
pressure to a
m ea surem ent stic k
c onvenient
rea ding level
No bubbles Standard Venturi

in pressure
tube s
Air vent
Plastic -Pipe Venturi
(o ptiona l)
Figure 11.4. Construction of plastic-pipe Venturi meter showing

differential head arrangement for direct discharge indication.
Venturi meters and orifice meters can be made more convenient for the user if a scale
is designed to indicate flow rate directly. For any given meter, a scale can be produced
that is used by placing the bottom at the water level in one leg of a manometer and read-
ing flow rate directly at the level of the other leg. No subtracting of readings and no table
look-up is needed (Replogle and Wahlin, 1994). The liquid levels in the manometer can
be raised or lowered to suit the observer without changing the differential (Figure 11.4).
Ultrasonic meters in irrigation applications. An exception to high costs that usu-
ally preclude on-farm applications is a single-path sonic meter in 76-cm concrete pipe
used on farm turnouts in some irrigation districts in Arizona. Such meters operated
satisfactorily in most installations (Replogle et al., 1990). Large random error fluctua-
tions were associated with partly opened vertical inlet gates combined with typically
45° horizontal pipe bends that tended to send jets in a spiral motion to the meter sec-
tion downstream. Three-minute averages were required to produce acceptable flow
rate readings. Totalized flow volume over 24 h or longer using a 16-s sample time
every 15 min was sufficiently accurate because the instantaneous errors were random
and tended to cancel. However, observing a single reading that was often 15% (and
once observed to exceed 30%) of average flow was not acceptable for on-farm water
management use. Hydraulic fixes, such as installing large-diameter orifices with open-
ings of 85% to 90% of the pipe diameter downstream from the last pipe bend before
the meter, appeared to cause enough cross mixing to mitigate the jets. The resulting
flow readings were more stable, fluctuating less than 5%.
Ultrasonic Doppler meters. In the early 2000s several commercial sources began
promoting ultrasonic Doppler meters that can be used in both canals and pipelines.
These devices and commercial offerings are changing rapidly, and the reader is ad-
vised to check manufacturers’ literature for operational applications and costs.
11.2.3 Delivery Systems Suited to Mechanized Irrigated Systems
With the increased appreciation of the value of higher on-farm irrigation efficiency
to save water, labor, and energy, there is an increased tendency for mechanizing and
automating supply, distribution, and application systems. To obtain these savings, the
whole system must be able to deliver water in a way that is flexible in frequency, rate,
and duration. For best operation, the flow must be controlled at the point of applica-
tion by automatic controls, or mechanized so that flow is easily controlled when the
irrigator is making on-the-spot decisions. Automatic systems are usually thought of as
being capable of initiating application of water to fields when irrigation is needed,
sequencing sets, and shutting off flow when irrigation is complete. Semiautomatic
(sometimes called mechanized) systems may need an operator’s periodic attention to
initiate a sequence or to reset components, such as drop gates.
The ability to control the rate and duration of flow at the point of application is essen-
tial to obtain high irrigation efficiency. For surface methods, large stream sizes reduce
the application time and labor requirements. For all irrigation methods, it is essential that
the water be turned off when the proper depth has been applied, because additional water
is wasted. Automated application systems have in-field controls responsive to flow vol-
umes, head, duration, soil moisture, etc., to apply the desired irrigation. Therefore, the
delivery system must also automatically make the desired flow conditions available.
Such automated application systems require a fair amount of supervision and main-
tenance. They require a large capital investment but have low nonsupervisory labor
requirements. Their flow capacity requirements are moderately large. The systems
may receive nearly continuous use because around-the-clock and weekend operations
are less likely to be avoided by users.
The semiautomated systems may require the irrigator to be in the field to operate
the controls, adjust the flow rates, head, durations, etc., in a fashion that can be pre-
planned as is done for the automated systems. Moreover, while applying water, the
irrigator can make on-the-spot decisions, repairs, etc. The in-field application equip-
ment for semiautomation consists of simple, standard types of controls requiring little
maintenance or supervision. Such systems require larger flow capacity because night-
time operation should be avoided, and the stream size should be as large as the irriga-
tor can handle with good manual equipment and design. Semiautomatic systems can
allow unattended nighttime irrigation, but may require the operator to check the opera-
tion, reset gates, or move components in the evening and again the next morning.
Some manual systems may approach or exceed the convenience of semiautomated
systems, for example, 2- to 5-ha level basins supplied through manually operated lift-
gates. These are especially practical on soils with low intake rates, require moderate
capital investment, and have low supervisory and moderately low non-supervisory
labor requirements (Dedrick et al., 1978; Dedrick, 1984, 1986, 1989, 1990; Dedrick
and Reinink, 1987; Clemmens, 1992; Clemmens et al., 1993b). Because of their large
capacity and low per-hectare labor requirement, they justify a well-trained and well-
paid irrigator. On average, one irrigator is employed to irrigate about 150 to 200 ha. In
a special arrangement of level basins, made possible by gently sloping topography in
southern Colorado, one irrigator serviced about 700 ha with manually operated gates
(Dedrick, 1984).
Highly automated microprocessor-controlled systems, like the liftgate method with
air cylinder actuators (Dedrick et al., 1978), require more capital investment and seem
most easily justified economically when replacing systems where very frequent irriga-
tions are necessary, using large quantities of labor and small flow rates, or where flow
rates are large relative to set size and application depth and water might need to be
switched at such short intervals that the irrigator would not have efficient blocks of
time for other chores if gates had to be opened and closed manually.
Mechanization without full automation is more readily justified with more normal
frequencies and larger flow rates, like the system mentioned above. A low-cost system
widely used in New Zealand consists mainly of dropgate structures tripped by manu-
ally preset timers (Stoker, 1978). Under favorable topographical and soil conditions,
as much as 100 ha can be irrigated per hour of labor with this system.
The on-farm water delivery system for both the automated and semiautomated sys-
tems usually consists of closed or semi-closed pipelines (Merriam, 1973, 1987a,b) or a
level-top canal with automatic controls (Merriam, 1977; Burt, 1987). Water should be
supplied from a flexible source like a reservoir if delivery is to operate effectively.
Closed pipeline systems are easy to operate but line pressure changes as the water
level in the reservoir fluctuates and as flow rates are changed. The latter causes head
variations due to increased or decreased friction loss. These variations require changes
in the field settings for high efficiencies. Outlet controls must be capable of regulating
flows at the high static pressure that occurs when flow is stopped, as well as the lower
pressure when the pipeline is flowing at capacity. When there are outlets to several
laterals that can be used simultaneously, the pipe sizes may be reduced in the down-
stream direction.
Semi-closed pipeline systems have the pressure broken into steps of 6.0 m or less
by downstream water level float-controlled valves. Other water level control systems
exist but are more sophisticated and expensive and are seldom justified for on-farm
distribution systems. The semi-closed systems may be used to automatically provide
variable flow, to downstream rate requirements, and still have small (0.1 to 0.3 m)
head variations at turnouts. The float valves eliminate water hammer, and head loss at
each float stand is about 1.5 times the velocity head for full flow.
Level-top ditches can deliver water effectively to automatic or semiautomatic on-
farm application systems. Such ditches permit a no-flow condition to exist. The no-
flow or variable flow condition is controlled by float-actuated gates maintaining con-
stant downstream water levels, or by more sophisticated devices better adapted to lar-
ger canals (Burt, 1987). These systems are actuated by variations in the water level
downstream created by the variable flow needs in the field. The bottom of the ditch
may be parallel to the level top or have some slope. In either case, the water surface
level decreases at a turnout, creating a negative wave at critical velocity back to the
gate, which then supplies the desired flow rate. The initial withdrawal pending the gate
operation is taken from canal storage as the water level drops.
The canal’s cross-section usually corresponds to a design velocity of about 0.3 to
1 m/s. The level reaches can be a kilometer long for typical on-farm conditions, and
longer on large canals. The length of these reaches must be shortened on steep gradients.
Mechanical gates for controlling flows in such ditches are available from the French
company ALSTHOM with their NEYRTEC Design equipment. Their AVIS gates are
suitable for canal drops of up to about 1 m (Goussard, 1987). Their AVIO gates for res-
ervoir outlets can control heads up to about 3 m. For greater heads, it may be feasible to
use a float valve system like the Harris float valve described by Merriam (1973, 1987b).
Canals controlled by the NEYRTEC gates have small variations in flow depth and
thus require only a small freeboard. Because of the reduced freeboard over normal
operations, canal costs are only a little greater than for low-gradient canal systems less
than about 1 km long. The additional cost for gates makes possible the automatic op-
eration of the delivery systems. Flow measurements need to be made by totalizing
meters because flow rates are expected to vary.
Microprocessors show promise of economical application to canal flow controls
and can be expected to improve canal operational capabilities. Labor, water, and en-
ergy savings are possible when additional capital is invested to provide automation or
semiautomation of the application system and to modify the delivery system to pro-
vide for the needed flexibility in frequency, rate, and duration (see Section 11.2.1).
11.2.4 Reservoirs in Farm Water Delivery Systems
Under many circumstances, small reservoirs are essential elements in a complete
farm water delivery system. They may be built and owned by the using farm or may
be designed and built by the irrigation district for canal regulation. They may be a
practical means for upgrading the scheduling system that can be offered by a district.
For example, to upgrade a fixed-amount, fixed-frequency schedule system, which
would characteristically have laterals too small to directly upgrade to one of the de-
mand schedule systems, strategically placed, off-canal reservoirs can frequently be
constructed at less cost than enlarging and expanding existing canals and right-of-way.
This would be especially true if the existing canal were already lined, and relining
costs were to be considered for a larger canal section.
Basically, reservoirs are management facilities whereby a variable or steady flow

can be regulated into a different variable or steady flow that is more convenient or
efficient in terms of water, labor, energy, and crop production. They may be used in
connection with pump or gravity systems and with all irrigation methods, and are of-
ten combined with other facilities like pipelines, wells, pumpback systems, and canal
delivery systems.
11.2.4.1 Reservoir use and management. Classification of reservoirs may be done
by major function: (1) long-term storage, (2) temporary storage, (3) “overnight” stor-
age, (4) regulation, (5) tailwater reuse, and (6) water table control. Farm reservoirs are
often designed to serve several of the above purposes as well as providing water for
frost control, fire protection, domestic use, spraying operations, or recreation.
Long-term storage reservoirs are usually associated with project operation and
year-to-year carryover. They have large capacity and basically are needed to provide
flexibility in frequency. Their occurrence on-farm is rare though they have been eco-
nomically used in conjunction with low-capacity wells where water can be pumped in
the spring season, stored, and then used to meet peak summer demands.
Temporary storage facilities are commonly used in the more humid areas to collect
runoff from rainfall for use during occasional dry periods or for frost or temperature
control. They are often used to provide a source for irrigation systems in areas where
rainfed cropping is possible but where irrigation can improve crop quality and yield.
“Overnight” reservoirs are common. They accumulate pump or canal deliveries for
24 h, more or less, to permit applications that are variable in rate and duration. They
may have a capacity as small as that needed to hold 12 to 16 h of flow from a well or
small surface source with the irrigation applied for only part of the 24-h period. How-
ever, this small size lacks reserve for unusual conditions and a larger capacity for up to
48-h storage is frequently more desirable, particularly in conjunction with a semiau-
tomated surface system. They often facilitate the continuous pumping of a well or the
24-h delivery flow from a water agency. The economical size is a function of the dura-
tion of irrigation, the area to be covered in a set, the capacity of the distribution sys-
tem, the economics of labor, and the water supply rate and duration of flow.
Regulating reservoirs are usually small. Their function is to permit relatively small
variations in flow and/or duration. They may be needed to accept flow from a water
agency or a well to absorb small variations in flow to sprinklers as the length of lateral
changes or time to move reduces flow needs for short periods. One may be needed
adjacent to a well supplying a sprinkler system so that variations in pumping rate due
to a falling water table can be equalized by the booster system to provide a steady rate
and a constant pressure. One may conveniently be used to accumulate enough water to
provide the larger initial stream desired for furrows. When the reservoir is empty, the
cutback stream size would be satisfied by the basic source such as a well, and the res-
ervoir could be used to accumulate the tailwater to be ready for the next set. Extra
capacity over the minimum volume is good design.
Tailwater reuse systems require a sump or reservoir at the lower side of a field. The
capacity needed may vary from complete storage of the runoff from one or more fields
to a small sump for a recycling system. If a cycling pump from a small sump is used it
should pump to a larger reservoir. If the pumped water is added to the irrigation
stream, the runoff that is already occurring will be aggravated.
A reservoir constructed to help maintain a water table is sometimes used in subirri-
gation systems or with ponded crops such as rice. These have no specific storage re-
quirement. Evaporation may cause significant water loss if the surface area is large.
Classification of reservoirs may also be done by location: gravity reservoirs serve
the irrigated area by gravity flow while low level reservoirs require pumped delivery.
Preferably, an irrigation reservoir is located at a high point on the farm so that all flow
can be taken out by gravity, particularly for surface irrigation systems. When water
supply agencies deliver water at an elevation below the desired water surface of a res-
ervoir, a small pump to fill the reservoir may be desirable. If the reservoir supplies
sprinklers, it might be most economical to let it fill by gravity. Then a single pumping
system can empty the reservoir and pressurize the sprinklers. The cost of pumping
may be compensated by the value of the irrigation of additional areas or by the labor
and water saved through more efficient application.
On large farms where topography permits, it may be practical to have two reser-
voirs at the same elevation at the upslope corners (more or less), interconnected by a
pipeline or a level-top ditch designed to permit flow in either direction. This permits
combining deliveries to the middle to obtain large flows for efficient labor utilization
from smaller, less expensive conveyance. A similar condition using flow to the center
from two ends exists when the reservoir is at the high side and a well is located at an
alternate side.
A gravity reservoir, in conjunction with large capacity closed or semi-closed pipe-
lines or level-top ditches, provides a source for deliveries that are flexible in rate and
duration and provides large stream sizes that greatly reduce labor costs.
Low-level reservoirs may permit the utilization of low swampy areas, may be part
of a tailwater reuse system, or may accept gravity delivery from low-head supply ca-
nals or groundwater or surface flows. In shallow water table areas, they may eliminate
the need for lining to control seepage losses. The use of a pump to deliver the water
from the reservoir restricts the flexibility of irrigation stream size, but permits a wider
choice of delivery location. When a low or swampy area otherwise unusable for agri-
culture is converted to a reservoir, the reservoir can sometimes be managed to enhance
wildlife habitat, albeit this may be at the expense of wildlife otherwise supported by
the swamp.
Depending upon the head, pumps used for surface irrigation deliveries from farm
reservoirs are usually low lift, high volume propeller pumps or centrifugal pumps.
They may feed directly into pipe conveyance systems which in turn may feed pipe-
lines or ditches for water distribution. Higher pressure centrifugal pumps, including
close-coupled vertical turbines, are commonly used to supply sprinkler systems di-
rectly from reservoirs.
When variable flows are desired in the field, using internal combustion engines to
power pumps, rather than electric motors, can provide such flexibility. To avoid hav-
ing to frequently change pump speed to closely match field demand, it is often practi-
cal to have an overflow bypass back into the reservoir, designed to maintain the
needed head on the pipeline for delivery. In operation, the pump is set to supply the
desired field flow with a small excess overflowed through the bypass standpipe, to
maintain a constant head. In semiautomated systems with closed or semi-enclosed
pipelines or level-top canals, a bypass stand beside the pump and reservoir is often
used. Without a bypass, the irrigator must take the full unregulated flow pumped and
this almost always results in less-efficient utilization of water and labor. With a bypass
a fine readjustment of the flow rate in the field to match the precise need will cause the
small rejected flow to go directly back into the reservoir rather than being wasted in
the field or continuing to a tailwater reservoir. The extra pumping cost for the small
head on the rejected flow is justified by the conserved water in the field and is less
than the cost of repumping tailwater.
In level areas, the reservoir can consist of an extensive level-top canal system to
provide storage, distribution, and sometimes drainage, depending on the drainage wa-
ter quality. The reservoir may then border on several fields, and gravity outlets or
portable pumps, if necessary, can be located adjacent to the field for which water is
being withdrawn, minimizing conveyance piping.
When the groundwater table is about the same elevation as the canal, there is little
or no seepage loss. Where elevation differences preclude a single-level canal system,
several steps in the system can be created by the use of mechanical gates, such as the
NEYRTEC AVIS automatic constant downstream level canal gates referred to in Sec-
tion 11.2.3 (Merriam, 1977).
It is practical to use such depressed canals for project transmission canals. Pumping
costs may be compensated by the small seepage loss from an unlined canal; and, more
importantly, by the elimination of the need for on-farm reservoirs to provide flexibility
of rate and duration. Pumps, portable or permanent, from the canal may serve small
elevated on-farm canals or pipelines. The pumps should have variable flow and bypass
capabilities.
An indirect way of obtaining on-farm flexibility in rate and duration is through the
use of reservoirs in the project supply system. The management and the project canal
system must be able to permit variable flows and duration at farm turnouts, which are
under the control of the irrigator. The farm turnouts can be automated by the use of float
valves to pipeline systems or NEYRTEC AVIO gates to canals. The irrigator can then
use the project system as his reservoir. With this variable outtake, the canal flows vary.
To overcome these variations, the canal system may be automated, or a sloping ca-
nal section can terminate in a reservoir for re-regulation. The next sections below the
reservoir must be treated the same way, and the final section should serve the farms
through a system of closed or semi-closed pipelines or level canals. In this way, the
project canals and reservoirs can provide the needed on-farm flexibility in rate and
duration to permit a limited-flow demand schedule. Flexibility of frequency should
still be under the control of the canal operation to match canal capacities. Increased
canal capacities may be obtained by canal automation because closer control will re-
duce freeboard requirements. This permits the limited-flow demand type of schedule.
11.2.4.2 Investigations and design. In many states, laws and regulations may con-
trol, limit, or even require the installation of farm reservoirs. Water rights may be af-
fected or public safety endangered by their construction. All applicable state statutes
and regulations must be met, and often permits are required before construction is au-
thorized. For such reasons, it is usually best that a reservoir be designed and con-
structed by qualified professionals. Design services may be provided by appropriate
federal agencies or consulting engineering firms. The following checklist includes
some considerations for irrigation reservoir design:
The purpose for the reservoir must be clearly defined.
Adequate consideration must be given to prevention of danger to life or damage
to property, both in event of reservoir embankment failures and in the course of
day-to-day usage.
Adequate capacity should be available for sediment inflows.
Capacity provides for potential seepage and evaporation losses.
Excessive inflows can be safely bypassed.
Outlet capacity is adequate to meet irrigation demands.
11.2.4.3 Evaporation and seepage losses. The net depth of the water lost through
evaporation during short periods in shallow ponds can be satisfactorily estimated by
assuming it to be equal to the evapotranspiration minus precipitation. Because the vol-
ume of water lost is dependent upon the surface area, evaporation losses can be mini-
mized by providing a maximum water depth and minimum surface area consistent
with an economical balance between construction, maintenance costs, and water loss.
Evaporation losses from storage reservoirs that are full year around may exceed
2500 mm per year under adverse climatic conditions and little compensating rainfall.
At the other extreme, a regulating reservoir covering a small area may lose 4 to 6 mm
per day of operation and up to 10 mm under very adverse climatic conditions. This
may represent a very small percentage of the water controlled; e.g., 0.2 ha surface ×
5 mm evaporated per day per 5000 m3 of water controlled per day is about 0.2% loss.
Even if the reservoir is kept full, though used only one-third of the time, it is still a
negligible loss. An overnight storage reservoir with large surface area and a moder-
ately severe climatic condition might lose about 2% of its controlled water.
Seepage losses are more difficult to estimate than evaporation losses, but in most
instances an adequate appraisal can be based on the permeability of the foundation
materials. Local experience with existing reservoirs built in similar soils is perhaps the
best guide as to the need for pond sealing. The method of seepage control, for given
site conditions and operation requirements, can be determined during the design proc-
ess and implemented during reservoir construction. Options include compaction of
existing earthen materials, treatment with bentonite or soil dispersants, or installation
of membrane, concrete, asphalt, or imported earth liners. It may also be practical to
delay the installation of seepage control methods until a period of evaluation has de-
termined a need. Seepage losses can seldom be completely eliminated and control
methods vary greatly in their effectiveness and cost.
Draining an earth-covered membrane-lined reservoir for a few days between uses, to
limit evaporation or seepage, will save very little water over retaining it partially full.
This is because the water in the earth covering above the membrane will evaporate. Un-
covered linings such as soil cement may be emptied but may develop cracks due to tem-
perature changes. Most linings should be kept covered with water to reduce maintenance
problems, even though evaporation will be greater than if the reservoir is emptied.
Where the economic justification for the reservoir is principally in labor saving be-
cause of the larger stream of water made available, or the seepage loss returns to a
shallow groundwater basin for repumping, it may not be practical to reduce seepage.
Increased irrigation efficiency made possible by the reservoir may easily save more
water than the seepage loss. A 200-mm per day seepage from 0.5 ha reservoir area is
less than 7% of 150 mm applied on a 10 ha field. However, if the economic value of
the water is high in the terms of its productivity or direct cost, very little seepage can
be tolerated. In general, the value of the water conserved by lining the reservoir, or of
preventing a high water table problem, whether for economic, aesthetic, or environ-
mental reasons, must exceed the cost of the lining and its maintenance.
11.2.5 Mosquito Control in Farm Delivery Systems

Irrigation systems can frequently increase existing mosquito problems or create a
new problem unless special control measures are exercised. Mosquito populations can
be increased if irrigation practices cause the free water surface conditions needed for
mosquitoes to complete their life cycle of egg, larval, and pupal stages. Conditions
favorable for mosquito production include reservoirs with floating vegetation and de-
bris, water standing for several days on fields or in tailwater ditches, and seepage areas
along canals or through leaky dams. Mosquito control planning should be coordinated
with other basic irrigated farm needs such as soil and water conservation, flood con-
trol, wildlife management, recreation, and, occasionally, hydroelectric power. Proper
water management and maintenance of facilities can achieve reduction or elimination
of manmade mosquito sources by applying knowledge of both irrigation principles and
mosquito biology. Further elaboration on details of mosquito control in irrigation pro-
jects is presented in Engineering Practice EP267.6 (ASAE, 2005a).
11.3 FARM WATER DISTRIBUTION SYSTEMS
11.3.1 Introduction
This section discusses lined open ditches and pipelines to convey irrigation water
on the farm. When buried pipe is used instead of ditches, the conduit can follow the
most direct route from the water supply to outlet points rather than following field
contours. Weed problems and loss of productive land are eliminated because crops can
be planted up to or over the buried pipeline and the pipe does not interfere with plant-
ing, cultivating, and harvesting. Seepage and evaporation losses are eliminated when
water is transmitted in a well-constructed pipeline. Pipelines present fewer safety haz-
ards than open ditch systems. Pipelines may not be desirable, however, if the irrigation
supply contains large amounts of sediment and the flow conditions in the line are such
that the sediment would tend to settle out and reduce the carrying capacity of the line.
Portable pipe systems laid on the soil surface have many of the advantages of bur-
ied pipe systems. If adequate labor is available, they can be removed from fields to
allow unimpeded cultural operations.
Seepage losses in unlined farm ditches often range from one-fourth to one-third of
the total water diverted. In extremely sandy or gravelly ditches, half the irrigation sup-
ply can be lost in seepage on the farm. Unlined ditches, because of their potential for
high seepage and maintenance expense, should only be considered for temporary use
or in cases where money is not available for lined ditches or pipelines. Reducing seep-
age by using better conveyance facilities may increase water available for crop needs,
allow irrigation of additional land, and prevent waterlogging or salination of soils be-
low the ditches. In areas with a history of failure or rapid deterioration of pipelines or
lined ditches, the guidelines in this chapter should be evaluated and modified on the
basis of local experience. The design of farm pipelines with high pressure ratings is
discussed in Chapter 15 and 16. Information on design and construction of unlined
ditches can be found in older irrigation engineering texts.
11.3.1.1 Capacity. The necessary capacity of farm conveyance conduits depends
upon the water available and design irrigation efficiencies, which in turn depend on
soil, topography, method of irrigation, crop, type of irrigation system, and farm man-
agement practices. If water is delivered to a field or farm at a fixed rate throughout the
season, the conveyance capacity can be less than if water is delivered periodically in
larger streams. Larger available stream sizes often allow for more efficient use of wa-
ter and labor. Irrigation with borders or basins and with high intake rate soils requires
larger streams than furrow irrigation and low intake rate soils. When designing conduit
size, keep in mind that even with good maintenance, sediment, trash deposits, or vege-
tative growth may reduce a ditch’s capacity. Pipeline roughness may be increased and
cross-sectional area reduced by encrustations or deposits.
Conduits normally are designed for a capacity equal to the crop water requirements
during peak demand plus irrigation and operational losses. Water losses vary with the
irrigation distribution system, method of irrigation, soil, crops grown, farm manage-
ment practices, and other factors, which are discussed in other chapters. For well-
managed, conventional furrow and border irrigation systems, conduit design should
allow for water application efficiencies of 60% or more. Level basins and surface sys-
tems with tailwater recovery are capable of 80% to 90% efficiencies. For reasons of
economy, conduits should not be oversized. However, they should be large enough to
provide adequate water for crop requirements under existing and anticipated irrigation
methods. The amount of surface water intercepted from precipitation and irrigation
waste must also be considered in determining capacity.
11.3.2 Lined Ditches and Structures
Lining ditches is an effective way to prevent ditch erosion, control rodent damage,
and reduce seepage at reasonable cost. Lining also reduces maintenance requirements,
controls weed growth, and ensures more dependable water deliveries. The seepage
reduction helps protect neighboring land from waterlogging and salt accumulations.
Linings must be properly designed and installed to avoid damage from vegetative
growth, fluctuating water tables, livestock traffic, and freezing and thawing.
11.3.2.1 Ditch capacity and design. Principles of open channel flow hydraulics
especially applicable to farm irrigation ditches are presented by Replogle et al. (1990:
Section 10.1.2).
After the necessary capacity has been determined, ditch size can be designed using
the Manning equation:
C
ν = u R 2 / 3S1 / 2 (11.4)
n
where V = mean velocity, m/s or ft/s
n = Manning roughness coefficient (see Table 11.3)
R = the hydraulic radius, m or ft
S = the energy-line slope
Cu = 1.0 for SI units and 1.486 for foot-pound (customary) units.
Efficient methods to design ditches to convey a given flow rate can be found in
standard hydraulics texts (Chow, 1959: Chapter 6; Schwab et al., 1993: Chapter 13).
Table 11.3. Appropriate n values for use in Equation 11.4 for common farm ditch linings.
Material Manning n for Capacity
Slip-formed concrete 0.015
Brick 0.017
Shotcrete, concrete panels 0.016
Sheet metal, flex membrane 0.013
Compacted earth Same as unlined ditches; see Chow (1959)
11.3.2.2 Lining materials. Selection of the lining material should be governed by

availability of the material and installation equipment, ditch size, climate, foundation
conditions, and type of irrigation stream, whether continuous or intermittent. Concrete
is probably the most popular lining material, but asphaltic materials, bricks, mem-
branes, metals, chemical sealants, and impermeable earth materials are also used. All
of these materials make good linings if they are properly selected for the site condi-
tions and are installed correctly.
Lined ditches can be installed in any soil, but many soils have limitations that must
be overcome before a durable, low-maintenance lining can be installed. Rigid linings,
such as concrete, should be installed on drained soils, particularly in colder climates.
For soils that have poor internal drainage or low bearing capacity, the concrete must
be thicker or steel reinforcement must be added.
11.3.2.3 Elevation and slope. Lined ditches should be designed so the water sur-
face at field turnout points is high enough to provide the required flow onto the field
surface. If ditch checks or other control structures are used to provide the necessary
head, the backwater effect must be considered in computing freeboard requirements.
The required elevation of the water surface above the field surface will vary with the
type of turnout structure used and the amount of water to be delivered. A minimum
head of 0.12 m above the field water surface should be provided.
The required freeboard varies with the size of ditch, the velocity of the water, the
horizontal and vertical alignment, the amount of storm or wastewater that may be inter-
cepted, and the change in the water surface elevation that may occur when control struc-
tures are operating. The minimum freeboard for any lined ditch should be 0.075 m.
For concrete lining, design velocities in excess of 1.7 times the critical velocity
should be restricted to straight reaches that discharge into a section or structure de-
signed to reduce the velocity to less than critical velocity. This will avoid unstable
surge flows. The maximum velocity in these straight reaches should be 4.5 m/s.
Maximum allowable velocities in ditches with sprayed-on membrane linings should be
limited to 1 m/s. Buried membrane installations are only effective where the flow ve-
locity is limited to that allowable for the covering earth material.
11.3.2.4 Concrete lining and construction. Concrete linings made from Portland
cement can withstand high stream velocities and therefore are particularly suitable for
controlling erosion as well as preventing seepage. They are more resistant to mechani-
cal damage than most other linings. However, their use is best suited to sites having
non-expanding soils with low sulphate concentrations and good internal drainage.
Local consultants or NRCS offices may be able to provide information on concrete
mixes and lining thickness. Special mixes may be required where soils are high in
sulphate or climatic conditions are severe.
An irrigator who wishes to excavate and line his own ditches can use concrete
panel-formed lining. This lining requires only semiskilled labor and a minimum of
equipment, and, for small jobs, it may cost less than slip forming. After excavating the
ditch to grade, guide forms are set about 3 m apart. Concrete is then poured in alter-
nate 3 m sections. When the concrete is set, the forms are removed and the skipped
panels are poured. The bottom in each section is poured first, and then additional con-
crete is screeded up both sides.
11.3.2.5 Other lining materials. Other linings that may prevent erosion and reduce
seepage in farm ditches include masonry, pneumatically applied concrete, prefabri-
cated sheet metal, various plastic and asphaltic compositions, compacted earth, clay
and buried membranes. Many of these are not available or are not practical in all irri-
gated areas.
11.3.2.6 Structures. Canal, lateral, and reservoir turnout structures that deliver wa-
ter to farm ditches are beyond the scope of this monograph. A wide variety of such
structures are described by Kraatz and Mahajan (1975). These are permanent struc-
tures, considered part of the canal delivery or supply system, and should include pro-
visions for controlling and measuring flow rates. Other structures for irrigation deliv-
ery and distribution are considered below.
Control structures. One of the most important factors in the efficiency and ease of
operation of irrigation systems is the character of the control structures used. Surface
irrigation characteristically has a high labor requirement, but good control structures
can reduce labor and simplify irrigation. Use of mechanized structures is increasing.
Combining two or more functions in a single structure can often minimize cost. For
example, checks, drops, turnouts, divisors, and measuring structures can be used in
various combinations. Flow measuring structures suitable for use in farm ditches are
described in Section 11.2.2 of this chapter.
If straight, suitably lined ditch sections are used to transmit water at supercritical
velocity, this velocity must be reduced before approaching bends, distribution reaches,
or through turnout structures into erosive channels. Energy-dissipating structures for
such channels must be individually designed. The general configuration of a suitable
structure and results of laboratory tests for discharges of 0.05 to 0.60 m3/s (1.6 to 20
ft3/s) have been published by the Bureau of Reclamation (USBR, 1963).
Sediment deposited in irrigation ditches and structures necessitates frequent ditch
cleaning and often results in inaccurate flow measurement. Trash in irrigation water is
a source of weed infestation on the farm and clogs irrigation structures. When trash is
a problem, special structures are often needed to remove it from the irrigation water
supply.
Drop structures. Drop structures are not needed for velocity control in lined ditches
unless the estimated velocities will be supercritical. If ditch water needs to be carried
across a road, a drop can be combined with a culvert, as shown in Figure 11.5. Con-
crete pipe (or PVC, if placed in a suitable trench, carefully bedded, and covered with
0.75 to 1.25 m of well-compacted fill) might be used in place of the corrugated metal
pipe shown. More care with the bedding and backfill is needed to prevent crushing of
the PVC, or joint leakage with the concrete pipe, than with the corrugated metal pipe.
This drop is rather easily plugged by trash.
Check structures. A check is any structure used to maintain or increase the water
level above the normal flow depth in an open channel. The structures in Figure 11.6
can be fitted with flash board slots and can be used as combination checks/drops.
Other designs are given by Skogerboe et al. (1971), Booher (1974), Kraatz and Maha-
jan (1975), and USDA.1 Commercial prefabricated and modular metal, concrete, and
masonry structures are commonly used.
1
USDA Natural Resources Conservation Service (NRCS) Conservation Standards can be located at
http://www.nrcs.usda.gov/technical/standars/nhcp.html. NRCS warns that these standards should not be
used to plan, design, or install a conservation practice. A user must have the conservation practice standard
developed by the state in which the practice is to be used to insure that all state and local criteria, which may
be more restrictive than national criteria, are met.
CMP
1:1
1.5
:1
D
CMP
PLAN
ISOMETRIC VIEW OF
Top of Ditch Bank CONCRETE SLAB
Concrete Slab Top of Ditch Bank
Ditch D Corrugatied Metal Pipe

Bottom D
Level Line
Watertight Welded Joint Ditch Bottom
SECTIONAL ELEVATION ON CENTER LINE
NOTE: Capacities range from 65 L/s for 254 mm pipe and 0.3m drop, to
229 L/s for 381 mm pipe and 0.9 m drop.
Figure 11.5. Corrugated metal pipe drop (USDA-SCS, 1978).
A Capacity of Width of
W
Ditch Opening, W H C A
Crest L/s Ft3/s m m m m
H
57 2 0.30 0.30 0.15 0.60
170 6 0.60 0.30 0.15 0.60
227 8 0.75 0.38 0.15 0.60
D+C 283 10 0.90 0.45 0.20 0.75
396 14 1.05 0.45 0.20 0.90
Stilling
Basin Drop, D Apron Length, L
L m m
0.30 0.75
0.45 0.90
C 0.60 1.25
0.90 1.80
Suggested Drop
Structure
Figure 11.6. Common types of drop structures used in farm irrigation ditches.
Long-crested weirs. To maintain a constant upstream water level for uniform diver-
sion of discharge into another channel or onto a field, an overflow weir-type check
should be used. The flow over such a check may be estimated from the general weir
equation, Equation 11.2, which in the general form is:
n
Q = CLy (11.5)
This equation is satisfactory for design purposes but not for exact water measure-
ment through the opening. The value of n for most overflow checks is approximately
1.5, and C varies from 1.55 to 1.82 for SI units and 2.8 to 3.3 for L and y in feet (val-
ues of 1.66 m or 3.0 ft are frequently used for design purposes). When the crest length,
L, is large, variations in discharge result in relatively small changes in the upstream
water level.
For a channel of given width, the fluctuation of upstream water surface as flow
rates vary may be greater than desirable, even with a straight weir crest perpendicular
to the channel center line. The water level fluctuation can be reduced even further by
lengthening the weir crest, either with a diagonal weir (a straight weir crest con-
structed diagonally to the channel center line) or a duckbill weir (Figure 11.7). The
effective length of a duckbill weir can be several times the channel width. Kraatz and
Mahajan (1975) and Walker (1987) illustrate several types of diagonal and duckbill
weirs, and give design information.
In about 1990, engineers at California Polytechnic State University started modifi-
cations to early designs of canal flap gates, which dated from the 1940s in The Nether-
lands (ITRC, 2001; Burt et al., 2001). Raemy and Hager (1998) published a version
that they claim is stable and uses a shock absorber to reduce oscillations (Figure 11.8).
The IERC design also currently uses a shock absorber to reduce oscillations. The flap
gate is a simple, inexpensive hydraulic gate for automatic upstream water level con-
trol. It is basically a rectangular plate placed at the end of a section of canal in which a
constant upstream depth is desired. In its simplest form this is at an overfall that al-
lows no backwater against the downstream side of the plate. The plate pivots about a
hinge at its top that is offset upstream from the plane of the gate. Thus, with properly
designed offset length, the weight of the gate and additional weights in the plane of the
gate are able to maintain a constant upstream depth, passing excess flows for uses
downstream.
Gates for automatically controlling the water level in canals can also be used in
some farm ditches (Kraatz and Mahajan, 1975). Semiautomatic timer-controlled
checks have been used to a limited extent in farm ditches. They are normally used
with timer-controlled outlets for releasing flow onto a field or into another ditch (Cal-
der and Weston, 1966; Humpherys, 1969, 1986; Hart and Borrelli, 1970; Robinson,
1972; Evans, 1977). Automation of farm water control structures can reduce the irriga-
tion labor requirement and improve the efficiency of surface irrigation.
Figure 11.7. Plan view of diagonal weir and duckbill weir.

Figure 11.8. Side view of counterweighted flap gate, from Raemy and Hager (1998).
Figure 11.9. Precast concrete farm turnout (from USBR, 1951).
Outlets and discharge controls. Field turnouts are used to control the release of wa-
ter from a farm ditch into basins, borders, furrows, or another irrigation ditch. The
turnout may be a fixed opening in the side of a ditch or one equipped with check
boards, gates, or other devices to adjust the opening area. If only a portion of the total
flow is to be delivered through a given turnout, discharge through an orifice-type de-
vice is more constant than through an overflow or weir-type structure. One of the most
frequently used turnouts is a concrete, plastic, or metal pipe with a slide gate on the
inlet, as shown in Figure 11.9. For lined ditches the headwall and slide gate are usually
constructed flush against the side of the ditch. The capacity of short pipe turnouts can
be determined from Equation 11.1.
For free flow, the head differential is the upstream head measured from the center
of the orifice opening; for submerged flow, it is the difference between upstream and
downstream water levels. The coefficient of discharge Cd varies widely, depending on
the position of the orifice in relation to the sides and bottom of the structure and the
type of orifice opening, i.e., sharp edged, rounded, short tube, etc. (Brater and King,
1976).
Humpherys (1978) developed timer-controlled gates for border turnouts where the
turnout must first be automatically opened to begin irrigation and then closed to termi-
nate irrigation while the water level in the head ditch remains constant.
It is often necessary to divide water from a farm lateral into two or more ditches for
distribution to different parts of the farm or to other farms. This may be accomplished
with a divisor at the ditch junction. For accurate flow division it is best to measure the
flow in each channel. Some divisors are designed to give a fixed proportional flow
division, whereas others have a movable splitter to change the flow proportions. Divi-
sors that give accurate proportions often divide the flow at a control section where
supercritical flow exists, such as at the nappe of a free overfall (Figure 11.10).
Flow can be divided accurately without creating supercritical flow if: (a) the ap-
proach channel is long and straight for at least 5 to 10 m upstream so that the water
flow approaches the divisor in parallel paths without crosscurrents, (b) there is no
backwater effect that would favor one side or the other, and (c) the flow section of the
structure is of uniform roughness. Detailed divisor designs are described by Bos
(1989) and Kraatz and Mahajan (1975).
Debris and weed seed control structures. Trash and weed seed in the water cause
problems in irrigation systems. Besides spreading weeds, trash and debris plug siphon
tubes, pipe gates, and sprinkler nozzles. Trash racks are needed at entrances to pipe
Figure 11.10. Simple fixed proportional flow divisor.

Inlet Box
Screen Frame
Frame Box
Outlet
Figure 11.11. Structure with screen for trapping debris in irrigation streams
(adapted from Bergstrom, 1961).
and underground crossings. Various types of screens and trash racks have been devel-
oped for screening irrigation water (Couthard et al., 1956; Bergstrom, 1961; Pugh and
Evans, 1966). One of the simplest devices is a screen placed below a weir or drop, as
in Figure 11.11. If the screen is taut and the water drops from a sufficient height, say
at least 0.2 m, the vibrations of the screen will move the trash far enough away from
the overfall to allow the water to pass through. Screen mesh size will depend on the
type of debris in the irrigation stream and the susceptibility to plugging of downstream
water control devices. Very fine screens may require more frequent cleaning, but may
also have advantages, such as removing weed seed.
A different configuration of a screen using much the same principle has been de-
veloped in Idaho. The “turbulent fountain” screen is circular, surrounding the opening
of a vertical pipe with flow discharging from the pipe’s open top. The screened water
is collected below the fountain screen, while the turbulence of the water pushes trash
off its sides (Kemper and Bondurant, 1982; Bondurant and Kemper, 1985).
Screens that use water wheels or electric motors to power brushes that pass repeat-
edly over the screen surface, moving trash to one side, are commercially available.
11.3.3 Low-Pressure Pipe Systems
11.3.3.1 Applications. Low-pressure pipe systems for irrigation water distribution
have been used extensively since the 1950s. The availability of relatively low-cost,
lightweight, rigid plastic pipe has made buried pipe systems especially popular. Ad-
vantages and disadvantage of buried pipelines are summarized in Section 11.3.1. Pipe-
lines with large diameter risers may require safety precautions, either adequate height
above the ground or a screen or cover to keep children from entering.
There are three general types of on-farm irrigation pipelines. The first is the com-
pletely portable surface system where water enters the line at the supply (a well, reser-
voir, or ditch turnout) and the water is applied to the field from the open end of the
pipeline or from gated outlets along the line. The second system is a combination of
buried and surface pipe where buried permanent line is used to transmit water from the
source to the field to be irrigated. Then water is supplied to gated surface pipe through
one or more risers. The third system, generally used for border dike or basin irrigation,
eliminates the need for surface pipe. Water is released onto the portion of the field to
be irrigated from risers on the buried line (see also Section 11.2.3).
11.3.3.2 Location. The irrigator can exercise much flexibility in locating surface
pipelines between the source of water supply and the point of water application. The
lines should be located to provide a minimum of interference with traffic on farm
roads or with field cultural operations. A pipe network should be designed so that the
shortest possible length of pipe is used to deliver water to all fields to be irrigated.
More care is needed in locating buried systems. If possible, the line should be located
where it can be easily buried with uniform trench depth and the minimum necessary
cover. It is best to avoid routes that will be crossed by heavy surface traffic. Also, be-
cause air relief must be provided at high points in a buried pipeline, the number of
these high points should be minimized. Trenching across fields to be irrigated should
be avoided if possible. Irrigation water may enter poorly compacted backfill in the
trench and cause piping along the line, simultaneously causing incomplete irrigation of
the field being watered.
11.3.3.3 Pipeline capacity. Pipeline capacity must be great enough so that ade-
quate water can be delivered to meet crop requirements. It generally will cost less to
use large pipe with the available head than to use smaller pipe and booster pumps for
such systems. When an entire irrigation system is to be designed, including pump,
power unit, and piping, irrigation pipe sizes should be chosen so as to minimize total
annual costs. Both fixed and operating items must be considered, including the initial
cost of the piping, the number of years’ service expected from the pipe, the initial cost
and expected years of operation of different sized pump units, the energy cost of
pumping, and the rate of return desired on the funds invested. Keller (1975) has pre-
sented a procedure for constructing design charts from which economical pipe sizes
can be selected, based on the above considerations for non-looping distribution sys-
tems having a single pump station. More complicated systems can be designed using
procedures given by authors cited by Keller.
The Darcy-Weisbach Formula (Brater and King, 1976) expresses head loss of tur-
bulent flow in pipelines as:
L V2
H1 = f (11.6)
D 2g
where Hl = the loss of head in equivalent height of water in a length of pipe L
D = inside pipe diameter
V = mean velocity
g = gravitational acceleration
f = a resistance coefficient.
Equation 11.6 is dimensionally consistent and can be used with the same f values
for either SI or foot-pound units. Values of f have been related to boundary roughness
dimensions for certain types of pipe surfaces and determined empirically for others
and are tabulated in most hydraulic handbooks. The Hazen-Williams equation (Brater
and King, 1976) is used extensively for determining friction losses in irrigation pipe-
lines. Hazen-Williams resistance coefficients are readily available, having been deter-
mined empirically for tubing commonly used in irrigation. The equation can be written:
V = CR 0.63 S 0.54 (11.7)
where R = the hydraulic radius and S = the friction slope in meters per meter or feet
per foot. When R is in mm, C = 0.0109 C1; when R is in feet, C = 1.318 C1; where C1
is the Hazen-Williams resistance coefficient. Values of C1 range from 144 to 146 for
aluminum tubing. C1 values for other materials are given in standard hydraulic hand-
books.
Energy losses through fittings and valves also have to be considered in the design
of an irrigation pipeline. These so-called “local” losses are frequently estimated by
applying a coefficient to the velocity head at the fitting. The sum of all local losses is
then added to the estimate of pipe wall friction to give total loss in the pipeline. Local
loss coefficients, K, to be used in the equation:
Hf = KV 2/2g (11.8)
are listed in Table 11.4. The coefficients in Table 11.4 have been determined for stan-
dard pipe diameters. Coefficients for special irrigation pipe dimensions and materials
are not available. However, because local losses occur as a result of flow separation
and changes in velocity, and are not greatly affected by the roughness of the fitting,
values in Table 11.4 should be valid for irrigation pipe fittings with similar geometry.
Table 11.4. Value for K for use in Equation 11.8 to estimate local head losses in pipelines.
Nominal Diameter
75 mm 100 mm 125 mm 150 mm 175mm 200mm 250 mm
Fitting or Valve (3 inch) (4 inch) (5 inch) (6 inch) (7 inch) (8 inch) (10 inch)
Standard pipe
Elbows
Regular flanged 90° 0.34 0.31 0.30 0.28 0.27 0.26 0.25
Long radius flanged 90° 0.25 0.22 0.20 0.18 0.17 0.15 0.14
Regular screwed 90° 0.80 0.70
Tees
Flanged line flow 0.16 0.14 0.13 0.12 0.11 0.10 0.09
Flanged branch flow 0.73 0.68 0.65 0.60 0.58 0.56 0.52
Screwed line flow 0.90 0.90
Screwed branch flow 1.20 1.10
Valves
Globe flanged 7.0 6.3 6.0 5.8 5.7 5.6 5.5
Gate flanged 0.21 0.16 0.13 0.11 0.09 0.075 0.06
Swing check flanged 2.0 2.0 2.0 2.0 2.0 2.0 2.0
Foot 0.80 0.80 0.80 0.80 0.80 0.80 0.80
Strainers, basket type 1.25 1.05 0.95 0.85 0.80 0.75 0.67
Other
Inlets or entrances
Inward projecting 0.78 (all diameters)
Sharp cornered 0.50 (all diameters)
Slightly rounded 0.23 (all diameters)
Bell mouth 0.04 (all diameters)
2
Sudden enlargements [ ]
K = 1 − (d12 / d s 2 ) , where d1 = diameter of the smaller pipe.
2
Sudden contractions [ ]
K = 0.7 1 − (d12 / d2 2 ) , where d1 = diameter of the smaller pipe.
11.3.3.4 Surface pipelines. Portable surface pipelines for low-pressure irrigation

water distribution are commonly constructed of aluminum or rigid or flexible plastics.
All can be used solely for transmission of water or can be provided with gated outlets
for water distribution to fields. Low-pressure aluminum pipe is generally available in
127-, 152-, 203- and 254-mm (5-, 6-, 8- and 10-in.) nominal diameters. The pipe is
rolled from 1.3-mm (0.051-in.) aluminum sheet and closed with a welded seam. A
rolled or cast female coupling is provided on each pipe length to contain flexible, pres-
sure sealing gaskets. Mechanical clamps are sometimes provided for positive pipe
coupling.
Aluminum gated pipe sections of large diameters are generally manufactured in 6-
m lengths, whereas those of smaller diameters are 9 m long. Standards for aluminum
and PVC tubing for gated pipe are published by USDA (2006). Rigid, extruded PVC
surface pipelines are also available. The PVC compound contains an ultraviolet inhibi-
tor that allows extended exposure to sunlight without deterioration. Diameters and
couplings are compatible with aluminum-gated pipe.
Conduits of thin-walled, flexible vinyl tubing for water transmission and distribu-
tion to furrows became available in the mid-1970s. This tubing is sold in lengths of
several hundred meters, packaged in compact rolls. Wall thickness is 0.25 mm or less
and diameters from 250 mm to 500 mm are available. A tubing installation implement,
which mounts on the three-point hitch of a farm tractor, can plow a shallow groove to
provide a bed for the tubing and unreel and lay the tubing simultaneously. Plastic fit-
tings are available for couplings, tees, and connections to the water supply. Gates are
inserted in the field, at the exact locations necessary to best serve furrows or corruga-
tions, using a simple insertion tool.
Flow capacity of the flexible vinyl tubing is about one-fourth less than aluminum
tubing of the same diameter, because at the low allowable operating pressures the tub-
ing is not fully round. When gated tubing is laid on sloping land, and the water source
is from the upslope end, the upslope end may be at such low pressure that adequate
streams will not flow from the gates. As a practical solution, some irrigators shovel
soil on the tubing at intervals in order to restrict through-flow and maintain adequate
pressure in the upper reaches. Initial cost of flexible vinyl tubing is a fraction of that
for comparable aluminum or rigid plastic pipe. However, the material is not durable
and usually has to be replaced for each irrigation season.
Some portable pipelines, especially aluminum lines that have low resistance to
crushing, can be flattened by the pressure differential created if sub-atmospheric pres-
sure occurs in the line. Sub-atmospheric pressures might occur if a deep well pump
that supplies a pipeline without check valves is suddenly stopped and the open valves
on the pipeline are inadequate for vacuum relief. Rapid closing of an upstream valve
in a line carrying high velocity flow may also lower pressures below the valve. Ap-
propriately placed vacuum relief valves and careful operating procedures can mini-
mize the chance of such damage occurring.
Also, pressure surges in the line can cause joints to separate or pipe sections to rup-
ture. Valves controlling flow into surface pipelines should be opened carefully so that
the lines fill slowly before they are brought up to full pressure. If water is to be re-
leased from the pipeline through gates, one set of gates should be open at the time the
line is filled. Likewise, shutting off the flow to the pipe before all gates are closed will
help prevent a vacuum from developing in the line. If the pipeline is supplied directly
Table 11.5. Friction loss in aluminum pipe as estimated by the Hazen-William formula.[a]
Joint losses are included[b]; assumes joints at 9-m (30-ft)[c] intervals
(adapted from USDA-SCS, 1978).
Nominal Outside Diameter
102 mm 127 mm 152 mm 203 mm 254 mm
(4 in.) (5 in.) (6 in.) (8 in.) (10 in.)
Discharge G[d] NG G NG G NG G NG G NG
(L/s) (ft3/s) Head Loss (m/100 m or ft/100 ft)
10 0.35 1.89 1.80 0.65 0.62 0.27 0.26 0.07 0.07
15 0.53 4.07 3.86 1.40 1.31 0.59 0.55 0.15 0.14
20 0.71 6.94 6.57 2.41 2.26 1.01 0.94 0.26 0.24 0.09 0.09
30 1.06 14.86 14.02 5.16 4.82 2.18 2.01 0.57 0.52 0.20 0.18
40 1.41 8.87 8.26 3.74 3.45 0.98 0.89 0.35 0.31
50 1.77 5.71 5.25 1.50 1.35 0.54 0.48
75 2.65 12.24 11.23 3.24 2.92 1.16 1.03
100 3.53 5.57 5.00 2.01 1.77
125 8.54 7.65 3.08 2.71
150 12.02 10.74 4.35 3.82
[a]
Hazen-Williams coefficients: C1 = 145 for 102- to 152-mm pipe; C1 = 144 for 203- to 254-mm pipe.
[b]
Local loss coefficient, Kc = 0.3 for gated pipe; 0.2 for pipe without gates.
[c]
For 6-m (20-ft) lengths, increase loss by 10% for gated pipe and by 7% for pipe without gates; for 12-m
(40-ft) lengths, decrease loss by 5% for gated pipe and by 4% for pipe without gates.
[d]
G = gated, NG = not gated.
from a pump, a standpipe near the pump outlet, with diameter at least half the pipe
diameter, serves to even out pressure surges and to relieve vacuum as well. A check
valve near the pump discharge prevents flow of water from the line back through the
pump if the pump stops. Low-pressure portable pipelines are generally not installed
over terrain with wide variations in elevations. However, if extreme high points exist
in the line, some method of air release at these points may be needed to avoid restric-
tion in pipe capacity.
Values of friction loss for aluminum pipe with and without outlet gates are given in
Table 11.5 for pipe sizes commonly available in the U.S. The head loss values in these
tables include a component to account for local losses at pipe joints.
Water is commonly released from surface pipelines through small gates installed in
the pipeline at intervals equal to the furrow or corrugation spacing. Head loss in a sec-
tion of pipe discharging water through open gates will be one-third to one-half of what
it would be if the pipe were transmitting the full flow from one end of the discharging
section to the other.
Pipe gate discharge. Pipe gates are supplied in many sizes and styles by a number
of manufacturers. Gate discharge is governed by the orifice flow equation (Equation
11.1) when there is no flow in the pipeline past the gate. When there is flow, gate dis-
charge decreases as velocity increases. Because the discharge coefficient varies sig-
nificantly for different types of gates, discharge information should be obtained di-
rectly from the manufacturer of the gates used.
When gated pipe is laid on too great a slope, head is greatest on the downslope
gates. These gates must be nearly closed to maintain constant furrow streams along the
pipeline. With the small gate openings, velocities are large and may cause erosion
damage where the streams impinge on the furrow surface. Flexible tubes or “socks”
attached to each gate still the high-velocity streams and minimize erosion. Simple but-
terfly valves or overflow stands (similar in principle to that shown in Figure 11.12b)
may be installed at intervals along the pipeline to maintain more nearly constant
heads. Orifice plates can also be placed in the bell ends of the pipe to dissipate excess
energy. The size and number of orifices used depends on the head to be dissipated.
11.3.3.5 Buried low-pressure systems. Non-reinforced concrete was the most
economical and commonly used material for buried low-pressure irrigation pipelines
before thermoplastic piping came into use. Concrete pipe sections are readily available
in diameters from 150 to 600 mm. Most types of concrete pipe are designed to be
joined with rubber gaskets; others are joined using a cement mortar. Either joining
method is more labor intensive than for plastic pipe. Nitrogen fertilizers should be
used with care in concrete pipelines. When ammonia is added to hard waters, calcium
carbonate may precipitate and adhere to the walls of the pipe. Ammonium sulphate
concentrations should not exceed 0.1% and the line should be flushed immediately
after use. Fertilizer manufacturers should be consulted on damage prevention measures.
In recent decades, the ready availability and ease of handling of lightweight, semi-
rigid thermoplastic piping has made buried low-pressure pipe systems an attractive
alternative to concrete pipelines for irrigation water transmission. The plastic com-
pounds commonly used in constructing such pipe are polyvinyl chloride (PVC), acry-
lonitrile butadiene styrene (ABS), and polyethylene (PE).
PVC pipe is manufactured in two general size classifications: iron pipe size (IPS)
and plastic irrigation pipe size (PIP). IPS pipe has the same outside diameter as iron or
steel pipe of the same nominal size. PIP sizes were developed primarily for irrigation
use. PIP is smaller in actual diameter than IPS pipe of the same nominal diameter and
therefore has a lower flow capacity. Difference in capacity ranges from about 18% for
100-mm pipe to 8% for 300-mm.
Plastic pipe is commonly available in a range of pressure ratings. The ratio of the
allowable tensile stress in the pipe wall to the maximum pressure that the fluid in the
pipe can exert continuously with a high degree of certainty that the pipe will not fail is
related linearly to the standard dimension ratio, SDR (the ratio of pipe diameter to
minimum wall thickness). Plastic pipe commonly used for irrigation is manufactured
with several SDRs. Schedule 40, 80, or 120 PVC pipe, with dimensions corresponding
to steel pipe, is also available.
For a given pipe material and standard temperature, pipes of all diameters with the
same SDR will have the same pressure rating. For PVC and ABS the average outside
diameter is used in calculating SDR. For PE the average inside diameter is used. Poly-
vinyl chloride pipe for low-head systems is usually labeled “low head,” “SDR = 81,”
or “SDR = 64.” Operating temperature also affects pressure ratings. Pressure ratings
for pipes of different materials and a range of SDRs, and rating service factors for dif-
ferent temperatures, are given in ASAE Standard S376 (ASAE, 2005b ).
Head losses in low-head plastic pipe are determined using the Hazen-Williams
equation (Equation 11.7). A resistance coefficient of C1 = 150 is commonly used. Lo-
cal losses are calculated using the same K factors as listed in Table 11.4.
Trenches for plastic pipe should have a relatively smooth, firm, continuous bottom
and be free of rocks. Where rough rock edges cannot otherwise be avoided, the trench
should be over-excavated and filled to the bedding depth with sand or finely graded
soils. The trench below the top of the pipe should be just wide enough to provide room for
joining the pipe and compacting the initial backfill. Minimum and maximum trench widths
for low-head and SDR-81 pipe are listed in ASAE Standard S376 (ASAE, 2005b).
The pipe trench should be excavated deeply enough so that from 0.75 to 1.20 m of
cover can be placed over low-head irrigation pipes. This cover will normally protect
the pipe from traffic crossing, freezing temperatures, or soil cracking, but is not so
great that it will cause excessive soil loading on the pipe. Low areas can be crossed
with shallow trench depths and then extra fill placed over the pipeline to provide the
minimum cover depth. In such cases, the fill should have a top width of no less than
3 m and 6:1 side slopes. The pipe should be uniformly supported over its entire length
on firm stable material in the trench.
Solvent-welded PVC pipe is often flexible enough to be assembled alongside the
trench and then lowered into the trench after the solvent welds are well set. PVC pipe
with SDRs of 81.0 and lower can be obtained with solvent-weld or gasketed joints.
Inexperienced workers can assemble leak-free pipelines more easily with gasketed
than with solvent-weld joints.
Wherever the flow direction changes (at elbows and tees), reaction forces on the
piping may exist. These forces, if unrestrained, may cause the joints to separate. Con-
crete thrust blocking is commonly used to provide resistance to pipe movement where
side thrust might occur. Concrete pads, poured around pipe fittings in the trench, allow
forces on the pipe to be resisted by a large area of soil along the trench surface. ASAE
Standard S376 (ASAE, 2005b) illustrates those fittings that require thrust blocks, ap-
proved configuration of the blocks, and gives tables for estimating thrust magnitude.
Water packing can be used to settle the backfill around low-head pipelines. Before
backfilling, the pipe is filled completely with water and all joints observed for leakage.
The pipe should remain filled with water during the entire backfill operation to keep it
from floating or deforming under the weight of the saturated fill. After the pipe is
proven to be leak free, 0.3 to 0.45 m of backfill are placed over the pipe and water is
added to the trench until the fill is thoroughly saturated. The wetted fill is then allowed
to dry until firm enough to walk on before the final backfill is placed.
Polyethylene pipes are joined using rigid insert fittings and clamps or bell joints
with rubber gaskets. Polyethylene up to 150 mm in diameter is available for buried
lines. For details of plastic pipe pressure rating, material characteristics, and installa-
tion specifications and procedures, see ASAE Standard S376 (ASAE, 2005b).
Protection from vacuum and pressure surges is essential for buried thermoplastic
pipelines. See Pressure Relief Valves in Section 11.3.3.6 below for details.
11.3.3.6 Structures. All buried pipelines require an inlet and one or more outlet
structures. In addition, structures for air release, relief of excess pressure, or head con-
trol may be necessary (Robinson et al., 1963; Seipt, 1974). Frequently two or more
functions can be combined in a single structure (Figure 11.12). Construction methods
and materials are similar for any given structure, regardless of pipeline material. Rein-
forced concrete pipe is often used for inlet and head control risers because of its dura-
bility and resistance to impact damage. Special openings in pipe sections, beveled
ends, etc., are most easily made while the concrete is green and can be specially or-
dered from pipe casting plants.
Inlets. Where water is supplied to a buried pipeline from an open channel, a short,
vertical section of pipe can be used as the basis of the inlet structure. If all downstream
outlets on the pipeline can pass small pieces of trash without clogging, trash bars
spaced on 8-cm centers can be used to keep large trash and animals out of the pipeline.
Figure 11.12. Typical standpipes used for head control, pressure relief, and water release
from low-pressure pipelines. Several variations on these designs are possible for specific sites.
For smaller outlets, especially on gated pipe, and a water supply containing trash, a
finer inlet screen is necessary. Finer screens usually need to be self-cleaning, or they
will clog and cause the irrigation stream to overflow the inlet structure. Even self-
cleaning screens require frequent, perhaps daily, visits to remove trash accumulation
and check mechanical functions.
A flow measuring device is usually needed near the pipeline inlet, especially where
the pipeline is supplied directly from a well or an irrigation district canal. Metering
allows the irrigator to determine the volume of water applied and is essential if irriga-
tion scheduling recommendations are to be followed. Changes in well capacity and
pump efficiency can also be detected if metered flow rates are compared periodically.
If the pipeline is supplied from an open channel, a critical depth flume upstream of the
pipe inlet may be used for flow measurement. See Section 11.2.2 for details of flow
measurement with meters and flumes.
Pump stands. When a pipeline receives water directly from a pump, a stand similar
to that shown in Figure 11.13 is necessary. Pump stands can be constructed from rein-
forced concrete or steel. The inside diameter (D3) should be great enough to limit
downward velocity to 0.6 m/s. The vent pipe, if of reduced size as shown, should be
large enough to limit velocities to 3 m/s with the full pump outflow flowing through
the vent. The connection between pump outlet and stand must prevent transmission of
vibrations to the stand or pipeline.
D1
D1 + D2
Minimum
D2
D3
Figure 11.13 Surge control/air release stand for a pump discharging into a buried pipeline.
Vents, gate stands, and turnout risers. The requirements for these various types of
pipeline structures are given in ASAE S376.1 (ASAE, 2005b). Figure 11.12 shows
two variations of structure that can be used for venting air, and/or controlling head in
low-pressure buried pipelines.
Many buried pipelines are designed to release water through risers capped with al-
falfa valves, which can control releases into the atmosphere or into portable surface
pipelines. Procedures and equipment have been developed for automatic timed water
releases from alfalfa valves, using auxiliary pneumatic valves (sometimes in gated
pipe hydrants) and low-pressure air supply. Such systems can increase irrigation effi-
ciencies and reduce labor requirements on many fields, if they are carefully designed
and if the field is otherwise prepared for efficient irrigation (Haise et al., 1965, 1978;
Haise and Kruse, 1966; Fischbach and Gooding, 1970; Humpherys, 1986).
Safety. All pipeline structures having openings within 2 m of the soil surface should
be high enough, or they should be covered, so that small children cannot enter the
pipeline. Covering also prevents small animals and wind-blown trash from entering
the pipeline.
Pressure relief valves. Open stands can be replaced by pressure relief valves so that
the pipeline need not be open to the atmosphere. Pressure relief valves must be marked
with the pressure at which they start to open. Adjustable valves should be capable of
being locked or sealed so that the proper pressure setting is not easily changed. The
valve should be capable of releasing the design flow of the pipeline without elevating
pipeline pressure more than 50% above the permissible working head of the pipeline.
The number and spacing of relief valves on a pipeline is determined by the grade and
the allowable working head of the pipe. For pipelines of constant slope, one valve, at
the lowest point, may be adequate. If examination of the hydraulic head line shows
that allowable pressures could be exceeded, additional relief valves must be added.
Some manufacturers supply pressure relief valves that also provide air and vacuum
relief functions.
11.3.4 Operation and Maintenance

11.3.4.1 Lined ditches. If lined ditches are not properly maintained, the result can
be damage to the ditch lining and structures, and erosion of the field surface, as well as
substantial economic loss to growing crops. Material needed to perform maintenance
should be stockpiled on the farm. The small investment required for the stockpiled
materials will be more than offset by the capability to make timely repairs.
The ditches should be inspected routinely and maintenance requirements assessed.
The timing of the inspections depends somewhat on the type of ditch lining and the
soil and climatic conditions at the site. However, as a minimum, ditches should be
thoroughly inspected before every irrigation season. Where rigid linings are used in
northern climates, the lined ditch should be inspected just before ground freezing to
make sure that drainage is adequate.
Cracks in concrete and masonry structures should be caulked to prevent water entry
and minimize freeze damage. Masonry structures are particularly susceptible to crack-
ing in cold climates. Metal structures should be painted and rubber-type seals replaced
as needed. Metal gates should be lubricated and, where possible, left partially open
during the winter. This prevents them from sticking or rusting closed. Ditches and
structures should be drained when not in use. Water ponded in or near structures con-
tributes to frost damage and structure deterioration.
The lifespan and annual maintenance costs of irrigation ditch lining will vary de-
pending on site conditions and the quality of the initial construction. In the absence of
reliable local data, a lifespan of 15 yr and annual maintenance costs of 5% of the in-
stallation cost are recommended for nonreinforced concrete, flexible membrane, and
galvanized steel ditch linings.
11.3.4.2 Pipelines. Pipelines should be inspected for leakage at least once a year.
Leaks may be spotted from wet soil areas above the line that are otherwise unex-
plained. Small leaks in concrete pipelines can be repaired by carefully cleaning the
pipe exterior surrounding the leak, then applying a patch of cement mortar grout. For
larger leaks, one or more pipe sections may have to be replaced. Longevity of concrete
pipelines can be increased by capping all openings during cold winter months to pre-
vent air circulation.
Small leaks in plastic pipe, except at the joints, can sometimes be repaired by press-
ing a gasket-like material tightly against the pipe wall around the leak and clamping it
with a saddle. Leaky joints or damaged pipe sections must be replaced with a new pipe
section. Special gasketed steel or plastic fittings must be used to fasten a section of
replacement pipe into the line if the pipe cannot be displaced horizontally to join the
original belled joints.
Where water is supplied from a canal to portable surface pipe, sediment often ac-
cumulates in the pipe. This sediment should be flushed out before the pipe is moved.
Otherwise, the pipe will be too heavy to be moved by hand and may be damaged if it
is moved mechanically.
Buried plastic pipelines can be expected to have a usable life of about 15 years if
well maintained. The annual cost of maintenance can be estimated as approximately
1% of the installation cost.
Acknowledgements
Much of Section 11.3 of this chapter is based on Chapter 11, Farm water distribu-
tion systems, of Design and Operation of Farm Irrigation Systems (1980), to which
A. S. Humpherys and E. J. Pope were major contributors.
REFERENCES
Abt, S. R., and K. J. Staker. 1990. Rating correction for lateral settlement of Parshall
flumes. J. Irrig. Drain. Eng. 116(6): 797-803.
Abt, S. R., K. Thompson, and K. Staker. 1989. Discharge corrections for longitudinal
settlement of Parshall flumes. Trans. ASAE 32(5): 1541-1544.
Ackers, P., and A. J. M. Harrison. 1963. Critical-depth flumes for flow measurements
in open channels. Hyd. Res. Paper No. 5. Wallingford, Berkshire, England: Dept.
Sci. and Indust. Res., Hyd. Res. Sta.
Ackers, P., W. R. White, J. A. Perkins, and A. J. M. Harrison. 1967. Weirs and Flumes
for Flow Measurement. New York, N.Y.: John Wiley & Sons. 327 pp.
ASAE. 2005a. EP267.6: Principles and practices for prevention of mosquito sources
associated with irrigation. St. Joseph, Mich.: ASAE.
ASAE. 2005b. S376.1: Design, installation and performance of underground,
thermoplastic irrigation pipelines. St. Joseph, Mich.: ASAE.
ASME. 1971. Fluid Meters: Their Theory and Applications. New York, N.Y.:
ASTM. 2005a. Standard specification for reinforced concrete culvert, storm drain, and
sewer pipe. ASTM C76-05, Vol. 04.05. West Conshohocken, Pa.: American
Society for Testing Materials.
ASTM. 2005b. Standard specification for concrete pipe for irrigation and drainage.
ASTM C188-05, Vol. 04.05. West Conshohocken, Pa.: American Society for
Testing Materials..
Bergstrom, W. 1961. Weed seed screens for irrigation systems. Ext. Publ. Bull. 43.
Moscow, Idaho: Univ. of Idaho, Pacific Northwest Coop. Agr. Ext. Serv.
Bondurant, J. A., and W. D. Kemper. 1985 Self-cleaning, non-powered trash screens
for small irrigation flows. Trans. ASAE 28(1): 113-117.
Booher, L. J. 1974. Surface irrigation. FAO Agri. Dev. Paper No. 95. Rome, Italy:
FAO.
Bos, M. G., ed. 1989. Discharge Measurement Structures. Publ. No. 20. 3rd rev. ed.
Wageningen, The Netherlands: Int’l. Inst. for Land Reclamation and Improvements
(ILRI).
Bos, M. G., J. A. Replogle, and A. J. Clemmens. 1991. Flow Measuring Flumes for
Open Channel Systems. St. Joseph, Mich.: ASAE.
Brater, E. F., and H. W. King. 1976. Handbook of Hydraulics. 6th ed. New York,
N.Y.: McGraw-Hill Book Co.
Burt, C. M. 1987. Overview of canal control concepts. In Planning, Operation,
Rehabilitation and Automation of Irrigation Water Delivery Systems. Proc. Symp.,
ASCE, 81-108. D. D. Zimbelman, ed. New York, N.Y.: American Soc. Civil
Engineers.
Burt, C. M., R. Angold, M. Lehnkuhl, and S. Styles. 2001. Flap gate design for
automatic upstream water level control. J. Irrig. Drain. Eng. 127(2): 84-91.
Discussion/closure in 128(4): 264-265.
Calder, G. G., and L. H. Weston. 1966. Automatic system of farm irrigation. New
Zealand J. Agr. 112(2).
Chow, V. T. 1959. Open Channel Hydraulics. New York, N.Y.: McGraw-Hill Book
Co.
Clemmens, A. J. 1987. Delivery system schedules and required capacities. In
Planning, Operation, Rehabilitation and Automation of Irrigation Water Delivery
System. Proc. Symp., ASCE, 18-34. D. D. Zimbelman, ed. New York, N.Y.:
Clemmens, A. J. 1992. Feedback control of a basin irrigation system. J. Irrig. Drain.
Eng. 118(3): 480-496.
Clemmens, A. J., M. G. Bos, and J. A. Replogle. 1993a. FLUME Design and
calibration of long-throated measuring flumes. Version 3.0, Pub. #54, WCL# 1751.
Wageningen, The Netherlands: Int’l. Inst. for Land Reclamation and Improvement
(ILRI).
Clemmens, A. J., A. R. Dedrick, and R. J. Strand. 1993b. Basin 2.0 for the design of
level-basin irrigation systems. In Proc. ASCE Nat. Conf. on Irrig. & Drain., 875-
882. New York, N.Y.: American Soc. Civil Engineers.
Clemmens, A. J., T. L.Wahl, M. G. Bos, and J. A. Replogle. 2001. Water
measurement with flumes and weirs. Publ. #58. Waginingen, The Netherlands:
Int’l. Inst. for Land Reclamation and Improvement (ILRI).
Clemmens, A. J., and J. A. Replogle. 1980. Constructing simple measuring flumes for
irrigation canals. USDA Farmer’s Bull. No. 2268. Washington, D.C.: USDA.
Coulthard, T. L., J. C. Wilcox, and H. O. Lacey. 1956. Screening irrigation water. Agr.
Eng. Div. Bull. A.E. 6. Vancouver, BC, Canada: Univ. Brit. Columbia.
Dedrick, A. R. 1984. Special design situations for level basins. In Proc. Twelfth
Congress on Irrigation and Drainage, 1(B), Question Q39, R29(4): 465-474. New
Delhi, India: ICID.
Dedrick, A. R. 1986. Control requirements and field experience with mechanized level
basins. Trans. ASAE 29(6): 1679-1684.
Dedrick, A. R. 1989. Improvements in design and installation features for mechanized
level basin systems. Appl. Eng. Agric. 5(3): 372-378.
Dedrick, A. R. 1990. Level-basin irrigation: An update. In Visions of the Future, Proc.
Third Nat’l. Irrigation Symp., 34-39. St. Joseph, Mich.: ASAE.
Dedrick, A. R., and Y. Reinink. 1987. Water ponding on level basins caused by
precipitation. Trans. ASAE 30(4): 1057-1064.
Dedrick, A. R., J. A. Replogle, and L. J. Erie. 1978. On-farm level-basin irrigation:
Save water and energy. Civil Eng. 48: 60-65.
Dethridge, J. S. 1913. An Australian water meter for irrigation supplies. Eng. News
(New York) 7(26).
Evans, R. G. 1977. Improved semiautomatic gates for cut-back surface irrigation
systems. Trans. ASAE 20(1): 105-108, 112.
Fischbach, P. E., and R. Gooding II. 1970. An automated surface irrigation valve.
Agric. Eng. 52(11): 584-585.
Goussard, J. 1987. NEYRTEC automatic equipment for irrigation canals. In Planning,
Operation, Rehabilitation and Automation of Irrigation Water Delivery Systems,
Proc. Symp., ASCE, 121-132. D. D. Zimbelman, ed. New York, N.Y.: American
Soc. Civil Engineers.
Haise, H. R., and E. G. Kruse. 1966. Pneumatic valves for automation of irrigation
systems. In Proc. 6th Congr. Int’l Comm. on Irrig. and Drain. Specialty Sessions.
Report No. l: S-1 to S-8. New Delhi, India: ICID.
Haise, H. R., E. G. Kruse, and N. A. Dimick. 1965. Pneumatic valves for automation
of irrigation systems. USDA-ARS 41-104. Washington, D.C.: USDA.
Haise, H. R., E. G. Kruse, M. L. Payne, and H. R. Duke. 1978. Automation of surface
irrigation: Summary of 15 Years USDA research and development at Fort Collins,
Colo. Prod. Res. Report #179. Washington, D.C.: USDA, SEA-AR.
Hart, W. E., and J. Borrelli. 1970. Mechanized Surface Irrigation Systems for Rolling
Lands. Contr. No. 133. Berkeley, Calif.: Univ. of Calif., Water Resources Center.
Humpherys, A. S. 1969. Mechanical structures for farm irrigation. J. Irrig. Drain.
Div., ASCE 95(IR4): 463-479.
Humpherys, A. S. 1978. Improving farm irrigation systems by automation. Preprint,
10th Congr. Int’l Comm. on Irrig. and Drain. Rpt. 5, Q. 35: 35.90-35.98. New
Delhi, India: ICID.
Humpherys, A. S. 1986. Automated Farm Surface Irrigation Systems: Worldwide.
New Delhi, India: Int’l Commission on Irrigation and Drainage.
Humphreys, A. S., and J. A. Bondurant. 1977. Cast-in-place, 2-foot concrete
trapezoidal flow-measuring flumes. Tech. Bull. 1566:43. Washington, D.C.: USDA.
ITRC. 2001. Flap Gate. Report No. R 01-003.San Luis Obispo, Calif.: California
Polytechnic State Univ., Irrigation Training and Research Center.
Keller, J. 1975. Economic pipe size selection chart. In ASCE, Irrig. and Drain. Div.
Spec. Conf. Proc., 109-121. Logan, Utah: American Soc. Civil Engineers.
Kemper, W. D., and J. A. Bondurant. 1982. Turbulent flow self cleaning trash screens.
In Proc. Irrigation Association Technical Conference, 75-84. Falls Church, Va.:
Irrigation Assoc.
Kindsvater, C. E., and R. W. Carter. 1959. Discharge characteristics of rectangular
thin-plated weirs. ASAE Paper No. 3001. St. Joseph, Mich.: ASAE.
Kraatz, D. B., and I. K. Mahajan. 1975. Small hydraulic structures. FAO Irrig. and
Drain. Paper No. 26, Parts No. 1 and 2. Rome, Italy: FAO.
Merriam, J. L. 1973. Float valve provides variable flow rate at low pressures. In Proc.,
ASCE Irrig. and Drain. Spec. Conf., 385-402. New York, N.Y.: American Soc.
Civil Engineers.
Merriam, J. L. 1977. Level-top canals for semi-automation of on-farm irrigation and
supply systems. In Proc., ASCE and Drain. Spec. Conf., 217-224. New York, N.Y.:
Merriam, J. L. 1987a. Pipelines for flexible deliveries. In Planning, Operation,
Rehabilitation and Automation of Irrigation Water Delivery Systems. Proc. Symp.,
ASCE, 208-214. D. D. Zimbelman, ed. New York, N.Y.: American Soc. Civil
Engineers.
Merriam, J. L. 1987b. Design of semi-closed pipeline systems. In Planning,
Operation, Rehabilitation and Automation of Irrigation Water Delivery Systems.
Proc. Symp., ASCE, 224-236. D. D. Zimbelman, ed. New York, N.Y.: American
Parshall, R. L. 1950. Measuring water in irrigation channels with Parshall flumes and
small weirs. Circular 843. (Revision of USDA 1932 and 1941 Farmer’s Bull. No.
1683.) Washington, D.C.: U.S. Soil Conservation Service.
Pugh, W. J., and N. A. Evans. 1966. Weed seed and trash screens for irrigation water.
Bull. S-522S. Fort Collins, Colo.: Colorado State Univ. Agri. Exp. Sta.
Raemy, F. and W. H. Hager. 1998. Hydraulic level control by hinged flap gate. Proc.
Inst. Civil Engineers Water Maritime and Energy 130: 95-103.
Replogle, J. A. 1968. Discussion of “Rectangular cutthroat flow measuring flumes” by
G. V. Skogerboe and M. Leon Hyatt. J. Irrig. and Drain. Div., ASCE 94(IR3): 359-
362.
Replogle, J. A. 1970. Flow meters for water resources management. Water Res. Bull.
6(3): 345-374.
Replogle, J. A. 1977. Compensating for construction errors in critical-flow flumes and
broad-crested weirs. Flow measurement in open channels and closed conduits. Natl.
Bureau Stds. Spec. Publ. 484(I): 201-218.
Replogle, J. A. 1984. Some environmental, engineering, and social impacts of water
delivery schedules. In Proc. of 12th Congress Irrigation and Drainage, 965-978.
New Delhi, India: ICID.
Replogle, J. A. 1986. Some tools and concepts for better irrigation water use. In Proc.
Irrigation Management in Developing Countries Current Issues and Approaches,
an Invited Seminar Series, 117-148. K. C. Nobe, and R. K. Sampath, eds. Ft.
Collins, Colo.: Int’l School for Agricultural and Resource Development, Colorado
State Univ.
Replogle, J. A. 1987. Irrigation water management with rotation scheduling policies.
In Planning, Operation, Rehabilitation and Automation of Irrigation Water
Delivery System, I&D Div., ASCE Proc. Symp., 35-44. D. D. Zimbelman, ed. New
York, N.Y.: American Soc. Civil Engineers.
Replogle, J. A. 1999. Measuring irrigation well discharges. J. Irrig. Drain. Eng.
125(4): 223-229.
Replogle, J. A., and A. J. Clemmens. 1979. Broad-crested weirs for portable flow
metering. Trans. ASAE 22(6): 1324-1328.
Replogle, J. A., A. J. Clemmens, and M. G. Bos. 1990. Chapt. 10: Measuring
irrigation water. In Management of Farm Irrigation Systems, 313-370. G. Hoffman,
T. A. Howell, and K. H. Solomon, eds. St. Joseph, Mich.: ASAE.
Replogle, J. A., and J. L. Merriam, 1980. Scheduling and management of irrigation
water delivery systems. In Proc. 2nd Nat’l. Irrig. Symp., 112-126. St. Joseph,
Mich.: ASAE.
Replogle, J. A., and B. T. Wahlin. 1994. Venturi meter constructions for plastic
irrigation pipelines. Appl. Eng. Agric. 10(1): 21-26.
Replogle, J. A., and B. T. Wahlin. 2000. Pitot-static tube system to measure discharges
from wells. J. Hydraulic Eng. ASCE 126(5): 335-346.
Replogle, J. A., B. J. Fry, and A. J. Clemmens. 1987. Effects of non-level placement
on the accuracy of long-throated flumes. J. Irrig. Drain. Eng. 113(4): 584-594.
Robinson, A. R. 1968. Trapezoidal flumes for measuring flow in irrigation channels.
ARS 41-141. Washington, D.C.: USDA-ARS.
Robinson, A. R., C. W. Lauritzen, D. C. Muckel, and J. T. Phelan. 1963. Distribution,
control and measurement of irrigation water on the farm. USDA Misc. Publ. #926.
Washington, D.C.: USDA.
Robinson, E. P. 1972. Automatic control of farm channels. 8th Congress, ICID.
Question 28.2, Report No. 21: 28.2.293-28.2.296. New Delhi, India: ICID.
Schwab, G. O., D. D. Fangmeier, W. J Elliot, and R. K. Frevert. 1993. Soil and Water
Conservation Engineering. New York, N.Y.: John Wiley & Sons.
Seipt, W. R. 1974. Waterhammer considerations for PVC piping in irrigation systems.
Trans. ASAE 17(3): 417-423.
Skogerboe, G. V., R. S. Bennett, and W. R. Walker. 1972. Generalized discharge
relations for cutthroat flumes. J. Irrig. Drain. Div., ASCE 98(1R4): 569-583.
Skogerboe, G. V., V. T. Somoray, and W. R. Walker. 1971. Check-drop-energy
dissipator structures in irrigation systems. Water Management Tech. Report 9. Fort
Collins, Colo.: Colorado State Univ.
Stoker, R. 1978. Simple, efficient, automatic irrigation. Irrig. Age 12(6): 76-77.
USBR. 1951. Irrigation Advisors Guide. Denver, Colo.: Bureau of Reclamation, U.S.
Dept. of Interior.
USBR. 1963. Hydraulic design of stilling basins and bucket energy dissipators. Eng.
Mono. No. 25 (revised). Denver, Colo.: Bureau of Reclamation, U.S. Dept. of
Interior.
USBR. 1997. Water Measurement Manual. 3rd ed. Denver, Colo.: Bureau of
Reclamation, U.S. Dept. of Interior.
USDA-SCS. 1978. Furrow irrigation. Chapter 5, Section 15, in National Engineering
Handbook. Washington, D.C.: USDA Soil Conservation Service. 175 pp. (These
standards are being regularly revised and updated by the USDA Natural Resources
Conservation Service; see http://www.info.usda.gov/CED/. Detailed irrigation
structure design information applicable to a given state may be obtained from the
USDA-NRCS, State Engineer. Readers outside the USA should contact USDA-
NRCS, Washington, D.C.)
USDA. 1979. Field manual for research in agricultural hydrology. In Agricultural
Handbook No. 224. D. L. Brakensiek, H. B. Osborn, and W. J.Rawls, coordinators.
Washington, D.C.: USDA, Science and Education Administration.
USDA. 2006. Above ground, multi-outlet pipeline (ft.). Natural Resources
Conservation Service. Conservation Practice Standard, Code 431. August. 3 pp.
Wahl, T. L., A. J. Clemmens, J. A. Replogle, and M. G. Bos. 2005. Simplified design
of flumes and weirs. Irrig. and Drain. 54: 231-247.
Walker, R. E. 1987. Long crested weirs. In Planning, Operation, Rehabilitation and
Automation of Irrigation Water Delivery Systems. Proc. Symp., ASCE, 110-120. D.
D. Zimbelman, ed. New York, N.Y.: American Soc. Civil Engineers.
Wang, C., G. Guan, W. Cui, and J. Fan. 2005. Modification for the formula of head
loss for long-throated flume and the optimize of bodily form. In Int’l Commission
on Irrigation and Drainage, 19th Congress, Question 52; Report 11. New Delhi,
India: ICID.
CHAPTER 12
PUMPING SYSTEMS
Harold R. Duke (USDA-ARS,
Abstract. Pumping plants are frequently a major part of the installation and oper-
ating costs of a modern irrigation system. Whether designing a pumping plant or
evaluating a proposal from a dealer, the irrigator is well served by basic knowledge of
how pumps work and the necessary factors to be considered when choosing and oper-
ating the irrigation pumping plant. Operating characteristics, as detailed in literature
supplied by the manufacturer, include relationships between flow rate, pumping head,
rotational speed, and power required. Optimum pump selection depends on considera-
tion of the soil type, climate, and crops to be grown. Special attention must be given to
the pumping plant when the irrigation system changes, as may occur when groundwa-
ter levels decline or operating pressure changes due to changing the type of irrigation
system. Such changes may result in temptation to add a “booster” pump or connect
two or more pumps together into a common distribution pipeline. These changes re-
quire a careful assessment if the desired result is to be achieved. Economic operation
of the entire irrigation system can depend upon proper selection of the pump, power
unit, and fuel type as well as proper routine maintenance, testing, and adjustment.
Keywords. Affinity equations, Centrifugal, Efficiency, Engine, Head, Head-
discharge curve, Impeller, Motor, Net positive suction head, Power, Pump curve,
Pump stage, Pumping rate, System head, Total dynamic head, Vertical turbine.
12.1 INTRODUCTION
All pumps are mechanical devices which receive energy from a power source and
convert that energy to either velocity or pressure energy of a fluid. As commonly used
on farm irrigation or drainage systems, pumps lift water to a higher elevation and/or
increase the pressure of the water as necessary to operate the system.
12.1.1 Types of Pumps
Pumps are generally categorized into two basic types based on the method by
which energy is imparted to the fluid. Positive displacement pumps displace the fluid
using such mechanical devices as gears, pistons, screws, diaphragms, vanes, or rollers.
The pumping rate is nearly independent of pressure. This type of pump is not com-
monly used for irrigation or drainage. Fertilizer injection and pesticide spray pumps,
however, are often positive displacement pumps and thus provide accurate control of
chemical application.
Irrigation and drainage pumps, on the other hand, typically transfer energy by ki-
netic principles such as centrifugal or viscous forces or momentum. These are generi-
Positive
Displacement
1000
Total Head, m
Vertical
100 Turbine and
Centrifugal
Mixed Flow
10
Axial Flow
1
1 10 100 1000 10000
Pumping Rate, L/s
Figure 12.1. Applicability of various pump types.
cally called centrifugal pumps and include centrifugal, jet, vertical and submersible
turbines, axial flow, and mixed flow, which has characteristics of both turbine and
axial flow. Within these subcategories, the size, shape, speed, number of sequential
pump units, and mode of operation can vary significantly. In general, there are suffi-
cient commercially available models to permit proper selection to meet any on-farm
irrigation or drainage need. This group of pumps will be discussed in this chapter.
Figure 12.1 illustrates generalized applicability of various pump types as deter-
mined by total head and pumping rate required. Note that positive displacement pumps
produce the highest total heads, but are usually limited to small pumping rates. Most
centrifugal pumps can operate over a range of conditions. In general, as the total head
increases, the pumping rate decreases as a result of fluid slippage. The most common
irrigation and drainage pumps are centrifugal and vertical turbine. For high pumping
rates at low heads, such as pumping from a river diversion into a canal, propeller or
axial flow pumps are commonly used. Mixed flow pumps use impellers which are
shaped to impart the energy by both axial and centrifugal forces. Many of the newer
high volume, high efficiency turbine pumps are actually mixed flow pumps.
Since both irrigation and drainage waters often contain sediment and many existing
wells produce sand, it may be necessary to select a pump that is minimally affected by
abrasives. Sediment will wear any pump, and it is desirable, but not always practical,
to assure that pumped water is clear. Both propeller and centrifugal pumps can handle
a reasonable amount of sediment, but wear caused by sediment will require periodic
replacement of impellers and volute cases. Turbine pumps are more susceptible to
sediment damage.
12.1.2 Scope of Chapter
In this chapter, I emphasize how to select the proper pump for a specific job. The
theory of pumps is treated rigorously by Karassik et al. (1976) or can be found in the
books by Zipparo and Hasen (1993) or Addison (1966). These authors also discuss the
theories used to design new pumps, which is of lesser importance to those charged
with selecting an appropriate irrigation or drainage pump.
This chapter briefly describes the types of pumps most commonly used for irriga-
tion and drainage applications. I discuss the performance characteristics of pumps,
394 Chapter 12 Pumping Systems
considerations to determine the desired pumping rate, development of the system head
curve, and finally pump selection. Information is also provided to aid in selecting a
power unit and controls and to determine initial investment and operating costs. Field
testing of existing systems is discussed and some suggestions are offered on pump
maintenance and operation.
Chapters 14, 16, and 17 describe the selection and operation of specific irrigation
equipment that normally use pumps. Because of these detailed descriptions, I will not
discuss piping on the discharge side of the pump. The construction of drainage wells
(sometimes called relief wells) is briefly referenced in Chapter 9. A rigorous treatment
of well drilling and development can be found in Ground Water and Wells (Johnson
Division UOP, 1975) and is included in such texts as Todd and Mays (2004), DeWeist
and Davis (1966), or McWhorter and Sunada (1977). It is important that the designer
understand groundwater hydraulics to properly select a pump that will perform satis-
factorily when well drawdown changes or adjacent wells interfere with one another.
12.2 PUMP COMPONENTS AND CHARACTERISTICS
Texts such as Hicks (1957) or Colt Industries (1979) provide background informa-
tion and nomenclature for those not familiar with pumps. These books and manufac-
turers’ literature typically provide cutaway drawings and descriptions of the various
components. A brief description of the principle components of centrifugal and turbine
pumps follows.
12.2.1 Pump Descriptions and Components
12.2.1.1 Centrifugal pumps. A centrifugal pump is normally located above the
water surface or in a dry well adjacent to the water supply. It has one or more impel-
lers fastened to a rotating shaft that turns inside a spiral-shaped volute case. Water
enters the eye (center) of the rotating impeller and is forced outward along the vanes
of the impeller by centrifugal force. The input energy is converted to velocity head as
water is forced outward. On leaving the impeller, part of this velocity head may be
converted to pressure.
Commonly used centrifugal pumps include: end suction, double suction, and multi-
stage. These descriptors refer to how water enters the eye of the impeller or how many
impellers are present. If two or more impellers are installed so that the discharge from
one is allowed to flow into the eye of the next, the pump is multistage. Water enters
the impeller eye from only one side in an end suction pump, while double suction
means that water enters the impeller from both sides.
Generally, the impellers are fixed onto the drive shaft which is then coupled to the
drive unit—either an electric motor or engine. Electrically powered systems may have
a common shaft for the motor rotor and the pump impeller. Because they are compact,
centrifugal pumps are often stocked as completely assembled units.
Almost all pumps have moving parts which require lubrication to reduce wear. In
some instances the bearings are lubricated and sealed at the time of manufacture. In
other instances oil or grease must be added periodically or continuously. Even the wa-
ter itself may be used as the lubricant.
Most centrifugal pumps have a packing gland around the drive shaft where it exits
the volute case. Some newer pumps use a mechanical silicate seal to prevent leakage
at this point and to allow an occasional drop of water to act as a lubricant. Older mod-
els use beeswax or graphite-impregnated hemp or Teflon packing, which provides
some support to the shaft, acting as a bearing but not permitting large leakage. A very
small amount of leakage through this packing gland is desirable to act as a lubricant.
Packing glands must occasionally be tightened or the packing replaced if leakage can-
not be controlled. Care should be taken to not over-tighten packing glands, thus cut-
ting off the lubricating water or increasing the torque necessary to turn the pump, as
this can result in pump failure or motor burnout.
12.2.1.2 Jet pumps. Jet pumps are a variation of the centrifugal pump, with modi-
fications to allow lifting water from deeper depths than can be realized by direct suc-
tion. The “jet” is located near the bottom of the suction pipe. A portion of the water
pumped to the surface is diverted back down the well through a secondary pipe for
delivery to the jet. The return water is passed through a Venturi where pressure is con-
verted to velocity. The high-velocity water jet transfers part of its momentum to water
entering the pump, which increases the pressure near the bottom of the suction line,
allowing a higher suction lift than otherwise possible (see discussion of net positive
suction head, Section 12.2.2.4). Deeper pumping lifts require a larger percentage of
the pumped water to be returned to the jet. Because energy is required to lift and pres-
surize the water returned to the jet, and this water is not discharged from the pump, jet
pumps tend to have low operating efficiencies. Therefore, they are generally used only
in small sizes, such as for domestic water supplies, where the cost of pumping is not a
major concern. Lubrication of the mechanical parts above ground is similar to that of a
centrifugal pump, and should be according to manufacturer’s recommendations.
12.2.1.3 Turbine pumps. The turbine pump also operates on the principle of cen-
trifugal force, with water discharged from the impeller nearly perpendicularly to the
axis of rotation. Turbine pumps are often used in deep wells or to raise water from
rivers, lakes, or sumps where space is limited or high head is required. These pumps,
illustrated in Figure 12.2, operate submersed in water, thus do not require priming.
Like the centrifugal, both the vertical turbine and submersible turbine impart energy to
the liquid by rotating one or more impellers inside the bowls. One impeller and its
bowl are referred to as a stage. When several of these bowl assemblies are coupled so
that the flow from the lower impeller moves directly to the eye of the next impeller
above, they form a multistage pump. For example, a four-stage pump has four bowl
assemblies coupled together.
The impellers are fixed on a shaft, called a line shaft. The line shaft of the vertical
turbine pump extends through the eye of each impeller, inside the column pipe to the
surface where it couples to a motor or gearhead. The line shaft is supported in the cen-
ter of the column pipe by bearings, typically at 2- to 3-m intervals. Water discharged
from the bowls also exits vertically upward and is carried within the column pipe. Oil-
lubricated pumps also have an oil tube surrounding the line shaft to direct lubricant to
line shaft bearings.
The column pipe supports the bowls beneath the pump head as well as serving as a
conduit to direct water to the surface. The distance from the bowls to the pump head is
referred to as the pump setting. The pump head serves as an elbow to redirect the flow
horizontally and to support the weight of the column pipe, bowls, impellers, and line
shaft. It generally rests on a foundation and serves as a mounting bracket for the motor
or gearhead.
The vertical turbine pump may be either oil or water lubricated. An oil-lubricated
pump should be used if sand or sediment is being pumped because the encasing oil
tube protects line shaft bearings from exposure to abrasives. A lightweight turbine oil
must be allowed to drip into the annulus between line shaft and oil tube continuously
while the pump is operating. Some of this oil passes out the weep hole located in the
casting of the upper stage. For irrigation, it is not critical whether the water contains
small amounts of oil; however, most domestic pumps are water lubricated. The bear-
ings in the bottom of the pump are generally lubricated and sealed when manufac-
tured.
Water-lubricated pumps generally use rubber bushings that are lubricated by water
flowing up the column. When the pump setting exceeds about 13 m, it is advisable to
install a water prelubrication system which will dump several liters of stored water
along the line shaft just prior to starting the pump. Otherwise, the top rubber bushings
may overheat and seize to the line shaft before water rises in the column pipe to pro-
vide lubrication.
Adjusting Nut
Thrust Bearing
Hollowshaft Motor
Discharge Head
Line Shaft
Column Pipe
Bearing
Bowl
Impeller
Suction Bell
Figure 12.2. Major components of a typical electrically powered

vertical turbine pumping plant.
Gearhead or motor bearings also need lubrication. Motors are available with either
an oil reservoir or a grease fitting. Some larger motors and gearheads route a small
part of the pumped water through the motor or gearhead castings as a coolant.
The submersible turbine was developed for irrigation use in the early 1960s and is
becoming quite common. Submersibles are generally cheaper for smaller sized pumps
and deeper settings than the vertical turbine. They are popular where the entire pump-
ing plant needs to be below ground for esthetic or security reasons, such as in parks
and golf courses. In many instances, submersibles can be used in a crooked well where
it is impossible to use a vertical turbine if the line shaft is not straight.
A submersible turbine pump has a watertight electric motor coupled beneath the
bowl assemblies. The entire motor-pump assembly is submerged so that the water
provides cooling. Water enters the first stage from below as in the vertical turbine. The
discharge exits the top stage and flows to the surface in a pipe which is unobstructed
by bearings or line shaft. This pipe also serves to support the motor and pump. There
are no rotating parts except for the rotor in the electric motor, the short line shaft, and
the attached impellers.
Some disadvantages of submersible turbines are that the motor must have adequate
water flow around it to provide cooling, and the length of motor below the pump re-
duces the maximum pump setting and may reduce the capacity of wells with little
saturated thickness. Wells that pump sand may cause bearing problems in the pump
and motor. Submersible electric motors of 50 kW capacity and larger are quite expen-
sive, and if there are any problems with the motor, the entire pump and motor must be
pulled.
12.2.1.4 Axial flow pumps. Axial flow impellers have a very small component of
radial flow, with most of the flow paralleling the axis of rotation. The impeller of axial
flow pumps is shaped like a propeller, and is most useful where flow rates and pump
speed are relatively high compared to the head delivered. Axial flow pumps are typi-
cally used for low-head, high pumping rate applications, such as lifting water from a
river diversion into a canal or secondary pumping station, pumping water from a run-
off recovery pit to the head of a field, and emptying animal waste storage lagoons.
These pumps deliver high thrust loads on the main shaft, and often use heavy brass
bearings to support the propeller and main shaft. Because of the shallow setting of
most axial flow pumps, the bearings are readily accessible, and they usually require a
heavy grease applied with a grease gun.
12.2.1.5 Pump impellers. Three styles of impellers, open, semi-open, and en-
closed, are used in centrifugal or turbine pumps. These terms refer to whether the im-
peller vanes are enclosed by shrouds. An open impeller has no shrouds and is often
used to pump liquids that contain organic matter, such as sewage sludge, paper pulp or
water containing animal wastes. The semi-open impeller has a single shroud on the
side opposite the impeller intake, and must be adjusted so the operating clearance be-
tween the open side of the impeller vanes and the bowl or volute face is within very
close tolerance of about 0.1 mm. If this clearance is too great, then water simply recir-
culates from the outer edge to the inlet of the impeller. The closed impeller has
shrouds on both sides of the impeller vanes and clearance adjustment is not as critical.
Hicks (1957) shows a photographs illustrating the types of impellers and the wear
rings which restrict recirculation within the bowl.
12.2.2 Pump Characteristic Curves

Pumps, because of their mechanical nature, have certain well-defined operating char-
acteristics. These characteristics vary between types of pumps and between manufactur-
ers and their various models. Most pump manufacturers publish information describing
how each pump performs over a range of operating conditions. This information, in the
form of graphs or tables, is called the pump characteristic curve. Because of minor
variations in manufacturing, a new pump may not perform exactly as the characteristic
curve indicates. Some manufacturers’ curves represent average performance for the ini-
tial test group, but others prepare their curves for the pump having the poorest perform-
ance. In some instances, the manufacturer may add a factor of safety to assure that all
pumps will provide at least the published head at a specific pumping rate.
Four characteristic curves are commonly provided by a manufacturer. The four
types of curves are combined in Figure 12.3 and described below. The pump model,
indicated in Figure 12.3 as 10XY, typically indicates the outside diameter (10 inches)
of the bowl assembly and a specific pump design, indicated by the letters (XY).
12.2.2.1 Pumping rate variation with total dynamic head. This curve is often re-
ferred to as the TDH-Q curve and relates the head produced to the pumping rate. Gen-
erally, the head produced decreases as pumping rate increases, although some pumps
have more complex curves. The most common irrigation pump curves have shapes
similar to that in Figure 12.3. The total dynamic head, TDH, which a pump must im-
part to the liquid can be computed from Bernoulli’s equation. The specific operating
point will be illustrated in Section 12.3.3.
The TDH-Q curve for a single stage pump is shown in Figure 12.3. Section 12.3.5
describes multistage pumps. The curve can be used to evaluate how the pumping rate
will vary due to fluctuations in the total dynamic head.
If a pump is operated against a closed valve, the head generated is referred to as the
shutoff head, as illustrated in Figure 12.3. The efficiency of the pump at this point is
zero because it still consumes energy. The pipe on the discharge side of turbine or
20 65 70 75 TDH-Q
80
Total Dynamic
82
83
Head, m
15
Shutoff head, 19.2 m 82
80
10 75
E-Q
70
5
Power, kW
Output
7 BP-Q
6
10
NPSHR,
m
5 NPSHR-Q
0
30 35 40 45
Pumping Rate, L/s
Figure 12.3. Pump characteristic curve for a 1760 rpm, three-stage,
vertical turbine pump, model 10XY. Add two percentage points to efficiency
for 4 or more stages, subtract two percentage points for single stage.
centrifugal pumps must be capable of withstanding the shutoff head in case a dis-
charge valve is inadvertently closed.
12.2.2.2 Efficiency variation with pumping rate. Efficiency is the ratio of output
work delivered by the pump to input work. The relation between efficiency and pump-
ing rate is drawn as a series of envelope curves and labeled E-Q in Figure 12.3. There
is generally only one peak efficiency, which occurs at a specific pumping rate. Effi-
ciencies vary between types of pumps, manufacturers, and models, with larger pumps
generally having higher efficiencies. The efficiency also depends on the materials of
construction, finish on the castings or machining, and the type and number of bearings
used. For example, enameled impellers are smoother than bronze or steel and will give
a higher efficiency.
The output work the pump imparts to the liquid is called the water power, WP (or
water horsepower in foot-pound units), and is given by:
WP = Q × TDH/C (12.1)
where Q = the pumping rate, L/s (or gpm)
TDH = total dynamic head, m (or ft)
C = coefficient to convert units, 102.0 (or 3960)
WP = output power, kW (or hp).
The pump efficiency, Epump, is determined by dividing Equation 12.1 by the input
power, BP, expressed in kW (brake horsepower):
Epump = WP/BP = Q × TDH/(C × BP) (12.2)
where the terms are defined as before.
Knowing the pumping rate, total pumping head, and pump efficiency, one can use
Equation 12.2 to compute the input power required. Published curves for turbine
pumps do not include such losses as line-shaft bearing and gearhead friction. The
manufacturer’s reported efficiency for these pumps is for a specific number of stages.
It is necessary to adjust the reported efficiency upward or downward, depending on
the number of stages.
12.2.2.3 Input power variation with pumping rate. As mentioned previously, the
input power is the brake power required to drive the pump, expressed in kilowatts, and
the relationship to pumping rate is commonly called the BP-Q curve. The peak power
requirement is usually at an intermediate pumping rate as shown in Figure 12.3. Other
pumps may have the highest power demand at the lowest pumping rate and the re-
quired input power declines as Q increases. The shape of the BP-Q curve is a function
of the TDH-Q and E-Q curves. The vertical scale for most BP-Q curves is small,
which limits accuracy. Manufacturers suggest computing the BP from Equation 12.2
rather than the curves.
12.2.2.4 NPSHR variation with pumping rate. The fourth characteristic curve is
the net positive suction head required, NPSHR, as a function of the pumping rate, i.e.,
the NPSHR-Q curve. As water is drawn into the pump impeller, its pressure is re-
duced. NPSHR expresses the amount of energy required to move water into the eye of
the impeller. If the absolute pressure in the water drops below the vapor pressure of
the water at its current temperature, then the water will vaporize, creating vapor bub-
bles which can collapse violently when pressure increases, a condition called cavita-
tion. Cavitation must be avoided, as extended operation under cavitating conditions
can physically destroy the pump. NPSHR is a function of the pump design, pump
speed, impeller shape, liquid properties, and pumping rate and varies for different
types of pumps, manufacturers, and models. Values are determined by the manufac-
turer from laboratory tests.
A thorough discussion of how to use NPSHR-Q values is given in below. It is nec-
essary to consider NPSHR only for the first stage of a multistage turbine pump. How-
ever, for other pumps operated in series, NPSHR must be checked for each pump.
12.2.3 Changing Impeller Speed
Although most pumps can be operated over a range of speeds, the performance is
affected by the rotational speed of the impeller. The characteristic curves described by
the manufacturer are determined for a specific impeller speed. The speed is generally
shown on the pump curve.
Most modern electrically driven irrigation pumps are directly coupled to the motor
shaft and turn at the same speed as the motor, thus once the motor is selected, operat-
ing speed is not a variable. Common nominal operating speeds for electric motors in
the U.S. (60 Hz) are 1160, 1760 or 3520 rpm, and pump manufacturers publish char-
acteristic curves for speeds near these. Many smaller pumps operate at 3500 rpm while
larger pumps operate at the lower speeds. The most common speed for electrically
driven well pumps is near 1750 rpm.
Belt-driven electric or internal combustion engine-driven pumps can operate over a
range of speeds. Speed of belt-driven electric pumps can be varied by changing pulley
sizes on the motor or pump. Internal combustion engine-powered vertical turbine
pumps are usually connected through a drive shaft and right-angle gearhead to trans-
mit power to the pump line shaft. Speeds for engine-driven pumps can be varied by
changing the throttle setting, changing the gearhead ratio on a vertical turbine, or a
combination.
Variation in pump speed can be used advantageously to adapt a single pump to a
range of system head curves (i.e., changing conditions) or to provide variable design
pumping rates. The necessity for a single pump to operate efficiently at two different
speeds increases the complexity of pump selection several fold.
Mathematical expressions known as affinity laws are described in the Colt Indus-
tries Handbook (1979). These are three equations which relate how the pumping rate,
head, and input power vary with change in pump speed, and are useful to estimate
change in pump performance with small changes in speed. The equations are:
Q1/Q2 = RPM1/RPM2 (12.3)
2
H1/H2 = (RPM1/RPM2) (12.4)
3
BP1/BP2 = (RPM1/RPM2) (12.5)
where Q, H, BP, and RPM can be in any consistent units, and the subscripts 1 and 2
refer to respective values of the parameters at the two speeds.
The affinity laws indicate that the pumping rate varies linearly with speed, the head
as the square of the speed, and the brake power as the cube. Thus, a small increase in
pump speed will produce slightly more water and a higher head but will require con-
siderably more power. The affinity laws are valid only as long as the pump efficiency
remains constant. They do not address how the pump efficiency will change with
speed. NPSHR also changes rapidly with changes in speed, but cannot be determined
from these equations.
60%
70%
80%
84%
87%
2500 RPM noisy
120
110
k
80 2000 95 k W
W
Total Head, m
75 k
W
1800
55 k
W
40 k
40 1200 W
800 20 k
W
7.5
kW
0
0 40 80 120 160
Pumping Rate, L/s
Figure 12.4. Effect of pump speed on TDH-Q, efficiency, and power curves.
Manufacturers test their pumps at various operating speeds and sometimes print
complex characteristic curves as a function of pump speed as shown in Figure 12.4.
Note that the change of head from the TDH-Q curve as speed changes from 2000 to
1200 rpm is not linear. Also note that the efficiency envelope curves have a shape
similar to a system head curve. Because of this similarity in shape, the pump effi-
ciency changes little with small variations in pump speed. However, it is safer to util-
ize pump curves determined at several speeds to interpolate operating characteristics
rather than the affinity laws.
The input power, BP, on Figure 12.4 is plotted as a family of lines that indicate the
size of motor or engine required to drive the pump, disregarding line shaft and gear-
head losses. This plot is useful for estimating power requirements as speed changes,
and is a safer alternative than the affinity laws.
12.2.4 Changing Impeller Diameter
As mentioned earlier, the operating characteristics of a given pump are also de-
pendent on the diameter of the impeller. It is common practice to reduce the diameter
of the impeller in both centrifugal and turbine pumps by machining, thus changing the
pump’s performance to better match a specific job. Figure 12.5 shows five TDH-Q
and four BP-Q curves. The 28-cm curve is for a full diameter impeller, the 26-cm
curve is for a small standard trim in impeller diameter, and the remaining curves are
for even greater standard trims. The pump supplier can order the pump with the next
larger impeller than required, and reduce the diameter in a lathe before final assembly.
Three additional mathematical equations relate impeller diameter to pump charac-
teristics. The Colt Industries (1979) handbook describes these relationships in detail
and gives examples of how they can be used. The equations are:
Q1/Q2 = D1/D2 (12.6)
2
H1/H2 = (D1/D2) (12.7)
3
BP1/BP2 = (D1/D2) (12.8)
where Q, H, and BP are defined as before and D is the impeller diameter. Note that the
pumping rate varies linearly with diameter, pumping head with the square of impeller
diameter ratio, and input power as the cube.
60
60%
70 %
80%
dia =
8 7%
28 cm
87%
80 %
Total Head, m
40 26
24
45
22 kW
20 30
20 cm 22 kW
15 .5
kW
kW
0
0 40 80 120 160
Pumping Rate, L/s
Figure 12.5. Effect of impeller diameter on TDH-Q curve.
Impeller diameter may only be reduced, as it is not feasible to increase the diameter
by adding additional material. The applicability of the affinity laws with respect to
diameter is limited to a maximum trim of about 10%. As with changes in impeller
speed, changes in efficiency limit application of these equations.
Pump curves often show performance with several diameter impellers as in Figure
12.5. It is common practice to interpolate required trim amounts from characteristic
curves such as Figure 12.5, and using the manufacturer’s curves to determine specific
trim is preferred over Equations 12.6 through 12.8.
12.2.5 Hydraulic Thrust
As energy is transferred to the water, forces act upon the impeller and other com-
ponents. In a vertical turbine pump, for example, these forces can be sizeable and can
stretch the line shafts. Bearings must be designed into the pump, motor, or gearhead to
support these forces. The line shaft must be of sufficient diameter to withstand not
only the torque required to turn the impellers, but also to limit the amount the line
shaft stretches as pressure builds. It is beyond the scope of this text to describe compu-
tation of thrust loads in a pump and selection of line shaft diameter. Most pump manu-
facturer’s catalogs contain an engineering section with the necessary graphs, tables,
and sample calculations to illustrate how to compute thrust loads. Line shaft stretch
will be discussed briefly in Section 12.3.5.3.
12.2.6 Pumps in Series
If a single pump provides insufficient head, then two or more pumps may be con-
nected in series. Pumps are connected so that the discharge from the first pump or
stage is piped into the inlet side of the second pump. If more than two pumps are in
series, they are connected so all the flow passes successively from one pump to the
next. Each successive pump simply adds more energy to the water. Series operation is
common with centrifugal pump systems where the operating conditions have changed
and the old pump is no longer capable of delivering the required head, such as after a
surface irrigation system has been converted to sprinkler. In such a case, a second
pump is connected in series to serve as a booster.
A multistage turbine or submersible turbine pump consists of single stages con-
nected in series. The same pumping rate passes through all stages and each one adds
additional head to the water. Unlike the centrifugal pump with a booster, such a multi-
stage pump is usually made up of identical stages.
A+B
Total Head, m
B
HA + H B
HB
HA
Pumping Rate, L/s
Figure 12.6. Combination of TDH-Q curves for pumps in series.
The combined head of two pumps operating in series is the sum of the individual
heads for each pump at the common pumping rate, as shown in Figure 12.6. By select-
ing several pumping rates and the respective heads, the combined head, HA plus HB, is
obtained to develop the new HA+B-Q curve. If each pump is identical, as in a multi-
stage turbine or submersible turbine pump, it is necessary only to multiply the head for
a particular pumping rate by the number of stages to get the combined value.
For pumps in series the flow from one pump is identical to the flow in the others,
i.e., QA = QB = QC. The combined input power for a series of pumps is the sum of the
input power for each pump for the specific Q plus any power needed to overcome fric-
tion. A combined BP-Q curve can be developed in a similar fashion to the combined
TDH-Q curve.
The equation to compute the combined efficiency value is:
Q × (TDH A + TDH B )
E ser = (12.9)
102 × ( BPA + BPB )
where Q is in L/s, TDH is in m, and BP is in kW.

The net positive suction head criteria must be satisfied for each pump in series, but
a combined curve has no meaning. Since a pump delivers water at a higher pressure
than at its inlet, for identical pumps in series (e.g., a vertical turbine), meeting the
NPSH requirement of the first stage will assure that there is no problem with subse-
quent stages. Such is not necessarily the case with dissimilar pumps, pumps connected
by long pipelines, or pumps at significantly different elevations. In such cases, the
NPSHR of a downstream pump may be greater than the upstream pump is capable of
delivering, especially if the downstream pump has significantly greater pumping ca-
pacity than the upstream.
12.2.7 Pumps in Parallel
Two or more pumps may also be operated in parallel. A typical example is several
small wells with small turbine pumps discharging into a common pipeline. Pumps are
Total Head, m
A
A+B
QA
QB
QA + Q B
Pumping Rate, L/s

Figure 12.7. Combination of TDH-Q curves for pumps in parallel.
often operated in parallel if the system requires wide variations in pumping rates at
approximately the same heads. Sprinkler irrigation systems having varying capacities
on different circuits, such as parks or golf courses, often require pumps in parallel.
To properly select pumps to be operated in parallel, it is necessary to develop their
combined operating characteristic curves. The procedure for developing a combined
TDH-Q curve is illustrated in Figure 12.7. First, select a particular total head, and then
determine QA and QB. The combined flow is the sum of the two and is plotted against
the originally selected head. New values of head are selected and the process repeated.
Only if the two pumps are identical will QA equal QB for all values of head.
To determine the BP-Q relationships, it is necessary to add the input power re-
quired by each pump for the same total head, add BPA to BPB, and then plot against the
combined flow, QA+B. Again, it is necessary to repeat the procedure for different
heads. The efficiency calculated for two pumps in parallel is:
(Q A + QB ) × TDH
E par = (12.10)
102 × ( BPA + BPB )
The computation sequence requires selecting a specific total head; determining val-
ues for QA, QB, BPA, and BPB; calculating efficiency; and plotting against the com-
bined pumping rate. The resulting E-Q curve may have more than one peak efficiency.
For more than two pumps in parallel, it is necessary to add the pumping rates in the
numerator and BP values in the denominator.
If parallel pumps are not taking water from a common source, different pumping
heads are generated by each pump and the analysis becomes more difficult. In practice,
there are many situations where two or more improperly selected pumps operate in par-
allel. The consequence is invariably poor efficiency, and one pump may provide virtu-
ally all the water while the other operates very inefficiently or even with reversed flow.
12.3 SELECTING A PUMP

Proper selection of a pump requires knowledge and use of the pump characteristic
curves described in previous sections. An equally important aspect of pump selection
is to determine the operating characteristics required of the system. The procedure for
selecting a pump is to first determine the desired pumping rate and the relation be-
tween pumping rate and head required by the water delivery system. Only then can
pump manufacturers’ characteristic curves or tables be used to select a pump that will
operate efficiently at or near the design pumping rate.
Engineers, pump installers, and irrigation equipment salespeople have historically
determined a single design pumping rate and total dynamic head to use to select the
pump. However, few irrigation or drainage systems have fixed operating conditions. A
more realistic approach is to select the pump based on a range of pumping rates about
the design pumping rate, which will result in satisfactory performance over the ex-
pected operating conditions.
12.3.1 Pumping Rate Required
The major role of pumping in crop production is to control the soil-water environ-
ment. In humid or lowland areas, this may mean pumping to improve soil aeration or
drainage. Where precipitation is not reliable or sufficient, pumps are used to increase
soil water. Chapters 8 and 9 detail evaluation of requirements for irrigation and drain-
age, respectively, and define the minimum pumping rate for certain applications.
Physical, economic, and legal considerations constrain the maximum pumping rate,
particularly for irrigation. The maximum rate at which a well can deliver water is lim-
ited by both the aquifer properties and the construction and development of the well.
Thus, it is imperative that tests be conducted to determine the capacity of the well and
the relation between pumping rate and water level prior to pump selection. The pump-
ing rate from surface supplies may be limited by ditch or pipeline conveyance capac-
ity. Legal rights to water use limit pumping rates in many areas.
The consequences of improper selection of pumping rate are costly. Large pumping
rates may result in pump surging or wells that produce sand, either of which will dam-
age the pump. Excessive design pumping rate also results in greater than necessary
capital cost, higher demand charges for electrically powered units, and may cause ex-
cessive runoff and erosion. Selection of the optimum pumping rate requires considera-
tion of both crop water needs and soil characteristics.
For drainage conditions, it is generally desirable to pump at a rate to lower the wa-
ter table a preselected amount within a given time while minimizing pumping cost.
Too low a pumping rate, however, results in insufficient drainage and possible crop
damage or salt accumulation.
12.3.1.1 Crop water requirements. Crop water use depends on many factors, in-
cluding climate, type of crop, and stage of crop growth. Methods of estimating crop
water use are discussed in Chapter 8. The total seasonal water use is not to be con-
fused with the amount of water that must be supplied by irrigation. Some precipitation
occurs in most areas, and the amount that is useful for crop production depends on
when it falls, how much infiltrates the soil, and how much can be stored in the root
zone. Thus, the total irrigation requirement is less than the total crop requirement by
an amount equal to the effective precipitation.
Crop water use is not a constant throughout the growing season, but increases with
increasing temperature, solar radiation, wind, and crop cover and with decreasing hu-
midity. As discussed in the next section, it is not usually necessary to pump irrigation
water at the rate of peak crop water use.
12.3.1.2 Soil water storage. One of the major functions of soil in the agronomic
environment is that of a water storage medium. Chapter 6 details soil water transport,
plant water uptake, and the concept of available water. All soils are capable of storing
some amount of water until it can be used by the plant. This available water capacity
may range from 0.05 cm/cm for very sandy soils to 0.2 cm/cm for clay-loam soils. The
available water storage capacity in the root zone is typically from three to twenty
times the maximum daily crop water use. Thus, the soil serves as a buffer against
short-term extremes in crop water use rates. This means that an irrigation pump can
provide satisfactory performance even when the pumping rate is somewhat less than
the peak crop water use rate.
To determine the risk associated with a given irrigation rate and soil type, Heer-
mann et al. (1974) analyzed 60 years of climatic data from eastern Colorado. They
present the results in terms of net application rate per unit area (L/s per ha) required to
meet crop water requirements for different values of available soil water at several
probability levels. It is important to note, however, that the pumping rate must be
greater than the net application rate because no irrigation system distributes water with
perfect efficiency or uniformity. The design pumping rate must be increased to com-
pensate for application efficiency and lack of uniformity. Application efficiencies may
range from 95% or more for drip or in-canopy spray irrigation systems to considerably
less than 50% for surface irrigation on coarse-textured soils. The impact of application
uniformity on net irrigation water requirements depends on management goals. If the
irrigator desires that almost all the field be adequately irrigated (mean low quarter
concept, see Chapter 13), then even if the application uniformity is 85%, the total
pumping must be 30% greater than the net requirement. Thus, the required pumping
rate may be two or more times the required net application rate.
The pumping capacity must account for application inefficiency and downtime that
may result from system maintenance, load control management for electrically pow-
ered pumps, and other interruptions to pumping. The gross capacity needed for such
systems can be computed by:
Qn
Qg = (12.11)
Ea (1 − Dt )
where Qg = gross pumping capacity needed
Qn = net capacity needed
Ea = application efficiency, expressed as a decimal fraction
Dt = fraction of the pumping period that the pumping systems is inoperable
(i.e., the downtime), as a decimal fraction.
12.3.1.3 Soil intake rates. The rate of water infiltration is an important consideration
for any irrigation system. It is probably the most important factor for surface irrigation
systems because intake rate controls advance of water across the soil surface. The in-
fluence of intake rates on surface irrigation systems is covered in detail in Chapter 13.
The role of intake rates is somewhat different for sprinkler irrigation systems.
Sprinkler systems are capable of high application efficiency and uniformity only if the
system, rather than the soil, controls infiltration. The application rate must be suffi-
ciently low that the water applied is either infiltrated immediately or stored in small
surface depressions until it infiltrates. Otherwise, water runs across the surface to low
areas, and the uniformity advantages of the sprinkler are lost.
The application rate of a moving sprinkler depends on the water flow rate, the
shape of the sprinkler distribution pattern, and the area wetted by the sprinkler. The
area wetted by a given sprinkler head is relatively independent of the flow. Thus, the
application rate is almost proportional to the sprinkler flow rate. For moving sprin-
klers, the depth of application depends on speed of movement but peak application
rate does not.
Thus, for sprinkler systems, too low a pumping rate will provide insufficient water
for optimum crop growth, while too large a pumping rate will result in surface runoff.
Excessive surface runoff leads to reduced uniformity and necessitates increased pump-
ing to adequately irrigate the “dry spots.”
12.3.1.4 Management criteria. The pumping rate is also affected by operational
decisions such as the hours of irrigation per day and irrigation days per week. Most
irrigators anticipate some shut-down time to maintain the system. Most surface irriga-
tion systems require labor to change sets, and pump schedules are often modified to
coincide with labor availability. When the hours of operation are reduced to make
these allowances, then the pumping rate must be increased to provide adequate water
during the critical crop period.
Other operating decisions that impact the design pumping rate are the cropping pat-
tern and the area to be irrigated. Preplant irrigation to fill the root zone can reduce the
design pumping rate for certain crop, climatic, and soil conditions.
12.3.2 Total Pumping Head
12.3.2.1 System head curve. The system head curve describes the total head re-
quired to move water through the pumping system as a function of water flow rate.
The total head required is independent of the pump, except for the friction loss that
occurs in the column pipe of a vertical turbine pump. Figure 12.8 shows a typical sys-
tem head curve which illustrates the parameters contributing to the total head and how
they vary as the pumping rate increases. Not all of the parameters illustrated are appli-
cable to every system. The total head is the sum of the static lift, static discharge head,
well drawdown, friction, and operating pressure head.
Static lift is the vertical distance between the centerline of the pump discharge and
the water source elevation when the pump is not operating. If the water source is be-
low the pump elevation, the static lift is positive. If the pump is located below the wa-
System Head
Pressure
Head
Head, m
Friction
Head
Well
Drawdown
Static Discharge Head
Static Lift
Pumping Rate, L/s

Figure 12.8. Components of the system head curve.
ter, for example in a sump, the static lift is negative. Static lift is independent of the
pumping rate, although it may vary with time. In that case the system head curve is
also time dependent.
The static pumping head is the elevation difference between the centerline of the
pump and the eventual point of use. When a pump discharges directly into a canal
through a level pipe, the static pumping head is zero. If, however, the pump supplies
water to a distant point at a different elevation, then it is necessary to compute the
static pumping head. To obtain this value, subtract the elevation of the pump from the
elevation of the final point of delivery.
The static discharge head is independent of the pumping rate, but may also be time
dependent. For example, the static discharge head will change with time if pumping to
a center pivot that moves over uneven terrain.
Pumping water from an aquifer causes the water level in the well to decline. This
well drawdown depends on the pumping rate, the aquifer properties, the well radius,
and the time the pump has operated. The drawdown can be computed using one of
several equations, depending on the conditions. Methods for calculating drawdown are
given by McWhorter and Sunada (1977), Todd and Mays (2004), and DeWiest and
Davis (1966), among others, and are beyond the scope of this chapter.
The most satisfactory method to determine the relationship between well draw-
down and pumping rate is to conduct a well test. To design a pump, the well test is
best conducted by the well driller or pump installer, using the contractor’s own pump.
The well is pumped at various rates, drawdown allowed to stabilize, and the draw-
down is measured. The well test can be conducted most quickly by beginning pump-
ing at the highest flow rate, allowing the drawdown to stabilize, and then successively
reducing the pumping rate. Care must be taken to pump at a constant rate for sufficient
time that the drawdown approaches a steady value.
Friction causes loss of head when water flows through a pipe system. The friction
head can be determined from the Darcy-Weisbach or Hazen-Williams equations,
which are described in Chapter 15 and in most hydraulics texts. Friction tables, nomo-
grams, or curves provided by pipe manufacturers are convenient to evaluate pipe fric-
tion. Friction tables for steel pipe can be found in the Colt Industries (1979) and Inger-
soll-Rand (1977) hydraulic handbooks. Friction tables for plastic, gated aluminum,
and concrete pipe can be obtained directly from the manufacturer.
The pump must add sufficient energy to the water to overcome friction. The water
velocity in a pipeline increases linearly with the pumping rate, but friction increases as
the velocity squared. Irrigation systems with long pipelines or undersized pipe may
have very significant friction. It is often economically feasible to select a larger pipe
size in order to reduce the cost of overcoming friction.
Friction loss must be determined on both the intake and discharge side of the pump.
It is necessary to compute the friction on the suction side of centrifugal pumps, espe-
cially, to assure that there is sufficient net positive suction head available to prevent
cavitation. This will be discussed more later in Section 12.3.6.
The friction loss in the pump head and column pipe of a vertical turbine pump must also
be computed. Manufacturer’s tables or curves report friction losses of column pipe as a
function of the pumping rate and account for the presence of the line shaft and supporting
bearings or bushings. Friction head in the column should be kept below 5 m/100 m. If it
exceeds this amount, consideration should be given to the next larger column pipe size.
Discharge head is the pressure head required at the point of discharge, such as the
outlets to a gated irrigation pipe, the head of water above the pump outlet in a drainage
ditch, or the pressure on a sprinkler. Except where water is discharged directly into an
open ditch or partially filled pipeline, all irrigation pumps require some discharge
head. The operating pressure and its relation to pumping rate are discussed in Chapters
16 and 17. Total head curves may be continuously increasing functions as illustrated
in Figure 12.8, or they may be step functions, such as with a center pivot having an
end gun which starts and stops.
Proper selection of a pump for a specific system requires knowledge of how the to-
tal head changes. The total head-flow rate relationship should be evaluated over the
entire range of operating conditions. If the water level in a well drops during the
pumping season or a center pivot sprinkler moves over significant elevation differ-
ences, then attention must be given to select a pump that can satisfy all conditions.
Most pumps will not operate at peak efficiency over wide ranges in total heads. If a
particular condition is prevalent, such as center pivot operation with the corner swing
arm operating, then the pump should be selected to operate efficiently for that condi-
tion yet satisfactorily for the other conditions encountered.
12.3.2.2 Variations in system head curve. The system head curve is time depend-
ent because well drawdown, friction, operating conditions, and static water level
change. Figure 12.9 illustrates two conditions which change the system head curves.
Figure 12.9a illustrates that the friction increases, i.e., the slope of the curve increases,
as steel or iron pipe corrodes. Steel or cast iron pipes may become partially blocked by
development of oxidized ferrous nodules or the pipes may rust through and begin to
leak. Vertical turbine pump column pipe is usually steel and is quite susceptible to
deposition or failure due to rust. Aluminum and polyvinylchloride (PVC) pipes, which
are commonly used in recent years, do not deteriorate by rusting, but the friction factor
for these materials can be increased by deposits on the walls. When groundwater
pumping exceeds the aquifer recharge, then the static water table is expected to de-
cline. Static water level decline, as illustrated in Figure 12.9b, increases the system
head curve. The rate of drawdown increases when the water table declines, because
saturated thickness is reduced. Proper selection of a pump for these conditions requires
consideration of how pumping conditions will change over the life of the pump.
(a) (b)
2000
Pumping Lift, m
Old Pipe
Friction Head, m
1980
Drawdown
New Pipe 2000
1980
Static Lift
Pumping Rate, L/s Pumping Rate, L/s

Figure 12.9. Effects of (a) pipe deterioration and (b) declining water levels
on system head curves.
For example, if additional wells are being drilled and new land irrigated, then one
might expect the rate of water level decline to accelerate. On the other hand, if pump-
ing is limited by regulation or artificial recharge is initiated, then static water levels
may stabilize or even recover.
It is important to know whether there will be well interference between the depres-
sion cones of nearby wells or whether underground boundaries impact a well’s long-
term performance. Failure to consider these factors can result in installation of an
oversized pump, which will begin to surge as the water table declines more rapidly
than expected.
12.3.3 Matching Pump Characteristics and System Head
It is possible to superimpose a pump’s characteristic curves of TDH-Q and E-Q
upon a system head curve, as illustrated in Figure 12.10. The point at which the TDH-
Q and system head curves cross is the operating point. The pump in Figure 12.10 will
produce 60 L/s at a dynamic head of 17 m.
In Sections 12.2.3 and 12.3.2 above, I discussed how system head curves may
change with time, operating conditions, or deterioration of components. It is necessary
to estimate the two extremes and the average system head curves as illustrated in Fig-
ure 12.11a. A pump should be selected to operate efficiently under the dominant oper-
ating conditions, but it must also deliver water at the required head under the extreme
circumstances. As illustrated in Figure 12.11a, the pump will have to produce signifi-
cantly different dynamic heads for a desired pumping rate of 60 L/s. In an earlier sec-
tion I showed that pumping rate decreases as head increases for both centrifugal and
turbine pumps. Referring to Figure 12.10, the pumping rate will drop from 60 L/s to
56.5 L/s when the head is increased 1 m. If the system head decreases from 17 m to 15
m, the pumping rate increases to 65 L/s.
Figure 12.11b illustrates the pumping rate change due to a change in the system
head curve. The curves sloping upward to the right are system head curves for spring
and fall. Perhaps the spring curve is lower due to a higher static water level prior to the
pumping season. The water level declines through the season to a level represented by
the fall curve.
TD
H-Q
18
Head, m
17
16
Efficiency, %
15
ead E-Q
em H 80
Syst
75
55 60 65
Pumping Rate, L/s
Figure 12.10. Superposition of system head curve and
TDH-Q curve to determine pump operating point.
(a) (b) Pump B

High
Total Dynamic Head, m

Avg Pump A Fall
Low
Spring
60 50 55 60
Pumping Rate, L/s Pumping Rate, L/s
Figure 12.11. The system head curve should be determined for expected extremes (a) so that
the effect of seasonal changes on pumping rate can be determined for prospective pumps (b).
The flow rate of pump B will change much less than that of pump A with seasonal changes.
The two curves labeled A and B which slope downward to the right are TDH-Q
curves for two possible pumps. The spring pumping rate for pump A is 60 L/s, but will
drop to 50 L/s by the end of the season because of the increased head. Pump B will
also deliver 60 L/s in the early season, but will only drop to 55 L/s in the fall.
Superimposing the TDH-Q and E-Q curves upon the system head curves (if all
have the same vertical and horizontal scales) for several pumps being considered pro-
vides a visual means of comparing each pump. The pump with the steepest TDH-Q
curve will have the least fluctuation in pumping rate when water levels or discharge
heads vary. If water is provided from a well in which the pumping level varies over
time, then it is desirable to select the pump with the steepest curve to maintain a nearly
constant pumping rate. Conversely, the flow rate to a center pivot with a corner swing
arm must vary as additional sprinkler heads turn on or off. This system will benefit
from a relatively flat pump curve.
12.3.4 Efficiency Considerations
Superimposing both TDH-Q and E-Q curves on the system head curve also illus-
trates how pump efficiency will change as the operating points change. The shape of
the E-Q curve is not the same for each pump and this shape has a large influence on
which pump will operate most efficiently over the range in operating conditions.
Ideally, the operating point on the TDH-Q curve of the selected pump should fall
slightly to the right of the peak efficiency. As the pump wears, the TDH-Q curve
moves downward in a manner similar to that of a smaller diameter impeller. The oper-
ating point will move to the left along the TDH-Q curve, and the efficiency will in-
crease. The net result is a decrease in input power required. If the new pump is se-
lected to operate to the left of the maximum efficiency, pump wear will move the op-
erating point to the left, further reducing the efficiency.
The importance of efficiency on power consumption can be computed with Equa-
tion 12.2. A simple example using Figure 12.12 may also be informative. Assume an
irrigator contacts two pump dealers to obtain a pump that will produce 60 L/s (1000
gpm) with a TDH of 30 m (100 ft). Dealer A selects a 25 cm (10 in.) pump, which will
produce 10 m of head per stage at the desired pumping rate. Thus, it will require three
stages of this pump to match the system head requirement. The cost of the three-stage
pump is $3000. Note the operating efficiency of this pump is just on the right-hand
side of the peak and has a value of 83%. Dealer B selects a 35 cm (14 in.) pump,
which will produce 15 m head at the required pumping rate. It is necessary to have
only two stages to meet the required 30-m head. The pump costs only $2400 because
of the one less stage, but is only 63% efficient and operates far left of the peak effi-
ciency. Since both pumps will produce exactly the same amount of water and meet the
system head required, the farmer may select the cheaper pump. Some pump dealers
have also emphasized that the simpler pump (fewer stages) is also better. Until recent
years, little effort has been made by pump dealers to calculate the impact of efficiency
on total pumping costs. Table 12.1 compares the operating cost for the two pumps
under a typical irrigation scenario. Note that the savings in power costs alone is $975
per year, which will more than offset the higher purchase price. Because of the lower
efficiency of pump B, it requires a 37.5 kW motor instead of the 30 kW motor on
pump A. Many electrical suppliers impose a demand charge to compensate for higher
costs of servicing large loads, which adds further to operating costs.
Dealer A Dealer B
20 55 Pump diameter 25 cm 20 Pump diameter 35 cm
60 60 70
75
80
Total Head, m
70
Total Head, m
75 84 85 80 75
80 70
10 80
75 10 60
70
60
20 40 60 80 60 80 100
(a) Pumping Rate, L/s (b) Pumping Rate, L/s
Figure 12.12. Comparison of the characteristics of two pumps,
each of which will deliver the design pumping rate and head.
Table 12.1 Comparative costs[a] for the two pumps of Figure 12.12.
Item Pump A Pump B
Initial investment $3000 $2400
Stages 3 2
Motor size 30 kW (40 hp) 37.5 kW (50 hp)
Annual demand charge $450 $560
Energy consumed 45,000 kWh 58,900 kWh
Annual power bill $3150 $4125
Average annualized cost $3750 $4805
[a]
Cost figures based on demand charge of $15 per kW, assumed 395,000 m3 (320 ac-ft) pumped annually,
power cost $0.07 per kWh, and 20-year expected life of the pumping plant.
12.3.5 Selecting Specific Pump Components

12.3.5.1 Number of stages. The previous section described a procedure to obtain
characteristic curves for two or more pumps in series. Superimposing these combined
curves on the system head curve, as illustrated in Figures 12.10 and 12.11b, allows
evaluation of a given pump operation. Multiple pumps operated in series are necessary
when the required pumping heads are greater than what a single pump or stage can
develop. In this section, I discuss how to determine the number of stages or series
pumps needed.
The selection procedure is similar to that for a single stage pump, and requires that
the system head curve and design pumping rate be known. The manufacturers’ cata-
logs are checked to determine pump models that will produce the design pumping rate
efficiently. Two such candidate pumps are shown in Figure 12.13. Single stage TDH-
Q curves for pumps A and B are shown in the lower part of the figure. The design Q is
60 L/s with a required system dynamic head of 120 m.
To determine the number of stages of a pump required, it is necessary to divide the
120-m head requirement of the system by the head produced by each pump for the
design Q of 60 L/s. Pump A produces 20 m for a single stage, thus will require six
stages. Pump B produces 60 L/s at 40 m per stage, thus only three stages are required.
Once the number of stages is known, the multistage TDH-Q curve can be developed
by simply summing the TDH for the same values of Q.
Curves of TDH-Q for the six-stage A and three-stage B pumps are plotted on Fig-
ure 12.13. Note that the single stage pump curve for B is steeper than for A, but the
multistage TDH-Q curve for the six-stage A curve is steepest. Thus, the slope of the
multistage TDH-Q curve is a function of both the number of stages and the slope of
the single stage curves.
Division of the 120-m system head by the single stage pump head results in exactly
six and three stages, respectively, for the pumps illustrated in Figure 12.13. Such is
160 A - 6 Stage
B - 3 Stage
120
d
Hea
s t em
Sy
80
B-
1S
tag
e
40 A-1
Stag
e
50 60 70
Pumping Rate, L/s
Figure 12.13. Comparison of multistage pump curves for two
bowl assemblies with different TDH-Q characteristic curves.
C-
C
3
-3
St
g
St
160 C- w/
ag
2 Tr
im
e
Sta
ge

120
ead
em H
Syst
80 C- 1
Sta
ge
40
0
50 60 70
Pumping Rate, L/s
Figure 12.14. Illustration of impeller trim to match a multistage pump
to the desired pumping rate and head.
seldom the case, however, as illustrated by pump C in Figure 12.14. Division of 120 m
head required by 50-m head produced per stage of pump C indicates 2.4 stages are
required. Two stages will produce 57 L/s at 112 m head, while three stages will pro-
duce 63 L/s at 130 m.
A solution to the lack of proper match is to select a three-stage model C pump but
reduce the diameter of the impellers. Figure 12.5 illustrates how the TDH-Q curves
change with impeller diameter. Manufacturer’s curves or the affinity laws (Equation
12.6) are used to compute the required trim.
The pump industry uses two different approaches to solve the problem represented
by pump C in Figure 12.14. One approach is to trim all three impellers to the same
diameter as described above to meet the required operating conditions. The other ap-
proach, often used with closed impeller designs, is to mix similar bowl models or im-
peller sizes to obtain the match. There seems to be no consensus among manufacturers
as to which is best. If only one impeller is trimmed, it should be assembled as the top
stage on a vertical turbine pump, to minimize the risk of cavitation. Whenever impel-
lers are trimmed, it may be necessary to reshape the vanes of the impellers to mini-
mize turbulence as water exits a blunt, thick vane tip.
12.3.5.2 Centrifugal pumps. The selection criteria described are also applicable
for centrifugal pumps. It is common to trim impellers of these pumps to match the
specific use. The decisions that must be made for pump selection are which model of
pump to use, how much to trim the impeller, and what speed to turn the impeller. As
described in the preceding section, manufacturer’s characteristic curves contain infor-
mation needed to make these decisions.
A centrifugal pump has suction and discharge fittings or flanges by which it is con-
nected to suction and discharge pipes. Care must be taken to minimize friction loss on
the suction side to avoid cavitation. Although some are capable of self-priming at
small suction head, most centrifugal pumps used for irrigation or drainage require a
foot valve on the intake end of the suction pipe to maintain prime when not operating.
12.3.5.3 Vertical turbine pumps. The selection of a specific pump model, deter-
mination of the number of stages, and need for impeller trim have been discussed pre-
viously. Other components that must be selected are the lineshaft, column pipe, pump
head, and drive units.
Line shaft size must be selected to assure that the necessary power can be transmit-
ted to the impellers and that the shaft will not stretch excessively under load. Manufac-
turers’ curves and tables account for the rotational speed, shear strength of the shaft
material, and the brake power transmitted to determine the shaft size required to with-
stand shear stresses.
The lineshaft is subjected to vertical forces that stretch the shaft. These vertical
forces are called thrust load and are the sum of the weight of the impellers, the weight
of the lineshaft, and the hydraulic thrust on the impellers. The hydraulic thrust is com-
puted by multiplying a thrust constant, provided by the manufacturer for each impel-
ler, by the total dynamic head generated by the pump. Lineshaft elongation is com-
puted from thrust load using tables or graphs. Lineshaft stretch may not exceed the
impeller clearance within the bowl, else an impeller adjusted to avoid rubbing on top
of the bowl when started under no load will rub against the bottom of the bowl as
pressure increases. When this occurs, it is necessary to either select a larger shaft or
have a greater clearance machined into the bowls.
The column pipe transmits water from the bowls to the pump discharge head. The
size, length of the column, and type of pipe connection must be selected. The column
pipe diameter should be large enough that the design pumping rate will result in fric-
tion losses less than 5 m/100 m. Manufacturers’ friction loss tables give the head loss
per 100 m of column pipe as a function of pumping rates, and are different for water
and oil lubricated pumps and for each size of lineshaft. The column pipe must be long
enough to keep the bowls submerged, which requires that the depth to water be known
under pumping conditions. The pumping depth is the sum of the depth to the static
water level and any drawdown, as described previously.
A turbine pump should have a suction pipe to direct water vertically before it enters
the bottom stage. The length of suction pipe may be dictated by the depth of the well
or sump and the necessity to set the pump near the bottom to permit pumping during
low water conditions. Special belled inlet sections minimize head losses in the suction
line, and in any case, suction pipes must not be so long that they reduce NPSH below
the required value.
A strainer or screen is not commonly installed on a pump operating in a well with a well
screen. However, pumps operating in open sumps, rivers, or lakes should have strainers or
screens on the intake side to prevent entrance of objects that might damage the impellers.
These strainers or screens must be kept clean, as plugging may result in pump cavitation.
The pump discharge head serves to support the weight of the column, lineshaft, and
bowl assemblies, acts as an elbow to redirect the flow to a horizontal direction, and
provides a mounting bracket for the motor or gearhead. The discharge head must be
compatible with the size of the column and the physical dimensions of the motor or
gearhead. The discharge flange or threaded coupling size is also a consideration. Dis-
charge and column pipes are commonly the same size.
The pump head should be sufficiently large that the friction loss is less than 1 m.
Manufacturers’ tables show the friction loss as a function of the pumping rate, as well
as the physical dimensions of several types of pump heads.
The electric motor or gearhead must match the physical dimensions of the dis-
charge head, and must also accept the lineshaft size selected and have sufficient bear-
ings to carry the thrust load. A short length of lineshaft, known as a head shaft, may
need to be fabricated to length to assure that the threaded sections are in the correct
location to permit tightening of the lineshaft nut for impeller adjustment. Gearheads
are available with a range of gear ratios to match the optimum speed of the power
source to the design speed of the pump.
12.3.6 Checking NPSHA to Prevent Cavitation
Energy is required to move water into the eye of the impeller. This was defined
previously as the net positive suction head required, NPSHR. The pump and its suc-
tion piping must be analyzed to assure that the net positive suction head available
(NPSHA) is greater than that required (NPSHR). If the NPSHA does not exceed the
NPSHR, then the pump will cavitate.
The positive suction head comes from either atmospheric pressure or a pressure in-
duced on the intake side of the pump, such as by an upstream pump or a higher water
level. NPSHA can be computed by
NPSHA = Pb – Hs – hf – Pv (12.12)
where Pb = barometric pressure
Hs = pumping suction lift
hf = friction head in the suction pipe
Pv = the vapor pressure of the liquid being pumped at the operating temperature.
All parameters are expressed in meters.
References such as the Colt Industries (1979) handbook and many physics texts
give values for atmospheric pressure at various altitudes. The barometric pressure
head, Pb, in m of water can be estimated as:
Pb = 10.33 – 0.00108 E (12.13)
where E is the elevation in m. The vapor pressure and specific gravity of water as a
function of temperature can be found in similar references. Vapor pressure, Pv, can be
estimated from Table 12.2.
Table 12.2. Water vapor pressure as a function of temperature.
°C 0 5 10 15 20 25 30 35 40 45 50
Pv, m 0.06 0.09 0.13 0.17 0.24 0.32 0.43 0.58 0.76 0.99 1.28
Friction loss in the suction pipe depends on pipe length, diameter, and fittings as
well as the pumping rate and must be determined accurately using friction tables or the
Darcy-Weisbach or Hazen-Williams formulas.
Example. Compute the NPSHA for a pump that is operating at an elevation of

1525 m (5000 ft) mean sea level (MSL) where the water temperature is 12°C (60° F).
Assume the pumping rate is 65 L/s (1030 gpm) and the suction line is a schedule 40
steel pipe 20 cm (8 in.) in diameter and 200 m (656 ft) long. The centrifugal pump is
located 2 m (6.56 ft) above the water surface, and has an NPSHR of 1.2 m (3.9 ft).
The atmospheric pressure head, from Equation 12.13, is 8.68 m (28.5 ft). From Ta-
ble 12.2, the water vapor pressure is 0.15 m (0.49 ft) and the specific gravity (Colt
Industries, 1979) is 0.999. The friction loss from Table 1, page 47, of the Colt Indus-
tries handbook is 1.65 m/100 m length of pipe or 3.3 m for 200 m. Substituting these
values into Equation 12.12, as shown graphically in Figure 12.15, results in a com-
puted NPSHA of 3.23 m, which is greater than the NPSHR of 1.16 m. Therefore, the
pump will operate safely without cavitation. If NPSHR exceeded NPSHA, then the
pump could be lowered nearer the water surface or the suction pipe could be enlarged
to reduce friction loss. Changes in the system head or in water temperature may cause
a well-designed pump to cavitate, thus these calculations should be based on the maxi-
mum expected NPSHA over the system life.
Atmospheric Pressure
Atmospheric Pressure
NPSHA, m
Vapor Pressure, m
Head
Friction Head, m
Suction Head, m
Absolute Zero Pressure

Figure 12.15. Determination of net positive suction head available
to assure pump will not cavitate.
12.3.7 Variable Pumping Rates

It is not always desired that irrigation systems operate at a constant pumping rate.
For example, if several center pivots are supplied from a single pumping station, the
desired pumping rate depends upon which pivots are to be operated. When a center
pivot corner arm extends, additional sprinkler heads are turned on, necessitating a
higher pumping rate or a slower angular speed. Different zones of a solid set sprinkler
or microirrigation system may require different pumping rates. Over the longer term,
declining water tables during the season or increased friction caused by deteriorating
pipes may decrease pumping rates.
There are several methods by which pumping rates can be matched to variable flow
requirements. Where water is supplied from a multiple pump station, it may be feasi-
ble to change the number of pumps operating to match the desired flow rate and pres-
sure. Less desirable options to reduce pumping rate and pressure include valving to
reduce irrigation system pressure (which increases pump discharge pressure), partial
recirculation of pumped water, and applying water when not needed. Each of these
alternatives results in a steeper system head curve than necessary and may result in
little or no reduction of power requirements, and thus energy cost.
A much more desirable alternative is to change the speed of the pump. The effects
of pump speed on pumping rate, pressure developed, and power required are described
in Section 12.2.3. With units powered by combustion engines, a speed change is read-
ily accomplished simply by changing the throttle setting. However, electrically pow-
ered units operate at a nearly constant speed, dependent primarily on the frequency of
the alternating current supplied. Thus, changing speed of electrically powered pumps
is more complex than for combustion-powered units.
Adjustable speed or variable frequency drives (VFD) allow adjustment of the fre-
quency of the electrical supply to the pump motor, thus allowing the speed of the
pump to be adjusted to match water flow requirements. Such VFD have a distinct ad-
vantage over other methods for adjusting the pumping rate of electrically powered
units because reducing the pumping rate (by reducing speed) also reduces power re-
quirements. From the affinity equations (12.3 through 12.5), a 20% reduction in pump
speed (and pumping rate) reduces the power consumption by 40% to 50%, depending
on the change in pump efficiency that accompanies the speed change. Thus, a pump-
ing plant can be designed to deliver the maximum pumping rate and the speed can be
reduced to obtain significantly reduced rates. VFD pumps make it easier to implement
daily management decisions, may reduce the equipment requirements if a single VFD
pump can replace several fixed capacity pumps operating in parallel, and have the
potential to significantly reduce energy consumption.
Variable frequency drives are not without disadvantages, however. In order to
change the frequency of the power, the AC current must be rectified to DC, then an
AC current of the desired frequency must be produced with an inverter. The solid state
electronic devices necessary to process the very high currents needed for a large pump
are quite expensive to design, purchase, and install (Poole, 1993). In rural areas, it may
be difficult to find qualified service personnel since these devices are most widely
used in industry. The square wave power produced (rather than the sine wave of con-
ventional line power) may cause problems with pump motors. The solid state compo-
nents of the VFD are especially vulnerable to overheating in the severe conditions to
which irrigation pumps are often exposed.
The ideal irrigation application for a VFD is a system in which a single pump
serves a large area planted to varied crops such that pumping rate requirements vary
widely (Nesbitt, 1995). Much of the operation should be expected to be at a reduced
pumping rate, and the irrigation season should be long. Systems serving a small area, a
single sprinkler system served by a single pump, monoculture cropping systems, or
low annual operating hours (typical for irrigation) are poor candidates for VFD.
The irrigator should expect little savings in electrical demand charges from install-
ing a VFD, although the VFD does provide a “soft start,” or reduced current inrush
upon starting the motor, as compared with conventional single- and multi-phase mo-
tors. One should also remember that pump efficiency is not a constant as speed is
changed, so careful design and pump selection are necessary.
VFD provides a potential energy savings of 20% to 50% under favorable condi-
tions, if annual operating hours are high (an industrial plant operating around the clock
will run for more than 8700 hours per year, compared to 1500 to 2000 hours typical
for irrigation systems in temperate areas). Hanson et al. (1996) found that VFD re-
duced power requirements of five pumping plants tested by 32% to 56% during the
times reduced pumping rates were needed. The pump load characteristics must be
variable, and the pump must have a moderate to high nameplate power rating, particu-
larly if it operates at more than about 75% of full load most of the time. A VFD may
be particularly attractive for installations where the initial pump installation is over-
designed to allow for possible future expansion of the irrigation system.
Before deciding on a VFD, the irrigator must compute energy use based on flow
rate, head, efficiency, and hours of operation at each level of flow and compare the
power savings of a constant speed system with the capital cost of a VFD. Cost of
equipment, engineering, and installation for a VFD may range from $2,500 for a
7.5 kW unit to more than $16,000 for 110 kW (Hanson et al., 1996). The Electric
Power Research Institute (Poole et al., 1989) has developed a computer program
(ASCON-I) to evaluate the economics of VFD.
12.4 POWER UNITS
Pumping plant power units convert electric energy or fuels into mechanical energy.
Electricity is widely used as a power source for pumping, as are natural gas, diesel,
liquified petroleum (l-p) gas, and gasoline. Irrigation and drainage pumps may be per-
manently located at a site, permitting engines or motors to be permanently connected
to a power supply. Portable pumps are generally driven either by an engine mounted
on the pumping unit or by the power take-off of a farm tractor, and utilize a temporary
fuel supply.
The cost of energy in the U.S. for pumping prior to the mid-1970s was, in many ar-
eas, minimal. However, regardless of the source, energy costs have risen dramatically
since that time, and have become a major agricultural production cost. Thus, it is im-
perative that the entire pumping plant be efficient to keep pumping costs as low as
possible. This requires careful selection of the pump and power plant for the specific job.
In Sections 12.2.2 and 12.3.4, I discussed the power required to drive a pump.
When using Equation 12.2 to compute the required output of a power unit, it is impor-
tant that the TDH value used include all friction head losses. Having determined the
pump input power required, the designer is ready to select a specific electric motor or
combustion engine. Selecting too small a unit will overload the power unit, which may
result in inadequate pump speed, reduced power unit efficiency, and reduced motor or
engine life from overheating and excessive bearing wear. Oversizing the power unit
results in higher than necessary investment costs and may result in lower efficiencies
and higher power costs.
It is generally not necessary to compute the power consumed by bearing losses for
centrifugal and short-set turbine pumps. When turbine pumps are set more than 20 m,
however, bearing losses can be significant and must be estimated using manufacturers’
curves or nomographs. Both lineshaft bearing and motor or gearhead thrust bearing
losses must be added to the brake power computed with Equation 12.2.
12.4.1 Electric Motors
Prior to 1945, most electrically driven irrigation and drainage pumps were belt
driven. Most modern installations couple the motor directly to the pump drive or line
shaft. The impeller of many centrifugal pumps is mounted directly to the extended
rotor shaft of the motor. Vertical turbine pumps commonly use hollowshaft electric
motors with a tubular rotor shaft through which the pump lineshaft or headshaft
passes. The direct coupled motor and pump is much safer than belt drives because
exposed rotating parts are minimized. They are also more efficient, because belt slip-
page and power to flex the belts are eliminated. The major advantage of belt driven
pumps was that pump speeds could be easily changed by changing pulleys.
12.4.1.1 Motor selection. The size of electric motor must be large enough to de-
liver the required energy to the pump. In years past, it was common practice to allow
overloading electric motors as much as 10% to 15% above their nameplate rating.
Most motors will operate under such excess load, but overloading results in higher
operating temperatures. Older motors typically had cast iron frames, copper windings,
were rated at service factors of 1.15 or greater, and performed satisfactorily with mod-
erate overloads. During the late 1970s, the electric motor industry began using lighter
motor frames, aluminum windings, different insulation material, and typically changed
the service factor to 1.0 or 1.05. This lower service factor indicates that the motor will
no longer tolerate the 10% to 15% overload. In general, motors available on the mar-
ket today should not be operated continuously above their rated load.
Because air density is dependent on altitude, motors operating at high altitudes may
not have sufficient air passing through them to prevent overheating. When motors are
used at altitudes over 1100 m MSL, most suppliers recommend derating the load ca-
pacity to prevent overheating. It may also be necessary to derate motors when operat-
ing at high temperatures such as in agricultural fields during the summer. Consult the
motor manufacturer’s literature for specific recommendations on how to compute the
lower rated power when the motor is to be used at high altitudes or where air tempera-
tures exceed 37°C.
Whether elevated temperatures arise from overloading, lack of ventilation, or high
ambient temperature, operation for significant periods of time at temperatures exceeding
the design operating temperature will reduce motor life significantly. A common rule of
thumb is that each 10°C above rated operating temperature will reduce motor life by 50%.
Thus, motor temperature should be carefully controlled to obtain a long service life.
12.4.1.2 Motor efficiency. Manufacturers provide data on the operating efficiency
of their product as a function of the load factor. Figure 12.16a is a typical motor effi-
ciency curve which shows that the efficiency is zero at no load but rises steeply so that
at 50% of rated load the efficiency is over 80%. When the load increases to 75% of the
rated load, the efficiency is near its maximum and will stay at that level until it begins
to drop slowly when the motor is loaded to 125% to 150% of its rated load. Thus, the
efficiency is nearly constant over a fairly wide range of load.
Smaller electric motors have lower operating efficiencies. The maximum efficiency
of 7.5 kW or smaller motors is usually below 88%. The range in peak efficiencies for
75 kW or larger motors is from 90% to 92%.
Unlike combustion engines, electric motors have few internal components to wear
and reduce operating efficiencies. When motor bearings wear, heat is generated and
the bearings generally seize, causing motor failure. It is safe to assume that if a motor
runs at all, it operates at or near its peak efficiency if the load is between 75% and
125% of rated load.
100 (a) (b)
Fuel Consumption, kg/kWh
Efficiency, percent
80
18
00
15
RP
00
M
60
12
RP
00
.78
M
RP
40
M
.72
20 .66
0 50 100 150 200 80 100 120

Percent of Rated Load Output Power, kW
Figure 12.16. Relation between load on power plant and (a) efficiency of electric
motor or (b) fuel consumption per unit energy output of combustion engine.
12.4.1.3 Electrical service to motors. Most electrical utilities require that motors
larger than 7.5 to 18 kW (10 to 25 horsepower) be connected to three-phase service.
Single phase motors draw six to ten times the run current when starting, as well as
higher running current per conductor than three-phase motors. The lack of three-phase
service or the prohibitive cost of adding the service may make electrical power infea-
sible for pumping in some areas. A new design of single-phase motor, however, has
the potential to significantly relax restrictions on use of larger single-phase motors.
The Written Phase motor runs 10% to 25% more efficiently than a conventional single
phase, and because it only draws twice the operating current when starting, allows
three to four times larger motors to operate on single phase lines. These motors are
commercially available up to 45 kW, with larger sizes being designed.
Proper selection of electrical conductors depends on the current requirement, volt-
age, location of service conductors, and the proximity of conductors to one another.
Wire insulation is specifically designated for the type service, with types designed for
direct burial in soil, for use specifically in conduit, and many special purpose types.
The current capacity of aluminum conductors is only about 60% that of the same size
copper conductor because aluminum has greater electrical resistance. If aluminum
conductors are used, all connections must be certified for use with aluminum to avoid
corrosion and thermal expansion problems which can lead to poor connections.
Most electrical suppliers recommend allowing a line voltage drop after the meter of
5% or less. Many pump loads are sufficiently large that voltage drop cannot be accu-
rately calculated simply using Ohm’s Law for direct current resistance. Particularly as
conductor sizes increase, spacing between conductors increases, and power factors
decrease, reactive losses can significantly exceed DC resistance losses. Several meth-
ods can be used to estimate reactive losses, including the Mershon diagram and drop
factor tables (Croft and Summers, 1992).
12.4.1.4 Overall efficiency. Overall pumping plant efficiency, Epplant, for electri-
cally powered plants is a product of the pump efficiency, Epump, and motor efficiency,
Emotor:
Epplant = EpumpEmotor = Q × TDH/C × Pm (12.14)
where the other terms are described in Equations 12.1 and 12.2, except that Pm is the
input power to the motor measured in kilowatts. This efficiency is commonly called
wire-to-water efficiency. The maximum theoretical value of this overall efficiency
ranges from 72% to 77%. For example, a pump with a maximum efficiency of 84%
coupled with a 90% efficient motor has a maximum overall efficiency of 75.6%.
Farmers are being advised that pumping plants should attain at least 65% effi-
ciency, and every new pumping plant should be tested to determine efficiency. Some
advisors suggest that if new pumping plants do not meet this level in field tests, then
payment to pump installers should be withheld until the installation is corrected. Such
conditions should be spelled out in a purchase contract.
12.4.1.5 Motor operating problems. Problems develop when motors are subjected
to unbalanced voltages, high voltages, or low voltages. Any of these problems will
result in motor overheating, reduced motor life, and in severe cases the motor will not
operate. Generally, the electric company is responsible for correcting these problems.
An exception is when wires connecting the power company’s main lines to the motor
are too small or become damaged, resulting in excessive voltage drop and low voltage
at the motor. Submersible turbine pumps are particularly susceptible to unbalanced
voltages or voltage surges. The major reasons for motor failure include overheating,
lightning strikes, destruction of insulation by rodents, and lack of proper lubrication.
12.4.1.6 Motor options. A number of different types of motor enclosures are avail-
able including open frame, weather protected, dust proof, and explosion proof. In ad-
dition, low-, medium-, and high-thrust motors are available, as are those with anti-
reverse ratchets and other less important options. Generally, irrigation and drainage
applications use weather-protected motors of medium to high thrust. Most motors used
on turbine pumps also have an anti-reverse ratchet to prevent the lineshaft from turn-
ing backward when the motor stops.
Motors which are not weather protected require construction of a pumphouse to
prevent rain or other moisture from entering the motor causing a short circuit. A pump
house must provide sufficient ventilation for motor cooling. At minimum, sun shields
are recommended to prevent sunlight from directly striking the motor and control pan-
els; this also provides some protection from rain or water from overhead sprinklers.
Some large motors are available with a water cooling option. A small part of the
pumped water is circulated through the motor casting, thus dissipating motor heat. Either
the motor or control panel should have a thermal breaker, which stops the motor if it be-
comes overheated or draws current in excess of the nameplate rating for extended periods.
12.4.2 Internal Combustion Engines
Internal combustion engines, and even steam engines, were used to power pumps
as early as the 1920s. The annual Irrigation Survey, published in the Irrigation Journal
(2001), documents the types of power used in the U.S. to pump irrigation water. In
some states, more than 90% of the pumps are internal combustion engine driven. In
others, such as California, Oregon, and Washington where electricity has been readily
available, the pumps are almost all electrically driven.
Natural gas, where available, was long a cheap source of power for pumping. Large
numbers of pumps in Texas, Oklahoma, Kansas, Nebraska, Colorado, New Mexico,
and Arizona relied on this source of power for a half century. Deregulation in the
1980s resulted in large increases in natural gas cost and reduced availability.
Diesel, l-p gas, and, to a lesser extent, gasoline have been used as fuel sources in
remote areas. Price structure has also made diesel competitive with electricity in some
areas. Because they are convenient to transport, these liquid fuels are used to power
most portable pumps.
Direct-coupled, engine-powered centrifugal pumps have the pump impeller
mounted on the engine shaft or a shaft coupled to the engine by a clutch. Most turbine
pumps powered by engines use a right-angle gearhead and drive shaft to transmit
power from the engine to the pump line shaft.
12.4.2.1 Engine selection. Many sizes, makes, and models of engines are available.
Each has a set of operating characteristics which describe how it will perform over a
wide range of conditions, and each has its most efficient zone of operation. Figure
12.16b illustrates how fuel consumption and output power are related to engine speed.
Note that the engine is most efficient at a particular speed and power output. The en-
gine in Figure 12.16, when operating at 1500 RPM, will operate most efficiently produc-
ing about 100 kW with a diesel fuel consumption of about 0.60 kg/kWh (0.36 lb/BP-h).
Engine speed is controlled by the throttle setting and the load, and may be adjusted
automatically by a governor. A detailed description of the process for matching an
internal combustion engine and gearhead to a turbine pump is beyond the scope of this
chapter. The necessary steps include (1) selecting the pump that will operate effi-
ciently while producing the desired pumping rate against the system’s total dynamic
head; (2) computing the required brake power to drive the pump using Equation 12.2
and then adding bearing losses; (3) consulting engine manufacturers’ catalogs to select
an engine that will produce the desired driveshaft power while operating efficiently;
(4) determining the pump speed and the speed of the selected engine at the most effi-
cient point of operation; then (5) selecting a gearhead with gear ratios to couple the
two units together with each turning at the proper speed.
Both lightweight and heavy duty engines are available. Engines may also be
classed as low, medium, and high compression. The performance, cost, efficiency, and
expected life of these engines vary significantly. Lightweight industrial or automotive
engines are usually least expensive, have a short lifetime, are low compression, and
thus have low efficiencies. Heavy duty engines may initially cost three to five times as
much as lightweight engines, but they have high compression ratios which permit
more efficient burning of the fuel. This is particularly important when natural gas or l-
p gas are used as fuels, because these fuels have a very high octane rating and can util-
ize much higher compression ratios than automotive type engines. Heavy duty engines
generally run much slower, and thus have a longer life. Heavy duty engines are de-
signed to be overhauled repeatedly, but it is often economical to scrap a lightweight
engine rather than overhaul it because components wear to the point that repair is
more costly than buying a new lightweight engine.
Manufacturers’ engine specifications are generally for operating conditions at sea
level and with air temperature below 30°C. Engine performance must be derated for
both altitude and temperature where field applications exceed those values. Engine
derating is much more important than derating electric motors, because the density of
air decreases rapidly with altitude and temperature, and engine output is directly pro-
portional to the mass of oxygen in the air combined with fuel. To increase air density
at high elevations and temperatures, turbochargers are often used. The manufacturer
should be consulted for specific information on how each engine will perform.
Care must be taken to assure that continuous operating curves are used for engine
selection. Irrigation pumps may operate for extended periods, which requires derating
curves which describe intermittent output, which are commonly available for light
service engines. Continuous rated power output may be 15% to 20% below the rated
intermittent output.
12.4.2.2 Engine efficiency. Engine efficiency is the ratio of the output power to the
input power. The ordinate of Figure 12.16b shows the amount of fuel consumption
expected as a function of power delivered. Thus, the low point on the horseshoe curve
is the point of best efficiency at each speed. This figure illustrates that even when the
engine is new there can be a wide range in efficiencies depending on the operating
point. Ignition timing and fuel injection adjustments also significantly affect effi-
ciency. Fuel consumption typically increases, driveshaft power decreases, and effi-
ciency drops as engine components wear. Continued maintenance of engines is essen-
tial to maintain high efficiency. This includes changing spark plugs and air filters and
regular oil and oil filter changes as well as cleaning fuel injectors.
Engine efficiency is determined in the field by installing a torque meter in the
driveshaft and measuring engine speed to determine power transmitted to the pump,
while also measuring fuel consumption. Such measurements are needed to determine
Table 12.3. Representative energy content of fuels.

Fuel Approximate Energy Content
Diesel 39,000 kJ/L (140,000 Btu/gal)
L-p gas 26,300 kJ/L (94,500 Btu/gal)
Gasoline 34,600 kJ/L (124,000 Btu/gal)
Natural gas 36,300 kJ/m3 (975,000 Btu/1000 ft3)
Electricity 3600 kJ/kWh (3410 Btu/kWh)
when equipment changes or engine overhauls should be undertaken. Although overall

pumping plant efficiency can be determined without the torque meter, use of the
torque meter permits separating pump and engine efficiencies.
Pumping plant efficiency is a measure of the overall performance of an engine
powered pumping plant. Pumping plant efficiency is the product of efficiencies of the
individual components, which for a turbine pump is:
Epplant = Epump × Egearhead × Edriveshaft × Eengine (12.15)
and for a direct coupled centrifugal pump,
Epplant = Epump × Eengine (12.16)
The pumping plant efficiency may also be evaluated by dividing the energy trans-
ferred to the water by the input energy (IE) to the engine,
Epplant = C ×Q ×TDH/IE (12.17)
where C is a conversion coefficient, 35.30 when energy consumed by the engine, IE,
is in kJ/h (33.46 when Q and TDH are in foot-pound units and IE is in Btu/h). For
electrically powered pumps, C is 0.00981 for Q in L/s, TDH in m, and IE in kW (1.89
× 10-5 for Q in gal/min, TDH in ft).
Liquid fuel consumption for diesel, butane, propane, and gasoline is measured in li-
ters per hour (gallons per hour). Natural gas consumption is measured in cubic meters
per hour (thousands of cubic feet per hour) at a standard temperature and pressure. The
energy content of fuel varies from dealer to dealer and with time. Representative val-
ues are given in Table 12.3.
12.4.3 Wind, Solar, and Other Sources of Power
In some countries, both animals and humans are used to power a variety of irriga-
tion and drainage pumps. The importance of these types of pumps is recognized, but
detailed descriptions of their use are beyond the scope of this chapter.
Wind has been widely used as a direct power source to provide both domestic and
livestock water, and in some places for irrigation and drainage. Wind power is gaining
favor for electrical generation, and is becoming economically competitive with other
methods. Most wind generation systems in the U.S. provide power to the electrical
grid for general distribution.
Studies by the U.S. Department of Agriculture near Amarillo, Texas, demonstrated
that wind can provide a significant portion of the energy required to irrigate crops in
that area, but a back-up power source is needed. This system uses a clutching mecha-
nism to mechanically transfer wind energy to a vertical turbine pump when wind
speeds are adequate. When wind energy is lacking, the clutching mechanism will shift
the power source to an electric motor or combustion engine. During some periods of
operation, energy is derived from both wind and the standby power supply.
The sun’s energy has been harnessed to power irrigation pumps in several experi-
mental installations in the U.S. Both photovoltaic cell arrays to charge batteries and
parabolic reflector arrays to concentrate the sun’s energy have been studied. Standby
power is still needed, but technical feasibility has been demonstrated. The costs asso-
ciated with utilization of solar energy remain considerably greater than for other
sources, except for small remote installations.
12.5 PUMP CONTROLS
The subject of irrigation controls is covered in considerable detail in a companion
publication by ASAE (Duke et al., 1990) on management of irrigation systems. Dis-
cussion here is confined to generalized controls for larger pumping plants, particularly
those aspects that relate to protection of equipment and safety of personnel.
12.5.1 Safety Controls
Because pumps often run unattended, combustion engines require automated pro-
tection against conditions that are most likely to damage the engine. Control units are
available which either ground the ignition system (for spark ignition engines) or cut
off the fuel supply (for compression ignition engines). Sensors connected to this con-
trol unit may include those to assure safe levels of coolant temperature, exhaust tempera-
ture, lubricant pressure, manifold pressure, engine speed, fluid levels, and vibration.
Additional sensors may protect other components, such as pump lubricant sensors.
Combustion engines provide many opportunities for personnel injury from contact
with rotating parts. It is particularly important that drive shafts be shielded to prevent
entanglement of clothing or hair, as such accidents are frequently fatal. Other areas
needing shielding are cooling fans, accessory drive belts, and exhaust systems.
Electric motors also require protection from damage. First, with all but the smallest
installations, the equipment used to start and stop the electric motor must be power
rated to equal or exceed the power rating of the motor(s) to be controlled. Motor cur-
rent must be started and stopped with magnetically controlled contacts which make or
break the circuit. In addition, motor controls must provide automatic emergency inter-
ruption of electrical service in case of short circuit or other severe overload. This
emergency interruption is provided with fuses. Fuses must be sized to pass the starting
current required for the motors served, but to fail if the current significantly exceeds
that required for normal operation over an extended period. Because starting current is
considerably greater than running current, fuses of size adequate to allow normal start-
ing provide no protection against motor overload. However, because an overloaded
motor draws higher current than normal, heater elements which activate thermal over-
load devices can provide protection against high current for longer than required to
start the motor. Dual element fuses sized to 125% of motor current rating will provide
short circuit protection and will also provide motor overload protection.
Even more important than protection of motors is protection of personnel. Because
the local environment of pumping plants is often wet, ground path distances may be
large, and voltages of 240, 480, or even greater are common, the potential for injury or
fatal accident is significant. A ground conductor must be provided from the meter
point to the point of service. Because distances are frequently large, ground path resis-
tance, even with a good ground rod, can be several ohms (enough to allow lethal cur-
rent to pass through the operator). This ground conductor must not be fused, because a
blown fuse here can leave the entire pump and control system at line voltage.
Ground-fault circuit devices, which detect current leakage of a few milliamperes,

are seeing use on certain irrigation equipment. However, the high currents often in-
volved with electric powered pumps and other irrigation equipment sometimes induce
sufficient capacitive coupling and other forms of leakage to cause nuisance tripping of
the devices.
Controls should be located so that the operator does not have to climb or kneel to
operate them. Disconnects must be between 1 and 2 m from the ground or operator
platform. Belt-driven pumps, common with older units, must have belt guards in place
to prevent personnel contact with moving belts and pulleys.
12.5.2 Automatic and Integrated System Controls
Electrically powered pumping plants are particularly adapted to automatic controls
which can start or stop pumping plants. These controls present a particular hazard to
service personnel who may not be aware of their existence or intended times of opera-
tion. Thus, it is especially important that provisions be made for temporary lockout
devices to prevent power from being applied when servicing of these units is necessary.
Automatic start devices allow the pump to be started by some other device. They
are commonly used with self-propelled sprinklers to allow both sprinkler and pump to
be controlled by the sprinkler panel. Auto-restart devices are sometimes installed to
restore pump operation without intervention after a power outage.
Time controls are very common on solid set sprinkler irrigation systems, such as
those used for parks and golf courses. These controls allow preselection of irrigation
timing and sequence, and may include sensors to initiate irrigation when soil water
becomes depleted, or to cease irrigation if significant precipitation occurs.
Time-of-day controls are sometimes used on large irrigation pumps to aid the elec-
trical supplier in spreading the electrical demand more evenly over time. Typical uses
are to encourage pumping during off-peak times (perhaps at night after industrial
plants have closed) by switching from one meter billed at a low rate to a second meter
billed at a higher rate during peak demand periods. More sophisticated peak load con-
trol systems utilize coded signals sent by radio or over transmission lines to address-
able receivers at individual pumps, which can interrupt power to the pump whenever
the power supplier needs to reduce demand.
Most manufacturers of self-propelled irrigation systems provide computerized con-
trols which allow very sophisticated monitoring and control of the irrigation systems
and the pumps that serve them. These control systems utilize radio, telephone, cellular
phone, and even satellite data transmission to control the pump and irrigation systems.
They are capable of remote operation from virtually anywhere in the world, so sys-
tems equipped with this type control are especially subject to unexpected operation.
12.6 ECONOMIC CONSIDERATIONS
The decision to install an irrigation or drainage system should include an economic
analysis. The returns from increased crop sales, value of landscape, or drainage bene-
fits drainage must be large enough to repay both investment and operating costs.
Pumping cost includes both fixed and variable components. Fixed costs do not de-
pend on the annual operating hours, and include such items as initial investment,
taxes, insurance, and annual connect charges for electric or natural gas service. Vari-
able or operating costs include those items which depend on the amount of pumping
each year. Variable costs are dominated by energy cost, but include lubricants, labor,
repair parts, and maintenance costs.
When energy costs were low, annual investment costs were typically larger than
variable costs in the U.S. Although pumping equipment costs have increased signifi-
cantly over the past two decades, power cost has increased several fold. Because oper-
ating costs typically exceed amortized investment costs, more expensive pumping
equipment may have significantly lower overall costs due to higher operating efficien-
cies. Power costs for electricity can be very complicated and typically depend on the
amount of energy used, time of use, and the connected load.
12.6.1 Investment Costs
The total investment cost can be obtained by simply adding costs of the individual
components, but annualized investment cost is more informative. Annualized invest-
ment costs can be computed by several methods. A simple approach is to determine
the average annual cost by dividing the cost of a particular item by the expected life.
However, this method disregards the impact of interest on the investment.
Computing the amortized cost, considering the expected life of the item and the in-
terest rate, is a more appropriate method. The capital recovery factor, CRF, can be
used to compute the annual amortized cost. CRF is calculated by
I × ( I + 1) n
CRF = (12.18)
( I + 1) n − 1
where I is the current annual interest rate (fraction), and n is the anticipated life of the
equipment (years). Multiplying the CRF by the initial cost results in the annual amor-
tized cost.
The investment cost should be calculated on the actual installed cost of all durable
equipment, which may include the well, pump, gearhead, motor or engine, drive shaft,
and fuel tank, gas pipeline, or electric power lines. The total annualized cost of a
pumping plant is the sum of the annualized costs of each component.
The expected life of various components will vary depending on the amount of use,
protection from weather provided, frequency of maintenance, and the quality of the
original components. If the manufacturer does not suggest an expected life, then it
must be estimated. The expected life for engines, gearheads, and pumps may be ex-
pressed as hours of operation. For example, automotive engines may have an expected
life as short as 3000 hours while some heavy duty engines are expected to operate for
15,000 to 20,000 hours with periodic overhaul. Table 12.4 shows life expectancy for
various pumping plant components. Quality of the water being pumped can signifi-
Table 12.4 Estimated useful life of pumping plant components (years).

Annual Hours of Operation
Component 500 1000 2000 3000
Well 25 25 25 25
Pump 15 15 15 10
Gearhead 15 15 15 10
Drive shaft 15 15 7 5
Engine (heavy duty) 15 15 10 7
Engine (automotive) 5 3 2 1
Gas pipeline 25 25 25 25
Engine foundation 25 25 25 25
Electric motor 25 25 25 25
Electric controls and wiring 25 25 25 25
cantly impact pump and well life by corrosion of ferrous components or deposition on
well screens and pump impellers, even when unused.
Annualized cost may also be calculated using a graduated depreciation schedule.
This method is often used to accelerate credit of investment costs for tax purposes.
12.6.2 Other Fixed Costs
Taxes, insurance, and power demand charges are paid annually and are independent
of pump operation. The taxes are generally assessed on the value of the equipment, but
may also be based on the volume of water pumped, which is a variable cost.
Liability insurance and insurance against fire, vandalism, and acts of nature (e.g.,
lightning), protect the owner. Both electric and gas utilities may charge for the privilege
of connecting. These may be billed as a one-time hookup fee (which should be annual-
ized), divided over a minimal recovery period, or as annual connect or demand charges.
12.6.3 Variable Costs
Power costs usually dominate annual operating or variable costs. Annual power
costs can be determined by estimating the energy consumption per hour of operation
(kW for electricity, m3/h for natural gas, or L/h for liquid fuels). Energy demand can
be estimated from one of Equations 12.15 through 12.17. This method requires that
the pumping rate, total dynamic head, and estimated pumping plant efficiency be de-
termined. Typical ranges of pumping plant efficiency are given in Table 12.5. Multi-
plying the hourly consumption by estimated annual hours of operation and by unit cost
of energy gives the annual cost. Natural gas and electric computations may require
application of a complex rate structure rather than a single unit price.
Other variable costs, such as lubrication and labor, can be estimated as a cost per hour
of operation. Only the labor required to service the pumping plant is included. Deter-
mine the manufacturer’s recommended lubrication and maintenance interval, then
multiply by the per unit cost and the number of maintenance periods annually. Costs
for minor repairs and overhauls should be estimated as a function of hours of operation.
Engine overhaul costs are generally prorated over the years between the overhauls.
Table 12.5. Representative pumping plant efficiencies.
Power Source Maximum Acceptable Average Observed
Electric 72–77 65 45–55
Diesel 20–25 18 13–15
Natural gas 18–24 15–18 9–13
L-P gas 18–24 15–18 9–13
Gasoline 18–23 14–16 9–12
12.6.4 Total Pumping Costs

Total pumping costs can be expressed either as annual cost or as total life cycle
cost. The annual cost is the sum of the annualized investment cost, other annual fixed
costs, and variable costs for a year. Life cycle costs are the sum of the initial invest-
ment costs and the other fixed and operating costs for each year, including interest on
the money invested. Whatever the method of estimating pumping costs, it is important
that pumping plant selection be made on the basis of total cost, not solely on initial
purchase price.
12.6.5 Minimizing Pumping Costs
The cost of pumping is dependent not only on the input power requirement of the
pumping plant, but also on the total amount of water pumped, and the costs associated
with supplying energy for pumping. Input power requirements have been shown to
depend on the efficiency of the power plant, drive mechanism, and pump, and on the
pumping lift, pumping rate, friction losses in the delivery system, and operating pres-
sure head. Therefore, optimizing these components will minimize pumping costs.
Pumping lift depends on the static water level from which the pump operates, and
for wells, on drawdown within the well. Although these factors are often beyond the
farmer’s control, proper development of a well and appropriate well screen when in-
stalled can reduce the resistance to water flow into the well bore, reducing drawdown.
Periodic measurement of static lift will determine whether changes are occurring in
regional water levels, which may eventually require modification to the pump. Meas-
uring pumping lift will help detect changes, such as incrustation of well screens or
clogging by microorganisms that can be remediated by chemical or mechanical treat-
ment to reduce drawdown.
Pumping rate can be reduced in some circumstances. A reduced rate will allow a
smaller drive motor to be used, which may reduce energy demand or hookup charges.
Reducing pumping rate may require pumping for a longer period, which may allow
the irrigator to take advantage of the reduced power rates characteristic of many rate
schedules which are based on energy consumption per unit of power connected. Be-
fore committing to pumping plant changes to reduce pumping rates, the farmer must
realize that such changes may increase the risk of crop damage from inadequate irriga-
tion or insufficient drainage. Reduced pumping rates also make the reliability of the
pumping plant a more critical factor, as less time is available for preventive mainte-
nance or repairs before crop stress begins.
Friction losses can be minimized by initial selection of larger diameter pump col-
umn pipes and delivery piping. Minimizing pipe fittings such as elbows, valves, and
abrupt changes in pipe diameter, will keep friction loss at a minimum. Use of smooth
materials, such as polyvinyl chloride (PVC) or epoxy-coated pipes can significantly
reduce friction loss. Periodic interior scaling and coating to restore interior dimension
and smooth the surface of aging pipe can be a cost-effective maintenance procedure,
particularly for large pipelines.
Total head is the remaining factor of power costs that can be controlled. Any valv-
ing of the discharge piping system simply reduces the pressure at the point of dis-
charge, but increases the head against which the pump operates. Therefore, reducing
pressure or flow rate by valving is not an economically viable alternative.
Sprinkler and microirrigation system manufacturers have developed many alterna-
tives in recent years to allow use of lower pressures. Low angle trajectory impact
sprinkler heads with unique orifice designs operate at lower pressures than conven-
tional impact heads. These heads reduce the wind distortion of sprinkler patterns and
minimize soil compaction problems associated with large water droplets. However,
the reduced diameter of the spray pattern increases the water application rate and may
result in water runoff and poor uniformity. Spray heads have largely replaced impact
heads on self-propelled sprinkler irrigation systems. The first of these spray heads had
very small pattern diameters, with high application rates and fine water droplets sub-
ject to considerable wind drift. New spray head designs increase the water pattern di-
ameter, produce moderate droplet sizes, and function at very low pressure.
With careful attention to soil characteristics, use of low pressure sprinkler heads or
microirrigation emitters can significantly reduce total head and possibly pumping
costs. Low pressure devices are most effective where elevation differences across the
irrigated area are small because the flow rate is more influenced by changes in eleva-
tion than for devices operating at higher pressure (see Chapter 16). Poor uniformity
resulting from low-pressure operation can increase the amount of water that must be
pumped to adequately irrigate a field. The cost of this increased pumping can easily
exceed savings realized by reducing pressure.
Operating head of irrigation systems requiring filtration can also be minimized by
careful maintenance of filters. Frequent cleaning of screens or back-flushing of filters
will reduce the head loss associated with moving water through the devices.
When changes are made that impact the total dynamic head of a pumping plant,
such as larger piping, low pressure sprinkler heads, or more efficient filtration sys-
tems, the pumping plant must be reevaluated to determine whether these changes war-
rant pump modifications.
One of the most effective ways to reduce pumping costs is to reduce the amount of
water pumped. Low water application uniformity requires pumping more water than
required by the crop to provide adequate water over all the field. Improvements to the
irrigation system, such as adaptation of surge irrigation or level basins for surface irriga-
tion or renozzling a center pivot to attain greater uniformity can significantly reduce the
necessary pumping (Chapters 14, 16, and 17). Avoiding excess irrigation by applying only
the amount needed at the time needed (Chapter 22) will directly reduce pumping costs.
Other management decisions may impact total energy charges, if not total energy
consumption. Where electrical demand charges are based on motor nameplate power,
use of undersized motors may reduce energy costs, but at the risk of shortened motor
life. Rate structures are often stepped, which tends to favor reduced pumping rates
(smaller motors) which in turn may require increased hours of pumping. Where elec-
trical demand charges vary with time, it may be advantageous to increase pumping
capacity to allow the pump to be off during times of high cost energy or to participate
in shutdown programs to control the supplier’s peak demand.
12.7 IMPACT OF PUMPING SYSTEM MODIFICATION
Any modification to the irrigation or pumping system, such as declining groundwa-
ter levels or converting from open ditch to gated pipe delivery, will change the system
head curve. In Section 12.3.3 I showed the relation between the system head curve and
the pump TDH-Q curve and how the two determine the operating head and pumping
rate. Thus, changes in the irrigation system require evaluating their impact on pump-
ing plant performance.
12.7.1 Changing Operating Conditions
Many factors can change the operating conditions of a pump. The system head
curve may be changed intentionally as irrigators seek new methods to contain costs.
Areas historically surface irrigated have been converted to sprinkler or trickle irriga-
tion to increase the efficiency of irrigation. Sprinkler heads may be replaced by mod-
els that operate at lower pressures to reduce total heads. Other changes to the system
head may be unintentional, such as deterioration of pipe walls, which may increase
friction and even reduce pipe internal diameters. Severe cases may result in rupture of
pipes resulting in loss of part of the water pumped or recirculation of water within a
well casing. Each of these conditions will change the total head of the pump, thus
shifting the operating point along the pump curve, with increased head generally re-
ducing pumping rates. Whatever the direction of change, the operator should be aware
of the potential impact of changes on the efficiency of the pumping plant. Any signifi-
cant change to an irrigation system must include evaluation of the impact of that
change on pumping costs.
Other changes may be beyond the direct control of the irrigator. Declining water
tables will likewise increase the system head curve. Sediment, whether from improper
well development, deterioration of well screens, or pumping of sediment-laden surface
waters, will accelerate wear of pump components, producing an effect similar to re-
ducing impeller diameter, thereby reducing both operating pressure and pumping rate.
If water levels decline, well screens become encrusted, or other conditions cause the
water level at the pump suction to decline, air may be introduced directly into the
pump, a condition referred to as breaking suction. This will reduce the pumping rate,
which may allow the water level to recover and the pump to begin operating again.
With centrifugal pumps having a positive suction lift, breaking suction will often
cause pumping to cease until the operator intervenes to evacuate the air from the suc-
tion side (i.e., prime the pump). Turbine and submersible pumps may recover from
this condition, but such surging of the flow can severely damage the pumping plant.
Declining water levels may also result in cavitation, which can quickly destroy a
pump. Similar conditions are created when water cascades into a well bore or sump.
Air bubbles may be entrained with the water, and these bubbles expand and collapse
rapidly as pressures change within the pump impeller.
12.7.2 Pump Modifications
Whenever pumping conditions change sufficiently to warrant modification to the
pumping plant, a number of options may exist. Pumps powered by combustion en-
gines can sometimes be adjusted to account for minor changes in operating conditions
by simply changing the engine throttle setting. In fact, deterioration in combustion
powered pumps often goes unnoticed until severe because the engine throttle can be
adjusted to give the desired pumping rate and pressure. In Section 12.2.3 I described
the relation between the pump head-pumping rate curve and pump speed. It is impor-
tant to keep in mind that changing pump speed not only changes the pumping rate and
discharge pressure, but it also changes the power requirement. As seen from Equation
12.5, the input power changes as the third power of pump speed. Thus, increasing
pump speed by only 10% will increase the power requirement by one third. Addition-
ally, significant changes in pump speed will undoubtedly change the efficiency of the
pump, and may result in significantly greater impact on power requirement than indi-
cated by Equation 12.5. The overload resulting from increasing pump speed may well
lead to severely shortened engine life, and the need to increase throttle setting should
be a signal to the irrigator to evaluate the pumping plant.
Those electrically powered pumps that utilize belt drives may sometimes be modi-
fied in a similar manner by changing pulley sizes to change pump speed, but the elec-
trical load must be carefully checked to assure against overheating the motor.
In general, significant changes to the pumping or irrigation systems will necessitate
some modification to the pump to assure peak performance. The amortized cost of
making such changes must be compared with the consequences of leaving the pump-
ing plant unmodified. Pump modification may be as minor as adjustment of the bowls
to compensate for wear, as discussed in the next section. However, if significant
changes have been made to the pumping or irrigation system, more extensive pump
modification may be warranted.
A pump test, described in the following section, will determine whether pump repairs
are needed. If the system head has been reduced, such as through conversion of a center
pivot from a high-pressure to lower-pressure sprinkler package, and the pump is in oth-
erwise good condition, it may be possible to modify the existing pump so that it will
work efficiently under the new conditions. Such modification requires evaluation of the
pump to determine the suitability of the existing bowl for the new application. If the
pump can be matched to the system head curve with reasonable efficiency by either re-
moving bowls, trimming impellers, or a combination, that may be an economical alter-
native. A decision to modify an existing pump for new conditions should always be ac-
companied by a comparison of repair or replacement costs with long-term power costs.
12.8 MAINTENANCE AND TESTING
12.8.1 Routine Maintenance
As with any mechanical equipment, a good pumping plant maintenance program is
necessary to maintain pump efficiency, minimize power costs, improve dependability,
reduce operating costs, and extend the life of the equipment. The major item of pump-
ing plant maintenance is proper lubrication of both the pump and drive unit. The line
shaft of a turbine pump must be lubricated, by water or oil as designed, prior to start-
ing. Pump head bearings and electric motor bearings must be greased or oiled as the
manufacturer recommends. Other maintenance items include periodic adjustment of
impeller clearance in turbine pumps and adjustment of packing glands or pump shaft
seals. Engine oil, coolant, and filters must be changed regularly. Regular ignition and
fuel system adjustments will help maintain engine efficiency. The motor or engine,
pump head, and gearhead should be cleaned periodically to help dissipate heat. Any
unusual noise or vibration is cause for immediate action.
Pumps must be drained after the irrigation season in cold regions or else be en-
closed in heated shelters if they must operate during periods when freezing is possible.
Protective screens must be installed on electric motors to prevent rodents from damag-
ing windings and these screens must be kept clean to permit air circulation.
12.8.2 Pump Adjustments
The primary adjustment to be made on centrifugal pumps is to periodically tighten
the packing gland that seals the input shaft. Although excessive leakage of water from
this seal may cause inconvenience, it should be remembered that water leakage pro-
vides the only seal lubrication for many pumps. In any case, the seal must never be
tightened so that the shaft is difficult to rotate by hand.
Vertical turbine pumps also have a line shaft seal at the point the shaft exits the
pump head. In addition, the vertical turbine pump allows for adjustment of the clear-
ance between impellers and bowls. For open and semi-open impeller types, adjustment
of this clearance is critical to attaining the maximum pump efficiency. The line shaft is
threaded on the top, with a key to assure that the line shaft rotates with the motor or
gearhead. The line shaft nut has a setscrew or similar locking device to hold the ad-
justment once it is set. The line shaft nut is backed off until the impeller just drags on
the bowl and the line shaft cannot be turned by hand. Then the nut is tightened to lift
the impeller from the bowl. The number of turns will depend upon the pump setting,
line shaft diameter, pump thrust characteristics, and total head, all of which influence
the amount of line shaft stretch when operating. Consult the manufacturer’s instruc-
tions for the appropriate adjustment. In any case, the line shaft nut should not be tight-
ened so far that the impeller drags on the top of the bowl.
Adjustment for installations with electric motors can be checked using an ammeter.
If the pump is operable prior to adjustment, use a clamp-on ammeter to determine the
current in each of the conductors. After adjustment, again measure the current in each
conductor. Although the current may change because pump performance is affected, a
large change in current should arouse suspicion that the impellers are dragging, and
the pump should be stopped immediately for professional service. Dragging impellers
may also cause unusual noise, vibration, or obvious overloading of an engine.
Some farmers with open or semi-open impeller pumps are inclined to increase im-
peller clearance to reduce the pumping rate and discharge head when the irrigation or
drainage system is modified. Although such adjustments may effectively change
pumping rate and discharge head, the efficiency of the pump is invariably affected. If
operation is to continue with increased impeller clearance, a pumping plant evaluation
should be conducted to determine whether the efficiency has been sufficiently changed
to warrant pump modification.
12.8.3 Evaluating Pumping Plant Performance
Field testing of both new and old pumping plants is recommended to evaluate per-
formance. New plants should be tested to confirm that the pump is operating as ex-
pected, and to rectify problems while the equipment is under warranty. The pump in-
staller should be responsible to insure that the pump performs as designed.
Every pumping plant should be performance tested after several years to develop a his-
tory of performance change. These tests will identify pumps with low efficiency or those
producing less flow or head than expected. Updated efficiency values should be used to
conduct a cost analysis to determine whether the pumping plant should be repaired.
A field test should determine the static lift, pumping rate, drawdown, discharge
head, and input power. Using a torque meter on engine-driven plants will allow sepa-
rate evaluation of engine and pump performance. As mentioned earlier, the efficiency
of electric motors can be assumed to remain constant so long as they run without ex-
cessive noise or vibration.
Equipment required to conduct a field test includes an accurate water flow meter, a
stopwatch, a device to measure the water level if pumping from a well, accurate pressure
gauges when the pump discharge is pressurized, a fuel flow meter and torque meter for
engine driven units, and current and voltage meters for electrical units. Because this
equipment is expensive and must be well-maintained and regularly calibrated, few farm-
ers can justify their own. In most areas, pumping plant tests are available for nominal
cost from consultants, pump installers, gas suppliers, or electrical utility companies.
12.8.4 Making Repair Decisions
Many repair decisions are not so apparent as those to replace broken parts that dis-
able the pumping plant. Gradual deterioration of the pumping plant may not be notice-
able until quite severe. At this point, the cost of pumping a unit of water may be sev-
eral times what it could be under optimum conditions. For this reason, the periodic
pumping plant test is a good economic investment. Following the pumping plant test,
one must make the same calculations of operational cost as were made to select the
proper components on installation. Comparing the present cost of operation with the
cost of a high-performance pumping plant, and the potential savings with the cost of
making necessary repairs, is the most realistic method of deciding whether to repair.
Acknowledgements
The author wishes to express appreciation to Mr. Robert A. Longenbaugh, Consult-
ant Water Engineer, retired from the Office of the Colorado State Engineer of Arvada,
Colorado. Mr. Longenbaugh was senior author of this chapter in the first edition, and
much of the current chapter is taken from that edition. Mr. Quintin E. Nesbitt, Agricul-
tural Engineer, Idaho Power Company, Boise, Idaho, provided materials from which
the information on variable speed pumps was prepared.
LIST OF SYMBOLS
BP brake input power to pump, kW
C coefficient to convert energy units,
CRF capitol recovery factor
D pump impeller diameter, m
Dt downtime as a decimal fraction of pumping period
E elevation above mean sea level, m
Ea application efficiency, decimal fraction
Edriveshaft driveshaft efficiency, decimal fraction
Eengine engine efficiency. decimal fraction
Egearhead gearhead efficiency, decimal fraction
Emotor motor efficiency, decimal fraction
Epplant overall pumping plant efficiency,
Epar efficiency of pumps in parallel,
Epump pump efficiency
Eser efficiency of pumps in series
hf friction head
Hp head imparted by pump
I interest rate
NPSH net positive suction head, m
NPSHA net positive suction head available, m
NPSHR net positive suction head required by pump, m
RPM pump speed, revolutions per minute
TDH total dynamic head, m
WP water power delivered by pump, kW
n anticipated life of equipment, years
Q pumping rate, L/s
Qg gross pumping capacity, ,L/s
Qn net pumping capacity, L/s
Pb barometric pressure, m
Hs pumping suction lift, m
Pv vapor pressure of pumped fluid, m
IE input energy to engine or motor, kJ/hr or kW
Pm input power to motor, kW
REFERENCES
Addison, H. 1966. Centrifugal and Other Rotodynamic Pumps. London, U.K.: Chap-
man and Hall.
Colt Industries. 1979. Hydraulic Handbook. Kansas City, Kans.: Colt Industries-
Fairbanks Morse Pump Division.
Croft, T., and W. I. Summers. 1992. American Electrician’s Handbook. 12th ed. New
York, N.Y.: McGraw-Hill, Inc.
DeWiest, R. J., and S. N. Davis. 1966. Hydrogeology. John Wiley and Sons, Inc.
Duke, H. R., L. E. Stetson, and N. C. Ciancaglini. 1990. Chapter 9: Irrigation system
controls. In Management of Farm Irrigation Systems, 265-312. G. J. Hoffman, T.
A. Howell, and K. H. Solomon, eds. St. Joseph, Mich.: ASAE.
Hanson, B., C. Weigand, and S. Orloff. 1996. Performance of electric irrigation pump-
ing plants using variable frequency drives. J. Irrig. Drain. Div. 122(IR3): 179-182.
Hicks, T. G. 1957. Pump Selection and Application. New York, N.Y.: McGraw-Hill.
Heermann, D. F., H. H. Shull, and R. H. Michelson. 1974. Center pivot design capaci-
ties in eastern Colorado. J. Irrig. Drain. Div. 100(IR2): 127-141.
Ingersoll-Rand. 1977. Cameron Hydraulic Data. Woodsliff Lake, N.J.: Ingersoll-Rand.
Irrigation Journal. 2001. Irrigation survey, 2000. Irrig. J. Jan/Feb.
Johnson Division UOP. 1975. Groundwater and Wells. St. Paul, Minn.: Johnson Div.. UOP.
Karassik, I. J., W. C. Krutzsch, W. H. Fraser, and J. P. Messina. 1976. Pump Hand-
book. New York, N.Y.: McGraw-Hill, Inc.
McWhorter, D. B., and D. K. Sunada. 1977. Groundwater Hydrology and Hydraulics.
Highlands Ranch, Colo.: Water Resources Publications
Nesbitt, Q. 1995. Variable frequency drives and irrigation pumps. In USDA-NRCS
Western States Irrigation Workshop. Washington, D.C.: USDA-NRCS.
Poole, J. N. 1993. Economics of adjustable speed drives. In Motor/Adjustable Speed
Drive Users Workshop, Northwest Power Quality Service Center, Electric Power
Research Institute. Palo Alto, Calif.: EPRI.
Poole, J. N., D. L. Seitzinger, and G. R. Stengl. 1989. Adjustable speed drives analysis
(ASCON-I) energy savings and return on investment. Raleigh, N.C.: CRS Sirrine,
Inc., User’s Manual for Electric Power Research Institute, Palo Alto, Calif.
Todd, D. K., and L. W. Mays. 2004. Groundwater Hydrology. John Wiley & Sons, Inc.
Zipparro, V. J., and H. Hasen. 1993. Davis’ Handbook of Applied Hydraulics. 4th ed.
New York, N.Y.: McGraw-Hill, Inc.
CHAPTER 13
HYDRAULICS OF
SURFACE SYSTEMS
Theodor S. Strelkoff (USDA-ARS,
Maricopa, Arizona)
Albert J. Clemmens (USDA-ARS,
Maricopa, Arizona)
Abstract. The goals of surface irrigation are defined and the physical processes in-
volved are described. Field variables such as slope, infiltration, and roughness;
physical design variables like furrow or border cross-section and length of run; and
management variables such as inflow and outflow control are discussed. The role of
modeling as a tool for subsequent design and management is emphasized, as is the
pivotal role of physical laws such as conservation of mass and momentum. Hydro-
logic, or lumped-parameter, modeling is described in detail as a relatively simple
technique sufficiently accurate for many purposes. The errors that do arise are ex-
plained as a consequence of some of the assumptions made to achieve the simplifica-
tion. The hydrodynamic, or distributed-parameter, approach, itself amenable to a se-
ries of simplifications, is described with sufficient detail to afford a basic understand-
ing of the assumptions and solution techniques. A user-friendly software package al-
lowing easy data entry and simulation graphical output is introduced. The chapter
closes with a discussion of techniques for estimating infiltration and roughness in the
field for entry into simulation and design software.
Keywords. Efficiency, Hydrodynamics, Modeling, Parameter estimation, Simula-
tion, Surface irrigation, Volume balance.
13.1 INTRODUCTION
13.1.1 History and Current Status of Surface Irrigation
Surface irrigation is accomplished by introducing water at some point in a field and
allowing it to flow over the soil surface to reach the receiving plants. It has been prac-
ticed in many areas of the world since ancient times. At least 4000 years ago, farmers
in Egypt, China, India, and countries of the Middle East irrigated agricultural lands.
As early as 200 B.C. the Hohokam farmers diverted water from what is now known as
the Salt River through 200 km of hand-dug canals in and around the site of modern
Phoenix, Arizona (Salt River Project, 1960). The long history, marked by successes and
failures, is valuable. Evidences of sustained agriculture in arid lands illustrate the contri-
bution that successful irrigation can make to society. Evidences of salted soils and aban-
doned systems emphasize the need to irrigate wisely or face similar consequences.
Shortly after the close of World War II, technology, energy, and industrial support
combined to promote a rapid rise in sprinkling; this was followed by micro- and LEPA
systems. Still, in spite of the popularity of pressurized systems, the USGS estimates
that of the roughly 25 million ha irrigated in the U.S., 48% is served by surface means
(Hutson et al., 2004). The percentage increases to well over 90% in other parts of the
world. It is in most cases the cheapest method of applying water to crops, and usually
requires the least expenditure of energy, but it is often viewed as inherently wasteful
of water. It is also considered more labor intensive and less convenient than pressur-
ized systems. Yet, under appropriate field conditions, proper design and management
of surface systems can lead to high application efficiencies, on a par with other meth-
ods. There is potential for automation and reducing labor requirements. Furthermore,
in many areas of the world, labor is relatively inexpensive, while availability of water
and capital are low.
13.1.2 Irrigation Scheduling vs. Uniformity of Application
A key aspect of efficient irrigation is application of water to the soil in the amounts
needed at the time needed, all over the irrigated field. Determination of the plant water
requirements is examined at length in Chapter 8. Still, once the requirements and tim-
ing are established, actually placing that amount at the plant roots is not a trivial mat-
ter and is the subject of hydraulic studies of the on-farm system that distributes water
over the cropped area. It is the variation of pressures in a pressurized system or water
surface elevations in a surface system, in response to and influencing the flows pre-
vailing at the various locations in the system, that determines the uniformity of distri-
bution of water to the individual plants. With a complete understanding of the system
hydraulics and the ability to influence it, rational decisions can be made balancing
system costs against the monetary and other costs of local under- and overirrigation.
13.1.3 The Role of Simulation in Design and Operation
Design of surface systems has traditionally been based on experience with success-
ful projects, intuition, empirical compilations, and not very much theory. Now, how-
ever, many of the analytical problems associated with the prediction of surface irriga-
tion flows have been solved, and it is now possible to develop designs and operating
recommendations based on simulations, i.e., solutions of the pertinent hydraulic equa-
tions with trial values of the design parameters in the search for an optimum.
These equations express basic laws of physics: conservation of mass and momen-
tum. In view of our limited ability to predict theoretically the interaction of the soil
and plants in the system with the hydraulics of the irrigation stream, some empirical
input is still necessary. However, by limiting the empirical aspects of the study, we
can hope for greater universality of the results than is possible with more purely em-
pirical attempts at predicting stream behavior or irrigation performance.
13.1.4 Scope of the Chapter
This chapter provides an overview of the behavior of irrigation streams as influ-
enced by soil and crop hydraulic parameters, physical design parameters, and irriga-
tion management parameters. These determine the outcome and merit of the irrigation.
The focus is on simulations of surface irrigation flow for use in development of design
and management recommendations.
Simulations are viewed from both a hydrologic and hydrodynamic point of view.
The hydrologic, or lumped, approach makes an assumption for the shape of the irriga-
tion-stream profile. One or another approximation of momentum conservation estab-
438 Chapter 13 Hydraulics of Surface Systems
lishes the depth at some point in the stream, typically the upstream end. The principle
of mass conservation then leads to ordinary differential equations (or an equivalent
convolution integral) in stream length as a function of time. The hydrodynamic, or
distributed, approach makes no shape assumptions utilizing, instead, the principles of
mass and momentum conservation to solve for the variation of depth and discharge
within the stream. This leads to partial differential equations in distance and time, or
their equivalent.
The equations governing the flow are presented, along with initial and boundary
conditions. Simulation software based on numerical solution of the equations is de-
scribed. The governing equations are also applied to an inverse problem: estimation of
field parameters from observed behavior of an irrigation stream. When needed, surface
irrigation hydraulics are presented in dimensionless form to allow a maximum amount
of information to be transmitted with the least computational effort and volume of
data; databases of simulations for design are typically calculated and stored dimen-
sionless.
The goal is the development of surface irrigation systems that can be operated in an
efficient, effective, sustainable, and environmentally sound way.
13.2 BASIC CONCEPTS OF
SURFACE IRRIGATION HYDRAULICS
13.2.1 Characteristics of Surface Irrigation
In surface irrigation, water is conveyed to the upper, or inlet, end of a horizontal or
sloping field and is released there to flow over the field by gravity. An obvious, over-
riding characteristic of surface irrigation is that the same flow channels which convey
water to their distal ends are also used all along the way to draw off water for plant
needs. Another clear characteristic is that the cross-section of the flow is not con-
strained by fixed boundaries such as a pipe wall; the depth of flow is free to seek its
own level. In a pipe, even if it is not full at the time irrigation starts, water moves
quickly to the end, and the time of operation of a lateral port at the distal end is much
the same as at the near end. But in surface irrigation, water typically travels much
more slowly, sometimes requiring hours to traverse a single furrow.
Irrigation streams, when confined by furrows, flow in specific, known directions.
The transverse dimensions of such streams are much smaller than their longitudinal
extents, and transverse variations in water-surface elevation and transverse compo-
nents of velocity are not significant. Furthermore, in well-constructed basins or border
strips, cross slope is negligible, and once flow is introduced uniformly across the head
end, it advances in a sheet with little transverse variation. In these cases a one-
dimensional analysis of the flow is adequate. Hydraulic variables like flow velocity
and depth can be viewed as functions of distance along the channel and time, i.e., in
one space dimension.
However, in some surface irrigation systems, the conditions for an adequate one-
dimensional representation are not met. It is customary in large level basins, for exam-
ple, to introduce water from essentially a point source, at a corner or in the middle of a
side. The unconfined flow spreads out from the source in all possible directions. No
matter what direction is selected for a one-dimensional simulation, the results can be
considered only a very rough approximation; a two-dimensional analysis is required.
Furthermore, in an imperfectly leveled or graded field, unless the stream size is so
large that water depths are much greater than the variations in the field surface eleva-
tions, water will concentrate in the low-lying areas, both in advance and during reces-
sion, with consequent reductions in uniformity of application over the field. Two-
dimensional simulations can predict the resulting distribution of infiltrated water.
Whereas in pressurized flow, the relationship between line pressure and lateral out-
flow through a nozzle or emitter is relatively well known, the prediction of lateral out-
flow by soil infiltration is far more poorly grounded. Though the theory of infiltration
is well established (Chapter 6), the results are dependent on the physical and chemical
characteristics of the soil, including the geometry of the soil particles and their con-
figuration in the matrix and the ambient water content within. Furthermore, in furrows,
a varying depth of flow can influence the wetted area through which infiltration into
the surrounding soil is effected. Characterizing all of these factors as they vary from
place to place along the irrigation stream can be a formidable task, and major simplify-
ing assumptions are commonly made.
In pressurized lines, resistance to flow is afforded by smooth or slightly rough
walls and is relatively easy to characterize. The relationships between flow rate and
pressure drop in a pipe, while empirical, are well known and measurable with good
repeatability. In surface irrigation, on the other hand, resistance to flow can be exerted
by bare soils, sometimes smoothened by previous irrigations, or by clods which are
large compared to the depth of the stream, by mulch at the bottom of furrows, or by
plant stems and leaves growing through the irrigation stream. The form drag exhibited
by these elements is difficult to characterize, even if their shapes, sizes, and density of
occurrence were known. As in the case of infiltration, major simplifying assumptions
are commonly introduced into the characterization of flow resistance in the simulation
equations.
13.2.2 The Surface Irrigation Process
A complete irrigation takes place in phases, starting with an advance phase, in
which the water moves out over the field from the point of entry. Advance can slow
and stop spontaneously before completion, at some intermediate point along its run.
This condition will occur if the intake rate of the soil is too great for the stream size or
duration. Otherwise, advance ends when the stream front encounters the field bounda-
ries. In a basin, runoff is prevented by a berm, and water ponds behind it; otherwise,
flow runs off at the end. After advance is complete, during a soaking phase, inflow
continues until sufficient water has been introduced to satisfy the requirement in the
field. After inflow cutoff, in the depletion phase, the surface stream is reduced in
depth as infiltration and, possibly, runoff continues. The soil surface below the stream
is eventually exposed as the recession phase begins, usually from the upstream end.
With a sufficiently small slope, recession can start from the downstream end; the
stream can also contract from both ends. The time interval between stream arrival and
departure at a point is known as the infiltration-opportunity time there.
Advance can be characterized as aggressive if the profile of the stream ends in a
blunt front advancing relatively rapidly, with only gradual reduction in velocity. This
is often true as the stream starts out. From the standpoint of simulation, this is a well-
posed mathematical problem: small changes in any of the postulated conditions lead to
small changes in calculated advance. But as more and more of the flow is drawn off by
infiltration—as the length of the stream increases, especially with small stream sizes
and relatively permeable soils—the profile of the front tends more to a feathered edge,
advancing very slowly (at perhaps 0.2 m/min or less) or in starts and stops. This re-
gime is badly posed: small changes in inflow, infiltration characteristics, soil-surface
slope, or roughness can produce large differences in the time to advance to a given
point. Simulation of this condition is not very robust1, with the possibility of spurious
calculation of negative depths and premature front-end recession.
Recession can also physically occur at the leading edge of the water stream, either
prior to completion of advance or after. Any reduction in inflow rate in the course of
system operation can lead to front-end recession. While this occurs most often after
cutoff in horizontal or nearly horizontal fields, it can also arise in sloping fields as
well, and before inflow is cut off. It is only necessary that the inflow rate, Q0, be ex-
ceeded by the rate of growth of the total infiltrated volume, Vz, i.e., dVz/dt be so much
greater than Q0 that the disparity cannot be made up by decreases in surface-water
depth alone. In theory, an infiltration rate increasing with wetting time can also cause
such behavior. Typically, infiltration rates decrease with time, or remain constant;
however, increases have been documented (Trout and Johnson, 1989).
A more common cause for front-end recession before cutoff is cutback operation of
a furrow system. In this mode of operation, in order to prevent excessive runoff from
the end of a field, the inflow is reduced when the streams reach field end. If the re-
maining inflow is insufficient to sustain infiltration over the entire length of the
stream, not only will runoff halt, but the stream will recede from the field end, the
length contracting until the inflow rate balances the total rate of infiltration from the
furrow. If infiltration rates decrease sufficiently prior to cutoff, the stream eventually
begins again to advance. In theory, if the infiltration rate of the soil and inflow rate to
the border or furrow were constant, advance would continue, i.e., stream length would
increase until the total infiltration along the length just matched the inflow rate; then
advance would halt, and stream length would remain constant until conditions
changed.
When inflow is cut off before the stream has reached the end of the field, advance
usually continues for a time, if the stream is deep enough, the slope great enough, and
the infiltration, roughness, and vegetative drag small enough. This is a common oper-
ating condition for high-efficiency irrigation in level basins and in sloping border
strips, where it reduces the amount of runoff.
But if front-end recession is initiated by premature stream cutoff, the leading edge
recedes upstream, while at the same time, if the field slope is sufficient, the trailing
edge recedes from the inlet end, until all surface water has disappeared into the soil.
Unless a series of surges is planned, each advancing farther than its predecessor, this is
clearly not a normal operating condition, for a portion of the field is left entirely dry,
and other areas receive very little infiltrated water. Appropriate mathematical models
of the surface irrigation process, such as SIRMOD (ISED, 1989) and SRFR (Strelkoff
et al., 1998), can predict the onset of front-end recession.
The entire phenomenon represents a case of unsteady, nonuniform, spatially varied
flow over a porous bed, and is similar to other cases of overland flow. As such, it can
be described either through a succession of water-surface profiles, or through hydro-
graphs depicting variation of surface discharge rate or depth with time at a series of
stations along the length.
1
A model is said to be robust when a wide range of conditions can be modeled without user intervention to
prevent an aborted simulation.
13.2.3 Irrigation Performance

The merit of an irrigation event defines how well the water is applied. Several de-
scriptors are in common use. The uniformity of the post-irrigation areal distribution of
infiltrated water (longitudinal distribution in one dimension) is often expressed
through a uniformity coefficient, the ratio of minimum depth in the distribution to av-
erage depth of infiltration,
d
DU min = min (13.1)
Dinf
or the ratio of the average of the low-quarter of the depths to the average depth,
~
d LQ
DU LQ = (13.2)
Dinf
Application efficiency is the percentage of the applied water volume contributing to

the plant requirement (Burt et al., 1997). When the minimum in the distribution just
equals the target (required) depth, the efficiency is known as the potential application
efficiency of the minimum,
DREQ
PAEmin = × 100% (13.3)
DQ
with the volumes expressed in terms of field-wide equivalent depths. When the re-
quirement is just matched by the average of the low-quarter depths, the application
efficiency is known as the low-quarter potential application efficiency, PAELQ. De-
pending upon circumstances, other terms of interest in describing merit are volumes,
equivalent depths, or percentages of runoff and deep percolation (infiltration in excess
of the requirement). See also Chapter 14.
13.3 VARIABLES THAT INFLUENCE THE
SURFACE IRRIGATION PROCESS
The independent variables, governing stream behavior in the field or determining
the solutions to the governing equations in a simulation, can be grouped into three
categories:
(1) prevailing field conditions—soil surface elevations, roughness, and infiltration;
(2) geometry and dimensions of the system; and
(3) operation of the system.
Below, all geometrical aspects of the independent variables are lumped into one
group, and the remaining field conditions, namely, resistance to the surface flow and
infiltration into the soil, into two more. Management of the system can influence all of
these factors: grading a field reduces irregularities in bottom elevations or levels the
surface as well; torpedoing (dragging a heavy, smooth cylinder along the length of a
furrow) or driving farm machinery in a furrow (making the wheel rows) compacts and
smoothes the soil surface reducing both drag and infiltration; while applications of
polyacrylamides (Trout et al., 1995) increase infiltration, as well as decrease the
erodibility of the soil surface. The last category of independent variables describes the
inflow hydrograph.
13.3.1 System Geometry

In the simplest case, the geometry of the channel conveying the irrigation stream is
defined by the cross-section of the furrow, border strip, or basin and its length and
slope, as well as the upstream and downstream boundary conditions, i.e., whether flow
is free to enter and leave each end.
At the upstream end of each flow channel, the inflow configuration determines if the
discharge can be given as an independent quantity. With gated pipe and free flow out of
the openings, inflow is independent of the depth in the furrow. Inflow through siphon
tubes or spiles depends somewhat upon the flow depth, but is usually considered known.
With a transverse head ditch feeding a set of furrows, possibly characterized by different
bottom elevations—for example, with primitive construction methods—the inflows are
directly dependent on the water surface elevation in the ditch. Thus, the layout of a fur-
row system can have a significant effect on surface irrigation performance.
Traditional level basins are sometimes modified by constructing conveyance chan-
nels, large furrows around the perimeter of the basin, designed to conduct the water
from the inlet quickly to the downstream end, so that the basin in effect is irrigated
from both ends. Under idealized conditions, the efficiency of such a basin approaches
that for a basin of just half the length.
To allow light applications with good uniformity and provide an outlet for rainfall,
level basins are sometimes constructed to allow both inflow and outflow from a side
of the basin (level basin with drainback; Dedrick, 1991). In a recent extension of the
surface-drainage concept, increasingly popular in the humid lower Mississippi Valley,
a large laser-leveled field is crisscrossed by a grid of small shallow ditches serving
both supply and drainage functions. Each small sub-area is quickly inundated leading
to good uniformity. Excess water, likewise, freely drains off the planted area.
At the downstream end of a traditional basin or set of furrows, the simplest con-
figuration is a blocked end with zero outflows. Alternately, there may be an overfall
into a drainage ditch. A transverse collector ditch or furrow, with a water level higher
than the bottoms of the longitudinal furrows, can influence the outflow from those
furrows, in effect coupling the flows within them.
Bottom slopes can vary along the length of run, and if furrow flows are coupled,
the bottom configurations of the neighboring furrows in a set will also affect the flow
in any given furrow. In any event, both longitudinal changes in grade and transverse
variations in bottom configuration can have an appreciable effect on field-wide distri-
bution uniformity.
Spatial variability in the bottom configuration is also likely in two-dimensional
flows. In basins, even a small amount of variability can have an appreciable effect on
the post-irrigation distribution pattern of infiltration. A newly laser-leveled basin typi-
cally exhibits a standard deviation in elevation of about 10 mm, while more traditional
grading yields standard deviations from a plane surface of perhaps up to 20 to 25
mm—or twice that in some areas of the world (Fangmeier et al., 1999). See Chapter 5
for estimating the influence of spatial variability in soil-surface elevations on distribu-
tion uniformity in basins. The effect in sloping border strips has not been well docu-
mented, but, qualitatively, it can be expected that the significance of deviations from a
plane surface will diminish with high flow rates, slopes, and roughness.
If furrows are constructed by hand, there can be appreciable longitudinal variation
in cross-section as well as in bottom slope. But even with machine-made furrows, if
the irrigated soils are erodible and stream velocities are high, there can be temporal
and spatial variation in cross-section as the result of erosion and deposition (Trout,
1992). Simulations of sediment transport in surface irrigation are in their infancy, but
the importance of such cross-sectional variations can be studied by giving reasonable
numbers as a function of space and time as input data to simulation models able to
accommodate such variability. Comparisons of output data with and without variation
will point up the need for movable-bed modeling of surface irrigation. It should be
noted, however, that the current interest in modeling sediment transport phenomena
(Strelkoff and Bjorneberg, 2001; Bjorneberg et al., 1999) stems primarily from the
potential for design and management recommendations aimed at reducing soil move-
ment on and from irrigated land.
13.3.2 Resistance to Flow
The drag of the soil surface and inundated vegetation upon the irrigation stream
flow is usually expressed through the friction slope Sf, defined as drag force in ratio to
stream weight (each, per unit length of channel). It depends on the geometry of the
resisting elements (size, shape, distribution, etc.), the cross-section of the flow, and the
average flow velocity. The following definition expression for the Chézy C expresses
that relationship,
QQ
Sf = (13.4)
Ay 2 C 2 R
Here, Ay is the cross-sectional area of flow, Q is the flow rate (with absolute value
signs to account for flow—and drag—reversal), and R is hydraulic radius, the ratio of
cross-sectional area to wetted perimeter. Alternatively, a conveyance K can be defined
such that
Q= K Sf with K = Ay C R (13.5)
written for flow in the positive direction.

13.3.2.1 Surface drag. The Chézy C depends primarily on relative roughness, the
ratio of roughness-element size to hydraulic radius. The widely used empirical Man-
ning formula,
R1 / 6
C = cu (13.6)
n
is characterized by the absolute roughness, n, expressed in m1/6 making the relative
roughness dimensionless; the units coefficient, cu, in the metric system is 1.0 m1/2/s.
To allow the same numerical value of n to be used in the foot-pound system, cu =
1.486 ft1/3m1/6/s.
Devised without the modern theory of resistance to turbulent flow, the Manning
formula, in order to correctly predict C when flow depths vary over a significant
range, requires n to vary with R, contrary to its designation as an absolute roughness.
Tinney and Bassett (1961) postulated a power-law variation of n with depth. More
theoretical studies (Sayre and Albertson, 1961) in very rough open channels, following
Colebrook and White’s pipe-flow analyses with equivalent sand-grain size, led to the
logarithmic formula,
⎛R⎞ (13.7)
C = 6.06 g log10 ⎜⎜ ⎟⎟
⎝χ⎠
in which χ is the absolute roughness, measured in m or ft; g is the ratio of weight to

mass (about 9.81 m/s2 in the metric system, 32.2 ft/s2 in the foot-pound system). In a
comparison study featuring large roughness elements (Strelkoff and Falvey, 1993) the
small variation of the Sayre and Albertson χ compared to the large variation in n testi-
fies both to the superiority and the imperfection of the Sayre and Albertson formula in
this case. To allow estimation of χ from absolute-roughness measurements, extensive
laboratory and field comparisons in the mid-1960s (Kruse et al., 1965) of measured
surface micro-geometry and hydraulically determined values of χ led to the approxi-
mate result:
χ = 12.9σ 1.66 (13.8)
in which σ is the standard deviation of surface elevations in the micro-geometry. More
information on roughness and drag can be found in the extensive bibliographic review
of Maheshwari (1992).
13.3.2.2 Vegetative drag. The purely empirical, classical studies of Ree and
Palmer (1949) have been supplanted by more fundamental, physically based ap-
proaches, depending on whether the vegetation is wholly submerged or protrudes
through the water surface. For the submerged case, the flexibility of the plant parts,
bending into evermore prone positions with increasing flow velocities, has been de-
scribed by a vegetation stiffness parameter, MEI. Here, I is a representative moment of
inertia of the plant-stem cross-section, E is the elastic modulus of elasticity of the plant
material, and M is the number of stems per unit boundary area (Kouwen et al., 1981).
Form drag on vegetation penetrating the flow is distributed over the entire channel
cross-section and is the sum of the drag forces on the individual plant parts. In a bor-
der strip (Petryk and Bosmajian, 1975) with plant density ap (projected area per unit
volume of flow) varying with height η above the bottom, the friction slope chargeable
to vegetation in flow at velocity V and depth y is
V2 1 y
2 g y ∫0
S f veg = C D a p dη (13.9)
with CD a drag coefficient (of value, typically, around 1.0). In the especially simple,
laboratory studies with vertical wires of diameter D, placed in the flow at a density of
M “stems” per unit plan area, ap = MD, a constant. Evaluations of vegetative density
can proceed either geometrically, or hydraulically, i.e., calculating backwards from
measurements of friction slope, average velocity, and flow depth. The hydraulic
measurements would necessarily include the soil-surface drag, a significant compo-
nent if the vegetation is sparse, and which must be estimated to isolate ap. Characteri-
zation of variations in ap can be found in Dudley et al. (1996) and Dunn et al. (1996).
13.3.2.3 Combining soil-surface drag and plant drag. A dense, submerged
growth makes the velocity distribution in the vertical more uniform, and hence, at the
boundary, leads to a steeper gradient and greater shear. Furthermore, the nonuni-
formity in velocity distribution induced by wall shear affects the drag on the plant
parts as a function of depth. But in a first approximation, the drag of the soil surface
and that of the plant parts can be considered independent. Then the two components of
drag can be individually calculated and simply added (note that while drag forces and
Sf are additive, C, n, and ι are not).
Neither the Sayre and Albertson χ, nor the vegetative ap, are in common use today
for surface irrigation simulation or design (SRFR can accommodate values of χ and
ap, constant within a cross-section). While both can be established in hydraulic ex-
periments, there is no body of experience suggesting appropriate values a priori. The
common measure of resistance is, currently, the Manning n in all cases. NRCS in its
design manuals recommends values for specified border-strip conditions (see Chapter
14).
13.3.2.4 Two-dimensional considerations—Sf as a vector. With the drag force a
vector whose direction is opposite the flow-velocity vector, the Sf vector is oriented,
by convention, opposite the drag-force and, in the one-dimensional case, in the same
direction as the flow. This is evident in Equation 13.4, valid for one-dimensional flow
in either direction. In two-dimensional flow, corresponding relationships need to be
expressed more generally because the irrigation stream, in flowing around high spots
in the field surface can come from any direction.
For an arbitrary flow direction, but for isotropic roughness, independent of the di-
rection of flow, in a two-dimensional x1-x2 coordinate system, Equation 13.5 can be
written
q1 = K S f cos θ1 ... q2 = K S f cosθ 2 (13.10)
in which, with i = 1, 2, qi is the component of the vector discharge per unit width, and
Sf is the magnitude of the friction-slope vector. K in this case is a positive scalar, with-
out directional attributes. The angle θi is defined as the angle between the Sf vector and
the xi-direction.
A field can exhibit greater resistance in one direction than another, for example,
when it has been tilled with corrugations parallel to one side, or when drilled grain is
grown, with greater resistance to flow across the rows than along them. In this case,
the conveyance must be viewed as a tensor whose principal values follow directly
from the largest and smallest roughness values encountered by flows in various direc-
tions. When these are not the same, the friction slope will no longer be collinear with
the discharge vector, but will be skewed (Strelkoff et al., 2003a).
13.3.3 Infiltration
Cumulative infiltration is expressed as a depth of surface-water, z (the actual loca-
tion of the wetting front below the soil surface depends upon the porosity and degree
of saturation of the soil). The accumulated intake in a border strip is generally as-
sumed to be a function of intake opportunity time alone, for any given soil at any
given initial water content. In furrows, the accumulated volume infiltrated per unit
length of furrow, Az, has the units of area. Local field depth z is given by z = Az/B, in
which B is the furrow spacing. And because the soil-surface area through which infil-
tration occurs—the wetted perimeter—depends upon depth, the latter, too, in princi-
ple, affects cumulative intake.
In theory, infiltration is described as a boundary condition on the Richards equation
for unsaturated flow in the soil medium (see Chapter 6). In surface irrigation, prior to
recession, bulk water available at the soil surface is drawn into the soil medium. A key
factor in surface irrigation simulation is the infiltrated volume per unit stream length.
It is possible to couple the surface-flow equations to some variant of the Richards
equation (Tabuada et al., 1995; Skonard 2002; Zerihun et al., 2005), but such a theo-
retical approach still requires empirical soil data relating water content to piezometric
head and hydraulic conductivity. The mathematical complexity the coupling entails,
together with observed inhomogeneity and anisotropy of natural soils (see Enciso-
Medina et al., 1998, for an approximate treatment), suggests a more direct empirical
approach. The major issue involved is cumulative infiltration, as a function of inunda-
tion time, wetted perimeter, and furrow spacing. It is true, even in border or basin irri-
gation, that depth of flow plays a role by providing some positive pressure at the soil
surface, but this is small, on the order of a 2% increase in sorptivity (S, in Equation
13.13, below) per cm of depth (Philip, 1969). In addition, the effects of successive
periods of dewatering and rewetting are important in surge flow and, potentially, with
some other inflow-hydrograph patterns, such as cutback flow.
13.3.3.1 Cumulative infiltration as a function of opportunity time. The ob-
served features of the cumulative-infiltration function of time are a large rate of infil-
tration upon initial wetting, followed by a decrease in rate, typically approaching a
constant at large infiltration times. In certain circumstances (typically, with cracked
vertisols) a significant volume is infiltrated as soon as the water is made available at
the surface. A formula exhibiting all of these features is the following modification of
the Kostiakov formula (Kostiakov, 1932; Lewis, 1937):
z = kτ a + bτ + c (13.11)
in which z is the volume infiltrated per unit infiltrating area, τ is the opportunity time,
and k, a, b, and c are empirical constants. Perea et al. (2003) compared the solution of
the Richards equation in a homogeneous isotropic silt loam soil to a fitted equation of
the form of Equation 13.11. In this case the modified Kostiakov equation fit the theo-
retical results to within a half millimeter over a time span between 1 min and 20 h.
The original monomial Kostiakov formulation, z = kτa, is sometimes used for
cracking, swelling clay soils with a tiny value of the exponent a, but a better represen-
tation of physical reality is a nonzero value of c. Furthermore, numerical simulations
of surface flow with such a highly nonlinear infiltration function are inconvenient and
troublesome, as the computations attempt to track the sharp “dog-leg” in the infiltra-
tion profile. The power-law formula is often satisfactory for moderate times in loam or
sandy soils, but at large times, it is characterized by an anomalous, ever-decreasing
infiltration rate. An observed relatively large infiltration rate at large times is often
achieved with the basic formula by postulating a large value of a. A better physical
representation is achieved (Hartley, 1992) by incorporating a nonzero value of b and a
moderate value of a, closer to the theoretical value, 0.5. Indeed, Philip (1969) devel-
oped a theoretical infiltration equation based on unsaturated flow in a homogeneous
isotropic soil medium in the form
z = Sτ 1/ 2 + Aτ (13.12)
Although S and A are constants depending primarily on the soil characteristic
curves, Equation 13.12 falls within the group of functions described by Equation
13.11.
It is often assumed that the basic rate of infiltration is achieved near the end of the
irrigation and can be found by simultaneously measuring outflow Qro and inflow Q0,
i.e.,
Q − Qro
b= 0 (13.13)
LW
in which L is the furrow length, and W is a representative wetted perimeter. It should

be noted, however, the contribution of the Kostiakov power term in Equation 13.11 to
infiltration rate can be of major significance even after many hours (Strelkoff and
Clemmens, 2001).
The Kostiakov branch function (Clemmens, 1981) describes those soils which
achieve the basic infiltration rate sooner than indicated in Equation 13.11. Modified by
incorporation of a constant term, this is
z = c + kτ a L τ ≤ τ B ; z = c B + bτ L τ > τ B
1
⎛ ak
⎞ 1− a
τB = ⎜ ⎟ ; c B = c + kτ B a − bτ B (13.14)
⎝ b ⎠
In Equations 13.14, τB is the cross-over value of inundation time, at which the infil-
tration rate of the first branch matches the final, basic rate b, and cB is the intercept on
the z-axis of the second branch. In some cases, the entire first (power-law) branch can
be effectively assumed to take place in zero time (relative to the total time of simula-
tion), leading to an exceptionally simple, one-branch infiltration formula consisting of
a constant infiltration depth upon initial wetting followed by a constant rate (Collis-
George, 1974).
Midway in complexity between solutions of the Richards equation and the pure
empiricism of the Kostiakov formulations is the Green and Ampt equation (Green and
Ampt, 1911) (see also Chapter 6 and Skonard, 2002). This implicit equation for cumu-
lative infiltration represents an approximate solution to the Richards equation, under
the assumption that behind the wetting front, the soil is saturated, and that the soil wa-
ter content exhibits a discontinuity across this front. The three parameters can be de-
termined from the same kind of measurements as required for a general solution of the
Richards equation, but it is also possible to fit measured infiltration data for the for-
mula parameters.
In USDA (1974), the NRCS characterized infiltration into borders and basins by
membership in a family (extended to furrows in USDA, 1985). The concept implies
that the curve of cumulative infiltration vs. time for any soil at any initial water con-
tent will fit within one of the regions, or families, into which the z-τ plane is divided.
Each family is characterized by a function of the form z = kτa + c, passing through the
middle of the region, and named for the basic infiltration rate for the soil (in inches per
hour). The c-value in the SCS functional form (a constant for all families) does not
represent anything physical, but was introduced to better fit empirical data. While
many soils fail to fit any of the families (graphs of their cumulative infiltration vs.
opportunity time intersect a number of SCS families), some are successfully incorpo-
rated within the SCS group. If experience justifies describing a soil in this way, the
families allow a simple representation of infiltration, with prescribed values of k, a,
and c (see also Chapter 14). It follows that k and a are related,
⎡ 0.148 ⎤
k (a ) = 60 a ⎢14088 a 45 + ⎥ (13.15)
⎣⎢ (− ln a )1.652 ⎦⎥
as per the empirical curve fit to the NRCS-tabulated data (Valiantzas et al., 2001).
Merriam and Clemmens (1985) built upon the NRCS concept of infiltration fami-
lies and developed the time-rated intake families, arguing that the time required to
infiltrate a target depth is the key parameter characterizing infiltration. With a large
database of non-cracking soils, they found a general empirical relation between Kosti-
akov a and the time to infiltrate 100 mm,
a = 0.675 − 0.2125 log10 (t100 ) (13.16)
A single measurement of t100, the time in hours to infiltrate 100 mm, provides the en-
tire Kostiakov cumulative infiltration function, since, with k = 100/t100a,
⎛ 0.675 − a ⎞
2− a⎜ ⎟
k= 10 ⎝ 0.2125 ⎠ (13.17)
(see also Chapter 14).
13.3.3.2 Influence of wetted perimeter on infiltration. In borders and basins, the
wetted perimeter is essentially independent of flow depth. In furrows the infiltrating
surface depends on depth and varies with both distance and time. If infiltration is be-
ing calculated theoretically, e.g., by solution of the unsaturated flow equations in the
porous soil (Schmitz, 1993a,b; Šimunek et al., 1999) coupled to the hydrodynamic
equations in the surface stream (Enciso-Medina et al., 1998; Skonard, 2002), the wet-
ted perimeter would be taken into account automatically. Boundaries to the lateral flux
of soil water engendered by infiltration from neighboring wetted furrows are treated
by postulating a no-horizontal-flow boundary midway between furrows.
But infiltration in surface irrigation simulations is usually treated as an empirical
boundary condition on the surface stream. For furrow infiltration, two schools of
thought have emerged. Some models of the surface irrigation process expect as data
input the parameters of empirical infiltration formulas directly expressing cumulative
volume per unit length of furrow, AZ, as a function of time. This implies that inflow
and outflow measurements had been taken in the furrow, or a similar one, to yield a
time-rate of growth of infiltrated volume, thus incorporating the ramifications of
whatever wetted perimeter was present when the flows were measured. Infiltration-
formula parameters are generally obtained by measuring the increasing infiltrated vol-
ume in a short furrow section containing water at some constant depth and fitting the
data with a formula, now, for example, of the form AZ = Kτ a + Bτ + C (in contrast
with Equation 13.11), with the units of each term now comprising an area (volume per
unit length). The measured infiltration is then viewed as a consequence of both soil
properties (including initial soil water content) and the details of the irrigation system
(furrow size, shape, spacing, flow depth, etc.) But how these data should be adjusted
to apply to water at, say, a different depth, even a constant one, is not clear.
In other models, the infiltration-data input is considered a function of soil proper-
ties alone, with the contribution of the irrigation system then calculated by the model.
The data input consists of formula parameters for volume infiltrated per unit infiltrat-
ing area (i.e., z, cumulative depth of infiltration, as in Equation 13.11), like basin infil-
tration. The volume infiltrated per unit length of furrow, AZ, is then calculated for the
varying flow depths and wetted perimeters of the simulated irrigation stream (NRCS,
1984; Perea et al., 2003).
In any event, at present, it is essential to recognize that the parameter values in the
infiltration formulas are strongly dependent on the accompanying wetted perimeter
assumption. The important physical quantity is Az—for a given value of Az, the pa-
rameters will be much smaller for a wetted perimeter assumed equal to the furrow
spacing rather than, say, on the smaller actual wetted perimeter, dependent on the
stream hydraulics. Thus, for a given physical infiltration, the parameter values should
be roughly inversely proportional to the average wetted perimeter assumed encoun-
tered in the simulation. It should be noted, in passing, that there is some experimental
evidence that as wetted perimeter increases, infiltration increases more than in propor-
tion (or less), suggesting that wetted perimeter be raised to a power somewhat greater
than unity (Blair and Smerdon, 1985), or less (Trout, 1992).
The significance of variations in wetted perimeter, along the length of an advancing
stream, depends on circumstances. If the advance constitutes a significant fraction of
the total application time, the variation in wetted perimeter can have a significant ef-
fect on the final distribution of infiltrated water. This effect is minimal if there is a
long period of runoff. But there may also be a long-term difference in wetted perime-
ter between the head and tail ends of the furrow, particularly with porous soils. In a
tight soil, over a long period of runoff, if the flow in the furrow is essentially uniform,
the wetted perimeter will be practically constant in the furrow, nearly equal to the wet-
ted perimeter at normal depth for the inflowing discharge. With extensively cracked
soils drawing water off laterally, the variations in wetted perimeter with depth, again,
may have little overall effect on the distribution of infiltrated water, the effective wet-
ted perimeter basically just the furrow spacing. A review of extensive field data
(Walker and Kasilingam, 2004) indicated that measured wetted perimeters failed to
show any consistent variation with either time or distance during the course of irriga-
tions, leading to doubts that incorporating theoretical changes based upon the assump-
tion of constant cross-sections (in the dry) in the course of a simulated irrigation, pro-
vides any improvement in accuracy.
13.3.3.3 Influence of rewetting on infiltration. It has long been noted that follow-
ing recession of water from a furrow, upon rewetting, the infiltration rate is smaller
than it would have been had the furrow been wet continuously. Various theoretical
reasons have been suggested for this behavior, including redistribution of soil water
after the initial wetting, but a likely significant cause is the tightening of the soil struc-
ture near the surface under the influence of the negative soil pore pressures as water
disappears from the surface. From the early eighties onward, irrigation by pulses of
applied water, with recession in between, has been used to take advantage of the re-
duced infiltration rates for the successive surges and achieve faster advance rates and a
more uniform application than with continuous inflow until cutoff. The issue of rewet-
ting arises also in the case of cutback furrow streams, if the reduced flow rate is inca-
pable of maintaining infiltration over the entire length of furrow previously wetted at
the full flow rate.
From the standpoint of simulation, some albeit approximate estimate of infiltration
changes under rewetting is desirable, in lieu of totally new measured values for the con-
stants in the infiltration formulas for each surge, or renewed advance. Theoretical solu-
tions of the Richards equation would likely fail to provide good estimates unless some
recognition of the tightened soil structure were incorporated into those solutions. Vari-
ous empirical models have been proposed (e.g., Walker and Humphreys, 1983; Blair and
Smerdon, 1984; Izuno and Podmore, 1985; Purkey and Wallender, 1989). More recent
investigations (Palomo et al., 1996) suggest a simple, accurate (for the soils tested)
modification of Izuno and Podmore (1985) for soils described by Equations 13.14. The
Kostiakov formulation of the first branch, in effect for the first surge, is replaced for all
subsequent surges by the constant, final (basic) infiltration rate, the second branch.
13.3.4 The Inflow Hydrograph

In the simplest case, inflow is held constant until cutoff. In cutback operation, the
inflow is cut, usually in half, when the stream reaches the end of the run. Other frac-
tions of initial stream size and other times to effect the cutback might prove more ef-
fective and can be studied with simulations. A survey of several different inflow pat-
terns as they affect uniformity and efficiency (Alazba and Fangmeier, 1995) provides
an example of necessary studies on the influence of hydrograph shapes on perform-
ance and how these can be achieved in a practical way, without extensive labor re-
quirements.
In surge operation, pulses of inflow (usually of equal magnitude and of increasing
duration) are released, with the intent to cut off each surge as successive constant in-
crements of total length of run are wetted. Commercial surge valves are designed with
a butterfly valve to shunt water to one or the other of two sets of furrows, the on time
of one set corresponding to the off time of the other. These valves, which operate on a
timer, are programmed to estimate the increasing pulse durations necessary. After ad-
vance is complete, they start a more or less rapid pulsing (on the order of 30-minute
cycles) between the two sides in what is called a cutback mode, until the final cutoff.
The two furrow sets are not served with identical inflow hydrographs, because one
side starts cutback immediately after advance is complete, while the other pauses after
its last advance, corresponding to the last advance increment of its counterpart (see,
e.g., sample surge simulations in Strelkoff et al., 1998).
Cablegation (Kemper et al., 1981) provides a gradually decreasing inflow to a set
of furrows fed from gated pipe through which a plug is pulled by a cable. A computer
program (Kincaid and Stevens, 1992) modeling the resultant furrow inflow hydro-
graph was incorporated into simulation software (see, e.g., sample cablegation simula-
tions in Strelkoff et al., 1998).
13.3.5 Estimation of the Independent Parameters
Geometrical parameters for input to simulations or design can be measured in the
field with rulers, tapes, levels, and other surveying equipment. It is worthwhile to note
that any measurement of a level in a survey of field-surface (or furrow-bottom) eleva-
tions with standard surveying equipment is likely to be in error by at least 3 mm.
Measurements of inflow are conveniently and accurately obtained by means of
long-throated critical-depth flumes easy to construct, commercially available in many
sizes, and requiring no field or laboratory calibration (Clemmens et al., 2001b).
Theoretical predictions of infiltration require measurement of the soil characteristic
curves (Chapter 6), but in most cases, empirical determinations are made of infiltrated
depth, or volume per unit length of furrow, as a function of time. Direct measurements
are possible with ring infiltrometers (preferably buffered), but their indications, at one
or a few points in a field, are of limited value in predicting what the average field infil-
tration will be, under flowing water in particular. In general, because of the influence
of running water on the soil structure, flowing-furrow infiltrometers are preferable
over ponded-furrow infiltrometers, which typically underestimate infiltration during
an actual irrigation. Where it is necessary to measure water depths to estimate infiltra-
tion, say in large level basins, the double-bubbler apparatus of Clemmens and Dedrick
(1984) allows this to be done from a distance, accurately and conveniently.
A considerable body of experience has been gained in the last twenty years in esti-
mating infiltration and roughness from measurements of complete border and furrow
irrigations, or during the advance period. Because these techniques often utilize one or
another simulation method to accomplish the evaluation, presentation is delayed until
Section 13.7, after mathematical simulation of surface irrigation has been presented.
13.4 CONSERVATION LAWS FOR MASS AND MOMENTUM
13.4.1 Water Volume Balance
As irrigation water flows down its channel, part of it infiltrating into the surround-
ing soil and part of it advancing onward—to be infiltrated farther down the channel or
to drain off the end of the field as surface runoff—a volume balance is necessarily
maintained: all of the water entering the field at the inlet is accounted for at all times.
At any stage of the process, some of the total inflow volume has infiltrated into the
soil, some has evaporated2, some is (temporarily) stored on the surface, and some,
possibly, has run off. This accounting is a manifestation of the physical principle of
conservation of mass. Because water exhibits so little compressibility in the context of
surface irrigation, conservation of mass translates into conservation of volume. In
terms of total volumes, the balance can be expressed mathematically as follows:
VQ (t ) = VY (t ) + VZ (t ) + VRO (t ) (13.18)
in which VQ is the volume of inflow, VY is the volume on the surface, VZ is the volume
infiltrated, and VRO is the volume of runoff, all at any time t from the start of the irriga-
tion. Equation 13.18 is equally valid for a one-dimensional or two-dimensional formu-
lation. In Equation 13.18,
t
VQ (t ) = ∫ Q0 dt (13.19)
0
in which Q0 is the inflow rate. After advance is complete, runoff volume begins to
accumulate, at the rate QRO(t):
t
VRO (t ) = ∫ QRO dt (13.20)
0
In contrast to inflow, runoff is usually extremely variable. Zero during the entire ad-
vance, it typically rises quickly to a shoulder, continues to rise gradually as infiltration
rates decrease with time, and finally returns to zero more or less gradually after cutoff.
In a one-dimensional context, the surface volume is given by
x A (t )
VY (t ) = ∫ AY ( x, t ) dx (13.21)
0
in which AY is the volume per unit length, or cross-sectional area, of the surface
stream, and xA is the advance distance. After advance is complete, xA = L, the length of
run. Similarly, the infiltrated volume is
x A (t )
VZ (t ) = ∫ AZ ( x, t ) dx (13.22)
0
2
Evaporation is generally ignored; the influence of air and soil temperature is much more pronounced in
affecting water viscosity and infiltration rate than in hydraulically significant evaporation.
in which AZ is the infiltrated volume per unit length. If infiltration is assumed a func-
tion only of wetting time (with variable hydrostatic-pressure or wetted-perimeter ef-
fects neglected), Equation 13.22 can be written
x A (t )
VZ (t ) = ∫ AZ (t − t A [ x]) dx (13.23)
0
(provided xA never decreases with t); tA is the time to advance to x.

13.4.2 Water Dynamics
Of equal importance in governing the flow of the irrigation stream is the dynamic
interplay between the forces of gravity and flow resistance. The gravitational forces
are evident in two ways: directly, as the down-field component of fluid weight, de-
pendent on both water depth and bottom slope, and indirectly (through the hydrostatic
pressure distribution in the vertical) as the consequence of a depth gradient, present if
the water surface is not parallel to the bottom. The two aspects of the gravitational
influence are reflected in the water-surface slope, S0: if this is horizontal, the net
downfield gravitational force is zero; the water is ponded, and the flow is zero.
When the downfield gravitational pull on a stream cross-section, ρgAYS0 per unit
length (with ρ the mass density), is just balanced by the drag upslope of the channel
walls and vegetation, ρgAYSf per unit length, the flow is in equilibrium and uniform, at
constant depth, called normal depth. In a given channel, normal depth is directly re-
lated to flow rate; with some empirical formula for the Chézy C, e.g., Equation 13.6,
cu
Q= AY R 2 / 3 S 0 (13.24)
n
in accord with Equation 13.4, with Sf = S0.
If the gravitational and drag forces are not in equilibrium, there is a depth gradient,
and the water accelerates or decelerates. In most cases of surface irrigation, however,
flow velocities are so small that their temporal changes are negligible, and thus, the
forces of fluid weight, depth gradient, and boundary and vegetative drag are essen-
tially in equilibrium (see Section 13.6.7).
Integral relationships similar to those developed for mass conservation can be writ-
ten for momentum conservation in the entire stream, but the dynamic equations are
usually used to describe small slices of the surface stream in a hydrodynamic approach
to simulation (see Section 13.6). In addition, hydrologic modeling commonly assumes
normal depth at the inlet.
13.5 HYDROLOGIC MODELING OF THE
SURFACE IRRIGATION PROCESS
An irrigation stream is subject to much the same analyses as in simulating canal
and river flows, the major differences being the strong influence of infiltration and
advance over a dry bed. But to avoid the complex mathematics of the hydrodynamic
approach, reasonable assumptions on size and shape are sometimes imposed on the
surface-stream or infiltration profiles, as described below. Then, application of mass
conservation alone leads to the stream behavior. To the extent that the assumptions are
justified, the results can be as good as those of hydrodynamic simulations, with far
less computational effort and a far more robust model. Furthermore, the simplicity of
the equations makes them better candidates than the hydrodynamic equations for solv-
ing inverse problems—estimation of infiltration and resistance from flow observa-

tions, or finding design variables for a given performance level. However, if there isn’t
a good physical basis for the assumptions, the hydrologic model can be grossly in er-
ror, though it remains robust and economical. In the present section, the techniques of
hydrologic modeling are presented; the validity and accuracy of the assumptions made
are, for the most part, deferred until hydrodynamic modeling has been discussed, as
the results thereof are taken as a standard for comparison.
13.5.1 Introduction to Hydrologic Simulation—a Simple, Explicit Model
As noted, conservation of mass or volume, i.e., a volume balance, lies at the heart
of any rational simulation of surface irrigation flows. In a one-dimensional formula-
tion, during advance, Equation 13.18 can be expressed
~ ~
Q0 t = x A AY + x A AZ (13.25)
~
in which xA is the advance distance of the stream up to time t, AY ( x, t ) is the average
~
stream volume per unit length (cross-sectional flow area), and AZ ( x, t ) is the average
infiltrated volume per unit length; time averages are shown barred, while distance av-
erages over the stream length are expressed through a tilde. While inflow continues,
i.e., as long as the upstream depth does not fall to zero, Equation 13.25 can be written
for the advance function as
Q0 t
xA = (13.26)
rY AY (0, t ) + rZ AZ (0, t )
in which rY and rZ are shape factors for the surface and subsurface profiles, respec-
tively, i.e., ratios of average cross-sectional area to upstream cross-sectional area. If
estimates are made for the four terms in the denominator, Equation 13.26 becomes a
simple, explicit formula for advance. AZ(0,t) can be calculated, for at x = 0, t = τ. In a
moderately steep flow channel, the upstream depth rises quickly to normal depth,
yielding AY. Given the generally convex form of the surface-stream and infiltration
profiles, the range of possibilities for the shape factors is 0.5 to 1. Both surface and
subsurface profiles can be assumed monomial power laws in distance back from the
wave front. During advance, the wave profile is quite blunt near the front. This sug-
gests a rather small exponent in the power law. The arbitrary choice of 1/4 for the
power yields a shape factor rY = 0.8 (see Section 13.6.12 for theoretical values). With
z(τ) given by the Kostiakov equation, assumption of a constant advance rate leads to
the exponent a for the infiltration profile, and the corresponding shape factor is
1/(1+a) (theoretically correct for small distances from the tip of the advancing profile:
Katopodes and Strelkoff, 1977a). The right-hand side of Equation 13.26 is now deter-
mined for any advance time t.
As an example, these assumptions are applied to a field test in a border strip (AR-9:
Roth et al., 1974). The irrigation parameters (see Figure 13.1) lead to y0 = 0.0285 m,
~y = 0.023 m, and rZ = 0.74. A plot of Equation 13.26 for this case over a range of time
is shown by the solid line in Figure 13.1, with the measured advance trajectory shown
as open circles. Solutions by more complex methods outlined in subsequent sections
are also shown. Evidently in this case, there is not a great deal of difference with the
various methods of solution, and all agree well with observation.
However, the above assumptions are not always justified. Test ID-NB-12AS-8-60
(Bondurant, 1971) was performed in a border strip of very small slope and a heavy
growth of alfalfa. This case (Figure 13.2) is characterized by y0 = 0.3253 m, rY = 0.8,

rZ = 0.6. Solution of Equation 13.26 yields the solid curve, greatly at variance with the
open circles of observed advance. The great discrepancy between calculated and ob-
served results has prompted much of the research effort devoted to surface irrigation
simulation. Certain results of that effort are shown as dashed and dot-dash curves on
the figure, to be explained in subsequent sections (13.5.2, 13.6.12).
Observed
ry=0.8, rz=1/(1+a)
40
ry=0.8, rz varies in
accord with Lewis-Milne Equation
Hydrodynamic theory (zero-inertia)
30
Time (minutes)
20 AR-9
q0=2.40 Lps/m
S0=0.001
10 n=0.035 m1/6
k=61.8 mm/hra
a=0.358
0
0 25 50 75 100
Advance (m)
Figure 13.1. Comparison of theory and experiment in a border on a moderate gradient.
Figure 13.2. Comparison of theory and experiment in a nearly level border.

13.5.2 Lewis and Milne Equation

Lewis and Milne (1938) recognized that if infiltration is a function of wetting time
alone, the subsurface profile depends solely upon the infiltration and advance func-
tions of time. Then, rZ is a function of xA(t), and with the infiltration function of time
given, Equation 13.25 can be viewed as an implicit expression for the advance, elimi-
nating some of the previous arbitrariness. Indeed, with wetting time, τ (x,t) = t – tA(x),
then,
t
~ dx
Q0 ⋅t = AY ⋅ x A + ∫ AZ (t − t A ) dt A (13.27)
0
dt A
Although Equation 13.27 is exact for the few conditions postulated, it contains too
many unknowns to allow for solution of the advance function, specifically, the un-
~
known variation, AY (x,t) from which the distance-average, AY (t ), could be obtained.
Without a rational (or arbitrary) estimate of this, or of the two factors Ay0(t) and ry(t),
further progress is impossible.
13.5.3 Assumed Average Stream Section
In applying Equation 13.27, in the often misnamed volume-balance approach
(every rational approach to simulation incorporates a volume balance), a constant
~
value for Ay is usually assumed, based on normal depth at the inflow and known a
priori—as in the simple, explicit calculation of Section 13.5.1. It was shown theoreti-
cally (Katopodes and Strelkoff, 1977a) that close to the front of a stream with negligi-
ble infiltration and Manning roughness, the exponent of a power-law surface profile is
3
/7. Tinney and Bassett (1961) found a terminal profile shape, with constant inflow and
no infiltration, following a hyperbolic tangent law for its entire length, somewhat
blunter than the 3/7 power law and closer to a power of 1/4. Soil with substantial infil-
tration, on the other hand, can lead to nearly linearly increasing depth with distance
from the stream front. In any event, the greater the infiltration, the less critical is the
~
estimate of Ay. Indeed, if advance proceeds for a sufficiently long time in a sloping
channel, AY0 and rY are both limited, while AZ, typically, grows without bound. In such
~
cases, advance becomes relatively insensitive to the assumption for AY .
Lewis and Milne proposed solutions for several simple infiltration functions z(τ);
the simplest case, for I = dz/dτ = constant had already been solved by Israelsen
(1932). In this case, Equation 13.27 can be supplanted by the differential equation,
~
Q0 ⋅ dt = AY ⋅ dx A + W ⋅ I ⋅ x A ⋅ dt (13.28)
⎛ W ⋅I t
− ~ ⎞
Q0 ⎜ A ⎟
with the solution xA = ⎜1 − e Y ⎟ (13.29)
W ⋅I ⎜ ⎟
⎝ ⎠
exhibiting an advance necessarily limited to Q0/(W· I).
But more realistic infiltration like the Kostiakov z = kτ a was not incorporated for
over 20 years. One of the earliest schemes, for general forms of infiltration formula
was presented by Hall (1956), who dealt with the (convolution) integral in Equation
13.27 numerically.
y
t1
t2
t3 t4
xA1 xA 2 xA3 xA4 x
t
t t
=4
t 2 3
t 1=1 t 2= t 3=
t4
z01
z02
z03
z04
xA1 xA2 xA3 xA4
z
Figure 13.3. Cumulative infiltration during advance. Time lines separated by constant δt.
13.5.4 Hall Recursive Technique
Figure 13.3 shows how z(x,t) is related to z(0,t). The profiles are drawn for succes-
sive instants of time, a constant interval δt apart. Once z(τ) is known, the horizontal
lines labeled z01, z02,..., etc. separated by the constant time interval can be drawn. The
verticals xA1, xA2,..., etc., represent successive locations of the stream front (albeit un-
known), at t = δt, t = 2δt, t = 3δt, etc. It follows (Hall, 1956) that the successive sub-
surface profiles can be constructed by joining the opposite corners of the resulting
rectangles as shown (furrow irrigation requires neglecting the effect of variable wetted
perimeter upon Az or z). Indeed, with zk,i representing z(xk,ti), and the xk chosen the
same as the successive positions of the advancing front xAi, then zk,i = z0,i-k. It follows
that the change, in one time step, of the convolution integral can be approximated nu-
merically, so that the increment of advance in the time step is
i −1 AZ 0 ,i−k +1 − AZ 0 ,i−k −1 ⎛ 1⎞
Qi ⋅ δt − ∑ δxk + ⎜ rzT − ⎟Az0,1 δx Ai−1
k =1 2 ⎝ 2⎠ (13.30)
δx Ai =
ry Ay0 + rzT Az0 ,1
(Hall 1956, corrected as per Alazba and Strelkoff 1994), in which rzT is the shape fac-
tor for the tip cell only of the subsurface infiltration profile.
13.5.5 Solutions of the Lewis and Milne Equation
with Constant Average Stream Section
One standard approach in treating convolution integrals is with Laplace transforms
(Boas, 1966; Philip and Farrell, 1964; Lenau, 1969). Amongst other efforts to solve
the Lewis and Milne integral equation, Hart et al. (1968) derived a dimensionless ver-
sion for the Kostiakov equation, solved by a numerical solution of controllable accu-
racy, while Collis-George (1974) developed a closed-form solution for the second
branch of the Kostiakov-Clemmens branch function.
Figure 13.4 provides an overview of solutions to the Lewis and Milne equation
with constant average surface-water depth and Kostiakov infiltration formula. The
dimensionless representation allows all solutions to be represented in a single family

of graphs. In this representation, t* = t/TR, and xA* = xA/XR, in which
⎛ A~ ⎞1/ a
⎜ y ⎟ Q T
TR = X R = 0~ R (13.31)
⎜ B0 k ⎟ Ay
⎝ ⎠
Each point was obtained by numerically integrating a definite integral derived as
per Lenau (1969), except for the curve labeled a = 1.0, representing a constant infiltra-
tion rate. For this, the dimensionless counterpart of Equation 13.29 was graphed. The
lower limit of a, 0.0, is not shown, because in this case the reference time TR becomes
undefined, and the selected reduction to dimensionless form breaks down. In dimen-
sional terms, the advance for a = 0.0 is given by
Q
xA = ~ 0 (13.32)
Ay + B0 k
in which k is the instantaneous infiltration depth when water is applied (as into
cracked soils over hardpan).
Figure 13.5 is a similar overview, showing the corresponding variation in infiltra-
tion-profile shape factor (which would yield a more accurate estimate of advance, with
Equation 13.26, than Figure 13.4). Theoretically derived limiting values of rz are (Hart
et al., 1968),
1 a π (1 − a )
rz = at small time rz = at large time (13.33)
1+ a sin (aπ )
The two known limits have suggested a number of curve fitting approaches to re-
place the information given graphically in Figures 13.4 and 13.5 by analytic proce-
dures. Valiantzas (1997a) took advantage of the partial collapse of the curves of Fig-
ure 13.4 towards a single curve when they are normalized and the independent vari-
able is changed to xA*/t* (which varies in the range 1 to 0, as t* varies from 0 to ∞ ).
He derived an approximate analytic fit by a simple polynomial; a few iterations prove
necessary to construct a point x*(tA*) on the advance curve because of the form of the
independent variable in the polynomial. Alazba (1999) sought an explicit solution by
fitting a rational expression to rz(t*), which then requires solution of a quadratic equa-
tion for a point x*(tA*). Both approaches are computationally less intensive than the
Lenau (1969).
It is useful to consider the conditions of Figures 13.1 and 13.2 in the light of Figure
13.5 in order to judge the magnitude of the error introduced by the assumption of the
small-time value for rz appearing in Equation 13.33. In Figure 13.1, the dimensionless
time associated with a 40-minute advance is 10.9, which in Figure 13.5 is seen (by
interpolating between the 0.3 and 0.4 a-curves) to increase the small-time value of rz
of 0.74 by about 4%, to 0.77. With Az about 1.7 times Ay at that time, correcting rz
would decrease xA by about 2.5%, i.e., to the dashed line, which represents the solution
to the Lewis and Milne equation. Similar corrections in the case of Figure 13.2 also
move the calculated advance from the solid line to the dashed line, which still exhibits
a fundamental divergence from the measured values. In this case, then, the assumption
of small-time conditions for rz was not significant (for a real time t = 60 min, t* = 10-5),
and some other factor is responsible for the discrepancy (in this extreme case, up-
10 000
0.2
0.3
1000 0.4
0.5 .1
0.6 a=0
0.7
0.8
100.
0.9
10.0
a=1.0
1.00
0.10
0.01
0.01 0.10 1.00 10.0 100. 1000.
Figure 13.4. Advance with constant average surface depth (Kostiakov infiltration).
1.00
a=0.1
0.2
0.3
0.75 0.4
0.5
0.6
0.7
0.8
0.9
a=1.0
0.50
0.01 0.10 1.00 10.0 100. 1000. 10 000
Figure 13.5. Infiltration-profile shape factor (Kostiakov infiltration).
stream depth during the hour of advance shown is very much less than normal depth;
see Section 13.6.12). It cannot be generally said, however, that errors in assuming the
first of Equations 13.33 throughout the advance are negligible: Figure 13.5 shows that
while at low values of a, variations in rZ with time are indeed insignificant, the larger
values show variations of up to 100%. The commonly used, theoretical value, a = 0.5,
shows a variation of about 16% over the entire possible range of irrigation times.
Figures similar to Figures 13.4 and 13.5 could be drawn for infiltration formula-
tions other than the Kostiakov. For example, with the theoretical Philip formula,
Sτ 1/2 + Aτ, dimensionless variables could be based on the reference time,
~ ⎞2
⎛ A
TR = ⎜ y ⎟
, instead of that in Equations 13.31, and the parameter identifying the
⎜W ⋅S ⎟
⎝ ⎠
A
curves would then be A* = 2 ~ . Infiltration formulas with more than two pa-
S W / Ay
rameters would require pages of graphs to cover the potential range.
13.5.6 Power-Law Advance
The curves on logarithmic paper in Figure 13.4, solutions of the Lewis and Milne
equation for advance (an expression of mass conservation) with a constant average
surface water depth and power-law infiltration, exhibit straight-line (power-law) ad-
vance,
xA = f t h ⋅ (13.34)
at both very small times and very large times, but not in between. However, measured
advance as well as hydrodynamic simulations often show a good power-law fit over
moderate intervals, say from 10% of field length to field end. Then, assumption of
both z = kτ a and xA = f · th leads to significant simplifications in the Lewis and Milne
expression for infiltrated volume, namely,
t
⋅ ⋅ 1
B0 ∫0 k (t − t A )a hf t A h −1dt A = B0 kt a f t h h ∫0 (1 − α )a α h −1 dα (13.35)
in which the definite integral on the right-hand side, a constant for given a and h, in-
dependent of time, is the well-known beta function, expressible in terms of gamma
functions (Abramowitz and Stegun, 1964, formulas 6.2.1, 6.2.2). The resulting shape
factor for the full infiltration profile is
(13.36)
Γ (1 + a) Γ (1 + h)
rz =
Γ (1 + a + h)
completely independent of time. The algebraic expression (Christiansen et al., 1966),
1 + a + h − ah
rz = (13.37)
1 + a + h + ah
is a close empirical fit over the entire range of 0 < a < 1 and 0 < h < 1.
At very small advance times, when the infiltration volume is negligible compared
to the surface volume at constant depth, h = 1, and rZ reduces to
Γ (1 + a ) Γ (1 + h) Γ (2) Γ (1 + a ) Γ (1 + a) 1
rZ = = = = (13.38)
Γ (1 + a + h) Γ (2 + a) (1 + a )Γ (1 + a) 1 + a
confirming the limiting values on the left of Figure 13.5. At very large advance times,
the surface volume is negligible compared to the infiltrated volume, and a + h = 1
(Hart et al., 1968); then,
(13.39)
Γ (1 + a)Γ (2 − a) π a(1 − a )
rZ = =
Γ (2) sin(π a )
(derived from Abramowitz and Stegun, 1964, formula 6.1.17), a function of a alone
and confirming the values approached at large times in Figure 13.5.
The more general Modified Kostiakov infiltration formula (Equation 13.12) can
also be coupled to power-law advance to yield
(
V z = f ⋅ t h rz1 kt a + rz 2bt + rz 3c ) in which :
1
rz1 = rz of eq. 36/37; rz 2 = ; rz 3 = 1 (13.40)
1+ h
The power-law approach in this section can be utilized either without any assump-
tions regarding the average surface-water depth, as in the inverse problem of estimat-
ing the Kostiakov parameters in Section 13.7, or as an approximation to the solution of
the Lewis and Milne equation with an assumed constant average surface-water depth,
as follows.
13.5.6.1 Two-point method for advance. Assumption of mass conservation with a
constant average surface-water depth and approximate power-law advance provides an
approximate alternative to the other hydrologic methods of estimating advance times,
~
given infiltration in the form kτa, slope, roughness, and inflow rate (leading to AY). For
example, in a field of length L, if advance times to the halfway point tL/2 and endpoint
tL were known, the entire advance curve, f · th, would be known. Then, simultaneous
solution of three nonlinear transcendental equations for tL/2, tL, and h solves the ad-
vance problem. With Q0 and xA known at the two unknown times, the two correspond-
ing values of VZ are known as functions of the two times; the pertinent simultaneous
equations are:
⎛L⎞
G1 = B0 kt L / 2 a ⎜ ⎟ rZ (a, h) − VZ L / 2 = 0
⎝2⎠
G2 = B0 kt L a L rZ (a, h) − VZ L = 0 (13.41)
1 h
G3 = t L −t L / 2 h = 0
2
~ L
in which VZ L / 2 = Q0 t L / 2 − AY (13.42)
2
~
VZ L = Q0 t L − AY L
For solution of the system by the Newton-Raphson procedure, the required derivative
of rz with respect to h is most simply found by differentiating the Christiansen expres-
sion (13.37).
13.5.7 Hydrologic Modeling —Conclusions
A fundamental goal of surface irrigation simulation is to predict the distribution of
infiltrated water along the length of run. If the advance and recession of the surface
stream were known, the opportunity times for infiltration would be known, and with a
known infiltration function independent of flow depth, the problem would be solved.
For infiltration that is dependent only on wetting time and not upon flow depth, the
Lewis and Milne integral equation is an exact statement of mass conservation during
irrigation-stream advance, and if some way were found to estimate the variation in
average surface-water depth with time, it would not lead to the kind of errors exhibited
in Figure 13.2. Most existing methods for solution of the equation, however, have as-
sumed that the average depth of surface water is a known constant. The inaccuracies
engendered by this assumption have led to the development of methods for simulating
advance by utilizing the laws of open-channel hydraulics governing the flow in the
surface stream, as in the next section.
It is possible to examine other phases of a surface irrigation, for example, deple-
tion, runoff, and recession by making plausible assumptions regarding the surface-
water profile and infiltration rates (Shockley et al., 1964; Wu, 1972; Strelkoff, 1977;
Singh and Yu, 1989a,b) and solving an equation of continuity for mass conservation.
But with the generally increasing tenuousness of the necessary assumptions on stream-
profile shape for the phases of an irrigation following advance, and the widespread
availability of fast personal computers, these approaches have, by and large, been sup-
planted by simulations—numerical solutions of the governing hydraulics equations,
which are equally applicable to all phases of an irrigation event.
Still, the simplicity of the hydrologic approach suggests using it to solve the inverse
of simulation, for field-parameter estimation and for design. Data for the appropriate
assumptions, then, could be gleaned from parallel hydrodynamic solutions, rather than
guesswork. Likely there will be a continuing interest in this oldest attempt at simulat-
ing surface irrigation.
13.6 HYDRODYNAMIC MODELING
OF THE SURFACE IRRIGATION PROCESS
13.6.1 Physical Laws of Open Channel Hydraulics—
One-Dimensional Solution Grid
The major errors that can be introduced by an assumed shape and depth of the sur-
face stream have led to calculating its length and depths on the basis of conservation
of mass and momentum applied over a series of time steps to elements of the surface
stream and infiltrated water. With this approach it is unnecessary to assume that infil-
tration is independent of furrow depth (wetted perimeter), for stream depth is now a
solution variable. But both infiltration and drag must still be estimated empirically.
Recent theoretical developments in turbulent flow calculations, and techniques for
coupling unsaturated flow of soil water to flow in the surface stream (Tabuada et al.,
1995; Bradford and Katopodes, 1998; Skonard, 2002; Zerihun et al., 2005) are making
such purely empirical contributions unnecessary, but these methods are not yet practical
or sufficiently accurate in their theoretical infiltration results for routine calculations.
In a typical hydrodynamic approach, the surface stream is subdivided into enough
slices, or cells, that the shape of each cell is closely approximated by a trapezoid, as in
Figure 13.6 (for the tip cell, a power-law shape is often assumed); the computational
Figure 13.6. Subdivision of stream length into cells.

L R
J M
Figure 13.7. Computation network in x-t plane.
field expands as the stream lengthens. The total simulation time is divided into a suffi-
cient number of intervals that the variation of pertinent variables over a time step can
also be considered linear (Figure 13.7). The distance and time steps are often program
controlled, with the user supplying a target cell density, the approximate number of dis-
tance steps to cover the entire length of run. In this case the time steps constitute part of
the solution for stream advance. The cell faces—nodes in the solution grid—are best
fixed with time along the length of run, as the stream increases in length, in order to
avoid interpolation in the infiltration profile from time step to time step3. The nodes do
not need to be equidistant, but should coincide with field locations where there is a break
in slope, or step change in infiltration or roughness parameters. At such locations, a dou-
ble node is needed because of possible step changes there in surface or infiltrated depth.
Similarly, cutback and cut-off times or the start of a surge should coincide with a solu-
tion time level. When the time step size is given during the advance phase, the advance
increment constitutes part of the solution. The goal is always that the variables change
continuously and gradually over a time or distance step, without breaks in gradient, for
then a linear approximation within the cell is justified.
13.6.2 Mass Conservation
In Figure 13.8, the solid lines represent the cell profile at the beginning of a time
step, the dashed lines at the end of the time step. The cells for expressing mass conser-
vation include both the surface water and infiltrated water, and are bounded at the top
by the actual water surface, at the bottom by the depth of infiltration (averaged over
the transverse direction), and at the front and rear by faces normal to the channel bot-
tom.
3
In the mass-balance computations, interpolation errors in the calculated infiltrated volume can lead to
computed negative surface depths, when these are small compared to infiltrated depths, and an aborted
simulation (Strelkoff et al., 1998).
Figure 13.8. Mass conservation in a cell of surface and infiltrated water.

Solid line: beginning of time step; dashed line: end of time step.
The net inflow of mass into a cell must equal the increase in its mass. In terms of
the variables in Figure 13.7 and 13.8, the mass balance for a time step δt is (with con-
stant mass density),
[θQL + (1 − θ )QJ ]δt − [θQR + (1 − θ )QM ]δt =
{[φ y L Ay L + φ y R Ay R + φz L Az L + φz R Az R ] − (13.43)
[φ y J Ay J + φ y M Ay M + φz J Az J + φz M Az M ] } δxLR
The weighting factors θ andφ reflect any departures from a straight-line variation of
the variables with time or distance (the trapezoidal rule calls for the value 1/2). The
factor θ is given a value a little greater than 0.5 even when time steps are small enough
that the time variation is well approximated by a straight line. This proves necessary to
prevent instability of calculations proceeding over a large number of time steps. Oth-
erwise, numerical schemes like Equations 13.43 and 13.45 (below) magnify small
numerical errors of truncation and roundoff with each succeeding time step, so that
ultimately, these oscillatory errors completely mask the true solution (see, e.g., Cunge
et al., 1980).
13.6.3 Momentum Conservation
Unlike mass, momentum can be created. The impulse, or time integral, of the resul-
tant of all the forces acting on a water-filled cell increases momentum in the same
amount. That increased momentum either builds up inside the cell, or flows out
through the cell boundaries.
The forces acting on the surface water in the cell in the direction of flow are pic-
tured in Figure 13.9. These are the down-slope component of fluid weight, drag of the
walls and bottom and any vegetation in the flow, and the difference in hydrostatic
pressure forces on the left and right faces. In a nonprismatic furrow, the pressures on
the cell faces are augmented by an inline component of the water pressures on the flar-
ing side walls.
The weight component is equal to the volume of water in the cell, multiplied by γ,
the unit weight of water, and the slope S0 of the channel bottom. The drag is expressed
through the friction slope Sf. The net hydrostatic pressure force on the cell water, in-
cluding the contributions of nonprismatic side walls, is most simply expressed through
Figure 13.9. Momentum conservation: forces acting on cell of surface water.
the average cell cross-sectional area multiplied by the difference in water depth at the
two faces and the unit weight of water. In fact, the sum of pressure and gravitational
forces is well described by the average cell cross-sectional area multiplied by the dif-
ference in water-surface elevations at the front and rear faces of the cell. This formula-
tion insures that standing water in a trough of any shape will be computed with a hori-
zontal surface.
Thus, the impulse δIF of the resultant force acting on a cell of water in the direction
of flow over the duration of a time step is
δIF = [θ (φyL AyL + φyR AyR)(hL – hR – Sf LR δxLR)
(13.44)
+ (1 – θ) (φyJ AyJ + φyM AyM)(hJ – hM – Sf JM δxJM)] γδt
in which h is water-surface elevation.
Just as each element of volume possesses mass, each element of volume δV pos-
sesses momentum in the amount, ρVδV, oriented in the same direction as the vector V.
Thus, momentum enters and leaves the cell through its upstream and downstream
faces; the infiltrating water possesses virtually no momentum. The buildup and flux of
momentum δM for the cell is given by the following equations, paralleling Equations
13.43,
δM = [θρVR QR + (1 − θ ) ρVM QM ]δt − [θρVL QL + (1 − θ ) ρV J QJ ]δt
(13.45)
( ) ( )
+ φ y LVL Ay L + φ y RVR Ay R ρδx LR − φ y J V J Ay J + φ y M VM Ay M ρδx JM + δM I
The last, negligibly small term represents momentum carried out of the surface
stream by infiltration. On the basis of energy considerations and reasonable physical
assumptions (Strelkoff, 1969), the momentum flux δMI contributed by infiltration in a
time step is
⎛ ρ ρ ⎞
δM I = ⎜θ VLR W ⋅ iLR δx LR − (1 − θ ) V JM W ⋅ i JM δx JM ⎟δt ≅ 0 (13.46)
⎝ 2 2 ⎠
The first overbar in each expression is the average velocity in the cell; the second is
the average volumetric rate at which water infiltrates out of the cell per unit length,
consisting of a representative wetted perimeter W and a representative infiltration rate
i per unit area.
Ultimately, the momentum equation for each cell is
δI F = δM (13.47)
13.6.4 Reduction to Partial Differential Equations

With specified roughness and infiltration characteristics and given inflow rate and
downstream boundary condition (advance, blocked end, or free overfall), the nonlinear
algebraic equations, 13.43 and 13.47, are ready for simultaneous numerical solution
for a given time step, over all of the cells. To orient the reader, however, it is worth
noting that they can be reduced to the momentum-conservation form of the partial
differential equations of open-channel flow (Saint Venant, 1871) by dividing both
equations by γδxδt and reducing δx and δt to zero. In the limit,
∂Q ∂Ay ∂Az
+ + =0
∂x ∂t ∂t
1 ⎡ ∂ ⎛ Q2 ⎞ ∂Q V ∂Az ⎤ ⎡ ∂y ⎤
⎢ ⎜ ⎟+ + ⎥ + Ay ⎢ − S 0 + S f ⎥=0 (13.48)
g ⎢⎣ ∂x ⎜⎝ Ay ⎟ ∂t 2 ∂t ⎥
⎠ ⎦ ⎣ ∂x ⎦
which reduces, with standard calculus, to the traditional non-conservative form,

∂V ∂y ∂y ∂A
Ay + VB + V Ay xy + B + z = 0
∂x ∂x ∂t ∂t
1 ∂V V ∂V V ∂Az ∂y
+ − = − S0 + S f (13.49)
g ∂t g ∂x 2 g ∂t ∂x
in which B is the top width at the depth y, and Ay xy is the rate of increase in Ay with
distance with the depth held constant in a non-prismatic furrow; in a prismatic furrow,
it is zero.
13.6.5 Initial and Boundary Conditions (Saint Venant)
Initial conditions for these equations, at t = 0, are zero depth and discharge every-
where. Upstream boundary conditions are usually a given inflow hydrograph, with the
solution yielding upstream depths. Alternately, upstream depths or water surface ele-
vations in a supply ditch can be given, and the resulting inflow calculated.
During advance, the downstream boundary condition, at the wave front, is zero
depth. Flow velocities approach a finite limit there (Whitham, 1955; Sakkas and
Strelkoff, 1974), so the discharge at the front is also zero.
When advance is complete, at the downstream field boundary, if the flow is
blocked, discharge is zero, and depth is calculated. For free overfall into a drainage
ditch, the flow is at critical (local Froude number is unity); if the water level in the
drainage ditch is above the inundated soil surface by an amount greater than critical,
that depth is the boundary condition, and the discharge is computed. To avoid the need
for specifying a depth, Sakkas et al. (1994) assume continuing advance into a hypo-
thetical field extension, the resulting discharge at the actual field end constituting the
runoff hydrograph.
At the trailing edge in recession, depth and discharge are both zero.
13.6.6 Solution of the Complete Saint Venant Equations
The acceleration terms (δM in Equations 13.47; the terms divided by g in Equations
13.48, 13.49) make the equations hyperbolic, i.e., wave equations, whose physical
behavior is truly mirrored only in a solution based on intersections of two families of
characteristic curves (Garabedian, 1964; Sakkas and Strelkoff, 1974; Strelkoff and
Falvey, 1993). As noted, subject to minor damping and phase shift, Equations 13.43
and 13.47, written in a Preissman scheme (Cunge et al., 1980; Walker and Skogerboe,
1987), are ready for solution without the need for any characteristic equations.
Solution of the St. Venant equations provides an accurate solution, sometimes nu-
merically sensitive, yielding jagged profiles or even failing with a computation of a
negative depth. Two common truncated forms of the St. Venant equations, zero-inertia
approximation and kinematic-wave approximation, are used to provide a more robust
simulation. The approximations find a broad range of applicability, though their limi-
tations should be recognized.
13.6.7 Zero-Inertia Approximation
The acceleration (inertial) terms in Equations 13.43 and 13.47, or 13.48 or 13.49,
though relatively small in surface irrigation because of the low water velocities, can
still be numerically troublesome and so reduce the robustness of a model. Justification
for truncating the governing equations to a zero-acceleration or zero-inertia form is
provided by quantifying the influence of the acceleration terms on surface irrigation
advance, as shown in Figure 13.10. Dimensionless representation, with each physical
variable viewed in ratio to a reference variable, allows a single-page overview of this
influence for a wide range of Kostiakov k, Manning n, and inflow rate. In a border
strip with unit inflow rate q0 and Kostiakov infiltration and Manning roughness, a con-
venient system is
q y t τ
q* = y* = t* = τ* = with
QR YR TR TR
1.00
a=0.5
S0* =1.0
Dimensionless time (t*)
el
od
m
tia
er
-in
0.10
ro
Ze
St. Venant Model

FN =0.1
FN =0.2
FN =0.3
FN =0.4
0.01
0.01 0.10 1.00
Dimensionless advance (xA*)

Figure 13.10. Effect of reference Froude number on stream advance.
1/ a
⎛Y ⎞
QR = q0 YR = y N (q0 , S 0 , n ) TR = ⎜ R ⎟ QR TR = YR X R (13.50)
⎝ k ⎠
Substitution into Equations 13.48 yields dimensionless equations with just three in-
dependent parameters during advance at constant inflow: a Froude number FN at nor-
mal depth for the inflowing discharge, a dimensionless slope S0*, i.e.,
QR 2 S0 (13.51)
FN = and S 0* =
g YR YR / X R
and the Kostiakov exponent a.

Figure 13.10 (drawn from Katopodes and Strelkoff, 1977b) demonstrates the very
small influence on the results if the inertial terms are discarded from the equation of
motion, leaving a statement of equilibrium amongst the forces acting on the surface
stream. The limiting curves, with FN = 0 were obtained with all inertial terms (those
containing g, in Equations 13.48, or 13.49) simply truncated from the equations (or
setting δM = 0 in Equation 13.47). The specific requirement for justification of this
procedure is that the Froude number at normal depth is small enough, say less than
about 0.4, a condition easily met in surface irrigation, with its small velocities.
The dimensioned zero-inertia equations, then, for borders, basins, or furrows are
Equation 13.43 and Equation 13.44 written for the end of a current time step and re-
duced to the following form,
(φyL AyL + φyR AyR)(hL – hR – Sf LR δxLR) = 0 (13.52)
a simple statement of equilibrium amongst the forces acting on the water in a cell,
since δM = 0.
13.6.8 Initial and Boundary Conditions (Zero Inertia)
Initial and boundary conditions for the zero-inertia equations are the same as for the
Saint Venant equations (Section 13.6.5), but it should be borne in mind that flows with
zero inertia are infinitely subcritical, i.e., critical depth is zero (Strelkoff and Katopo-
des, 1977a). At a free overfall, while the theoretical velocity is infinitely great, the
discharge remains bounded, and just a short distance upstream from the brink, the ve-
locity is down to realistic levels.
The common practice of specifying normal depth as the downstream boundary
condition for a freely draining furrow or border is inconsistent with the zero-inertia
concept. If the bottom slope is too small to warrant a kinematic-wave approach to the
simulation (see Section 13.6.11), i.e., too small to assume normal depth at all points in
the surface stream, then it is too small to assume it at the end.
13.6.9 Solution of the Zero-Inertia Equations
With the acceleration terms truncated from the Saint Venant equations (tantamount
to an infinitely great relative value of g in the partial differential equations), the char-
acteristic directions in the x-t plane for the two families degenerate into dt/dx = ±0.
This indicates that any disturbance introduced at the boundary travels instantly into the
interior of the stream, felt immediately everywhere (though with ever-decreasing in-
tensity with distance from the source, just as with the heat equation, which the zero-
inertia equations mimic). Thus, the zero-inertia equations represent a two-point
boundary value problem progressing with time, a natural for algebraic solution of
Equations 13.43 and 13.52 at each time step (Strelkoff and Katopodes, 1977b). While
most solutions of the zero-inertia equations have been numerical (Wallender and
Rayej, 1985; SRFR, Strelkoff et al., 1998; SIRMOD, ISED, 1989), including the fi-
nite-element approach of Shayya, Bralts, and Segerlind (1991), a few elegant and
complex analytic solutions exist (Schmitz and Seus, 1987,1990,1992).
In numerical solutions, the implicit system of nonlinear algebraic equations at each
time step together with the boundary conditions, constitute a closed nonlinear alge-
braic system. The Newton-Raphson technique is effective in obtaining a solution for y
and Q at each of the interior points, plus either one or the other at upstream and down-
stream field boundaries, and δxA (or δt) during advance. The matrix of coefficients for
the system of linear equations in the corrections to the successive approximations of
the solution variables is sparse, and is effectively inverted by double-sweep techniques
(e.g., Cunge et al., 1980). Either, the magnitude of the time step is given, and the ad-
vance increment is unknown, or, the advance to a given location is given, and the time
step is unknown. In the latter case, the matrix of coefficients contains a column of
coefficients for δt. An extension of the original double-sweep technique is available
for the solution (Strelkoff, 1992).
Convergence of the Newton-Raphson iterations is good most of the time, but de-
pends to some extent on the initial guesses. Furthermore, in poorly posed regions of
solution (e.g., very slow advance), the corrections developed by the technique can be
too large—and negative—for the small depths, and the simulation aborts. Usually,
satisfactory initial guesses are simply the values at the beginning of the time step. For
advance, a more satisfactory first guess comes from approximate mass-balance con-
siderations (SRFR, Strelkoff et al., 1998).
In general, if the correction terms prove so large as to disrupt the iterations, they are
arbitrarily limited to a fraction of the prescribed values. However the same fraction is
applied to all the terms in the correction vector so that its (Newton-Raphson) direction
remains unchanged (“backtracking,” Press et al., 2001). Convergence, even if slow, is
sufficient justification for the procedure.
Sometimes the nonlinear algebraic equations are linearized and the solution vector
at the end of δt found in one step (Cunge et al., 1980). Theoretically, the procedure,
which amounts to just the first iteration in the Newton-Raphson procedure, is justified
if the solution is continuous and δt is small enough, but pursuing the successive itera-
tions until the equations are satisfied to a prescribed tolerance ensures that continuity
and momentum are indeed satisfied in each cell at every time step (SRFR, Strelkoff,
1998).
13.6.9.1 Nearly stopped advance. As noted in Section 13.2.2, very slow advance
is a poorly posed problem, with premature recession a likely result in the course of the
iterations. An approach taken in SRFR (Strelkoff et al., 1998) is applicable both to
numerical wavering of advance and physical front-end recession stemming from a
temporary insufficiency of inflow. Upon calculation of negative advance or a negative
depth just behind the wave front, the computational boundary is backed up one cell.
The downstream boundary condition is set to a small depth, through which outflow
spills out and infiltrates onto the soil ahead. When the depth of this infiltration ahead
of the front is large enough in one time step to satisfy the infiltration equation for the
time step, the computational boundary is moved one cell downstream again. In this
way, advance can halt and restart, perhaps many times, as observed in the field with
nearly stopped advance.
13.6.10 Kinematic-Wave Approximation
While the zero-inertia approximation is valid provided only that the Froude number
of the flow is low enough, it becomes increasingly inefficient to implement as the
steepness of the bottom slope increases. Then the flow profile is almost uniform along
the length of the stream, and only at the end makes a sharp down turn to zero depth at
the very front of the wave. In order to apply the trapezoidal rule to the numerical inte-
gration over the slices comprising the surface stream, they need to be made smaller
and smaller toward the front. A typical numerical response to too coarse a grid near
the wave front is a blip or spike at the front, which if sufficiently great is accompanied
by a negative depth just behind, which aborts the computation. To avoid the problem
with the x-nodes stationary in a full hydrodynamic or zero-inertia context, all cells
must be small.
At the same time, with a large bottom slope, the forces stemming from the pressure
difference on the faces of a cell are dwarfed by both the component of stream weight
down the slope and the drag resisting the flow. The stream is thus essentially at normal
depth—for the local discharge—everywhere along its length. This is sufficient to meet
the conditions for the existence of a kinematic wave (Lighthill and Whitham, 1955),
viz., a unique relation between discharge and cross-sectional area at any location in the
stream, specifically, in open-channel hydraulics,
Q2
S f = S0 = (13.53)
K2
in which K is the conveyance of the flow channel (Equation 13.5), a function of x and
Ay, or Ay alone in a uniform, prismatic channel. Substitution of Equation 13.53 (the
second of Equations 13.48 with ∂y/∂x and acceleration terms deleted) into the first of
Equations 13.48 (continuity) yields the kinematic wave equation,
∂Ay ∂Ay ∂Az
+w =− − QxAY (13.54)
∂t ∂x ∂t
In a nonuniform channel, the relation between discharge and area is dependent on
location x, so that
∂Q( Ay , x ) A ∂Q( Ay , x )
w= while Qx y = (13.55)
∂Ay ∂x
describes the nonuniformity, whether variable slope, cross-section, or roughness. In a

A
uniform channel, Qx y = 0, and
dQ S 0 dK ( y )
w= = (13.56)
dAy B( y ) dy
Kinematic-wave solutions have been obtained analytically or semi-analytically, de-
pending on the infiltration function selected, for example, by Sherman and Singh
(1982), Singh and Ram (1983), and Weir (1983, 1985). A theoretically correct nu-
merical solution to Equation 13.54 would follow a single family of characteristic
curves in the x-t plane, each curve with inverse slope w(x,t) (Garabedian, 1964; Smith,
1972; Strelkoff, 1985), but at the expense of some numerical attenuation (Hromadka II
and DeVries, 1988), a practical numerical solution to the kinematic-wave approxima-
tion can be obtained in terms of the same rectangular grid (Figure 13.7) as with the
zero-inertia and full hydrodynamic forms. Analysis of the characteristics solution dis-
closes essential elements of kinematic-wave behavior that can be accommodated on a
rectangular grid. For example, no independent downstream boundary condition is pos-
sible; conditions at the downstream end of the channel are governed entirely by cir-
cumstances upstream. The front of the wave in advance constitutes a step in depth
(and cross-section area AyF), related to the discharge there, QF, through Equation
13.53. The velocity of the front, wF , advancing on a dry bed is given by
QF
wF = (13.57)
AyF + AzF
with AzF non-zero only if the constant c term in the cumulative infiltration function of
time, Equations 13.11 and 13.15, is not zero. In addition, in Equation 13.43, under-
scoring the importance of upstream conditions in the surface stream, φYL is set to unity,
and φYR is set to zero. Finally, a true kinematic-wave solution, in contrast to other solu-
tions, shows instantaneous reduction to zero depth at the upstream end upon cutoff.
Downstream from this point, the depths fall gradually, leading to a feathered trailing
edge. A finite-difference approximation to Equation 13.54 on an x-t grid can indicate a
nonzero lag time and a gradual approach to zero depth at the upstream end, but this
result is due to numerical-approximation error, and represents neither a true kinematic-
wave solution nor the corresponding zero-inertia or Saint Venant solution.
13.6.11 Applicability of the Kinematic-Wave Assumption
Kinematic-wave theory is useless in level basins or furrows. When S0 = 0, Equation
13.53 becomes untenable, and normal depth does not exist. Instead, the depth at the
inlet of constant inflow rises indefinitely, until cutoff. With nonzero, positive slopes,
Figures 13.11 and 13.12 illustrate conditions under which the kinematic-wave ap-
proximation yields results close to solutions of the full Saint Venant equations, or
zero-inertia approximation. The dimensionless variables depicted are the same as
those in Equations 13.50 and 13.51. To orient the reader in terms of typical dimen-
sioned variables, say in an irrigated border strip, with inflow q0 = 4 Lps/m, S0 = 0.001,
n = 0.04, k = 30 mm/hra, a = 0.5, it follows that the reference slope,YR/XR = 0.0000625,
and the relative slope, S0* = 16. From Figure 13.12, with XR = 671 m, the error in the
time to advance to 100 m using kinematic-wave theory would be about 5% high. Fig-
ure 13.13 gives an indication of how much time must pass, as a function of relative
slope S0*, before normal depth is achieved at the upstream end. Kinematic wave the-
ory assumes that at every point in the surface stream, the depth is instantaneously at
normal for the discharge there.
Clearly, a long run on a steep slope is the ideal application. The full complement of
curves of the type of Figure 13.12 with a in the range, 0.1 < a < 0.9, presented in Ka-
topodes and Strelkoff (1977b) shows that a value of S0*xA* > 20 is required for small
errors in advance. It follows that the ratio of bottom slope to the quotient of normal
depth and stream length should exceed 20, i.e.,
1.00
el
a=0.5
od
m
FN =0.4
ve
S0* =100
Dimensionless time (t*)
wa
ic-
at
m
el
ne
od
Ki
tm
0.10
n
na
Ve
int
Sa
el
od
m
tia
er
-in
ro
Ze
0.01
0.001 0.010 0.10
Dimensionless advance (xA*)

Figure 13.11 Comparison of simulation models in stream advance.
10
Dimensionless time, t*
1.0
a=0.5
n)
tio
0.10
lu
so
ve
wa
ic-
at
em
in
(K
01
00
0.0
*
S 001
S0
0.0 1
01
0
0 *=
0
0
1
10
10
1.
0.
0.
0.0
0.01
0.01 0.1 1.0 10.
Dimensionless advance, xA*
Figure 13.12. Influence of dimensionless bottom slope on advance:
applicability of kinematic-wave model.
S0*=100
1.0
10
Dimensionless upstream depth, y0 *

1.0
0.1
0.01
0.001
0.1 0.0001
S0*=0.00001
a=0.5
0.01
0.01 0.1 1.0 10.
Dimensionless time, t*
Figure 13.13. Growth of upstream depth as a function of relative bottom slope.
1.0
a=0.5
Surface shape factor, ry=y/y0
S0*=100
0.9
_
10
0.8
1.0
0.1
0.7 0.01
S0*=0.00001
0.6
0.01
0.01 0.1 1.0 10.
Dimensionless advance, xA*
Figure 13.14. Evolution of surface stream shape factor as a function of relative bottom slope.
y0 S 0
<
x A 20
(13.58)
for reasonably accurate computations of advance.
13.6.12 Average Depth of Surface Flow in Border Strips and Basins
Dimensionless forms of the zero-inertia equations, 13.43 and 13.52 (FN = 0), were
solved for stream advance with a series of S0* (Katopodes and Strelkoff, 1977b). The
more or less gradual rise in upstream depth with time is noted in Figure 13.13, while
the variation in surface-profile shape factor (Equation 13.26, Section 13.5.1) with ad-
vance is shown in Figure 13.14. Such curves constitute a criterion against which the
assumptions yielding an average stream cross-section (Section 13.5.3) can be meas-
ured. In particular, an examination of the conditions of Figure 13.1 and 13.2 in the
light of Figure 13.13 explains the sometimes significant discrepancy between field
observations and the simplified theory of 13.5.1 and 13.5.3. While Figures 13.13 and
13.14 were drawn for Kostiakov a = 0.5, and cannot be expected to represent condi-
tions for other values, a qualitative appreciation of the variations is nonetheless possi-
ble. Likewise, similar considerations can be expected to hold for furrows. Recently, a
dimensionless database of average surface-water depths in border irrigation has been
developed for use with hydrologic models (Monserrat and Barragan, 1998).
13.6.13 Modeling the Two-Dimensional Surface Irrigation Process
Some surface irrigation flows, for example in wide basins or a border strip with
cross slope, do not possess such a dominant longitudinal direction that transverse ve-
locities and variations in water-surface elevation are negligible. In this case, the basic
physical laws on which simulations are based are still conservation of mass and mo-
mentum with empirical infiltration and flow resistance, but now, these must be ex-
pressed in terms of two lateral coordinate directions, x1 and x2, as well as time t. To
limit the study to two space dimensions, the pressure distribution in the vertical is
again assumed hydrostatic. The velocity distribution in the vertical is assumed essen-
tially uniform, represented by an average flow velocity vector V (x1 , x2 ) oriented at an
angle to the x1 and x2 directions. The drag vector was discussed in Section 13.3.2; in-
filtration is assumed solely a function of opportunity time.
The pertinent equations have been solved in the context of dam-break flooding, a
problem similar to surface irrigation on a greater scale, by finite-element techniques
on a plane surface (Akanbi and Katopodes, 1988), and by finite differences applied to
arbitrary, convenient grid cells over measured topography (Bellos et al., 1991). In di-
rect application to surface irrigation, Playán et al. (1994) discretized the full Saint Ve-
nant equations with isotropic resistance and a level bottom in a centered leapfrog
scheme. For computational convenience, to allow solution of the wave equations eve-
rywhere (in a through calculation) rather than just in the region covered by the ad-
vance of the stream, a small depth was postulated in the nominally dry areas, with the
implied assumption that the advancing front constituted a hydraulic bore. In Playán et
al. (1996), the scheme was modified to accommodate an irregular soil surface.
Strelkoff et al. (2003a) solved zero-inertia equations on irregular topography, in good
agreement with measured advance and surface-depth hydrographs (Clemmens et al.,
2003). The parabolic nature of the governing equations allowed a through computation
in both wet and dry regions without assumption of a leading-edge bore (analogous to
heat propagation in an insulated rod, suddenly heated at one end [Carslaw and Jaeger,
1959]). To assure that water was not calculated as flowing from a higher dry point to a
lower point, a potential consequence of a through computation, water levels were
monitored and flow allowed only from wet cells. Likewise, infiltration was allowed
only from cells with sufficient water depth not to go dry in the course of the time step.
13.6.14 Non-Dimensional Overviews of Surface irrigation

Simulations and Performance
As noted earlier, expression of governing equations, boundary conditions, and solu-
tions in dimensionless form led to simplifications allowing general overviews of re-
sults. Figure 13.4, on a single sheet of graphs, shows all possible advance curves in
border strips, subject to the assumptions that the average surface-water depth is con-
stant with time and that cumulative infiltration follows some power law with time.
Figure 13.12 is one of just nine figures (one for each value of Kostiakov a = 0.1, 0.2,
..., 0.9), which show dimensionless pre-cutoff advance x*(t*) in sloping borders with-
out restrictive assumptions. Surface water depth is allowed to vary with time and dis-
tance in accord with the physical laws governing its behavior and, so, is a function of
relative slope S0*.
Once it is recalled that pre-cutoff advance in a border strip is a function of a, k, n,
q0, and S0, it is evident that to present the same information in dimensioned graphs of
xA(t) with five independent parameters would require shelves of books of graphs, and
the advantage of nondimensional representation is clear. There are many different pos-
sible systems of reference variables, each leading to results that point up one or an-
other aspect of the irrigation hydraulics (see Strelkoff and Clemmens, 1994, for an
extensive discussion of various schemes).
The space-saving opportunities afforded by use of nondimensional variables have
led to development of databases of pre-run simulations encapsulating overviews of a
particular phenomenon. For example, an overview of irrigation performance in level
basins in general has led to the design-aid software, BASIN (Clemmens et al., 1995a)
that allows a user to see virtually instantaneously what physical dimensions and oper-
ating conditions lead to given application efficiencies. At the heart of BASIN are tabu-
lations of DUmin (the same as PAEmin) as a function of Kostiakov a and dimensionless
length and unit inflow rate based on the primary reference variables Dreq, τreq, and Man-
30.0
q 0 (Lps/m)
10.0
Limit line
E
Completion of advance A
C
B
50
E)
60
95 (PA
70
a = 0.5
80
90
1.0
n
100 1000
mi
DU
L (m)
D
L
L* = 2
⎛ Cu ⎞ 3 7 2
⎜ ⎟ Dreq9 τ req3
⎝ n ⎠
Figure 13.15. Potential application efficiency (DU) in level basins. Dimensioned
and dimensionless scales: length, unit inflow. Kostiakov a = 0.5.
(a) limiting length for DU = 80%: 229 m; (b) practical length at DU = 80%: 200 m,
width = 63 m; (c) practical width = 75 m; DU = 81.1%.
ning n, field variables assumed known at the onset of the design process. Figure 13.15,
drawn from Strelkoff and Clemmens (1994) illustrates the case for a = 0.5 (except the
target depth in this case is set to 100 mm, the time to infiltrate that depth, 210 min, and
Manning n = 0.15 m1/6; see sample level-basin design below). Every point on the con-
tours is characterized by a minimum depth of infiltration in the basin of 100 mm. Two
sets of scales are shown: dimensionless ones, constituting the permanent, hardwired
database and ordinary, dimensioned ones for a particular set of field conditions. For a
given field, the two sets differ by constant multipliers (reference variables XRc and QRc,
Strelkoff and Clemmens, 1994). The curve labeled completion of advance represents
the operating condition of cutoff at the time the stream reaches the downstream end.
The curves show that for a given length of basin, performance can be enhanced by
increasing the unit inflow while initiating cutoff before completion of advance, allow-
ing the stream to “coast” to the end after cutoff. The steepening of the DU contours at
high inflow rates, however, shows that there is a limit to this procedure. Clemmens
and Dedrick (1982) suggested practical limitations on the process, arguing that over-
reliance on coasting to the end can leave that end high and dry if field conditions are
not precisely known; their suggested limits are shown as the limit line.
13.6.14.1 Sample level basin design. Drawing on the example physical design (p.
4-1, Clemmens et al., 1995a), a designer seeks to achieve a target DUmin of 80% in
each basin in a field with the aforementioned infiltration and roughness properties,
dimensions 600 m wide and 1200 m long, and an available water supply of 230 Lps.
He or she first looks for the limiting length. Point A in Figure 13.15 depicts that oper-
ating point, corresponding to a length of 229 m (the corresponding unit inflow, q0 =
3.65 Lps/m leads, with the given total available supply of 230 Lps, to a width of 63
m). With 1200 m evenly divisible by 200, that is chosen for the basin length. Point B
in Figure 13.15 is the operating point for a length of 200 m and a DUmin of 80%. This
corresponds to a width of 84 m. Points between D and E in Figure 13.15 (where D is
an indeterminate operating point of very low DUmin, and E represents a very determi-
nate limiting width) are available to the designer for selecting a practical width. Eight
basins, each of 75 m width and 200 m length, L, evenly divide up the available field
and lead to a DUmin of 81.1% (operating point C). The appropriate cutoff time, calcu-
lated from the simple volume balance, tco q0 = LDreq /DUmin, is 134 minutes. The di-
mensionless advance curves stored in the BASIN database yield the corresponding
advance at cutoff, 176 m, a useful guide in inflow management (Clemmens, 1998).
The user-friendly BASIN software allows menu-driven data input (dimensioned)
for field conditions and design objectives and then transparently performs all the con-
versions between dimensional and dimensionless data evident in Figure 13.15, as well
as the necessary interpolations and calculations to satisfy the given target depth by the
minimum in the post-irrigation distribution of infiltrated depths. The user never gets to
see the curves depicted in Figure 13.15; they remain in the background.
13.6.14.2 Sloping border strips with tailwater runoff. Sloping borders with tail-
water runoff are governed by a larger number of input parameters than level basins,
and a different strategy was employed for the dimensionless database lying at the heart
of the design-aid program, BORDER (Strelkoff et al., 1996; Strelkoff and Clemmens,
1996a). The reference variables, QR, YR, XR, TR, are, respectively, the unknown unit
inflow rate, normal depth at the given slope and roughness, normal depth divided by
bottom slope, and XR YR /QR. Several thousand dimensionless zero-inertia simulations
were run (Shatanawi and Strelkoff, 1984; Strelkoff and Shatanawi, 1984) with Kostia-
kov infiltration, i.e., with a range of K* = kTRa/YR, a, and dimensionless cutoff times in
hypothetical borders indefinitely long, so that advance ends before reaching the border
end. “Infiltrated” volume past the actual end of the border strip then closely represents
actual runoff. Within the border length, a stored shape factor for the post-irrigation
infiltration profile allows determination of the minimum or low-quarter-average depth.
In use, e.g. to aid in physical design, a grid of border lengths and widths spanning the
range of interest of the design variables is established, and successive approximations
utilizing interpolations within the data base yield the grid of cutoff times that just sat-
isfy the target depths, as well as a series of performance parameters.
The software presents performance contours as a function of field conditions,
physical design, and operating conditions. In this case, the user does not have to set a
desired performance level a priori; the contours show what the maximum performance
level can be for the given field, available water supply, and target depth, as well as
how to achieve a particular level by selection of design parameters. Figure 13.16a
(Strelkoff and Clemmens, 1996b) shows sample contours of potential application effi-
ciency for a field with the following characteristics:
Kostiakov k = 57.4 mm/hra Slope S0 = 0.001
a = 0.5 Target depth: Dreq = 100 mm
Manning n = 0.10 m1/6 Supply: Q = 60 Lps
Figure 13.16. Contours of low-quarter potential application efficiency as a function of bor-

der length and width for different soils. Kostiakov a = 0.5, Manning n = 0.10 m1/6, bottom
slope = 0.001, available water supply = 60 Lps, target depth (a-c) = 100 mm;
(a) Kostiakov k = 57.4 mm/hra; (b) sandier soil, k = 80 mm/hra;
(c) tighter soil, k = 4 0 mm/hra; (d) k = 40 mm/hra, target depth = 75 mm.
A companion graph (not shown) presents the corresponding pattern of cutoff times
required to achieve the performance. Additionally, contour maps of distribution uniform-
ity, runoff, deep percolation, stream advance at the time of cutoff, and water cost per unit
area can be displayed. The contour plot shows a ridge of relatively high potential appli-
cation efficiency (PAE) designs, with the lower unit inflow rates (larger widths) corre-
sponding to the shorter lengths of border. The slopes near the top of this ridge are gentle,
implying that errors in evaluating field conditions or executing the design may not affect
potential application efficiency very much; closely spaced contours would indicate large
changes in performance with small changes in input parameters. The efficiency ridge
climbs gradually as larger lengths, larger unit inflow rates, and shorter application times
are considered. At the highest and lowest unit inflow rates, however, the contours near
the ridge top are closer together than with more moderate unit inflow rates, say,
around, 1.5 Lps/m (40-m width), indicating increased sensitivity to input conditions.
Figure 13.16b shows the pattern for a sandier soil; with all other conditions remain-
ing the same, the Kostiakov coefficient has increased, so that the one-hour cumulative
infiltration depth is now 80 mm. Qualitatively, the same observations as made for the
original soil hold here, but at shorter lengths. Conversely, Figure 13.16c, with Kostiakov
k = 40 mm/hra, requires very long lengths to achieve comparable potential application
efficiencies (note the change in scales). However, with a decision to irrigate more fre-
quently, and lightly, say with the 75-mm target application of Figure 13.16d, the higher
efficiencies return to the shorter border-strip lengths. Alternative charts (not shown) pro-
vide an overview of the effects of inflow management with a given physical design.
13.6.15 User-Friendly Simulation Software; Transport of Constituents
As noted, the numerical tedium of solving Equations 13.43 to 13.47 et seq., as well
as supplying the variety of initial and boundary conditions for the range of one-
dimensional surface irrigation scenarios encountered has been taken over by user-
friendly, menu-driven software (SRFR 4.06, USDA, 1999; SIRMOD III, Walker,
2003). Figure 13.17 shows a frame of the automation depicting the stream behavior as
calculated by SRFR for a case of furrow irrigation with erosion. Shown are the surface
water profile, sediment transport capacity and load profiles, and infiltration profile
11/4 h into the irrigation. Figure 13.18 is a post-irrigation hydraulic summary of the
event, showing the inflow hydrograph and resulting advance and recession curves,
runoff hydrograph, and final distribution of infiltrated depths. The software also pre-
pares a list of performance indicators, from which the merit of the irrigation can be
judged (Figure 13.19), including, in this case, the off-site transport of sediment,
GS_O. The software, as released by USDA free of charge, comes with some 15 sam-
ple data-input files, illustrating several scenarios.
SIRMOD III contains some basic infiltration estimation capability as well as cer-
tain system design aids. Recent enhancements in SRFR have allowed researchers to
utilize its hydraulic output to study transport of constituents in surface irrigation flows,
e.g., in addition to sediment (Strelkoff and Bjorneberg, 2001), fertilizer in solution
(Strelkoff and Clemmens, 2006, by advection alone; and Zerihun et al., 2005, and
Perea et al., 2005, by advection and turbulent diffusion). Currently underway is a
USDA project (WinSRFR) to port the current stand-alone DOS programs, SRFR,
BASIN, and BORDER along with an irrigation and field-properties evaluation module
into an integrated WINDOWS environment allowing data interchange between the
various modules for input and output (USDA-ARS, 2006).
SURFACE STREAM
Transport capacity
SEDIMENT TRANSPORT
Sediment load
INFILTRATION PROFILE
Infiltration requirement
Figure 13.17. Frame of animation calculated by SRFR during advance with irriga-
tion-induced erosion.
Figure 13.18. Graphical output of SRFR simulation. Hydraulic summary shows inflow hydro-
graph, advance, recession, outflow hydrograph, and post-irrigation infiltration distribution.
Figure 13.19. Graphical output of SRFR simulation of furrow irrigation with

erosion. Performance synopsis shows post-irrigation infiltration distribution and
a series of performance indicators, including off-site sediment transport, GS_O.
13.7 ESTIMATION OF FIELD PARAMETERS

Neither simulation nor design can rationally be undertaken without entry of field
parameters such as infiltration and roughness. Infiltration in particular plays an enor-
mous role in the outcome of a surface irrigation. With the variability found in the field
and limited success of infiltrometers in predicting even local parameters for the condi-
tions of a surface irrigation, it is reasonable to deduce these parameters from observa-
tions of an actual irrigation. In addition, it might be necessary to estimate infiltration
and roughness in real time, i.e., during an irrigation—say during early advance—in
order to predict how successful the irrigation will prove to be, and what to do about it.
Generally, the greater the detail in measurement, the more precise the evaluation
can be. On the other hand, detailed measurements, of surface depths of flow, for ex-
ample, are time consuming and expensive. Thus, attempts are made to estimate field
parameters from simpler measurements, for example, stream advance.
In any case, whenever an estimate is made of field parameters to input into a simu-
lation or design procedure, an additional estimate of likely error in each parameter
should also be made. Then, the procedure can be executed for a range of input parame-
ters, with the totality of results lying within a confidence band. In other words, the
graphical results of simulation or design should not be thought of as a curve with the
thickness of a pencil line, but as a broad brush stroke, with the true results lying some-
where within. It is the task of the engineer to predict not only the centerline behavior
of the brush stroke, but also to estimate its thickness.
13.7.1 Mass Conservation—Volume Balance
Explicit expression of mass conservation forms the basis for all of the direct meth-
ods of parameter estimation. During advance, the inflow volume must equal the vol-
ume infiltrated plus the volume in surface storage, as given in Equation 13.18. In one-
dimensional flow, the surface storage is given by Equation 13.21, while the infiltrated
volume is given by Equation 13.23, which contains both the advance function (xA(t) or
tA(x), which can be measured) and the infiltration function (sought). Many of the direct
parameter-estimation methods are based on a known VZ(t), stemming from measured
inflow and outflow, and AY from either measured surface water depths or an estimate.
13.7.1.1 Post-irrigation determination of infiltration characteristics. A particu-
larly simple and theoretically exact application of the volume balance to evaluation of
the infiltration parameters stems from the post-irrigation analysis of measured advance
and recession curves. With the irrigation completed, all surface water has either
drained off or infiltrated, VY in Equation 13.18 is zero, and VZ is known from measured
inflow, VQ, and outflow, VRO. Thus the average depth of infiltration is known. Indeed,
following an initial suggestion by Merriam (1971), Clemmens (1981) developed equa-
tions for whole-border estimates of Kostiakov k, if the exponent a is known from other
sources. If the basic Kostiakov formula, kτoa, is assumed, in which τo(x) is the infiltra-
tion opportunity time between the advance and recession curves, and if a is taken, for
example, from ring data, k can be found from the equation for the total volume infil-
trated per unit width,
N
VZ = kW ∑τ o aj δx j (13.59)
j =1
in which the border length has been subdivided into N segments and τoj is the average
opportunity time for the jth segment; W is the effective wetted perimeter, assumed con-
stant, so that AZ = Wz.
Clemmens (1981) extended the technique of Equation 13.59 to the branch function
of Equations 13.14 for the common case that the basic infiltration rate has been
achieved within the smallest opportunity time measured. For example in a border strip,
with the post-irrigation depth of infiltration d at any station dependent solely on the
opportunity time there,
d = kτ B a + b(τ o − τ B ) (13.60)
the average depth of infiltration over the border (known from measurement of inflow
and outflow) is
(
d = kτ B a + b τ o − τ B ) (13.61)
With a and b estimated from ring data, the cross-over point τB is found in terms of k
from the third of Equations 13.14 and the result substituted into Equation 13.61, to
yield the following (simplified, corrected) whole-border estimate of k,
a 1− a
⎛b⎞ ⎛ d − bτ o ⎞
k =⎜ ⎟ ⎜ ⎟ (13.62)
⎝a⎠ ⎜ 1− a ⎟
⎝ ⎠
If τB is better estimated from the ring data than b, b follows from
ad
b= (13.63)
(1 − a)τ B + aτ o
obtained after eliminating k from Equation 13.61 and the third of Equations 13.14,
which then yields k.
The details of the infiltration function at opportunity times less than the observed
minimum play a minor role in the distribution of infiltrated water over the given
length of run, as long as the behavior of the function within the observed range of op-
portunity times is reasonably correct. The more nearly uniform are the opportunity
times distributed over the length of run, the more exact will be the evaluation of the
irrigation. At the same time, Clemmens et al. (2001a) caution that a narrow range of
opportunity times in the parameter estimation can lead to large errors in predicted per-
formance for the same soil but under different hydraulic conditions (e.g., slope, inflow
rate, length, etc.); ultimate stream behavior is dependent on the entire infiltration func-
tion of time right up to the end of recession.
It should also be possible to avoid independent assessment of a, if two borders with
the same soil conditions are irrigated, and their variations in opportunity time are ana-
lyzed. In principle, both k and a can be solved by a Newton-Raphson solution of the
nonlinear equation pair. Overconditioning the problem with more irrigations and equa-
tions would allow for a least-squares best fit for both k and a. This multiple-border
post-irrigation approach is based on that of Bouwer (1957), which in the absence of a
functional form for infiltration leads to a table of cumulative infiltration vs. time that
may, like the results of Finkel and Nir (1960) described below, tend to oscillate.
Independent determination of a is also unnecessary with soils that can be character-
ized by membership in empirically defined families (Section 13.3.3), e.g., the SCS
(NRCS, 1984) families or the Merriam and Clemmens (1985) time-rated families.
Both sets are characterized by an implied relationship between k and a, Equations
13.15 or 13.17. Then, a Newton-Raphson solution for k and a would utilize the partial
derivatives of Equation 13.59 (restated in the form G1 = VZ – f [k,a] = 0) with respect
to k and a, and those stemming from a restatement, say, of Equation 13.17,
⎛ 0.675 − a ⎞
2−a⎜ ⎟
G2 = k − 10 ⎝ 0.2125 ⎠ =0
∂G2
=1
∂k
⎛ 0.675− a ⎞
∂G2 2−a⎜ ⎟ 2a − 0.675
0.2125 ⎠
= − ln 10 ⋅10 ⎝ (13.64)
∂a 0.2125
Similar equations can be derived for Equation 13.15.
13.7.1.2 Parameter estimation requiring a closely spaced set of surface-depth
hydrographs. At the other end of the spectrum of required field data, measurement of
a complete set of depth hydrographs along the length of run during the irrigation
yields the time variation of the volume temporarily stored on the surface, VY(t) in
Equation 13.18. Thus, with inflow and outflow known, VZ(t) is also known. Together
with the advance and recession curves, xA(t), xR(t), (a byproduct of the depth hydro-
graphs) this information can be used to deduce the field infiltration parameters.
This class of methods requires the most extensive data collection, but is encum-
bered by the fewest assumptions, and so is the most direct and physically based of all
the techniques reflecting conditions during the entire irrigation. Based almost exclu-
sively on mass conservation, the principle problem lies in extracting the infiltration
parameters (and possibly roughness) from the measured data. The techniques, all of
which assume that VY(t) is a known, measured function of time, differ in the assump-
tions made on the functional forms of the advance and infiltration functions.
In an early development, Finkel and Nir (1960), making no assumptions on infiltra-
tion- or advance-function form, proposed a graphical inversion of Hall’s (1956) tech-
nique (Section 13.5.4). In principle, a tabulated infiltration function of time could be
constructed step by step, a Δz for each Δt. However, unless great care is taken that the
increments in surface volume are accurate, and the construction is limited to a small
number of Δt, the results for Δz begin to oscillate. This instability evidently stems
from the structure of the governing algebraic equation, the inversion of Hall’s. Any
error in measurement or calculation (roundoff, etc.) must be absorbed by the current
calculated Δz, and successive steps in the calculation magnify it. The forward calcula-
tion, advance, is stable with the Hall technique; the reverse, for infiltration, is not.
In a direct, interactive computer procedure (EVALUE), Strelkoff et al. (1999) super-
impose screen plots of measured VZ(t) and a theoretical VZ(t) calculated from estimates
of infiltration parameters in a selected functional form. Variable wetted perimeter is not
considered, and the unknown parameters are assumed constant over the entire length of
run. Numerical integration for the volume under the infiltrated profile is enhanced by
weighting factors based on the ratio of calculated infiltrated depths at each end of a pro-
file segment and on an assumed power-law (with exponent a) for depth vs. distance back
from the advancing front. The parameter values are modified at the keyboard until the
user is satisfied with the fit of the two curves. In addition to the function of time, the
final values of infiltrated volume are also plotted for matching. Various functional forms
can be investigated simultaneously with the parameters in those forms. This brute-force
approach is aimed at selecting those parameters best representative of the entire time
span of the irrigation. A by-product of the calculations is a determination of Manning n
values at the stations, determined from Equations 13.4 and 13.6 by calculating the local
discharge from continuity and setting Sf to the water-surface slope given by smoothed
profiles derived from the measured hydrographs. The average or median constitutes
the representative value of n for the irrigation.
Maheshwari et al. (1988) working with irrigation borders made in extensively
cracked clay soils (k and a of Equation 13.11 were essentially zero in many cases)
automated a similar procedure, by formally minimizing an objective function, Z*,
N
Z * = ∑ (VO i − VC i )2 (13.65)
i =1
with the Hooke and Jeeves pattern-search technique. VOi is an observed infiltrated vol-
ume, based on Equation 13.16 with measured inflow and outflow rates and VY given
by numerically integrated measured surface-water depths. VCi is calculated from cur-
rent values of infiltration parameters and the measured advance curve. N is the number
of times (separated by equal increments) that the comparison is made in the duration
of the study, ending, typically, when recession begins at the upstream end. In applica-
tion, the measured advance and infiltrated-volume functions of time are fitted with
mathematical expressions by regression. A number of different functional forms for
infiltration and advance were investigated. In 1997 (Esfandiari and Maheshwari,
1997), the method was extended to furrows. The modified-Kostiakov parameters for
furrows, K, a, and B, were sought, yielding intake directly (see Section 13.3.3), so no
explicit consideration of wetted perimeter in infiltration was undertaken.
Monserrat (1994) devised a volume balance for the advance phase that utilized
theoretical rather than measured values of average surface water depths. He minimized
an objective function, based on a balance between inflow, surface, and infiltrated vol-
umes in level or sloping borders calculated with measured data, assumed Manning n,
and unknown k and a. Post-irrigation conditions were also included in the objective
function. The optimization procedure allows solution for both k and a, as opposed to
~
the inversion technique of Clemmens (1991), as well as AY ; on the other hand, Mon-
serrat’s static database of solutions limits the analyses to the conditions—broad as
they are—of the pre-run simulations.
Other techniques have been designed to extract the infiltration parameters by one or
another manipulation of the integral in Equation 13.23. The best known of these is the
Lewis and Milne equation.
13.7.2 Lewis and Milne Integral Equation
As noted in 13.5.2, Lewis and Milne (1938) substituted a time integral for the dis-
tance integral in Equation 13.23 through a formal change in variable, leading to Equa-
tion 13.27. Some well-known applications thereof incorporate a simplifying assump-
tion for the functional form of the advance curve. These will be described in Section
13.7.2.1. A technique free of arbitrary assumptions on the advance function is the lin-
ear station advance procedure of Clemmens (1982), who proposed direct evaluation of
the theoretical VZ at a sequence of time levels ti in terms of increments j of the total
profile length at that time, i.e.,
N
VZ (ti ) = ∑ ΔVZ j (ti ) (13.66)
j =1
with the volume increments calculated by assuming a (different) constant rate of ad-
vance over each increment ΔxA. The advance rate in Equation 13.27 can be taken out-
side the integral and expressed as a quotient of differences, so that
tA j
Δx j
ΔVZ j (ti ) = ∫ AZ (ti − t A ) dt A (13.67)
Δt A j t A j −1
in which ΔtAj = tA(xj) – tA(xj-1). Now, with any given functional form for z(τ), the inte-
gral in Equation 13.67 can be evaluated for each j = 1,..., N, and the results summed
for a theoretical Vz(ti) in terms of the infiltration-function parameters. For example,
with the modified-Kostiakov formula of Equation 13.11, Equation 13.67 appears as
ΔVZ j ( ti ) =
⎡ ⎛
{
Δx j ⎢ ⎜ ti − t A j −1
k⎜
} {
a +1
− ti − t A j }
a +1 ⎞
⎟
⎢ ⎟
Δt A j ⎢ ⎜ a +1 ⎟
⎣ ⎝ ⎠
(13.68)
+ b⎜
⎛
{ 2
} { 2 ⎞⎤
⎜ ti − t A j −1 − ti − t A j ⎟⎥ }
⎟⎥ + cΔx j
⎜ 2 ⎟⎥
⎝ ⎠⎦
as can be followed in Figure 13.3. The first term of Equation 13.68 is identical to the
result presented by Clemmens (1982).
With the branch function (Equations 13.14), if t – tA < τB, the first and third terms
of Equation 13.68 are appropriate. For t – tA > τB,
ΔVZ j = b
⎛
{ } {
2
Δx j ⎜ t − t A j −1 − t − t A j } ⎞⎟ + c
2
(13.69)
⎜ ⎟ B Δx j
Δt A j ⎜ 2 ⎟
⎝ ⎠
The form of Equations 13.68 and 13.69 allow for spatial variability in the infiltra-
tion parameters, but, typically, the equations would be used to estimate whole-border
values. With two parameters, k and a in a power law, a minimum of two time levels is
required for solution. For the four parameters in the Modified Kostiakov formulation
of Equation 13.68, it would follow that four simultaneous equations of the type of
Equation 13.67, at each of four time levels, would be the minimum. Typically, many
more are used, with a best fit sought.
In a computational simplification, Clemmens (1982) guesses a, say 0.5, allowing a
direct, rather than simultaneous, solution for k at each t up to ti, with the same averag-
ing of opportunity times leading to a point on the plot. The slope of the line, on loga-
rithmic paper, provides a better guess for a, and the process is repeated until a con-
verges (with two trials typically necessary).
With the infiltration parameters found, Equations 13.68 or 13.69 provide the
changes in subsurface storage in each segment in each time step, while the depth hy-
drographs at the stations yield the changes in surface storage. A volume balance be-
tween stations and time steps yield the discharges at each station, while the profiles
derived from the water surface elevations yield the water surface slope. Thus Manning
n can be determined at each station at each time level. With the median values at a
station found over time, the median or distance average, as in Clemmens (1982),
yields a representative value for the border strip.
If the advance points leading to Equation 13.68 had been found at constant incre-
ments of time (either measured or interpolated) as in Figure 13.3, Equation 13.68, with
the Kostiakov power law, z = kτa, leads to a numerical evaluation of rZ through the
summation,
N Δx ⎛
V j {i − j + 1}
a +1
− {i − j}a +1 ⎞⎟
rZ i = aZ i = ∑ ⎜ (13.70)
kti x A i j =1 x A i ⎜⎝ i a (a + 1) ⎟
⎠
in which rZ is the ratio of average depth of infiltration to the upstream depth. The rela-
tionships tAj = jΔt, ti = iΔt, tAj – tAj-1 = Δt, etc. were used in the development of Equa-
tion 13.70.
13.7.2.1 Power-law advance. Significant simplification in evaluating the Lewis
and Milne integral in Equation 13.27 follows assumption of a functional form for ad-
vance as well as for infiltration, as discussed in Section 13.5.6 for power-law advance.
In fact methods for determining infiltration use the same concepts as were used for
determining advance (Sections 13.5.2 to 13.5.6).
The two-point method (Elliot and Walker, 1982)—In a popular variation of the
Christiansen et al. (1966) approach, advance time is measured only twice, typically
halfway to the end of the field and at field end. The method assumes power law advance
~
and, rather than base AY on the degree to which VZ follows a power law, makes an inde-
~
pendent estimate for AY based on assumed normal depth for the given cross-section, bot-
tom slope, Manning n, and inflow rate and an assumed constant shape factor rY, as
~
AY = rY AY 0 (13.71)
Then VY in Equation 13.18 is known at both xA. The two-point method shares with
similar methods any inaccuracies stemming from estimation of the average cross-
sectional area of the surface stream (Strelkoff et al., 2003b). It also shares the small
violation of mass conservation stemming from assumption of power laws for both
infiltration and advance along with a constant average cross-sectional flow area during
advance. Concern that one or both of the two advance measurements may be faulty
can be alleviated by plotting a line of best fit to the advance curve on logarithmic pa-
per and determining the two points from that. The two-point method is easily extended
to accommodate non-zero b values in Equation 13.11, once these are determined inde-
pendently. Note, however, the caveat regarding Equation 13.13. The outcome of the
method, for the two points at full length, L, and half length, L/2, are the Kostiakov k
and a, as follows:
log 2 L
h= f = h (13.72)
⎛ t ⎞ tL
log ⎜⎜ L ⎟⎟
⎝ tL / 2 ⎠
⎛ V ⎞
log ⎜⎜ Z L ⎟⎟
V
⎝ Z L/2 ⎠ VZ L
a+h = g= (13.73)
⎛ t ⎞ t L a+h
log ⎜⎜ L ⎟⎟
⎝ tL/ 2 ⎠
g
and k= (13.74)
f rZ (a, h)
Analogous equations yield K if desired for a direct empirical determination of infil-
trated volume per unit length.
Figures 13.20a,b illustrate the effect of an incorrect estimate of average surface-
stream cross-section for two different bottom slopes, (a) S0 = 0.0005, (b) S0 = 0.005.
The two-point method was applied to a simulated border irrigation. The infiltration
function input to the simulation (a time-rated family member requiring 4 h to infiltrate
100 mm) is labeled SRFR; the simulation then yielded both an advance curve (t2 = tL
shown) and rY, slowly varying with time. A representative value for rY was selected
for use in the two-point method, and the resulting infiltration function is labeled “rY
correct”. Then the representative value was changed by ±10% with the results shown.
As can be expected, with the surface volume at the small slope comprising a greater
fraction of the total inflow than at the larger slope, an error in rY leads to greater errors
in estimated infiltration with the small slope than with the large slope. In both cases
the deviations from the correct infiltration increase significantly for times extrapolated
beyond t2 (Strelkoff et al., 2003b).
In the event of a level field, normal depth has no meaning, but an extension of the
method to this case has been proposed (Zerihun et al., 2004). Instead of assuming AY0
constant in Equation 13.71, it is allowed to grow with xA in accord with Equations 13.4
and 13.6 and the approximation that Sf = y0/xA. Then, for a given furrow cross-
sectional shape, AY0 is implicit in the relationship,
AY 0 2 R0 4 / 3 y0 = x A Q0 2 n 2 (13.75)
with the right-hand side known at both advance distances (selected in Zerihun et al.,
~
2004, as L/3 and 2L/3). Then AY follows from Equation 13.71. To extend the method
to non-zero B requires depth measurements as the ponded water surface falls after
cutoff. B would be given by the measured rate of fall and the assumption that at this
point in time infiltration rate does not vary with location in the basin.
Alvarez (2003) utilized the assumptions of the two-point method to predict the ad-
vance and Kostiakov K for furrow discharges other than the particular one at which
the field test was run, to account for the different wetted perimeter. The technique is
based on the further assumption, supported by field observations, that with different
inflow rates, the exponent in the power law for advance varies very little.
ry correct
100.
ry 10% high
Border strip SRFR
Infiltration depth (mm)

S0=0.0050
t2
50.
ry 10% low
0.
(a) 0.00 1.00 2.00 3.00 4.00
Time (hours)
100.
Border strip
S0=0.0005
Infiltration depth (mm)
SRFR
ry correct
ry 10% low
50.
t2
ry 10% high
0.
(b) 0.00 1.00 2.00 3.00 4.00
Time (hours)
Figure 13.20. Effect on infiltration prediction of errors in assumed shape factor rY
for surface water profile. Two-point method. t2: advance time;
SRFR: z(τ) in simulation. (a): S0 = 0.0005; (b): S0 = 0.005.
One-point methods—One-point methods require measurement of the time to ad-

vance to a single point, typically field end. It is also assumed in this method, as usually
applied, that the flow channel is sufficiently steep to achieve normal depth at the up-
stream end by the end of advance, and that rY is known. This allows solution of the
Lewis and Milne (1938) integral equation. And, while two measured advance points
are generally required to yield independent values of the two Kostiakov parameters,
additional assumptions can yield the pertinent parameters with just one advance point.
In the original one-point method (Shepard et al., 1993), the reasonable assumption of a
Philip infiltration function, z = Sτ1/2 + Aτ (Philip, 1969), and a quite restrictive ad-
vance function, x = f·t1/2, lead to both constants, S and A, from a measurement of the
time to reach furrow end. The concurrent assumptions, instead, of NRCS (1984) (SCS)
infiltration families and power-law advance, x = f·th, with f and h site-specific con-
stants, lead approximately to the pertinent family, once the time for the stream to reach
a specified point is given (Valiantzas et al., 2001). This is made possible by the fact
that each family is defined by a particular value of k and a particular value of a; thus k
is a function of a (Equation 13.15). In an alternate method, introduced because some
non-cracking soils are fit better by the time-rated families of Merriam and Clemmens
(1985), a similar development (see Strelkoff et al., 2003b) also leads to the Kostiakov
k and a from a single advance point. In this connection, the empirical relationship be-
tween k and a (Equation 13.17) depends on the actual wetted area, and consequently,
in furrows k should be based on wetted perimeter rather than on furrow spacing. The
effects of errors in surface volume estimation are also discussed.
13.7.3 Direct Inversion of the Hydrodynamic Flow Equations
Clemmens (1991) was able to invert the hydrodynamic unsteady-flow equations
governing the irrigation-stream flow to the extent of one unknown field parameter.
With measured time and distance increments given in the double-sweep algorithm for
solving the linearized equations of the Newton-Raphson scheme, the column of coef-
ficients stemming from the gradients of depth and discharge with respect to time (see
Section 13.6.9) is replaced by the gradients with respect to some field parameter
(Clemmens, 1991). Then that field parameter comes out in the solution at each time step.
The method requires a priori specification of all the remaining field parameters,
usually, Kostiakov a and Manning n. At each time step, the resultant k is used to cal-
culate an opportunity time for the computed average depth of infiltration. Then the
resulting plot is used to fit a new Kostiakov function, allowing recalculation with the
more current value of a. The original paper provides a wealth of detailed comparisons
of results from various combinations of assumptions in the method.
13.7.4 Parameter Optimization through Repeated Simulation
When the unknown field parameters are so embedded in the pertinent equations
that the latter cannot be inverted to yield them, they can be sought in a program of
successive approximation. This typically involves repeated simulations, with changing
values of the parameters, in a formal search procedure known as optimization, aimed
at minimizing the discrepancies between simulated and measured values of selected
quantities, such as advance, water depths, runoff hydrographs, etc.
General concerns and approaches to securing an optimum can be found in Press et
al. (2001), which explains the concepts in the following discussion with more detail
and background information. In particular, they first discuss the problems encountered
and techniques available in a one-dimensional search for a minimum. The global, or
true, minimum of a general objective function F(x) of one independent variable x can
certainly be found, within a pre-specified permissible range of the independent vari-
able, by brute force, i.e., calculating F over sufficiently small increments, δx, that a dip
lower than all the others will not be missed. The fundamental problem in practical
one-dimensional optimization is in devising a technique that will lead to the global
minimum via a reasonable number of calculations. In multi-dimensional optimization,
in which there are a number of independent parameters, described, say by the vector p,
the additional problem to be confronted is in the selection of directions in which to
change p from a current estimate, as well as in how much to change it, so as to reach a
global minimum, all within the permissible range of p. In general, a good optimization
algorithm is robust, i.e., will inexorably converge to the global minimum, bypassing
local minima, regardless of the starting guesses for the independent parameters or the
nature of the objective function. It will be economical of computation and will not
require excessive computer storage.
In the gradient methods, the search proceeds in a series of one-dimensional steps,
with the direction in each step selected in response to the gradient of F (calculated, for
example, by simulation at two neighboring p). In the step from iteration i to i + 1,
p( i +1 ) = p( i ) − d = p i − α ( i )R ( i )g( i ) (13.76)
the change in p is made in the direction of the vector d, given by the matrix R premul-
tiplying the gradient g, and of length given by the scalar α. R = I, the identity matrix,
corresponds to the method of steepest descent, but this can be extremely inefficient
(Press et al., 2001), and other choices, twisting the direction of correction away from
the gradient, prove more suitable in the search for the bottom of a long, narrow valley
with spurious local minima representing the objective function (as visualized with two
free p components). A logical selection for R in the search for the minimum of the F
surface is the inverse H of the Hessian matrix for F,
−1
⎛ ∂2F ⎞
H=⎜ ⎟ (13.77)
⎜ ∂pi ∂p j ⎟
⎝ ⎠
for that is an indication of the local rate at which the gradient is approaching zero, i.e.,
where F is at a minimum. The expectation is that, except for points at the boundaries
of the permissible domain of independent variables, the minimum of the objective
function corresponds to a zero gradient there. Direct calculation of H is inconvenient,
partly because of the difficulty in determining the second derivatives and partly be-
cause of the need for inverting a potentially large matrix. A number of numerical al-
ternatives have been proposed, as noted in subsequent sections.
13.7.4.1 Optimization to match advance. The issue of how much information
about the field parameters can be deduced from measurements of advance alone (ob-
servability of advance) was examined theoretically, in the context of a linearized,
zero-inertia model, by Katopodes (1990). He concluded that resistance and infiltration
have indistinguishable effects on the rate of advance, but that if one is given, the other
can be found. With three field parameters controlling advance, say, Kostiakov k and a
and Manning n, if any two are known, the third can unequivocally be deduced from the
observed advance. Two parameters can be deduced if some measured surface depths are
available for comparison. The optimization can be difficult: in the case of an objective
function based on varying n and a, several local minima are observed in a very narrow
valley, each of which is higher than the global minimum. All three parameters can be
estimated with measurements of a single depth profile (Yost and Katopodes, 1998).
A series of methods, also, in a sense, constituting an inversion of the advance prob-
lem, consists of repeated simulations with a mathematical model controlled by one or
another search procedure. Walker and Busman (1990), with the aim of evaluating real-
time infiltration, i.e., while an irrigation event is in progress, minimized an objective
function, Y, based on measured and calculated advance,
m
Y= ∑ (T A j − t A j ) 2 (13.78)
j =1
in which the TAj are measurements of advance time to specified locations xj, while the
tAj are simulated times with specific values for the infiltration parameters, in a current
estimation. The roughness parameter is assumed known. The number of comparison
points, m, increases as the irrigation progresses, providing ever more data on which to
base the infiltration parameters. At any step of the irrigation m, the objective function
is minimized (the infiltration parameters are optimized) in successive approximations
via the downhill simplex method (Press et al., 2001), not requiring calculation of the
gradient of the objective function, but only of the objective function itself, at N + 1
estimates of the N infiltration parameters at each step of the iterative process, i.e.,
m
Yi = ∑ (TA j − t Ai j )2 i = 1,..., N +1 (13.79)
j =1
With two infiltration parameters k and a (N = 2), there are three vertices to the sim-
plex, in general a polyhedron with N + 1 vertices in N-dimensional parameter space,
reducing, in the case of Kostiakov parameters, to a triangle. With three, in a search for
best values of k, a, and b, the simplex is a tetrahedron in three-dimensional space.
Each vertex is characterized by a set of values for the N desired parameters. The sim-
plex method consists in a strategy (Press et al., 2001) for changing the locations of the
vertices in a manner designed to bring the optimum point into the interior of a very
small polyhedron.
Azevedo (1992) found the multidimensional modified Powell method for determin-
ing successive directions for solution-vector change coupled to the one-dimensional
line minimization of Brent at each direction change, together with minimum bracket-
ing (all in Press et al., 2001) much faster and more reliable than the downhill simplex
method. He also found that smaller minimums of the objective function prevailed
when all three Kostiakov-Lewis parameters, k, a, and b, were sought in the search—
field determinations of b, for example, were often in considerable error. Of note, the
solutions proved not unique, with different ultimate parameter values resulting from
different starting values. At the same time, advance calculated with the different sets
did not differ much.
Bautista and Wallender (1993) used the Marquardt method (Press et al., 2001), a
standard technique for non-linear least-squares curve fitting, to minimize the objective
function,
2
m ⎛ T A j − t A (p ) j ⎞
F (p ) = ∑ ⎜ ⎟ (13.80)
⎜
j =1 ⎝ σj ⎟
⎠
in which m is the number of comparison stations, TAj is measured advance to a given
station xj, while tAj is simulated, with field-parameter vector p having components, k,
a, and b; σ is a weighting factor, generally set to unity, to account for uncertainty in a
measurement. Roughness was assumed known, and ultimately, b also (field-measured
steady state infiltration rate), as the Marquardt algorithm exhibited difficulty in con-
verging to optimum values of three parameters. An alternate objective function,
couched in terms of advance velocity, proved more useful in identifying spatially
varying infiltration, but ultimately, those results were inconclusive.
All of these methods provide more reliable infiltration data if advance data from the
entire irrigation is used, rather than any portion of the advance curve, e.g., in an at-
tempt to discern the infiltration parameters as the irrigation progresses.
13.7.4.2 Optimization to match outflow hydrographs. Infiltration parameters
suitable for the entire duration of an irrigation should yield simulated runoff in agree-
ment with measured values. Accordingly, Ley (1978), incorporated a match of total
runoff volumes into his extension of Christiansen’s et al. (1966) match of advance
curves.
Walker (2005) developed a systematic procedure for selecting appropriate values of
K, a, B, and Manning n in a simulation that would match measured outflow hydro-
graphs, when measured inflow, furrow cross-sections, and bottom slope were given.
The search procedure differs from classical optimization of several parameters simul-
taneously in that the parameters are sought sequentially, albeit iteratively. The concept
of sequential determination of the parameters is based on the general observation in a
wide range of simulations that the four sought after parameters affect the results in
different ways. Specifically, the K value appears the most influential in determining
advance time (the time outflow begins), a is the most important factor affecting the
shape of the outflow hydrograph, and B is the most important in determining the abso-
lute values of runoff rate. The roughness n was seen as most influential in determining
the time recession ends (the time outflow ends). The necessity of iteration stems from
the imprecision of these observations and the interactive effects of parameter values.
In the multiple simulations comprising the method, the measured furrow geometry
and inflow hydrograph are augmented by successive approximations to the four pa-
rameters. Starting guesses follow from Equation 13.13 for B, the two-point method
(Equations 13.72–13.74) for K and a, and values in the literature for n. In the sequence
of nested optimizations, K is adjusted in the innermost nest of simulations to make
computed and measured advance times equal. The secant method provides the succes-
sive approximations as changes in tL are nearly proportional to changes in K. In the
next outer nest, a Fibonacci search (Kahaner et al., 1989) for B minimizes rmsRO the
root mean square of the differences between computed and measured runoff rates,
2
1 m⎛ ⎞
rms RO = ∑ ⎜ QRO i M − QRO i S
m i =1 ⎝
⎟
Kˆ , aˆ , Bˆ , nˆ ⎠
(13.81)
where m is the number of points in the hydrograph, the subscript M refers to measured
values, S refers to simulated values, and the hat above a variable, e.g., K̂, refers to the
current value that satisfies the search. At each value of B found in the process, the in-
ner nest of searches adjusts K to match the advance time. The next ring of the search is
for a, which also minimizes the root-mean square of the computed and measured val-
ues of runoff. At each a value, B̂ and K̂ are re-computed. The final, outer ring deter-
mines n̂ to match computed and measured recession times. A key advantage of the
method is that all data collection during the irrigation is confined to the inlet and outlet
of the furrow, with no need to enter the wet field.
13.7.4.3 Matching measured depths. An estimation technique based on a few se-
lected depth measurements at various locations or times during advance (not enough
to define stream volume, but only to compare with computed depths) was presented by
Katopodes et al. (1990). Several similar gradient procedures were applied by Playán
and Garcia-Navarro (1997) in a search for the n-k parameter pair and also the a-k pair,
with the third component of the parameter vector held fixed.
A quite different procedure for estimating whole-border infiltration and roughness
for the advance phase was developed by Valiantzas (1994) utilizing a combination of
measurements and theoretical (zero-inertia) simulations (Section 13.6.7 et seq.). The
field parameters sought are the Kostiakov k and a and Manning n (actually its equiva-
lent, normal depth, for the given inflow and bottom slope). Both the advance trajectory
xA(ti), i = 1,..., N, and a depth hydrograph yR(tj), j = 1,..., M, in a reference point xR at or
near the inlet end of the border, are measured.
REFERENCES
Abramowitz, M., and I. A. Stegun, eds. 1964. Handbook of Mathematical Functions.
Appl. Math. Series 55, June. Gaithersburg, Md.: Nat. Bur. Standards.
Akanbi, A. A., and N. D. Katopodes. 1988. Model for flood propagation on initially
dry land. J. Hydr. Eng. 114(7): 689-706.
Alazba, A. A. 1999. Explicit volume balance model solution. J. Irrig. Drain. Eng.
125(5): 273-279.
Alazba, A. A., and D. D. Fangmeier. 1995. Hydrograph shape and border irrigation
efficiency. J. Irrig. Drain. Eng. 121(6): 452-457.
Alazba, A. A., and T. S. Strelkoff. 1994. Correct form of Hall technique for border
irrigation advance. J. Irrig. Drain. Eng. 120(2): 292-307.
Alvarez, J. A. R. 2003 Estimation of advance and infiltration equations in furrow
irrigation for untested discharges. Agric. Water Mgmt. 60(3): 227-239.
Azevedo, C. A. V. 1992. Real-time simulation of the inverse furrow advance problem.
Ph.D. diss. Logan, Utah: Utah State Univ.
Bautista, E., and W. W. Wallender. 1993. Identification of furrow intake parameters
from advance times and rates. J. Irrig. Drain. Eng. 119(2): 295-311.
Bellos, C. V., J. V. Soulis, and J. G. Sakkas. 1991. Computation of two-dimensional
dam-break induced flows. Adv. in Water Resour. 14(1): 31-41.
Bjorneberg, D. L., T. J. Trout, R. E. Sojka, and K. K. Aase. 1999. Evaluating WEPP
predicted infiltration, runoff, and soil erosion for furrow irrigation. Trans. ASAE
42(6): 1733-1741.
Blair, A. W., and E. T. Smerdon. 1985. Effect of wetted perimeter on infiltration in
furrows. In Proc. Spec. Conf. Irrig. and Drain. Div., 162-169. Reston, Va.: Ameri-
can Soc. Civil Engineers.

Blair, A. W., E. T. Smerdon, and J. Rutledge. 1984. An infiltration model for surge
flow irrigation. In Proc. Spec. Conf. Irrig. and Drain. Div., 691-700. Reston, Va.:
Boas, M. L. 1966. Mathematical Methods in the Physical Sciences. New York, N.Y.:
John Wiley and Sons.
Bondurant, J. 1971. Proc. of the Annual Report to Regional Research Committee W-65
on the Hydraulics of Surface Irrigation. Washington, D.C.: USDA-ARS.
Bouwer, H. 1957. Infiltration patterns for surface irrigation. Agric. Eng, 38(9): 662-
664, 676.
Bradford, S., and N. D. Katopodes. 1998 Nonhydrostatic model for surface irrigation.
J. Irrig. Drain. Eng. 124(4): 200-212.
Carslaw, H. S., and J. C. Jaeger. 1959. Conduction of Heat in Solids. New York, N.Y.:
John Wiley & Sons.
Christiansen, J. E., A. A. Bishop, F. W. Kiefer, Jr., and Y. S. Fok. 1966. Evaluation of
intake rate constants as related to advance of water in surface irrigation. Trans.
ASAE 9(5): 671-674.
Clemmens, A. J. 1981. Evaluation of infiltration measurements for border irrigation.
Agric. Water Mgmt. 3: 251-267.
Clemmens, A. J. 1982. Evaluating infiltration for border irrigation models. Agric.
Water Mgmt. 5: 159-170.
Clemmens, A. J. 1991. Direct solution to surface irrigation advance inverse problem.
J. Irrig. Drain. Eng. 117(4): 578-594.
Clemmens, A. J. 1998. Level basin design based on cutoff criteria. Irrig. Drain.
Systems 12: 85-113.
Clemmens, A. J. and Dedrick, A.R. 1982. Limits for practical level basin design. J.
Irrig. Drain. Div., ASCE 108(IR2): 127-141
Clemmens, A. J., and A. R. Dedrick. 1984. Multi-purpose head detection and
monitoring unit. Trans. ASAE 27(6): 1825-1828.
Clemmens, A. J., A. R. Dedrick, and R. J. Strand. 1995a. BASIN: A Computer
Program for the Design of Level-Basin Irrigation Systems. Version 2.0. WCL
Report #19. Phoenix, Ariz.: USDA-ARS, U.S. Water Conservation Laboratory.
Clemmens, A. J., D. E. Eisenhauer, and B.L. Maheshwari. 2001a. Infiltration and
roughness equations for surface irrigation: How form influences estimation. ASAE
Paper No. 01-2255. St. Joseph, Mich.: ASAE.
Clemmens, A. J., and K. H. Solomon. 1995. Procedures for combining distribution
uniformity components. In Proc., First Int’l. Conf. Sponsored by the Water
Resources Eng. Div., 1531-1535. Reston, Va.: American Soc. Civil Engineers.
Clemmens, A. J., T. S. Strelkoff, and C. M. Burt. 1995b. Defining efficiency and
uniformity: Problems and perspectives. In Proc. First Int’l. Conf., Water Resources
Eng. Div., 1521-1525. Reston, Va.: American Soc. Civil Engineers.
Clemmens, A. J., T. S. Strelkoff, and E. Playan. 2003. Field verification of two-
dimensional surface irrigation model. J. Irrig. Drain. Eng. 129(6): 402-411.
Clemmens, A. J., T. L. Wahl, M. G. Bos, and J. A. Replogle. 2001b. Water
Measurement in Flumes and Weirs. Publication #58. Wageningen, The

Netherlands: Int’l. Inst. for Land Reclamation and Improvement (ILRI).
Collis-George, N. 1974. A laboratory study of infiltration advance. Soil Sci. 117(5):
282-287.
Cunge, J. A., F. M. Holly, and A. Verwey. 1980. Practical Aspects of Computational
River Hydraulics. Boston, Mass.: Pitman Advanced Publishing Program.
Dedrick, A. R. 1991. Surface-drained level basins for rice production. In Proc. Spec.
Conf. Irrig. and Drain. Div. Reston, Va.: American Soc. Civil Engineers.
Dudley, S. J., S. R. Abt, C. D. Bonham, and C. Fischenich. 1996. Efficiency of
estimating vegetative density using point frame. In Proc. North Am. Water and
Environ. Congr. Reston, Va.: American Soc. Civil Engineers.
Dunn, C., F. Lopez, and M. Garcia. 1996. Vegetation-induced drag: An experimental
study. In Proc. North Am. Water and Environ. Congr. Reston, Va.: American Soc.
Civil Engineers.
Elliot, R. L., and W. R. Walker. 1982. Field evaluation of furrow infiltration and
advance functions. Trans. ASAE 25: 396-400.
Enciso-Medina, J., D. Martin, and D. Eisenhauer. 1998. Infiltration model for furrow
Esfandiari, M., and B. L. Maheshwari. 1997. Application of the optimization method
for estimating infiltration characteristics in furrow irrigation and its comparison
with other methods. Agric. Water Mgmt. 34: 169-185.
Fangmeier, D. D., A. J. Clemmens, M. El-Ansary, T. S. Strelkoff, and H. E. Osman.
1999. Influence of land leveling precision on level-basin advance and performance.
Trans. ASAE 42(4): 1019-1025.
Finkel, H. J., and D. Nir. 1960. Determining infiltration rates in irrigation borders. J.
Geophys. Res. 65: 2125-2131.
Garabedian, P. 1964. Partial Differential Equations. New York, N.Y.: John Wiley &
Sons.
Green, W. H., and G. Ampt. 1911. Studies of soil physics. Part I: The flow of air and
water through soils, J. Agric. Sci. 4: 1-24.
Hall, W. A. 1956. Estimating irrigation border flow. Agric. Eng. 37: 263-265.
Hart, W. E., D. L. Bassett, and T. S. Strelkoff. 1968. Surface irrigation hydraulics:
Kinematics. J. Irrig. Drain. Div., ASCE 94(IR4): 419-440.
Hartley, D. M. 1992. Interpretation of Kostiakov infiltration parameters for borders. J.
Irrig. Drain. Eng. 118(1): 156-165.
Heermann, D. F., and N. A. Evans. 1969. Estimate of resistance from roughness
measurements. Unpublished contribution from the Northern Plains Branch, Soil
and Water Conservation Res. Div., ARS, USDA and the Colo. Agric. Expt. Stn.,
Fort Collins, Colo. to W. Reg. Res. Proj. W-65, USDA-CSRS.
Hromadka II, T. V., and J. J. DeVries. 1988. Kinematic wave routing and
computational error. J. Hydr. Eng. 114(2): 207-217.
Hutson, S. S., N. L.Barber, J. F. Kenny, K. S. Linsey, D. S. Lumia, and M. A Maupin.
2004. Estimated Use of Water in the United States in 2000. USGS Circular 1268.
Available at: water.usgs.gov/pubs/circ/2004/circ1268/.
ISED. 1989. SIRMOD, Surface Irrigation Simulation Software. User’s Guide. Logan,
Utah: Irrig. Software Eng. Div., Dept. Agric. and Irrig. Eng., Utah State Univ.
Israelsen, O. W. 1932 Irrigation Principles and Practices. New York, N.Y.: John
Wiley & Sons.

Izuno, F. T., and T. H. Podmore. 1985. Kinematic wave model for surge irrigation
research in furrows. Trans. ASAE 28(4): 1145-1150.
Kahaner, D., C. Moler, and S. Nash. 1989. Numerical Methods and Software.
Englewood Cliffs, N.J.: Prentice Hall.
Katopodes, N. D. 1990. Observability of surface irrigation advance. J. Irrig. Drain.
Eng. 116(5): 656-675.
Katopodes, N. D., and T. S. Strelkoff. 1977a. Hydrodynamics of border irrigation:
Complete model. J. Irrig. Drain. Div., ASCE 103(IR3): 309-324.
Katopodes, N. D., and T. S. Strelkoff. 1977b. Dimensionless solutions of border
irrigation advance. J. Irrig. Drain. Div., ASCE 103(IR4): 401-417.
Katopodes, N. D., J. H. Tang, and A. J. Clemmens. 1990. Estimation of surface
irrigation parameters. J. Irrig. Drain. Eng. 116(5): 676-695.
Kemper W. D., W. H. Heinemann, D. C. Kincaid, and R. V. Worstell. 1981.
Cablegation: I. Cable controlled plugs in perforated supply pipes for automatic
furrow irrigation. Trans. ASAE 24(6): 1526-1532.
Kincaid, D. S., and J. L. Stevens. 1992. CABLE: A Design Program for the
Cablegation Automatic Surface Irrigation System. Kimberly, Idaho: USDA-ARS.
Kostiakov, A. N. 1932. On the dynamics of the coefficient of water percolation in
soils and on the necessity for studying it from a dynamic point of view for purposes
of amelioration. In Trans., 6th Comm., Int’l. Soc. of Soil Sci. (In Russian) Part A:
17-21. Moscow: ISSS, Russian Federation.
Kouwen, N., R-M. Li, and D. S. Simons. 1981. Flow resistance in vegetated
waterways, Trans. ASAE 24(3): 684-690, 698.
Kruse, E. G., C. W. Huntley, and A. R. Robinson. 1965. Flow resistance in simulated
irrigation borders and furrows. Conserv. Res. Report 3: 1-56. Washington, D.C.:
USDA-ARS.
Lenau, C. W. 1969. Discussion of Hart et al. (1968), Surface irrigation hydraulics:
Kinematics. J. Irrig. Drain. Div., ASCE 95(IR4): 624-627.
Lewis, M. R. 1937. The rate of infiltration of water in irrigation practice. Trans. AGU
18: 361-368.
Lewis, M. R., and W. E. Milne. 1938. Analysis of border irrigation. Agric. Eng. 19:
267-272.
Ley, T. W. 1978. Sensitivity of furrow irrigation performance to field and operation
variables. M.Sc. thesis. Fort Collins, Colo.: Dept. of Ag. and Chem Eng., Colorado
State Univ.
Lighthill, M. J., and G. B. Whitham. 1955. On kinematic waves. I: Flood movements
in long rivers. Proc. Royal Soc. of London, Series A 229: 281-316.
Maheshwari, B. L. 1992. Suitability of different flow equations and hydraulic
resistance parameters for flows in surface irrigation: A review. Water Resour. Res.
28(8): 2059-2066.
Maheshwari, B. L., A. K. Turner, T. A. McMahon, and B. J. Campbell. 1988. An
optimization technique for estimating infiltration characteristics in border
irrigation. Agric. Water Mgmt. 13(1): 13-24.
Merriam, J. L. 1971. Adjusting cylinder infiltrometer data for field use. ASAE paper
No. 71-13. St. Joseph, Mich.: ASAE.
Merriam, J. L., and A. J. Clemmens. 1985. Time rated infiltrated depth families. In
Development and Management Aspects of Irrigation and Drainage, Spec. Conf.

Proc., Irrig. and Drain. Div., 67-74. Reston, Va.: American Soc. Civil Engineers.
Monserrat, J. 1994 Solucion al Problema inverso del riego por tablares mediante un
modelo hidrologico mixto, Tesis Doctoral. Universidad de Leida, Escuela Tecnica
Superior de Ingenieria Agraria.
Monserrat, J., and J. Barragan. 1998. Estimation of the surface volume in hydrological
models for border irrigation. J. Irrig. Drain. Eng. 124(6): 238-247.
NRCS. 1984. Section 15: Irrigation, Chapter 5: Furrow Irrigation. In National
Engineering Handbook. Washington, D.C.: USDA NRCS.
Palomo, M. J., N. A. Oyonarte, L. Mateos, and J. Roldan. 1996. Infiltracion en riego
por surcos mediante pulsaciones intermitentes. In Proc. XIV Congreso Nacional de
Riegos, 235-243. Aguadulce (Almeria). (in Spanish).
Perea, H. 2005. Development, verification, and evaluation of a solute transport model
in surface irrigation. Ph.D. diss. Tucson, Ariz.: Univ. Arizona.
Perea, H., T. S. Strelkoff, J. Šimunek, E. Bautista, and A. J. Clemmens. 2003.
Unsteady furrow infiltration in the light of the Richards equation. In Proc. 2nd Int’l
Conf. on Irrig. and Drain., 625-636. U.S. Com. on Irrig. and Drain.
Petryk, S., and G. Bosmajian. 1975. Analysis of flow through vegetation. J. Hydr.
Eng. 101(7): 871-884.
Philip. J. R. 1969. Theory of infiltration. In Advances in Hydroscience. Ven Te Chow,
ed. 5: 215-295. New York, N.Y.: Academic Press.
Philip, J. R., and D. A. Farrell. 1964. General solution of the infiltration advance
problem in irrigation hydraulics. J. Geophys. Res. 69: 621-631.
Playan, E., J. M. Faci, and A. Serreta. 1996. Modeling microtopography in basin
Playán, E., and P. García-Navarro. 1997. Radial flow modeling for estimating level-
basin irrigation parameters. J. Irrig. Drain. Eng. 123(4): 229-237.
Playan, E., W. R. Walker, and G. P. Merkley. 1994. Two-dimensional simulation of
basin irrigation. I: Theory. J. Irrig. Drain. Eng. 120: 837-856.
Press, W. H., S. A. Teukolsky, W. T. Vetterling, and B. P. Flannery. 2001. Numerical
Recipes in FORTRAN 77: The Art of Scientific Computing. 2nd ed. Cambridge,
England: Cambridge Univ. Press.
Purkey, D. R., and W. W. Wallender. 1989. Surge infiltration variability. Trans. ASAE
32(3): 894-900.
Ree, W. O., and V. J. Palmer. 1949. Flow of Water in Channels Protected by
Vegetated Linings, 1-115. Bulletin No. 967. Washington, D. C.: U.S. Soil Conserv.
Roth, R. L., D. W. Fonken, D. D. Fangmeier, and K. T. Atchison. 1974. Data for
border irrigation models. Trans. ASAE 17(1): 157-161.
Saint Venant, B. D. 1871. Théorie du mouvement non-permanente des eaux avec
application aux crues des riviéres et à l’introduction des marées dans leur lit.
Compte Rendus, Acad. Sci. 73: 148-154, 237-240.
Sakkas, J. G., and T. S. Strelkoff. 1974. Hydrodynamics of surface irrigation: Advance
phase. J. Irrig. Drain. Div., ASCE 100(IR1): 31-48.
Sakkas, J. G., C. V. Bellos, and M. N. Klonaraki. 1994. Numerical computation of
surface irrigation. Irrig. Sci. 15: 83-99.
Salt River Project. 1960. Major Facts in Brief. Phoenix, Ariz.: Salt River Project.
Sayre, W. W., and M. L. Albertson. 1961. Roughness spacing in rigid open channels.
J. Hydr. Div. 87(3): 121-149.

Schmitz, G. H., and G. J Seus. 1987. Analytical solution of simplified surge flow. J.
Irrig. Drain. Eng. 113(4): 603-610.
Schmitz, G. H. 1993a. Transient infiltration from cavities. I: Theory. J. Irrig. Drain.
Eng. 119(3): 443-457.
Schmitz, G. H. 1993b. Transient infiltration from cavities. II: Analysis and
application. J. Irrig. Drain. Eng. 119(3): 458-470.
Schmitz, G. H., and G. J. Seus. 1990. Mathematical zero-inertia modeling of surface
irrigation: Advance in borders. J. Irrig. Drain. Eng. 116(5): 603-615.
Schmitz, G. H., and G. J. Seus. 1992. Mathematical zero-inertia modeling of surface
irrigation: Advance in furrows. J. Irrig. Drain. Eng. 118(1): 1-18.
Shatanawi, M.R. and Strelkoff, T.S. 1984. Management contours for border irrigation,
J. Irrig. Drain. Eng. 110(4): 393-399.
Shayya, W. H., V. F. Bralts, and L. J. Segerlind. 1991. Zero-inertia surface irrigation
design: A finite element approach. ASAE Paper No. 91-2099. St. Joseph, Mich.:
ASAE.
Shepard, J. S., W. W. Wallender, and J. W. Hopmans. 1993. One-point method for
estimating furrow infiltration. Trans. ASAE 36(2): 395-404.
Sherman, B., and V. J. Singh. 1982. A kinematic model for surface irrigation: An
extension. Water Resour. Res. 18(3): 659-667.
Shockley, H. J., G. Woodard, and J. T. Phelan. 1964. A quasi-rational method of
border irrigation design. Trans. ASAE 7(4): 420-423, 426.
Šimunek, J., M. Šejna, and M. T. van Genuchten. 1999. The HYDRUS-2D software
package for simulating two-dimensional movement of water, heat, and multiple
solutes in variably saturated media. Version 2.0, IGWMC-TPS-53. Golden, Colo.:
Int’l. Grd.Water Modeling Ctr., Colorado School of Mines.
Singh, V. J., and R. S. Ram. 1983. A kinematic model for surface irrigation:
Verification by experimental data. Water Resour. Res. 19(6): 1599-1612.
Singh, V. P., and F. X. Yu. 1989a. An analytical closed border irrigation model I:
Theory. Agric. Water Mgmt. 15: 223-241.
Singh, V. P., and F. X. Yu. 1989b. An analytical closed border irrigation model II:
Experimental verification. Agric. Water Mgmt. 15: 243-252.
Skonard, C. 2002. A field-scale furrow irrigation model. Ph.D. diss. Lincoln, Nebr.:
Univ. of Nebraska.
Smith, R. E. 1972. Border irrigation advance and ephemeral flood waves. J. Irrig.
Drain. Div., ASCE 98(IR2): 289-307.
Strelkoff, T. 1969. One-dimensional equations of open channel flow. J. Hydr. Eng.
95(HY3): 861-876.
Strelkoff, T. 1977. Algebraic computation of flow in border irrigation. J. Irrig. Drain.
Div., ASCE 103(IR3): 357-377.
Strelkoff, T. 1985. BRDRFLW: A Mathematical Model of Border Irrigation, 1-100.
Publ. ARS-29. Washington, D.C.: USDA-ARS.
Strelkoff, T. 1990. SRFR: A computer program for simulating flow in surface
irrigation: Furrows–Basins–Borders. WCL Report #17. Phoenix, Ariz.: U.S. Water
Conserv. Lab.
Strelkoff, T. 1992. EQSWP: Extended unsteady flow double-sweep equation solver. J.
Hydr. Eng. 118: 735-742.
Strelkoff, T. S., E. Bautista, and A. J. Clemmens. 2003b. Errors inherent in simplified

infiltration parameter estimation. In Proc. 2nd Int’l. Conf. on Irrig. and Drain.,
735-745. Phoenix, Ariz: U.S. Com. on Irrig. and Drain.
Strelkoff, T. S., and D. L. Bjorneberg. 2001. Hydraulic modeling of irrigation-induced
furrow erosion. Selected papers from the 10th Int’l Soil Conserv. Organiz. Conf.,
ISCO ‘99. CD-ROM: 699-705.
Strelkoff, T. S., and A. J. Clemmens. 1994. Dimensional analysis in surface irrigation.
Irrig. Sci. 15: 57-82.
Strelkoff, T. S., and A. J. Clemmens. 1996a. Managing border irrigation for near-zero
discharge. In Proc., North Am. Water and Environ. Congr. ‘96, CD-ROM. Reston,
Va.: American Soc. Civil Engineers.
Strelkoff, T. S., and A. J. Clemmens. 1996b. Overviews of border irrigation
performance for optimization. ASAE Paper No. 962140. St. Joseph, Mich.: ASAE.
Strelkoff, T. S., and A. J. Clemmens. 2001. Data-driven organization of field methods
for estimation of soil and crop hydraulic properties. ASAE Paper No. 01-2256. St.
Strelkoff, T. S., A. J. Clemmens, M. El-Ansary, and M. Awad, M. 1999. Surface-
irrigation evaluation models: Application to level basins in Egypt. Trans. ASAE
42(4): 1027-1036.
Strelkoff, T.S., Clemmens, A. J., and H. Perea-Estrada. 2006. Calculation of non-
reactive chemical distribution in surface fertigation. Agric. Water Mgmt., 86:93-101.
Strelkoff, T. S., A. J. Clemmens, and B. V. Schmidt. 1998. SRFR, Version 3.31: A
Model for Simulating Surface Irrigation in Borders, Basins and Furrows. Phoenix,
Ariz.: USWCL, USDA-ARS.
Strelkoff, T. S., A. J. Clemmens, B. V. Schmidt, and E. J. Slosky. 1996. Border:A
design and management aid for sloping border irrigation systems. In Model and
Computer Program Software Documentation. WCL Report #21.
Strelkoff, T. S., and H. T. Falvey. 1993. Numerical methods used to model unsteady
canal flow. J. Irrig Drain. Eng. 119(4): 637-655.
Strelkoff, T. S., and N. D. Katopodes. 1977a. End depth under zero-inertia conditions.
J. Hydr. Eng. 103(HY7): 699-711.
Strelkoff, T. S., and N. D. Katopodes. 1977b. Border-irrigation hydraulics with zero
inertia. J. Irrig. Drain. Div., ASCE 103(IR3): 325-343.
Strelkoff, T.S. and Shatanawi, M.R. 1984. Normalized graphs of border-irrigation
performance, J. Irrig. Drain. Eng. 110(4): 359-374
Strelkoff, T. S., A. H. Tamimi, and A. J. Clemmens. 2003a. Two-dimensional basin
flow with irregular bottom configuration. J. Irrig. Drain. Eng. 129(6): 391-401.
Tabuada, M. A., Z. J. C. Rego, G. Vachaud, and L. S. Pereira. 1995. Modelling of
furrow irrigation. Advance with two-dimensional infiltration. Agric. Water Mgmt.
28: 201-221.
Tinney, E. R., and D. L. Bassett. 1961. Terminal shape of a shallow liquid front. J.
Hydr. Div. 87(HY5): 117-133.
Trout, T. J. 1992. Flow velocity and wetted perimeter effects on furrow infiltration.
Trans. ASAE 35(3): 855-862.
Trout, T. J., and G. S. Johnson. 1989. Earthworms and furrow irrigation infiltration.
Trans. ASAE 32(5): 1594-1598.
Trout, T. J., R. E. Sojka, and R. D. Lentz. 1995. Polyacrylamide effect on furrow
erosion and infiltration. Trans. ASAE 38: 761-765.

USDA. 1974. Chapt. 4, Section 15: Border Irrigation. In National Engineering
Handbook. Washington, D.C.: USDA, Soil Conservation Service.
USDA. 1985. Chapt. 5, Section 15: Furrow Irrigation. In National Engineering
Handbook. Washington, D.C.: USDA, Soil Conservation Service.
USDA. 1999. SRFR, Version 4.06. Maricopa, Ariz.: U.S. ALARC, USDA-ARS.
USDA-ARS. 2006. WinSRFR: Surface Irrigation Analysis, Design, and Simulation. V.
1.1. U.S. Arid Land Agricultural Research Center. Maricopa, AZ 85239. Available
at: http://www.ars.usda.gov/services/software/download.htm?softwareid=146.
Valiantzas, J. D. 1994. Simple method for identification of border infiltration and
roughness characteristics, J. Irrig. Drain. Eng. 120(2): 233-249.
Valiantzas, J. D. 1997. Surface irrigation advance equation: Variation of subsurface
shape factor, J. Irrig. Drain. Eng. 123(4): 300-306.
Valiantzas J. D., S. Aggelides, and A. Sassalou. 2001, Furrow infiltration estimation
from time to a single advance point. Agric. Water Mgmt. 52: 17-32.
Walker, W. R. 2003. SIRMOD III: Surface Irrigation Simulation, Evaluation, and
Design. Guide and technical documentation. Logan, Utah: Dept. of Bio and Irrig.
Eng., Utah State Univ.
Walker, W.R. 2005. Multilevel calibration of furrow infiltration and roughness, J.
Irrig. Drain. Eng., 131(2): 129-136
Walker, W.R. and Busman, J.D. 1990. Real-time estimation of furrow infiltration, J.
Irrig. Drain. Eng., 116(3): 299-318
Walker, W. R., and A. S. Humpherys. 1983. Kinematic-wave furrow irrigation model.
J. Irrig. Drain. Eng. 109(4): 377-392.
Walker, W. R., and B. Kasilingam. 2004. Another look at wetted perimeter along
irrigated furrows: Modeling implications. In Proc. World Water and Environ.
Resour. Congr. CD-ROM. Reston, Va.: EWRI/ASCE.
Walker, W. R., and G. V. Skogerboe. 1987. Surface Irrigation Theory and Practice.
Englewood Cliffs, N.J.: Prentice Hall.
Wallender, W. W., and M. Rayej. 1985. Zero-inertia surge model with wet-dry
advance. Trans. ASAE 28(5): 1530-1534.
Weir, G. J. 1983. A mathematical model for border strip irrigation. Water Resour. Res.
19(4): 1011-1018.
Weir, G. J. 1985. Surface irrigation advance motion after gate closure. Water Resour.
Res. 21(9): 1409-1414.
Whitham, G. B. 1955. The effects of hydraulic resistance in the dam-break problem.
Proc., Royal Society of London, Series A. 227: 399-407. London, England.
Wu, I. 1972. Recession flow in surface irrigation. J. Irrig. Drain. Div., ASCE 98(IR1):
77-90.
Yost, S. A., and N. D. Katopodes. 1998. Global identification of surface irrigation
parameters. J. Irrig. Drain. Eng. 124(3): 131-139.
Zerihun, D., C. A. Sanchez, A. Furman, and A. W. Warrick. 2005. A coupled surface-
subsurface flow model for improved basin irrigation management. J. Irrig. Drain.
Eng. 131(2): 111-128.
Zerihun, D., C. A. Sanchez, and K. L. Farrell-Poe. 2004. Modified two-point method
for closed-end level-bed furrows. In Proc. 2004 World Water and Environ. Resour.
Congr. CD-ROM. Reston, Va.: EWRI/ASCE.
CHAPTER 14
DESIGN OF
SURFACE SYSTEMS
Albert J. Clemmens (USDA-ARS,
Maricopa, Arizona)
Wynn R. Walker (Utah State University,
Logan, Utah)
Delmar D. Fangmeier (University of Arizona,
Tucson, Arizona)
Leland A. Hardy (H & R Engineering, Inc.,
Salem, Oregon)
Abstract. Surface irrigation design and operation are a challenge because soil in-
filtration and soil and crop resistance influence the movement of water over the field
and thus the water distribution. In the past, design for each surface irrigation method
was treated differently because of differences in the simplicity with which different
phases of the irrigation could be described. This has tended to make surface irrigation
analysis and design appear disjointed. In this chapter, we apply the same basic proce-
dures for the design of all surface systems, deviating where needed to make the proce-
dures both straightforward and sufficiently accurate. The basis for these designs is the
ability to predict advance, recession, the distribution of infiltrated water, and the per-
formance for a given set of conditions. Conservation of mass is the main concept, with
empirical approximations used where needed. Examples are provided for each method.
Keywords. Basin, Border, Design, Furrow, Irrigation, Surface irrigation.
14.1 INTRODUCTION
Surface irrigation methods are so named because water is distributed across the
field by flowing over the field surface. Thus, soil infiltration and soil and crop flow
resistance have major influences on the distribution of water. Because these conditions
change with both time and space, the design and operation of surface irrigation sys-
tems is often more difficult than pressurized irrigation systems. However, surface irri-
gation systems still comprise more than 90% of the world’s irrigated land (FAO,
2005) and almost half of the irrigated land in the U.S. (Hutson, 2004). While some
land is continually being converted to pressurized irrigation, a significant portion of
land is likely to remain surface irrigated for the foreseeable future. Surface irrigation
can be an effective irrigation method and in some cases can equal the efficiencies of
pressurized irrigation methods (Kennedy, 1994: 166). Surface irrigation is still the
most economical method for many situations.
500 Chapter 14 Design of Surface Systems
Table 14.1. Categories of surface irrigation.

Control of Inflow End
Method Lateral Flow Slope Control Conditions
Sloping furrow furrows steep or to individual open,
low gradient; furrows blocked, or
either can have group of furrows
cross slope blocked
Border strip flat planted or steep or distributed open,
corrugations low gradient across blocked, or
cross slope on upper end partially blocked
milder slopes
Level basins/ flat planted, zero in all can be point blocked
Level furrows furrowed or directions inflow if furrowed, all
bedded interconnected
furrows zero in direction to individual blocked, or
of run, can have furrows group of furrows
cross slope blocked
Surface irrigation methods can be categorized according to how they function hy-
draulically. Distinctions can be made by advance and recession curves, which describe
the time when the advancing stream reaches a particular location and the time when
standing water no longer occurs on the surface. This hydraulic comparison assumes
that water enters the field or irrigation set along one end and flows to the other end
uniformly across the set width. We distinguish three categories of surface irrigation:
sloping furrows, border strips, and level basins and furrows. The main physical differ-
ences are given in Table 14.1. These physical differences result in hydraulic differ-
ences, primarily the magnitude of the inflow rate (and pattern) and the general shape
of the recession curves and runoff hydrographs.
The purpose of this chapter is to present methods for the design of modern surface
irrigation systems, where modern implies a reasonable control over the water supply.
The chapter focuses on the three major categories of surface irrigation; design of rice
paddies and methods such as contour levee, contour ditch, wild flooding, etc. are not
discussed.
14.2 DESIGN CONSIDERATIONS AND APPROACHES
Many surface irrigation systems are ineffective and inefficient. This can be caused
by physical constraints (e.g., steep land slopes, shallow soils, poor water supplies,
etc.), by poor design and layout, or by improper operation and management. A thor-
ough discussion of the constraints and limitations of surface irrigation systems is be-
yond the scope of this chapter. For more details, see Walker and Skogerboe (1987),
Clemmens and Dedrick (1994), or Burt et al. (1999). One advantage of surface irriga-
tion over pressurized irrigation methods is that they often do not require a good, reli-
able water supply. They can be adapted to different rates of flow, flows that vary ran-
domly, and flows with poor water quality (sediment, debris, etc.). Efforts in surface
irrigation research and extension have focused on methods for providing better water
control—control over flow rate or control over volume applied. These generally focus
on how the system is operated. Of equal importance is field design and layout. Good
operation cannot make up for a poor field design. However, when surface irrigation
systems are properly designed and more modern operating procedures are used, irriga-
tion efficiencies and uniformities can be high.
14.2.1 Design Objectives

The amount of water to be supplied during an irrigation event, referred to as the
target or required depth of application, is a major design consideration. Typically,
surface irrigation systems have a narrow range of target depths for which they are rea-
sonably efficient and uniform. The irrigator must adjust irrigation practices (typically
flow rate and application time) to account for changing field conditions (infiltration
and roughness). Irrigators typically develop rules of operation that they use to make
adjustments. A poor field design will make these judgments difficult for the operator.
A good design will include guidance on system operation.
Surface irrigation systems are most applicable on mild to level slopes. On steeper
slopes, erosion can become excessive and the range of operating conditions can be
narrow (e.g., narrow range of target depths).
The design objectives are typically stated in terms of achieving some desired appli-
cation efficiency, Ea. This efficiency is called potential application efficiency, PEa,
here to distinguish it from a field-measured Ea. Design is typically based on supplying
the target depth of water everywhere in the field. The PEmin is the application effi-
ciency when the minimum depth infiltrated just equals the target depth:
d min d req
PE min = = (14.1)
da da
where dreq is the required depth, dmin is the minimum infiltrated depth, and da is the
depth of applied irrigation water, in this case, the depth applied that results in the
minimum depth just equal to the required depth. In practice, some under-irrigation is
usually allowed and operation is based on satisfying the low-quarter depth. Because
design does not take into account all of the variability which exists within the field, we
use a design based on satisfying the minimum depth, with the expectation that when
properly operated the low quarter depth would be satisfied.
14.2.2 Choosing Design Conditions and Parameters
Surface irrigation design requires the estimation of parameters that define the infil-
tration of water into the soil and the resistance to water movement caused by the soil
surface and vegetation. These are key factors in the design and are often very difficult
to determine accurately. These factors can change over the course of the growing sea-
son, further complicating design. Published values of the Manning roughness coeffi-
cient, n, are usually sufficient for design purposes. However, infiltration is much more
difficult to estimate. Whenever possible, estimates of these parameters should be ob-
tained from field measurements, preferably from irrigation events. Soil infiltration
conditions can change significantly when land is newly converted to surface irrigation.
Once a reasonable estimate of infiltration parameters is obtained and a design se-
lected, a sensitivity analysis should be performed to determine how the design might
change under different conditions of infiltration, flow resistance, and target infiltration
depth; conditions that will likely be faced by the irrigator.
14.2.3 Equations for Infiltration in Borders and Basins
Infiltration is one of the most important factors affecting the design and perform-
ance of surface irrigation systems. Unfortunately, estimating infiltration conditions is
one of the most difficult things to do in the field. Several equations have been used to
define infiltration. Philip (1957) developed equations for ideal soil conditions (e.g.,
uniform physical properties) for infiltrated depth versus time for short times:
1/ 2
d( τ ) = kτ + bτ i( τ ) = 1 2 kτ −1/ 2 + b (14.2)
and for long times:
d( τ ) = c2 + bτ i( τ ) = b (14.3)
where τ = infiltration time
k = sorptivity
b is related to the hydraulic conductivity
c2 = a derived constant.
Unfortunately, soil surface conditions are not ideal and many soils do not follow
Equation 14.2 very well at short times. Many do follow Equation 14.3 at long times,
but with different values for c2 than those derived theoretically.
Irrigation engineers have tended to use the Kostiakov or Kostiakov-Lewis equation,
or a variation thereof, defined by
d( τ ) = kτ a i( τ ) = akτ a −1 (14.4)
or in modified form,
d( τ ) = c + kτ a + bτ i( τ ) = akτ a −1 + b (14.5)
where b, c and k are all empirically derived. In Equation 14.4, the exponent a is often
below 1/2, indicating the poor fit to Equation 14.2. However, with Equation 14.4, the
infiltration rate approaches zero as time approaches infinity. Soils generally reach a
constant final infiltration rate. For course-textured soils, where the required infiltration
time is short, this occurs well after typical irrigation opportunity times. For fine-
textured soils, the final infiltration rate is reached during the irrigation and should be
included.
In keeping with the theoretical development, a Kostiakov branch function has been
used, where the first branch uses Equation 14.4 (or 14.5 with b = 0) for short times,
and switches to Equation 14.3 when the infiltration rate (derivative of equation with
respect to time) equals b. For soils that reach a nearly constant, final infiltration rate
during the irrigation, design and evaluation can be greatly simplified with use of this
equation. The Kostiakov branch function equations are:
d( τ ) = c + k τ a i( τ ) = ak τ a-1 for τ ≤ τ B
(14.6)
= c 2 + bτ =b for τ ≥ τ B
where τB is the time at which the infiltration depth and rate for the two branches
match. For cracking-clay soils, the depth infiltrated over the first few minutes of wet-
ting can be large relative to the total depth infiltrated during the irrigation. With Equa-
tion 14.4, this could be fit with a very small value of the exponent a. However, this
makes the problem unnecessarily difficult, since whether the depth infiltrates immedi-
ately or, say, during the first 10 min has very little effect on the hydraulics of the irri-
gation or on the distribution of infiltrated depths. For these soils, it makes more sense
to fit infiltration with an equation such as Equation 14.3 and ignore the initial part of
the infiltration curve (e.g., use second half of Equation 14.6).
The U.S. Soil Conservation Service (SCS, currently Natural Resources Conserva-
tion Service) developed a series of infiltration families to assist field personnel in de-
termining infiltration relationships. These families are defined according to Equation
14.5 with b = 0 and c = 7 mm. The infiltration relationships for many soils do not fol-
low the SCS families very well. The value of exponent a in this equation is around 0.7
for all the families. Non-cracking soils typically have exponents that range from 0.3 to
0.8. Merriam and Clemmens (1985) developed a relationship between the exponent a
and the time to infiltrate 100 mm from a wide range of field data. From this they de-
veloped time-rated intake families, based on the time to infiltrate 100 mm, a typical
required depth for surface irrigation. These time-rated intake families are given in Ta-
ble 14.2.
These infiltration relationships are entirely empirical. Emphasis should be placed
on the general characteristics of the infiltration relationship and not on the equation
chosen or the values of the various parameters. As long as the equation and constants
chosen fit the observed infiltration relationship and are useful, this is sufficient justifi-
cation for their use. In particular, it is most important to have the proper range of infil-
trated depths over the range of infiltration opportunity times. Table 14.3 shows the
parameters for five infiltration functions, representing the various equations fit to infil-
tration described by the 6-hr time-rated family (i.e., 6 h to infiltrate 100 mm with a
Kostiakov exponent of 0.51). This corresponds to an SCS intake family of about 0.45.
These infiltration functions are plotted in Figure 14.1. There is very little difference
between the infiltrated depths predicted by the three different Kostiakov equations. A
linear equation even fits well between 4 and 8 h.
Table 14.2. Time-rated intake families (from Merriam and Clemmens, 1985).
Time to Infiltrate Kostiakov Kostiakov Constant (k)
100 mm (hours) Exponent (a)[a] (mm/hr[a]) (in/hr[a])
0.5 0.739 167.0 6.57
1 0.675 100.0 3.94
2 0.611 65.5 2.58
4 0.547 46.8 1.84
8 0.483 36.6 1.44
16 0.419 31.3 1.23
32 0.355 29.2 1.15
[a]
Exponent a = 0.675 – 0.2125 log10(τ100), where τ100 is the time (in hours) to infiltrate 100 mm.
100
Depth (mm)
50
SCS
Kostiakov
Modified
Linear
Branch
0
0 2 4 6 8
Time (hours)
Figure 14.1. Comparison of five infiltration functions.
Table 14.3. Infiltration functions matching 6-hr time-rated family.

SCS Kostiakov Modified Kostiakov Linear
Parameters Equation Equation Kostiakov Branch Equation
k (mm/hr) 24.26 40.1 37.39 40.1 0
a 0.75 0.51 0.35 0.51 0
b (mm/hr) 0 0 5.0 8.5 8.9
c (mm) 7 0 0 0 49.0
14.2.4 Equations for Infiltration in Furrows

Infiltration in furrows can be significantly more complicated than in flat borders or
basins due to the two-dimensional nature of the furrow cross-section. Infiltration in
borders and basins is generally considered to be one dimensional: downward. Infiltra-
tion in furrows can be influenced by the wetted width of the stream and lateral flow
into the furrow bed. If gravitational forces dominate the infiltration process (e.g., a
very sandy soil), then infiltration may be directly proportional to the wetted width. If
furrows are closely spaced and the soil is very heavy such that capillary forces domi-
nate infiltration, the lateral infiltration from adjacent furrows will meet such that infil-
tration essentially occurs over the entire set width; i.e., for each furrow, infiltration
becomes a function of the furrow spacing. The main problem is that for many situa-
tions, infiltration conditions are in between these two extremes, and may differ signifi-
cantly over time or along the length of run. For example, infiltration at the head end of
the field, where the flow rate and water depth are high, may result in infiltration after,
say, 12 hours essentially governed by furrow spacing; while at the tail end where the
flow rate and wetted width are significantly reduced and the infiltration opportunity
time is short, infiltration may be strongly influenced by the wetted width.
The SCS design procedures for sloping furrows (USDA, 1984) use an infiltration
function based on the wetted perimeter at normal depth, plus an empirical constant of
213 mm to account for the lateral flow. This is reported as an average value from nu-
merous tests. For irrigation of every furrow, this width should not exceed the furrow
spacing. This width is then assumed constant over time and distance. The assumption
with this procedure is that the SCS intake family for a given soil used for border-strip
irrigation design can also be used for sloping-furrow design. In practice, this approach
has not been very successful.
For general use, infiltration really needs to be judged from field performance of the
irrigation system. If a constant width is to be chosen for infiltration, it is simpler to
express infiltration as a function of furrow spacing; recognizing that significant
changes in wetted perimeter (e.g., from furrow shape or flow-rate changes) or furrow
spacing (or whether all furrows are irrigated) may change infiltration constants based
on furrow spacing. The design procedures presented here do not consider the influence
of wetted width on furrow infiltration. Thus, these design procedures should be used
with some caution as they may overestimate irrigation uniformity.
Where no furrow infiltration data are available, it may be necessary to estimate fur-
row infiltration from ring infiltration data or from border irrigation trials. For mild
slopes, high flow rates, closely spaced furrows and heavy soils, the infiltration can be
assumed to be somewhat similar to that for borders (i.e., one-dimensional and based
on furrow spacing), except there is a tendency for the infiltration rates to start higher
and drop more quickly (i.e., a smaller value of the exponent a in the Kostiakov-Lewis
equation). For steeper slopes, low flow rates and coarse-textured soils, the border infil-
tration rates can be applied to the wetted width, with perhaps an increase in k to ac-
count for some lateral movement and a decrease in the exponent a. Such adjustments
must be considered just a guess since the effects of differences in tillage and compac-
tion from tractor traffic between borders and furrows can be substantial.
Furrow irrigation has the additional complexity in that equipment wheels travel
through some furrows and not others. Wheel traffic generally causes compaction and
smoothing of the soil surface, a reduction in infiltration, and a reduction in flow resis-
tance. In some cases with sloping-furrow systems, only wheel or only non-wheel rows
are irrigated. Where both are irrigated, the inflow rates to wheel and non-wheel rows
are typically adjusted by the irrigator so that advance in all rows is approximately the
same. These adjustments are not needed in level furrow or furrowed level-basin irriga-
tion systems. In other cases, non-wheel furrows are intentionally compacted by drag-
ging heavy objects (sometimes called torpedoes) through them. When considering also
changes in properties over the season, design procedures often have to be applied to a
wide range of conditions. Because of this, some irrigation evaluations only attempt to
determine field-average conditions (e.g., an infiltration function determined for an entire
set), which assumes that irrigation and tillage practices will be similar after design.
14.2.5 Equations for Flow Resistance
Resistance to flow is usually described by the Manning roughness coefficient. This
equation relates the flow rate, Q, to the flow area, A, the hydraulic radius, R (area over
wetted perimeter), the friction slope, Sf, and the Manning n:
AR 2 / 3 S 1f / 2 Cu
Q= (14.7)
n
The units coefficient, Cu, allows the same value of the Manning n in both foot-
pound and metric units. For units of m and m3/s, Cu has a value of 1.0. Expressed in
terms of flow rate per unit width, q, for border strips and basins, this becomes
y5/3 S1/2
f Cu
q= (14.8)
n
where y is the flow depth and q = Q/W, where W is the basin or border width. All units
in Equations 14.7 and 14.8 are in meters and seconds.
In theory, the Manning n should change with flow depth when vegetation is pre-
sent. However, the changes in the Manning n over the season and with changes in
planting density are as large as changes caused by flow depth. For design purposes, a
constant Manning n seems to be satisfactory. Published values of the Manning n for
various crops are shown in Table 14.4. Manning n values have been observed as high
Table 14.4. Recommended Manning n values.
Source n Conditions for Use
USDA (1974, 1984) 0.04 smooth, bare soil surfaces and furrows
USDA (1974, 1984) 0.10 drilled, small-grain crops if the drill rows run in the direction of
water flow and corrugations
USDA (1974) 0.15 alfalfa, broadcast small grains
Clemmens (1991) 0.20 dense alfalfa or alfalfa on long fields with no secondary ditches
USDA (1974) 0.25 dense sod crops and small grains drilled perpendicular to the flow
as 0.4 on closely planted wheat, a value significantly higher than that given in the ta-
ble. Also, furrow-irrigated crops with vegetation that hangs into the water can cause
much greater resistance than the Manning n of 0.04 given in Table 14.4. Here again,
field observations of advance and recession can be used to determine whether design
estimates of Manning n are reasonable.
14.2.6 Consideration of Operating Practices and Guidelines
We know that the conditions assumed in design will never be exactly encountered
in practice. The irrigator of a surface irrigated field will adapt operations to account
for differences from the assumed conditions. In addition to variations in infiltration
and roughness, the irrigator must adapt to changes in flow rate and target depth of ap-
plication. Frequently, the flow rate is only approximately known and changes in appli-
cation time to account for this are not easily made. The operator frequently uses in-
formation about the advancing stream to make these adjustments, either in inflow rate
or in application time.
Usually, these adjustments are based solely on experience. However, such experi-
ence takes time to accumulate and often experience on one type of soil does not trans-
late to another type of soil. To aid in this process, design should consider the operating
criteria to be used; either as recommendations after design or used explicitly in the
design procedure to help determine the field layout. In any event, a sensitivity analysis
can determine what will happen if the irrigator operates the system with the defined
operating criteria over the range of possible conditions.
14.2.7 General Design Approaches
The approach taken for design assumes that repeated trials are needed based on ex-
isting conditions, local farming and irrigation practices, and designer experience. The
procedures provided in this chapter allow the calculation of irrigation results and per-
formance for each individual trial. It is up to the designer to explore various options in
arriving at a final design recommendation. The methods presented here allow one to
compute advance and recession curves, the distribution of infiltrated water, the amount
of deep percolation, and runoff, if applicable. The approach assumes that the designer
knows the infiltration function of the soil. If infiltration is not well known or changes
over the season, the designer is responsible for evaluating the impact of variations in
infiltration on design and operations. This is beyond the scope of this chapter.
The design procedures assume that the depth to be infiltrated during a particular ir-
rigation event is known by the designer. This required depth of infiltration may be
different for different crops and can change over the season. It is the designer’s re-
sponsibility to take this into account during the design process.
Design is typically based on assuming that one end of the field or the other will re-
ceive the least infiltrated depth. Then, the inflow and application time are adjusted
such that the required depth is infiltrated at that location. The time to infiltrate the re-
quired depth, τreq, becomes an important design parameter. Typically, it has more in-
fluence on the design than the infiltration constants themselves. The infiltration oppor-
tunity time at any location, x, along the length of run, τopp(x), is defined as the time
between advance, tA(x), and recession, tR(x) or
τ opp ( x) = t R ( x) − t A ( x ) (14.9)
If the minimum depth is at the head end of the field (x = 0), the opportunity time at
that point is equal to the recession time, or
τ opp (0) = t R (0) = t co + tlag (14.10)

where tco is the time of cutoff or application time and tlag is the recession lag time, or
the time required for the water depth at the upstream end to drop to zero after cutoff.
Design based on meeting the requirement at the upstream end does not require ad-
vance and recession curves to be computed. However, an estimate of PEmin is required,
along with a method for estimating recession lag time (USDA, 1974).
For well designed systems, the minimum depth is typically at the downstream end
of the field, furthest from the water source (x = L). Then advance and recession curves
must be computed, and estimation of inflow rate and time required to achieve τreq is
more difficult. Specifically,
t R ( L) = t A ( L) +τ req (14.11)
Precise solutions of advance and recession are possible with solution of the Saint
Venant equations of continuity and momentum. However, this solution requires nu-
merical analysis and is done on a case-by-case basis. It does not directly produce gen-
eral equations for advance and recession. The time of cutoff is
t co = t A ( L) +τ req − [t R ( L) − t R (0) − tlag ] (14.12)
where the term in brackets is the time between cutoff and recession at the downstream
end, which is sometimes assumed to be zero (e.g., sloping furrows with runoff). In this
chapter, we provide methods for design based on satisfying the minimum depth at the
downstream end.
14.2.8 Assumed Surface-Volume Method
All design methods must use procedures that satisfy a volume balance, regardless
of how sophisticated they are. In this chapter, we make assumptions regarding the
surface volume in order to use a volume balance to determine an advance curve. This
has advantages over strictly empirical equations since the assumptions regarding the
surface volume can be verified with field observations or from computer simulation.
During advance, the cumulative infiltrated volume is equal to the difference be-
tween the accumulated inflow volume and the surface storage volume at that time.
This volume balance relationship may be expressed as:
V in (t ) = V y (t ) + V z (t ) (14.13)
where Vin(t) is the inflow volume at time t, Vy(t) is the volume in surface storage at
time t, and Vz(t) is the infiltrated volume at time t. Using this relationship to determine
advance time, tx, to distance x requires that the surface and subsurface volumes at any
time during advance are known. Typically, Equation 14.13 is put in the following form:
Qin t x = σ y A0 (t x ) x + σ z Wz (t x ) x (14.14)
where Ao(t) is the cross-sectional flow area at the inlet at time t, σy is the surface stor-
age shape factor, W is the width, z is the infiltrated depth, and σz is the subsurface
shape factor (all units are in meters and seconds). The surface shape factor is the ratio
between the average cross-sectional flow area and that at the head of the field. For
furrows, it is typically between 0.70 and 0.80.
The subsurface shape factor is the ratio between the average infiltrated cross-
sectional area (infiltrated depth times width), and the infiltrated cross-sectional area
(depth times width) at the head of the field. When infiltration is defined by Equation
14.5, the subsurface shape factor is difficult to determine. It is easier to rewrite the
subsurface volume in the following form:
⎛ h ⎞
V z = W ⎜ c + σ z1 k t xa + bt x ⎟ x (14.15)
⎝ 1+ h ⎠
where h is the exponent on the advance equation,
t x = s xh (14.16)
where s is a constant. Then σz1 can be found from (ASAE, 1991)
h + a(h − 1) + 1
σ z1 = = (14.17)
(1 + a)(1 + h)
Determining an advance curve requires knowledge of Qin, A0, σy, the infiltration
constants (c, k, a, and b), and the width. Defining this curve is still an iterative process
since the advance exponent h and advance times are interrelated and must be solved
for simultaneously.
The main problem with determining advance from Equations 14.14 through 14.17
is that the A0 and σy are not known in general. For steep slopes, this method is rea-
sonably good since after a short time, the surface volume is typically a small amount
of the total inflow volume. Also, A0 can be assumed equal to the flow area at normal
depth and σy can be estimated with sufficient accuracy. For milder slopes, the surface
volume is often a large portion of the inflow volume, even at the time of cutoff. Also,
A0 changes continuously during the irrigation event and in some cases never reaches
normal depth (see Chapter 13). Similarly, σy can vary (see Chapter 13). On a unit
width basis, A0 = y0 W.
14.2.9 Simulation Approach
One method for more accurately determining advance is to simulate it with the un-
steady flow equations, e.g., with the simulation programs SIRMOD (Walker, 1989) or
SRFR (Strelkoff, 1990). Recession can be found either from continued solution of the
equations, or from assuming that recession occurs at the same time everywhere. The
latter approach was used in the level-basin design program, BASCAD (Boonstra and
Juriens, 1988). BASCAD will search through various input parameters to arrive at a
useful solution. It does not allow PEmin as a design input, but displays it as a design
output. The above unsteady-flow simulation programs can also be used through trial
and error by the user to arrive at a design solution.
14.2.10 Dimensional-Analysis Approach
Because of the large number of variables involved in surface irrigation, it is virtu-
ally impossible to present the results of simulations for the range of possible field con-
ditions. Dimensional analysis methods have been used to reduce the number of pa-
rameters so that meaningful results can be displayed on a limited number of graphs.
Solutions have been developed for flat-planted level basins and open-ended sloping
border strips, and are available in the form of computer programs BASIN (Clemmens
et al., 1995) and BORDER (Strelkoff et al., 1996). No nondimensional solutions are
currently available for furrows. This methodology is presented in some detail in Chap-
ter 13.
14.3 SLOPING-FURROW IRRIGATION

With sloping-furrow irrigation, water advance must be sufficiently fast so that the
downstream end will receive adequate water while the upstream end is not excessively
over irrigated. However, advance that is too rapid can result in a large percentage of
the applied water running off the field, unless inflow after completion of advance is
reduced, for example with a cutback, surge, or cablegation system, or the runoff is
collected for reuse.
The Soil Conservation Service (USDA, 1984) provided the following guidance for
limiting irrigation-induced erosion from furrows. Furrow slopes in areas of high rain-
fall should be great enough to allow adequate drainage (> 0.03%), yet not so great as
to cause significant erosion. For erodible soils (e.g., silty soils), the maximum furrow
slope should be limited to 60/(P30)1.3, where P30 is the 30-min rainfall in mm for a 2-
year frequency. This limit can be exceeded by about 25% for less erodible soils (e.g.,
sandy and clayey soils). Further, erosion can be limited by placing a limit on the irri-
gation stream size. The maximum flow velocity should be limited to 8 and 13 m/min
for erosive and non-erosive soils, respectively. The relationship between velocity and
flow rate can be obtained from Equation 14.7, since velocity is flow rate, Q, divided
by flow area, A.
14.3.1 Design of Sloping Furrows with No Cutback or Tailwater Reuse
Design of sloping furrows with runoff is very straightforward under the following
assumptions. First, advance can be computed from Equation 14.13 or 14.14, which
give an accurate estimate of advance times when a good estimate of the surface-
volume shape factor is made. Second, we can assume that the flow depth at the up-
stream end is at normal depth for the entire period of inflow and the friction slope
equals the bottom slope, Sf = S0. These assumptions are particularly appropriate for
slopes greater than 0.05%. Next, recession along the entire furrow length is assumed
to occur immediately after cutoff for steeply sloped furrows, particularly above 0.5%
slope. For flatter slopes, the recession at the downstream end will take some time, re-
sulting in a greater infiltration depth there and underprediction of PEa. Adjustment to
the application time and recession curve are made based on the surface volume at the
completion of advance.
Finally, this design approach assumes that the wetted width does not vary over the
length of the furrow. That is, the infiltrated volume over a unit length of furrow is de-
pendent only on the infiltration opportunity time and not on the depth of flow or wet-
ted width. For heavy soils with significant lateral sorption, the wetted width from ad-
jacent furrows overlap, making the wetted width per furrow essentially equal to the
furrow spacing. For coarse-textured soils, wetted width can cause a significant reduc-
tion in infiltration along the furrow due to a reduction in flow depth as the flow rate
gets smaller with distance from the head end. For this latter case, field design should
not attempt to minimize runoff since this would drastically reduce furrow flow rates
toward the downstream end. Other measures (e.g., return flow) are needed to improve
performance in these cases.
Design for a specific set of field conditions (i.e., infiltration, roughness, and re-
quired infiltration depth) is essentially by trial and error for furrow length, slope, and
inflow rate. The furrow cross-sectional area and wetted perimeter must be specified as
a function of flow depth (i.e., normal depth). Any function can be useful provided that
it properly describes the cross-section. Trapezoidal and power-function shapes are
commonly used. Then the normal depth for a given discharge can be found by trial
and error from Equation 14.7. The flow area at the upstream end, A0, is then computed
from the cross-section definition. Since this does not vary during inflow, the surface
volume can be computed as a function of distance only.
The assumed surface-volume method provides a relationship between advance dis-
tance x and advance time tx from Equation 14.14, where W is the furrow spacing, un-
der the assumption that Qin, W, A0, σz, and σy are fixed. Fortunately, for most situa-
tions of interest σy does not vary significantly, generally falling between 0.7 and 0.8.
However, σz can vary considerably and should be adjusted for the particular case of
interest. For a power infiltration function, it can be estimated from Equation 14.17,
which required a power advance function (Equation 14.16). Estimation of the exponent h
in Equation 14.16 requires that at least two points on the advance curve be estimated.
The design procedure starts with estimates for the time for the advancing stream to
reach half the field length and the entire field length. From these two, the advance
exponent can be computed from
log(t L / 2 /t L )
h= (14.18)
log(1/2)
With Equations 14.15 and 14.17, we can compute the subsurface volume. A new
guess for the advance times can then be estimated from the volume balance (e.g.,
Equation 14.14) and new values for h calculated. The procedure converges fairly rap-
idly. An alternate procedure which is computationally faster for spreadsheets is to es-
timate h, allow the spreadsheet to iterate on the time to achieve a volume balance at
both distances independently, and manually iterate on h until it converges.
To provide the required depth at the downstream end, the cutoff time is computed
from Equation 14.12, with the last term (in brackets) assumed equal to zero. The ap-
plication efficiency is computed as the required volume (required depth × furrow spac-
ing × furrow length) divided by the applied depth (inflow rate × cutoff time).
The volume of runoff can be estimated by computing a distribution of infiltrated
depths, the associated infiltrated volume, and subtracting this volume from the inflow
volume. The deep percolation volume is the infiltrated volume minus the required
volume, since it is assumed that all points receive at least the required depth. The infil-
trated volume can be computed from the infiltrated depths at the head, middle, and tail
of the field, since the infiltration opportunity times are known there (i.e., recession is
assumed to occur at cutoff). Better estimates can be made by closer numerical integra-
tion, for example by computing additional points along the advance curve.
Example 1. Consider a field 400 m long with a slope of 0.002, infiltration (based
on a width equal to the 1 m furrow spacing) described by k = 40 mm/hra and a = 0.35,
and Manning roughness of 0.05. (Carefully smoothed and compacted furrows can
have a roughness as low as 0.03.) The furrow spacing is 1 m and the furrow shape is
defined as a trapezoid with a bottom width of 100 mm with 2:1 side slopes, horizontal
to vertical. Initial design will be made for a required depth of 80 mm. The resulting
required opportunity time is 435 min.
This design example starts by assuming an inflow rate of 1.0 L/s. Iterative solution
of Equation 14.7 gives; a normal depth of 53 mm, a wetted perimeter of 338 mm, a
velocity of 5.5 m/min (well within the acceptable range), and an upstream cross-
sectional flow area of 0.011 m2.
For sloping furrows, we assume that the upstream depth reaches normal depth
quickly and remains there. Then the surface volume during advance is not a function
of time, and Vy(t) becomes Vy(x). If we assume that σy = 0.75, then from the expres-
sion for Vy in Equation 14.14, we get surface volume of 1.65 and 3.29 m3 for advance
to 200 and 400 m, respectively (e.g., 0.75 × 0.011 × 200 = 1.65).
The design procedure starts with a guess for the advance time to half the field
length and to the field end. For this example, we arbitrarily guess 100 and 250 min,
respectively. For a flow rate of 1 L/s, this represents inflow volumes of 6.0 and 15.0
m3. From Equation 14.18, we get h = 1.32. With h = 1.32 and a = 0.35, Equation 14.17
gives σz = 0.777. (Note that for the Kostiakov equation σz and σz1 are the same.) The
infiltrated depths at the head of the field at 100 and 250 min are 47.8 and 65.9 mm,
respectively. Substituting these into the expression for Vz in Equation 14.14, we get
subsurface volumes of 7.43 and 20.48 m3 at these two times, respectively.
At 100 min, the computed surface and subsurface volumes total 1.65 + 7.43 = 9.08,
which is much larger than the inflow volume of 6.0. Thus the advance time to 200 m
must be much longer. Similarly, at 250 min Vy + Vz = 23.77 m3 > Vin = 15.0 m3. If we
adjust the advance times to remove the volume error, our next guess of advance times
would be 100 × (9.08/6.0) = 151 min and 250 × (23.77/15.0) = 396 min, respectively.
If we repeat the calculations, we get h = 1.39, σz = 0.783, Vin = 8.65 and 24.25, Vy +
Vz = 10.30 and 27.55, for the two advance locations, respectively. Clearly, this linear
adjustment is too small. This is due to the fact that the infiltrated depth also changes
with the change in time. From this point, if we take 11/2 times the correction we get
advance times of 181 and 490 min, h = 1.44, σz = 0.787, Vin = 9.27 and 26.27, Vy + Vz
= 10.92 and 29.56, for the two advance locations, respectively. Again the correction is
insufficient, but much closer.
The solution for advance finally converges as shown in Table 14.5. The cutoff time
is then 494 + 435 = 929 min (15.5 h). The required volume is 32 m3 (400 m × 1 m ×
0.080 m), while the applied volume is 55.7 m3 (0.0010 m3/s × 929 min × 60 sec/min),
giving an application efficiency of 57.4% (32/55.7). This assumed surface-volume
method can be used to compute the advance time to intermediate locations. The ad-
vance times computed from the power-function advance relationship derived from the
two-point method usually do not exactly satisfy a volume balance, but can be used as a
good first guess of the advance time. If we use the value of the advance exponent h
from the two-point method, the advance times calculated at intermediate points are
shown in Table 14.6.
If we assume that recession occurs everywhere at the time of cutoff, then the depths
infiltrated at the head, middle, and tail end of the field are 104, 97, and 80 mm, respect-
Table 14.5. Advance calculations for assumed surface-volume

sloping-furrow design example.
Q x tx Vin(t) A0 Vy(t) h σz Vz(t)
(L/s) (m) (min) (m3) (mm) (m3) (m3)
1.0 200 182.3 10.94 0.011 1.65 1.44 0.787 9.29
1.0 400 494.0 29.64 0.011 3.29 1.44 0.787 26.35
Table 14.6. Advance and recession for sloping-furrow design

example (Qin = 1 L/s), with recession times adjusted.
Distance Advance Time Recession Time Opportunity Time Infiltrated Depth
(m) (min) (min) (min) (mm)
0 0 874 874 102
50 25 881 856 101
100 67 888 820 100
150 121 894 774 98
200 182 901 719 95
250 251 908 657 92
300 327 915 580 89
350 408 922 514 85
400 494 929 435 80
tively. Numerical integration with the trapezoidal rule (i.e. numerically averaging
depths over the two intervals) gives an average infiltrated depth of 94.5 mm and infil-
trated volume of 37.8 m3. The runoff volume is the difference between the applied and
infiltrated volumes, or 55.7 – 37.8 = 17.9 m3, or 32.1%. The deep percolation volume
is 37.8 – 32 = 5.8 m3, or 10.4%. Better estimates for these volumes can be found by
dividing the field into several increments and computing advance either from the vol-
ume balance or from Equation 14.16 (the difference is insignificant). In this example,
dividing the field into eight increments (as shown for advance in Table 14.6) for ad-
vance and recession gave runoff and deep percolation volumes of 31.7% and 10.9%
(rather than 32.1% and 10.4%).
Our assumption with the method is that the volume of water on the surface at the
time of cutoff is insignificant. In this example, the slope is relatively shallow and the
volume on the surface at cutoff from Table 14.5 is 3.3 m3, or 5.9%, not an insignifi-
cant volume. If we compare these results with those from simulation, we find that re-
cession at the downstream end takes more than an hour, implying that we could cut off
1200
1000
800
Time (min)
Vol-Bal Advance
600 Vol-Bal Recesssion
SRFR Advance
SRFR Recession
400
SRFR-adjusted cutoff
Vol-Bal adjusted cutoff
200
0
0 100 200 300 400
Distance (m)
Figure 14.2. Advance and recession curves for sloping-furrow design example.
the stream much earlier. Figure 14.2 shows the advance and recession curves from the
volume balance and from simulation with the SRFR program. Note the excellent
agreement in the two advance curves, associated with the assumed surface-volume and
the unsteady-flow simulation. The computed runoff and deep percolation volumes
were 29.4% and 13.3%, respectively. For this relatively mild slope, the effect of the
recession curve is significant, as reflected by the shift in runoff and deep percolation
volumes.
An adjustment to the recession curve can be made by subtracting the volume of wa-
ter on the surface at the time of cutoff from the total applied volume. Then the applica-
tion time is found from
⎛ V y ( L) ⎞
t co = t A ( L) +τ req − ⎜ ⎟ (14.19)
⎜ Q ⎟
⎝ in ⎠
where Vy(L) is the steady state surface volume for the inflow rate after advance is
complete. For convenience, we use the surface volume computed when advance is just
complete. The recession time at the upstream end is tco and the infiltration opportunity
time at the downstream end is assumed equal to τreq. This results in a cutoff time of
856 min. (14.3 h), an applied volume of 55.7 – 3.3 = 52.4 m3, and an application effi-
ciency of 61.0%. Here the recession curve is not horizontal, but can be assumed linear
between the upstream and downstream ends (Table 14.6). This assumption, with ad-
vance and recession computed over eight intervals, gives runoff and deep percolation
of 28.3% and 10.7%, respectively. From simulation, this design is closer to matching
the required depth at the downstream end, as shown in Figure 14.2, than assuming
instantaneous recession. Further, simulation resulted in runoff and deep percolation
volumes of 26.9% and 12.2%, respectively. It appears that the inflow volume could be
reduced even more than the volume on the surface. However, this simple procedure
provides reasonable results and is generally on the conservative side. For furrows with
a steeper slopes, this adjustment will be small and is not necessary.
With the runoff percentage significantly higher than deep percolation percentage,
one would expect a higher potential application efficiency with a smaller flow rate.
Repeating these calculations at 0.7 L/s results in an application efficiency of 64.4%, tL
= 808 min (13.5 h), tco = 1182 min (19.7 h), and runoff and deep percolation of 18.1%
and 17.5%, respectively. A field is typically broken up into an integer number of irri-
gation sets, with the total inflow divided among the furrows. Thus furrow flow rate is
not a continuous variable, but typically can only take on discrete values. A field is also
broken into an integer number of lengths, making furrow length also not a continuous
variable for design.
14.3.2 Design of Sloping Furrows with Cutback

The efficiency of furrow irrigation systems can often be improved by reducing the
inflow rate after water has advanced to the end of the field. A high initial flow rate can
provide rapid advance, and thus more uniform opportunity time, while cutting back
the stream will reduce the amount of runoff. If the cutback stream is too small to keep
up with infiltration, recession can occur at the downstream end and move back up the
field. Rather than cutting off at the completion of advance, cutoff is typically delayed
until infiltration is somewhat reduced. This will also help assure that the downstream
end receives sufficient flow depth and wetted perimeter.
A common practice is to cut back to 50% of the inflow. Dividing the cutback in-
flow rate by the wetted field area gives the average infiltration rate that matches the
cutback inflow
Q
iCB ≤ CB (14.20)
WL
The average infiltration rate at any time after completion of advance can be com-
puted from numerical integration over the distance (e.g., the eight intervals used
above). A direct analytical solution for average infiltration rate is not possible, but a
numerical approximation can be found by integrating infiltration rate over distance.
Infiltration at any point and any time is a function of the infiltration time, or the cur-
rent time minus advance time. The advance time is a function of xh, from Equation
14.16. If this term is replaced with a truncated series expansion, with higher order
terms removed (e.g., xh = 1 + h(x – 1)), an analytical expression for the infiltration
rate, averaged over the field length, can be found, namely
⎡⎛ ⎧ a
⎞ ⎛ ⎞ ⎤
a
aktCB ( a −1) ⎢⎜ ⎨ tCB ⎫⎬ − 1 + h ⎟ − ⎜ ⎧⎨ tCB ⎫⎬ − 1⎟ ⎥
iCB ≈b+ (14.21)
ah ⎢⎜⎝ ⎩ t L ⎭ ⎟ ⎜ t
⎠ ⎝ ⎩ L ⎭ ⎠ ⎥⎦
⎟
⎣
where tCB is the time of cutback. The cutback time to achieve the necessary average
infiltration rate can be determined from Equation 14.21 by trial and error.
This is a very conservative estimate of cutback time since the reduction in flow re-
sults in less surface storage on the field, and this change in surface storage can con-
tribute to infiltration. A conservative estimate of the correction in cutoff time can be
found by dividing the change in surface volume by the cutback flow rate, which is
related to the distance averaged infiltration rate through Equation 14.20. The adjusted
cutback time is simply
V y (Qin ) − V y (QCB )
tCB adj = tCB − (14.22)
QCB
where the surface volume after completion of advance is now a function only of flow
rate.
The cutoff time is computed according to Equation 14.19, but with Vy(Qin) replaced
with Vy(QCB). Because less volume is on the surface during cutback, the cutoff time is
actually slightly longer. For some soils, the reduction in wetted perimeter caused by
cutback might require an increase in total application time.
Example 2. For a 50% cutback flow of 0.5 L/s, Equation 14.20 indicates that the
average infiltration rate would need to be below 4.5 mm/hr:
0.5L/s 3600sec 1m 3 1000mm
× × × = 4.5mm/hr
400m × 1m 1hr 1000L m
From Equation 14.22, the time of cutback is 582 min. (Trial and error solution with
numerical integration of the average infiltration rate over eight intervals gives tCB =
609 min.) The surface volume for the cutback flow is 1.98 m3. Equation 14.22 gives
an adjusted cutback time of 538 min. The adjusted cutoff time, from Equation 14.19
with Vy(QCB) replacing Vy(L), is 863 min. The total inflow volume is now 42.0 m3
rather than 55.7 m3. This gives an application efficiency of 76.1%. Runoff is reduced
to 4.53 m3, or 10.8%, while deep percolation is 13.1%. SRFR simulation gave 9.4%
and 14.4% for runoff and deep percolation, respectively. Advance, recession and the
distribution of infiltrated water are shown in Table 14.7.
Table 14.7. Advance and recession for sloping furrows with cutback design example (Qin = 1 L/s).
0 0 863 863 102
50 25 871 845 101
100 67 879 810 100
150 121 888 767 98
200 182 896 713 95
250 251 904 654 92
300 327 912 586 89
350 408 921 513 85
400 494 929 435 80
14.3.3 Design of Sloping Furrows with Tailwater Reuse

Furrow irrigation systems on sloping fields usually need some runoff in order to
provide a water depth at the downstream end to get sufficient wetted perimeter and
lateral water movement. The design procedures used here ignore this level of detail
and can sometimes give misleading results, particularly when cutback flows are used.
Tailwater reuse provides another mechanism for providing a sufficient flow rate and
water depth at the downstream end, while still improving furrow application effi-
ciency.
Tailwater can be reused in fields which are downstream from the field where the
tailwater is generated (e.g. cascaded from field to field). Alternately, the tailwater
from a field can be collected in a storage pond, pumped back, and reused on the same
field. Improvements in irrigation efficiency resulting from tailwater reuse depend upon
the type of system, the amount of the tailwater that is reused, and the number of times
it is reused.
Solomon and Davidoff (1997) developed equations for determining the combined
efficiencies associated with the reuse of water through a series of areas. Applied to
tailwater reuse systems, the application efficiency of the combined system with reuse,
Ea_RU, can be estimated from the application efficiency from a single irrigation set from
Ea
E a _ RU = (14.23)
⎛ n −1 ⎞
1 − RU ⎜ ⎟
⎝ n ⎠
where RU is the fraction of the applied water that is reused and n is the number of
subunits from which water is reused. For a cascaded reuse system, n is the number of
systems (e.g., individual fields, farms, projects, etc.) irrigated in sequence. For a recy-
cling reuse system, n is infinity and the term (n – 1)/n becomes unity.
With a recycling runoff reuse systems, a very high runoff fraction can result in ex-
cessive pumping. The volume of water pumped or recycled, VRC, relative to the inflow
volume can be estimated from the reused runoff fraction by

RU
VRC = Vin (14.24)
1 − RU
There are two main types of recycling runoff reuse systems: those that recycle run-
off at a rate nearly the same as the runoff rate, and those that recycle at a constant rate.
The advantage of the runoff-rate type is that storage volume is minimized. Two meth-
ods used for these systems are to cycle the pump on and off at short intervals, the ex-
act interval related to the rate of runoff. Maximum recommended cycle rates for these
systems are 15 times per hour. Sump surface area depends upon the maximum runoff
rate and the allowable depth fluctuations between cycles. The other method for runoff-
rate recycling systems is to use variable speed pumps (Trout and Kincaid, 1994).
For these systems, runoff should be applied to another field or a different irrigation
set, or pumped to a larger, upstream reservoir. It is not recommended to recycle water
back into the supply stream for the field that is creating the runoff, unless additional
furrows are started as return flow increases (ASAE, 1997). Using a variable return-
flow rate to irrigate additional furrows is difficult to manage and increases labor, and
therefore is not recommended. Sump sizes for these systems are on the order of tens of
cubic meters.
The intent of a constant flow recycling system is to provide a constant flow rate to
all furrows. This can be accomplished by pumping water from a reuse storage reser-
voir and adding it to the normal irrigation stream from the beginning of the irrigation.
The return-flow rate should be such that the return-flow volume for one set is ap-
proximately equal to the runoff volume for one set. Then the reservoir will be at ap-
proximately the same level at the end of the irrigation as it was at the beginning. This
essentially requires a full reservoir at the start of the irrigation. Also, there must be a
sufficient volume of water in storage to provide the return flow until the runoff rate
from the first set matches or exceeds the pumping rate. This operating scheme might
be practical where several fields share the same return-flow system and reservoir.
However, for many systems this is not practical.
Rather than starting with the reservoir full, the same general scheme can be used
with the reservoir initially empty. This is accomplished by splitting the width of the
first set into a portion irrigated with the supply inflow only and a portion irrigated with
the reuse portion only. The second part of the first set is actually irrigated after the
main supply flow is turned off and irrigated with return flow only (i.e., it is moved to
the end of the irrigation and the end of the field). This scheme was first proposed by
Stringham and Hamad (1975).
For a given field (length, width, slope, infiltration) and a given inflow rate Qin, the
design of these systems is based on balancing three parameters: the desired infiltrated
depth (e.g., average, di or low quarter, dlq); the individual furrow flow rate, q; and the
number of irrigation sets across the field width, N (an integer number). These cannot
be selected independently, but must assure that the set runoff and return-flow volumes
approximately match. Once these three are determined, the other design parameters
can be found. A number of alternative combinations can be examined to find a worka-
ble combination (e.g., optimum).
The number of sets, the individual furrow flow rate, Qf, and relationship between
the supply inflow rate, QS, and the return-flow or pumpflow rate, QP, should satisfy:
⎛Q Qp ⎞
F = N⎜ s +
⎜ Qf Qf ⎟
(
⎟ = N ns + n p ) (14.25)
⎝ ⎠
where F is the total number of furrows in the field, N is the number of sets, and ns and
nf are the number of furrows in one set irrigated with the supply and pumpback flows,
respectively. The selection of return-flow rate must assure that there is sufficient run-
off volume which is satisfied when
Qp ⎛ RO ⎞
RO ≥ or QP ≤ ⎜ ⎟QS (14.26)
QS + Q P ⎝ 1 − RO ⎠
where RO is the runoff fraction (i.e., fraction of inflow that runs off for an average
furrow). If this equation isn’t satisfied, there may not be enough water in the reservoir
to irrigate the last set. When this inequality is satisfied as an equality, the pumping rate
is the maximum allowable pumping rate, QP-max. If the pumping rate is less than the
maximum allowable, a portion of the runoff volume will not be returned to the same
field. Here, we assume that this portion of the runoff volume is unused or a loss.
The reservoir volume, VR, needed for operation of such systems is the amount of
water required in the reservoir to irrigate the last set with only return flow,
VR ≥ QP (1 − RO)t co ≤ Qs ROt co (14.27)
where tco is the application time for each set. When Equation 14.26 is an equality, this
volume equals the volume of runoff from the first set, irrigated without return flow,
and the right inequality of Equation 14.27 is an equality. Note that the volume re-
quired in the reservoir at the start of this last set is less than the volume needed to irri-
gate the last set. For that set, its own runoff is recirculated to provide the necessary
volume. The runoff volume lost is equal to the total volume applied times the differ-
ence between the actual runoff fraction and the reused runoff fraction Vin(RO –RU).
The method presented above for determining sloping-furrow advance and recession
curves can be used to determine the runoff fraction for a given furrow flow rate and
application time. This information can be used to determine whether this satisfies the
desired depth of infiltration and Equations 14.25 and 14.26. The application efficiency
can be adjusted according to Equation 14.23 with n = ∞ and RU = RO(QP/QP-max), or
some fraction thereof to account for return-flow system losses. This assumes that only
the water pumped during this irrigation event is reused.
Example 3. Consider the example given in the previous sections. In this case, there
are a total of 800 furrows (800 m at 1 m/furrow) and the available flow rate is 80 L/s.
An initial trial at 1 L/s/furrow gives the advance and recession curves of Table 14.6
and nS = 80 furrows. From Equation 14.26, the maximum return-flow pumping rate is
31.6 L/s. Choosing QP = 31.6 L/s results in nP = 31.6 and N = 7.2 (800/[80 +31.6]),
which is not an integer and so is not feasible. If the solution were feasible, we would
get RO = 28.3%, PEa = 85.1% (from Equation 14.23), and a reservoir volume of 1188
m3. One way to make this solution feasible is to choose a smaller pumping rate. At QP
= 20 L/s, nf = 20 and N = 8, and a smaller reservoir is needed. However, then only
20/31.6 or 63% of the runoff is reused. If only the reused volume were considered
useful, then PEa would drop to 74.3% with a corresponding runoff volume lost would
be 3480 m3 (Table 14.8). This emphasizes the need to satisfy the inequality in Equa-
tion 14.26 as close as possible to an equality.
A number of adjustments can be made in the other variables to produce an integer
number of sets, with a better fit to Equation 14.26. Table 14.8 provides the results of
changing various parameters on PEa, VR, and VRO-lost. If the inflow could be increased
to 82 L/s, a feasible solution would result (Table 14.8, run 3). The target application
depth could be changed, either increased slightly or decreased slightly (runs 4 and 5).
The lower efficiency for run 5 reflects the lower uniformity associated with the re-
duced target depth. Or, the individual furrow flow rate could be adjusted to arrive at
feasible solutions, as in run 6. In run 7, the flow rate decrease did not provide a very
useful solution without also a change in the required depth; run 8. Selection of the best
alternative from among these and other potential solutions should be based on eco-
nomic considerations; labor, reservoir construction, pumping plant facilities, etc. Ef-
fective operation of these systems requires a matching of runoff to return-flow pump-
ing rate. Monitoring of runoff volume can be used to adjust set duration and/or pump-
ing rate to assure efficient use of the runoff (i.e., a balancing of Equation 14.26).
Table 14.8. Examples of solutions for tailwater recovery system design.
QS dreq Qf tco RO QP_max QP nS nP N RU PEa T VR VRO-lost
Run (L/s) (mm) (L/s) (min) (%) (L/s) (L/s) (%) (%) (hrs) (m3) (m3)
1 80 80 1 874 28.3 31.6 31.6 80 32 7.2 28.3 85.1 104 1188 0
2 80 80 1 874 28.3 31.6 20 80 20 8 17.9 74.3 117 752 3480
3 82 80 1 874 28.3 32.4 32 80 32 7 28.0 84.7 102 1203 96
4 80 83 1 922 30.4 35.0 34.0 80 34 7 29.6 85.2 108 1309 256
5 80 68 1 713 20.1 20.1 20 80 20 8 20.0 79.5 95 684 21
6 80 80 1.2 720 34.3 41.8 40 67 33 8 32.8 85.9 103 1135 510
7 80 80 0.97 890 27.5 30.2 17.6 82 18 8 15.9 73.1 119 681 3952
8 80 85 0.97 973 31.0 35.9 31.4 82 32 7 27.1 81.9 114 1265 1268
14.4 BORDER-STRIP IRRIGATION

With border-strip irrigation, application efficiency typically is reduced when ad-
vance is either too fast or too slow. Thus design needs to provide a layout such that
inflow rate and time can be adjusted within reason to provide satisfactory perform-
ance. The Soil Conservation Service (USDA, 1974) provided the following recom-
mendations. The maximum recommended inflow rate to limit erosion on non-sod-
forming crops, such as alfalfa and small grain, is found from
qin = 0.00018 S 0−0.75 (14.28)

max
where S0 is in m/m and qin is in m2/s. For sod-forming crops, twice this value can be
used. A minimum inflow rate has also been suggested so that the water depth will be
sufficient to spread laterally:
0.000006 L S10/ 2
qin = (14.29)
min n
Border-strip irrigation is typically practiced on slopes less than 0.05 m/m (5%). On
fine-textured soils, slopes are typically less than 0.01 m/m. The maximum slope based
on the criteria for minimum flow depth (and discharge) can be found by solving Equa-
tion 14.29 for S0. This does not consider erosion potential.
14.4.1 Design of Open-Ended Border Strips
For short border strips and steep slopes, the minimum depth infiltrated can be at the
head end of the strip. However, numerous runs with the BORDER design program
indicated that for most situations when near the peak potential efficiency, the mini-
mum depth was at the downstream end. Even when the minimum was at the upstream
end, the range of low-quarter depths was split between the upper and lower ends. Thus
design based on the downstream end should give more consistent results over the
range of typical design conditions. Since, in addition, estimates for potential efficiency
are needed, recession lag-time design (or design based on satisfying the requirement at
the upstream end, as in USDA [1974]) is no longer recommended.
Simple volume-balance procedures can be used for border strip design, as was done
for furrow design, under the assumption of a surface shape factor. For sloping borders,
we can assume that the flow depth at the upstream end approaches normal depth. The
normal depth can be computed from Equation 14.8 with q as the unit inflow rate and Sf
set to the bottom slope, S0. Solving for y gives
3/5
qin (n / Cu )3/5
y0 = (14.30)
S 03/10
As with furrows, Vy = σy y0 W, where σy ≈ 0.7. The remaining problem in solving the

volume balance is computing the recession curve, where the recession time varies with
distance.
The recession time at the downstream end is set so that the required depth is just
satisfied there, as in Equation 14.12. An empirical relationship, adapted from Walker
and Skogerboe (1987), is used to determine the cutoff time for this required down-
stream recession time. An approximate upstream recession time is found from
0.666 n 0.4757 S 0y .2074 L0.6829

Δt R = t R ( L) − t R (0) = (14.31)
I 0.5244 S 00.2378
where all units are in meters and seconds, and Sy is

0.6
1 ⎡ (qin − IL)n ⎤
S y= ⎢ ⎥ (14.32)
L ⎢⎣ S 0.5
0 ⎥⎦
and I is the infiltration rate (m/s) at tR(0)S, averaged over the length. For the branch
infiltration function after the branch point, it is simply b. For the other infiltration
functions, the infiltration rate can be numerically integrated or approximated by aver-
aging the values at the upstream and downstream ends, τ(0) = tR(0)S and τ(L) = tR(L) –
tA(L). Equation 14.31 is essentially an empirical fit to a series of computer runs over a
wide variety of conditions. It is based on an estimate of the upstream recession lag
time from (Strelkoff, 1977), which results in a cutoff time of
y0 L
tco = t R (0) S − (14.33)
2 q in
Solution of Equations 14.31 and 14.32 for tR(0)S is essentially a trial-and-error
process if the average infiltration rate is not fixed (i.e., with the Kostiakov branch
function, the trial and error is not needed).
While the upstream recession time from Equation 14.31 is needed for the procedure
used to compute cutoff time, it does not give a very realistic estimate for the actual
upstream recession time at small slopes. The following equation can be used to esti-
mate the upstream recession time (adapted from Hart et al., 1980):
q0.2
in n
1.2
t R (0) = tco + 1.6
(14.34)
⎡ ⎛ 0.345 n q0.175 ⎞⎤
⎢S 0 + ⎜ in ⎟⎥
⎢ ⎜ τ 0.88 S 1/2 ⎟⎥
⎣ ⎝ req 0 ⎠⎦
where units are in meters and seconds. The recession lag times for steeper slopes are
generally very small and either equation gives reasonable results. For smaller slopes
(e.g., <0.004), Equation 14.34 generally gives the best estimate (thus it is recom-
mended over Equation 14.33). The procedure given above for computing cutoff times
appears to be valid over a wide range of slopes even if it gives poor estimates of reces-
sion lag time.
Once the advance and recession curves are computed, the infiltrated volume and
runoff volume can be determined. The above procedures give estimates for the two
ends of the recession curve. For steeper slopes, a straight line through two points gives
reasonable results. For milder slopes, a straight line generally underestimates the vol-
ume infiltrated and overestimates the volume of runoff. If a better estimate of the re-
cession curve for smaller slopes is desired, the following procedure can be used.
Compute the slope of the recession curve from
d tR ⎡ 3 σ y y n t R ( L) − t R (0) ⎤
= max ⎢ , ⎥ (14.35)
dx ⎣⎢ qin L ⎦⎥
with the restriction that the computed recession time at any distance not exceed the
recession time at the downstream end. The second term in the brackets of Equation
14.35 is just a straight line between the recession time, tR(L) computed from Equation
14.12 and tR(0) computed from Equations 14.31 and 14.32. The first term assumes that
recession progresses at a rate which removes the surface volume, linearly, at one-third
the inflow rate.
Example 4. Consider the same conditions as in the sloping-furrow design example,

but with the infiltration function from Table 14.3, with k = 40.1 mm/hra and a = 0.51.
With this infiltration function the infiltration opportunity time required to infiltrate
80 mm is 232 min. The other particulars are Manning n = 0.15, σy = 0.75, and q = 2.5
L/s/m. With a slope of 0.002 m/m the maximum and minimum recommended flow rates
from Equations 14.28 and 14.29 are 19 L/s/m and 0.7 L/s/m, respectively. The same
assumed surface-volume method is used to compute advance, as shown in Table 14.9.
Table 14.9. Advance calculations for assumed surface-volume

border-strip design example.
Q x tx Vin(t) y0 Vy(t) h σz Vz(t)
(L/s/m) (m) (min) (m3/m) (mm) (m3/m) (m3/m)
2.5 200 108.6 16.29 56.8 8.52 1.38 0.716 7.77
2.5 400 282.0 42.30 56.8 17.03 1.38 0.716 25.28
The desired recession time at the downstream end from Equation 14.11 is tR(L) =
232 + 282 = 514 min. The cutoff time is computed from Equations 14.31 to 14.33 as
follows. Start by assuming that tR(0)S = 514 min. Then compute the infiltration rate at
the downstream end of the field with τopp(L) = 232 min from Equation 14.4, I(L) =
10.5 mm/hr (or 2.93 × 10-6 m/s). Estimate the infiltration rate at the upstream end
equal to the infiltration rate at tR(0)S from Equation 14.4, I(0) = 7.14 mm/hr (or 1.98 ×
10-6 m/s). Next compute Sy from Equation 14.32 (with I = 2.45 × 10-6 m/s), as
0.6
[(0.025 m 2 /s − 2.45 × 10 −6 m/s × 400 m)0.15]
S y= = 0.000105
400 m × 0.002 0.3
From this compute the upstream recession time from Equation 14.31 as
0.666 × 0.15 0.47565 × 0.000105 0.20735 × 400 0.6829 1 min
t R (0) S = 514 − 0.52435
= 514 min − 150 min = 364 min
(2.45 × 10 − 6 ) S0
0 .237825
60 sec
which is different from our assumed value. We recalculate the average infiltration rate
with τ(0) = 364 min and τ(L) = 364 – 282 = 82 min, which gives us 8.5 and
17.6 mm/hr, averaging 13.0 mm/hr (or 3.61 × 10-6 m/s). Then recompute Sy and tR(0)S,
until they converge. The final solution is I = 11.9 mm/hr, Sy = 0.0000903, and tR(0)S =
386 min. (If we integrate the infiltration rate over eight intervals, the final solution is
tR(0)S = 376 min.)
In these iterations, only the average infiltration rate changes. The method assumes
that recession starts at the upstream end after completion of advance. If this is not the
case, then the infiltration at the downstream end is zero.
The cutoff time is determined from Equation 14.33 as
400 m × 0.0568 m 1 min

tco = 386 min − × = 386 − 76 = 310 min
2
2 0.0025 m /s 60 sec
The actual upstream recession time is found from Equation 14.34, as
0.0025 0.2 × 0.15 1.2 1 min

t R (0) = 310 + 1.6
× = 310 + 10 = 320 min
⎡ ⎛ 0.345 × 0.15 × 0.0025 0.175 ⎞⎤ 60 sec
⎢0.002 + ⎜ ⎟⎥
⎢⎣ ⎜ (232 × 60 ) 0.88 × 0.002 1/2 ⎟⎥
⎝ ⎠⎦
Figure 14.3 shows the advance and recession curves for this example. The advance
curves for the assumed surface-volume advance is a reasonably close match to ad-
vance from unsteady flow simulation (e.g., SRFR). The recession at the downstream
end is well predicted by Equations 14.31 and 14.32. The recession time at the up-
stream end is well predicted by Equation 14.34. However, the recession curve itself is
not all well predicted. Using a straight line from the recession lag time from Equation
14.33 (i.e., suggested by Walker) to the downstream recession time matches the curve
fairly well except at the upstream end. Equations 14.34 and 14.35 are reasonable for
this example, but are not satisfactory for other conditions. If the difference in the up-
stream recession time computed by the two methods is less than 10% to 15%, use the
simplified straight line, otherwise the recession curve should be computed.
The volume applied for this example is 46.6 m3/m, whereas the target volume is
32 m3/m, giving an application efficiency of 68.7% (32/46.6). Use of Equations 14.34
and 14.35 for advance and recession computed over eight intervals (Table 14.10) re-
sulted in deep percolation and runoff percentages of 14.4% and 16.9%, respectively.
(The simplified procedure of Walker resulted in 13.3% and 18% for these two per-
centages.) Unsteady flow simulation with a cutoff time of 312 min gave these same
percentages for deep percolation and runoff, however even with slightly more water
applied, the target depth was not quite satisfied at the downstream end (Ea was 68.4%
and minimum depth was approximately 79 mm). Either procedure produces reason-
able results in most cases.
600
500
400
Time (min)
300 Equation
Equation
200 Walker
SRFR
SRFR
100
0
0 50 100 150 200 250 300 350 400
Distance (m)
Figure 14.3. Advance and recession curves for border-strip design example.
Table 14.10. Advance and recession for border-strip design example.
0 0 321 321 94
50 20 363 344 98
100 45 406 361 100
150 75 448 374 102
200 109 491 382 103
250 146 514 368 101
300 188 514 326 95
350 233 514 281 88
400 282 514 232 80
14.4.2 Design of Blocked-End Border Strips

Improvement in application efficiency can be obtained by blocking the downstream
end of the border strip. This should only be done where the ponding depth, and associ-
ated infiltration time, will not cause crop damage. To limit crop damage, the end is
sometimes partially blocked to limit the maximum ponded depth (e.g., by the elevation
of overspill) or the maximum ponding time (e.g., by leaving a breach in the dike to
allow it to eventually drain).
Where all the runoff is contained, the distribution of infiltrated water can be modi-
fied by assuming that the volume that ran off is ponded at the downstream end. Ad-
vance, recession, and the distribution of infiltrated depths are computed as it is for an
open-ended border strip, then a ponded depth is simply added to the infiltrated depth at
each location. The length of ponding can be found from
2VRO
Lp = (14.36)
S0
where VRO is the runoff volume. The ponded depth is zero at a distance Lp from the
downstream end and is S0Lp at the downstream end.
In the design procedures of the Soil Conservation Service (USDA, 1974), the field
length is adjusted to account for the ponding. In this design procedure, the cutoff time
is adjusted, by trial and error, until the minimum depth matches the required depth.
This will either occur at the upstream end or at a distance Lp from the downstream end.
Example 5. For the example given above, the runoff volume is 7.79 m3/m. From
Equation 14.36, the ponding length is 88.3 m, or 311.7 m from the upstream end. The
additional depth of ponding at the downstream end is 177 mm. By interpolation from
Table 14.10, the depth infiltrated at this point is 94 mm, much greater than the 80 mm
required. Thus a significant decrease in application time can be made. The simplest
approach for this design procedure is to decrease the recession time at the downstream
end for the open-ended design procedure. If the decrease were proportionate to the
minimum depth, the decrease in application time would be roughly 46 min (310 min ×
(94 – 80)/94 = 46 min). The required recession time at the downstream end is reduced
from 514 to 468 min. The recession time at the upstream and downstream ends must
now be recomputed, as described above, with the solution tR(0)S = 352 min, tco = 277
min, tR(0) = 287 min. The runoff volume is now 5.11 m3/m, which results in a ponding
length of 71.5 m, and a minimum depth infiltrated there of 84 mm. In this case, the
minimum depth is still at the upstream end of the ponded section. In other cases it will
be at the upstream end of the field.
After several iterations, the minimum depth matches the requirement with a reduc-
tion in open-ended downstream opportunity time of 62 min, with solution tR(0)S = 342
min, tco = 266 min, tR(0) = 276 min, VRO = 4.33 m3/m, and Lp = 65.8 m. The final dis-
tribution of infiltrated water is given in Table 14.11. The very large depth infiltrated at
the downstream end, 200 mm, suggests that the slope on this field may be too steep for
complete ponding. While the application efficiency in this case is 80.2%, this large
ponded depth may be unacceptable.
Table 14.11. Advance and recession for ponded border-strip design example.
Open-Ended Open-Ended Open-Ended Ponded
Advance Recession Opportunity Infiltrated Infiltrated
Distance Time Time Time Depth Depth
(m) (min) (min) (min) (mm) (mm)
0 0 276 276 87 87
50 16 318 302 91 91
100 42 361 319 94 94
150 73 404 331 96 96
200 109 446 338 97 97
250 148 452 305 92 92
300 190 452 263 85 85
334 220 452 232 80 80
350 235 452 218 77 109
400 282 452 170 68 200
14.5 LEVEL-BASIN AND LEVEL-FURROW IRRIGATION

With level-basin irrigation, rapid advance will produce a high uniformity. The de-
sign of basin irrigation systems is based on providing rapid advance, but without ap-
plying excessive amounts of water. Since the field is level, soil erosion is only a con-
cern where water is turned into the field, i.e. locally. Minimum and maximum flow
rates are not specified, but are dictated by the hydraulic conditions. With level basins,
flow depth can become high and needs to be examined in design.
14.5.1 Design of Flat-Planted Level Basins
Walker and Skogerboe (1987) provide a straightforward volume balance approach
for level basins, again by assuming a surface-volume shape factor. In order to solve
the volume balance equation (Equations 14.13 and 14.14), they compute the upstream
depth, y0, for advance to some distance x from the Manning equation by assuming σy =
0.8 and by assuming
y0
Sf= (14.37)
x
Combining Equations 14.8 and 14.37 and solving for y0 gives the flow depth at the
upstream end as
3 / 13
2
y 0 = (qin ( n / Cu ) 2 x ) (14.38)
For a given flow rate, advance time can be found iteratively from
V y ( x) V z ( x,t x)
tL= + (14.39)
Qin Qin
where Vy(L) = σy y0WL and Vz is found from Equation 14.15, with x = L and is a func-
tion of advance time. The SCS (USDA, 1974) used a similar procedure, but ignored
the last term in Equation 14.39. Two advance distances and times are required so that
the advance exponent h can be determined, typically associated with the field length
and one-half the field length.
Yet to be determined are tco and PEmin. The SCS assumed a relationship between
advance ratio, AR = tL/τreq, and PEmin. With this assumption, they were able to generate
a series of design charts (USDA, 1974). These charts plotted curves of PEmin as a func-
tion of unit flow rate (ordinate) and length (abscissa). However, Clemmens and
Dedrick (1981) showed that PEmin varied widely with the advance ratio, particularly
when the infiltration exponent varied from that assumed by the SCS intake families.
Walker and Skogerboe (1987) assumed that after advance, the increase in infiltrated
depth throughout the basin is the same as the required depth. This actually results in a
very conservative estimate of PEmin.
However, if we assume that recession occurs at the same time throughout the basin,
we can integrate the infiltrated volume over distance. A direct solution is not possible,
but an approximation can be obtained by representing the power advance function
with a series expansion. Taking the first two terms in the expansion results in the fol-
lowing equation for the final infiltrated volume:
⎧⎪ a [(1+ hAR )
a +1
− 1] ⎛ hA ⎞⎪⎫
V z = LW ⎨c +kτ req + bτ req ⎜1+ R ⎟⎬ (14.40)
⎪⎩ hAR (a + 1) ⎝ 1+ h ⎠⎪⎭
However, if we use the branch infiltration function, we get an exact solution, pro-
vided that τreq > τB. The resulting equation is
⎧ bht A ( L) ⎫
V z = LW ⎨d req + ⎬ (14.41)
⎩ (h +1) ⎭
The cutoff time is found by dividing this volume by the inflow rate, Qin. For very
small values of AR, this equation is not appropriate, however in such cases the cutoff
time can be based solely on required volume (i.e., PEmin = 100%).
This procedure assumes that advance is complete prior to cutoff. For large level ba-
sins as used in the U.S., this is frequently not the case, particularly when flow resis-
tance is high (e.g., alfalfa or grass). One of the biggest errors associated with this pro-
cedure is that the surface volume during advance is large. This is particularly true
when cutoff precedes completion of advance.
Example 6. The assumed surface-volume method given above was applied to the de-
sign example for border strips, except that the slope is zero. (Such light application
depths are more difficult to apply with level basins unless the length is significantly re-
duced. Variations in field surface elevations also play a role in potential efficiencies if
attempts are made to apply too little water.) The length is reduced to 200 m and the flow
rate to 2.0 L/(s m). The surface shape factor is 0.80. The advance curve computed with
the assumed surface-volume method (Equations 14.38 and 14.39) is given in Table 14.12.
From Equation 14.40, the total volume applied is 20.23 m3, resulting in a cutoff
time of 169 min and PAEmin = 79.1%. If we assume that recession occurs at the time
required to infiltrated the desired depth at the far end (195.4 + 232.4 = 428 min), op-
portunity times at various points along the basin can be computed by subtracting the
advance time from this recession time, as shown in Table 14.13. Using these com-
puted points to estimate the volume infiltrated gives (with eight-point numerical inte-
gration) PAEmin = 80.5%, suggesting that Equation 14.40 is a reasonable estimate for
the required volume. (Walker and Skogerboe’s method gives PAEmin = 60.1%.)
Table 14.12. Results for assumed surface-volume level-basin design.

Q x tx Vin(t) y0 Vy(t) Vz(t)
(L/(s m)) (m) (min) (m3/m) (mm) (m3/m) h σz (m3/m)
2.0 100 72.2 8.66 68.5 5.47 1.44 0.723 3.18
2.0 200 195.4 23.45 80.4 12.86 1.44 0.723 10.58
Table 14.13. Advance and recession for level-basin design example.

0 0 428 428 109
25 8 428 420 108
50 21 428 407 106
75 38 428 390 104
100 58 428 370 101
125 80 428 348 98
150 104 428 324 95
175 131 428 297 91
200 195 428 232 80
500
400
Time (min)
Vol-Bal Advance
300 Vol-Bal Recession
SRFR Advance
200 SRFR Recession
100
0
0 50 100 150 200
Distance (m)
Figure 14.4. Advance and recession curves for level-basin design example.
For this example, the advance time exceeds the application time. This assumed sur-
face-volume method likely overestimates PAEmin. The BASIN design program for this
set of conditions gives PAEmin = 78.4%, tA(L) = 211 min, tco = 170 min, and cutoff at
advance to 88% of the basin length (R = 0.88). This is close to the limit line presented
by Clemmens and Dedrick (1982), which occurs at approximately R = 0.85, and be-
yond the limit in BASCAD for design. While the differences are small in this case
(2% change in PAEmin), errors increase rapidly as cutoff occurs at shorter advance dis-
tances. A comparison of advance and recession curves for the assumed surface-
volume method and unsteady-flow simulation are given in Figure 14.4.
14.5.2 Design of Level Furrows and Furrowed Level Basins

The same assumed surface-volume method can be used for level furrows or fur-
rowed level basins as was used for flat-planted level basins, except Equation 14.38 is
replaced with (from Equations 14.7 and 14.37)
A02 R04 / 3 y0 = qin

2
( n / Cu )2 x (14.42)
Since A0 and R0 are functions of y0 as defined by the furrow cross-section shape,

determination of A0 for the volume balance calculation is iterative rather than direct
(i.e., it adds one more internal iteration loop to the procedure). While the design pro-
cedures for the two systems are essentially the same (i.e., based on advance and reces-
sion of a single furrow), the cultural and irrigation operations for level furrows and
furrowed basins are quite different.
Example 7. The infiltration and roughness conditions for the sloping-furrow irriga-
tion example are used here: k = 40 mm/hra, a = 0.35, and n = 0.05. The furrow spacing
is 1 m, and as before, the furrow bottom width is 100 mm with 2:1 side slopes (hori-
zontal:vertical). In this case, we start the design assuming 2 L/s per furrow and a
length of 200 m.
The design starts with a solution for the depth at the upstream end as a function of
distance x, from Equation 14.42. This relationship does not depend on the advance
time. At 100 m, we guess a depth of y0 = 0.1 m. For a trapezoidal shape, this results in
an area A0 of 0.0485 m2, a wetted perimeter of 0.6935 m, and a hydraulic radius R0 of
0.0699 m. Solving Equation 14.42 for y0 gives 0.0148 m. This new estimate of y0 does
not provide a good value for the next guess (i.e., the solution diverges). Instead, we
use 0.8 times the original guess plus 0.2 times the value computed from Equation
14.42; 0.8(0.1) + 0.2(0.0148) = 0.1091 m. After three iterations, the solution con-
verges to 0.0891 m. At 200 m, solution with this procedure gives y0 = 0.1012 m.
The surface volume for advance to these two distances is found from Equation
14.14, with σy = 0.80, as 1.983 and 4.897 m3, respectively. The subsurface volumes
are a function of time and the advance exponent. Initial guesses of 60 and 250 min
give h = 1.32, and from Equation 14.15, subsurface volumes of 3.107 and 4.897 m3,
respectively. The inflow volumes at these times are 7.2 and 18.0 m3, respectively, or
volume errors of more than 20%. From Equation 14.39, new estimates of advance
time are 42 and 122 min, respectively. The solution eventually converges to 39 and
104 min, respectively, h = 1.415 and an advance ratio AR = 0.239.
From Equation 14.40 with a target depth of 80 mm, the ultimate infiltrated volume
is 16.89 m3, which occurs at a cutoff time of 140.7 min, or approximately 37 min after
the completion of advance. The potential application efficiency for this case is 94.7%.
Numerical integration with eight points on the advance and recession curve, assuming
recession occurs at the time needed to infiltrate 80 mm at the downstream end, gives
an infiltrated volume of 16.75 m3, an application time of 139.5 min and a potential
application efficiency of 95.6%. Clearly in this case, either a longer basin or a smaller
application depth can be applied. Further application of this procedure at different
lengths, furrow flow rates, and application depths should be explored to arrive at an
economic design.
14.6 SURFACE IRRIGATION SYSTEM HEADWORKS

AND CONTROL OF INFLOW
The headworks for a surface irrigation system are used to convey and distribute wa-
ter from the supply point to individual irrigation sets, furrows, etc. They provide a
transition from the supply to the field, and must be compatible with both the type of
supply available and the type of surface irrigation system in use. There are two main
types of supply: from a farmer-operated groundwater well and from an irrigation dis-
trict. The irrigation district supply may be from a pressure pipeline or from an open
canal or pipeline. Flow from a groundwater well is typically constant and non-
adjustable. Flow from an irrigation district with open canals or pipelines is typically
variable and not under control of the irrigator. Flow from an irrigation district pressure
pipeline may be adjustable by the irrigator (e.g., by applying backpressure on the out-
let).
Farm field conveyance and distribution works may be a pipeline or an open ditch.
Water from groundwater can be pumped into a field pipe distribution system or an
open canal. Where sufficient head is available, water from canals and open pipelines
can also supply field pipe distribution systems (e.g., gated pipe). However, debris must
be kept out of the pipeline since it could clog outlets. Turbulent fountain screens are
an effective way to achieve this (Bondurant and Kemper, 1985).
Where earthen farm ditches are used, water can be distributed to individual furrows
with siphon tubes or spiles; or to individual borders or basins through siphon tubes,
pipe outlets (such as large spiles that are closable) or cuts in the canal wall. For con-
crete ditches, options for water distribution include siphon tubes (for all methods),
pipe outlets (for borders and basins), or open gated outlets (e.g., with jack gates). For
farm pipe conveyance and distribution systems, alfalfa valves are common for borders
and basins, while gated pipe systems are most common for furrows.
The selection from among these various alternatives should be based on compati-
bility with the supply and water control to the irrigated field. This latter must consider
the available head of the water above the field surface, which may change substan-
tially over the farm, so that water will flow out of the distribution system into the de-
sired field area and not elsewhere. Systems should typically have 0.15 to 0.3 m of
head available at all points. Outlets should also be non-erosive. This can be a severe
problem with steep slopes and erodible soils, or where very high flow rates are used
(e.g., > 300 L/s).
The headworks must often supply fields that may irrigate both row and field crops,
such that, for example, both furrow and border-strip irrigation are practiced on the
same field. The headworks should not limit this possibility.
Selection of headworks is often dictated by common practices in the area and eco-
nomics. However, modernization or improvement in surface irrigation practices usu-
ally requires better control of inflow and may suggest other alternatives. Control of
inflow rate is necessary to help balance advance and recession curves and to minimize
runoff that is lost. Is some cases, some form of automation will help improve water
control.
Cutback furrow irrigation has not been widely practiced because of the labor in-
volved in making effective cutback. Several methods have been developed to accom-
plish cutback automatically or semi-automatically. Surge flow was developed in at-
tempts to apply cutback flow to gated pipe. With these systems, if the pipe gates are
set to give the desired distribution at full flow (i.e., gate openings vary due to slope,
wheel vs. non-wheel row, etc.), reducing the pipe flow in half will not give each fur-
row half of its previous flow. By cycling the full flow at a relatively short interval
(e.g., 10 min), a more effective cutback is achieved. Automatic surge valves that cycle
water have made surge flow, and the associated cutback, a practical method for im-
proving sloping-furrow irrigation performance.
Cablegation is another method of achieving an effective cutback flow. Under these
systems, gated pipe is placed on a uniform slope. A plug moves at a constant speed
down the gated pipeline. Immediately upstream from the plug, the outlets have essen-
tially full pressure. Further upstream, outlets have a reduced pressure. By allowing the
plug to move downstream, the flow gradually decreases, essentially producing a grad-
ual cutback. Different flow distributions can be produced with different supply inflows
and plug speeds (Kincaid, 1984). Automation only requires control over the plug
speed. These systems are not in common use.
Cutback flows from field ditches was first proposed by Garton (1966), where spiles
for subsequent irrigation sets are located at a lower elevation. When the next set is
irrigated, the current set has a lower head on it, producing the cutback flow. These
have not proven to be practical in the field, partly because of physical limitations (i.e.,
they need to have the correct cross-slope), their relative nonadjustability, and plugging
of the spiles with debris. Canal side weirs or ditch notches have been used effectively
in some areas. Here small weirs, one for each furrow, are made when the concrete
ditch is poured. Flow into a set of furrows is controlled by the water surface elevation
in the ditch. Blocking the canal gives full flow. Cutback flow can be achieved and
controlled by unblocking the canal to allow most of the flow to continue downstream,
but by maintaining the water level (e.g., with check boards). This system is also less
susceptible to weed plugging. These systems have not seen widespread use due to the
cost of construction. (Eftekharzadeh et al., 1987).
Improved management of borders and basins usually requires selection of the right
flow rate and duration. With level-basin irrigation, the tendency has been to increase
the flow rate to speed advance and reduce labor costs. Level basins as large as 15 ha
with flows above 1 m3/s have been used. Outlet erosion control becomes a major issue
for such systems. With fixed-width borders and basins, flow rate adjustments are not
always feasible, making control of duration the only operational control over perform-
ance. A number of methods have been proposed for the automatic control of duration
to such systems, including automatic control of jack-gates, pipe outlets, and drop-open
or drop-closed gates within a concrete-lined supply canal. Most of these use simple
timers to move water from one set to another. Field sensors have also been developed
to sense the arrival of water and signal the change in irrigation set (e.g., based on ad-
vance distance). These sensors are also not in common use.
Drain-back level basins were developed to reduce many of the limitations of the
more traditional, modern level basins used in southwestern Arizona. With these sys-
tems, an earthen ditch is placed below the field grade. When the ditch is blocked, the
water level rises and simply flows into the field. When the ditch is opened, the supply
water is conveyed past the field, below its surface elevation, to the next field. In addi-
tion, some of the applied water drains back off the field just irrigated, thereby reducing
the applied depth. The conveyance ditch is used as the turn row. This system signifi-
cantly reduces the cost of level-basin construction (i.e., no concrete ditch to construct),
makes lighter water applications more feasible, and places less restriction on machin-
ery operations (i.e., one can drive though ditches from one basin to the next) (Dedrick,
1984).
More recently, similar systems have been utilized for level-basin irrigation in high
rainfall areas, including paddy rice production. A common feature is below-grade
ditches on one to three sides. These replace the contour levee system, in that rather
than water flowing from levee to levee through the field, the water flows in the side
ditches from levee to levee. This actually promotes both more efficient irrigation and
fastest surface drainage (Clemmens, 2000).
The headwork facilities are an extremely important aspect in system design since
they often represent a major part of the development cost for a surface irrigation sys-
tem. The surface irrigation design procedures presented herein should be used in con-
junction with an economic analysis of all of the factors that influence the cost of de-
velopment and operations. Consideration should also be given to the future pressure to
reduce water applied and environmental protection. Designs that potentially allow
better water control offer more flexibility to meet these changing demands. More de-
tails on these the considerations in irrigation system selection can be found in Burt et
al. (1999).
REFERENCES
ASAE. 1991. EP419: Evaluation of furrow irrigation systems. St. Joseph, Mich.: ASAE.
ASAE. 1997. EP408.1: Design and installation of surface irrigation runoff reuse
systems. St. Joseph, Mich.: ASAE.
Bondurant, J. A., and W. D. Kemper. 1985. Self-cleaning, non-powered trash screens
for small irrigation flows. Trans. ASAE 28(1): 113-117.
Boonstra, J., and M. Jurriens. 1988. BASCAD A Mathematical Model for Level Basin
Irrigation. ILRI Publication 43. Wageningen, The Netherlands: Int’l. Inst. for Land
Reclamation and Improvement.
Burt, C. M., A. J. Clemmens, R. D. Bliesner, J. L. Merriam, and L. A. Hardy. 1999.
Selection of irrigation methods for agriculture. ASCE On-Farm Irrigation
Committee Report. Reston, Va.: American Soc. Civil Engineers.
Clemmens, A. J. 1991. Feedback control of a basin irrigation system. J. Irrig. Drain.
Eng. 118(3): 480-496.
Clemmens, A. J. 2000. Level basin irrigation systems: adoption, practices, and the
resulting performance. In Proc. 4th Decennial Nat’l Irrigation Symp., 273-282. St.
Clemmens, A. J., and A. R. Dedrick. 1981. Estimating distribution uniformity in level
basins. Trans. ASAE 24(5): 177-1180, 1187.
Clemmens, A. J., and A. R. Dedrick. 1982. Limits for practical level-basin design. J.
Clemmens, A. J., and A. R. Dedrick. 1994. Chapt. 7: Irrigation techniques and
evaluations. In Management of Water Use in Agriculture, 64-103. K. K. Tanji and
B. Yaron, eds. Adv. Series in Agricultural Sciences, Vol. 22. Berlin, Germany:
Springer-Verlag.
Clemmens, A. J., A. R. Dedrick, and R. J. Strand. 1995. BASIN: A Computer Program
for the Design of Level-Basin Irrigation Systems. WCL Report #19. Phoenix, Ariz.:
U.S. Water Conservation Lab., USDA-ARS.
Dedrick, A. R. 1984. Water delivery and distribution to level basins. In Water Today
and Tomorrow, Proc. Am. Soc. Civ. Eng. conf., 1-8. Reston, Va.: American Soc.
Civil Engineers.
Eftekharzadeh, S., A. J. Clemmens, and D. D. Fangmeier. 1987. Furrow irrigation
using canal side weirs. J. Irrig. Drain. Eng. 113(2): 251-265.
FAO. 2005. AQUASTAT online database, FAO’s Information System on Water and
Agriculture. Food and Agriculture Organization, United Nations, Rome, Italy.
Available at: www.fao.org/ag/agl/aglw/aquastat/dbase/index2.jsp.
Garton, J. E. 1966. Designing an automatic cut-back furrow irrigation system. Okla.
Agr. Bull. B-651.
Hart, W. E., H. G. Collins, G. Woodward, and A. S. Humpherys. 1980. Chapt. 13:
Design and operation of gravity or surface irrigation systems. In Design and
Operation of Farm Irrigation Systems. M. E. Jensen, ed. St. Joseph, Mich.: ASAE.
Maupin. 2004. Estimated Use of Water in the United States in 2000. USGS
Circular 1268. Washington, D.C.: USGS.
Kennedy, D. N. 1994. California Water Plan Update. Vol. 1. Sacramento, Calif.: State
of California Dept. of Water Resources.
Kincaid, D. C. 1984. Cablegation: V. Dimensionless design relationships. Trans.
ASAE 27(3): 769-722.
Merriam, J. L., and A. J. Clemmens. 1985. Time rated infiltrated depth families. In
Development and Management Aspects of Irrigation and Drainage, Spec. Conf.
Proc., 67-74. Reston, Va.: American Soc. Civil Engineers, Irrig. and Drain. Div.
Philip, J. R. 1957. The theory of infiltration: 4. Sorptivity and algebraic infiltration
equations. Soil Science 84: 257-264.
Solomon, K. H., and B. Davidoff. 1997. On the relationship between unit and subunit
irrigation performance. Draft copy.
Strelkoff, T. 1977. Algebraic computation of flow in border irrigation. J. Irrig. Drain.
Div., ASCE 103(1R3): 357-377.
Strelkoff, T. 1990. SRFR: A Computer Program for Simulating Flow in Surface
Irrigation Furrows-Basins-Borders. WCL Report #17. Phoenix, Ariz.: U.S. Water
Conservation Laboratory, USDA/ARS.
Strelkoff, T. S., A. J. Clemmens, B. V. Schmidt, and E. J. Slosky. 1996. Border: A
Design and Management Aid for Sloping Border Irrigation Systems. Version 1.0.
WCL Report #21. Phoenix, Ariz.: U.S. Water Conservation Laboratory, USDA-ARS.
Stringham, G. E., and S. N. Hamad. 1975. Design of irrigation runoff recovery
systems. J. Irrig. Drain. Div., ASCE 101(IR3): 209-219.
Trout, T. J., and D. C. Kincaid. 1994. Float-activated variable-speed irrigation
tailwater pump. ASAE Paper No. 942124. St. Joseph, Mich.: ASAE.
USDA. 1974. Chapt. 4, Sect. 15: Border Irrigation. In National Engineering
Handbook. Washington, D.C.: Soil Conserv. Serv., USDA.
USDA. 1984. Chapt. 5, Sect. 15: Furrow Irrigation. In National Engineering
Handbook. Washington, D.C.: Soil Conserv. Serv., USDA.
Walker, W. R. 1989. Guidelines for designing and evaluating surface irrigation
systems. FAO Irrigation and Drainage Paper 45. Rome, Italy: Food and Agriculture
Organization of the United Nations.
Walker, W. R., and G. V. Skogerboe. 1987. Surface Irrigation: Theory and Practice.
Englewood Cliffs, N.J.: Prentice-Hall, Inc.
CHAPTER 15
HYDRAULICS OF
SPRINKLER AND
MICROIRRIGATION
SYSTEMS
Derrel L. Martin (University of Nebraska,
Lincoln, Nebraska)
Dale F. Heermann (USDA-ARS, Fort
Collins, Colorado)
Mark Madison (CH2M Hill, Portland,
Oregon)
Abstract. This chapter presents and discusses the hydraulic characteristics of

closed-pipe irrigation systems needed for the design of sprinkler and microirrigation
systems. Several of the common equations for calculating friction head loss in closed-
pipe systems are presented. The accuracy and limitations of the methods are stressed.
A discussion of irrigation system valves and their proper use is provided for inclusion
in the design of closed-pipe systems that are frequently drained and refilled during
operation. Water distribution in the soil to enhance uniformity is also mentioned.
Keywords. Air and vacuum relief valves, Check valves, Friction resistance, Head
loss, Irrigation control valves, Pipe friction loss, Lateral design criteria, Surge valves.
15.1 INTRODUCTION
The fluid dynamics of sprinkler and microirrigation systems are complex. Water
moves dynamically from the water source through the pump into the pipe network.
Water often goes through a series of screens and filters depending on the source and
type of irrigation system. From the pipe network, water is supplied under pressure to
sprinkler systems and through a sprinkler nozzle into the air at a high velocity where it
breaks up into droplets and falls to the soil or crop surface and is redistributed. Sprin-
klers are mounted at various heights with many types of nozzles that experience di-
verse pressure and wind speeds. The diversity of operating conditions produces a large
range of drop sizes. When the network provides water to a microirrigation system the
water is discharged through emitters that are located on the surface or buried in the
soil. In this chapter, we discuss the flow within the closed-pipe network to either the
sprinkler or microirrigation emitter. The importance of the redistribution on the sur-
face and in the soil is discussed. We emphasize general design considerations with
limited focus on operation. Chapters 16 and 17 present detailed procedures for design-
ing and operating sprinkler and microirrigation systems.
15.2 HYDRAULICS OF PIPE SYSTEMS
Pressurized irrigation systems consist of main lines which convey the entire flow
from the water source to lateral pipelines positioned across the field. Some of the flow
in the main line branches into laterals which direct the discharge to emitters which wet
the immediate soil area or sprinkler heads, and nozzles, which direct the jet stream
into the air. The main line energy losses due to surface resistance can be determined
with uniform flow methods, whereas laterals and outlets with branching flow require
nonuniform flow analysis due to surface resistance and form losses.
15.2.1 Energy Losses in Pipes and Couplers
The Darcy-Weisbach formula, as discussed in Chapter 11, is recommended for cal-
culating energy losses in pipes. The Darcy-Weisbach equation for circular pipes flow-
ing full is given by
L V 2 ⎛⎜ 8 f ⎞⎟ LQ 2
hf = f = (15.1)
D 2 g ⎜⎝ π 2 g ⎟⎠ D5
where hf = the head loss, m
f = the Darcy-Weisbach resistance coefficient
L = the pipe length, m
D = the inside diameter of the pipe, m
V = the velocity, m/s
Q = the flow in the pipe, m3/s
g = the gravitational acceleration, 9.81 m/s2.
The resistance coefficient (f), also called the friction factor, is a function of the rela-
tive roughness (e/D) and Reynolds number (Re). The absolute roughness height (e) is a
constant for a given pipe material. The resistance coefficient is dimensionally consis-
tent and can be represented for most commercial materials by the semi-empirical
equation as
1 ⎛e 9.35 ⎞⎟
= 1.14 − 2 log⎜ + (15.2)
f ⎜D R f ⎟
⎝ e ⎠
where e = the absolute roughness of the inside of the pipe, mm
Re = the Reynolds number (VD/v)
v = the kinematic viscosity, m2/s.
The average flow velocity is equal to the discharge divided by the cross-sectional
area so the Reynolds number for circular pipes flowing full can be computed from
Re = 1.273×106 Q/D (15.3)
534 Chapter 15 Hydraulics of Sprinkler and Microirrigation Systems
where Q is the discharge in L/s, D is the inside diameter of the pipe in mm, and the
kinematic viscosity is 1.0×10-6 m2/s, which represents a water temperature of approxi-
mately 20°C.
The Reynolds number is generally between 300,000 and 700,000 at the inlet for
center pivot or lateral move systems designed with representative system capacities
(Figure 15.1). The Reynolds number decreases along a lateral as water exits through
nozzles or emitters. The flow in a center pivot lateral at a point 90% of the way along
the lateral should be about 20% of the inflow to the system. Thus, the Reynolds num-
ber exceeds 50,000 for most of the lateral for pivots with typical capacities. The Rey-
nolds number for many microirrigation and other types of sprinkler systems are usu-
ally less than for center pivots. Large diameter pipes used for main lines may have
Reynolds numbers larger than shown in Figure 15.1.
The resistance coefficient is dimensionless and consistent with customary (foot-
pound) or metric units; however, Equation 15.2 must be solved iteratively. Table 15.1
presents equivalent sand-grain size roughness values for determining the friction fac-
tor for various pipe materials. General resistance diagrams for determining the friction
factor are available in most fluid mechanics textbooks. Values for the resistance coef-
ficient typical of irrigation systems are presented in Figure 15.2.
The resistance coefficient varies along sprinkler or microirrigation laterals. The
relative roughness depends on the smoothness and inside diameter of the pipe; there-
fore, the relative roughness is constant for a given pipe. However, the flow in a lateral
decreases along the pipeline. For example, consider a center pivot with a galvanized-
steel lateral that has an inside diameter of 163 mm. The roughness (e) is about 0.11
mm from Table 15.1 which gives a relative roughness (e/D) of 0.00068. If the pivot
was designed for an inflow of 51 L/s the Reynolds number at the inlet would be about
400,000. The resistance coefficient at the lateral inlet would be about 0.019 (Figure
15.2). At a point 90% of the way along the lateral the flow should be about 20% of the
inflow to the lateral giving a Reynolds number of approximately 80,000 which pro-
duces a resistance coefficient of 0.022. This represents a change of approximately
Table 15.1. Values of roughness e (equivalent sand grain size)

for pipe materials for determining the Darcy-Weisbach friction factor
and roughness coefficient C for the Hazen-Williams equation
(adapted from Albertson et al., 1960, and Meadows and Walski, 1998).
Material e (mm) C
Aluminum with a coupler spacing of 9.2 m 0.10 - 0.30 115 - 125
Asbestos cement 0.0015 – 0.0025 135 - 145
Commercial steel or wrought iron 0.03 - 0.09 135 - 145
Concrete 0.20 - 3.00 100 - 140
Corrugated metal pipe 30.0 - 60.0 55 - 60
Galvanized iron 0.10 - 0.25 120 - 130
Galvanized steel for center pivot and lateral move systems 0.10 - 0.12 130 - 140
New wrought iron or steel schedule 40 pipe 0.04 – 0.050 140 - 150
Plastic, polyethylene and PVC 0.0015 - 0.0025 140 - 150
Riveted steel 0.90 - 9.00 100 - 110
700,000
650,000 168 203 254
600,000
550,000 PIPE DIAMETER, mm 152

500,000
REYNOLDS NUMBER
450,000
400,000
350,000
300,000
250,000
200,000
150,000
100,000
50,000
0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140
DISCHARGE, L/s
Figure 15.1. Reynolds number for pipes typical of center pivot and lateral move systems.
Figure 15.2. Resistance coefficient for flows used in irrigation, based on Equation 15.2
using 20°C water.
16% in the resistance coefficient along the lateral. Figure 15.2 also illustrates that the
resistance coefficient varies with the diameter of pipe. If 51 L/s flowed in a 254-mm
pivot lateral as in the above example the relative roughness would be about 0.00043
and the resistance coefficient would drop to 0.018 for a Reynolds number of 260,000.
The irrigation industry has generally accepted empirical formulae to determine fric-
tion losses. Commonly used empirical equations include the Scobey method and the
Hazen-Williams equation. The Hazen-Williams and Scobey equations are given by
LQ d1
hf = K (15.4)
Dd2
where hf = the friction loss in a pipe that conveys the flow throughout the length, m
L = the length of pipe, m
Q = the flow in L/s
D = the inside diameter of the pipe, mm.
The value of K varies for the two equations. For the Hazen-Williams equation K =
1.21 × 1010 C-1.852 where C is the pipe roughness coefficient. The exponents are given
by d1 = 1.852 and d2 = 4.87 for the Hazen-Williams equation. Typical values of C are
listed in Table 15.1 for various types of pipe. For the Scobey method K = 4.10 × 106 Ks
where Ks is Scobey’s coefficient of retardation and the exponent values are d1 = 1.9
and d2 = 4.9.
The empirical equations (Equation 15.4) are similar in form to the Darcy-Weisbach
equation when expressed in terms of flow. The exponents for the flow and pipe diame-
ter for the Hazen-Williams and Scobey equations were determined from experimental
evaluation and are smaller than for the Darcy-Weisbach method. However, the pri-
mary difference is that the roughness coefficient for the Hazen-Williams equation and
the Scobey retardation coefficient are often taken as constants for a given type of pipe.
In reality, the pipe size, type of piping material, coupler design, alignment of pipe, and
distance between couplers affect flow resistance. Normally, as pipe size increases,
resistance decreases (i.e., C should increase for the Hazen-Williams equation). As the
number of couplers decreases, C should increase. Piping materials with smoother in-
side walls will have less resistance and higher C values for the Hazen-Williams equa-
tion. The resistance coefficient for the Darcy-Weisbach equation depends on the pipe
roughness, pipe diameter, flow rate and the temperature of the water. Thus, the Darcy-
Weisbach equation has broader applications than the Hazen-Williams or Scobey equa-
tions.
Other friction loss equations have been proposed that have a similar form. Keller
and Bliesner (1990) combined the Reynolds number, Darcy-Weisbach, and Blausis
equations to develop functions similar to the Hazen-Williams but with exponents for
flow of 1.75 for plastic pipe with diameters smaller than 125 mm and 1.83 for pipes
larger than 125 mm. The diameter exponent was 4.75 and 4.83 for pipes smaller than
125 mm and larger than 125 mm, respectively.
The Hazen-Williams and other empirical equations should be used with caution.
Liou (1998) analyzed several datasets to establish limitations and recommended uses
of the Hazen-Williams equation. He demonstrated significant variations of C with
Reynolds number, relative roughness, and, separately, diameter or absolute roughness
of the pipe. The error resulting from applying the Hazen-Williams equation outside its
calibration range can be as large as 20%. This error is similar to the variation in the
resistance coefficient along pivot laterals. Variations in pipe diameter do not introduce
errors for the Darcy-Weisbach equation if the correct roughness height is used. Tables
and nomographs of energy loss with the Darcy-Weisbach equation are needed to en-
courage greater industry adoption.
The energy loss for pipe with couplers can be determined by increasing the friction
factor by 11% for aluminum irrigation pipe with couplers spaced 9 m apart. Decreas-
ing the spacing to 6 m would increase the energy loss by 17%, while increasing the
coupler spacing to 12 m should decrease the energy loss by 8%. Lyle and Wimberly
(1962) found these losses were the upper limits in their study of various styles of cou-
plers. Either manufacturers’ data or individual tests should be obtained for a given
coupler to determine energy losses for pipe and couplers.
Kincaid and Heermann (1970) used numerical techniques to calculate the pressure
distribution on a center pivot system with the Darcy-Weisbach equation. Their results
showed that the head loss in the sprinkler riser can be approximated by
V2 ⎛ 9.2q s ⎞
hr = exp⎜⎜ ⎟⎟ (15.5)
2g ⎝ Q ⎠
where hr = the head loss in the riser, m
qs = the sprinkler discharge, L/s
Q = the upstream lateral discharge, L/s.
The branching loss due to the change in direction and flow into the riser was generally
less than 0.6 m (6 kPa) and is often neglected in the design of high pressure systems
(>400 kPa).
Pressures along pipelines are computed in an iterative fashion. The outlet pressure
head Hi for section i along the pipeline is given by
Hi = Hi-1 – hfi + (Ei-1 – Ei) (15.6)
where Hi-1 = the pressure head at the inlet to the section
hfi = the head loss due to friction in the section
Ei = the elevation at the distal end of the section
Ei-1 = the elevation at the inlet to the section.
15.2.2 Lateral Pressure Distribution
Many systems are designed for nearly constant sprinkler discharge along the lateral
at a specified average nozzle pressure. The piping is then specified in terms of maxi-
mum lateral length and multiple sizes to maintain the pressure variation along the lat-
eral within specified limits. The laterals are usually designed first, then the main line is
designed, and the inlet pressure is set high enough to maintain the minimum desired
sprinkler pressure. In designing portable laterals, normally only one or two tubing
sizes are used per lateral. However, for permanent laterals several sizes of pipe along
the lateral may be the most economical design. In sizing main or supply and lateral
lines, the designer should not only consider uniformity of application, but also pump-
ing costs and elevation differences between the water source and the fields.
When water is removed at intervals along the lateral, the friction loss will be less
than if the flow was constant for the entire length of the pipeline. To accurately com-
pute friction loss in the lateral, start at the last outlet on the line and work back to the
supply line, computing the friction loss between outlets. This tedious process has been
simplified by a procedure developed by Christiansen (1942). He developed an adjust-
ment factor (F) which is the ratio of the friction loss in a lateral with multiple outlets
having equal spacings and discharges, to that calculated from the general formula that
assumes all of the water is carried to the distal end of the pipeline. The F factor as-
sumes that the first sprinkler is located one full sprinkler spacing from the main line.
Jensen and Fratini (1957) developed a modified factor F′ when the first sprinkler is
located one-half the sprinkler spacing from the supply line. Equations have been de-
veloped to predict the values for F and F′. The equation for F is based on
Christiansen’s formulation while the equation for F′ is based on the approximation
from Keller and Bliesner (1990) to the function developed by Jensen and Fratini:
1 1 m −1 2N ⎛ 1 m − 1 ⎞⎟
F= + + and F ′ = ⎜ + (15.7)
m + 1 2N 6N 2 2N − 1 ⎜ m + 1 6N 2 ⎟
⎝ ⎠
where N is the number of sprinkler outlets on the lateral and m is the exponent in the
friction loss equation. The exponent m = 1.852 for the Hazen-Williams equation, 2.0
for Darcy-Weisbach, 1.9 for the Scobey function, and either 1.75 or 1.83 for the func-
tions developed by Keller and Bliesner (1990). Values of F and F′ are listed in Table
15.2.
Table 15.2. Factors to use in computing head loss in a lateral with equal sprinkler discharge
and spacing for equations commonly used to compute friction loss (based on Equation 15.7).
m coefficient Keller and Bliesner Hazen-Williams Scobey Darcy-Weisbach
Number of 1.75 1.852 1.9 2
Sprinklers F F′ F F′ F F′ F F′
2 0.65 0.53 0.64 0.52 0.63 0.51 0.63 0.50
3 0.55 0.46 0.53 0.44 0.53 0.43 0.52 0.42
4 0.50 0.43 0.49 0.41 0.48 0.41 0.47 0.39
5 0.47 0.41 0.46 0.40 0.45 0.39 0.44 0.38
6 0.45 0.40 0.44 0.39 0.43 0.38 0.42 0.37
7 0.44 0.39 0.43 0.38 0.42 0.37 0.41 0.36
8 0.43 0.39 0.42 0.38 0.41 0.37 0.40 0.36
9 0.42 0.39 0.41 0.37 0.40 0.37 0.39 0.36
10 0.42 0.38 0.40 0.37 0.40 0.36 0.39 0.35
12 0.41 0.38 0.39 0.37 0.39 0.36 0.38 0.35
14 0.40 0.38 0.39 0.36 0.38 0.36 0.37 0.35
16 0.40 0.38 0.38 0.36 0.38 0.36 0.37 0.34
18 0.39 0.37 0.38 0.36 0.37 0.36 0.36 0.34
20 0.39 0.37 0.38 0.36 0.37 0.35 0.36 0.34
25 0.38 0.37 0.37 0.36 0.37 0.35 0.35 0.34
30 0.38 0.37 0.37 0.36 0.36 0.35 0.35 0.34
35 0.38 0.37 0.37 0.36 0.36 0.35 0.35 0.34
40 0.38 0.37 0.36 0.36 0.36 0.35 0.35 0.34
50 0.37 0.37 0.36 0.35 0.35 0.35 0.34 0.34
The relationship between pressure and discharge for sprinklers shows that the dis-
charge varies with the square root of the pressure. Since pressure varies along the lat-
eral due to friction and elevation differences, the sprinkler discharge also varies. The
pressure ratio on a lateral, P/Pd, is the ratio of the pressure at any point on the lateral to
the pressure at the end sprinkler (with constant elevation). The discharge ratio is equal
to the square root of the pressure ratio:
qs P
= (15.8)
qd Pd
where qs is the discharge of any sprinkler whose pressure is P and qd is the discharge
of the last sprinkler on the lateral with pressure of Pd. Thus, with 20% variation in
pressure along a lateral, the discharge varies by about 10%.
To obtain high application efficiencies and stay within economical pipe sizes, the
variation in lateral pressure must be held to a practical minimum. The variation should
be determined by the system designer with the approval of the purchaser. Generally
the variation in pressure should not exceed 20% of the average lateral pressure. Ulti-
mately the allowable pressure loss involves balancing capital cost of the pipe against
pumping costs due to friction. Keller and Bliesner (1990) provide methods to deter-
mine the most economical pipe size.
The variation of pressure along a lateral can be approximated by assuming a uni-
form discharge per unit length along the lateral. The total flow in the lateral (Qx) at
any point x along the lateral is given by
Qx = QL (1 – x/L) = N qs (1 – x/L) (15.9)
where QL is the total inflow to the lateral. If it is assumed that the slope of the lateral is
uniform, the head loss per unit length at a point along the lateral for the Hazen-
Williams equation is given by
dH
= − KQ1x.852 D − 4.87 − S f (15.10)
dx
where Sf is the uniform slope of the lateral from the inlet. A positive slope means the
lateral is running uphill. Combining these equations and integrating from the distal end
of the lateral to a point along the lateral gives the pressure head H(x) along a uni-
formly sloping lateral:
⎡ K L ⎛ x⎞
2.852 ⎤
H ( x) = H d + ⎢ ( Nq s )1.852 ⎜1 − ⎟ + S f ( L − x)⎥ (15.11)
⎢⎣ 2 .852 D .87
4
⎝ L ⎠ ⎥⎦
When this equation is applied to the inlet of the lateral (i.e., x = 0) the pressure loss
in the lateral is 35% (1/2.852) of the loss in an enclosed pipeline where all of the flow
is delivered to the distal end of the pipeline. This corresponds to an F value of 0.35
which would equate to an infinite number of sprinklers on a lateral. The values for the
appropriate F or F′ value can be substituted for 1/2.852 in Equation 15.11.
Chu and Moe (1972) derived an analytical approximation for calculating head loss
in a center pivot lateral. They assumed an infinite number of tiny sprinklers evenly
distributed along the lateral with the discharge proportional to the radius for a uniform
irrigation. The total head loss in the lateral is 54% of the head loss for the lateral oper-
ating as a supply line with the inflow discharge. This supply line head loss can be de-
termined from an energy-loss table or one of the previously discussed equations. The
distribution of the pressure loss was approximated by
⎧ 5 ⎫
⎪ ⎛ 15 ⎞ ⎡ x 2 ⎛ x ⎞ 1 ⎛ x ⎞ ⎤ ⎪
3
H x = H L + (H o − H L ) ⎨1 − ⎜ ⎟ ⎢ − ⎜ ⎟ + ⎜ ⎟ ⎥ ⎬ (15.12)
⎪⎩ ⎝ 8 ⎠ ⎢⎣ L 3 ⎝ L ⎠ 5 ⎝ L ⎠ ⎥⎦ ⎪⎭
where Hx = the pressure head at a radial distance of x from the pivot, m

Ho = the pressure head at the pivot, m
HL = the pressure head at the distal end of the lateral at radius L, m
x = the radial distance from pivot, m.
15.2.3 Average Pressure
The average pressure head along the lateral can be computed by integrating the
pressure distribution for the lateral and dividing by the length. This results in
Fh f Sf L
Ha = Hd + + (15.13)
3.852 2
where hf is the friction loss for an equal sized pipe with no outlets and the adjustment
factor can be equal to F or F′ depending on the spacing of the first sprinkler from the
main line.
The average pressure depends on the sprinkler discharge, the number of sprinklers
on the lateral, the diameter and type of pipe, and the slope along the lateral. The layout
of laterals determines the length and number of sprinklers on the lateral. The discharge
per sprinkler is determined by crop water requirements, soil characteristics, sprinkler
spacings, and management constraints. Once this information is known the average
pressure along the lateral can be determined. The average pressure and the required
discharge are used to determine the sizes of nozzles for the sprinkler. The selected
nozzle and pressure combination must provide adequate coverage to maintain uni-
formity. The average pressure may need to be raised to satisfy both conditions.
15.2.4 Pressure Variation
The pressure head at the inlet (Ho) and distal (Hd) ends of the lateral can be com-
puted using the pressure distribution equation:
Ho = Ha + 0.74 F hf + 0.5 Sf L and Hd = Ha – 0.26 F hf – 0.5Sf L (15.14)
The pressure variation for an example system is shown in Figure 15.3. The average
pressure generally occurs in the second quarter of the pipeline. The average pressure
point moves down the pipeline as the slope increases.
The guideline for maintaining the pressure variation to less than 20% of the average
pressure requires that the maximum and minimum pressure be identified. The maxi-
mum pressure occurs at the inlet to the lateral when the elevation increases along the
lateral (i.e., lateral runs uphill) or when the downward slope is less than the friction
loss along the lateral (i.e., –Sf L < F hf). Conversely, the maximum occurs at the distal
end of the lateral when the lateral slopes downhill at an angle that exceeds the friction
loss along the lateral (i.e., –Sf L > F hf). The maximum pressure can be computed us-
ing either Equation 15.14 or 15.15.
The minimum pressure can occur anywhere along the lateral depending on the fric-
tion loss and slope (Figure 15.3). The minimum pressure and its location can be de-
termined from Equation 15.11. The uniformity criteria require that the difference be-
tween the maximum and minimum head be less than 20% of the average pressure
head. The minimum acceptable average pressure is then five times the pressure varia-
tion along the lateral. Using this criterion and the pressure distribution, the minimum
acceptable average pressure is given by
1. For Sf ≥ 0: Hmax = Ho and Hmin = Hd, so Pmin = 5(F hf + Sf L) 9.8
⎡ 1 ⎤
⎢ ⎛ S f L ⎞ 1.852 ⎥
2. Sf < 0: xmin = L ⎢1 − ⎜ − ⎟ ⎥ so
⎢ ⎜⎝ h f ⎟⎠ ⎥
⎢⎣ ⎥⎦
1
⎡⎛ S f L ⎞⎤ ⎛ S L ⎞ 1.852
H min = H d + Fh f ⎢⎜ − ⎟ ⎥ + S f L⎜ − f ⎟ (15.15)
⎜ ⎟
⎢⎣⎝ h f ⎠⎥⎦ ⎜ hf ⎟
⎝ ⎠
For (– F hf /L) ≤ Sf ≤ 0: Hmax = Ho and xmax = 0
⎧ ⎡ 2.852 ⎤ ⎡ 1 ⎤⎫
⎪⎪ ⎢ ⎛ S f L ⎞ 1.852 ⎥ ⎢ ⎛ S f L ⎞ 1.852 ⎥ ⎪⎪
Pmin = 5⎨ Fh f ⎢1 − ⎜⎜ − ⎟ ⎥ + S f L ⎢1 − ⎜⎜ − ⎟ ⎥ ⎬9.8
⎪ ⎢ ⎝ h f ⎟⎠ ⎥ ⎢ ⎝ h f ⎟⎠ ⎥⎪
⎩⎪ ⎣⎢ ⎦⎥ ⎣⎢ ⎦⎥ ⎭⎪
and for Sf ≤ (– F hf /L): Hmax = Hd and xmax = L
⎧ 2.852 ⎡ 1 ⎤⎫
⎪⎪ ⎛ S f L ⎞ 1.852 ⎢ ⎛ S f L ⎞1.852 ⎥ ⎪⎪
Pmin = 5⎨− Fh f ⎜ − ⎟ − S f L ⎢1 − ⎜ − ⎟ ⎥ ⎬9.8
⎜ hf ⎟ ⎢ ⎜ hf ⎟ ⎥⎪
⎪ ⎝ ⎠ ⎝ ⎠
⎩⎪ ⎣⎢ ⎦⎥ ⎭⎪
where the Pmin (in kPa) is the minimum value that can be used for the average pressure
in selecting nozzle sizes.
The above equations can be used to design constant-sized laterals. Guidelines can
be developed for specific conditions such as shown in Figure 15.4. For this example
the number of sprinklers on the lateral and the discharge per sprinkler are known from
the system layout and preliminary calculations. The chart can be used to quickly esti-
mate the average pressure to use in selecting nozzles for the sprinkler, or the maxi-
mum number of sprinklers for the lateral. A series of charts can be made for varying
spacings, pipe sizes, and slopes. Slide rules and other tools have been developed and
are widely used for quick estimates of friction loss.
900
UPSLOPE FROM LATERAL LENGTH 400 m

INLET, m/m LATERAL INFLOW 16 L/s
800 ALUMINUM PIPE, C 120
0.06 PIPE I.D 99 mm
END PRESSURE 350 kPA
700
0.03
PRESSURE, kPa
600
AVERAGE
0.0 PRESSURE, kPa
524
500
465
-0.03
400 406
347
300 -0.06 288
200
0 50 100 150 200 250 300 350 400
DISTANCE ALONG LATERAL, m

Figure 15.3. Example pressure distribution along a sprinkler lateral versus slope
and the location along the lateral where the average pressure occurs.
700
650
SPRINKLER
600 DISCHARGE, L/s
1.2 1.0 0.8 0.6 0.5 0.4
MINIMUM AVERAGE PRESSURE, kPa
550
500
450
400
350
300
250
200
150
Sprinkler Spacing 9.1 m
100 Pipe Diameter, 101.6 mm
50
C value 120
0
0 10 20 30 40 50 60
NUMBER OF SPRINKLERS
Figure 15.4. Design chart based on the minimum average pressure
for a 101.6 mm lateral with sprinklers spaced at 9.1 m.
When multiple pipe sizes are used along the lateral, or topographic variations are
complex, an iterative section-by-section calculation scheme is required. With the
widespread availability of personal computers and spreadsheet software an iterative
technique is proficient.
Friction losses in valves and fittings such as sprinkler risers and lateral valves
should be included in pressure calculations. Losses in pipe couplers are commonly
included in the factor for friction losses (see Chapter 11 for details on computing
losses in fittings and valves). Valve flow coefficients (flow in L/s for a loss of 1 kPa)
are available from manufacturers. The actual head loss through a valve is usually pro-
portional to the square of the flow rate.
15.2.5 System-Head Curve
The design of a lateral only produces one flow rate and pressure head at the inlet.
For pump specification a head-discharge curve should be developed for the sprinkler
system. The head-discharge relationship can be developed from the pressure distribu-
tion function and the sprinkler discharge equation:
1 / b1
1 ⎛⎜ q s ⎞
⎟ L Sf L
Ho = + 0.74 FK ( Nq s )1.862 + (15.16)
9.8 ⎜⎝ C d Dna1 ⎟
⎠ D 4.87 2
where Cd is the discharge coefficient for the sprinkler used on the lateral, Dn is the
diameter of the nozzle, and parameters a1 and b1 are the coefficients for the discharge
equation:
b1
q s = C d Dna1 P (15.17)
Coefficients a1 and b1 frequently have the value of 2.0 and 0.5, respectively.
A system-head curve can also be developed by treating the lateral as a single circu-
lar nozzle. Using this relationship the inlet head will increase approximately in propor-
tion to the square of the flow rate (Ho = k QL2). The value of k can be computed from
the design values of Ho and QL and can be used for other flows.
If the pump will operate at various flow rates, for example with variable numbers
of laterals functioning, various head-flow combinations may be specified. The pump
pressure can be controlled if necessary, although it is desirable to operate the pump
close to full capacity. Multiple pumps operating in parallel would be a good choice for
systems in which the flow rate varies greatly but pressure head is nearly constant. A
head-booster pump can be used where a portion of the system is at a high elevation
relative to the pump.
It is desirable to apply water as uniformly as possible, whether the entire system or
only a segment of the system is operating. This requires that pressure loss due to fric-
tion in the main line be minimized or that flow control devices be used. Elevation dif-
ferences in a field must also be included in the calculations.
Normally, flow velocities should not exceed 3 m/s. For permanent systems in
which polyvinyl chloride (PVC) plastic pipe and asbestos cement (AC) pipe are used
for supply lines, velocities should not exceed 2.25 m/s, and most manufacturers cau-
tion against using velocities in excess of 1.6 m/s.
When more than one lateral on a system is operated at one time (especially adjacent
to each other), the supply line becomes multi-outlet. This means more water is flowing
through the portion nearer the pump and friction loss must be computed in each segment.
15.2.6 Pressure Regulation/Flow Control

The operating pressure of a sprinkler may vary from the design value. Variations in
pressure may result from changes in elevation as the sprinkler lateral is moved across
the field. Variations can also occur when the numbers of laterals or sprinklers per lat-
eral vary during the operation. When a lateral is shutoff for repositioning, the pressure
increases for the laterals that remain in operation. If the variation is too severe, pres-
sure regulators or flow control nozzles may be required to provide the desired uni-
formity.
Pressure regulators are devices that maintain a nearly constant downstream pressure
for varying inlet pressures as illustrated in Figure 15.5. The advantage of pressure
regulators is that the discharge from the sprinkler is nearly constant over a wide range
of inlet pressures. An indicator that regulators are needed is based on the concept that
the pressure at a sprinkler should not vary more than 20%. Guidelines such as shown
in Figure 15.6 can be developed from these criteria.
Pressure regulators are not without some limitations. First, there is a loss of pres-
sure across regulators. The pressure loss occurs within the regulator even when the
inlet pressure is below the design pressure. Commonly designers use a loss of 20 to
40 kPa across the regulator. For proper design the pressure in the lateral should be
adequate to satisfy the required pressure for the sprinkler, pressure losses in fittings
and riser used to connect the sprinkler to the lateral and losses through the regulator.
Pressure regulators increase the cost of the irrigation systems. The analysis by von
Bernuth (1983) compared the economic benefit of the increase in uniformity for using
pressure regulators to the amortized investment. This type of comparison is very site
specific and results cannot be generalized; however, the analysis provides a procedure
for localized evaluation.
300
250 FLOW, L/s

0.05
OUTLET PRESSURE, kPa
0.6
200 1.0
150
FLOW, L/s
100
0.05
0.6
50 1.0
0
0 100 200 300 400 500 600 700 800
INLET PRESSURE, kPa
Figure 15.5. Regulator control of outlet pressure with regulators set to maintain
70 and 210 kPa.
8.0
7.0
Pressure Regulators Recommended
6.0
Elevation Change, m
5.0
4.0
3.0
2.0
Pressure Regulators Not Recommended
1.0
0.0
50 100 150 200 250 300 350
Design Pressure, kPa
Figure 15.6. Elevation change equivalent to 20% variation in the design pressure.
Proper operation of pressure regulators is difficult to monitor. As von Bernuth and

Baird (1990) showed, some regulators display a hystersis when the inlet pressure
gradually changes. Thus, the regulator does not restrict flow to the extent desired for
high pressure situations and excessively restricts flow when the pressure drops. Pres-
sure regulators operate by restricting the opening for water flow when the pressure
increases. This requires relatively small openings that may plug if foreign material is
contained in the irrigation water. This makes utilization of the devices inconsistent for
some sources of water. If the regulators are plugged, have hystersis problems or sim-
ply stick in one position it is difficult to observe the inappropriate operation. The
sprinkler may appear to be working yet the pressure could be inappropriate.
Regulators are occasionally used when they are not needed. Regulators are habitu-
ally used on many center pivots that use low pressure sprinkler devices. Some design-
ers use regulators as insurance so that the system will operate at the appropriate pres-
sure even though the terrain would not merit pressure regulation. Obviously this drives
up the investment and operating costs and should be avoided.
If pressure regulators are used the minimum pressure in the system should be de-
termined. This usually occurs at a point near the end of the system where the elevation
is the greatest. The minimum pressure may vary during the growing seasons if the
supply capacity of wells or other sources of water declines. The system should be de-
signed for this pressure and the appropriate regulators should be used for this condi-
tion. The uniformity of the system can be reduced substantially if the inlet pressure
drops below the design pressure for the system.
Special nozzle designs are available that compensate for pressure variation in the
field. These nozzles, called flow control nozzles, are constructed so that the diameter
0.6
NORMAL 3.97 mm NOZZLE

0.5
DISCHARGE, L/s
0.4
FLOW CONTROL NOZZLE
0.3
0.2
0.1
0.0
0 200 400 600 800 1000
NOZZLE PRESSURE, kPa
Figure 15.7. Illustration of flow control nozzle compared to a regular nozzle.
of the nozzle varies with nozzle pressure. When the pressure rises the nozzle diameter
decreases and a relatively uniform flow rate is maintained as illustrated in Figure 15.7.
Flow control nozzles can be effective in providing uniform discharge and are cheaper
to purchase and operate than pressure regulators. Flow control nozzles, however, will
not compensate for as large of variation in pressure as regulators. As with pressure
regulators, it is difficult to determine if the nozzles are operating properly and flow
measurements are generally required to monitor operation.
15.2.7 Energy Losses in Regulators and Emitters
The industry recommends that the inlet pressure exceed the nominal regulated pres-
sure by 30 to 35 kPa. This in general will allow for the pressure loss through the regu-
lator at most design flow rates. As illustrated in Fig. 15.6, the regulated outlet pressure
is dependent on the flow rate. The head loss through the regulator increases as the
flow increases. Emitters also have head losses that are variable depending on pressure
and thus flow rate. It is recommended to search the literature and obtain manufactur-
ers’ recommendations on head loss for specific regulators and emitters.
15.2.8 Energy Losses in Filtration Systems
Filtration systems included screens and media type filters. The inlets to pumping
plants often use a rotating screen filter that has minimal head loss with the relatively
large screen area sized to the design flow. Filters are also used prior to the inflow into
the lateral systems particularly for microirrigation systems. In some cases filters are
used to clean the water to prevent the plugging of application devices for sprinkler
systems. The manufacturers can supply the required head for proper operation. Many
of the filter systems are designed to provide self cleaning when the pressure drop
across the filter exceeds approximately 50 kPa. The head loss depends on the flow rate
and manufacturers’ curves illustrate losses of 1 to 100 kPa for a range of discharges.
15.3 IRRIGATION SYSTEM VALVES

The following sections provide considerations for special valves when installing
pressurized pipelines. Specific technical data and installation recommendations should
be obtained from the equipment manufacturers. Often these data can be obtained from
internet sites.
15.3.1 Surge Control and Check Valves
Irrigation systems require valves for many functions. Some common valves and
their functions and applications are described below. Undesirable high and low pres-
sures can occur in a pipeline following power failure. Carefully timed and controlled
closure settings on the valves at each pump discharge will help reduce the high pres-
sure spikes that occur when the valves fully close. Low pressures are problematic if
the water begins to vaporize. Bubbles form in the liquid if the pressure in the pipeline
drops below the vapor pressure of the water. If the bubbles coalesce they may form a
pocket of vapor in the pipeline. This is called column separation and is a condition that
must be guarded against during flow. Column separation that occurs at high flow rates
and high head conditions cannot be corrected by adjusting valve closure times. Vac-
uum relief valves along portions of the pipeline will be required to minimize column
separation.
15.3.1.1 Surge chambers. A surge chamber (Figure 15.8) is the most reliable water
hammer control device. Factors used to determine the required surge chamber size—
which can be large—are pipeline length, maximum flow capacity, acoustic wave
speed in the line, pipe profile, and flow distribution. The required size can only be
determined with computer analysis, but an approximate size may be estimated by
QL
VT = (15.18)
100c
where VT = the total volume, m3
Q = the discharge, L/s
L = the length of the pipeline, m
c = the velocity of the pressure wave in the pipeline, m/s.
The velocity of the pressure wave as a function of pipe material and the diameter ratio
of the pipe are shown in Figure 15.9 for materials commonly used for irrigation. Equa-
tions to compute the velocity of the pressure wave for other pipe materials or tempera-
tures are provided in fluid mechanics textbooks. The temperature of the water has little
effect on the wave velocity. The diameter ratio used to determine the velocity of the
wave is the ratio of the inside diameter to the wall thickness of the pipe.
Surge chambers do not work properly with in-line control valves or surge anticipa-
tor valves. To prevent forcing flow back through the pumps, it is appropriate to use a
check valve in conjunction with surge chambers on the discharge side of pumps. Dif-
ferent outlet and inlet line sizes may significantly reduce the required surge chamber
size.
Surge chambers act as energy sources following power failure and as shock absorb-
ers during upsurge events such as a pump start or reverse flow following a pump shut-
down. Characteristically, surge chambers remove sharp pressure spikes and create
smooth, controlled pressure oscillations until friction damps out transient pressure
waves.
Figure 15.8. Picture of surge tank assembly with sketches of an air vent/vacuum
relief valve and a pressure release valve.
Figure 15.9. Velocity of pressure wave based on the diameter ratio of the pipe
(for 20°C water and pipes with expansion joints).
15.3.1.2 Combination air valves. Air admission/air release valves (see Figure
15.8) allow air into the line when the pressure drops below atmospheric pressure and
discharge air (if any) when the pressure becomes positive. Negative pressures in a line
during a surge may be better understood by visualizing the negative pressures as a
drooping string hanging below the pipe profile. The string is pulled up to the pipe at
control points, such as reservoirs or air valves. The string still droops between these
control points. The only way to prevent column separation is to make the valves large
enough to pull the pressures up to near atmospheric pressure at the valves and place
the valves close enough together to keep the drooping pressure between valves at a
safe level. Pressures should normally be above 50 kPa, which yields a safety factor of
about two in avoiding column separation.
Generally, the admission of large quantities of air into pipelines is not a good prac-
tice. Air cavities are created at low pressure; this limits the cushioning characteristics
of the cavities when water columns rejoin. High pressures can result when water col-
umns collide. In addition, air must be expelled to ensure a smooth, efficient pipeline
operation.
15.3.1.3 Surge anticipator valves. Surge anticipator valves open when the pres-
sure drops below a specified set point. They remain open for a pre-set period of time,
and then close in a manner that prevents the high pressure caused by rapid valve clo-
sure. Properly installed, surge anticipator valves prevent high pressures at pump sta-
tions but have little ability to alleviate column separation, especially in long pipelines.
Furthermore, valve movements must be set with care; otherwise, they may cause more
severe water hammer problems than they correct. Anticipator valves can also be set to
“open” when line pressures exceed a pre-set value, thus operating as a surge relief
valve.
15.3.1.4 Surge relief valves. Pressure relief valves consist of a movable seal plate
that is held in position by a spring and jam nut as illustrated in Figure 15.8. When
pressure in the pipeline exerts more force on the seal plate than the spring can resist
the seal plate rises and allows high pressure water to exit from the pipeline. Surge or
pressure relief valves can reduce maximum surge pressures if they are adjusted prop-
erly. They open at a pre-set pressure level and relieve excessive pressure at a critical
point in a pipeline, often at the discharge of pumps. Like surge anticipator valves,
surge relief valves cannot prevent column separation.
15.3.1.5 Check valves. Check valves prevent or limit reverse flow and should be
designed to prevent slamming. Excessive noise and pressure spikes upon final closure
are two negative results of slamming valves. Check valves are available in numerous
types, styles, and shapes. Below we summarize the desirable and undesirable features
of the five basic types of check valves: the swinging check valve (both single disc and
double disc), the slanting-disc check, the silent check (poppet, disc, ball), the control
disc check, and the foot check.
The various types of check valves may look different, but all check valves perform
the same basic function: they facilitate flow in one direction and control for a no-flow
(check) return. Check valves are commonly installed on the discharge side of the
pump. The most important role of a check valve is to act as the automatic shutoff
valve when the pump stops. This prevents the draining of the system that the pump is
filling. However, each check valve has different shutoff characteristics.
Pressure surge (water hammer) can be greatly reduced by selecting the correct
check valve. For example, silent-check valves shut off very quickly, in 1/10 to 1/20 of a
second; this makes them excellent surge protection devices when pumping short dis-
tances (less than 800 meters). Conversely, the control-disc check valve has extremely
slow and variable speed shutoff; it is an excellent surge protection device when pump-
ing long distances (more than 800 meters).
15.3.2 Check Valve Comparisons
15.3.2.1 Swinging check (single disc). Swinging check valves (Figure 15.10) de-
pend on gravity and reverse flow to close the valve and shut off reverse flow. Many
models include a spring to help close the valve. The pivot point of the swinging check
is outside of the periphery of the disc and the larger the head, the greater the possibil-
ity that the fluid will flow back through the valve before the disc can shut. To affect
complete shutoff, the disc of a swinging check valve must travel through a 90° arc to
the valve seat. Without resistance to slow the disc’s downward thrust, the shutoff re-
sults in slamming and potentially damaging water hammer. Caution should be used
when applying this type of check valve when velocities exceed 1 m/s reverse flow and
pressures are higher than 200 kPa.
Figure 15.10. Sketch and photograph of single-disc swinging check valve.
15.3.2.2 Double-disc swinging check (split-swing disc). The double-disc swing-

ing check valve (Figure 15.11) is similar to the conventional swinging check valve
except that the single disc is split for the double-disc valve. Splitting the disc reduces
the mass of the disc and the travel distance from the open to the shutoff position. Fur-
ther improvement results from torsion springs that force the double discs to shut with
minimal flow reversal. In comparison to the conventional swinging check, this valve’s
shutoff characteristic minimizes the slam potential. With double-disc checks, the hinge
pin is stationary and each disc swings freely when opening or closing. Multiple
springs are used on larger sizes to compensate for heavier discs and improve shutoff
speed.
Figure 15.11. Diagram of the components of a double-disc swinging check valve.
15.3.2.3 Slanting-disc check (with off-center pivot). Slanting-disc check valves

have an enlarged flow area through the disc and seat sections. The disc pivots off-
center, with approximately 30% of the disc area above the pivot point offering resis-
tance against the 70% disc area below the pivot point. Hence, this type of check valve
has built-in non-slam shutoff characteristics. The shutoff seating angle is 55°. The disc
swings open through this 55° seat angle, travels a short distance, and stops at 15° off
of the horizontal. Hence, the disc travels only a short distance, allowing minimal re-
verse flow and non-slam shutoff.
15.3.2.4 Silent check. Silent check valves have a center-guided poppet that is
spring loaded and is normally closed. The feature that results in the silent shutoff is the
relation of the poppet to the seat when in the open position. This distance is approxi-
mately 1/4 of the valve size; in other words, a 100-mm valve has a 25-mm distance to
shut off. It is the short poppet travel distance coupled with the spring force that results
in a silent shutoff. The approximate time of shutoff is between 1/20 and 1/10 of a second.
Silent check valves are furnished with springs, eliminating any possibility of reverse
flow and minimizing the water hammer or pressure surge.
15.3.2.5 Control-disc check. The control-disc check valve is a unique adaptation
of the slanting-disc check valve. All previously described check valves open and close
when flow from the pump starts or stops. The opening and closing of the disc, poppet,
or flapper is uncontrolled. The control-disc check valve is recommended where pump
flow control is essential to prevent pressure surges when flow starts or stops in the
pipeline. The valve disc can be electrically controlled to permit remote operation of
automatic pump stations. The valve functions as a shutoff valve, slowly opening after
flow starts and closing slowly when stopping the flow to close the control-disc check
valve. The valve shuts off automatically if electrical power is interrupted. This type of
valve is available with hydraulic controls and disc position indicators. Similar control
functions can be achieved with hydraulic cylinder-actuated or motor-actuated butterfly
valves.
15.3.2.6 Foot valve. A foot valve is a form of check valve installed at the bottom of
the pump suction line and below water. Foot valves are inexpensive ways to prime
centrifugal pumps. The foot valve is installed in the vertical position with the direction
of flow upward. In this position, the foot valve is normally closed. It is necessary to
fill the suction line with water prior to starting the centrifugal pump. Once the suction
line is filled, the foot valve takes over and opens while the centrifugal pump is running
and closes when the pump stops. This maintains a flooded suction and primed pump.
Since foot valves are continually submerged and are not readily accessible for inspec-
tion or repair, it is important to select high-quality, long-wearing valves.
15.3.3 Theory and Use of Air Valves
15.3.3.1 Air release valves. Air entrained in conveyed water will often separate
from the liquid and collect at high points within the pipeline. If provisions are not
made to remove air from high points, pockets of air will collect and grow in size. Air
pocket growth will gradually reduce the effective liquid flow area, creating a throttling
effect similar to a partially closed valve. Often, the velocity of liquid flow will be suf-
ficient to carry a portion of the enlarging air pocket downstream and the air will lodge
at another high point. The ability of flow velocity to trim the size of air pockets may
minimize the effect of reduced flow area in the pipe, but the throttling effect caused by
the presence of air will always reduce the flow rate below design. Air pocket problems
are difficult to detect, but if permitted to continue may cause a constant power drain.
Air release valves installed at high points of the system shut off when filled with
liquid and are then subjected to pressure. During system operation, small particles of
air will separate from the liquid and enter the valve. Each particle of air will displace
an equal amount of the liquid within the valve and lower the liquid level relative to the
float. When the liquid level lowers to a point where the float is no longer buoyant, the
float will drop. This action opens the valve orifice and allows the air that has accumu-
lated in the upper portion of the valve to be released into the atmosphere. As air is
released, the liquid level within the valve once again rises, lifting the float and closing
the valve orifice. This cycle repeats itself as often as air accumulates in the valve.
15.3.3.2 Air and vacuum valves. An air and vacuum valve (AVV) is float-
operated with a large discharge orifice equal in size to the valve’s inlet. This valve
allows great volumes of air to exhaust from or enter into a system as it fills or drains.
The following conditions prevail with AVVs used in pipelines:
Prior to filling, a pipeline may be thought of as empty, but this is not true. In re-
ality it is filled with air. The air must exhaust from the pipeline in a smooth, uni-
form manner to prevent pressure surges and other destructive phenomena.
To prevent potentially destructive vacuums from forming, air must re-enter the
pipeline in response to a negative pressure. Even in those instances where vac-
uum protection is not a primary concern, air re-entry is still essential to effi-
ciently drain the pipeline and prevent water column separation, which can be as
damaging as pressure surges.
To perform the functions outlined, an AVV must be installed at each high point or
change in grade, at the ends of pipelines, and at isolation valves.
As a pipeline fills air exhausts into the atmosphere through an AVV. As air is ex-
hausted, water enters the valve and lifts the float to close the valve orifice. The rate at
which air is exhausted is a function of the pressure differential that develops across the
valve discharge orifice. This pressure differential develops as water filling the pipeline
compresses the air sufficiently to provide an escape velocity equal to that of the in-
coming fluid. Since the size of the valve controls the pressure differential at which the
air is exhausted, valve size selection is very important.
Should the internal pressure of the pipeline approach a negative value at any time
during system operation due to column separation, draining of the pipeline, power
outage, or pipeline break, the float will immediately drop away from the orifice and
allow air to re-enter the pipeline. Air re-entry will prevent a vacuum from occurring
and protect the pipeline against collapse or water column separation. The size of the
AVV will dictate the degree to which a vacuum is prevented; therefore, selecting the
correct valve size is very important. The AVV, having opened to admit air into the
pipeline in response to a negative pressure, is now once again ready to exhaust air.
This cycle will repeat as often as necessary.
During system operation and while under pressure, small amounts of air will enter
the AVV from the pipeline and displace the fluid. Eventually, the entire AVV may fill
with air but it will not open because the system pressure will continue to hold the float
closed against the valve orifice. To reiterate, an AVV is designed to exhaust air during
pipeline fill and to admit air during pipeline drain. It will not open and vent air as it
accumulates during system operation—air release valves are used for this purpose.
15.3.3.3 Combination air valves. Combination air valves consist of a float that
moves vertically inside the valve to control the discharge and entry of air into the pipe-
line (Figure 15.12). As the name implies, combination air valves (CAVs) have operat-
ing features of both air and vacuum valves and air-release valves. CAVs are also
called double-orifice air valves. Combination air valves are installed at high points of
a system where it has been determined that dual-purpose air-and-vacuum and air-
release valves are needed to vent and protect a pipeline. Generally, it is sound engi-
neering practice to use CAVs instead of single-purpose air and vacuum valves at high
points.
Combination air valves prevent accumulations of air at high points within a system
by exhausting large volumes of air as the system is filled and by releasing accumu-
lated pockets of air while the system is operational and under pressure. Combination
Figure 15.12. Picture of a combination air control valve for when the float is in the
down position and when it is raised to operate in a continuous position.
air valves also prevent potentially destructive vacuums from forming by admitting air
into the system during power outages, water column separations, or a sudden drainage
of the pipeline.
15.4 WATER DISTRIBUTION TO THE SOIL
The objective of microirrigation is to apply water for individual plants at rates slow
enough to prevent water translocation on the surface. Tree crops are often irrigated
with the emitter applying the water under the canopy, whereas with row crops the wa-
ter is applied near the plant or in the row to provide the required water for production
but limit the evaporation from a wet soil surface. Buried microirrigation minimizes the
soil surface evaporation while providing sufficient water for plant growth. In contrast,
the objective of sprinkler irrigation is to uniformly apply water over the area for crop
use. Sprinkler irrigation systems should be designed to apply water at a rate less than
the intake rate to prevent surface runoff. The water application uniformity of a sprin-
kler system depends less on the dynamics of the individual sprinklers and more on the
spacing and operating pressure. The spacing of sprinklers and, on moving systems, the
travel speed and path affect irrigation uniformity. The effects of wind and evaporation
on uniformity are discussed in more detail in Chapter 16.
15.5 REDISTRIBUTION IN THE SOIL
The uniformity of sprinkler irrigation systems normally is evaluated on the basis of
distribution at the soil surface. The crop, however, is dependent on available water in
the root zone. Water will move within the soil profile due to potential gradients, and
plant roots have a horizontal distribution. Hart (1972) analyzed the subsurface redis-
tribution, using finite-difference solutions of the two-dimensional diffusion equation.
He concluded that there is substantial redistribution within 1 m and final soil water
distribution is not significantly affected by the nonuniformity on the soil surface
within that distance (as illustrated in Figure 6.6). An estimated 10% reduction in total
water requirements resulted due to the redistribution of soil water for one test field.
Where the nonuniformity exists over larger than 1- to 3-m distances, the subsurface
redistribution will have little effect on the final soil-water uniformity.
Seginer (1978) estimated a net loss of $180/ha for a 0.1 decrease in the uniformity
coefficient in cotton production due to nonuniform water application. The effective
nonuniformity for various crops is a function of the extent of horizontal root develop-
ment of individual plants. Seginer (1979) analyzed the increase in effective uniformity
of water distribution in orchards with various tree and sprinkler spacing and calculated
an increase in effective uniformity of 0.1. The “actual” uniformity is appropriate for
“point” plants that have a limited horizontal root distribution. He also found that when
rows are parallel to the shorter sprinkler spacing of solid set systems, a better effective
uniformity can be obtained.
Microirrigation systems generally apply the water directly to the root zone. How-
ever, in some cases the spacing of the emitters does rely on horizontal movement to
become available for the root system. This is particularly true for row crops where
emitters are not located in each row but in between or every other row spacing.
15.6 SUMMARY
Design and operation of sprinkler and microirrigation systems depends on proper
selection and sizing of components, which rely on the water pressure distribution
throughout the system. This chapter discusses the hydraulics of mainline and lateral
pipelines, performance of pressure regulation devices, and characteristics of valves
and pipeline protection devices commonly used in pressurized irrigation systems. Fig-
ures and charts are provided for use in design and operation. The ability of soils to
redistribute soil water and thereby affect the uniformity of water availability to plants
is briefly reviewed.
LIST OF SYMBOLS
a1,b1 coefficients for the sprinkler discharge equation
C pipe roughness coefficient for the Hazen-Williams equation
c velocity of a pressure wave in the pipeline, m/s
Cd discharge coefficient for sprinkler discharge
D inside diameter of the pipe, m
Dn diameter of the nozzle, mm
d1,d2 coefficients for the Hazen-Williams and Scobey equations
e absolute roughness of the inside of the pipe, mm
Ei elevation at the inlet section i of a pipeline, m
Ei -1 elevation at the outlet of section i of a pipeline, m
f Darcy-Weisbach resistance coefficient
F adjustment factor for lateral pressure loss when first sprinkler is one sprinkler
spacing from the inlet
F′ adjustment factor for lateral pressure loss when first sprinkler is one-half the
sprinkler spacing from the inlet
g gravitational acceleration, 9.81 m/s2
Ha the average pressure head along a sprinkler lateral, m
Hd the pressure head at the distal end of a sprinkler lateral, m
hf head loss, m
hfi head loss due to friction in the section
Hi-1 pressure head at the inlet to the section
HL pressure head at radial distance L from the pivot point, m
Hmax the maximum pressure head along the sprinkler lateral, m
Hmin the minimum pressure head along the sprinkler lateral, m
Ho pressure head at the inlet to a lateral or pivot, m
hr head loss in a sprinkler riser, m
Hx pressure head at a radius x from the pivot, m
Ks coefficient of retardation for the Scobey equation
L pipe length, m
L radius of the irrigated area, m
m exponent in the friction loss equation
N number of sprinkler outlets on the lateral
P pressure in the pipeline
Pd pressure at the distal end of a sprinkler lateral, kPa
Pmin minimum value for the average pressure on a sprinkler lateral, kPa
Q flow in the pipe, m3/s or L/s
qs sprinkler discharge; L/s
qd discharge of the last sprinkler on a lateral, L/s
Qx flow in the lateral at a point along the lateral, L/s or m3/s
Re Reynolds number
Sf uniform slope of the lateral from the inlet, m/m
V velocity of flow, m/s
v kinematic viscosity, m2/s
VT total volume required for a surge tank, m3
x distance from the inlet to a point along a sprinkler lateral, m
REFERENCES
Albertson, M. L., J. R. Barton, and D. B. Simons. 1960. Fluid Mechanics for
Engineers. Englewood Cliffs, N.J.: Prentice-Hall, Inc.
Chu, S. T., and D. L. Moe. 1972. Hydraulics of a center pivot system. Trans. ASAE
15(5): 894-896.
Hart, W. E. 1972. Subsurface distribution of nonuniformity applied surface waters.
Trans. ASAE 15(4): 656-661, 666.
Jensen, M. C., and A. M. Fratini. 1957. Adjusted “F” factor for sprinkler lateral
design. Agric. Eng. 38(4): 247.
Van Nostrand Reinhold.
Kincaid, D. C., and D. F. Heermann. 1970. Pressure distributions on a center-pivot
sprinkler irrigation system. Trans. ASAE 13(5): 556-588.
Liou, C. P. 1998. Limitations and proper use of the Hazen-Williams Equation. J.
Hydraulic Eng. ASCE 124(9): 951-954.
Lytle, W. F., and J. E. Winberly. 1962. Head loss in irrigation pipe couplers. Bull. No.
553. Baton Rouge, La.: Louisiana State Univ. and Agric. Mech. College, Agric.
Exp. Station.
Meadows, M. E., and T. M. Walski. 1998. Computer Applications in Hydraulic
Engineering. Waterbury, Conn.: Haestad Press.
Seginer, I. 1978. A note on the economic significance of uniform water application.
Irrig. Sci. 1: 19-25.
Seginer, I. 1979. Irrigation uniformity related to horizontal extent of root zone: A
computational study. Irrig. Sci. 1: 89-96.
von Bernuth, R. D. 1983. Nozzling considerations for center pivots with end guns.
Trans. ASAE 26(2): 419-422.
von Bernuth, R. D., and D. Baird. 1990. Characterizing pressure regulator
performance. Trans. ASAE 33(1): 145-150.
CHAPTER 16
DESIGN AND OPERATION

OF SPRINKLER
SYSTEMS
Derrel L. Martin (University of Nebraska,
Lincoln, Nebraska)
Dennis C. Kincaid (USDA-ARS,
Kimberly, Idaho)
William M. Lyle (Texas A&M University,
Lubbock, Texas)
Abstract. The design and operation of sprinkler irrigation systems are presented
and discussed in this chapter. Information is provided concerning the components of
irrigation systems and the characteristics of the system that affect water application
efficiency and uniformity. The interactions of system characteristics with soil and crop
properties are evaluated relative to design and operation of irrigation systems to
minimize runoff, deep percolation, and excessive evaporation losses.
Keywords. Application efficiency, Center pivot, Lateral move, Runoff, Side roll,
Solid set, Sprinkler irrigation, Towline, Uniformity.
16.1 INTRODUCTION
Sprinkler systems have revolutionized the development of irrigated agriculture. Ef-
ficient water application with sprinkler irrigation involves the design and operation of
pumps, pipes, and sprinkler devices to match soil, crop, and resource conditions. Thus,
sprinkler systems can be designed and operated for efficient irrigation for a wide array
of conditions. Sprinkler irrigation also facilitated the expansion of irrigated agriculture
onto lands classified as unsuitable for surface irrigation. Initially the labor required to
transport the system across the field impeded the adoption of sprinkler irrigation.
Through automation the labor required for sprinkler irrigation has been reduced sig-
nificantly. Reduced labor requirements enabled producers to irrigate more frequently
with smaller water applications which diminished unintentional leaching and increased
the potential to store precipitation in the crop root zone while satisfying crop require-
ments.
Morgan (1993) indicated that sprinkling devices were used as early as 1873. By
1898 seventeen patents had been issued for sprinkler devices. Since that early begin-
558 Chapter 16 Design and Operation of Sprinkler Systems
ning many developments have occurred. The patent for impact sprinklers as we know
them today was issued in 1934. Aluminum pipe with rubber gaskets were first pro-
duced in the late 1940s while an early version of the side roll machine was first pro-
duced in the 1950s. Self-propelled center pivot and lateral move systems were in-
vented in the late 1940s. Producers quickly recognized that controlling an irrigation
system was essential for proper performance. One of the first controllers for sprinkler
irrigation was installed in 1924.
These early developments laid the foundation for the growth of sprinkler irrigation.
In the late 1940s and early 1950s development began in earnest and continued with
large increases in the 1960s and 1970s when automated systems were commercialized.
The amount of land irrigated with sprinkler systems continues to increase. Most of the
development today is devoted to automated and semi-automated sprinkler systems.
According to the Farm and Ranch Irrigation Survey (USDA-NASS, 2003), sprinkler
irrigation was used on approximately 51% of the land irrigated in the United States
(Table 16.1). Currently, center pivots are used on approximately 79% of the land that
is irrigated by sprinkler. Slightly more low-pressure center pivots are used than me-
dium-pressure pivots. Relatively few high-pressure pivots were used in 2002.
Table 16.1. Distribution of sprinkler irrigated land in the United States in 2002
(source: USDA-NASS, 2003).
Land Area (hectares) by Pressure Range Total Area
% of Sprinkler-
System < 207 kPa 210 to 414 kPa > 414 kPa hectares Irrigated Land
Center pivot 3,925,883 3,909,860 784,943 8,620,685 79.0
Lateral move 60,553 78,784 139,337 1.3
Solid set 112,551 364,353 476,904 4.4
Side roll -- -- 739,231 6.8
Traveler or big gun -- -- 256,351 2.4
Hand move -- -- 673,498 6.2
Total sprinkler-irrigated land 10,906,006
Total land irrigated with all types of systems 21,288,338
16.2 COMPONENTS OF SPRINKLER SYSTEMS

Sprinkler irrigation systems generally consist of a pump used to lift and pressurize
water, a main line piping system to convey water from the pump to the lateral pipe-
lines used to transport water across the irrigated field, and sprinkler devices to apply
the water within the field (Figure 16.1). In some cases the main line is subdivided into
submains to convey water to portions of the field. Multiple laterals may be used if the
field is large or if the field needs to be irrigated frequently. Sprinkler devices are in-
stalled at intervals along the lateral. Often a pipe called a riser is used to adjust the
height of the sprinkler device to the desired height to avoid crop interference with the
jet of water discharged from the sprinkler. For some applications with center pivots or
lateral move systems, the sprinkler device is suspended below the lateral using drop
tubes to minimize evaporation and drift of sprinkler droplets.
Figure 16.1. Components and general layout of sprinkler irrigation systems.
The proper spacing of sprinklers along the lateral is crucial to providing enough
overlap of water from adjacent sprinklers to achieve uniform irrigation. Sprinkler lat-
erals can be permanently installed as with solid set systems, or the laterals can be
moved across the field. Laterals can be periodically moved from one location (often
called a “set”) to the next location as with hand move systems or the laterals can move
continuously across the field as with center pivot and lateral move systems. When the
sprinkler lateral is periodically moved across the field the distance between subsequent
sets of the lateral must be narrow enough to provide for adequate overlap.
Many kinds of sprinkler devices are available; some are shown in Figure 16.2. All
sprinkler devices use a nozzle to control the discharge from the sprinkler. The diame-
ter and shape of the nozzle orifice and the water pressure at the nozzle control the rate
of flow. The nozzle converts pressure within the piping system to velocity upon dis-
charge from the sprinkler. The velocity propels droplets through the air to provide a
wetted pattern about the sprinkler. The design of the nozzle and the sprinkler device
determines the diameter of coverage and the distribution of water within the wetted
region.
As with any irrigation system, the sprinkler system must be laid out to conform to
the boundaries of the field. The dimensions of the field physically or economically
often dictate the type of sprinkler system to be used. Ultimately the components of the
sprinkler system must be matched to irrigate efficiently and economically. Inappropri-
ate specification of any component will negatively affect the entire system.
Figure 16.2. Examples of sprinkler devices used on irrigation systems.
16.3 DESIGN FUNDAMENTALS

Sprinkler irrigation systems should be designed to satisfy crop water requirements
while applying water at a rate that minimizes runoff and excess leaching. To accom-
plish these objectives, the pressure distribution in the main line and sprinkler lateral
must be appropriate as discussed in Chapter 15. Efforts are underway to design sys-
tems that apply variable depths of water within a field for site-specific management.
However, emphasis is currently on systems to apply water uniformly throughout the
field. General concepts that apply to design and operation of sprinkler systems are
developed in this section. Details for specific systems are presented later.
16.3.1 Application Efficiency
The application efficiency is the fraction of the applied water that remains in the
crop root zone following irrigation. Water that infiltrates and remains in the root zone
is the net irrigation. The application efficiency is then the ratio of the net irrigation
depth (dn) to the gross depth of water applied (dg).
The possible losses of water in sprinkler irrigation are illustrated in Figure 16.3.
Some water may evaporate before it reaches the crop canopy or the soil surface. A
portion of the water is intercepted by the canopy while the remainder falls through the
canopy to the soil surface. Water that is intercepted by the canopy can be retained on
the canopy, may run down the stem of the plant to the soil surface, or may drip from
the plant. Steiner et al. (1983) showed that for a mature corn crop up to 50% of the
Figure 16.3. Diagram of water losses for sprinkler irrigation.
water applied with a center pivot flowed along the leaves and stems to the ground.
Water that is intercepted and/or retained on the canopy can evaporate during and after
irrigation. Water that reaches the soil surface can evaporate, infiltrate, run off, or be
stored in depressions along the soil surface.
When infiltration exceeds the available storage in the crop root zone, the excess ap-
plication will slowly drain through the soil profile inducing deep percolation. Water
stored on the soil surface has been called surface storage, depression storage, or reten-
tion storage. This water is available to infiltrate after the rate of application of water
decreases below the infiltration rate of the soil. Water that runs off the point of appli-
cation can infiltrate at a downstream location and may be used. If the runoff leaves the
field, the water is lost to that field. If the runoff accumulates in low-lying areas deep
percolation may occur.
The type of system, its design and operation, and the soil and climatological condi-
tions at the time of irrigation all affect the application efficiency. Design of sprinkler
systems is based upon average application efficiencies. Typical efficiencies are sum-
marized in Table 16.2.
16.3.2 System Discharge
The discharge required for the irrigation system is one of the first considerations in
the design process. The discharge must be sufficient to meet crop water requirements
for the climatological conditions of the region. We are using the net system capacity as
the indicator of the supply required to meet crop needs. The net system capacity (Cn) is
the rate water must be continually supplied to satisfy plant water requirements.
Heermann et al. (1974) used a daily soil water balance to simulate the effect of net
system capacities on the soil water content. For a given capacity, they determined the
maximum soil water depletion reached during a year. They simulated the soil water
balance for a 59-year period to develop data for a probability distribution of the net
Table 16.2. Characteristics of sprinkler irrigation systems (adapted from Keller et al., 1980).
Installation Annual Cost of Labor per Irrigation, Application Coefficient
Type of Cost ($ Maintenance (hour ha-1) Efficiency of
-1
System hour ) (% of purchase) Preseason In Season (%) Uniformity
Hand move 300 2 0.25` 1.73 65-75 70-85
Towline 500 3 0.25 1.00 70-80 70-85

Side roll 2 0.25 0.86
675 70-80 70-85
Side move 4 0.49 0.62
Solid set 2000 2 0.25 0.15 70-85 75-85
Center pivot
standard 800 5 0.05 80-92 85-95
0.12
with corner 925 6 0.06
Lateral move
ditch fed -- 0.12 0.10
85-92 85-95
hose fed 1250 6 0.15 0.15
pipe fed -- 0.12 0.07
Travelers 900 6 0.25 0.62 60-70 70-80
system capacity requirement. Results from von Bernuth et al. (1984) were used to de-
velop design guidelines for center pivots in Nebraska (Figure 16.4). Howell et al.
(1989), and Lundstrom and Stegman (1988) used similar procedures for other areas in
the Great Plains.
The inflow rate required for the system is the amount of water that must be sup-
plied to avoid water stress. The system flow rate is given by
⎛ C T ⎞⎛ A ⎞ ⎛ d g Ai ⎞
QS = 0.116⎜⎜ n i ⎟⎟⎜⎜ i ⎟⎟ = 0.116⎜⎜ ⎟
⎟ 16.1
⎝ Ea ⎠⎝ To ⎠ ⎝ To ⎠
where Qs = inflow to the sprinkler irrigation system (i.e. gross system capacity), L s-1
Cn = net system capacity, mm day-1
Ti = irrigation interval, days
Ea = application efficiency, decimal fraction.
Ai = area irrigated, ha, and
To = time of operation per irrigation, days
dg = gross depth of irrigation water applied, mm
The irrigation interval is the time from the start of the irrigation until the beginning
of the subsequent irrigation. The irrigation interval is often a nominal value represent-
ing normal practices during the peak water requirement period of the irrigation season.
The time of operation is the amount of time that water is applied during the irrigation
interval. There are times when the irrigation system may be shut down for mainte-
nance, repositioning sprinkler laterals, farming operations, or convenience. This time
can be referred to as the downtime. The downtime is the difference between the irriga-
tion interval and the time of operation. The selection of the irrigation interval and
downtime must be coordinated with the operator and should not be arbitrarily selected
by the designer.
1.0
NET SYSTEM CAPACITY, L/s/ha

0.9
WESTERN NEBRASKA
0.8
0.7
0.6
EASTERN NEBRASKA
0.5
25 50 75 100 125 150
ALLOWABLE SOIL WATER DEPLETION, mm
Figure 16.4. Net system capacity requirement for center pivots in Nebraska
(adapted from von Bernuth et al., 1984).
The total system capacity cannot exceed the available water supply. If the capacity
requirements are too high, the amount of area irrigated may have to be reduced, or
deficit irrigation will be necessary.
16.3.3 Sprinkler Discharge
The discharge required for a sprinkler can be determined by the density of sprin-
klers and the number of laterals within the field. For solid set and moved-lateral sys-
tems the density is determined by the spacing between the sprinklers along a lateral
and the distance between sets along the main line (Figure 16.2). The representative
area for a single sprinkler is the product of the sprinkler spacing along the lateral (SL)
and the distance between laterals or the set width (Sm). The number of sprinklers on a
lateral (N) is determined by the length of the field (FL), the number of laterals used
along the field length (NL) and the sprinkler spacing (i.e., N = FL/NL SL). The number
of sets per lateral (n) is determined by the width of the field (FW), the number of later-
als along the width of the field (Ns) and the width of an individual set (n = FW /Ns Sm).
The number of sprinklers per lateral and the number of sets per lateral must be inte-
gers.
The irrigation interval depends on the number of sets per lateral (n), the operational
time per set for a lateral (Ts), the time required to move the sprinkler later between sets
(Tm), and downtime between successive irrigations:
Ti = n (Ts + Tm) + Td (16.2)
The combination of the operational time and the time required to move from one
set to the next is often referred to as the set time. The set time must be selected in con-
sultation with the owner/operator to match labor availability. The minimum downtime
is the time required to reposition laterals from the last set in the field to the beginning
set for the next irrigation. Additional time may also be required for maintenance and
farming operations.
The discharge from an individual sprinkler for such systems depends on (1) the
depth of water that must be applied per irrigation, (2) the representative area for one
sprinkler, and (3) the time that water is applied for an individual set, as
1 ⎛ CnTi ⎞⎛ S L Sm ⎞ d g S L Sm
qs = ⎜ ⎟⎜ ⎟= (16.3)
3600 ⎜⎝ Ea ⎟⎠⎜⎝ Ts ⎟⎠ 3600Ts
where qs = discharge from a sprinkler, L s-1

SL = spacing between sprinklers, m
Sm = distance between laterals, m
Ts = operational time per set for a single lateral, hours
Part-circle sprinklers can be placed at the end of the lateral to provide uniform irri-
gation without applying water beyond the field boundaries. The discharge for an end
sprinkler should be a fraction of the discharge of sprinklers along the lateral as based
on the proportion of a full circle that is watered with the end sprinkler.
Some solid set systems are designed so that the area between adjacent sprinklers
forms an equilateral triangle. For this arrangement the distance between sprinklers is
equal to the spacing along the lateral (SL) and the distance between laterals is Sm =
0.866 × SL. Relationships are developed in later sections for the discharge from indi-
vidual sprinklers on other types of systems.
The pressure available to the sprinkler device and the effective diameter of the noz-
zle determine the discharge from a sprinkler. For circular nozzles the relationship is
given by
qs = 0.00111 Cd Dn2 P (16.4)
-1
where qS = discharge per nozzle, L s
Cd = discharge coefficient
Dn = inside diameter of the nozzle, mm
P = pressure at the base of the sprinkler device, kPa.
The discharge coefficient, which incorporates characteristics of the sprinkler head
and nozzle, is unique for sprinkler devices and nozzles. Heermann and Stahl (2006)
developed equations to estimate the discharge coefficient for selected sprinkler de-
vices:
Cd = d0 + d1 Dn2 + d2P (16.5)
where d0, d1 and d2 are empirical parameters. Typical parameter values are listed in
Table 16.3 for selected sprinkler devices. The discharge from the range nozzle and the
spreader nozzle should be added for sprinklers equipped with two nozzles. For some
nozzles the orifice is not circular and the effective diameter of the nozzle should be
used in Equations 16.4 and 16.5.
16.3.4 Diameter of Coverage
The diameter of coverage, also called the wetted diameter, (Wd), is the size of the
circular pattern that is wetted by an individual sprinkler. The diameter of coverage has
a large influence on sprinkler and lateral spacing and thus the cost of a system. The
diameter of coverage depends on the velocity of the sprinkler jet, the angle of the noz-
zle axis above the horizon and the size of droplets that form when the sprinkler jet
breaks up. Up to a point, as pressure increases, the diameter of coverage increases and
more uniform application may result. The pressure should be increased as the nozzle
sizes increase to ensure that the spray breaks up adequately.
Sprinkler manufacturers provide performance data that includes the discharge and
diameter of coverage as a function of operating pressure and nozzle sizes as illustrated
in Table 16.4. The diameter of coverage provided by the manufacturer generally speci-
fies the height of the nozzle above the ground when the device was tested. Test results
are usually for no-wind conditions and include the recommended operating pressure
range for the listed configurations.
Table 16.3. Coefficients for the discharge coefficient (adapted from Heermann and Stahl, 2006).
Minimum
Model Pressure
Company Number (kPa) d0 d1 d2
Nelson Irrigation Corp. F32AS 245 0.954 5.061 × 10-3 1.859 × 10-5
Nelson Irrigation Corp. F32S 245 0.954 5.061 × 10-3 1.859 × 10-5
-3
Nelson Irrigation Corp. F33A 245 0.976 -1.617 × 10 -2.567 × 10-6
Nelson Irrigation Corp. F33AS 245 0.976 -1.617 × 10-3 -2.567 × 10-6
-3
Nelson Irrigation Corp. F33S 245 0.976 -1.617 × 10 -2.567 × 10-6
-4
Nelson Irrigation Corp. F33DN 196 1.011 -2.267 × 10 -3.441 × 10-4
-3
-5
Nelson Irrigation Corp. F43AP 245 0.945 -3.363 × 10 -2.928 × 10-6
-3
Nelson Irrigation Corp. F43P 245 0.975 -1.390 × 10 -8.234 × 10-6
-3
-3
Nelson Irrigation Corp. F80A 392 0.991 -1.002 × 10 3.213 × 10-5
-3
Nelson Irrigation Corp. R30 147 1.035 -2.845 × 10 -1.359 × 10-4
Nelson Irrigation Corp. R3000 196 0.294 0 0
Nelson Irrigation Corp. R30D4 294 0.969 0 0
Nelson Irrigation Corp. R30D6 294 0.957 0 0
Nelson Irrigation Corp. SPR1 98 0.986 -2.692 × 10-3 -9.696 × 10-5
Rain Bird Corp. 20 196 0.966 -7.768 × 10-3 7.394 × 10-5
-2
Rain Bird Corp. L20 245 1.059 -2.341 × 10 -4.154 × 10-6
-2
Rain Bird Corp. L2020 245 1.084 -1.585 × 10 -6.466 × 10-6
Rain Bird Corp. 30 245 0.995 -3.149 × 10-3 -5.449 × 10-6
-3
Rain Bird Corp. L30 245 1.009 -9.980 × 10 -8.572 × 10-6
-3
Rain Bird Corp. L3030 245 1.081 -4.962 × 10 -6.456 × 10-5
-3
Rain Bird Corp. 35 294 0.952 2.904 × 10 -4.927 × 10-6
-3
Rain Bird Corp. 40 245 1.000 -2.656 × 10 4.358 × 10-5
-3
Rain Bird Corp. 65 392 0.984 -3.061 × 10 7.541 × 10-5
-3
Rain Bird Corp. 70 392 1.002 -3.780 × 10 6.085 × 10-5
-3
Rain Bird Corp. 85 245 0.960 -1.278 × 10 3.300 × 10-5
-3
Rain Bird Corp. 8X 98 0.922 -9.870 × 10 1.005 × 10-4
-4
Senninger Irrigation Inc. S4006 245 0.963 -6.183 × 10 1.269 × 10-5
-3
Senninger Irrigation Inc. S5006 245 0.990 -2.502 × 10 -7.896 × 10-6
-4
Senninger Irrigation Inc. SSPR 59 0.983 -6.662 × 10 -8.100 × 10-5
Valmont Industries, Inc. SPR06 59 0.865 0 0
Valmont Industries, Inc. SPR10 98 0.816 0 0
Table 16.4. Performance data for an example sprinkler.

Nozzle Sizes (mm)
4.37 × 2.38 4.76 × 2.38 4.76 × 3.18 5.16 × 3.18 5.56 × 3.18 5.95 × 3.18
Nozzle Diam. Diam. Diam. Diam. Diam. Diam.
Pressure Flow cover Flow cover Flow cover Flow cover Flow cover Flow cover
(kPa) (L s-1) (m) (L s-1) (m) (L s-1) (m) (L s-1) m (L s-1) (m) (L s-1) (m)
173 0.35 26.8 0.40 27.4 0.47 27.4 0.52 27.7 0.58 28.4 0.64 28.7
207 0.38 28.0 0.44 29.0 0.51 29.0 0.58 29.9 0.64 30.5 0.71 30.8
242 0.41 29.0 0.47 29.9 0.56 29.9 0.63 30.8 0.70 31.4 0.76 32.0
276 0.44 29.6 0.51 30.5 0.60 30.5 0.67 31.4 0.75 32.3 0.82 32.6
311 0.47 30.2 0.54 31.1 0.63 31.1 0.71 32.0 0.80 32.9 0.87 33.5
345 0.50 30.5 0.57 31.7 0.67 31.7 0.75 32.6 0.84 33.5 0.92 34.1
380 0.52 30.8 0.60 32.3 0.70 32.3 0.79 33.2 0.88 34.1 0.96 34.8
Several types of nozzles are available for sprinklers. Conventional straight bore
(with round orifices) brass or plastic nozzles produce the maximum pattern radius and
good patterns at medium to high pressures. Noncircular orifice nozzles (with diffuse
jets) produce good patterns at reduced pressures but the wetted radius is usually
smaller than for straight bore nozzles. Kincaid (1982) developed an equation to predict
the diameter of coverage based on the nozzle pressure and discharge from the sprin-
kler. Heermann and Stahl (2006) used the diameter of the nozzle and the sprinkler
pressure to predict the wetted radius for a range of sprinklers:
( ) ( )
Wr = r0 + r1 Dn2 P + r2 Dn2 P
2
(16.6)
where Wr= radius of coverage, m

Dn = nozzle diameter, mm
P = nozzle pressure kPa,
r0, r1, r2 = empirical constants.
Values for the empirical parameters are listed in Table 16.5 for selected devices.
16.3.5 Time of Operation
The time of operation per set (Ts) should be selected to satisfy crop water require-
ments throughout the irrigation interval while avoiding deep percolation. The opera-
tional time must also be acceptable to the operator. The appropriate set time depends
on the rate of water application.
The average application rate is the average rate that water is applied to the crop/soil
surface. The application rate depends on the discharge and the representative area for
an individual sprinkler. The average application rate for solid set and periodically
moved systems is given by
3600 qS
Ra = (16.7)
S L Sm
where Ra = the average application rate, mm hour-1.

The rate of application is important since runoff may occur when the application
rate exceeds the infiltration rate. The recommended maximum and minimum applica-
tion rate for typical sprinkler layouts for periodically moved and solid set systems are
listed in Table 16.6. These values represent average conditions and should be adjusted
when local information is available or where runoff is prevalent.
Table 16.5. Coefficients for the wetted radius for selected sprinklers
(adapted from Heermann and Stahl, 2006).
Model Coefficients
Company Number r0 r1 r2
Nelson Irrigation Corp. F32AS 7.66 1.0188 × 10-3 -5.9017 × 10-8
Nelson Irrigation Corp. F32S 11.16 4.4340 × 10-4 -8.8466 × 10-9
Nelson Irrigation Corp. F33A 9.95 6.7152 × 10-4 -2.0504 × 10-8
Nelson Irrigation Corp. F33AS 9.95 6.7152 × 10-4 -2.0504 × 10-8
Nelson Irrigation Corp. F33S 10.74 6.5956 × 10-4 -1.7156 × 10-8
Nelson Irrigation Corp. F33DN 9.67 4.2715 × 10-4 -1.1985 × 10-8
Nelson Irrigation Corp. F43AP 9.36 3.9817 × 10-4 -6.0400 × 10-9
Nelson Irrigation Corp. F43P 12.34 4.5998 × 10-4 -7.2575 × 10-9
Nelson Irrigation Corp. R30 9.15 0 0
Nelson Irrigation Corp. R3000 6.40 0 0
Nelson Irrigation Corp. R30D4 8.54 0 0
Nelson Irrigation Corp. R30D6 7.62 0 0
Nelson Irrigation Corp. SPR1 3.73 7.7219 × 10-4 -4.2217 × 10-8
Rain Bird Corp. 20 10.54 5.2870 × 10-4 -3.3920 × 10-8
Rain Bird Corp. L20 8.30 9.5893 × 10-4 -8.8110 × 10-8
Rain Bird Corp. L2020 9.48 4.0092 × 10-4 -2.6504 × 10-8
Rain Bird Corp. 30 10.75 6.4496 × 10-4 -1.6816 × 10-8
Rain Bird Corp. L30 10.23 1.8772 × 10-4 -4.2494 × 10-9
Rain Bird Corp. L3030 10.27 1.6917 ×10-4 -3.1560 × 10-9
Rain Bird Corp. 35 10.64 6.9710 × 10-4 -1.9120 × 10-8
Rain Bird Corp. 40 12.00 5.1893 × 10-4 -9.1826 × 10-9
Rain Bird Corp. 65 13.09 3.6063 × 10-4 -3.0844 × 10-9
Rain Bird Corp. 70 16.52 2.7829 × 10-4 -1.2432 × 10-9
Rain Bird Corp. 85 16.69 1.9212 × 10-4 -6.1428 × 10-10
Rain Bird Corp. 8X 6.47 3.4306 × 10-4 -6.4353 × 10-9
Senninger Irrigation Inc. S4006 11.02 5.0444 × 10-4 -1.3677 × 10-8
Senninger Irrigation Inc. S5006 11.49 3.8105 × 10-4 -3.3509 × 10-9
Senninger Irrigation Inc. SSPR 3.83 4.1793 × 10-4 -1.3894 × 10-8
Valmont Industries, Inc. SPR06 3.52 5.2760 × 10-7 0
Valmont Industries, Inc. SPR10 2.97 7.6539 × 10-4 -2.9334 × 10-8
Table 16.6. Maximum application rate and depth of infiltration based on soil type
and management criteria (partially from Keller and Bleisner, 1990).
Maximum Application Rate Available Maximum Depth of Infiltration
(mm h-1) Water- (mm) for a root zone
Holding 1 m deep with allowable
Soil Slope (%)
Capacity depletions of:
Soil Texture 0-5 5-8 8-12 >12 (mm m-1) 35% 50% 65%
Coarse sand 50 40 30 20 50-70 21 30 39
Fine sand 40 32 24 16 75-85 28 40 52
Loamy fine sand 35 28 21 14 85-100 32 45 59
Sandy loam 25 20 15 10 110-125 42 60 78
Fine sandy loam 20 16 12 8 130-150 49 70 91
Very fine sandy loam 15 12 9 6 145-165 53 75 98
Loam 13 10 8 5 150-170 56 80 104
Silt loam 13 10 8 5 160-200 63 90 117
Sandy clay loam 10 8 6 4 140-170 53 75 98
Clay loam 8 6 5 3 150-180 56 80 104
Silty clay loam 8 6 5 3 140-180 56 80 104
Clay 5 4 3 2 130-180 53 75 98
The depth of water applied per irrigation (i.e., the volume of water per unit area of
land) for solid set and moved-lateral systems can be computed from
3600 qS TS
d g = RaTS = (16.8)
S L Sm
The depth of water applied and the corresponding application rate for other types of
sprinkler systems will be presented in later sections.
The net depth of application (dn = Ea dg) should not exceed the soil water depletion
at the time of irrigation; yet, the net depth must be large enough to satisfy the crop
water use during the irrigation interval. To satisfy these constraints the time of opera-
tion per set must be consistent with the irrigation interval and the net system capacity.
If the depth required for the net capacity exceeds the depletion at the time of irrigation
more laterals are generally required along the field width. For design purposes the
depletion is computed as
AD = MAD RD TAW (16.9)
where AD = allowable depletion, mm
MAD = management allowed depletion, decimal fraction
RD = root depth during peak water use period, m
TAW = total available water per unit depth of soil, mm of water per m of soil.
Representative values for the parameters in Equation 16.9 are available in Chapter
3 and Table 16.6. The net depth of water must then fit within the following range:
3600 qS TS Ea
CnTi ≤ d n = ≤ AD = MAD RD TAW (16.10)
S L Sm
where dn is net application depth (mm). This equation can be revised to give the
maximum and minimum limits for the operational time per set:
n Tm + Td MAD RD TAW S L Sm
≤ TS ≤ (16.11)
86400 qS Ea 3600 qS Ea
−n
Cn S L Sm
The results show that the ability of the soil to hold the net application provides the
maximum set time while the time required to satisfy the net capacity provides the
minimum set time (Figure 16.5). For conditions between the limits the system can be
shut off for more than the one day of downtime as used in the figure. These results
illustrate that the amount of downtime increases as the sprinkler discharge increases. If
there are no feasible combinations the number of laterals along the field width should
be increased. While the equation for the time of operation is complex, it summarizes
the considerations required for a given area and soil. The equation is easily analyzed
with a spreadsheet program. Solution often involves a trial and error procedure.
16.3.6 Drop Size Distribution
The size of sprinkler droplets profoundly affects the performance of sprinkler irri-
gation systems. The analyses by Edling (1985), Kohl et al. (1987), Kincaid and Long-
ley (1989), and Thompson et al. (1993) show that small droplets evaporate faster than
large droplets (Figure 16.6). Individual droplets are often assumed to be spherically
shaped. Therefore the surface area is given by (πd2) while the mass of the droplet is
given by (πρd3/6) where d is the diameter of the droplet and ρ is the density of water.
20
Sprinkler Spacing 12.2 m
Lateral Spacing 18.3 m
Applic. Efficiency 0.75
Root Depth 1.2 m
MAD 0.50
16 Sets / Lateral 20
Time to Move Lateral 1h
TIME OF OPERATION PER SET, hours
Downtime / set 24 h
TOTAL
AVAILABLE
12 WATER, mm/m
200
8 160
6 7 8 120
4 NET SYSTEM
CAPACITY,
mm/day
0
0.4 0.5 0.6 0.7 0.8 0.9 1.0
SPRINKLER DISCHARGE, L/s
Figure 16.5. Relationship of acceptable time of operation to sprinkler discharge
when net capacity water requirements form the minimum operational time
and the soil water holding capacity constrains the maximum operational time.
25
Impact Sprinklers
38˚C Air Temperature
20
20% Relative Humidity
EVAPORATION LOSS, %
20˚C Water Temperature
15
10
0.2 0.4 0.6 0.8 1.0 1.5 2.0 3.0 4.0 5.0
DROPLET DIAMETER, mm
Figure 16.6. Illustration of the effect of droplet size on relative evaporation rates.
Accordingly, the ratio of the surface area to the mass of the droplet varies inversely
with the diameter of the droplet. Hence, a higher portion of the water in a small drop-
let is exposed to the atmosphere which results in larger relative evaporation rates. De-
vices that produce large drops should reduce evaporation loss and would seem to in-
crease the application efficiency.
Research has also shown that small drops are very prone to drift in windy condi-
tions. The drag force caused by the difference in velocity between the droplet and the
air is proportional to the projected area of the droplet. The momentum is proportional
to the mass of the droplet. Therefore, drag forces are more significant for small drops
than large drops. The large influence of drag forces causes small drops to decelerate
rapidly in still air. Since momentum is more significant for large drops, they are less
affected by drag and decelerate more slowly than small drops. These processes result
in a variation of droplet sizes with distance from the sprinkler device. In still air, small
drops fall closer to the sprinkler while large drops travel further. In windy conditions
the additional drag force from the wind transports droplets downwind. This process is
called drift. Drift can reduce irrigation efficiency when the water is deposited outside
of the field. Uniformity can be reduced if drift results in significant distortion of the
application pattern for a prolonged period.
The momentum of large drops resists drift in windy conditions; however, it nega-
tively affects the soil surface. Levine (1952) and others have shown that the impact
energy contained in water droplets leads to breakdown of soil aggregates and the for-
mation of a seal on the soil surface which reduces infiltration and can lead to runoff.
Levine (1952) showed that sandy soils were less affected by droplet impact than finer
textured soils. The kinetic energy of a drop can be expressed as πρd3v/12 where v is
the velocity of the droplet.
Stillmunkes and James (1982) summarized literature showing that sealing is related
to the amount of kinetic energy per unit area at impact and the accumulation of the
energy over time. The kinetic energy per unit area increases with droplet diameter, but
reaches a plateau where the kinetic energy per unit area is nearly constant as the drop
size increases. Similar results were presented by Kohl et al. (1985). The kinetic energy
per unit area is proportional to the diameter of the droplet while the drag coefficient as
presented by Seginer (1965) is inversely proportional to the diameter of the droplet.
These interactions are such that there is relatively little difference in the velocity of
droplets as the size of the drop exceeds 2 or 3 mm (Figure 16.7). Stillmunkes and
James (1982) concluded that the kinetic energy per unit area is insensitive to the drop
size if the drops are larger than 3 mm and the kinetic energy per unit area is dependent
on the application rate and the length of time that water is applied. The equation from
Stillmunkes and James (1982) for the kinetic energy per unit area is given by
Ke ρdv 2 ρRTv 2
= = (16.12)
a 2 2
where Ke/a = kinetic energy per unit area
ρ = density of water
d = depth of water applied
v = droplet velocity.
R = average application rate
T = exposure time
10
20
9 8
6
5
8
4
7 3
VELOCITY, m/s
6 2
5
1
4
3 0.5
2 FALL HEIGHT, m
0
0 1 2 3 4 5 6 7
DROP DIAMETER, mm
Figure 16.7. Effect of drop size and height of fall on the velocity of droplets
(adapted from Stillmunkes and James, 1982).
Data from several sources were used by von Bernuth and Gilley (1985) to estimate
the relative infiltration rate for bare soils compared to soils protected with mulch or
crop cover (Ir):
0.683 1.271 − 0.353 0.237
I r = 1 − 0.0354 d50 v Sa Si (16.13)
where Ir = rate of infiltration for bare soil relative to protected soil, decimal fraction
d50 = volume median drop size, mm
v = velocity of the volume mean size droplet, m s-1
Sa = sand content of the soil, %
Si = silt content of the soil, %.
The effect of the drop size and velocity on infiltration is illustrated in Figure 16.8.
While the evaporation and drift of small drops are significant, the relative volume of
water in a particular size range of drops must be considered. For example, the mass of
water in a 5-mm diameter drop is 1000 times the volume in a 0.5-mm droplet. The
drop size distribution must be considered to determine if losses could be significant.
1.0
2
RELATIVE INFILTRATION RATE

4
6
0.8 8
Droplet Velocity, m/s
0.6
0.4
0.2
SANDY LOAM
0.0
0 1 2 3 4 5 6 7
DROP SIZE, mm
1.0
2
RELATIVE INFILTRATION RATE
0.8 4
6
0.6
8
0.4
Droplet Velocity, m/s
0.2
SILT LOAM
0.0
0 1 2 3 4 5 6 7
DROP SIZE, mm
Figure 16.8. Effect of droplet impact energy on infiltration rates.
1.0
STATIONARY PAD
0.8 - Smooth Plate
RELATIVE FREQUENCY OF DROP SIZE

STATIONARY PAD
- Medium Grooved Plate
0.6
STATIONARY PAD
- Coarse Grooved Plate
0.4
IMPACT SPRINKLER
- Straight Bore Nozzle
- 4 mm Diameter
0.2 - 400 kPa
0.0
0 1 2 3 4 5 6
DROP SIZE, mm
Figure 16.9. Examples of drop size distributions for a stationary pad sprinkler device
with three styles of plates and an impact sprinkler with a straight bore nozzle.
This has led to methods to determine the distribution of drop sizes from sprinkler
devices operated at varying pressures and heights. Various methods have been used to
measure the size of water droplets (see Solomon and Bezdek, 1980; Eigel and Moore,
1983; Kohl and DeBoer, 1984; and Dadiao and Wallender, 1985 for descriptions of
the various techniques). Examples of drop size distributions are illustrated in Figure
16.9. Bezdek and Solomon (1983) analyzed various mathematical functions to de-
scribe drop size distributions. They concluded that the upper-limit lognormal distribu-
tion was adequate for many experimental data sets. The disadvantage of this distribu-
tion is that the solution is difficult to derive from experimental data.
Until recently, the limitation on considering drop size distributions in design has
been the scarcity of experimental data for commonly used sprinklers. Kincaid et al.
(1996) developed methods to estimate the distribution of drop sizes for fourteen sprin-
kler devices. They analyzed four models of impact sprinklers that were equipped with
straight bore, flow control, or square nozzles. They also examined ten spray head de-
vices equipped with jets that impinged onto fixed or moving plates. An exponential
model was used for the drop size distribution. Li et al. (1994) had indicated that the
exponential model was comparable in accuracy to the upper-limit lognormal model
but much easier to use. The exponential distribution model is given by
⎧ ⎡ η ⎫
⎪ ⎛ d ⎞ ⎤⎪
Pv = 100⎨1 − exp ⎢− 0.693⎜⎜ ⎟ ⎥⎬
⎟ (16.14)
⎪⎩ ⎢ ⎝ d 50 ⎠ ⎥⎦ ⎪
⎣ ⎭
where Pv = percent of the total drops that are smaller than d
d = drop diameter, mm
d50 = volume mean drop diameter, mm

η = dimensionless exponent.
Data were developed for the values of d50 and η for each nozzle and sprinkler com-
bination for a range of operating pressures and nozzle sizes. The exponential model
accurately represents the drop size distribution for drops larger than 3 mm and overes-
timated percentage volumes for smaller drop sizes.
Kincaid et al. (1996) developed a procedure to estimate the parameters required for
the exponential model. The procedure used the effective diameter of the nozzle and
the nozzle pressure. They showed that the ratio of the nozzle diameter to the pressure
head at the nozzle could be used to describe the volume mean drop diameter and the
empirical parameter η. The relationships that they developed are given by
d50 = ad + bd Ω and η = an + bn Ω (16.15)
where Ω = ratio of the nozzle diameter to the pressure at the base of the sprinkler device
ad, bd, an, bn = regression coefficients.
Results were grouped into seven categories as describe in Table 16.7. These results
provide a more general method to estimate the effects of the design and operation of
irrigation systems on the drop sized generated.
Kincaid (1996) presented data on the kinetic energy of spray droplets. Relation-
ships were developed to predict the kinetic energy per unit mass based on the ratio of
the nozzle diameter to the pressure head:
⎛ N e2 ⎞
Ek = e0 + e1⎜ de ⎟ + U 1.5 (16.16)
⎜H 3 ⎟
⎝ n ⎠
where Ek = kinetic energy, kJ kg-1
Nd = diameter of the nozzle, mm
Hn = nozzle pressure head, m
U = wind speed, m s-1
e0,e1,e2,e3 = regression constants.
Values for the regression constants are summarized in Table 16.8. Kincaid (1996) also
correlated the kinetic energy to the mean volumetric drop size and the percentage of
the drops that exceed a diameter of 3 mm:
Ek = 2.79 + 7.2 d50 and Ek = 10.41 + 0.249 P3 (16.17)
where P3 = the percentage of the drops that are smaller than 3 mm.
Table 16.7. Coefficients for estimating drop size distribution parameters.
Type of Sprinkler Device ad bd an bn
Impact sprinkler,
small round nozzles (3-6 mm) 0.31 11900 2.04 -1500
Impact sprinkler,
large round nozzles (9-15 mm) 1.30 2400 1.82 300
Wobbler 0.78 1870 2.08 -630
Rotator, 4 groove 1.07 3230 1.70 -830
Rotator, 6 groove 0.81 1480 2.07 -1300
Concave, 30 groove plate 0.82 620 2.68 -750
Flat smooth plate 0.66 680 2.74 -920
Table 16.8. Coefficients for estimating energy from sprinkler devices.

Sprinkler Device e0 e1 e2 e3
Impact, large round nozzles 14.1 45.1 0.5 1.0
Impact, small round nozzles 6.9 132 0.5 1.0
Rotators, 4 groove plates 12.1 50.7 0.5 1.0
Rotators, 6 groove plates 8.9 38.0 0.5 1.0
Spinners, 6 groove plate 6.9 36.9 0.5 1.0
Low drift nozzle (LDN) 10.4 0.57 2 0.5
Medium groove stationary plate 6.2 0.45 2 0.5
Smooth groove stationary plate 5.4 0.40 2 0.5
The wind speed has a major impact on the kinetic energy, increasing the energy per
unit mass fivefold for wind speeds of 10 m s-1 compared to still air. Moldenhauer and
Kemper (1969) showed that the infiltration rate decreased by one order of magnitude
when the cumulative drop energy per unit area exceeded 500 J m-2. Spray devices that
produce small drops with 5 J kg-1 could apply 100 mm of water before reduction oc-
curs. For devices that produce larger drops with impact energies of 20 J kg-1 the
threshold would be reached with 25 mm of water. These results allow designers to
evaluate the potential runoff problems for various sprinkler devices. The concepts pre-
sented to this point can be applied to the design of sprinkler systems.
16.3.7 Sprinkler Placement
Sprinklers must be properly placed and aligned. Occasionally sprinklers are not
placed high enough to provide an unobstructed path for the sprinkler jet. The canopy
that interferes with the water jet reduces the diameter of coverage and leads to poor
water distribution. For row crops, sprinklers should be at least 0.5 m above the tallest
mature crop that will be irrigated. For orchard crops the sprinkler should be placed to
provide the wetted soil zone without causing degradation of fruit quality by wetting
leaves and fruits. If the sprinkler is to be used for frost control, the sprinklers must be
high enough to provide coverage of the crop canopy. On center pivots and lateral
move systems the sprinkler devices may be placed below the lateral. The sprinklers
should be placed so that the water stream does not impact on structural components of
the machine. If the sprinkler devices are placed below the top of the crop canopy, they
should be placed close enough along the pipeline to provide plants with equal access
to water. This often requires that devices be positioned above alternate furrows.
The vertical alignment of the sprinkler riser is also important. Nderitu and Hills
(1993) showed that the diameter of coverage is reduced if the sprinkler riser is tilted
from vertical. They showed that the precipitation profiles were nearly the same when
the sprinkler riser was within 10° of vertical. However, the application uniformity
decreased when the riser was tilted by 20°. Sprinklers that are firmly supported pro-
duced higher uniformities than unsupported risers.
16.4 APPLICATION UNIFORMITY

16.4.1 Single-Leg Distribution
The uniformity of application from a sprinkler system depends on the distribution
of water from individual devices. The distribution is measured with catch containers
placed around a sprinkler as shown in Figure 16.10. The sprinkler is operated long
enough to measure the depth of water applied at distances from the sprinkler. Meas-
urement containers should be large enough and deep enough to provide an accurate
measurement (Kohl, 1972). Fisher and Wallender (1988) showed that the measure-
ment accuracy was directly related to the diameter of the catch container. If the con-
tainers are used in an open field, evaporation should be estimated. Various methods
are available including using oil in the containers to suppress evaporation (see Heer-
mann and Kohl, 1980). If the irrigation system will operate in calm and windy condi-
tions, it is best to measure the distribution under both wind conditions.
SPRINKLER LOCATION
COLLECTOR LOCATION
AREA WETTED
BY A SINGLE
SPRINKLER
DIAMETER OF
COVERAGE
DEPTH OF WATER APPLIED
ELLIPTICAL PATTERN
TRIANGULAR PATTERN
WETTED
RADIAL DISTANCE FROM SPRINKLER
RADIUS
Figure 16.10. Arrangement to measure the single-leg water distribution of a sprinkler.

Elliptical and triangular distribution patterns for the single-leg distribution are shown.
The water application pattern about a single sprinkler is the single-leg distribution.
The most commonly used equations are for a triangular and an elliptical distribution
(Figure 16.10). Equations for the distributions are given by
3q S To
d (r ) = (Wr − r ) for a triangular pattern and
πWr3
(16.18)
3q S To
d (r ) = Wr2 − r 2 for an elliptical pattern
2πWr3
where d(r) = the depth of water applied at a radial distance r from the sprinkler,
Wr = the radius of coverage or wetted radius of the sprinkler
To = the duration of the sprinkler operation.
16.4.2 Overlapping (Stationary Laterals)
Several sprinklers should apply water to a location in the field to achieve an accept-
able uniformity of application. The total depth of application can be estimated by
overlapping the depth determined from the single-leg distribution for individual sprin-
klers. An illustration of the procedure is shown in Figure 16.11. This example is based
on an impact sprinkler with 4.76 × 3.18 mm nozzles operated at 350 kPa as listed in
Table 16.4. An elliptical distribution and an irrigation set time of 10 hours were as-
sumed for the example. These conditions produce the single-leg distribution given by
d (r ) = 3 16 2 − r 2 . A distribution of catch containers is assumed to compute the uni-
formity. In this case the containers are placed on a 3 m × 3 m grid. The first container
is located half the grid spacing from the sprinkler. To compute the depth of water ap-
plied at each container the radial distance from a sprinkler to the container (r) is com-
2 2
puted by r = x + y where x is the horizontal distances from the lateral and y is the
vertical distance to the sprinkler. The first four values for each container in Figure
16.11 are arranged by sprinkler starting with the upper-left sprinkler and continuing to
the lower-right sprinkler. The top and bottom rows of containers receive water from
upstream and downstream sprinklers that are not shown in the figure. The fifth and
sixth rows of the data shown in the figure represent the contribution for those sprin-
klers. The total depth of water applied is shown as the last value in the column for
each container. For this example the maximum depth applied was 145 mm while the
minimum was 89 mm. The coefficient of uniformity was computed to be 87 using
procedures described in Chapter 4.
Overlapping provides a means to evaluate sprinkler spacing in the design of
moved-lateral or solid set systems. Distribution data are available from the manufac-
turer or testing organizations for many sprinkler devices. Actual data from a single-leg
test conducted for normal wind conditions provide more representative information for
assessing the uniformity. The effect of overlapping applications can also be computed
for moving systems. The procedure is somewhat more involved and is discussed in a
later section for those systems.
DISTANCE FROM LEFT LATERAL, m
0 1.5 4.5 7.5 10.5 13.5 16.5 18
48 43 37 27 0
1.5 46
DISTANCE FROM TOP SPRINKLER, m

0 27 37 43 46 48
37 35 30 20 0 0
0 0 20 30 35 37
27 24 15 0 0 0
0 0 0 15 24 27
112 132 145 145 132 112
4.5 46 45 41 35 24 0
0 24 35 41 45 46
43 41 37 30 15 0
0 15 30 37 41 43
89 125 143 143 125 89
7.5 43 41 37 30 15 0
0 15 30 37 41 43
46 45 41 35 24 0
0 24 35 41 45 46
89 125 143 143 125 89
10.5 37 35 30 20 0 0
0 0 20 30 35 37
48 46 43 37 27 0
0 27 37 43 46 48
27 24 15 0 0 0
12 0 0 24 27
0 15
112 132 145 145 132 112
SPRINKLER
LATERAL CU = 87
CATCH CONTAINER
Figure 16.11. Distribution of water from overlapping the single-leg distribution for each
sprinkler at each container location.
16.4.3 Wind Effects

Wind has a pronounced effect on the distribution of water from sprinklers. The
work by Christiansen (1942), Vories and von Bernuth (1986), and Seginer et al.
(1991a) and others showed how the application pattern of a single sprinkler is dis-
torted in the wind. The general effects of wind are illustrated in Figure 16.12. The pat-
tern is transported downwind as would be expected; however, the diameter of cover-
age perpendicular to the wind is reduced. The narrowing of the pattern perpendicular
to the wind has an impact on the layout of sprinkler laterals. As illustrated the narrow-
ing of the diameter of coverage can lead to poor uniformity if laterals are not placed
closer together when the wind is parallel to the lateral. To compensate for the smaller
diameter, sprinklers must be located closer together in the perpendicular direction. It is
more economical to space sprinklers closer together along the lateral rather than space
laterals closer together. Thus, the general recommendation is to orient laterals perpen-
dicular to the prevailing wind and to space sprinklers closer together along the lateral
than the spacing between the laterals. General guidelines have been developed for the
maximum spacing to maintain acceptable uniformities (Table 16.9).
Models have been developed to predict the distribution of water about a sprinkler
(Vories et al., 1987). The models treat water droplets as ballistic objects and the equa-
tions of motion which include the effects of gravity and the drag on the drops are
SPRINKLER
WIND DIRECTION
Wd
PATTERN
WITHOUT WIND
PATTERN
WITH WIND
LATERAL
WIND DIRECTION
LATERAL PERPENDICULAR LATERAL PARALLEL

TO WIND TO WIND
Figure 16.12. Effect of wind on sprinkler distribution and resulting water application
uniformity. Note that orienting laterals perpendicular to the prevailing wind
direction is generally the most economical arrangement.
Table 16.9. Maximum sprinkler and lateral spacing as a percentage of the effective wetted
diameter for sprinklers operating at the average pressure along the lateral.
Wind Condition Sprinkler Spacing Lateral Spacing
No wind 45 65%
8 km h-1 40 60%
8-16 km h-1 35 50%
> 16 km h-1 30 30%
solved. The increased drag resulting from high wind speeds causes droplets to move
down wind. Seginer et al. (1991b) showed that the distribution could be simulated if
the drag coefficient was adequately described. The formulation they used for the drag
coefficient was sensitive to the type of sprinkler used. The adjustment to the drag coef-
ficient requires outdoor measurements for acceptable accuracy. The coefficient of uni-
formity was quite sensitive to the adjustment made to the drag coefficient.
Han et al. (1994) developed a mathematical model of the effects of wind on the
single-leg distribution. They used an ellipse to describe the effect of wind on the hori-
zontal shape of the distribution pattern and shape patterns across four principal sec-
tions of the pattern to predict the depth of application. They also conducted tests of
numerous combinations of sprinkler devices, nozzle types, operating pressures and

wind speeds. While their model is more empirical than the work by Seginer et al.
(1991b) it does provide a means to predict the three dimensional distribution of water
about a sprinkler in windy conditions.
16.5 SOLID SET SYSTEMS
One method to minimize labor and automate sprinkler irrigation is to permanently,
or for a single season, install laterals at intervals across the field. This type of system
is called a solid set system (Figure 16.13). Solid set systems are adaptable to a wide
range of soils, crops, topography and field shapes. Being relatively expensive, they are
commonly used for high-value crops (orchards, turf, and seedling establishment) to
save labor and for environmental control. Solid set systems can be temporary or per-
manent. Temporary systems use aboveground aluminum or plastic laterals placed in
the field at the beginning of the season, left in place for at least one irrigation and re-
moved prior to harvest. Permanent systems use buried plastic, asbestos cement, or
coated steel pipe for main and laterals with risers or riser outlets aboveground.
Controllers allow complete flexibility in operation of solid set systems. Sophisti-
cated solid state controllers are available to control normally open diaphragm valves
on sprinkler laterals. Two types of controls are commonly used. For one method, low-
voltage (24 VAC) electrical solenoids are used as pilot valves to control water pres-
sure to a diaphragm. The second method uses a hydraulic system with small tubing to
supply air or water pressure directly to the diaphragm to close the valve. Solid set sys-
tems are frequently designed with a control valve at the inlet to the lateral. The valves
are normally closed and are opened by supplying an electrical excitation for the length
of time that water is to be applied. This allows each lateral to operate independently. A
controller is used to determine how long each lateral is operated. When multiple later-
als are operated simultaneously, the combination is called a circuit or a zone. Solid set
systems provide excellent control of the amount of water applied. Some characteristics
of solid set systems are summarized in Table 16.1.
The disadvantages of to solid set systems include:
The high cost to install and maintain—More laterals are required than for a peri-
odically moved system, which increases costs substantially. The electronic
valves and the controller also increase the costs. Costs increase substantially
when the main line, submains and laterals are buried. All of the working parts
also require maintenance for proper operation.
The inflexibility due to installation of the system at specific locations in the
field—If production practices change, such as changes in implement width or
row spacing, it is very difficult to modify the layout of the solid set system to fa-
cilitate the new management practices.
Its inconvenience to farm around—The systems are difficult to farm around
unless perennial crops are planted and remain in the same location for prolonged
periods.
The advantages of the solid set system are that:
The systems apply water uniformly across the field.
The systems provide push-button control.
Properly designed solid set systems can satisfy auxiliary needs such as frost con-
trol.
Figure 16.13. Diagram of solid set sprinkler irrigation system.
16.5.1 Sprinkler Selection, Performance, and Spacing

In general, solid set systems are designed to use low-flow, medium-pressure sprin-
klers. However, large sprinklers may be used if they are manually moved or individu-
ally valved. Sprinkler spacings will vary from 9 m by 9 m, to 73 m by 73 m. Nozzle
sizes can be as small as 1.59 mm to as large as 36 mm, pressures range from 172 to
620 kPa. The sprinkler spacing depends upon the sprinkler and nozzle combination,
operating pressure, desired coefficient of uniformity (CU), wind speed, and use of the
system. For certain high-value crops, it may be desirable to design for a high CU. A
crop of lesser value may not justify the cost of a design with a high CU. Since it is not
possible to design for all wind conditions, the system should be designed for average
conditions. A system designed for frost and freeze protection may not require as high a
CU as a system for soil moisture control. A system that is used to supplement rainfall
may not need as high a CU as one in which crop production totally depends upon irri-
gation.
16.5.2 Designing for Constant Sprinkler Discharge

The general design procedure for solid set systems where sprinkler nozzle and pipe
sizes are uniform has been described. As indicated there is variation of flow along the
lateral. The following procedure can be used to design individual sprinkler laterals, or
groups of laterals and associated main line or submain sections where there is less
variation once the sprinkler spacing and discharges have been specified. The flow
within any section of pipe can be determined. The procedure is iterative for selection
of pipe sizes and input pressure. The procedure can be used to design complete sys-
tems or subsections of large systems and includes the following steps:
1. Assuming the desired sprinkler spacing has been selected, the first step is to lay
out a system of laterals and main line on a topographic map of the field, or
measure the elevation of each proposed sprinkler location, as well as the loca-
tion and elevation of the system inlet. The elevation of each sprinkler outlet po-
sition is determined.
2. Calculate the flow in each pipe section as the total of all sprinkler downstream
of that section. Flows need to be recalculated whenever the sprinkler flow or the
number of operating sprinklers changes.
3. Pipe sizes are selected for main line and lateral sections. Initially, a large, uni-
form pipe size can be selected for main line and laterals, and then pipe sizes can
be reduced in certain areas as the design is optimized.
4. Establish an assumed pressure head at an initial point in the system, the inlet be-
ing a convenient point. The initial pressure may be set low or equal to the mini-
mum adequate sprinkler pressure.
5. Calculate pressures at all points by working upstream or downstream one pipe
section at a time or by using the pressure distribution relationship previously
presented.
6. Evaluate the pressure distribution. If some sprinkler pressures are inadequate,
increase the input pressure and go back to Step 5 until the minimum sprinkler
pressures are obtained. If all sprinkler pressures are within desired limits, the de-
sign may be acceptable, but some pipes may be oversized. If the range of pres-
sures exceeds the desired limit, some pipes may be undersized, or elevation dif-
ferences may be too large.
7. Reduce pipe sizes in selected areas, usually near the ends of laterals, or low ele-
vation areas, and go back to Step 5. Repeat as necessary until pipe sizes and
pressure distribution are optimized.
When the pressure distribution has been calculated, the required nozzle size for
each sprinkler can be calculated using the specified flow and calculated pressure. If
the range of pressures is sufficiently narrow, one nozzle size can be used. Alterna-
tively, pressure regulated sprinklers or flow-control nozzles can be used.
If only a fraction of the laterals are to be operated at one time, the main line design
should be checked with each set operating to ensure adequate pressure for all sets. The
main line size(s) can usually be minimized by spreading the operating laterals as uni-
formly as possible over the whole main line. However, it may be desirable for cultural
reasons to concentrate the operating laterals, in which case the set farthest from the
inlet will usually dictate the main line design.
The variability of topography within each group of sets must be considered. If ele-
vation differences are great in the direction perpendicular to the lateral, pressure regu-
lated sprinklers or flow-control nozzles are desirable. Also, if different numbers of

laterals may be operated at different times, inlet pressure may change.
The sprinkler flow, input pressure, pipe roughness, or any of the pipe sizes can be
changed recomputed when using a spreadsheet. If the spacings are changed, the layout
and thus the elevations would need to be changed accordingly. The laterals can be
shortened by setting downstream pipe diameters to zero.
In the case of uphill flows, large unavoidable pressure differences may be encoun-
tered. Pressure regulating valves can be located at lateral inlets or other points to re-
duce and limit the pressure to a specific value. Such controlled-pressure points can be
used as the starting point for pressure calculations. Alternatively, individual pressure
regulators can be used on sprinklers to limit the nozzle pressure and the inlet pressure
can be increased to maintain the minimum pressure throughout the system. The fol-
lowing examples will illustrate the design procedure.
Example 1. A 160-m × 180-m field is to be irrigated as shown in Figure 16.14. A

spacing of 17 m along the main line and 15 m along the laterals is selected. The sprin-
kler flow is 0.8 L s-1. The inlet is at the pump, located one main line section upstream
of the first lateral. The pump elevation is 10 m. Figure 16.15 shows the calculations
using a spreadsheet program. Numbers in bold type indicate the required input data.
Part A gives the pipe roughness (Hazen-Williams C), sprinkler flow, spacings, pump
(inlet) pressure and elevation, total flow and computed average and percent variation
in sprinkler pressure.
WATER SOURCE AND PUMP
MAINLINE
LATERAL VALVE
1 2 3 4 5 6 7 8 9 10
v v v v v v v v v v
1
SECTION
3 9
8
5
6
7
7
8 6
9 5
10 4
ELEVATION 3
SPRINKLER
CONTOUR, m
FIELD BOUNDARY
Figure 16.14. Plan view of field irrigated with a solid set sprinkler system.
Figure 16.15. Design of mainline and laterals for a solid set system with 10 laterals, where:
Pm is the pressure head at the main line outlet or lateral inlet, Pmin is the minimum pres-
sure and DP is the pressure difference on a lateral, Ns is the number of sprinklers on a
lateral (ON:OFF = 0,1), Ql is the total lateral flow, Qtot is the total flow in the mailine sec-
tion, and EL is the elevation of the inlet of a main line section (m).
Part B gives the number of sprinklers or pipe sections on each lateral, elevations
and diameters of the main line section upstream of each lateral, and computed main
line pressures, minimum lateral pressures, pressure difference on each lateral, flow in
each main line section and lateral, and main line velocities. Laterals are designated as
being on or off by entering 1 or 0 in the ON:OFF column (Figure 16.15). Individual
laterals can be turned off to simulate smaller sets or movable lateral systems.
Parts C to F give the elevations, pipe diameters, sprinkler pressure heads, and com-
puted nozzle sizes in grid form corresponding to the grid layout of Figure 16.15. The
sprinklers are located at the downstream end of each lateral section.
Following the above procedure, the elevations and pipe sizes were input (Parts C
and D). The flows in the main and laterals are computed by summing the sprinkler
flows downstream of each section. The assumed input pump pressure of 50 m be-
comes the inlet pressure for the first mainline.
All 10 laterals are to be operated simultaneously, so the flow in the first section is
80 L s-1. The diameter is 200 mm, and friction loss is 0.6 m. The outlet pressure for the
first section is 50 – 0.6 + 10 – 9.9 = 49.5 m, which becomes the inlet pressure for the
second main line section and also the inlet pressure for the first lateral. Pressures were
calculated down the first lateral one section at a time, and the minimum was found to
be 39.9 m (in section 8 since the laterals are running down slope). Successive main
and lateral sections were calculated to the end of the last lateral. The overall minimum
pressure head was 39.6 m in lateral 2.
Pipe size selection and readjustment is the main iterative process in this procedure.
In this example pipe sizes were adjusted so the pressure difference within the laterals
is less than 20% of the minimum pressure. The pipe size on the first section of the first
2 laterals was reduced to decrease the lateral pressures and thus reduce the overall
pressure variation to less than 20%. The computed nozzle sizes are thus nearly uni-
form, and one nozzle size can be selected. If variable nozzle sizes are needed to main-
tain uniform flows, the designer can select the available nozzle size closest to the
computed size.
Example 2. This example uses the same field and lateral layout in Example 1, op-
erated by running two adjacent laterals per set. Results are given in Figure 16.16 for
laterals 9 and 10, the laterals farthest from the pump. The total flow at the pump de-
creases to 16 L s-1. The main line sizes are reduced accordingly. The lateral pipe sizes
are the same as in Example 1. The average pressure, Pav, and variation Pvar, are for the
operating laterals only.
16.5.3 Operation and Maintenance Guidelines

The operating mode for a solid set sprinkler system depends upon the design and
use of the system, available labor, water supply, and available capital. The system can
be designed using the lateral or area (zone) design method. With lateral design indi-
vidual laterals are controlled by valves and each lateral may be operated as desired.
Normally, more than one lateral is operated simultaneously, but the operating laterals
usually are widely separated in the field. The lateral design method minimizes the
main or supply line pipe size, but it increases the number of valves required and also
Figure 16.16. Design of mainline and laterals for a solid set system with 2 laterals o
perating, where Pm is the pressure head at the main line outlet or lateral inlet, Pmin is the
minimum pressure and DP is the pressure difference on a lateral, Ns is the number of
sprinklers on a lateral (ON:OFF = 0,1), Ql is the total lateral flow, Qtot is the total flow in
the mailine section, and EL is the elevation of the inlet of a main line section (m).
the time to open and close valves when a manual valve system is used. With the area
(or zone) design method, a contiguous portion of the field is irrigated at one time.
Usually a submain is installed to supply water to that portion of the field.
For frost and freeze protection, the entire system may be operated at one time. De-
pending upon the crop being protected, the application rate will be 2 to 5 mm per hour.
In the eastern U.S., most orchard systems are designed to apply water over the crop. In
the western U.S., both undertree and overtree systems are used; however, with saline
water only undertree systems can be used successfully.
If the system is being used strictly for irrigation, only a portion of the system is
normally operated at one time. Where several hours are required for irrigation, control
may be manual or automatic. For an irrigation system on a shallow-rooted crop grown
on a coarse-textured soil, or in a container nursery operation where daily or frequent
irrigation is required, it is best to automatically control the sequencing of the system.
Where labor is very limited, automatic control may be desirable regardless of irriga-
tion frequency, however this will increase the initial investment. Conversely, limited
capital may require a totally manual system. A limited water supply, such as a well or
stream, may mean that only a portion of the system can be operated at once.
16.6 PERIODICALLY MOVED LATERALS
Sprinkler systems in this category have laterals that are moved between irrigation
settings. They remain stationary while irrigating. The lateral is drained prior to moving
to the next set. Set-move laterals are used extensively because of their relatively low
cost and adaptability to a wide range of crops, soils, topographies and field sizes.
Equipment cost is largely dependent on the number of sets irrigated by each lateral.
They are well suited to soils with high water holding capacity, deep-rooted crops, low-
growing crops, supplemental irrigation, and deficit irrigation management. They can
be classified as hand move or mechanical-move systems. Mechanical-move systems
are similar to hand move except that pipe types and sizes are often dictated by me-
chanical rather than hydraulic considerations.
The system consists of laterals, a pipeline with outlets for distributing irrigation wa-
ter to sprinkler devices that are periodically moved across the field. The lateral is made
of pipe sections that are from 50 to 150 mm in diameter and 6 to 18 m long (Figure
16.1). A coupler is installed on one end of the pipe section. The other end of the pipe
section is inserted into the upstream coupler and fastened with either a hook that
latches into the coupler or a ring assembly. New types of couplers are currently being
developed that use different mechanisms to connect the sections of pipe. Gaskets are
installed in the coupler to prevent leaks when the system is pressurized. Small pipes,
called risers, convey water from the lateral to the sprinkler. Water is supplied to the
laterals by main lines, or submains which branch from the main line.
The most common system has a single center main line with one or more laterals
which irrigate on both sides of the main line. If there are multiple laterals, they are
spaced equally, so by the time any lateral reaches the starting position of the lateral
ahead of it, the entire field has been irrigated once. The spacing of the sprinklers on
the laterals and between subsequent sets of each lateral is such that the water distribu-
tion patterns from the sprinklers give almost complete overlap. Large systems often
require more complex designs with multiple main lines, although simple systems are
possible on rectangular fields up to at least 64 ha.
Valve-tees are usually placed in the main line at the desired interval for spacing be-
tween lateral settings. The valve-tees are controlled by manipulating a valve-opening
elbow that makes the connection between the main line and laterals. Common valve
spacings are 12.2, 15.2, 18.3, and 24.4 m. Since common pipe lengths are 6.1, 9.1,
12.2, and 15.2 m, the desired valve spacing is obtained by using various combinations
of lengths. For many systems buried main lines are preferred. The pipe should be
placed safely below plow depth and should also be below frost depth, unless provi-
sions are made for draining the pipeline.
The advantages of periodically moved systems are that:
investment costs are minimal;
the systems offer a great deal of flexibility;
the systems are easy to understand and operate; and
the sprinklers and nozzles are generally the same size, which maximizes inter-
changeability.
The disadvantages of periodically moved systems are:
the high labor requirements;
the relatively large applications of water each irrigation;
the reduced uniformity when a uniform size of nozzle is used on long laterals or
rough terrain; and
the time required for laterals to drain before moving.
The water application and other characteristics of the hand move, towline and side
roll systems are very similar. Characteristics of the system are presented in Table 16.1.
16.6.1 Hydraulics
The hydraulic design of moved laterals has been described in Chapter 15 and pre-
viously in this chapter. With these systems the sprinkler and nozzle sizes are generally
constant and the pipe diameter for the lateral is uniform. Thus, the procedures devel-
oped in earlier sections apply. The general procedure is to layout the field boundaries,
water supply and pump location. The main line is frequently laid down the center of
the field to minimize pressure losses in long laterals. The spacing between sprinklers
and laterals is selected to fit the field. With an initial layout the discharge for the
sprinkler and lateral are determined.
The minimum average pressure is determined for the selected size of lateral pipe.
From these data the nozzle sizes can be determined. The diameter of coverage of the
sprinkler must be large enough to provide adequate overlap. If all components are
adequate the inlet pressure for each lateral position should be determined. The main
line should be selected to provide the highest feasible uniformity. The cost of design
alternatives should be determined. For the final design, product specifications and an
operational plan should be developed and discussed with the client. Special considera-
tions for each type of periodically moved system are discussed below.
16.6.2 Hand Move Laterals
The earliest periodically moved systems involved laterals that were moved by car-
rying sections of pipe across the field. Between moves the lateral operates for a period
time (the set time) and applies water to a portion of the field (a set) (Figure 16.17).
This is called a hand moved system. An extensive amount of labor is required to move
laterals from one set to the next set, which discourages frequent movement resulting in
large applications of water per irrigation.
Figure 16.17. Picture and operational sketch of a hand move irrigation system.
Common pipe diameters range from 51 to 152 mm, and pipe lengths are 6.1, 9.2, and
12.2 m. The most popular aluminum lateral length is 9.1 or 12.2 m. Shorter lengths
mean more walking during the move. Longer lengths are more difficult to transport
and do not provide proper spacing for the common sprinkler sizes.
Hand move systems are the lowest equipment cost alternative for moved-lateral
systems, but are labor intensive. Lack of available labor is the main reason farmers are
tending toward center pivot or other automated systems. Most hand move sprinkler
systems now use aluminum laterals, although plastic is available. Most lateral pipe
couplers contain a chevron-type rubber gasket which seals under pressure, and are
designed to latch automatically when the pipes are pushed together. Many couplers
contain an optional adjustment for easier unlatching. Hook and latch type couplers can
be fixed so they will unlatch automatically when the irrigator pushes and twists the
pipe. Ball and socket couplers automatically latch when the pipe is under pressure, and
release when the pressure is off. Drop-lock couplers have hooks that engage as the
pipe is lowered to the ground. The automatic unlatch saves some walking, but could
increase the hazard of an accidental unlatching. In contrast to the main line gaskets,
the lateral gaskets are designed to release their tight contact with the pipe when the
pressure on the water is reduced. This permits the water to drain from the pipe when
the pressure has been turned off so the pipe can be moved easily to the next setting.
Each coupler is threaded to receive a sprinkler riser pipe, usually 25 mm in diame-
ter. If both the coupler and the riser are aluminum, it is customary to connect them
with a zinc alloy fitting or Teflon tape to avoid thread seizure. The riser should extend
at least to the top of the crop canopy, but the uniformity of water distribution is im-
proved if it is extended another 0.5 m.
The uniformity of water distribution can usually be improved by using an offset
pipe with a 90° elbow every second irrigation. The length of the offset should be half
of the spacing between lateral settings. Using the offset pipe permits the lateral to be
placed midway between the positions used during the previous irrigation. Thus, con-
sidering two irrigations added together, a 12.2-m by 18.3-m spacing, for example, is
effectively reduced to 12.2 m by 9.2 m.
A good procedure for the irrigator to follow when moving the lateral from one set-
ting to the next is to start by moving the valve opening elbow and the section of pipe
connected to it. As soon as these pieces are in place at the new location, the valve is
slightly opened so a very small stream of water runs out the end of the first pipe sec-
tion. As each subsequent section of pipe is put into place, the small stream of water
runs through it, flushing out any soil or debris that may have been picked up during
the move. The last section of pipe with its end plug in place can be connected before
the stream of water reaches the end and builds up pressure. Then the irrigator walks
back along the lateral, correcting any plugged sprinklers, leaky gaskets, or tilted risers.
After returning to the main line, the valve is opened further until the desired pressure
is obtained. A quick check with a pitot gauge on the first sprinkler confirms the valve
adjustment. To save time on each lateral move, there is a tendency to completely open
the valve and fill the line as quickly as possible. This causes water hammer at the far
end of the line, so a surge plug at that end may be needed. The sprinklers commonly
used on hand move systems may have either one or two nozzles. Typically, individual
sprinkler capacities range from about 0.06 to 0.63 L s-1. Operating pressures range
from 240 to 415 kPa.
Certain crops, such as orchards, require specially designed sprinklers. When the
sprinklers are used over the tops of the trees, conventional models can be used. How-
ever, when they are used under trees, sprinklers with low water trajectory must be
used. Lowering the trajectory reduces the uniformity, unless the spacing is reduced.
Hedge-rowed trees present a difficult problem, especially if it is desirable to irrigate
through the skirts of the rows.
16.6.3 Towline Systems
Towline or skid-tow systems were developed to reduce labor required to reposition
laterals. The most efficient layout for towlines is to divide the field in half so the lat-
eral can be pulled in a zigzag fashion across the field (Figure 16.18). Towline sprin-
kler laterals have relatively rigid couplers fitted with skids or wheels so the line can be
moved by pulling it from the end. The skids consist of a flat metal plate held on the
Figure 16.18. Picture and operational sketch of a towline or skidtow system.
underside of the pipe by one or more clamps. In one type, the skid is placed under the
coupler and clamped at both ends. This makes the skid take the major part of the end
thrust at the coupler when the pipe is towed. If relatively long sections of pipe are
used, a second skid may be needed under the center of each section to reduce abrasion
from soil contact. Stabilizers, outriggers, or wheel-mounts were used to keep the lat-
eral from tipping. Two or three outriggers along the line are needed to keep the pipe
oriented with the skids on the bottom and the sprinklers upright.
For wheel-type units, a pair of wheels mounted on a simple U-frame is clamped to
each section of pipe. The wheels are oriented so the entire length of lateral pipe can be
pulled endwise. The pipe itself stands only 0.3 to 0.5 m above the ground. The flexi-
bility of the pipe and the articulation of the couplers permit the lateral to curve slightly
while being moved to a new position. In one type, however, the lateral stays straight.
The wheels are fixed so they shift to a 45° angle from the lateral when pulled in one
direction, and then shift back to a 45° position the other way when pulled from the
other end of the lateral. Thus, by pulling alternately from both ends, the entire length
of lateral is shifted the desired distance to the next set position.A coupler and hitch are
attached to each end of the lateral so that it can be towed in either direction. An end
cap is used to plug the downstream end of the pipeline. Drain plugs are installed along
the lateral to empty water from the pipeline before moving. Often a flexible hose is
used to connect the lateral to the main line. If more than one lateral is used in a field,
provisions need to be made to split the main line while moving the lateral. A telescop-
ing coupler can be used for this purpose. When the lateral reaches the edge of the
field, the lateral must be disassembled and transported to the starting location. If the
surrounding property is amenable, the lateral can be towed to the starting position.
The traditional way of moving a towline lateral is to snake it past the main line in
an S-shaped curve to a new position on the other side (Figure 16.18). For the next set-
ting, the lateral is pulled in the other direction past the main line in an opposite S-
shaped curve. With this procedure, each move needs to advance the lateral only half
the distance between adjacent sets.
Towline systems are the least expensive of the mechanically moved systems. How-
ever, they are not used extensively because the moving process is tedious, requires
careful operation, and damages many crops. Towline systems have been used success-
fully in some forage crops and in row crops. The moves are made easier if the main
line is buried.
16.6.4 Side Roll Systems
The side roll, or wheel-move, system is a third type of periodically moved lateral.
With this system, wheels are mounted on the sprinkler lateral to carry the pipeline
above the crop (Figure 16.19). A cart provides power to rotate the wheels. The cart
and the water feed to the side roll can be positioned anywhere along the lateral. Fre-
quently the pipeline is used as the axle for the torque; however, a separate drive shaft
can be used to rotate the pipe. Multiple laterals are often used in a field. A special ap-
paratus with a swivel and weight is used to keep sprinklers vertical when the pipeline
rotation is not precise. The mainline may be above or below ground. Flexible hoses
are often used to connect the side roll lateral to the mainline.
Rigid couplers permit the entire lateral to be rolled forward by applying torque at
the center while the pipe remains in a nearly straight line. Aluminum pipe having a
100 or 125 mm diameter is commonly used. To have sufficient strength, the aluminum
pipe wall thickness should be at least 1.8 mm. A motorized drive unit, usually near the
center of the lateral, provides torque to move the lateral and holds the pipe in place
during operation. Normally, the drive unit contains a gasoline engine and a transmis-
sion with a reverse gear. Electric motors or hydraulic motors are also used. Typical
lateral length is 400 m, but longer lengths are made with two drive units spaced about
½ the lateral length apart, and connected by a drive shaft. Since the greatest torque is
applied to the pipe near the drive unit, 125-mm pipe is sometimes used near the center
of the lateral for greater strength. The pipe is flexible enough so that these systems can
be used on rolling topography with mild slopes.
The most popular sprinkler spacing (and pipe length) is 12.2 m. The wheels are
usually placed in the center of each length of pipe, with sprinklers located halfway
between the wheels. Thus, a standard 400-m lateral contains 32 pipe lengths and 36
wheels because four wheels are required for the drive unit. Sometimes an extra wheel
is provided for the last pipe section at each end.
Figure 16.19. Picture and operational sketch of a side roll irrigation system.
The rigid couplers can be quickly disconnected to shorten the lateral in the case of
odd shaped fields. Often the sprinklers are provided with self-levelers so they will
right themselves if the lateral is not stopped where the riser would be exactly upright,
i.e. on variable topography, and to aid in aligning the laterals with main line valves.
Labor requirement is approximately five minutes per lateral per move. Some side roll
drive units can be controlled from the end of the lateral, eliminating the need to walk
to the center drive unit. At least one manufacturer has developed an automated side
roll system that can be programmed to drain and move itself and irrigate up to five
sets. Water is supplied through a flexible hose.
The wheel diameter must be large enough so the pipe will pass over the crop with-
out damaging it, and the crop will not prevent the lateral from being rolled to the next
position. Common wheel diameters commonly are 1.17, 1.47, 1.63, and 1.93 m.
A spring-loaded drain valve also is located about midway between the wheels, near
the pipe coupler and near the sprinkler. This valve opens automatically when the pres-
sure is off, so the pipe will drain quickly and permit moving the lateral forward to the
next set without much time loss. (Attempting to roll the pipeline when it is full of wa-
ter will damage the equipment).
The most popular side roll spacing along the main line is 18.3 m. Two popular op-
erating schemes are used. In one, the lateral is connected to every outlet valve along
the main line, and when the lateral reaches its destination and completes its last set, it
is rolled back to the starting point. In the other, the lateral is connected to every even-
numbered outlet valve on the main line while the lateral is moved across the field, and
then connected to the odd- numbered valves while the lateral is moved back to the
starting position. For the latter case the lateral interval between irrigations is longer at
the two ends of the field than in the center.
As with hand move laterals, there is a tendency to completely open the hydrant
valve and fill the line as quickly as possible, causing a water hammer at the far end of
the line. Therefore, a surge plug at the closed end is recommended. The use of offsets,
especially for the 12.2 m by 18.3 m spacing, is also recommended. Side roll laterals
are highly susceptible to wind damage when they are empty, and should be staked
down during the off season. Special braces, which allow the lateral to roll in one direc-
tion only, help protect the laterals during the irrigation season.
A lateral with 32 sprinklers is commonly designed with 100-mm diameter pipe,
even when the water is introduced into it from one end and the friction loss is 55 to 60
kPa. However, if the water is admitted to it at the center of the lateral, the friction loss
is reduced to about 1/5 as much. A 125-mm pipe would have only about 1/3 the fric-
tion loss of the 100-mm pipe when the water is admitted from one end. The best
method will depend on the future price and availability of energy. End-feed laterals
are usually preferred because a drive-through roadway can be maintained along the
main line for easy accessibility to the valves.
16.6.5 Side Move with Trail Lines
Side move laterals with trail lines are supported on wheel-mounted A-frames. The
pipe does not serve as the axle of the wheels and can be higher above the ground. Each
A-frame carriage is driven from a drive shaft that extends the length of the pipeline.
The drive shaft can be turned from the center of the line or from one end. One version
uses a continuous-move (center pivot) type lateral operating in stationary set-move
mode. The wheels are powered by electric or hydraulic motors supplied from an on-
board generator or hydraulic pump.
The small diameter trail lines can each carry several sprinklers. Usually, short
sprinkler risers are used as they are easier to keep upright than tall risers. Outriggers at
the last sprinkler on each trail line are used to keep the risers upright; however, these
may damage some crops.
This system is sometimes called a “movable solid set.” It greatly reduces the num-
ber of moves necessary to cover the field, thus saving labor. Sprinkler spacings and
nozzle sizes that give low application rates at an acceptable uniformity can be used.
With the low rates, 24-h set times may be practical for some soils and crops, thus per-
mitting a normal daytime work schedule for the irrigator. When a trail line system
reaches the end of the field, the trail lines are uncoupled, and the lateral is moved to
the opposite ends of the trail lines, which are then recoupled to irrigate on the way
back across the field. The wheels on most side move systems can be turned 90°, per-
mitting the lateral to be pulled endwise to another field.
16.7 CENTER PIVOTS

In 1948 Frank Zybach invented the self-propelled irrigation system. His invention
used towers with wheels to carry a pipeline around a pivot point in the field. Even
though his invention has undergone numerous changes, the basic concept is still used.
A span of pipe is supported by a tower and a truss system (Figure 16.20). Today most
systems are propelled by electrically or hydraulically powered motors mounted on
each tower. A system of switches on each tower energizes the motor when the tower
needs to move. The depth of water application is generally controlled by selecting the
speed of the last or end tower. For many electric systems a one-minute timer is used to
control the velocity. If the timer is set to 100%, the motor on the end tower is ener-
gized the entire minute causing the end tower to move at a constant velocity equal to
the maximum speed of the system. If the timer is set to 50% the motor is only ener-
gized for 30 seconds; thus, the end tower is stationary for 30 seconds and then moves
for 30 seconds at the maximum speed for the end tower. Some hydraulic and electric
systems provide continuous movement of the end tower at variable speeds to apply the
desired depth of application.
The interior towers are controlled by switches or valves mounted on the tower. One
switch or valve is set to energize the motor if the downstream tower has moved far
enough to exceed a start angle. The switch or valve energizes the motor and causes the
tower to move at a constant velocity. The tower moves until the angle between adja-
cent spans exceeds a stop angle. For continuously moving systems the interior towers
move continuously at varying speeds to maintain alignment. The interior towers may
be designed to move at a faster velocity than the end tower allowing them to catch up
with the end tower to maintain alignment.
Controllers are presently available to change the speed and/or direction of rotation
of the pivot lateral as the system circles the field. This is useful if different crops are
planted under a pivot or if obstructions are located in the field. Small center pivots are
also made that can be moved within a field, or from field to field. This allows for
semi-automated irrigation of irregularly shaped fields and small tracts of land.
Center pivots can be equipped with an end gun to increase the portion of a field that
is irrigated (Figure 16.20). The end gun is a large sprinkler similar to that used on a
traveler that is mounted on the end of the pivot lateral. The gun throws water a long
distance thereby increasing the amount of land irrigated for a given length of lateral. A
valve is attached to the end gun so that the end gun only operates in the corners of the
field. When the pivot lateral reaches a preset angle of rotation, the valve opens and
water is supplied to the end gun. In some cases a booster pump is attached to the valve
to increase the pressure for the end gun.
A corner system can be used to irrigate an even larger portion of a square field. A
special span is attached to the end of a typical system (Figure 16.21). The corner span
revolves about the end of the pivot lateral. The corner span is tucked behind the main
lateral when the boundary of the field is close to the end of the pivot lateral. The
sprinklers on the corner span begin to irrigate when the pivot rotates to an angle where
the sprinkler package on the primary lateral does not reach to the field boundary.
Sprinklers on the corner span are attached to a series of valves. As the pivot lateral
rotates toward the corner of the field, the corner lateral extends and opens valves as
the corner lateral is extended. An end gun can be attached to the corner lateral to throw
Figure 16.20. Components and field layout for typical center pivot irrigation systems.
water further into the corner. Many corner systems use an underground cable and an
antenna on the corner tower to follow a desired path. The cable can be positioned to
irrigate irregular shapes and to move around permanent obstructions.
Center pivots have many advantages including:
Automated operation—Center pivots can be operated with minimal labor often
making several revolutions without stopping. They can also be controlled re-
motely from farm vehicles or computers.
Ability to apply small irrigation depths—Since the systems are automated, they
can apply small irrigations to match crop needs without leaching nutrients.
Very high uniformity—Since the lateral moves slowly and because there is a
great deal of overlap of water application from successive sprinklers on the lat-
eral, center pivots apply water very uniformly.
Chemigation—Pivots can be managed to quickly and uniformly irrigate a field
with small amounts of water, which provides the opportunity to apply fertilizer,
herbicides, and insecticides.
PRIMARY
PRIMARY LATERAL
LATERAL
CORNER
CORNER SPAN
SPAN
CORNER
CORNER
TOWER
TOWER
FIELD BOUNDARY
+ UN
M G
STE ND
E
SY D
SIC AN
M
BA R
S A
IU ER
D N
RA OR
C
RADIUS OF BASIC SYSTEM
Figure 16.21. Picture and operational sketch for a center pivot

equipped with a corner watering system.
Little annual setup is required—Once the system is constructed, the pivot can be
operated anytime that water is needed. This is advantageous for germination of
crops or for preparing seedbeds.
Some disadvantages of center pivot systems are its:
Cost—Depending on the reference, the cost of a center pivot system is moder-
ately high. The cost per unit of land for a typical center pivot is approximately
20 to 30% of that for solid set and microirrigation systems. However, the cost
exceeds that for systems requiring more labor such as hand move or towline sys-
tems. The cost per unit area decreases as the length of the lateral increases; how-
ever, several factors limit the maximum length of the lateral.
High application rates—The rate of water application at the outer end of the lat-
eral is quite high, which can cause runoff.
Circular pattern—Center pivots irrigate about 80% of a square field. Lost pro-
duction in the corners may be a consideration when land is expensive or high-
value crops are produced.
The self-propelled irrigation system as named by inventor Frank Zybach in 1948
has transformed irrigation. Many lands are now efficiently irrigated that were once
labeled as unsuitable for irrigated agriculture. Center pivot systems as developed by
the irrigation industry have reduced labor requirements, improved water application
efficiency and contributed to the economic viability of many regions. Nevertheless
there have been failures with center pivots. Problems have arisen due to poor design
and improper management. In some locations excessive development has led to the
overdraft of water resources causing declines in ground water supplies. Without
proper management leaching of agricultural chemicals can occur especially under
shallow rooted crops grown on sandy soils. Other chapters discuss the planning and
operation of irrigation systems to fit within environmental, economical and resource
constraints. The focus in this section is on the design of center pivot systems to effi-
ciently and uniformly apply irrigation water.
16.7.1 Sprinkler Discharge
The discharge from each sprinkler on a center pivot must be determined to apply
water uniformly. The area midway between the adjacent upstream and downstream
sprinklers defines the representative area for a sprinkler on a pivot (Figure 16.22).
FIELD
BOUNDARY
RS
R
SL
SPRINKLER SPACING SL SL
RADIAL DISTANCE PIVOT LATERAL

TO SPRINKLER (R)
SL
R – SL / 2 R + SL / 2
Figure 16.22. Diagram of representative area for determining

the discharge for sprinklers on a center pivot lateral.
This area is area is given by

AR = 2π R SL (16.19)
where AR = representative area for the sprinkler at distance R from the pivot point
SL = local spacing between sprinklers on the pivot lateral, m.
It is common for the spacing between sprinklers to vary in intervals along the pivot
lateral. Often the spacing near the center of the pivot is double the spacing near the
outer end of the pivot. The average discharge per unit area for the pivot is the ratio of
the flow into the pivot system divided by the circular area irrigated by the primary
system (i.e. where the end gun or corner system is not operating). Combining these
relationships provides the discharge required for the sprinkler located at radial distance
R from the pivot point:
2QS RS L
qR = (16.20)
Rs2
where qR is the discharge required for a sprinkler located a distance R from the pivot
point and QS is the total discharge into the center pivot system. The expression can be
revised to include the gross system capacity as
qR = 2 × 10-4π Cg R SL (16.21)
To design the sprinkler package the nozzle sizes for each sprinkler must be deter-
mined. This requires information on the distribution of pressure along the lateral. For
level land the relationship developed by Chu and Moe (1972) can be used for the dis-
tribution:
⎡ ⎧ 3 5 ⎫⎤
15 ⎪ R 2 ⎛ R ⎞ 1 ⎛ R ⎞ ⎪⎥
PR = PS + PL ⎢1 − ⎨ − ⎜⎜ ⎟⎟ + ⎜⎜ ⎟⎟ ⎬ (16.22)
⎢ 8 ⎪ Rs 3 ⎝ Rs ⎠ 5 ⎝ R s ⎠ ⎪⎥
⎣ ⎩ ⎭⎦
where PR = the pressure in the lateral at point R
PS = pressure at the distal end of the pivot lateral
PL = pressure loss from the inlet into to the distal end of the pivot lateral.
Keller and Bleisner (1990) and Scaloppi and Allen (1993) presented methods to
compute the pressure distribution along pivot laterals. Their results produce the same
pressure distribution along the lateral as that from Chu and Moe (1972). The relative
distribution of pressure loss along the lateral is shown in Figure 16.23.
The difference between the methods by Chu and Moe (1972) and Scaloppi and Al-
len (1993) is in the procedures used to compute the friction loss from the inlet to the
distal end of the lateral (i.e., PL). Chu and Moe used the Hazen-Williams equation.
Scaloppi and Allen used the Darcy-Weisbach equation. Scaloppi and Allen considered
the variation in the velocity head in the lateral as well while Chu and Moe considered
it to be negligible. In Chapter 15 the Darcy-Weisbach method was recommended for
hydraulic calculations. We concur with that recommendation for applications using
computer spreadsheets and/or programs to compute friction loss. Applying the Darcy-
Weisbach method for sprinkler laterals results in changing friction factors along the
pipeline. This is difficult to represent in this chapter. Thus, the hydraulics for center
pivot pipelines is illustrated using the Hazen-Williams equation for this chapter.
1.0
FRACTION OF PRESSURE LOSS ALONG LATERAL

0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
FRACTION OF DISTANCE ALONG LATERAL

Figure 16.23. Distribution of pressure loss along a center pivot lateral.
The pressure loss along pivot laterals is traditionally computed by the industry us-
ing the Hazen-Williams equation (see Chapter 15). Typical C values for center pivot
materials are provided in Table 16.10. Since the discharge from sprinklers varies along
the lateral for a center pivot the F value for pivot laterals is different than for other
sprinkler laterals. Results from Chu and Moe (1972) show that the F value for pivots
without end guns is 0.54. Pair et al. (1983) show that the F value is 0.56 when the end
gun is operating. These values have been compared with a procedure that computes
the pressure loss between each sprinkler along the pivot lateral as used by Kincaid and
Heermann (1970). Comparisons show that the method by Chu and Moe (1972) is suf-
ficiently accurate.
Based on these procedures the pressure distribution along pivot laterals can be
computed using Table 16.10. These results provide the friction loss along laterals that
only have one diameter of pipe. Figure 16.23 shows that over half of the total loss
along the lateral occurs within the first third of the lateral and that 80% of the pressure
loss in the lateral occurs in the first half of the pipeline. Experience has shown that
operating pressures and therefore operating costs can be reduced by using larger di-
ameter pipe for the first portion of the lateral.
Computing the friction loss for laterals that contain two diameters of pipe can be
computed for R ≥ Rc as
⎡ ⎧⎪ R 2 ⎛ R ⎞3 1 ⎛ R ⎞5 ⎫⎪ ⎤
PR = PS + PLs ⎢1 − 1.875 ⎨ − + ⎬⎥ (16.23)
⎢ Rs 3 ⎜⎝ Rs ⎟⎠ 5 ⎜⎝ Rs ⎟⎠
⎣ ⎩⎪ ⎭⎪ ⎥⎦
and for R ≤ Rc: ⎡ ⎧⎪ Rc 2 ⎛ Rc ⎞3 1 ⎛ Rc ⎞5 ⎫⎪ ⎤
PC = PR + PLs ⎢1 − 1.875 ⎨ − + ⎬⎥
⎢ Rs 3 ⎜⎝ Rs ⎟⎠ 5 ⎜⎝ Rs ⎟⎠
⎣ ⎩⎪ ⎭⎪ ⎥⎦
⎡ ⎧⎪ R ⎛ Rc3 R 3 ⎞ 1 ⎛ Rc5 R5 ⎞ ⎫⎪ ⎤
PR = PC + PLL ⎢ 1.875 C − R − 2 ⎥
⎨ ⎜ 3− 3⎟ + ⎜ 5 − 5 ⎟⎬
⎢ ⎜ ⎟ 5 ⎝ Rs Rs ⎟⎠ ⎪
⎜ ⎥
⎣ ⎪⎩ RS RS 3 ⎝ Rs Rs ⎠
⎭ ⎦
Table 16.10. Friction loss (m/100 m) in galvanized steel pipe for center pivot and lateral
move systems. Results assume a C value of 140 for the Hazen-Williams equation and
a single pipe size. To compute losses for a pivot multiply values times the lateral length
and the appropriate F factor.
System Outside Diameter of Pipe (mm)
Discharge
(L s-1) 102 114 127 141 152 168 178 203 219 254
5 0.57
10 2.06 1.12
15 4.36 2.38 1.39
20 7.42 4.06 2.37 1.38
25 11.2 6.13 3.58 2.09 1.42
30 15.7 8.60 5.02 2.92 1.99 1.21
35 20.9 11.4 6.68 3.89 2.65 1.61 1.22
40 26.8 14.6 8.56 4.98 3.40 2.06 1.56
45 33.3 18.2 10.6 6.19 4.22 2.56 1.94 1.00
50 22.1 12.9 7.53 5.13 3.12 2.36 1.21
60 31.0 18.1 10.6 7.20 4.37 3.31 1.70 1.16
70 24.1 14.0 9.57 5.81 4.41 2.26 1.55
80 30.9 18.0 12.3 7.44 5.64 2.89 1.98
90 22.4 15.3 9.25 7.02 3.59 2.47 1.18
100 27.2 18.5 11.3 8.53 4.37 3.00 1.43
120 38.1 26.0 15.8 12.0 6.12 4.20 2.01
140 34.6 21.0 15.9 8.14 5.59 2.68
160 26.9 20.4 10.4 7.16 3.43
180 33.4 25.3 13.0 8.91 4.26
200 30.8 15.8 10.8 5.18
220 36.7 18.8 12.9 6.18
240 22.1 15.2 7.26
260 25.6 17.6 8.42
F factor for pivots: Without end gun: 0.54 With end gun: 0.56
Adjustment for pipes with C value 100 110 120 130 140 150
different roughness Multiplier 1.86 1.56 1.33 1.15 1.00 0.88
where RC = location along the lateral where the pipe diameter changes
PLS = pressure loss from the inlet to the distal end for the small diameter pipe
PLL = pressure loss from the inlet to the distal end for the large diameter pipe.
An example of the use of this procedure is included in the example shown in Figure
16.24. The example illustrates that the investment in larger pipe for the first 40% of
the lateral would reduce the inlet pressure by approximately 20% while using larger
pipe along the entire lateral would only reduce the pressure by 29%. A cost analysis is
required to determine which alternative is optimal.
The above computations for pressure distribution along the lateral neglected the ef-
fects of elevation changes. If there is a uniform slope along the lateral the pressure can
be adjusted appropriately and the proper nozzle sizes can be selected to provide the
desired discharge. However, the performance of the pivot across a sloping field varies
once a specific set of nozzles is installed. Often design along a flat slope is a reason-
able compromise for sloping fields. For many fields irrigated with pivots the terrain is
450
400 168 mm DIAMETER PIPE
MIXED PIPE SIZES
PRESSURE IN LATERAL, kPa

350
300
250
200
203 mm DIAMETER PIPE
150
100
80 L/s FLOW RATE

50
0
0 50 100 150 200 250 300 350 400 450 500
RADIAL DISTANCE FROM PIVOT, m

Figure 16.24. Distribution of pressure along center pivot laterals with external diameters of
168 and 203 mm, and for a mixed lateral with 200 m of 203-mm and 300 m of 168-mm pipe.
neither uniform nor flat. Since the design of sprinkler packages is usually accom-
plished with computer programs that start at the end of the lateral and determine pres-
sure loss and nozzles sizes for each outlet along the lateral, the elevation at each loca-
tion can be used in the calculation. Pressure regulators are commonly used if the ele-
vation differences in the field lead to pressure variations that would reduce the uni-
formity of application below an acceptable level.
The procedures from Chu and Moe (1972) are useful in understanding the hydrau-
lics of sprinkler laterals and for simple analysis of pivots. However, center pivot sys-
tems are typically designed using computer programs that start at the distal end of the
pivot and sequentially compute the pressure available at each upstream sprinkler outlet
(Heermann and Stahl, 2006). In these programs the friction loss in the portion of the
lateral between outlets is computed using either the Darcy-Weisbach or Hazen-
Williams equations for a pipe without outlets. The pressure losses through fittings
used to connect the sprinkler to the lateral are also computed. This has become more
important as the operating pressure for sprinkler devices continues to decrease to save
energy and as the sprinkler device is installed further from the pivot later using drops.
Using the sprinkler discharge equation for the required sprinkler discharge gives the
required flow at each outlet. With this information and the available pressure at the
base of the sprinkler device, the most appropriate nozzle size can be determined. Since
a finite number of nozzle sizes are available, it is not possible to exactly match the
required discharge. Many designers maintain the cumulative error between the actual
total flow in the system and the required flow. At each outlet the nozzle is selected that
closely matches the required discharge for the outlet and that minimizes the cumula-
tive flow error. These types of programs provide very detailed specification of system
characteristics and allow for consideration of the terrain of the land.
16.7.2 Uniformity of Application

As with other types of sprinkler systems the uniformity of application beneath cen-
ter pivots can be computed by considering the overlapped depth of application from
individual sprinklers. This analysis provides guidance for the most appropriate spacing
of sprinklers devices along the lateral. The procedure is more complex for pivots since
individual sprinklers are not stationary but travel in circular paths about the field. Bit-
tinger and Longenbaugh (1962) developed the framework for analyzing the distribu-
tion of water from moving sprinkler systems. They considered the case of sprinklers
moving in a straight line and in a radial pattern. They showed that sprinklers that
travel on an arc produce a skewed application pattern that is somewhat difficult to
analyze. However, the effect of the skewed pattern is negligible when the distance
from the pivot point to the sprinkler is more than five times the wetted radius of the
sprinkler. For center pivots the inaccuracy of uniformity calculation incurred from
assuming linear travel of sprinklers will be small, thus only the solutions for linear
travel are presented here.
The framework for the analysis by Bittinger and Longenbaugh, which considered
triangular and elliptical water application patterns for an individual sprinkler, is illus-
trated in Figure 16.25. The precipitation rates for the triangular and elliptical applica-
tion rates are given by
Ppt
P= (Wr − s ) for triangular patterns with 0 ≤ s ≤ Wr
Wr
Ppe
and P= Wr2 − s 2 for elliptical patterns with 0 ≤ s ≤ Wr (16.24)
Wr
where P = precipitation rate at distance s from the sprinkler, mm h-1
Pp = peak precipitation rate at the sprinkler location, mm h-1
Wr = wetted radius of the sprinkler, m
s = distance from the observation point to the sprinkler, m
t, e = subscripts denoting triangular and elliptical patterns.
The peak application rate for the triangular and elliptical patterns can be computed
from the discharge of the sprinkler located at radial distance R and the wetted radius of
that sprinkler:
3q R 3q R
Ppt = for triangular patterns and Ppe = for elliptical patterns (16.25)
πWr2 2πWr2
The solutions for the depth of water applied at a point from an individual sprinkler
as the pivot lateral moves over the point were developed by Bittinger and Lon-
genbaugh (1962). Heermann and Hein (1968) used this procedure to compute the
depth of water along a radial line from the pivot point to the end of the center pivot.
The depth applied at a point is the summation of water applied by all sprinklers that
apply water to the point and is given by
1 N Wri Ppti ⎡⎢ 2
⎛ 2
2 ⎜ 1 + 1 − ui
⎞⎤
⎟⎥
for triangular patterns: d = ∑ 1 − u i − u ln
i ⎜ ⎟⎥
ω i =1 Ri ⎢ ⎜ ui ⎟
⎢⎣ ⎝ ⎠⎥⎦
TRIANGULAR
APPLICATION RATE
ELLIPTICAL
AVERAGE
Wr
DISTANCE FROM SPRINKLER (s)
OBSERVATION
PATH OF POINT
SPRINKLER
(R)
ER
NKL
SPRI
O
ET
ANC
DIST Wr
IAL s
RAD x
u Wr
PIVOT SPRINKLER
LATERAL
PIVOT
POINT
AREA WETTED
BY A SPECIFIC
SPRINKLER
Figure 16.25. Illustration of parameters used to compute uniformity of

water application for center pivot systems.
and for elliptical patterns: d =

π N Wri Ppei 1 − ui
∑
( 2
) (16.26)
2ω i =1 Ri
where d = depth of water applied at a point

u = ratio of the distance from the sprinkler to the point relative to the wetted
radius of the sprinkler,
ω = angular velocity of the pivot lateral,

Ri = radial distance from the pivot point to sprinkler i,
N = number of sprinklers that apply water to the point of interest.
The angular velocity can be determined from the system capacity and the average
depth of water applied (da):
2πCg
ω= (16.27)
da
In using these equations a series of points along a radial line is selected that repre-
sents the density of observations desired for computing the uniformity. At each point
Equation 16.26 is used to compute the aggregate depth of application. Heermann and
Hein (1968) used this technique to compute the coefficient of uniformity for pivots as
given by
⎡ ⎛ ∑ d j X j ⎞⎟ ⎤⎥
⎢ ⎜
j
⎢ ∑⎜ X j d j − ⎟⎥
⎢ j ⎜
⎜
∑ X j ⎟⎥
⎟⎥
⎢ j
UC p = 100 ⎢1 − ⎝ ⎠
⎥ (16.28)
⎢ ∑ j j
d X ⎥
⎢ j ⎥
⎢ ⎥
⎢ ⎥
⎣ ⎦
where UCp = the uniformity coefficient for center pivots
Xj = distance from the pivot point to the point that the depth is computed
dj = depth of water applied at point j.
Field results and simulation models have shown that the uniformity of application
is very high for well-designed and operated center pivots. It is common to find uni-
formities greater than 90 for center pivots. This is higher than most other types of
sprinkler systems or alternate irrigation methods.
The previous analysis assumed that the angular velocity was constant. Of course
center pivot lateral do not rotate at a constant angular velocity. Some pivots are de-
signed so that the towers move at a constant velocity for a period of time and then the
tower is stationary until a signal is received for the tower to move again. Other pivots
are designed to move continuously but not at a constant velocity. If a pivot tower is
stationary at one point the area watered during that time receives a larger application
than areas irrigated when the lateral is moving. If sprinkler devices are used that have
a small wetted radius, the start-stop motion of the later can lead to nonuniformity, as
illustrated by Hanson and Wallender (1985). With this program they were able to
simulate the performance of the sprinkler package for any terrain and for the adjust-
ments of the pivot system that control the start-stop nature of the pivot. This procedure
allows for efficient evaluation of many design alternatives.
The development presented is based on triangular or elliptical application patterns
for a single sprinkler. Devices that have been introduced recently have a different ap-
plication pattern that those analyzed by Bittenger and Longenbaugh (1962). The pro-
cedure presented here should be combined with the single-leg distribution for the
packages that are of concern to assess the performance of sprinkler packages.
At this point the nozzle size for each sprinkler along the lateral can be computed for
alternative sprinkler packages. The operating pressure at the pivot point can be esti-
mated for the field terrain and each sprinkler package. Ultimately, the uniformity of
application can be computed (Heermann and Stahl, 2006). However, two additional
features, runoff and evaporation losses, have a bearing on the suitability of a design.
16.7.3 Runoff Avoidance
For center pivots to operate efficiently the applied water must be available for crop
use. If the water runs off sloping lands or evaporates while in the air, the design effi-
ciency will not be achieved. Keller and Bleisner (1990) provide a simple diagram of
the suitability of center pivots based on the soil texture (Figure 16.26). While this
helps with the general suitability, it is not adequate for design of the center pivot.
The water application rate beneath the outer spans of a center pivot is very high.
The application rate may exceed the ability of the soil to infiltrate water. Some of the
water applied at rates exceeding the infiltration rate can be stored temporarily on the
soil surface. This is called surface or retention storage. Once the local surface storage
is filled, the excess application begins to flow across the field. Some infiltration occurs
as the water flows across the field. Ultimately, the runoff water either accumulates in
low areas in the field leaves the field. In either case the distribution of water that infil-
trates is different than the distribution of water applied and some water is lost as runoff
or deep percolation. Thus, both uniformity and efficiency are reduced when runoff
becomes significant.
Kincaid et al. (1969) investigated the potential for runoff from center pivot irriga-
tion systems. They developed a procedure to adjust the infiltration rate measured with
systems that pond water on the soil surface for conditions that occur under center piv-
ots. Dillion et al. (1972) were the first to develop design procedures that considered
the infiltration characteristics of the soil. A combination of these procedures was used
by Gilley (1984) to develop guidelines for the selection of sprinkler packages based on
E
NC
D
ITE
MA
LIM
10
OR
90
RF
L
NA
20
PE
80
RGI
MA
T
30
IVO
70
LA
RP
40
TC
60
EN
TE
PE
CLAY
D
50
O
RC
EN
GO
RC
50 SILTY
PE
DC
EN
CLAY
60
40
TS
SANDY
TE
CLAY SILTY
ILT
CLAY LOAM CLAY 70

IPA
30 LOAM
T
EN
80
TIC
SANDY CLAY LOAM

LL
20
CE
AN
LOAM
EX
90
LO SANDY LOAM SILT LOAM
10 AM
Y
SA SILT
ND
SAND
90 80 70 60 50 40 30 20 10
PERCENT SAND
Figure 16.26. Anticipated performance of center pivot systems based on soil texture
(adapted from Keller and Bleisner, 1990).
70
INFILTRATION RATE
60 PEAK APPLICATION RATE
50
RATES, mm / hour
APPLICATION
40 SURFACE RATE
STORAGE POTENTIAL
RUNOFF
30
20
10
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
TIME, hours
Figure 16.27. Illustration of the runoff potential when the application rate of the pivot
exceeds the infiltration rate of the soil.
the potential for runoff. With this method an elliptical pattern is assumed for the sprin-
kler package. The combined distribution of water from the overlap of individual sprin-
klers provides an elliptical application rate perpendicular to the lateral (Figure 16.27).
This application rate is given by
Pp
P (t ) = 2 t t p − t2 (16.29)
tp
where P(t) = rate of water application as a function of the time (t) that water has been
applied
Pp = peak application rate for the elliptical pattern of the sprinkler package
tp = time after initial wetting that the peak application rate is reached.
With this distribution for the sprinkler package the peak application rate can be
computed as
⎛ Q ⎞ R
Pp = 4 ⎜ S ⎟
⎜ π R 2 ⎟ Wr
⎝ S ⎠ (16.30)
This demonstrates that the peak application rate for the sprinkler package depends
entirely on the design of the system. Once a flow rate is established for the irrigation
system and the sprinkler package is installed on the pivot, the peak application rates
for points along the system are established. The quantity within the parentheses in
Equation 16.30 is the gross system capacity.
The peak application rate increases linearly with the system capacity and the dis-
tance from the pivot point and inversely with the wetted radius of the sprinkler pack-
age. The time required to reach the peak application rate (tp) can be computed from:
100
INFILTRATION RATE
75
RATES, mm/hr
50
APPLICATION
25
DEPTH
20 mm 30 mm 40 mm 60 mm
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
TIME, hr
Figure 16.28. Illustration of the effect of the application depth on potential runoff.
Wr d g
tp = (16.31)
⎛ Q ⎞
2π R ⎜ s
⎟
⎜ π R2 ⎟
⎝ S ⎠
Thus, the time to reach the peak application rate is linearly related to the wetted ra-
dius of the sprinkler package and the gross depth of water applied, and inversely re-
lated to the distance from the pivot point and the system capacity. The total time that
water is applied to a point is twice the time required to reach the peak application rate
for the elliptical pattern which is symmetrical about the peak rate. Thus, the time that
water is applied to a point is affected by design variables (i.e., Wr, R, QS, and RS) and
by management (dg). The example in Figure 16.28 shows how the potential for runoff
increases for increasing water application depths.
Gilley (1984) used the infiltration rates provided by the intake family of curves
provided by the Natural Resources Conservation Service to estimate the maximum
depth of water that could be applied before runoff began. The surface storage volumes
from Dillion et al. (1972) were used to incorporate the effects of slope on the runoff
process (Table 16.11). Results from Gilley are presented in Figure 16.29 for the four
soils that he analyzed. These results provide a means to begin assessment of the suit-
Table 16.11. Allowable surface storage as a
function of slope (adapted from Dillion et al., 1972).
Slope Range, % Allowable Surface Storage, mm
0-1 12.7
1-3 7.6
3-5 2.5
>5 0.0
50 50
0.3 INTAKE FAMILY
MAXIMUM APPLICATION DEPTH, mm

0.1 INTAKE FAMILY

40 40
30 30
12.5
20 20
12.5
7.5
10 7.5 10
2.5
0 SURFACE STORAGE, mm
2.5 SURFACE STORAGE, mm
0
0 0
0 50 100 150 200 250 0 50 100 150 200 250
PEAK APPLICATION RATE, mm/hr PEAK APPLICATION RATE, mm/hr
50 50
1.0 INTAKE FAMILY

0.5 INTAKE FAMILY

40 40
30 30
12.5
12.5
20 20
7.5
7.5
2.5
10 10
2.5 0
0 SURFACE STORAGE, mm
SURFACE STORAGE, mm
0
0
0 50 100 150 200 250
0 50 100 150 200 250
PEAK APPLICATION RATE, mm/hr
PEAK APPLICATION RATE, mms/hr
Figure 16.29. Maximum depth of application to avoid runoff as a function of the peak ap-
plication rate for the sprinkler package installed on a center pivot for four NRCS intake
families and surface storage values shown in Table 16.11.
ability of sprinkler packages on fields with various soils and slopes. The method by
Gilley was extended to consider an entire field by Wilmes et al. (1994). Von Bernuth
and Gilley (1985) and Martin (1991) adapted the method from Hachum and Alfaro
(1980) to use the Green-Ampt model instead of the family method to simulate infiltra-
tion. These developments allow for the consideration of surface sealing and initial soil
water distribution on the performance of sprinkler packages.
The variables given above for the description of the performance of center pivot
sprinkler packages shows that five system variables are involved. Considering these
variables shows that runoff increases as the system capacity, depth of application, and
the length of the lateral increase. The runoff is inversely related to the wetted radius of
the sprinkler package. Problems can occur if the length of the pivot lateral is excessive,
especially for soils with low infiltration rates and steep slopes. Unfortunately, some op-
erators attempt to rectify runoff problems by reducing the system capacity. This in-
creases the potential for water stress during dry periods and encourages management that
applies excess water during time of the year when water demands are low in an effort to
guard against periods of inadequate supply. The methods such as those by Heermann et
al. (1974), von Bernuth et al. (1984), and Howell et al. (1989) should be employed to
determine the system capacity required for a specific location and soil. These methods
included the utilization of stored soil water to carry crops through periods of high evapo-
rative demand. Users should not arbitrarily reduce capacity to avoid runoff problems.
The sprinkler package and the depth of application are the most feasible adjust-
ments once a system is in place. The depth of application per irrigation can be easily
adjusted by operators to manage applications to reduce runoff. This can be an iterative
procedure based on field observation of runoff problems. The second most feasible
method to solve runoff problems is to replace the sprinkler package (perhaps only on
the outer portions of the pivot) to sprinkler devices that provide a larger wetted radius
for the same pressure as the previous package. Boom systems are available to spread
water over a wider distance to also mitigate problems.
There have also been a series of special tillage systems developed to increase the
surface storage to reduce runoff. One implement that has been used creates implanted
reservoirs. The machine used to create the implanted reservoirs uses a subsoiler fol-
lowed by a paddlewheel assembly that creates the reservoirs. Research by Oliveria et
al. (1987), Kranz and Eisenhauer (1990), Cuelho et al. (1996), and others has helped to
quantify the benefits of systems that develop implanted reservoirs. The implanted res-
ervoirs provide 5 to 10 L of storage per reservoir. For the reservoir density and row
spacing used by Cuelho et al. (1996) the storage volume is equivalent to the water use
for two days during the middle of the season if water is applied to every furrow and all
furrows are specially tilled. The ability to sustain the reservoirs throughout the grow-
ing season varies greatly depending on the cohesiveness of the soil, the slope in the
field and rainfall patterns. Others have used basins within the furrows to store water.
The basins are generally from 1 to 3 m long. Generally basins can store more water
than implanted reservoirs but they are not well suited to sloping soils. In the end, the
designer must weight the potential savings of using low-pressure sprinkler devices that
have smaller wetted radii than devices requiring more pressure against the costs of
special tillage to store water that may run off. In some locations there is an additional
benefit of special tillage in reduction of runoff and erosion from rainfall.
16.7.4 Devices to Irrigate the Corners
Many center pivots are equipped with special sprinklers attached to the end of the
lateral to increase the amount of land irrigated in the corners of the field. The end gun
is operated when the lateral reaches an angle where the end gun throw will stay within
the boundaries of the property (Figure 16.30). The central angle (β) during which the
pivot operates was presented by von Bernuth (1983):
⎛R ⎞
β = cos −1 ⎜ S ⎟
(16.32)
⎜R ⎟
⎝ E⎠
where RE is the total radial length irrigated when the end gun operates. The area irri-
gated in each corner (AE) is given by
( ⎡π
) ⎤
AE = RE2 − RS2 ⎢ − cos −1 (β )⎥
⎣4 ⎦
(16.33)
These relationships show that there is a trade-off between the radial length of the
area irrigated with the end gun and the central angle that is irrigated. When the cover-
age of the end gun is short, the central angle is larger but the area gained per unit rota-
tion of the lateral is small. The area irrigated in one corner relative to the area irrigated
with the primary system is shown in Figure 16.31. These results show that the area in
FIELD BOUNDARY AREA IRRIGATED
WITH ENDGUN
END GUN
RADIUS
RE
β RADIUS OF AREA
IRRIGATED WITH
ENDGUN ON
Rs
RADIUS OF AREA
PIVOT IRRIGATED WITH
POINT ENDGUN OFF
Figure 16.30. Illustrations of the parameters used to describe the
amount of land irrigated in pivot corners.
0.6 6
End Gun Discharge / System Discharge
r ge
Area in One Corner as % of System Area

ha 5
isc
0.5
D
0.4 4
0.3 3
Area
0.2 2
0.1 1
0.0 0
0 0.05 0.10 0.15 0.20 0.25
End Gun Radius / System Radius

Figure 16.31. Discharge required for an end gun as a percentage of the flow for the
main system and the area in one corner of the field relative to the area in the main field.
the corners is maximized when the length of the end gun coverage (RE –RS) is ap-
proximately 18% of the system radius (RS).
The discharge for an end gun depends on the length of the coverage by the end gun:
q E = QS
(R 2
E − RS2 ) (16.34)
RS2
where qE is the discharge required for the end gun.
The discharge for the end gun is approximately 35% of the discharge required for
the primary system if the optimal area is irrigated (Figure 16.31). While the results in
Figure 16.31 show that the area is maximized at a ratio of 18% of the system length, it
is difficult to find end guns that provide the wetted radius required at the needed dis-
charge. The area in the corner does not change appreciably for radii larger than 12% of
the system length; therefore, the radius of the end gun can be less than the optimal
with little loss in area.
Solomon and Kodoam (1978) provided data on the pattern for end sprinklers. They
show that the depth of application from typical end guns decreases near the edge of the
pattern wetted by the end gun. Because the application tapers off at the end, it may be
necessary to apply some water beyond the edge of the field to ensure that crops
planted there receive an adequate water supply. In windy climates this is especially
important. Application of water beyond the boundary of the planted cropland reduces
the application efficiency of the end gun below that for the primary system and affects
the central angle if the overthrow cannot be applied to the adjacent land.
Many center pivots used today utilize a low-pressure sprinkler package for the pri-
mary system. Although end guns have been developed that require less pressure than
earlier, end guns often require more pressure for adequate operation than is available
at the end of the lateral for many sprinkler packages. A booster pump is generally
needed on these systems to provide for adequate operation. The booster pump is usu-
ally located near the end of the lateral, often at the last tower. As special supply line is
provided from the lateral to the booster pump and ultimately to the end gun. The
power required for the end gun is computed from the discharge for the end gun and the
difference in pressure between that for the sprinkler lateral at the distal end and that
required for the end gun. Any pressure losses in the supply system to the end gun must
also be considered.
As the above information illustrates the discharge through the end gun represents a
substantial portion of the flow for the primary system. This results in more pressure
loss along the lateral while the end gun operates, which can cause the discharge from
the sprinkles on the lateral to decline. Ultimately, two systems curves need to be con-
structed to determine how the center pivot interacts with the pump used to supply irri-
gation water. A pump should be selected that provides for efficient operation for both
conditions. It would be most desirable for the difference in the discharge for the pri-
mary system when the end gun operates and when it is off to be small.
To irrigate a larger portion of the area in the corners of a field a corner span can be
attached to the end of the primary lateral. When the lateral has rotated to the proper
angle, the corner span begins to move into the corner of the field. As the corner span
rotates into the corner, a series of sprinklers turns on. More sprinklers operate as the
corner span rotates further into the corner. The hydraulics of the corner system is very
complex requiring computer modeling to predict the performance of corner machines.
16.8 LATERAL MOVE SYSTEMS

The components of center pivot systems have been used to develop a system that
travels in a straight line. These systems are called linear or lateral move machines. The
towers, pipe material and the truss systems are very similar to center pivots. The dif-
ference is that instead of pivoting about a permanent base where water is supplied, the
water supply to the lateral is available across the field. Water application amounts and
the frequency of irrigation for lateral move systems are similar to that for center piv-
ots. Thus, guidelines and limitations for design and operation of pivots generally apply
for lateral move systems.
Water can be supplied to the lateral move by one of three methods. A supply canal
can run parallel to the direction of travel of the lateral move (Figure 16.32). A pump
mounted on the supply tower lifts water from the canal and pressurizes the water for
the system. Although a portable dam can be attached to the suction system to block
flows on sloping fields; the canal system is limited to fields that are relatively flat in
the direction of travel. A second choice is to drag a hose across the field similar to that
for a traveler (see Section 16.10). Water is supplied from the main pump to a riser
along the travel path of the supply tower. A hose is attached to the riser and to the inlet
on the supply tower. As the machine moves it pulls the hose across the field. Many
Figure 16.32. Examples of lateral move systems

that are supplied by a ditch and a hard-hose system.
times the valve is placed in the middle of the field and the supply hose is long enough
to irrigate the entire field without stopping. If the field is too long, several risers may
be needed. Some lateral moves were designed to automatically connect to valves con-
nected to an underground main line. This type of system is equipped with carts that
lead and trail the lateral. The carts automatically connect to valves installed on risers
from the buried main line. Water is supplied by one or both carts. This system is ex-
pensive and complex; thus, few of these systems have been produced.
The lateral move requires a guidance system to control the direction of travel.
Three types of systems have been used. One uses an aboveground cable that runs
across the field parallel to the direction of travel. An assembly on the cart follows the
cable and keeps the lateral move on course. A second option uses a signal from a low-
voltage buried cable with an antenna guidance system on the control tower. The third
option uses a trench cut across the field to provide the direction. A guide follows the
trench to steer the lateral move system across the field.
The guidance and alignment system on lateral move machines is interfaced to
maintain proper alignment and to ensure that the system follows the intended path.
The speed of the last tower is controlled by the irrigator to apply the desired depth of
water. The alignment of individual towers for lateral move systems works similar to
that for center pivots. However, this does not ensure that the lateral move progresses
parallel to the guidance cable. The guidance assembly is designed to reduce the speed
of the interior towers if the lateral move begins to travel away from the guidance ca-
ble. If the lateral move system is progressing on an angle toward the guidance cable,
the velocities of the interior towers are increased causing the lateral move to change
the direction of travel.
Lateral move systems have characteristics similar to pivots. They also have the fol-
lowing advantages:
a larger portion of a square field is irrigated than for pivots;
a rectangular field can be irrigated; and
the rate of water application is less than for pivots, leading to fewer problems
with runoff.
The disadvantages of lateral move systems are that:
the cost per unit area irrigated is substantially higher than for pivots;
more labor is required to move the system to the starting point or to reverse the
system so it can irrigate in the opposite direction;
the hose used to supply the system can be difficult to move and attach; and
aboveground guidance and supply systems interfere with farm operations.
Utilization of lateral move systems lags significantly behind the use of center piv-
ots. The higher investment costs and increase labor required to arrange the water sup-
ply and to reposition the lateral move system deter wider use. However, the systems
do offer substantial promise. Low energy precise application (LEPA) systems and
other types of sprinkler packages that are capable of applying water below crop cano-
pies are well suited to lateral move systems. Such systems essentially become moving
microirrigation systems. If the development of site-specific irrigation develops, lateral
move systems are logical choices for supplying irrigation water, crop nutrients, and
other agricultural chemicals. Thus, there are good reasons to assume that the use of
lateral move irrigation systems will grow.
16.8.1 Layout
Since the cost per unit length of the lateral move usually exceeds the linear cost of
the water supply system, investment costs are minimized by laying out the system so
that the lateral pipeline is parallel to the shortest side of a rectangular field. This orien-
tation results in longer travel distances that produce longer irrigation intervals. The
length of the field, water supply capacity, and the depth of application determine the
irrigation interval. If the lateral move is repositioned to a starting location after each
irrigation the downtime required to move the lateral and water supply system must be
included in the irrigation interval. In some cases irrigators simply reverse the direction
of travel at one edge of the field and irrigate in the opposite direction for the subse-
quent irrigation. There is some downtime to reposition the water supply system for
this mode of operation but it is not as long as required to return the lateral move to the
starting position. With this type of operation the depth of water applied and the irriga-
tion interval at the ends of the field is essentially twice that at the center of the field.
Care must be taken to avoid deep percolation and/or water stress at the edges of the
field when operating in such a manner.
Lateral move systems are not capable of negotiating steep lands or severely rolling
terrain. The maximum recommended slope along the lateral can be as high as 6%, but
should generally be less than 2%. The type of water supply system and the direction of
operation determine the maximum slope in the direction of travel. For canal-fed sys-
tems the maximum slope is about 0.5%, while it is 1% for canal-fed systems with a
movable dam in the canal and 3% for hose-supplied systems.
The pressure loss in the lateral can be minimized by placing the water supply sys-
tem in the center of the field. However, this may present obstacles to farming opera-
tions that are less severe if the water source is placed along the boundary of the field.
The consequences of each alternative should be discussed with the producer.
As with all sprinkler systems the main line must be designed after the lateral move
system is designed. The head-discharge curve for matching pumps to the sprinkler
system can be developed by treating the lateral move as a single large sprinkler or the
pressure distribution function can be used directly. The head-discharge relationship
depends on the losses in the conveyance system, supply carts, and canal suction sys-
tem. The losses in various components for typical systems are shown in Figure 16.33.
Information should be obtained for specific models for final design.
16.8.2 Water Application
Lateral move systems have characteristics similar to center pivots and periodically
moved laterals. The discharge from sprinklers should be constant similar to moved
laterals. The discharge is given by
Q S
qS = S L
FW
(16.35)
where the field width (Fw) is parallel to the lateral.
The pressure distribution along the lateral is similar to that for moved laterals, yet
the pipe materials are the same as for center pivots. The friction per unit length along
the lateral can be computed from Table 16.10; however, the F value for lateral move
systems should be determined as for moved laterals.
50
64 76 89 102 114 127
45
NOMINAL
40 DIAMETER
OF HOSE,
35
H EA D LOSS, m
30
HARD HOSE 152
25
20
15
10
5 LOSS IN SUPPLY CART
0
0 10 20 30 40 50 60 70 80 90 100
INFLOW, L/s
50
64 76 89 102 114
45
40
35 127
HEAD LOSS, m
30
25
NOMINAL
20 DIAMETER
OF HOSE, mm
15
10
SOFT HOSE
5 @ 1000 kPa
0
0 10 20 30 40 50 60 70 80 90 100
INFLOW, L/s
Figure 16.33. Head loss in hard and soft hoses used on lateral move and
traveler irrigation systems. Head loss in the supply cart for travelers is also shown.
The pressure and required discharge for a sprinkler are used to select the sprinkler
package and the nozzle sizes. The lateral move is different than moved laterals in that
the size of nozzle may vary along the lateral as the pressure changes. If pressure regu-
lators are used along the entire machine, the nozzle sizes would be constant.
The application rate is uniform along the lateral of a lateral move system because
the representative area is the same for all sprinkler The application rate can be com-
puted using the elliptical or triangular patterns as previously presented. The peak ap-
plication rate for a lateral move system is given by
4QS
Pp = (16.36)
πWr FW
and the time to the peak application rate is
tp = Wr/v (16.37)
where v is the average linear velocity of the lateral move and Wr is the radius of cover-
age for the sprinkler package.
For equal net system capacities and application efficiencies, the peak application
rate for a lateral move is equal to the application rate at a point 65% of the way along
the lateral of a pivot. The depth of water applied per irrigation is given by
QS
dg = (16.38)
vFW
Runoff is less of a problem with lateral move systems than for center pivots be-
cause field slopes are less than sometimes found for center pivots and the peak water
application rate is less than for pivots. The runoff relationships for center pivots in
Figure 16.29 can be utilized to determine if runoff is a potential problem.
Lateral move systems provide the opportunity of very high uniformity of applica-
tion. Part-circle sprinklers can be used to improve the uniformity at field boundaries.
The effect of sprinkler spacing on the uniformity can be estimated using the overlap-
ping procedures developed for center pivots.
Automated sprinkler systems are increasingly replacing surface irrigation systems.
When systems are replaced irrigation management must change. A frequent problem
is that irrigators attempt to apply the same depth of water per irrigation with auto-
mated sprinkler systems as was applied with surface systems. This negates the poten-
tial of the automated system and is generally unsuccessful. Runoff and traction prob-
lems generally occur with such management. Applying smaller depths per irrigation
capitalizes on the potential of automated sprinkler systems to provide high application
efficiencies. The maximum depth of water applied with lateral move systems should
be less than 50 mm with 25 mm per application being typical.
16.8.3 Water Supply System
Design considerations vary depending on the type of water supply used for the lat-
eral move. For systems that drag a supply hose, considerations involve the friction loss
in the hose and other elements of the water supply system and the force required to
drag the hose across the field.
The pipe roughness for the Hazen-Williams equation (i.e., the C value) is approxi-
mately 150 for the hard or soft hoses used to supply water. The inside diameter of soft
hoses varies with the pressure inside the pipe and additional information is required
for operating pressure other than shown in Figure 16.33. There are limits on the short-
est bending radius for each hose. Short bends cause the soft hose to kink, which tem-
porarily blocks the flow, but there is no long-term damage to the hose. The hard hose
can be damaged if it is bent too severely.
The force required to drag the hose limits the maximum diameter and length of
hose and may require a special design of the tower that pulls the hose. Hose lengths of
200 m are common as that length matches well with the property dimensions common
in the U.S. Hoses longer than 200 m are not common and should only be specified if
in consultation with manufacturers. The normal towers used to pull the hose are not
usually capable of pulling 200 m of 203-mm diameter hose. Even for smaller diameter
hose, the supply tower can be equipped with four wheels that are supplied power
rather than the normal two. This improves traction for wet or slippery surfaces.
Systems that are supplied from a ditch require that the channel capacity, system ca-
pacity, water supply source, and the onboard pump and power unit be matched. Once
the topography is known the ditch can be designed using procedures described in ear-
lier chapters on conveyance systems. The minimum depth of the ditch is usually about
1 m. The minimum bottom width is about 0.3 to 0.6 m. Erosion, weed control, and
trash in the ditch are issues that must be considered in the design and operation. If the
soils at the site are highly permeable, the canal may need to be lined with flexible
membranes or concrete. Either type of lining significantly increases the cost of the
system. The sideslope of the canal should not be so steep that erosion occurs or that
humans and wildlife cannot climb from the canal should they enter the waterway. Out-
flow structures may be required in locations where storm water enters the canal. Spe-
cial features are needed for canal supply systems to ensure that the ditch does not
overtop if the lateral move stops. Controls are also needed to stop the lateral move if
the water supply is interrupted.
Systems that automatically connect to valves attached to buried main lines require
special considerations in design. When the water flow changes between the supply
carts some water hammer may develop in the supply system. The amount of pressure
surge depends on the flow and the time required for valves to open and close. Pressure
surges problems are most prevalent on long lateral moves that require a large inflow.
Pipeline protection is essential for these systems. The supply system for these lateral
moves is very complex and is usually designed by the manufacturer.
16.9 LOW ENERGY PRECISION APPLICATION
(LEPA) SYSTEMS
LEPA irrigation systems are either center pivots or lateral move systems that are
modified with extended length drop tubes and application devices designed to apply
small frequent irrigations at or near ground level to individual furrows (Lyle and Bor-
dovsky, 1981). The primary purpose of LEPA systems is to minimize evaporation
from spray droplets, foliage, and soil surfaces. The LEPA concept involves soil sur-
face management to increase surface storage (Lyle and Bordovsky, 1983). Selected
tillage methods and/or crop residue management are used to increase retention of both
rainfall and irrigation.
Nozzle pressure requirements for LEPA are low since wide dispersal of water is not
necessary. Much of the pressure head at the nozzle (located near ground level) is ob-
tained by the elevation head differential between the lateral and the nozzle. This re-
sults in reduced lateral design pressures as compared to overhead sprinkler packages
and reduced energy requirements.
LEPA systems discharge water beneath the plant canopy and therefore offer an op-
portunity to irrigate with poor-quality water that might cause leaf burn when applied
through sprinkler systems. LEPA systems are also advantageous for crops susceptible
to fungal diseases that may thrive with frequent wetting of the foliage.
16.9.1 General Considerations
There are numerous considerations in the design, installation and management of
LEPA systems that are not required for sprinkler systems (Lyle, 1994). Since water is
applied as a narrow band or stream it is important that the LEPA drop tubes and dis-
charge devices be positioned so that each plant within a field has equal opportunity for
irrigation water delivery. This is best accomplished if water is applied in the furrow
between crop rows. It is, therefore, recommended that plant row position for both lat-
eral move systems and center pivots be established using the system’s tower tire tracks
as a guide for row crop establishment. This guideline results in circular rows for center
pivots and linear rows for lateral move systems.
It is highly desirable that the system span length be an even multiple of the row
spacing and also of equipment width to facilitate consistency in row establishment and
applicator placement along each span. For alternate furrow application, the drop tubes
and applicators should be positioned in the “soft” furrows or those not compacted by
tractor or equipment wheel traffic so that infiltration is maintained. This requires that
equipment gage wheel location in relation to tractor wheels be such that every other
furrow remains free of traffic.
LEPA systems are designed to be essentially independent of soil intake rates.
LEPA systems are best suited to soils that maintain structural integrity throughout the
season to retain surface storage formed by enhancing tillage practices.
Topography (slope) is the primary limiting factor in choosing LEPA irrigation. Al-
though system speed can be adjusted to accommodate slopes of up to 2% without irri-
gation surface redistribution, slopes should be limited to 1% for circular rows in cli-
mates with high-intensity rainfall to minimize rainfall runoff and potential soil ero-
sion. In all instances, normal practices to prevent erosion from rainfall runoff, such as
terracing and/or grassed waterways, should be used in conjunction with LEPA.
16.9.2 Surface Storage and Water Application Devices
The establishment of improved soil surface storage and its maintenance throughout
the irrigation season is a function of both tillage methods and type of applicator cho-
sen for irrigation. The combination should maintain seasonal surface water storage
capable of holding the entire water volume of each irrigation pass without surface wa-
ter redistribution. Recommended surface modification practices include basin tillage,
reservoir tillage, interfurrow chiseling or subsoiling, or any combination of these prac-
tices with thick standing stubble or residue. These practices may also beneficially in-
crease the infiltration rate.
To insure surface containment of applied irrigation, LEPA systems should be oper-
ated at a sufficiently high speed so that the application volume is less than or equal to
the surface storage. For some situations and locations this might require daily irriga-
tion.
The surface storage should be minimally diminished during LEPA irrigation. In
addition, the design and installation of the nozzle and regulator system should deliver
maximum uniformity for all operating conditions within the field and throughout the
irrigation season. Ideally, LEPA applicators should have a narrow profile to minimize
crop contact and drag. It is desirable for applicators to also possess spray capability in
addition to single furrow discharge, although this is not the primary mode of opera-
tion. Examples of spray application needs include temporary use for seed germination,
herbicide application, and when close-seeded crops are included in cropping rotations.
Currently there are two basic LEPA applicator designs that minimize erosion to
furrow dikes or other enhanced soil storage conditions (Figure 16.2). One consists of a
nozzle and shroud assembly which causes water to be discharged as a continuous sheet
or bubble. The other design uses removable drag socks, designed to minimize furrow
dike erosion as water is delivered directly to the soil surface.
Nozzle selection and hydraulic design are similar to that of other pressurized irriga-
tion systems but careful consideration of elevation head gain and friction losses in
each drop tube between the lateral and the nozzle is required to accurately determine
the pressure available to the applicator.
16.9.3 Drop Tube Design
The drop tube should be sufficiently long to position the application device when
irrigating from 10 to 45 cm above the soil surface depending on type of applicator and
topography. The diameters of all materials making up the drop tube should be suffi-
cient to supply necessary operating pressure to an applicator or pressure regulator with
an overhead lateral pressure at the end of the system of no greater than 70 kPa. A pres-
sure of 21 kPa above the regulator rating is normally necessary at the regulator inlet
for correct operation of LEPA systems. Drop tube diameters normally range from 15
to 20 mm.
Drop tubes are attached to the overhead pipeline outlets by means of furrow arms
(extended length gooseneck connectors) which are normally 30 to 50 cm in length.
Although pivots and lateral move systems may be ordered with outlets spacings the
same as the desired furrow spacing, they are often not in satisfactory alignment. The
furrow arm is therefore rotated to center the drop over the furrow. When determining
friction loss between the pipeline and nozzle, entrance and friction losses in the furrow
arm and associated fittings as well as the various components of the drop tube must be
calculated using appropriate head loss equations.
Several material options are available for drop tube construction and configuration.
The major (largest) portion of the drop tube, normally the upper portion, must be suf-
ficiently rigid to prevent significant movement in high winds during preplant irrigation
and prior to full canopy establishment. Rigidity should also be sufficient to insure that
the LEPA applicator remains within the crop canopy after full canopy development.
The entire drop tube should exhibit sufficient flexibility to allow travel over terraces or
soil mounds, or encounters with other objects without breakage. The upper rigid sec-
tion of drop tubes may consist of galvanized steel, UV-protected PVC, or extruded
polyethylene. Galvanized steel may also be used for a lower section (approximately 70
cm in length) located above the applicator for weight and rigidity to help maintain
applicator position within the canopy. Slip weights of galvanized metal, concrete, or
poly material may also be slipped over light and/or flexible materials to make up the
lower section of drop tubes. A short section (0.5 to 0.67 m) of flexible reinforced or
non-reinforced vinyl-covered PVC hose or tubing should be used to couple the more
rigid upper and lower section to provide necessary flexibility to the drop. It also allows
adjustment of applicator height above the soil surface by modifying the hose length
after filling the system with water.
16.9.4 Evaluation and Performance
The coefficient of uniformity is used for evaluating overhead irrigation systems but
it is not applicable to the individual furrow application of LEPA systems. Instead,
LEPA nozzle package design and performance should be evaluated by nozzle dis-
charge uniformity. The nozzle discharge uniformity describes the uniformity of LEPA
nozzle discharge rate converted to equivalent application depth along the length of the
system. The nozzle discharge uniformity is calculated for pivots with the Heermann-
Hein equation and with the Christiansen’s equation for lateral move systems. How-
ever, the depth of applied water in the equation is replaced with equivalent application
depth, which equals the individual nozzle discharge (measured by timed volumetric
catchment) divided by coverage area per nozzle.
LEPA nozzle package design should result in nozzle discharge uniformities of 96
or greater. A measured drop in uniformity, to 94, due to elevation and/or flow rate
changes in the field is the lowest value recommended. Modification of design parame-
ters (operating pressure, pipe size, pressure regulator use, etc.) should be considered if
uniformity drops below 94 over significant portions of an irrigated field.
16.10 TRAVELERS
Another method to automate and save labor is the traveler irrigation system. Trav-
eler systems consist of a large sprinkler, commonly referred to as a “gun,” mounted on
a moving cart (Figure 16.34). Water is supplied to the cart by a hose. With early de-
signs the cart was connected to a cable and winch system which was attached to an
anchor. The winch, which was powered by water pressure, retracted the cable pulling
the cart across the field. Many current designs use the water supply hose to pull the
cart across the field (Figure 16.34). A large spool is anchored at one end of a travel
lane. The spool rotates winding up the hose and pulling the cart toward the spool.
Whether connected by cable or hose, the spool assembly is designed to provide a
nearly constant travel velocity. This requires that the spool rotate faster when small
amounts of cable or hose have been retracted.
The traveler system is operated as shown in Figure 16.34. The cart is positioned at
one edge of the field and is pulled to the center. The system then shutdowns and the
cart is moved to the opposite side of the field. Once a strip is irrigated, the system is
moved to the next lane. Occasionally, the system can be pulled across the entire field
to avoid reconnecting the hose at the middle of the field. The gun on the cart dis-
charges a large volume of water and produces a large wetted radius. Therefore, the
travel lanes are often spaced up to 100 m apart.
The disadvantages of traveler systems include:
High pressure requirements—Travelers require higher pressures than other
sprinkler systems. The pressure at the cart can exceed 700 kPa. Pressure loss in
the hose and water supply components add to the pressure requirement. There-
fore, traveler irrigation systems have high operating costs.
Labor required to move the cart—Traveler systems should be considered semi-
automated since the cart, hose, and spool must be manually moved.
Loss of land in the travel lanes—The carts generally have a low clearance, so if
tall crops are grown, the lane must be planted to a low-growing crop.
Nonuniform application at the edges and center of the field—The gun is oper-
ated to irrigate part of a circular pattern (Figure 16.34). The exact angle depends
on the lane spacing, the overlap of adjacent passes of the traveler, and the design
of the gun. However, it is common to require more than 180° rotation, as shown
in Figure 16.34. If the traveler cannot apply water beyond the boundaries of the
property, there are sections along the edges of the field that may receive less wa-
ter than the bulk of the field. A series of dry diamond-shaped regions may also
occur near the center of the field. If too much overlap occurs at the center there
can be areas of excess irrigation.
HOSE REEL
AND CART
SUPPLY
HOSE
TRAVELING
GUN
AREA WETTED IN
ZONE WETTED WHEN STATIONARY ONE PASS
TRAVELING GUN
SUPPLY HOSE HOSE REEL/CART
TRAVEL LANE
FIELD MAINLINE
BOUNDARY
WELL & PUMP
Figure 16.34. Picture and operational sketch of traveler or big-gun irrigation system.
Poor uniformity in windy conditions—Since the traveler throws water a long

distance, wind can distort the application pattern. To maintain uniformity in lo-
cations with high winds the travel lanes must be closer together, which increases
labor requirements and leads to longer times between irrigations.
Large droplets—Traveler systems often produce large droplets with a high ve-
locity producing a large amount of energy when reaching the soil/crop surface.
The impact energy can exacerbate runoff and soil compaction. Therefore, travel-
ers are best suited to cropping and farming systems that provide plant or residue
material to absorb the energy in the droplets before reaching the soil surface.
There are several advantages for traveler systems, including their:
Flexibility—Since the lanes can vary in length, traveler systems can be used to
irrigate nearly any shape of field. The systems can be moved from field to field
to irrigate several tracts of land. This is especially attractive in semi-humid areas
where the system may not be necessary for one field for the entire season or
where cropping rotations require that irrigation be shifted from one field to an-
other.
Wide range of irrigation depths—Since the sprinkler moves automatically, the
depth of water applied per irrigation can be set to the desired amount. If a small
irrigation is needed the system can be set to move quickly across the field. Con-
versely, for deep-rooted crops the cart can move slowly.
Factory assembly—The system is assembled during manufacturing, thus the sys-
tem can be quickly installed. This allows traveler systems to be operational with
a minimum amount of preparation.
Traveler irrigation systems are mostly used in semi-arid and semi-humid climates
where the needs for irrigation are not as large and consistent as for arid lands. The
systems are more popular in locations where the size and shape of fields are not ame-
nable to center pivot or lateral move systems, or where multiple fields can be irrigated
from the same water source. The high pressure requirements for travelers cause oper-
ating costs to be high. The major considerations for travelers are the hydraulic design
of the water supply system and the arrangement of travel lanes to match field bounda-
ries while achieving acceptable uniformities of application.
Users should be very careful around traveler systems. Travelers operate at high
pressures and large amounts of tension develop in the hose or cable system used to
move the traveler. Both of these conditions contribute to the danger. Special care
should be taken to avoid applying water onto electrical power lines. Kay (1983) rec-
ommends travel lanes be located at least 30 m from electrical power lines.
16.10.1 Field Layout
As with any irrigation system, the initial design step is to lay out the field bounda-
ries and arrange the system on the landscape. Topographical information is needed to
ensure that the desired uniformity is achieved. Small elevation changes in the field are
of less concern with travelers because of the high operating pressure. Slopes only
slightly affect the ability of the traveler to maintain the uniform speed of travel re-
quired for acceptable uniformities. Therefore, the topography within a field is not of
major concern to the operation of the traveler. The pressure required to reach the high-
est elevation in the field should be the primary interest.
The arrangement of travel lanes is the principal consideration in laying out the sys-
tem. The maximum length of the supply hose is generally 200 m. If the field is wider
than 200 m the traveler will generally have to be pulled toward the center of the field
requiring that the main line be positioned through the center of the field. The distance
between lanes should be the same across the field. Also, the lanes must be placed close
enough together to provide the overlap needed for acceptable uniformity. The spacing
of the travel lanes depends on the wind speed and the diameter of coverage of the
sprinkler. Recommended maximum spacings are given in Table 16.12. Iteration may
be needed to find a travel lane spacing that provides uniform irrigation while fitting
within the field dimensions. It is generally recommended that the travel lanes be ori-
ented perpendicular to the prevailing wind direction. The first travel lane may be
Table 16.12. Maximum spacing of lanes for traveler systems

(adapted from Addink et al., 1980 and Kay, 1983).
Wind Speed, m s-1
Diameter of 0 0 to 2.5 2.5 to 5 >5
Coverage, m (80% of Wd) (70% of Wd) (60% of Wd) (50% of Wd)
50 40 35 30 25
60 48 42 36 30
70 56 49 42 35
80 64 56 48 40
90 72 63 54 45
100 80 70 60 50
120 96 84 72 60
140 112 98 84 70
160 128 112 96 80
180 144 126 108 90
located a half of the spacing from the boundary of the field if the water from the sprin-
kler can be applied beyond the edge of the field. Otherwise the first travel lane should
be a full spacing from the field boundary. The sprinkler can be positioned at the edge
of the field if the water can be applied beyond the edges of the field that are perpen-
dicular to the travel lanes. Otherwise the sprinkler cart should be positioned to avoid
throwing water beyond the property boundaries. If multiple travelers are used in one
field it is desirable to supply water to the center of the field and to place a traveler on
each half of the field. This minimizes the difference in pressure available to each
sprinkler.
16.10.2 Hydraulic Design
The discharge from the sprinkler on the travelers should be equal to the flow re-
quired for the field divided by the number of travelers used in the field. The guns used
on travelers can be equipped with either tapered or ring nozzles. Tapered nozzles gen-
erally produce larger diameters of coverage but do not provide as much breakup of the
sprinkler jet.
The depth of water applied per irrigation can be determined from
qS
dg = (16.39)
vWT
where WT is the width of the travel lane (i.e. the distance between travel lanes) and v is
the linear velocity of the sprinkler cart.
The discharge from the sprinkler and the spacing between travel lanes determines
the pressure required for the sprinkler. The pressure must be high enough to provide
the discharge and diameter of coverage required.
The total pressure loss in the systems consists of losses in the main line, fittings,
supply hose, and the sprinkler and hose carts. The pressure loss in the supply hose can
be determined from Figure 16.33. The diameter of the soft hose increases as the pres-
sure of the system increases. Thus, if the pressure is different than shown in Figure
16.33 the head loss will vary accordingly. There is additional loss within the sprinkler
and reel carts. Losses increase when the hose is wrapped around the reel rather than
when fully extended. Oakes and Rochester (1980), Rochester et al. (1990), and Roch-
ester and Hackwell (1991) provided methods to compute the friction loss for varying
sizes of travelers. The head loss in the components of the traveler was computed from
hl = kV 2/2g (16.40)
where hl = the loss
k = the loss coefficient
V2/2g = the velocity head.
The loss coefficients in the sprinkler cart and hose-reel cart were found to be 1.76
and 3.91, respectively. The coefficient for the loss due to coiling on the reel was repre-
sented by a bend coefficient which was found to be 0.09 m-1. The bend coefficient is
multiplied by the length of hose coiled on the reel. The loss coefficients vary with the
size of the system and the design of the equipment; therefore, they should be used as
an initial estimate.
Portable pipe or underground pipelines with risers and valves can be used for the
main line used to supply water to the traveler. The friction loss in the main line is
similar to that for any other type of sprinkler system. Special care is required to protect
pipelines against pressure surges with traveler systems. Travelers require high operat-
ing pressures so pumps used to supply the water are usually designed for high pres-
sures. When more than one traveler is supplied by the same main line or if the supply
hose should kink or some other form of obstruction occurs, the pressure in the convey-
ance system can increase rapidly. The pipeline should be protected to avoid damage
under such conditions. In many cases it is best to select pumps where the shutoff head
is less than the pressure rating of the supply system. Water hammer can also be a prob-
lem since the pressure builds quickly when water reaches the sprinkler. The elasticity
of the hose helps to minimize some of the pressure surge problem but water hammer
should be evaluated in designing the supply system.
16.10.3 Uniformity
The uniformity of application with traveler systems depends on a constant velocity
of travel, constant sprinkler discharge and proper spacing to provide adequate overlap.
Travelers are now specially designed to vary the speed of the reel used to retract the
sprinkler cart so that variations in travel speed along the travel lane are not severe.
Long supply hoses exert considerable resistance when fully extended as compared to
when only a portion of the hose must be moved. The increased resistance can cause
variations in the speed of travel. Thus, designers should check with manufacturers for
the maximum length of hose for local conditions. The pressure at the inlet to the
sprinkler varies with the distance of travel of the sprinkler cart, thus calculations
should be made for varying amounts of retracted hose.
The uniformity can be estimated using the overlapping procedure for a constantly
moving sprinkler. The single-leg distribution of the guns used on travelers will not
usually fit the elliptical or triangular functions that were presented, thus the procedure
must be modified for the appropriate distribution (Rochester et al., 1989). The over-
lapping process can be used to assist in selecting lane spacings. The single-leg distri-
bution for average wind conditions should be used if available. However, the overall
field uniformity and application uniformity are difficult to estimate due to the variabil-
ity along field boundaries and at the center of the field.
The operator must check to ensure that the sprinkler device is operating at the ap-
propriate pressure. Low pressure leads to an inadequate breakup of droplets which
produces a doughnut-shaped water application pattern. Excessive pressure causes the
sprinkler jet to break into to small drops that do not travel as far as intended and that
are subject to evaporation and drift.
16.11 AUXILIARY USES OF SPRINKLER SYSTEMS
Sprinkler irrigation systems can perform several auxiliary uses in addition to sup-
plying crop water requirements. Permanently installed systems can be operated
quickly to meet these additional demands. If the system must be positioned in the field
and periodically moved the ability to perform auxiliary uses is diminished.
The application of effluent and agricultural chemicals are two common uses. Since
the sprinkler system is designed for high uniformity and operates during the rapid
growth period of crops, sprinkler systems provide excellent capabilities for these ap-
plications. However, these applications are regulated in most states to ensure protec-
tions of soil and water resources. In addition, special hydraulic designs are required to
guard against the backflow of chemicals into the water source. Details of the use of
sprinkler systems for chemigation are discussed in more detail in Chapter 19.
Due to the high heat of fusion for ice, sprinkler systems can be used for frost and
freeze protection in some applications. Generally water must be applied frequently to
the crops during periods of frost or freeze danger. This may require a higher flow rate
for the field than needed for crop water requirements. The requirement also eliminates
those systems that cannot irrigate the entire field in a very short time. Very careful
management is required to ensure that the objectives are accomplished for frost or
freeze protection. Successful practices vary considerably depending on the water
source, wind speeds, and other local conditions. Local guidelines should be followed
for success. Care must also be taken to avoid damage to the irrigation system when
applying water during cold periods. Ice may form on structural members of the irriga-
tion system and the added weight may cause components to fail.
In some locations wind erosion, before plants are large enough to shield the soil
surface, is a major concern. Irrigation during such periods may increase the cohesion
between soil particles, which increases aggregate stability and reduces erosion. Care
must be taken, however, because the impact from water droplets can dislodge soil par-
ticles and contribute to increased wind erosion when the soil dries. Irrigation with
small applications is usually adequate to stabilize the soil surface for a period of time.
The efficiency of water use for erosion control is low. The small application does not
wet the soil to a very large depth which leads to evaporation of water from the soil
surface with little storage of water in the soil profile. In many ways irrigating to con-
trol wind erosion is a last-ditch effort, as it is better controlled through residue man-
agement and other production practices.
16.12 SAFETY
Producers, service technicians, and other individuals that work around sprinkler ir-
rigation equipment must be very careful. Sprinkler irrigation equipment is often con-
nected to high-voltage electrical supplies, has numerous moving parts, requires high
water pressures, and operates in a wet and slippery environment. The systems are oc-
casionally used to apply chemicals that could be toxic. While many standards and op-
erational guidelines have been developed for the proper design, manufacturing, instal-
lation, and operation of this equipment, not all systems adhere to the intended prac-
tices. Anyone designing or operating these systems must be aware of the appropriate
laws, codes, and engineering standards that apply to the equipment and the intended
use. The standards from ASABE are continually updated and should be consulted rou-
tinely for proper practices. Local ordinances and other regulations should be deter-
mined before the system is designed.
16.13 SUMMARY
This chapter describes the fundamentals of sprinkler irrigation, performance of
sprinkler systems including uniformity and efficiency of application, types and charac-
teristics of sprinkler systems currently used, and design and management procedures
for specific types of sprinkler systems. Information is provided to enhance design and
management of sprinkler systems which are the most rapidly growing form of irriga-
tion today.
LIST OF SYMBOLS
AD allowable depletion
AE area irrigated in a corner of a center pivot
Ai area irrigated
AR representative area for a sprinkler on a center pivot
Cd discharge coefficient
Cn net system capacity
d depth of water applied or effective diameter of water droplet
d50 volume mean drop diameter
dg gross depth of irrigation water applied
Dn inside diameter of a nozzle
Ea application efficiency
Ek kinetic energy
FL length of the field
FW width of the field
Hn nozzle pressure head
Ir rate of infiltration for bare soil relative to protected soil
Ke/a kinetic energy per unit area
MAD management allowed depletion
n number of sets per lateral
N number of sprinklers on a lateral
Nd diameter of the nozzle
NL number of laterals along the field length
NS number of laterals along the width of the field
P pressure in a sprinkler lateral or at a sprinkler nozzle
P(i) rate of water application as a function of time
P3 percentage of drops smaller than 3 mm
PL pressure loss from the inlet into to the distal end of the pivot lateral
PLL pressure loss from the inlet to the distal end for the large diameter pipe
PLS pressure loss from the inlet to the distal end for the small diameter pipe
Pp peak precipitation rate at the sprinkler location
PR pressure in the center pivot lateral at a point R from the pivot point
PS pressure at the distal end of the pivot lateral
Pv percent of the total drops that are smaller than a specified size
qE discharge required for an end gun
qs discharge from a sprinkler
Qs flow into the sprinkler system (equal to the gross system capacity)
R average application rate or radial distance from pivot point
Ra average application rate
RC location along the lateral where the pipe diameter changes
RD root depth during peak water use period
RE total radial length irrigated when the end gun operates
s distance from the observation point to the sprinkler
Sa sand content of the soil
Si silt content of the soil
SL sprinkler spacing along the lateral
Sm distance between laterals along the main line (equal to the set width)
T exposure time
TAW total available water per unit depth of soil
Td downtime between successive irrigations
Ti irrigation interval
Tm time required to move the sprinkler later between sets
To time of operation per irrigation
tp time after initial wetting that the peak application rate is reached.
Ts operational time per set for a lateral
u distance from a sprinkler to a point relative to the wetted radius of the sprinkler
U wind speed
UCp the uniformity coefficient for center pivots
v droplet velocity or linear velocity for a traveler system
Wr radius of coverage or wetted radius of the sprinkler
WT distance between lanes for a traveler system
β angle of operation for center pivots and end guns
ρ density of water
ω angular velocity of a pivot lateral
Ω ratio of the nozzle diameter to the pressure at the base of the sprinkler
REFERENCES
Addink, J. W., J. Keller, C. H. Pair, R. E. Sneed, and J. W. Wolfe. 1980. Design and
operation of sprinkler systems. In Design and Operation of Farm Irrigation
Systems, 621-660. M. E. Jensen, ed. St. Joseph, Mich. ASAE.
Bezdek, J. C., and K. Solomon. 1983. Upper limit lognormal distribution for drop size
data. J. Irrig. Drain. Eng. 109(1): 72-88.
Bittinger, M. W., and R. A. Longenbaugh. 1962. Theoretical distribution of water
from a moving irrigation sprinkler. Trans. ASAE 5(1): 26-30.
Chu, S. T., and D. L. Moe. 1972. Hydraulics of a center pivot system. Trans. ASAE
15(5): 894-896.
Cuelho, R. D., D. L. Martin, and F. H. Chaudry. 1996. Effect of LEPA irrigation on
storage in implanted reservoirs. Trans. ASAE 39(4): 1287-1298.
Dadiao, C., and W. W. Wallender. 1985. Drop size distribution and water application
with low-pressure sprinklers. Trans. ASAE 28(2): 511-514, 516.
Dillon, Jr., R. C., E. A. Hiler, and G. Vittetoe. 1972. Center pivot sprinkler design
based on intake characteristics. Trans. ASAE 15(5): 996-1001.
Edling, R. J. 1985. Kinetic energy, evaporation and wind drift of droplets from low
pressure irrigation nozzles. Trans. ASAE 28(5): 1543-1550.
Eigel, J. D., and I. D. Moore. 1983. A simplified technique for measuring raindrop
size and distribution. Trans. ASAE 26(4): 1079-1084.
Fisher, G. R., and W. W. Wallender. 1988. Collector size and test duration effects on
sprinkler water distribution measurement. Trans. ASAE 31(2): 538-541.
Gilley, J. R. 1984. Suitability of reduced pressure center pivots. J. Irrig. Drain. Eng.
110(1): 22-34.
Hachum, A. Y., and J. F. Alfara. 1980. Rain infiltration into layered soils: Prediction.
J. Irrig. Drain. Eng. 106(4): 311-319.
Han, S., R. G. Evans, and M. W. Kroeger. 1994. Sprinkler distribution patterns in
windy conditions. Trans. ASAE 37(5): 1481-1489.
Hanson, B. R., and W. W. Wallender. 1986. Bidirectional uniformity of water applied
by continuous-move sprinkler machines. Trans. ASAE 29(4): 1047-1053.
Heermann, D. F., and P. R. Hein. 1968. Performance characteristics of self-propelled
center pivot sprinkler irrigation system. Trans. ASAE 11(1): 11-15.
Heermann, D. F., and R. A. Kohl. 1980. Fluid dynamics of sprinkler systems. In
Design and Operation of Farm Irrigation Systems, 583-618. M. E. Jensen, ed. St.
Heermann, D. F., H. H. Shull, and R. H. Mickelson. 1974. Center pivot design
capacities in eastern Colorado. J. Irrig. Drain. Eng. 110(2): 1127-141.
Heermann, D. F., and K. M. Stahl. 2006. CPED: Center Pivot Evaluation and Design.
Available at: www.ars.usda.gov/Services/docs.htm?docid=8118.
Howell, T. A., K. S. Copeland, A. D. Schneider, and D. A. Dusek. 1989. Sprinkler
irrigation management for corn-southern Great Plains. Trans. ASAE 31(2): 147-
160.
Kay, M. 1983. Sprinkler Irrigation Equipment and Practice. London, UK: Batsford
Academic and Educational Ltd.
Van Nostrand Reinhold.
Keller, J., F. Corey, W. R. Walker, and M. E. Vavra. 1980. Evaluation of irrigation
systems. In Irrigation: Challenges of the 80’s. Proc. Second Nat’l Irrigation Symp.,
95-105. St. Joseph, Mich.: ASAE.
Kincaid, D. C. 1982. Sprinkler pattern radius. Trans. ASAE 25(6): 1668-1672.
Kincaid, D. C. 1996. Spraydrop kinetic energy from irrigation sprinklers. Trans. ASAE
39(3): 847-853.
Kincaid, D. C., and D. H. Heermann. 1970. Pressure distribution on a center pivot
sprinkler irrigation system. Trans. ASAE 13(5): 556-558.
Kincaid, D. C., D. F. Heermann, and E. G. Kruse. 1969. Application rates and runoff
in center pivot sprinkler irrigation. Trans. ASAE 12(6): 790-794.
Kincaid, D. C., and T. S. Longley. 1989. A water droplet evaporation and temperature
model. Trans. ASAE 32(2): 457-463.
Kincaid, D. C., K. H. Solomon, and J. C. Oliphant. 1996. Drop size distributions for
irrigation sprinklers. Trans. ASAE 39(3): 839-845.
Kohl, R. A. 1972. Sprinkler precipitation gage errors. Trans. ASAE 15(2): 264-265,
271.
Kohl, R. A., and D. W. DeBoer. 1984. Drop size distributions for a low pressure spray
type agricultural sprinkler. Trans. ASAE 27(6): 1836-1840.
Kohl, K. D., R. A. Kohl, and D. W. DeBoer. 1987. Measurement of low pressure
sprinkler evaporation loss. Trans. ASAE 30(4): 1071-1074.
Kohl, R. A., R. D. von Bernuth, and G. Heubner. 1985. Drop size distribution
measurement problems using a laser unit. Trans. ASAE 28(1): 190-192.
Kranz, W. L., and D. E. Eisenhauer. 1990. Sprinkler irrigation runoff and erosion
control using inter-row tillage techniques. Applied Eng. Agric. 6(6): 739-744.
Levine, G. 1952. Effects of irrigation droplet size on infiltration and aggregate
breakdown. Agric. Eng. 33(9): 559-560.
Li, J., H. Kawano, and K. Yu. 1994. Droplet size distributions from different shaped
sprinkler nozzles. Trans. ASAE 37(6): 1871-1878.
Lundstrom, D. R., and E. C. Stegman. 1988. Irrigation scheduling by the checkbook
method. Extension Circular No. AE-792. Fargo, N.D.: North Dakota State
Extension Service.
Lyle, W. M. 1994. LEPA defined: More control with less consumption. Irrig. J. 44(6):
8,11.
Lyle, W. M., and J. P. Bordovsky. 1981. Low energy precision application (LEPA)
irrigation system. Trans. ASAE 24(5): 1241-1245.
Lyle, W. M., and J. P. Bordovsky. 1983. LEPA irrigation system evaluation. Trans.
ASAE 26(3): 776-781.
Martin, D. L. 1991. Effect of frequency on center pivot irrigation. In Proc. Nat’l Conf.
Irrigation and Drainage, 38-44. ASCE.
Moldenhauer, W. C., and W. D. Kemper. 1969. Interdependence of water drop energy
and clod size on infiltration and clod stability. Soil Sci. Soc. America Proc. 33: 297-
301.
Morgan, R. M. 1993. Water and the Land: A History of American Irrigation.
Washington, D.C.: The Irrigation Association.
Nderitu, S. M., and D. J. Hills. 1993. Sprinkler uniformity as affected by riser
characteristics. Applied Eng. Agric. 9(6): 515-521.
Oakes, P. L., and E. W. Rochester. 1980. Energy utilization of hose towed traveler
irrigators. Trans. ASAE 23(5): 1131-1138.
Oliveira, C. A. S., R. J. Hanks, and U. Shani. 1987. Infiltration and runoff as affected
by pitting, mulching and sprinkler irrigation. Irrig. Sci. 8: 49-64.
Pair, C. H., W. H. Hing, K. R. Frost, R. E. Sneed, and T. J. Schiltz. 1983. Irrigation.
5th ed. Washington, D.C.: The Irrigation Association.
Rochester, E. W., C. A. Flood, Jr., and S. G. Hackwell. 1990. Pressure losses from
hose coiling on hard-hose travelers. Trans. ASAE 33(3): 834-838.
Rochester, E. W., S. G. Hackwell, and K. H. Yoo. 1989. Pressure vs. flow control in
traveler irrigation evaluation. Trans. ASAE 32(6): 2029-2034.
Rochester, E. W., and S. G. Hackwell. 1991. Power and energy requirements of small
hard-hose travelers. Applied Eng. Agric. 7(5): 551-556.
Scaloppi, E. J., and R. G. Allen. 1993. Hydraulics of center pivot laterals. J. Irrig.
Drain. Eng. 119(3): 554-567.
Seginer, I. 1965. Tangential velocity of sprinkler drops. Trans. ASAE 8(1): 90-93.
Seginer, I., D. Kantz, and D. Nir. 1991a. The distortion by wind of the distribution
patterns of single sprinklers. Agric. Water Mgmt. 19: 341-359.
Seginer, I., D. Nir, and R. D. von Bernuth. 1991b. Simulation of wind-distorted
sprinkler patterns. J. Irrig. Drain. Eng. 117(2): 285-306.
Solomon, K., and J. C. Bezdek. 1980. Characterizing sprinkler distribution patterns
with a clustering algorithm. Trans. ASAE 23(4): 899-906.
Solomon, K., and M. Kodoma. 1978. Center pivot end sprinkler pattern analysis and
selection. Trans. ASAE 21(5): 706-712.
Steiner, J. L., E. T. Kanemasu, and R. N. Clark. 1983. Spray losses and partitioning of
water under a center pivot sprinkler system. Trans. ASAE 26(4): 1128-1134.
Stillmunkes, R. T., and L. G. James. 1982. Impact energy of water droplets from
irrigation sprinklers. Trans. ASAE 25(1): 130-133.
Thompson, A. L., J. R. Gilley, and J. M. Norman. 1993. A sprinkler water droplet
evaporation and plant canopy model: I. Model development. Trans. ASAE 36(3):
735-741.
2003. Farm and Ranch Irrigation Survey and the 2002 Census of Agriculture.
National Agricultural Statistics Service (NASS), Agricultural Statistics Board,
USDA.
von Bernuth, R. D. 1983. Nozzling considerations for center pivots with end guns.
Trans. ASAE 26(2): 419-422.
von Bernuth, R. D., and J. R. Gilley. 1985. Evaluation of center pivot application
packages considering droplet induced infiltration reduction. Trans. ASAE 28(6):
1940-1946.
von Bernuth, R. D., D. L. Martin, J. R. Gilley, and D. G. Watts. 1984. Irrigation
system capacities for corn production in Nebraska. Trans. ASAE 27(2): 419-424, 428.
Vories, E. D., and R. D. von Bernuth. 1986. Single nozzle sprinkler performance in
wind. Trans. ASAE 28(6): 1940-1946.
Wilmes, G. J., D. L. Martin, and R. J. Supalla. 1993. Decision support system for
design of center pivots. Trans. ASAE 37(1): 165-175.
CHAPTER 17
MICROIRRIGATION SYSTEMS
Robert G. Evans (USDA-ARS,
Sidney, Montana)
I-Pai Wu (University of Hawaii,
Honolulu, Hawaii)
Allen G. Smajstrala (University of
Florida, Gainesville, Florida)
Abstract. Microirrigation, the slow rate of water application at discrete locations
at low pressures, includes trickle or surface drip, subsurface drip, microsprinklers and
bubblers. It has made tremendous strides in the past three decades, and has become
the modern standard for efficient irrigation practices for water conservation and op-
timal plant responses. Microirrigation is an extremely flexible set of technologies that
can be economically used on almost every crop, soil type and climatic zone, but it re-
quires a high level of management. These particular systems and their unique equip-
ment and components have specialized needs and problems. This chapter discusses
many of the advantages as well as disadvantages of various microirrigation technolo-
gies and their applications to horticultural and agronomic crops. Water quality con-
cerns, filtration and management are addressed in detail
Keywords. Bubblers, Design, Drip, Irrigation, Management, Microsprinklers,
Trickle.
17.1 INTRODUCTION
The development of modern drip irrigation technologies in the 1960s marked a sig-
nificant step in the history of irrigation science and technology. The first attempts
were plagued with problems; however, most of these have been solved and almost all
aspects of microirrigation have greatly matured since that time, especially in the areas
of filtration, water treatment, and emitter technology. Bucks (1995) has provided a
knowledgeable and concise summary of the history of microirrigation for those who
would like additional information.
Microirrigation includes any localized irrigation method that slowly and frequently
provides water directly to the plant root zone. Drip irrigation, trickle irrigation, bub-
blers, localized small microsprinklers, microspinners, and microsprayers are included
in the general term. The slow rate of water application at discrete locations with asso-
ciated low pressure and the irrigation of only a portion of the soil volume in the field
can result in water delivery systems at relatively low cost, as well as reduce water di-
versions, compared to other irrigation methods.
Microirrigation offers the potential for precise, high level management and is an
extremely flexible irrigation method. It can be adapted to almost any cropping situa-
tion and climatic zone. Microirrigation can be used over a wide range of terrain condi-
tions, and it has allowed expansion of irrigated crop production into areas with prob-
lem soils (such as either very low or very high infiltration rates) and poor water quality
that could not be used with other irrigation methods. It can be installed as either a sur-
face or subsurface water application system.
Microirrigation can be used on most agricultural crops, although it is most often
used with high-value specialty crops such as vegetables, ornamentals, vines, berries,
olives, avocados, nuts, fruit crops, and greenhouse plants. In many cases, it can be
economically used for field crops, golf greens, fairways, cotton, and sugarcane. How-
ever, the requirements for appropriate designs and management in humid areas can be
considerably different from those in arid areas and the technology and techniques suit-
able in one area may not work in the other. Microirrigation will not be the most appro-
priate or economical irrigation method in all situations.
The use of microirrigation is rapidly increasing around the world, and it is expected
to continue to be a viable irrigation method for agricultural production in the foresee-
able future. With increasing demands on limited water resources and the need to
minimize environmental consequences of irrigation, microirrigation technology will
undoubtedly play an even more important role in the future. Microirrigation provides
many unique agronomic and water and energy conservation benefits that address many
of the challenges facing irrigated agriculture. Farmers and other microirrigation users
are continually seeking new applications, such as wastewater reuse, that will continue
to provide new challenges for designers and irrigation managers.
Any irrigation system must be compatible with cultural operations associated with
a specific crop. Adoption of microirrigation may require new or innovative adapta-
tions to various cultural practices and even the development of new harvest and tillage
equipment. For example, surface lateral lines can hinder traditional harvest operations,
requiring pre-harvest removal of the tubing or development of a new harvester and
harvesting techniques. Lateral lines can be buried but this generally requires moving to
minimal-tillage or permanent bed systems.
An in-depth understanding of the unique benefits and limitations of microirrigation
systems is needed to successfully design and manage these systems. As with all other
irrigation methods, there are trade-offs with both positive and negative impacts on
irrigation scheduling, efficiency, and uniformity, as well as environmental impacts,
crop responses, and economics.
17.1.1 Advantages and Disadvantages of Microirrigation
Microirrigation has advantages as well as disadvantages to consider and understand
before adopting the technology. Advantages include water conservation and reduced
deleterious water quality impacts due to high application efficiencies, automation ca-
pabilities, improved or increased yields, ease of chemical applications, and potential
sustainability. Disadvantages include a high potential for emitter plugging, high sys-
tem costs, and required high levels of management.
17.1.1.1 Advantages. Microirrigation is commonly used in areas with limited wa-
ter and high water costs, but it has great value in other areas as well. Properly de-
signed, installed, and managed microirrigation systems can eliminate surface runoff
and associated soil erosion, efficiently and uniformly apply water-soluble fertilizers,
634 Chapter 17 Microirrigation Systems
and achieve high uniformity and efficiency of water application. They generally tend
to have smaller wetted areas, reduced deep percolation, and lower evaporation losses
than other irrigation methods. There can be water and chemical savings because of
increased efficiency, reduced weed control costs because a limited surface area is wet-
ted, and better productivity can be achieved due to improved control of water and nu-
trients in the root environment.
Microirrigation generally has high production efficiencies whether expressed as
yield per unit water, yield per unit nutrient input, or yield per unit land area. Advanced
cultural practices such as the use of plastic or sheet paper mulches to reduce weed
growth, heat soils, and decrease soil evaporation are also facilitated by drip irrigation.
Due to relatively small pipe and valve sizes, microirrigation systems are easily and
inexpensively automated, which reduces labor costs and improves general manage-
ment flexibility.
Because microirrigation methods can apply water in small amounts that nearly
match evapotranspiration, soil characteristics such as hydraulic conductivity and wa-
ter-holding capacity are usually not limiting. Less salt may be applied with the irriga-
tion water because less water is needed with these potentially highly efficient systems.
Low soil matric potentials reduce salinity hazards, improve the ability to manage sa-
line or sodic soils, and permit the use of poorer-quality water than can be used with
other irrigation methods. Because of its potential to be highly efficient, microirrigation
is often specified as a best management practice for reducing groundwater contamina-
tion due to irrigation.
The ability to precisely manage soil water deficits and to make prescription applica-
tions of nutrients and other chemicals through the irrigation system often leads to in-
creased yields. In general, a body of research has consistently shown increases in
vegetative growth and yields compared to more traditional irrigation techniques as a
function of scheduling and management. Perennial crops may also experience more
rapid growth and earlier production under microirrigation because water and nutrient
stresses on young plants with small root zones may be greatly reduced. Uniformity in
plant growth across a field, due to uniform water and nutrient distributions, also con-
tributes to overall yield increases.
Microirrigation must be managed as both a water and a nutrient application system.
Fertilizers and other water soluble chemicals such as pesticides (e.g., nematicides,
systemic insecticides, herbicides) and soil amendments (e.g., acids, polymers, pow-
dered gypsum) can be efficiently and effectively applied through microirrigation sys-
tems. Buried drip irrigation systems are particularly amenable to the application of soil
fumigants as well as other chemicals that tend to be fixed by the soil particles (e.g.,
some pesticides and phosphorus fertilizers).
If designed and managed properly, microirrigation systems can reduce off-site im-
pacts of irrigation on wildlife habitat and aquatic ecosystems compared to other meth-
ods. Pesticide use is often reduced because the efficacy of systemic pesticides is en-
hanced. In arid areas, herbicide expenses are usually less because only a portion of the
area is wetted so weed germination is reduced; and because the soil and plant canopy
are generally drier, there is often lower fungal disease pressure and fungicide use is
generally less (Scherm and van Bruggen, 1995). Plastic films (biodegradable and non-
biodegradable), large sheet paper, and other mulches often work very well in drip irri-
gated crop culture to control weeds (and eliminate herbicide use) and reduce soil
evaporation losses.
Microirrigation systems may enhance long-term sustainability of an agricultural
operation because of their potential for maximizing water application efficiencies and
minimizing chemical applications. Limited water supplies (quality and quantity) can
be utilized more efficiently for agricultural production, thus providing more water for
competing uses or reducing withdrawals from aquifers. In addition, microirrigation
methods are low-pressure systems that typically use less total energy compared to
sprinklers.
17.1.1.2 Disadvantages. Because of their relatively small orifice sizes, microirriga-
tion emitters can be easily plugged due to physical, chemical, and biological factors.
Clogging adversely affects uniformity, and can negate the benefits and effectiveness
of microirrigation.
Microirrigation systems are generally expensive to install and maintain but are
similar in costs to most other advanced irrigation methods. For larger systems (e.g.,
>10 ha) with relatively close plant spacing, their cost is comparable to permanent solid
set sprinkler systems covering the same area although the filtration and chemical
treatment systems are major expenses that can vary widely depending on conditions
and system size. High-density plantings requiring large amounts of tubing may not be
economical. Operational costs will be high due to the need for chemical treatment,
filtration, and labor for routine flushing of lines, although lower energy costs and wa-
ter savings may offset some of this increase. There can also be significant costs asso-
ciated with the retrieval and disposal of tape/tube and non-biodegradable plastic
mulches.
A high level of management is required to operate and maintain a microirrigation
system. Managers require a greater level of training and proficiency than for surface or
sprinkler systems. They command higher salaries and are usually employed year-
round because of the need to retain their skills, however, they can generally cover
three to four times as much cropped area as an irrigator using more traditional meth-
ods, primarily due to automation. The higher level of management also requires adop-
tion of ancillary technologies (with their associated costs), such as irrigation schedul-
ing, soil water monitoring, and frequent detailed plant tissue nutrient analysis for ferti-
gation programs.
As a general rule, microirrigation systems are less forgiving of mismanagement or
poor design than methods that irrigate a much larger portion of the root zone. These
problems range from overirrigation and excessive leaching of chemicals to severe
drought, salinity, or nutrient stresses.
Uneven distributions of water, nutrients, and roots across a field can create prob-
lems unique to microirrigation. The restricted wetted soil volume may affect the extent
of the rooting system and the physical stability of a plant. Smaller rooting volumes
also limit the amount of soil water available to buffer the plant against drought in the
event of an irrigation system failure. In addition, the small wetted soil volumes in-
crease the difficulty of maintaining an optimally balanced soil nutritional status be-
cause access to nutrients stored in adjacent nonirrigated soils is limited.
Pest problems may change because the frequent irrigation may create environ-
mental and moisture conditions favorable to fungal diseases or pests that may not be
concerns under other irrigation technologies (e.g., mites that favor dry, dusty condi-
tions). Frequent irrigations may also create optimal conditions for some plant diseases
requiring special management (e.g., fumigation before and/or after the growing season
to minimize inoculum).
Polyethylene microirrigation tubing can be physically damaged by a number of
mechanical and natural causes. Damage by farm equipment commonly occurs. Coyo-
tes, rodents, and other animals may damage tubing, especially when looking for water
in arid areas. Gophers and other rodents may cut through buried tubing as they bur-
row. Woodpeckers have been reported to peck holes in tubing. Insects and spiders can
plug emitters, but may also enlarge orifices when searching for water. Tall grass,
weeds, spider webs, and large insects can stop the rotation of microspinners.
17.1.2 Soil, Water Quantity, and Water Quality Considerations
17.1.2.1 Soils. A microirrigation system must be designed and managed to match
the soils on which it is used. Deep sandy soils often have little lateral spreading of
water requiring several small irrigations each day and/or microsprinklers to expand the
wetted root volumes. Improper scheduling due to poor system design or management
can result in excessive deep percolation and leaching of nutrients. When application
rates exceed infiltration capacity, soils become saturated, weeds and other problems
may be enhanced due to large wetted areas, and runoff may even occur. In addition to
environmental pollution considerations, soil waterlogging can result in increased plant
disease and induce plant physiological disorders.
Soil salinity will affect system design and management because salts accumulate at
the edges of wetted areas and on the soil surface. Deficit irrigation may lead to exces-
sive salt levels in the soil profile. These salts need to be periodically leached, which
can be complicated by the development of preferential flow paths. The use of plastic
film mulches that reduce soil evaporation have also been found to reduce soil salinity
directly under the mulches. Injection of acids (e.g., sulfuric) may sometimes be re-
quired to increase the solubility of salts to facilitate leaching.
Maximum leaching of salts will occur near the emitters, with effectiveness decreas-
ing with distance from the emitters. Seed germination of some crops (e.g., lettuce)
may require sprinkler irrigation to move salts below the seed bed. In areas of low rain-
fall, it may be necessary to sprinkler irrigate periodically to drive salts below the tub-
ing depth. Another salinity-control technique is to irrigate during rains, pushing salts
to the outside of the wetted volume. Bed shaping—forming beds higher than neces-
sary—has also been used. The drip system is then operated to push the salts to the
surface. The salty surface soils are then scraped to the side and the crop planted in the
less-saline soil lower in the bed.
A complete soil chemical analysis should be conducted as an initial part of the
planning process, for water application as well as for cultural decisions. Salination
and/or changes in soil pH may also develop because of the quality of the water supply
or as a result of various water treatment and chemical/fertilizer management programs.
Soil pH can have major effects on the availability of soil nutrients to the plants and in
some cases can cause toxicity. Soil amendments (e.g., gypsum or lime) should be ap-
plied prior to planting to ameliorate existing or anticipated problems, although some
supplemental gypsum can be injected through the microirrigation system. Producers
should annually monitor the soil chemistry in the rooting volume throughout the life
of the irrigation system.
17.1.2.2 Water quantity. Timing, availability, and reliability of long-term water

supplies must be determined. Seasonal irrigation requirement depths for many crops can
range from 100 to as much as 2000 mm. Canal and on-farm delivery systems in many
areas of the world are often designed to satisfy the requirements of surface or sprinkler
irrigation (e.g., 1.0 L s-1 ha-1 based on the farm’s total irrigated area) or deliveries may
be based on calendar rotations (e.g., every 7 days) that are inappropriate for microirri-
gation designs. ET may be higher with microirrigation due to reduced drought stress,
although most of the time it will be less than under other irrigation methods due to
reduced soil evaporation losses. Supplemental wells and storage ponds may be re-
quired to utilize microirrigation technologies effectively under some conditions.
17.1.2.3 Water quality. Physical, biological, and chemical water quality, including
salinity, is a major concern in the management of all microirrigation systems (Naka-
yama and Bucks, 1991; Lamm et al., 2000). The physical, biological, and chemical
characteristics of the water supply from all sources (e.g., wells, canals, reuse ponds)
must be considered. The potential for emitters to become plugged by physical, chemi-
cal, or biological contaminants can create significant problems. Success hinges on
filtering and treating the water to match actual water quality conditions throughout the
year with both surface and groundwater. It is sometimes not economically feasible to
treat a water source to make it suitable for microirrigation and other irrigation methods
should be considered.
A successful water treatment program must accommodate worst-possible condi-
tions while meeting high microirrigation water quality standards. The potential for soil
salination due to the water supply must be assessed. Fertigation and injection of other
chemicals require knowledge of the water chemical constituents to ensure compatibil-
ity between injected chemicals as well as to help determine suitable chemical water
treatment needs and procedures.
Laboratory tests are necessary to determine the nature and composition of inorganic
contaminants, as well as the relative proportions of each that may create significant
problems in the long-term management of the system or affect the crop’s utilization of
water and nutrients. Specific concerns may be pH, salinity (electrical conductivity),
calcium, magnesium, sodium, iron, manganese, carbonates, bicarbonates, and sulfur.
Organic contaminants may sometimes be problematic, but these are usually controlled
by good filtration and chlorine treatments.
A suitable treatment program may consist of several progressive steps or phases in-
cluding: settling basins, gravity screens, centrifugal separators, screen filters, disk fil-
ters, and/or media filters, plus the injection of chlorine, acids, or other water treatment
chemicals. Appropriate design and management of each stage of the treatment system
provides the capability of maintaining high water quality standards throughout the life
of the project despite variations in the physical and chemical properties over time.
More specific and detailed information on water quality concerns related to microirri-
gation is discussed in several sections later in this chapter
17.1.3 Environmental Considerations
Substantial environmental advantages can result from properly designed, main-
tained, and managed microirrigation/chemigation systems. Environmental advantages
result from reduced diversions of water, reduced chemical usage, and reduced ground-
water contamination by reducing leaching of salts and other chemicals below the root
zone.
Although microirrigation can achieve highly uniform water applications, it is in

fact always less than perfectly uniform. As a result, some portions of the field will be
overirrigated and some will be underirrigated. Overirrigation will waste water and
cause deep seepage which may contaminate underlying groundwater resources. Deep
seepage losses occur due to overestimating ET causing excessive applications; nonuni-
formity of irrigation; overirrigation because of poor scheduling or lack of automation
and feedback; and leaching for soil salinity management. Underirrigation will result in
a yield reduction and may cause undesirable soil salinity accumulations.
Improper application of some pesticides and fumigants may negatively affect bene-
ficial soil biota, including earthworms, bacteria, fungi, and insects. Disposal of flush
water from filters and lateral lines may sometimes be a problem, especially if pesti-
cides or fertilizers are in the effluent.
17.1.4 Economic Considerations
The ultimate goal of any agricultural activity is to achieve maximum net economic
return. The irrigation system uniformity, irrigation scheduling practices, cost of water,
yield price, yield reductions by deficit irrigation, and damage caused by overirrigation,
including possible groundwater contamination, are all very important factors affecting
the economic return from a microirrigated production system. Expected economic
returns and the required system uniformity can also affect emitter selection.
17.2 MICROIRRIGATION SYSTEMS
17.2.1 Methods
Microirrigation methods are generally defined by the water emission device. Emis-
sion devices range from thin-wall plastic tube with simple orifices, microsprinklers,
orifices and long-path laminar flow emitters and microtubing, to more elaborate and
efficient turbulent-path and pressure-compensating emitters. Some emission devices
are manufactured as an integral part of the plastic tubes and tapes while others are
attached during installation.
Surface applicators include emitters (drippers), microsprinklers/ microsprayers, and
bubblers, all of which apply water on or above the surface of the soil. Subsurface drip
involves the use of point-source emitters or line-source emitter tubing and tapes to
apply water below the soil surface at depths depending on the soil type and crop. Sur-
face and subsurface drip have also been used for water table control in some humid
areas as a variation of subirrigation, primarily on vegetable crops. Drippers and bub-
blers are designed to apply water at or slightly above atmospheric pressure, whereas
microsprinklers apply water from about 70 to more than 250 kPa.
Two general categories of microirrigation laterals are polyethylene tape and tubing.
Tapes are collapsible, thin-walled, low-pressure polyethylene tubes with built-in emit-
ters or orifices. Tubing is more rigid and more expensive than tape, has thicker walls,
and may or may not have pre-installed emitters. Tapes and tubing may also be divided
into five classes depending on use: (1) disposable, thin-walled surface tape (1-year
life); (2) shallow, buried tapes (1-5 year life); (3) reusable/retrievable surface tapes (1-
to 3-year life); (4) retrievable surface tubing (multi-year life); and, (5) buried tubing
(multi-year life). Tapes are most commonly used on annual or seasonal row crops
while tubing is used more often on perennial crops. ASAE Standard S435, Polyethyl-
ene pipe used for microirrigation laterals, presents manufacturing and testing require-
ments for tubing (ASAE, 2005c).
Tapes typically have wall thicknesses ranging from 0.1 mm to 0.64 mm and inside
diameters may range from 9.5 to 28.6 mm. Emitters usually have close, uniform spac-
ing (e.g., 3.5-60 cm) along a lateral line and emitters are simple orifices, long path,
labyrinth flow paths. or a combination. Emitters may be embossed within the welded
seam of the tape or they may be separate, pre-molded devices installed during fabrica-
tion. Currently, emitters on tapes are not pressure compensating, and water discharge
rates of individual tape emitters range from 0.2 L h-1 to over 7.6 L h-1. Tape operating
pressures range from 20 kPa to about 140 kPa. They should always be installed with
the emitters facing up because of plugging problems due to sediment accumulations in
the bottom.
Tubing has wall thicknesses typically ranging from 0.25 mm to over 0.9 mm with
inside diameters from 9 mm to over 35 mm. Pre-installed emitters on tubing have uni-
form spacing, however, in contrast to tapes, point-source emitters and microsprinklers
can be installed in the field at any spacing on the tubing to meet specific irrigation
requirements. Emitters are either non-pressure compensating or pressure compensating
with water discharge rates from 1.5 to over 20 L h-1. Microsprinkler discharge rates
range from about 5 to over 40 L h-1. Operating pressures range from about 40 to over
250 kPa.
17.2.1.1 Drip and microsprinkler emitters. Water distribution by drip and mi-
crosprinkler emitters can be characterized as line-source or point-source applications
for both tapes and tubing. Line sources apply water in a continuous or near-continuous
pattern along the length of the lateral. In this category are soaker hoses or porous pipes
(line-source emitters) in which the entire pipe wall is a seepage (and filtration) surface,
as well as drip tapes with closely spaced (e.g., 15-30 cm) emission points whose water
application patterns overlap. Point-source emitters can be grouped based on their flow
characteristics into long-path emitters (microtubing, laminar-flow, and turbulent-path
emitters), short-path emitters (microsprayers and other orifice emitters), orifice-vortex
emitters, and pressure-compensating emitters. These devices apply water at discrete
points and overlap between wetting patterns may or may not occur, depending on
emitter spacing, irrigation duration, and emitter flow rate.
Orifices and microtubing emitters are the two simplest emission devices. They
were common in the early development of drip irrigation, but are currently only used
on tapes. Plugging is usually a serious problem for orifice emitters due to small (less
than 0.3 mm) outlet diameters and low discharge rates. Orifice-vortex emitters are
orifice emitters that have been modified so that water enters the emitter with an angu-
lar velocity such that the circular vortex motion provides additional energy loss so that
the orifice can be larger and less prone to plugging.
Microtubing is a long-path emitter inserted into the lateral line. Different lengths of
microtubing can produce various flow rates depending on their dimensions and water
pressure. The size of microtubing typically ranges from 1 to 10 mm in diameter, and
the flow characteristics can be either laminar or turbulent, as a function of tube size.
The laminar-flow (small-diameter) microtubing tends to accumulate small deposits
and is quite susceptible to partial plugging. The long-path or spiral emitter is basically
a microtubing emitter that is wrapped around a short, larger plastic tube to make a
more compact unit. Larger-sized microtubing with turbulent flow produces a bubbler
effect and usually has few plugging problems (Rawlins, 1977).
The labyrinth emitter, designed with long intricate passageways, will create turbu-
lent flow at normal operating pressures and is often called a turbulent or turbulent-path
emitter. The turbulent flow resists plugging by allowing the flow path to be as large as
possible and preventing small particles from settling or becoming lodged in the pas-
sageway. Flow rates from turbulent path emitters are also relatively insensitive to tem-
perature fluctuations (Wu and Phene, 1984; Rodriguez-Sinobas et al., 1999), thus
avoiding a major cause of nonuniform water application under field conditions. Cur-
rently, most point-source emitters utilize turbulent-flow paths to control the applica-
tion of water from tubing and some tapes.
Emitters can be inserted or molded into the tubing or tape during the manufacturing
process. With internal “in-line” emitters, there are no protrusions to interfere with me-
chanical installation or retrieval of the tubing or tape. Alternatively, emitters (and mi-
crosprinklers) can be attached to the outside of the tubing when the system is installed,
usually by manually punching a hole and inserting the barbed end of the emitter. This
procedure requires more labor but it allows a system to be customized to match the
needs of widely or unevenly spaced plants.
Microsprinkler or minisprinkler emission devices are generally simple orifices and
include small, low-pressure minisprinklers, foggers, spitters, jets, and sprayers that are
installed in the field on tubing. These typically apply water (at 35 to 70 L h-1) to larger
areas than drip emitters, but do not uniformly cover the entire cropped area. They are
used to irrigate tree crops, shrubs, widely spaced plantings, and localized grass areas
with extensive root systems, especially on sandy soils where lateral movement of soil
water is limited by soil hydraulic properties or other areas with greatly restricted root
zone depths. Nozzle sizes typically range from 0.5 mm to 2 mm; plugging problems
are greatly reduced with nozzle sizes larger than 0.75 mm (Wu et al., 1991) combined
with adequate filtration and chemical treatment of the water. Microsprinklers are in-
stalled after the lateral tubing or pipe has been laid in the field. They may be inserted
with barbed fittings directly into the tube but are more commonly mounted on stake
assemblies and connected to the lateral lines with 4-6 mm tubing. Some may also be
mounted directly on threaded PVC fittings on the lateral. The state of the art for mi-
crosprinklers is advancing rapidly and improved microsprinklers (e.g., pressure-
compensating and self-cleaning) are being developed and tested.
A variation of the microsprayer pulses the water jet in short bursts of up to 60-70
cycles per minute, which serves to minimize application rates while maximizing the
wetted radius. These can be an advantage on heavy soils with low infiltration rates or
soils where poor lateral water movement may be a concern.
Both drip emitters and microsprinklers are available as pressure-compensating de-
vices. These use a flexible orifice that changes its diameter depending on the pressure,
thereby regulating the flow. Pressure-compensating devices are used to provide uni-
form flows from each emitter along a lateral whenever elevation differences or exces-
sive pressure losses to long lateral lengths cause flow variations to exceed design stan-
dards. However, these devices are more costly than standard emitter devices.
17.2.1.2 Low-head bubbler irrigation systems. Bubblers are large-orifice, low-
pressure emitters that apply water at discrete points but at considerably higher rates
than common drip or microsprinkler emitters. Filtration requirements are greatly re-
duced, but flow rates are often so high that basins or very flat terrain may be required
to prevent runoff. Some bubblers are designed to operate on gravity flow or low-head,
high-flow pumps (1 to 8 m of head), while others use pressurized (e.g., 200 kPa) pipe-
lines to distribute water. The higher-pressure bubbler systems use special devices with
large openings where flow rates can be mechanically adjusted at each location.
Low-head bubbler irrigation systems use microtubing emitters to deliver water to
the crop. These systems consists of a main line connected to a water source, a constant-
head device, manifolds, laterals, and small-diameter (e.g., 4 mm to 26 mm inside di-
ameter microtube) delivery hoses. Laterals are usually buried and laid between rows.
A sufficiently long, large-diameter (e.g., 5 to 25 mm) delivery hose is inserted into the
lateral pipe to deliver water to a plant. The delivery hoses are anchored to a tree or
stake, and the outlet elevations are adjusted to the hydraulic energy gradient so that
water flows or “bubbles” from all hoses at equal rates.
Bubblers are well-suited for economical irrigation of trees and vine crops and are
being developed for turf and landscape applications. Bubbler systems do not usually
require elaborate pumping and filtration systems, but are not widely adopted (Yitayew
et al., 1995). Design considerations and installation of low-head bubbler systems are
discussed by Yitayew et al. (1995), Yitayew et al. (1999), Thorton and Behoteguy
(1980), and Rawlins (1977).
17.2.2 Wetting Patterns
The applied water moves through the soil largely under unsaturated flow conditions
at the wetting front. The distribution of water and the shape of the wetted volume can
be predicted from the physical laws of capillary movement for either point sources or
line sources (Warrick and Lomen, 1983; Clothier et al., 1985; Philip, 1991; Or and
Coelho, 1996; and many others).
A point-source emitter will provide a wetted volume in the soil, which is affected
by the initial soil water content, emitter flow rate, irrigation frequency and duration,
capillary movement of water and the water-holding capacity of the soil. In arid areas,
the emitter creates wetting patterns in the soil that determine the size and shape of the
crop root zone.
“Point sources” refer to individual emitters with discrete application points. A
point-source emitter or a group of emitters forming a point source are generally used
for tree crops or other widely spaced plantings. Microspray emitters with large spacing
such that their wetting patterns do not overlap are also point-source emitters. Even
groups of emitters with overlapped wetting patterns but designed as a unit, such as
around individual trees, can be considered to be a point source.
“Line source” wetting patterns develop when emitter applications along a lateral
merge and form a half-cylinder wetting pattern or trough of wetted soil in the field.
High-density row crops are usually irrigated with line sources by lateral lines with
closely spaced emitters or microsprayers.
The wetting pattern for a point-source emitter in a homogeneous soil is a three-
dimensional hemispherical shape with a water gradient from the center (point source)
to the edge of the sphere. The wetting pattern for a line-source application by closely
spaced emitters will form a two-dimensional half-cylinder shape in homogeneous soil.
In layered soils, the wetting patterns will tend to be confined within the top layer so
the bottom of the hemisphere or half-cylinder will be relatively flat and form a wetting
pattern shaped like a disk or rectangle.
Hardpans in many soils have smaller pore spaces than the material above or below.
The hard pan serves as a barrier because capillary water movement does not readily
occur from smaller to the larger pores below the hardpan. Likewise, where a fine soil
is underlain by a coarser material, irrigation must saturate the upper soil before water
will enter the coarser layer.
17.3 DESIGN FACTORS
17.3.1 General Considerations
The design of a water application system will determine the maximum potential
performance level for any proposed crop use, whereas management dictates the actual
benefits received and the magnitude of any ecological impacts. High-quality installa-
tions are more easily maintained and are much less expensive to operate over time
than a substandard design that requires frequent repairs and has a shorter operational
life. Minimum requirements for the design, installation, and performance of microirri-
gation systems are presented in ASAE Engineering Practice EP405.1, Design and In-
stallation of Microirrigation Systems (ASAE, 2005b).
The first rule of microirrigation design is the same as for all irrigation systems:
keep it as simple as possible. The system must be designed to meet the users’ level of
expertise and it must fit within their perceived needs and cultural practices. It must be
reliable and sustainable, able to manage salts, easy to maintain, and allow for needed
tillage and harvest operations.
The design and installation must be site-specific. They are governed by soil type
and depth distributions, topography, climate, water quality, water quantity, the pro-
posed crops and cropping systems, as well as the preferences of the irrigator. How-
ever, the fundamental aspects of high-frequency irrigations, limited wetted rooting
volumes, filtration and chemical treatment of the applied water, and extraordinary
consideration of the spatial uniformity of water applications per emission device are
common to all microirrigation systems. Designs should facilitate maintenance. Ponds
and chemigation installations should be fenced for safety of workers, children, and
animals. Water treatment, filtration, and lateral line flushing must be high priorities.
Due to low operating pressures and chemigation requirements, hydraulic variables
are more rigorous for microirrigation systems than for other types of systems operat-
ing at higher pressures. Total system pressures should normally not be permitted to
vary by more than 20% unless pressure-compensating emitters are used.
The total allowable pressure loss of the whole system, which provides the desired
design uniformities, is selected at the start of the design process and depends on the
preferred pressure regulation strategy (e.g., optimal combinations of valves, pressure-
compensating emitters, and topographic layouts). Lateral, subunit, submain, mainline,
and control head system pressure losses are assigned so that the sum does not exceed
the total system design criteria.
As much as possible, the systems should be designed based on anticipated actual
installed emitter discharges, which are often different from the manufacturer’s litera-
ture due to factors such as unit-to-unit variation in manufacturing, system elevation
changes, system pressure variations, emitter wear, pressure losses in stake assemblies
(microsprinklers), and varying lengths of small-diameter (e.g., 4- to 6-mm diameter)
supply tubing from the lateral to the emission devices. The coefficient of flow varia-
tion of the emission devices should always be less than 10%. Distribution uniformities
should normally be greater than 90%, especially when chemigation will be used.
Pipelines (usually PVC) should be placed at sufficient depths to avoid damage from
farm and construction equipment. Concrete blocks to prevent pipeline movement
(thrust blocks) need to be placed at appropriate locations to prevent failure of pipe-

lines, valves, and other components. Adequate air relief, vacuum breakers, and pres-
sure relief valves must be sited appropriately to ensure proper operation. Information
and procedures for installation of PVC pipe can be found in ASAE Standard S376.1,
Design, Installation and Performance of Underground, Thermoplastic Irrigation Pipe-
lines (ASAE, 2005a). Some emitters are also designed to facilitate air relief in the lat-
erals used on highly variable topography.
Designing for ease of maintenance is critical. It is important to ensure that there is
easy access to all equipment and various system components for maintenance and re-
placement whether buried or on the surface. All aboveground components are typi-
cally at least 0.4 m but not more than 1.5 m above concrete or soil surfaces for ease of
maintenance. All pipe nipples should have sufficient length for wrenches. Fenced ar-
eas should have provisions for equipment access.
Valves should be installed to hydraulically isolate components requiring frequent
cleaning, repair, or replacement, such as inline filters. Likewise, it should be possible
to isolate blocks within a field for maintenance without shutting down the entire sys-
tem. Unions, flanges, or Victaulic couplings should be provided for easy removal of
the affected components. In areas of cold climate, there must be the capability to win-
terize the entire system, including drain valves, infiltration pits and, if necessary, pro-
visions for using compressed air to remove water when pipelines do not have appro-
priate slopes for gravity drainage. Pumps, filters, flow meters, gauges, tanks, and
valves may also require special fittings or removal for cold-temperature protection.
Pumps and electrical panels should be protected from exposure to the sun by covers or
shading to reduce heating and maximize their useful life.
A lack of understanding of the fundamental benefits and limitations of microirriga-
tion has resulted in many systems that are unintentionally under-designed. The most
common signs of an under-designed system are the inability to fully provide for the
water needs of the crop during peak water use periods and inadequate line flushing
velocities. Operational flexibility may also be limited by a poor design.
External factors such as soil salinity, soil hydraulics, crop sensitivity, water quan-
tity, water quality, and any environmental concerns must be addressed from the begin-
ning of the design process. These concerns will guide the selection of the tubing, emit-
ters, and emitter spacing.
17.3.2 Field and Crop Considerations
Microirrigation distributes water directly into the root zone of crops, so the selec-
tion of emitters must consider the rooting characteristics of the crop, the expected vol-
ume of soil to be wetted in the field, the total amount of water to be applied, and the
estimated total allowable time per irrigation per day. The selection of emitters and
spacing will be based on the maximum application amounts, estimated irrigation
times, water supply considerations, and hydraulic capacities.
Perennial crops may require one to five emitter laterals per plant row to adequately
supply water needs depending on soil types, water emission device, size of plants and
climate. Established, widely spaced plantings, such as pecan trees, should have at least
two lines, 2 to 3 m on either side of the row. More closely spaced perennial crops,
such as asparagus, grapes, and hops, may need only one lateral per row or bed. Plant-
ings in humid regions may require more laterals and microsprinklers due to the exten-
sive rooting systems that are stimulated by recurrent rainfall to ensure adequate deliv-
ery of water and nutrients to the roots.

Crops using line-source drip tapes, such as row crops and shallow-rooted or closely
spaced perennials, are usually designed based on flow rate per unit lateral length.
Widely spaced permanent crops using tubing with discrete emitters are usually de-
signed using discharge per outlet.
Point-source emitters with non-overlapping wetting patterns, including microsprin-
klers, are usually selected for low-density plantings of trees, vines, or vegetables.
Emitter locations should ensure that the wetting patterns are within the plant’s root
zone and that 25% to 50% of the potential root zone is irrigated. High-density row
crops are generally irrigated by tapes or line-source lateral lines. Line-source systems
should be designed so that the entire root system of high-density crops is within the
wetting pattern.
Microsprinklers may be required to increase the wetted area to maximize soil water
availability and avoid leaching on light, highly permeable soils or areas with shallow
root zones. Widely spaced permanent crops, such as citrus, may require microsprin-
klers to irrigate a relatively large fraction (e.g., > 50%) of the root zone for peak pro-
ductivity, especially in humid areas.
For the same lateral diameter, emitters with higher flow rates will have a larger
pressure variation per lateral length compared to low-flow emitters. High flows per
unit length of tubing will also limit lengths of runs.
It is generally desirable to have the highest discharge rates that meet soil hydraulic
conditions, because higher-flow tapes and emitters have larger orifices and are less
subject to plugging and thus provide higher uniformity and maximum operational
flexibility in scheduling. Buried drip systems are strongly affected by the saturated
hydraulic conductivity of the soil and emitter flow rates are selected depending on
whether the grower wants to be able to wet the soil surface.
17.3.3 System Considerations
System hydraulic capacity for irrigation should be based on peak evapotranspira-
tion demands for the most critical period for a mature planting, usually in the range of
5 to 10 mm per day depending on the crop, climate, and application efficiencies. If
economical and practical, a design should aim to supply about 120% of peak ET to
provide the capacity to catch up in the event of maintenance down times, line break-
age, equipment failures, electrical outages, or other problems.
17.3.3.1 Pipe systems. Main and submain lines utilizing PVC pipe that is not UV-
protected should be buried. It is advisable to keep control valves above ground to fa-
cilitate maintenance and keep submains full of water to minimize system drainage and
decrease startup times. Each block should have isolation valves so that it can be main-
tained without shutting down the entire system. Pipes, fittings, and valves should have
sufficient pressure ratings to withstand waterhammer surges and static pumping heads.
The proper size of mainlines, submains, headers, manifolds, and valves, as well as
operating pressures, may be dictated by flushing requirements. Most systems will not
have sufficient hydraulic capacity to flush the entire system at once. More often, the
system will be flushed in zones, with other zones shut off so that sufficient pressure
and flow will be available to flush each zone.
Even small changes in elevation at the low pressures common to microirrigation
can cause large flow variations. Pressures can be managed by proper pipe sizing, spe-
cial valving, and/or carefully controlling elevation differences within blocks. Pressure
regulation must be designed to ensure uniform water distribution to various parts of

the microirrigation system and to reduce fitting failure due to excess pressure build-
ups. Non-pressure-compensating emitters and tapes must operate at consistent inlet
pressures and be properly sized with appropriate run lengths and lateral diameters
while considering field slope to optimize application uniformities. Inlet pressure regu-
lation with very low-pressure systems (e.g., <50 kPa) are often adjusted by varying the
length and diameter of the tubing connecting the tape to the manifold. As a minimum,
pressure gauges should be placed at the pump discharge, at both sides of the filters,
and at the entrance to each zone.
The maximum allowable flow velocities in main and submain pipes should nor-
mally be less than 1.5 m/s at the maximum expected flow rates to avoid excessive fric-
tion losses and surge problems. When unconstrained pipe fittings are used, thrust
blocks should be used at all elbows, tees, control valves, dead ends, etc.
Lateral lines should always run across or down slope to obtain high uniformity and
ensure adequate flushing velocities to minimize plugging problems. They should not
run up slope if possible because particulates tend to settle at the lower elevations and
cannot be easily flushed. Likewise, lateral-flushing requirements mandate that emitter
lines be supplied water from only one direction because it is desirable to have particu-
lates accumulate at the distal ends of the laterals for flushing.
Systems of pressure-compensating emitters generally follow the same guidelines as
systems of non-compensating emitters. These are mostly used when topographic con-
ditions cause significant pressure differences due to elevation changes. Pressure-
compensating emitters are also used when lateral lengths or elevations are such that
pressure losses from friction and other losses are too high for uniform flow rates from
emitters. In this case, line pressures can be boosted to match the operating range for
the pressure-compensating emitters to ensure more uniform flows. These devices are
not always economical and their use needs to be evaluated on a case-by-case basis.
17.3.3.2 Filtration. A well-designed and well-managed filtration system is critical
for every microirrigation installation. It must be supported by appropriate chemical
and biological water treatments (Bucks et al., 1979). These systems are required be-
cause open ditches, canals, and ponds generally contain organic contaminants and
particulate matter. Pumps on wells may also introduce sand particles into the water
stream. Suspended solid concentrations greater than 100 mg L-1 will require extraordi-
nary filtration systems. These contaminants must be removed before entering the irri-
gation system. Filter operation requires pressures in excess of the normal system op-
erational pressures, which reduces potential energy conservation benefits of low-
pressure systems. Specific filtration devices and concerns are discussed in Section
17.8.2, and in much more detail in Nakayama et al. (2007).
17.3.4 Tubing Selection and Spacing
Tubing should be selected based on the proposed economic life of the installation
as well as cultural practices, harvesting equipment, and environmental conditions. For
example, buried installations that will be subjected to direct equipment traffic should
specify heavy-walled hose rather than thin-walled tape products. Highly variable to-
pography or extraordinary long runs may require large-diameter hose with high burst-
ing strengths.
The lengths of run will depend on the specific hydraulic properties of each tubing
type, emitter flow requirements, and field slope. To maintain high uniformities, lateral
lengths of low-flow, low-pressure, thin-walled (e.g., <0.25 mm thickness) tapes should

normally not exceed 200 m. Thin-walled, high-flow tapes are normally less than
150 m in length. Large-diameter, thicker-walled (>0.25 mm thickness) hoses and tapes
can sometimes be as long as 400 m or more. Pressure-compensating emitters may be
required for small-diameter tubing, high operating pressures, long runs or steep slopes.
Placement of tubing with respect to plants depends on expected cultural operations,
emitter spacing, emitter flow rate, size of the root zone, and soil hydraulic characteris-
tics. Typically, row crops have the tubing placed within 0.05 to 0.10 m of the plant for
seed germination and tillage. Placement on permanent crops can vary from 0.1 to 2 m
although most are within 0.5 m of the plant row. Tubing can be moved as the plant
matures. Tubing can also be suspended above the ground on trellised crops allowing
for easier maintenance, weed control, and less damage by mechanical operations.
The spacing between drip lines depends on the crop being irrigated, the lateral ex-
tent of the crop root zone, and water redistribution in the soil. Installations in humid
areas may require that a larger portion of the root zone be wetted to match crop root-
ing characteristics and increase uptake efficiency of water and nutrients. Widely
spaced trees and vines typically have one or two dripper lines per crop row. Sandy
soils with little lateral movement as well as areas susceptible to wind and water soil
erosion may require several drip lines or the use of microsprinklers to cover a larger
portion of the root zone.
17.3.5 Emission Device Selection and Spacing
The selection of the specific water-emission device determines many of the opera-
tional characteristics of a microirrigation system. For example, labyrinth turbulent-
flow emitters are commonly used because they have an equivalent hydraulic diameter
of 0.75 mm and, thus, a low plugging potential.
Emitter spacing should be selected to irrigate a sufficient root zone volume to pro-
vide for the water needs of the crop. Low-density plantings, such as tree crops, may
have several emitters per tree, but emitter spacing can be flexible as long as an ade-
quate volume of root zone is irrigated (e.g., >25% in arid regions, >50% in humid re-
gions). For high-density plantings such as row crops, the root systems merge along the
row, and closely spaced emitters or line-source systems should be used to apply water
uniformly along the row length.
17.3.6 Emitter Plugging Potential
The characteristic low application rates, low pressures, and small orifice openings
that are unique to microirrigation systems also create emitter plugging problems. Se-
lection of the proper type and size of emitters will reduce the potential for plugging,
although all emitter types can completely or partially plug. Less plugging will occur
when irrigation water is applied with proper filtration and water treatment.
When a number of emitters are considered as a unit, such as several emitters
grouped to irrigate a single tree, the uniformity of water application will be much im-
proved as compared to using a single emitter per tree (Wu et al., 1988b). System vari-
ability can be controlled within 10% when plugging is zero and at least two emitters
are used in a group.
The effect of plugging on water application uniformity in the field can be mini-
mized by closer emitter spacing or emitter grouping (Bralts et al., 1987b). A study of
contiguous, random plugging (Wu et al., 1991) showed that even with as much as 20%
to 30% total plugging, only 1% of the plugging consisted of four to five consecutively
plugged emitters. When the coefficients of hydraulic and manufacturer’s variation are
both less than 10% and emitters are placed in groups of four per tree, 10% to 20%
random plugging will still maintain an overall coefficient of variation (CV) of 17% to
25%, respectively (Wu, 1993a). For high-density row crops when the coefficients of
hydraulic and manufacturer’s variation are less than 10% and the emitter spacing is
half of the wetted diameter, a 10% to 20% plugging will produce an overall CV of
about 20% to 30%, respectively.
A superposition technique (Wu et al., 1989) was used to evaluate the spatial uni-
formity along the lateral by adding soil-water patterns from all emitters at various
specified spacings. This work showed that plugging, followed by emitter spacing,
were the most significant factors affecting the spatial uniformity. Wu (1993a) showed
that emitter grouping was as significant as spacing. Both the hydraulic design and
manufacturer’s variation were less significant than plugging, grouping, and emitter
spacing as long as their individual coefficients of variation were less than 10%.
Neither the soil-wetting patterns nor the plugging distributions were highly signifi-
cant when soil-wetting patterns overlapped by 50% along a lateral. When the hydrau-
lic design of a drip irrigation system is based on a 20% emitter flow variation, a 90%
and 70% uniformity coefficient can be achieved for 0% and 20% plugging, respec-
tively, as long as the emitter spacing is designed for 0.5 of its wetted soil diameter
(Wu, 1993b).
17.4 HYDRAULICS OF EMITTERS
AND EMITTER DESIGN VARIATION
The basic relationship between emitter flow rate and pressure is given as:
q = chx (17.1)
where q = flow rate
c = emitter discharge coefficient
h = pressure head
x = emitter discharge exponent.
The x value is used in emitter selection because it characterizes the flow type in the
emitter and varies from 1 to near zero. When x = 1 the emitter is a laminar flow emit-
ter (capillary) whereas x will have a value around 0.85 for microtubes, 0.65 for long or
spiral path emitters, 0.5 for fully turbulent flow emitters, about 0.4 for a vortex emit-
ter, and near zero for a fully pressure-compensating emitter. For a given hydraulic
variation, less flow variation will occur with turbulent flow emitters than laminar flow
emitters. When the hydraulic design is based on a 20% pressure variation in the mi-
croirrigation system, fully turbulent flow emitters (x = 0.5) will produce only about
10% emitter flow variation while the laminar flow emitters (x = 1.0) will produce 20%
emitter flow variation.
17.4.1 Hydraulics of Orifice Emitters
The discharge rate from an orifice or a short-length nozzle is determined by the hy-
draulic pressure inside the lateral line at the orifice and the orifice dimensions. When
the flow path is fixed and the flow cross-section area is constant, the emitter flow rate
will be affected by only the hydraulic pressure.
The flow rate from an orifice or nozzle type emitter can be theoretically expressed as:
q = c1a 2 gh (17.2)
where q = emitter flow rate

c1 = discharge coefficient
a = cross-sectional area of flow
g = acceleration due to gravity
h = pressure head at the base of the emitter.
When the cross-sectional area is constant, Equation 17.2 can be rearranged as a simple
power function similar to Equation 17.1:
q = C h0.5 (17.3)
where C is the emitter discharge coefficient and equal to c1a 2 g .
17.4.2 Hydraulics of Microtube Emitters
A microtube is a small pipe. The hydraulics of microtubes is the same as for pipe
flow. Thus, the equation for energy drop by friction in the microtube can be expressed
by a simple power function:
qm
h=K n L (17.4)
D
where h = head loss due to friction and is also the pressure head at the inlet of a microtube
K = emitter discharge coefficient
q = emitter flow
m = flow rate (q) exponent (m is 1 for laminar flow, 2 for complete turbulent
flow and 1.75 for turbulent flow in a smooth pipe)
D = inside diameter
n = diameter exponent
L = length of microtube.
Microtube discharge can be determined by rearranging Equation 17.4 as:
1
⎛ n ⎞m
q=⎜ D h⎟ (17.5)
⎝K L ⎠
For a given microtube emitter in which the length, L, and diameter, D, are fixed, the
emitter flow and hydraulic pressure head is often presented as a simple power function
(Equation 17.1) where C is a coefficient and is a constant, x = 1/m and is 0.5 for turbu-
lent flow and 1 for laminar flow. Depending on turbulent flow conditions, x will be
between 0.5 and 1.0.
17.4.3 Hydraulics of Long-Path and Labyrinth Emitters
The relationship between flow rate, q, and pressure head, h, can be expressed by the
same relationship given above for long-path (including spiral-path) and turbulent-flow
emitters. The two power function coefficients, C and x for individual long-path and
labyrinth emitter, are determined by hydraulic laboratory testing. Values for x will
range between 0.5 and 1.0 but are typically 0.65 to 0.85.
17.4.4 Hydraulics of Pressure-Compensating Emitters
When an emitter is designed so that the cross-section area decreases with respect to
pressure,
a = b h− y (17.6)
where a is the cross-sectional area of the emitter flow path, a nozzle, or microtube; b
and y are coefficients. By introducing Equation 17.6 into Equation 17.2 and rearrang-
ing, the emitter flow and water pressure can be shown as:
q = C2 h0.5− y (17.7)
where C2 is a coefficient. Equation 17.7 shows that the exponent, 0.5 – y, can be
smaller than 0.5, which indicates a reduced effect of pressure head on emitter flow as
compared to a turbulent flow emitter. If the y value is 0.5, the exponent will be zero
and no changes in emitter flow will occur. When this occurs, an increase of water
pressure will cause a decrease in cross-sectional area of flow that exactly compensates
for the increase in pressure, and the emitter will be fully pressure compensating. How-
ever, if y is greater than 0.5, the exponent will be negative and flow rate will decrease
with increasing pressure. The concept of pressure compensation can also be applied to
microtube emitters in a similar manner.
17.4.5 Emitter Flow Variation
Microirrigation is characterized by frequent water applications at low application
rates. Thus, even small variations in the magnitude of the emission device’s flow rate
may cumulatively represent relatively large changes in the total seasonal water appli-
cations. Factors that may affect emitter flow rate include manufacturing variation,
temperature effects, plugging, emitter wear with time, elevation changes, and micro-
sprinkler stake assembly losses.
17.4.5.1 Manufacturer’s flow variation of emitters. The basic emitter flow, q,
and water pressure, h, relationship (Equation 17.1) shows that if there is no pressure
variation in the microirrigation system, all emitter flows should be constant and the
emitter flow variation will be zero. However, in an actual field situation, there will
always be emitter flow differences even under constant water pressure conditions.
This variation is caused by small errors in the manufacturing process that result in
flow differences from one emitter to the next. Any deviation in the flow passage area
or shape from a standard size will cause emitter flow variation.
The manufacturer’s variation is the variation in emitter flows from a random sam-
ple of emitters operated at the same pressure, and is expressed statistically as the coef-
ficient of variation of emitter flow, CVM, which is the standard deviation of emitter
flow, S, divided by the mean value of emitter flows, q .
S
CVM = × 100 (17.8)
q
Test results show that the coefficient of variation for microirrigation emitters typi-
cally range from 3% to 20% (Solomon, 1979). Microsprinklers, microsprays, and
minisprinklers usually have low manufacturer’s flow variations of less than 3%, al-
though there can be large differences in the uniformity of water application patterns.
Emitters with CVM values greater than 0.20 are not acceptable for microirrigation sys-
tem design.
17.4.5.2 Temperature effects on emitter flow. The water temperature in a mi-
croirrigation line will be affected by the temperature of the air and soil surrounding the
line. Exposed laterals and water will also be heated by solar radiation. Water tempera-
ture in a lateral line showed a 12°C to 17°C increase in bright sun (Gilad et al., 1968;
Parchomchuk, 1976).
The temperature effect on a microirrigation system can be considered in two parts:

the effect on emitter flow and the effect on lateral-line hydraulics. The first one de-
pends on the design and shape of the emitter and the second one depends on the fric-
tion situation in the line caused by the increase or decrease of viscosity of water due to
temperature changes. The hydraulics of a lateral line determines the pressure distribu-
tion and water movement in the system which, in turn, affects the temperature varia-
tion in the microirrigation system. Emitter flows will be affected by the water tem-
perature at the base of the emitters along the lateral lines.
The effect of temperature on emitter flow depends on the type of emitter (Keller
and Karmeli, 1974; Parchomchuk, 1976; Moser, 1979; Zur and Tal, 1981; Wu and
Phene, 1984; Rodriguez-Sinobas et al., 1999). Temperature effects can be considered
as insignificant when using both orifice and labyrinth-type turbulent-flow emitters
(Wu and Phene, 1984; Rodriguez-Sinobas et al., 1999). Long-path laminar-flow emit-
ters show an increasing emitter flow with increasing temperature. Vortex-type emitters
show a decreasing emitter flow with increasing temperature (Rodriguez-Sinobas et al.,
1999). The relationship between the emitter flow and temperature change can be ex-
pressed as a linear function.
The temperature profile along a lateral line can be shown as a power function
(Solomon, 1984) or a straight line when the temperature of the last point was consid-
ered to be caused by end effect and neglected (Wu and Phene, 1984). A theoretical
evaluation of friction drop along a lateral line with a linear temperature gradient
showed that the shape of the energy gradient line is not affected (Peng et al., 1986).
17.4.6 Emitter Variation
The emitter exponent, x, also affects the relationship between emitter flow varia-
tion, qvar, and pressure variation, hvar, of a microirrigation system. This can be derived
from Equation 17.1 (Wu et al., 1979) as:
qvar = 1 − (1 − hvar ) x (17.9)
where qvar can be simply expressed as the range of variation (Wu and Gitlin, 1974):
qmax − qmin
qvar = (17.10)
qmax
where qmax is the maximum emitter flow and qmin is the minimum emitter flow. The
pressure variation, hvar, is derived in the same fashion as emitter flow variation, but
should be within ±10% of the average emitter pressure.
Equation 17.9 shows that emitter flow variation is zero when x = 0, regardless of
the pressure variation in the system. When x = 1, the emitter flow variation will have
the same variation as pressure variation. This indicates that when the pressure varia-
tion, hvar, is 20%, the emitter flow variation for laminar flow emitters will also be
20%. But, for turbulent flow emitters, x equals 0.5, and a pressure variation of 20% in
a microirrigation system will produce only about 10% emitter flow variation.
17.4.6.1 Combined variation. In the field, emitter flow rate variations are due to
the combination of hydraulic variation and manufacturer’s variation. The relationship
between them was first determined statistically (Bralts et al., 1981; Bralts et al., 1987a)
and then verified by computer simulation (Wu et al., 1985). The total emitter flow
variation caused by both hydraulic and manufacturer’s variation can be expressed by:
(CV ) 2HM = (CV ) 2H + (CV ) 2M (17.11)
where (CV)HM is the coefficient of variation of emitter flows caused by the combined
effects of both hydraulic and manufacturer’s variation; (CV)H and (CV)M are the coef-
ficient of variation of emitter flows caused by hydraulic design and manufacturer’s
variation, respectively.
17.4.6.2 Grouping effects. When a number of consecutive emitters are grouped
and considered as a unit (e.g., several emitters to irrigate a single tree), the uniformity
of water application per tree will be improved (Wu et al., 1988b). The improvement
depends on the magnitude of emitter flow variation caused by the hydraulic and manu-
facturer’s variations. For the case in which the emitter flow variation is caused by hy-
draulics only and the manufacturer’s variation is zero, there will be no grouping effect;
that is,
[(CV ) H ] g = (CV ) H (17.12)
where [(CV)H]g is the coefficient of variation of emitter flow by hydraulics after group-
ing and (CV)H is the coefficient of variation of emitter flow from only hydraulics.
For the case in which emitter flow variation is caused by manufacturer’s variation
only and hydraulic variation is zero, the emitter flows will follow a normal distribution
and the grouping effect will be shown by the relation
(CV ) M
[(CV ) M ] g = (17.13)
N
where [(CV)M]g = coefficient of variation of emitter flow by manufacturer’s variation
after grouping
(CV)M = coefficient of variation of emitter flow by manufacturer’s variation only
N = number of emitters grouped together.
When the emitter flow is affected by both hydraulic and manufacturer’s variations,
the grouping effect can be expressed by the regression equation (Wu et al., 1989),
A
[(CV ) HM ] g = + 1.2487 B − 5.3935 B 2 + 7.6749 B 3 + 2.3113 AB (17.14)
N
(R2 = 0.99)
where [(CV)HM]g = coefficient of variation of emitter flow caused by both hydraulic
and manufacturer’s variations after grouping
A = (CV)M
B = (CV)HM – (CV)M
N = number of emitters grouped together.
Equation 17.14 can be used for up to 17 emitters per group.
17.4.7 Effects of Plugging on Design
A major problem encountered in drip irrigation is the plugging or clogging of emit-
ters. Emitter plugging can adversely affect the rate of water application and the uni-
formity of water distribution. The combined effect of hydraulics, manufacturer’s
variation, and plugging was evaluated statistically (Bralts et al., 1981) and verified
through computer simulation (Wu et al., 1988a). The coefficient of variation of emitter
flow caused by hydraulics, manufacturer’s variation, and plugging can be expressed as:
(CV ) 2HM P
(CV ) 2HMP = + (17.15)
1− P 1− P
where (CV)HMP is the total emitter flow variation affected by all three factors; hydrau-
lics, manufacturer’s variation, and plugging (complete plugging only) and P is the
fraction of emitters completely plugged. For the case where (CV)HM is zero, the coeffi-
cient of variation caused by plugging alone can be expressed simply as a function of P:
P
(CV ) P = (17.16)
1− P
where (CV)P is the coefficient of variation of emitter flow caused by plugging alone.
Equation 17.16 shows that plugging can affect the uniformity tremendously. For ex-
ample, 10% plugging will produce (CV)P of 33% for emitter flow while the ranges of
(CV)H and (CV)M are 0.03 to 0.07 and 0.03 to 0.20, respectively, for the same impact.
Similar to the situation of grouping emitters, an evaluation of contiguous plugging
(Wu et al., 1991) showed that effect of four or more emitters plugged together is less
than 1% for 10% and 20% plugging, respectively. Therefore, if four or more emitters
irrigate a tree, the chances that a tree would receive no water because of plugging are
greatly reduced.
17.5 MICROIRRIGATION DESIGN
17.5.1 Performance-Based Criteria
Design calculations can proceed once the emission device, the required average
emitter flow rate, emitter spacing, emitter variation, allowable pressure losses, and
other criteria are determined. Microirrigation systems are designed based on the uni-
formity of water application with respect to crop needs. Two primary uniformity con-
siderations used for performance based designs are the system emission uniformity
and the spatial uniformity of the irrigation water in the crop root zone. However, the
design procedures are basically the same regardless of the selected uniformity criteria.
Emission uniformity (EU) describes how uniformly the overall system can distrib-
ute water from each emission device in the field and should be designed for at least
80% (90% with chemigation). The design criteria affecting the system emission uni-
formity include hydraulic design, manufacturer’s variation, temperature, plugging, and
the number of emitters per plant. System emission uniformity is usually the most ap-
propriate for design for microirrigation systems designed for widely spaced trees.
Spatial uniformity is a measure of the distribution of the irrigation water in the crop
root zone across the field. The primary design criteria affecting spatial uniformity in-
clude system uniformity, pattern of soil wetting, and emitter spacing. Spatial uniform-
ity is more meaningful than system uniformity for irrigation of high-density plantings
and is often used for designs where emitter wetting patterns overlap.
Other uniformity metrics that may also be useful as design criterion include
Christiansen’s uniformity coefficient, CU, and the coefficient of variation, CV. These
uniformity measures are related as expressed by the following regression equations
(Wu and Irudayaraj, 1987):
CU = 1.0865 CV (r2 = 0.999) (17.17)
CV = – 0.0095 + 0.4288 qvar (r2 = 0.97) (17.18)
2
CU = 1.0085 – 0.3702 qvar (r = 0.97) (17.19)
The high correlation between any pair of the uniformity measures expressed in
Equations 17.17, 17.18, and 17.19 indicates all three uniformity measures can be used
as criteria for hydraulic design. This justifies using the simple uniformity value, qvar,
which is determined by only the maximum and minimum emitter flows for a lateral
line or submain unit.
17.5.2 Emission Uniformity
ASAE Engineering Practice EP405.1 (ASAE, 2005b) defines the emission uni-
formity (EU), also often referred to as distribution uniformity (DU) (Burt and Styles,
1994), of a microirrigation system for design purposes as:
⎡ ⎛ CVM ⎞ ⎛ Qm ⎞⎤
EU = 100 ⎢1 − ⎜⎜1.27 × ⎟⎟ × ⎜⎜ ⎟⎟⎥ (17.20)
⎣⎢ ⎝ n ⎠ ⎝ Qa ⎠⎦⎥
where CVM is the manufacturer’s coefficient of variation for point- or line-source
emitters expressed as a percent. However, it is appropriate to use a combined represen-
tative CV including values presented above as well as other factors such as uneven
drainage (during shutoff) and unequal spacing for initial design calculations because
the actual EU will usually be less than the design EU. Qm is the minimum flow rate at
the minimum pressure in the system, Qa is the average (or design) emitter flow rate at
the average (or design) pressure, and n is the number of emitters per plant or the root-
ing diameter of plants divided by a given length of lateral line (often equals 1). The
first factor represents the flow rate variation due to manufacturing (or combined)
variation, and the second factor, Qm/Qa, expresses the variation resulting from system
pressure changes. It should be noted that for evaluation purposes, EU = 100 (qLQ /qa)
where qLQ is the average measured discharge in the lowest quarter of the measured
field values while qa is the average of all measured values. The correlation coefficient
(CV) is also related to the EU by the following relationship:
CV = 0.77 × (1 – EU) (17.21)
The general recommendation is that the selected combined flow variation and the
flow variation ratio (Equation 17.20) should always result in a design EU above 80%.
However, the actual selection of CV, CU, or DU depends on a number of factors in-
cluding the cost of the system, the cost of water and related costs; the sensitivity of the
crop (yield and quality) to stresses caused by nonuniform irrigation; the market value
of the crop; and environmental concerns (e.g., leaching of agrichemicals to groundwa-
ter). Table 17.1 presents suggested ranges of uniformity values to use in design based
on these factors.
Table 17.1. Suggested range of design criteria for different uniformity expressions
based on various economic, water supply, and environmental conditions.
Design Considerations CV CU EU
Abundant water and no environmental
20%–30% 75%–85% 60%–75%
pollution problems
Abundant water but environmental protection
10%–30% 80%–90% 75%–85%
considerations are important
Limited water resources but no environmental
15%–25% 80%–90% 70%–80%
concerns
Limited water resources combined with the
5%–15% 85%–95% 80%–95%
need for environmental protection
17.5.3 Hydraulic Design of Laterals

The hydraulic design of a microirrigation subunit is based on the energy relations in
the drip tubes, including the friction losses and energy changes due to slopes in the
field (Bralts and Segerlind, 1985). Direct calculations of water pressures along a lat-
eral line or in a subunit can be made by using an energy-gradient line approach (Feng
and Wu, 1990; Wu and Gitlin, 1974). All emitter flows along a lateral line and in a
manifold can be determined based on the corresponding water pressures along the
lateral and manifold pipeline. When all the emitter flows are determined, the emitter
flow variation, qvar, can be expressed by Equation 17.10.
In general, the emitter flow variation, qvar, is used for hydraulic design. The design
criterion for emitter flow variation, qvar, for drip irrigation design is recommended to
be 10% to 20%, which is equivalent to a coefficient of variation from 0.033 to 0.076,
respectively (ASAE Standard EP405.1, ASAE, 2005b). However, if justified, values
from Table 17.1 may be used for design.
Microsprinklers are often designed with orifice sizes over 1 mm in diameter to re-
duce plugging. These emitters can usually achieve system uniformity coefficients in
the field above 90%. However, the individual water distribution patterns of these de-
vices can be quite variable, which would not be evident from a uniformity coefficient
based on flow rates.
17.5.4 Hydraulic Design for Subunits
A microirrigation subunit is a fraction of the microirrigation system than is usually
operated separately from other subunits but may be operated simultaneously with
other subunits. For reasons of economy and water availability, microirrigation systems
are often designed in four or more subunits. Thus, the irrigation pump, power supply,
filtration system, and other water supply components can be smaller than if the entire
production system was irrigated as one unit. However, it is sometimes desirable or
necessary to operate the entire microirrigation system as one unit, such as when mi-
crosprinklers are used for frost/freeze protection (Evans et al., 1988; Evans, 1994), but
this increases capital costs because many of the various system components must be
considerably larger.
A subunit consists of an irrigation manifold (or header) pipeline with laterals that
are supplied water from the manifold. A valve (usually a solenoid valve) is used at the
entrance to the manifold to control water applications to the subunit. A pressure gauge,
flow meter, pressure regulator, and chemical injection port may also be located at the
manifold entrance as needed.
If there are several smaller blocks in a subunit, the design should prevent drainage
from blocks at higher elevation causing excess applications in lower blocks by eleva-
tion control or the use of spring-loaded check valves. This also provides for more
rapid pipe filling and better system uniformity because the piping system does not
have to be recharged for each irrigation.
Water is provided to the subunit by main or submain pipelines that are hydrauli-
cally much different from subunit pipelines. Subunits consist of multiple outlet pipe-
lines with uniformly spaced outlets removing water along their lengths. Conversely,
mainlines and submains have uniform flow along their lengths, leading to greater fric-
tion losses for the same pipe diameters and inflow rates.
Subunit pipelines are designed to meet two criteria: high uniformity and low cost.
However, these criteria often oppose each other because high uniformity is achieved
by uniform pressures and minimal pressure losses that are accomplished by increasing
pipe sizes, additional control valves, and other measures causing higher costs. For this
reason, standards have been developed to define an acceptable degree of uniformity of
water application (e.g., ASAE Standard EP405.1, ASAE, 2005b). These standards
were written with the realization that it is expensive to achieve very high uniformities.
Emission uniformities in the range of 70% to 95% are generally acceptable (Wu and
Irudayaraj, 1992). Higher values may be appropriate on flat surfaces where it is less
costly to achieve higher uniformities, while lower values are more acceptable on
steeply sloping areas unless chemigation is used.
Subunits must be designed considering head losses in both the manifold and lateral
pipelines. Because some subunit head losses occur in both the manifold and the later-
als, it is not appropriate to base the design on flow uniformity only in the laterals
unless pressure is regulated at the entrance to each lateral. These differences may be
especially significant when field slopes are large and the manifold is positioned up-
and-down slope. Subunits can be designed including both the laterals and manifolds
by the following steps:
1. Select the design emission or spatial uniformity based on ASABE Standards or
other sources.
2. Calculate the allowable flow rate variation within the subunit from the emission
uniformity equation.
3. Calculate the allowable pressure variation within the subunit from the emitter
hydraulic characteristics and the allowable flow rate variation.
4. Design the lateral using a fraction (e.g., 60% on first approximation) of the al-
lowable pressure loss within the subunit (which is a proportion of the allowable
whole system pressure losses).
5. Design the manifold using the remaining allowable head loss in the subunit that
was not used in the lateral design.
6. Repeat steps 4 and 5 in a trial-and-error procedure that changes the split between
the lateral and manifold pressures losses until a minimum-cost solution is ob-
tained.
Lateral calculations can be based on the energy gradient line (EGL) or revised en-
ergy gradient line (REGL) method (Wu, 1992a; Wu and Yue, 1993). Design charts
were developed for lateral and submain designs for simple EGL (Wu and Gitlin, 1974)
and REGL (Wu, 1992b) methods. The design criteria for the lateral line is calculated
based on design emission uniformity criteria. Subunits have also been designed using
a finite element approach (Bralts and Segerlind, 1985).
Flow conditions in the laterals and manifold decreases steadily along their length,
but can vary spatially depending on the layout (Howell and Hiler, 1974; Anyoji and
Wu, 1987). A lateral line with hundreds of emitters makes a step-by-step (SBS) calcu-
lation for all sections between emitters very tedious. An energy-gradient line (EGL)
approach was applied for determining pressure variation along the lateral (Wu and
Gitlin, 1974). The concept of energy-gradient line offers a direct calculation of emitter
flows along the lateral line because simple equations can be derived to determine all
emitter flows along a lateral.
There are several good commercial programs and spreadsheets available for com-
puter-aided design of these systems. For example, computer programs have been de-
veloped for microirrigation design using a finite-element approach (Bralts and Seger-
lind, 1985) and step-by-step calculations (Pitts et al., 1986; Meshkat and Warner,
1985). The energy-gradient line concept has also been applied in developing com-
puter-aided design for microirrigation systems (Feng and Wu, 1990).
17.5.5 Main and Submain Pipeline Design
Main and submain pipelines must deliver the necessary amount of water and en-
ergy (pressure) to the entrance of each subunit to meet emission uniformity criteria.
They must be properly pressure-rated to withstand operating plus surge pressures. This
generally requires that pipeline velocities be limited to values that limit surge pres-
sures to acceptable levels (e.g., 1.5 m/s or less).
Thus, main pipelines are designed based on economics using pipe and fittings that
meet the required pressure ratings. A cost analysis of materials and energy use is re-
quired to determine the lowest-cost pipeline for the required flow rate and estimated
hours of operation per analysis period. These analyses are usually made on an annual
basis to amortize initial capital and installation costs and compare them with estimated
annual operating costs.
17.6 DESIGNING THE SYSTEM CONTROL HEAD
The microirrigation system control head is defined as all of the pumps, valving, fil-
ters, injectors, controllers, monitoring equipment, and other facilities required to de-
liver water at sufficient pressures and appropriate quantity and quality to the irrigation
system. The irrigation system control head must be located with convenient access for
maintenance and operation. Figure 17.1 presents a schematic of the various compo-
nents and their placement in a typical microirrigation system control head.
Site preparation should ensure drainage of excess storm water from the control
head area as well as providing reliable access under adverse climatic conditions. The
Figure 17.1 Schematic representation of the all the components typically required in
the system control head for a microirrigation system.
control head should be installed on a level concrete pad of sufficient strength and size
to mount all pumps, filters, flow meters, electrical control panels, valves, injectors,
and other equipment. This provides a stable foundation to which equipment can be
bolted to reduce vibration, avoid structural stresses, and facilitate maintenance. Thrust
blocks may also be required at inlet and outlet pipes. Suitable supports must be pro-
vided beneath heavy components such as flow meters, control valves, and filters.
Protect the installation from accidental mechanical damage by agricultural imple-
ments, vehicles, and tractors. The control head, ponds and settling basins, and chemi-
cal tanks should be fenced to keep children, unauthorized people, or animals from
damaging the components or being exposed to dangerous chemicals.
Drainage and/or spill-containment facilities should be provided around any chemi-
cal or fertilizer supply tanks to prevent direct contamination of any surface water from
spills. Likewise, the wellhead should be protected to prevent contamination of the sub-
surface waters from bacterial and/or chemical sources. Filter backwash or other poten-
tially contaminated water should be disposed of by land spreading (e.g., dust suppres-
sion) and not allowed to flow into drainageways, especially if it contains injected
chemicals.
A suggested checklist for considerations during structural design of the system con-
trol head is:
1. Design the height of the control head installation for convenient dismantling
and assembly of the various components for cleaning, repair, and replacement,
while minimizing the potential for debris or other contaminants to enter the sys-
tem. Components should generally have a minimum height of about 0.4 m
above the concrete surface to provide adequate working space.
2. Maintain appropriate distances between various components to ensure reliable
function of meters and gauges and facilitate operation, maintenance, and clean-
ing of filters, and dismantling and replacement of defective parts. Make sure
that the direction-of-flow arrows on components such as flow meters and check
valves are in the direction of flow.
3. Ensure that the components can be isolated by valves for repair and mainte-
nance work and that sufficient unions, Victaulic couplings, and/or flanges are
installed to facilitate dismantling and repair of components. Locate all valves
for easy access, opening, maintenance, and removal. Avoid directly joining dis-
similar metals without a dielectric union to prevent electrolysis and corrosion of
fittings. The exposed length of threaded steel pipe nipples after assembly
should enable convenient access for a pipe wench.
4. Select resistant materials for all pipes and components that may come in con-
tact with concentrated chemicals, including fertilizers. Special coatings or lin-
ings may sometimes be required to protect hydraulic components from direct
chemical effects.
5. Pressure gauges or pressure-measurement taps should be provided immediately
upstream and downstream of all major components that modify pressure (e.g.,
pressure regulators, filtration devices, fertilizer injectors, pressure sustaining
valves, etc).
6. Electrical and hydraulic interlocks are required for injection equipment to pre-
vent backflows from contaminating water supplies and to prevent chemical in-
jection when the main water-supply pumps are not operating. Likewise, chemi-
cals (except biocides such as chlorine) should not be injected during filter
backwashing and flushing activities. Installation of backflow-prevention system
interlocks, injection-line check valves, and other safety devices must be in ac-
cordance with local standards or regulations.
7. Solenoid valves should have manual-override or hydraulic-bypass capabilities.
They should take 1 to 5 seconds to open and close to avoid waterhammer prob-
lems.
8. Pump-control valves that slowly bring the system online without water hammer
and protect the pumps on shutdown are recommended, especially for turbine
pump installations. Pumps should have low-pressure switches to prevent dam-
age to the pump in case of water loss.
9. Protect workers from electrical hazards by installing and maintaining proper
shielding, interlocks, and by providing adequate grounding of all electrical
equipment. Ground-fault circuits should be provided for all injection pumps.
10. Pump motors or engines should have a cover to shield them from direct
sunlight. This will increase the life of the components and reduce overheating.
11. Electric pump motors should have ammeters installed at the panel and the read-
ings recorded as part of normal regular maintenance record-keeping programs.
17.7 INSTALLATION
One of the most important considerations for installation of a microirrigation sys-
tem is worker safety. Adequate room and suitable topography for equipment to operate
safely without endangering the operator and other workers must be ensured. Trenches
and other excavations deeper than 1 m must be sloped or special protective measures
provided to protect workers from side-wall failures.
The contractor should implement a quality control program during installation to
ensure that all connections are made correctly and avoid entry of soil and debris into
pipes and tubing. All mains and submains should be thoroughly flushed before hook-
ing up tubing. Microirrigation systems should be thoroughly flushed immediately after
installation is completed as well as after any new construction or repairs. The contrac-
tor should ensure that valves, pumps, and filters are properly installed and adjusted.
The contractor should also test the system for proper pressure and flow distribu-
tions and ensure that there are no leaks. The emission uniformity of the new system
should be evaluated to determine if the new system meets design specifications. Addi-
tional guidelines for the installation and post-installation evaluation of new microirri-
gation systems are presented in ASAE Engineering Practice EP405.1, Design and In-
stallation of Microirrigation Systems (ASAE, 2005b).
The irrigator should become familiar with the controls and characteristics of the
new system. A suitable water management program should be implemented and new
cultural and harvesting practices adopted as necessary.
17.8 MAINTENANCE
Implementation of a diligent and rigorous maintenance program is central to the
long-term success of microirrigation. A good maintenance program involves imple-
menting good record keeping, an appropriate chemical water-treatment program and
regular flushing to keep pipelines clean.
Many maintenance problems can be circumvented by consistent records of flow
meters and pressure gauges. This process can be greatly facilitated by remote commu-
nication technologies tied to local computers. Frequent visual inspections are also re-
quired to ensure that all system components are functioning properly. Regular field
inspections will help find emitters that are plugged, identify improperly working flush
valves, and locate pipes and tubing damaged by coyotes, small rodents, insects, and
other causes, including farm equipment.
17.8.1 Field Evaluation of Microirrigation Operation and Uniformity
Water distribution uniformity measurements should be made on newly installed
microirrigation systems to confirm that the system has been properly designed and
installed as well as to provide a basis for later comparisons. Measurements of the uni-
formity of water distribution should be made before each crop season and compared to
the new system evaluations. If this is too labor intensive or is impractical (as may be
the case for subsurface systems), then, as a minimum, the irrigator should compare
actual system flows and pressures at the inlet and distal ends of the system with the
initial evaluations. Additional tests may be required for evaluation and adjustment of
maintenance and operational procedures during the growing season, particularly where
emitter plugging problems are severe.
Decreases in distribution uniformity over time are a cause for concern. Although
regular visual inspections will locate emitters that are completely or almost completely
plugged, they will not identify small changes in emitter flow rates from partial plug-
ging. Frequent examinations of flow-meter records and periodic field measurements of
emitter flow rate and pressure variations will help ascertain changes in system per-
formance. Early identification of problems should indicate the need for special chemi-
cal water treatments to clean partially plugged emitters before the problem becomes
more serious. Subsequent comparisons, where partial emitter plugging may be present
due to chemical precipitation, algae, or other causes, may be made using the
Christiansen uniformity coefficient (CU) or other statistical measures of uniformity
discussed in Chapter 5 and in Pitts et al. (1996). Emitter damage and wear will also
affect flow rates as the emitters age.
If the uniformity is low, additional samples should be taken to improve the statisti-
cal confidence. If the additional samples of emitter flow rate indicate that the distribu-
tion uniformity is still low, pressure distribution tests should be conducted to assist in
identifying the cause(s). Crimped or leaking pipelines and laterals, improperly ad-
justed pressure regulators, improperly sized pipelines or fittings, and valves that fail to
operate properly are all factors that can result in high hydraulic pressure variations.
Conversely, if the hydraulic variation is low, then the poor water distribution uniform-
ity is likely a problem with plugging or incorrect emitter selection.
The water application by the drip irrigation system as well as the water infiltration
in the field can usually be considered as a normal distribution as long as the coefficient
of variation of emitter flow or spatial uniformity is less than 30% (Wu, 1988). The
cumulative frequency distribution of a normal distribution can be approximated by a
straight line. The linear distribution of irrigation application will produce both under-
and overirrigated areas, which can be quantitatively determined by simple mathemati-
cal equations (Karmeli, et al., 1978; Seginer, 1978; Sammis and Wu, 1985; Wu,
1988).
ASAE Engineering Practice EP-458, Field Evaluation of Microirrigation Systems
(ASAE, 2005d), defines general procedures for field emitter evaluations. EP-458 as-
sumes a normal distribution of emitter flow rates measured in the field. However, even
with partial plugging of emitters, the uniformity will usually be overestimated because
plugging and pressures are not normally distributed throughout the system. In general,
criteria for emitter flow variation are: 5% or less, excellent; 5% to 10%, very good;
10% to 15%, fair; 15% to 20%, poor; and greater than 20%, unacceptable. Hydraulic
coefficient of variation criteria are: 10% or less, excellent; 10% to 20%, very good;
20% to 30%, fair; 30% to 40%, poor; and greater than 40%, unacceptable. Bralts et al.
(1987a; 1987b) discussed statistical considerations in the determination of distribution
uniformities. Camp et al. (1997) evaluated several different distribution uniformity
evaluation techniques and discussed the limitations and advantages of each.
Because many of the economic and environmental impacts are functions of climate,
topography, and crop production systems, guidelines for acceptable uniformity often
exist for specific locations. Smajstrla et al. (1997) presented specific steps for field
evaluations in humid regions (also assuming statistically normal emitter-flow distribu-
tions). They presented tabular and graphical procedures that simplify data analysis for
both flow rate and pressure variation to identify nonuniformity problems, to determine
the required number of emitters to test, and to determine the cause of any nonuni-
formity observed. The graphical procedures require that a minimum of 18 emitters be
randomly sampled in each subunit evaluated.
17.8.2 System Maintenance
Manufacturer’s recommendations for maintenance should be followed for all com-
ponents. Each component should be routinely inspected and tested to ensure that it
functions properly. Consistency in all aspects of the maintenance program is the key to
successful microirrigation. Keeping detailed records of irrigation schedules, chlorina-
tion, chemical treatments, chemigation, and maintenance activities is critical to docu-
ment maintenance problems, properly schedule required maintenance, conduct finan-
cial analysis, and plan for future improvements.
Inspection of buried pipelines and equipment is difficult. Therefore, indirect moni-
toring and evaluation by routinely charting flow-meter readings and pressures can be
used to check the performance of subunits or the entire microirrigation system.
Changes in system performance will indicate maintenance needs, even when such
changes occur slowly. Monitoring pressures and flows identifies leaks or emitter-
plugging problems and documents how fast the problems are progressing. For exam-
ple, gradually decreasing flow rate and increasing pressure may indicate gradual emit-
ter plugging, while rapidly increasing flow rate and decreasing pressure can indicate
leaks or broken pipelines. Instrumentation such as tensiometers or other soil-water
sensors can also help call attention to field distribution problems. Periodic calibration
or replacement of flow meters and pressure gauges will also be required.
Mainlines, submains, and laterals should be flushed to remove sediments at least
once each month or as needed during the season, depending on water conditions.
Mainlines should be flushed first, then submains, manifolds, and finally laterals. Sys-
tems can be manually or automatically flushed. The whole system should be flushed at
seasonal startup, at the end of the season, and whenever repairs are made. Flushed
materials should be inspected for signs of chemical precipitations, algal buildups, or
root intrusion.
Monitoring the frequency of primary filter backwashing and pressure drops can di-
rect attention to developing filtration problems. Filtration media should be replaced as
needed. All filters should be manually inspected and cleaned on a regular basis. Ponds,
canals, and settling basins require periodic mechanical or chemical cleaning to control
sediment accumulations, aquatic weeds, and algae.
Weed control with microirrigation can be a challenge because both wet and dry soil
conditions exist over short distances. Widely different weed species, requiring differ-
ent herbicides for control, will inhabit small areas, but most weeds will be in the fringe
areas between the wet and dry soil zones. Fortunately, herbicides labeled for direct
application through emitters tend to work well. However, high soilwater conditions
can cause rapid leaching or degradation of many herbicides. Weeds are often success-
fully controlled using plastic mulches or multiple spray applications of glyphosate or
other herbicide depending on the crop and location. It is also important to keep the
control area free from weeds, brush, vines, or other materials that might block access
or hinder maintenance activities.
As mentioned earlier, coyotes, rodents, and other animals may damage tubing. This
generally occurs when they are looking for water. Daily monitoring of system flow
rates and visual inspections will help reduce resulting water distribution problems. If
these are chronic problems, bitter oils can be periodically injected and strategic place-
ment of water dishes around the field may be beneficial.
Other, less regular, maintenance activities include flushing injection equipment
with clean water after each use for safety and to avoid corrosion. Pumps, filters,
valves, gauges, injectors, tanks, pipelines, and other hydraulic components must be
protected from freezing in winter by removal or draining in cold climates. Insects may
also enter air vents and cause them to leak. Electrical panels need to be kept free of
moisture and dust.
17.9 MANAGEMENT
In general, microirrigation requires higher levels of management than other irriga-
tion methods because decisions must be made daily or more frequently. Specific man-
agement decisions will depend on crop, site, soil, and environmental conditions.
Schwankl et al. (1995) discussed water management of microirrigated tree and vine
crops while Hanson et al. (1994) presented a similar discussion for row crops.
The questions concerning microirrigation management generally center on when to
irrigate, how much to apply, how to accurately evaluate the water status of the plant,
and integration of other cultural activities with irrigation needs. These decisions are
facilitated by adoption of a sound irrigation scheduling program, which may be sup-
ported with automation and monitoring instrumentation. Chemical treatment of water,
filter cleaning, routine flushing of pipelines and laterals, and a good overall mainte-
nance program are also fundamental to good management.
17.9.1 Management in Arid Areas
One of the most important microirrigation management considerations for arid ar-
eas is that active rooting volumes are small because water is often applied to 30% or
less of the total potential rooting area. This can physically limit water and nutrient
uptake causing stress during high ET-demand periods. Thus, management must focus
on optimizing the use of a limited wetted soil volume for both water and nutrients.
Increased sizes of wetted areas by microsprinklers instead of drippers may be required
on sandy soils to improve soil water and nutrient availability.
Reduced wetted volumes compared to other irrigation methods can affect manage-
ment decisions regarding soil salinity and leaching, applied water quality (e.g., salts),
plant nutrition, soil pH, and micronutrient availability. Because of the restricted root
zone, however, fertigation programs work well on a high-frequency basis where nutri-
ents can be applied as needed in small amounts with rapid uptake and minimal leach-
ing, although foliar applications of micronutrients may be required in some cases.
Water table contributions to plant ET may be a significant factor in irrigation water
requirements for microirrigated crops depending on seasonal variations in aquifer
depth. The benefits of maintaining slightly reduced root zone soil water levels for
storage of precipitation are limited because soil water levels are already substantially
reduced outside the wetted root zone areas, thus providing abundant storage for any
precipitation.
17.9.2 Management in Humid Areas
The crop-rooting volume is not limited to the irrigated zone in humid areas. Water
supplies are plentiful and frequent rains allow root development and associated water
and nutrient uptake to occur outside of the irrigated zone. Thus, it is important that the
irrigated rooting volume be large enough to minimize stress because roots are not con-
centrated near emitters. It is generally recommended that a microirrigation system be
designed to irrigate at least 50% of the crop root area in these situations. When drip-
pers are used, low-flow emitters are relatively closely spaced with close lateral spacing
to ensure optimum crop yields.
There are tremendous advantages to microirrigation in humid areas. The cost of
water applications is normally low because of the small amounts of water applied over
a season and the low pressure requirements of these systems, but crop yield and qual-
ity increases can be substantial by avoidance of short-term drought effects. Irrigations
can also be applied without wetting the plant foliage and maximize the time that the
foliage remains dry between rainfall events which greatly reduces the incidence of
foliar plant diseases that require the use of fungicides or other agrichemicals.
Fertigation is highly effective in humid areas; however, leaching of nutrients is of-
ten a significant problem due to both heavy rainfall and overirrigation. Thus, microir-
rigation systems must be properly managed to avoid leaching and the associated prob-
lems of contamination of groundwater or surface water systems. This requires that
both water and chemicals be applied in small doses so that the leachable quantity of
chemicals is limited in anticipation of large rainstorms.
As contrasted to microirrigation management in arid areas, excess water applica-
tions are rarely needed for salinity control in humid regions. Exceptions occur where
very poor-quality irrigation water is used, when very salt-sensitive crops are grown, or
during extended drought periods. The relatively frequent occurrence of large rain-
storms normally provides adequate leaching and soil salinity management in humid
areas. Boman and Parsons (1998) discuss the selection and design of microsprinkler
systems for tree crops in humid regions.
As in arid areas, water table contributions to crop water requirements can be sig-
nificant. The amount will be site specific and must be considered in scheduling irriga-
tion applications. These shallow aquifers are also easily contaminated, and irrigation
schedules must be developed that avoid leaching.
Both surface and subsurface drainage systems are often required in humid areas,
especially on heavy soils or light soils in flat areas with restrictive layers that perch
water tables near the soil surface. These drainage needs are reduced but not eliminated
by microirrigation because of numerous high rainfall events.
There is little benefit from delaying or reducing irrigation applications in anticipa-

tion of rainfall, thus full irrigations are normally applied for most crops to optimize
yield and quality. The primary exception to this practice is microspray or microsprin-
kler irrigation of tree crops where the irrigated volume of soil is normally large
enough to significantly increase effective rainfall by delaying or reducing the amount
of irrigation.
17.9.3 Controlled Root Volumes
Efficient root concentration within a limited wetted soil volume is readily achiev-
able with microirrigation in arid areas. Benefits of maintaining concentrated root sys-
tems under an emitter may be: (1) improved water availability due to the reduced im-
portance of soil hydraulic conductivity; (2) efficient application of water by minimiz-
ing losses due to evaporation and deep percolation; (3) efficient application of fertiliz-
ers and other water-soluble chemicals, particularly those which tend to be fixed by the
soil particles (e.g., potassium and phosphorus); and (4) inducing physiological root
restriction or drought effects on perennial plants to cause reduced vegetative-to-
reproductive growth ratios and better light penetration into the canopy. There are basi-
cally three variations of controlled root zone strategies, including regulated-deficit
irrigation, controlled-deficit irrigation, and partial root zone drying. These are dis-
cussed below.
17.9.3.1 Regulated-deficit irrigation. One controlled root-volume technique is
regulated-deficit irrigation (RDI), which is limited to relatively arid areas. This tech-
nique deliberately imposes specific plant water stresses during specific growth stages
(usually early in the season) using daily irrigations but only replacing 10% to 30% of
the plant’s daily water use. The wetted soil volume contracts from the sides and bot-
tom of the root zone. At the end of the stress period (as indicated by various physio-
logical markers), water application amounts are increased (e.g., up to 85% to 100%
daily actual evapotranspiration), but soil water profiles are not refilled and the size of
the small wetted soil volume remains constant. Vegetative growth must not be reiniti-
ated by excess soil water conditions. RDI requires that adequate allocations of
late-season water be available to “finish the crop” and that the system be designed to
apply at least peak crop water use on a daily basis throughout the entire growing sea-
son. Automated microirrigation is highly desirable.
To date, RDI has only been investigated on perennial crops. Research in Australia
on peaches (Chalmers et al., 1981) and pears (Mitchell et al., 1984), Washington on
apples (Proebsting et al., 1977; Middleton et al., 1981; Peretz et al., 1984; Evans et al.,
1993, Ebel et al., 1995; Drake and Evans, 1997) and grapes (Evans et al., 1990; Wam-
ple, 1996, 1997) have produced beneficial responses. Additional work in California,
Israel, Australia, Chile, and other arid locations on several crops has also shown that
carefully managing the severity and duration of a uniform, constant level of water
stress on fruit trees, wine grapes, and some other perennial crops can be advantageous.
RDI has been found to control vegetative growth, increase fruiting, advance fruit ma-
turity, increase precocity, and increase soluble solids in fruits. Annual water diversions
can be reduced by 20% or more. The key to successful RDI is rigid control of soil wa-
ter volumes to control vegetative growth. It is made possible by the practical ability to
achieve high-frequency irrigation regimes and the capacity to carefully restrict soil
water by controlling the application amount and the size of the wetted volume of soil
available to the roots.
17.9.3.2 Controlled-deficit irrigation. Controlled-deficit irrigation (CDI) gener-

ally refers to irrigation strategies that apply less than actual water use late in the grow-
ing season. For example, CDI is often used as a water conservation technique with
perennial crops in arid areas, such as peaches, plums, or cherries, which are harvested
in early to mid-summer Similarly, carefully managed CDI may also be used to induce
a plant physiological response, such as inducing winter hardiness of perennial crops
that are not physiologically adapted for growth in arid areas with cold winters.
Irrigations are maintained at a reduced level after harvest for the remainder of the
season; however, drought stress is not allowed to reach severe levels that would affect
next year’s crop. The saved water is then used for other crops.
17.9.3.3 Partial root zone drying. Partial root zone drying is a simple drip irriga-
tion technique on perennial tree and vine crops that utilizes drip lines located near the
middle of the alley between plant rows (Leib et al., 2006). The systems are used in a
manner similar to RDI in terms of timing; however, only one drip is irrigated at a time
allowing the soil volume covered by the line on the other side of the plant to dry out.
The next irrigation will apply water through the second drip line allowing the other
side to dry to low soil water levels. Physiological responses are similar to those re-
ported for RDI. This practice is commonly used on European (Vitis vinifera) wine
grapes in south-central Washington, west-central Idaho, western Colorado, and north-
central Oregon.
17.10 SCHEDULING MICROIRRIGATION
The basic philosophy of microirrigation is to replace water in the root zone in small
increments as it is used by a plant at intervals ranging from several times a day to once
every two to three days, rather than refilling a much larger soil water reservoir after
several days or weeks. Thus, the major concern for scheduling microirrigation systems
is how much to apply during an irrigation because the irrigation interval is often fixed
by other factors.
The estimated crop water use (ET), combined with the percent of the area that is ir-
rigated, will determine the total volume of water to be applied (Clark, 1992). The
maximum interval between irrigations is primarily controlled by soil hydraulic charac-
teristics, soil profile layering, and tubing placement. The depth of soil, saturated hy-
draulic conductivities, and soil water-holding capacities will control the volume ap-
plied in a single irrigation to avoid runoff or excessive deep percolation.
It is sometimes not possible to achieve optimum irrigation schedules because of ir-
rigation system limitations. These may include inflexibility in controls and instrumen-
tation, inadequate system hydraulic capacities (including fill times and system drain-
age), and the quantity and quality of available water throughout the season.
Management considerations such as the quality and quantity of available labor can
affect the ability to implement scheduled irrigations. Likewise, timing, amount, and
label requirements for chemigation may influence irrigation timing and depth of appli-
cation that can affect prior as well as subsequent irrigation schedules. Excess applica-
tions may have to be periodically scheduled to leach salts. Irrigation schedules may
also have to be adjusted because of other cultural or harvesting considerations.
Once the above factors are considered, irrigations can be scheduled whenever an
estimated allowable depletion level has occurred, or to replace estimated or measured
ET each day. Alternatively, irrigations can be automatically initiated and stopped
whenever soil water matric potentials at selected points in the wetted soil volume
reach predetermined levels as measured by soil water sensors. An economically opti-

mal microirrigation schedule (Wu, 1995) can also be developed based on irrigation
system uniformity and costs associated with supplying water, value of the yield, and
the costs related to groundwater contamination due to seepage.
17.10.1 ET Estimation
Irrigation scheduling must be a dynamic process because ET will vary spatially and
temporally during the growing season. ET depends on the plant, soil, local environ-
mental conditions, the percentage of the root zone irrigated, planting density, rooting
characteristics, and canopy size. Pest and disease problems also reduce crop water use,
and are usually variably distributed across a field, creating opportunities for localized
excess applications and leaching. Likewise, nutrient availability and the plant’s uptake
can strongly influence canopy growth affecting the total and the spatial distribution of
water use. The calculation of ET is covered in Chapter 8.
Rainfall will lessen the irrigation water requirements by the effective rainfall
amount (Kopec et al., 1984). The contributions to ET from shallow water tables will
likewise reduce irrigation requirements. The most reliable ET estimates are based on
field irrigation experiments conducted with a wide range of irrigation treatments (e.g.,
Doorenbos and Kassam, 1979), lysimetry, or determined by calibrated ET models us-
ing weather variables (Jensen et al., 1990).
ET under nonstressed conditions, such as those commonly found under microirriga-
tion, may be higher than previously determined values which were developed under
more traditional forms of irrigation. Conversely, total ET from widely spaced tree and
vine crops may be reduced because soil evaporation and transpiration from grass un-
der the crop canopy are less. Any deviations in estimating ET different from actual can
result in reduced yields, wasted water, undesirable plant physiological responses, or a
combination of these and other factors. Consequently, these uncertainties require that
soil water or plant water potentials be monitored under all microirrigation methods for
proper irrigation scheduling.
The irrigated area, in general, is taken as the total surface area for row crops and
other high-density plantings, considering that eventually most of the area is shaded
when the crop matures. However, for low-density or very young plantings with small
root zones, water applications and schedules should use projected canopy area or other
measures of the affected cropped area.
17.10.2 Irrigation Frequency
More than one water application per day may be required because either the total
actual daily evapotranspiration cannot be stored in the limited wetted root zone vol-
ume (e.g., small vegetable crops) or single large water applications at a point may
cause excessive deep percolation losses and leaching. Conversely, on heavier soils
with high water-holding capacities or poor drainage, optimal irrigations might be only
every second or third day.
Some crops (e.g., cantaloupe, cotton, and many perennial crops) may perform bet-
ter with less frequent irrigations (e.g., every 2 to 4 days) especially on heavy soils,
whereas more water-sensitive crops (e.g., tomatoes, watermelons, lettuce) may need at
least daily irrigations for best yields and quality.
If leaching of fertilizers or other chemicals is a major concern, then sensor-
controlled, ultra-high frequency (e.g., 8 to 10 times/day) pulsed irrigation systems may
be an option. Shallow-rooted crops often benefit from light, high-frequency irriga-
tions. High-frequency microirrigation techniques have been shown to increase lateral

wetting and reduce water and nutrient stresses, especially when used with fertigation,
but the small wetted soil volume will reduce the plant’s ability to endure a drought
stress of even short duration. However, growers need to be aware of management and
soil water problems caused by frequent filling and localized draining of lines under
high-frequency strategies.
On the other hand, less-frequent irrigations may be beneficial to manage so that
humidity levels in dense crop canopies are lowered to reduce incidence of fungal dis-
eases (e.g., Botrytis on grapes). Irrigation deficits may be intentionally imposed at
times to attain certain desirable crop quality or other plant physiological responses.
17.10.3 Monitoring Soil Water
Soil water sensors make point or small-volume measurements in a field to monitor
soil status and to control irrigations. All soil water monitoring devices should be
placed at appropriate depths and locations to ensure that irrigation scheduling will be
appropriate to optimize yields, minimize water usage, and minimize leaching to the
groundwater. However, microirrigated soil water distributions are highly variable and
there are major questions on determining suitable sensor locations and the correct in-
terpretation of the readings. Preferential flow of soil water is often a major, but largely
unquantifiable, factor in soil water distributions. Consequently, microirrigation sched-
uling is often “calibrated” to particular sensor placements with respect to a water-
emission point that is correlated with plant water potential measurements or other in-
dependent variables. Calibration is typically required to optimize both water and nutri-
ent utilization (Smajstrla and Locascio, 1996). The number of required sensors can be
minimized by choosing representative plants and soil types across a field.
Optimal sensor location will also be influenced by irrigation interval because a soil
water gradient will develop from the emitter to the perimeter of the wetted volume
during irrigation. This gradient decreases after irrigation due to water redistribution
and the wetted soil volume approaches relatively uniform water content. Thus, sensors
to control daily or more frequent irrigations are generally located within 10 to 15 cm
of the emitters but may be located further away for less-frequent irrigations. Electronic
soil matric potential sensors are often appropriate for these applications.
Sensors that determine when to irrigate are normally placed in the upper one-fourth
to one-half of the root zone within the most active areas of water and nutrient uptake.
Sensors located in the lower portion of the root zone can be used to control the amount
of water applied and avoid excessive applications.
17.10.4 Scheduling Criteria for Design
Microirrigation systems are commonly designed for a 90% uniformity coefficient,
especially when agrichemicals are to be injected through the system. Even with a 10%
emitter plugging the uniformity will still be greater than 70%. Consequently, a field
“scheduling” efficiency of 80%, which includes effects of emitter and hydraulic non-
uniformities, is often used. The efficiency relationships between gross and net applica-
tions are discussed in Chapters 5 and 21 and will not be expanded here.
17.11 PLUGGING OF MICROIRRIGATION SYSTEMS
Partial or total plugging of emitters is a chronic problem and the most serious con-
straint to the long-term operation of any microirrigation system. Inadequate considera-
tion of the physical, biological, and chemical characteristics of the water supply will
result in serious plugging problems. The most critical design factors affecting plug-
ging are emitter design, filtration, and the chemical water treatment system. System
operation and maintenance, including inadequate flushing of pipelines, will have ma-
jor effects on microirrigation plugging problems. Improper installation, such as plac-
ing tapes with emission points facing down, may also contribute to plugging problems
due to sediment accumulations along the bottom of the tape.
17.11.1 Causes of Plugging
Plugging of microirrigation systems may occur from single or multiple factors.
Physical factors such as suspended colloidal clays, silts and other materials passing
through filters, broken pipes, root intrusion, and aspiration of soil particles into the
emitter orifices are common physical causes of plugging. Chemical factors such as
precipitation of carbonates and iron oxides, and precipitates from chemical injections
are also significant causes of emitter plugging. Likewise, organic and biological fac-
tors such as oils, algae, aquatic weeds, bacterial slimes, fungi, as well as spiders, in-
sects, worms, fish, frogs, snails, clams, and their eggs or larva can be major contribu-
tors. Low system pressures and flow rates will exacerbate plugging problems.
17.11.2 Sediment
Routine flushing of pipelines is required to prevent emitter plugging from the grad-
ual accumulation of particles which are too small to be filtered, but which settle out or
flocculate at the distal ends of pipelines. Flushing velocities should be about 0.6 m s-1
to ensure transport and discharge particulate matter from the pipelines. This requires
flow rates at the ends of the lines of about 0.12 L s-1 for 15-mm tubing and 0.22 L s-1
for 22-mm tubing. Flushing frequency should be at least once a month, but will vary
through the season depending on the rate debris and particulates accumulate. Applying
surfactants or dispersing agents such as sodium hexametaphosphate through the mi-
croirrigation system may reduce some plugging problems by preventing the floccula-
tion of silts and colloidal clays, allowing them to easily pass through the emitters or be
flushed from pipelines.
Automated flush valves are sometimes used at the ends of the laterals to help flush
fine particulates at the start of each irrigation; however, periodic manual flushing is
still required. Use of these valves is generally not recommended because they tend to
leak and waste water, requiring extra maintenance in addition to the added purchase
cost. Use of these valves with chemigation may also be problematic due to leaks and
the potential for chemicals to accumulate in these locations.
17.11.3 Algae and Bacterial Slimes
Chlorine injection is the most common and least expensive method to prevent
clogging by biological growth (algae, colonial protozoa, sulfur bacteria, and other mu-
cous organisms). Bacteria that precipitate iron, sulfur, and manganese can also be ef-
fectively and economically controlled by chlorine treatments.
Copper sulfate, chlorine, and organosulfur compounds are used to control algae
and/or bacterial slimes in drip systems as well as in ponds or canals. The degree of
control will vary with light and water temperature conditions. Some chemicals such as
quaternary ammonium are effective when algal growth is slow to moderate, but will
fail under conditions of rapid growth. Organic growers may be limited to copper sul-
fate at 100 to 200 mg L-1 concentrations for controlling algae, depending on local
regulations.
Soil bacteria can sometimes be aspirated back into an emitter and produce slimes
that glue small particles together and plug the orifices. These have been fairly success-
fully, but expensively, removed by strong solutions of hydrofluoric acid (1000 mg L-1)
combined with a surfactant (Wuertz, 1992).
17.11.4 Chemical Plugging
Chemical precipitates can cause plugging of irrigation systems (Nakayama and
Bucks, 1986; Hills et al., 1989, Burt and Styles, 1994; Burt et al., 1995). Precipitates
of iron oxides (reddish-brown color), iron sulfides (black color), calcium carbonates
(white color), and manganese oxides (black color) in irrigation water can clog emit-
ters. Changes in water pH, temperature, pressure, dissolved oxygen levels, and injec-
tions of chlorine and other chemicals (e.g., some fertilizers) can induce chemical pre-
cipitations. Groundwater supply systems tend to have greater problems with chemical
precipitates than surface water, and the insoluble precipitates that form may not dis-
solve even after treatments such as lowering the water pH or injection of special sol-
vents .
Concentrations of 0.15 to 0.22 mg L-1 of iron (>2 mg L-1) in irrigation water may be
problematic when water pH exceeds 5. Manganese and iron oxides greater than
2 mg L-1 will need treatment if water pH is 4 or greater. Chlorine injection will cause
oxidation and precipitation of iron and manganese (plus kill any iron and other pre-
cipitating bacteria that are present). The general recommendations are to inject
1 mg L-1 of free chlorine per 0.7 mg L-1 soluble iron or 1.33 mg L-1 free chlorine per
mg L-1 of soluble manganese prior to the filtration system. Calcium and magnesium
problems are best addressed by the injection of acids to maintain a water pH between
6.0 and 6.6. Temporary storage of water in ponds or other open containers with agita-
tion is a recommended pretreatment for elevated levels of iron (>4 mg L-1) and man-
ganese to facilitate more oxidation and precipitation before entering the irrigation sys-
tem.
Plugging by other chemical precipitates can often be reduced by acid treatment to
lower pH and prevent precipitates from forming, avoiding the injection of insoluble
fertilizers or incompatible fertilizer/chemical mixes, and regular flushing of lines. It is
possible to reduce the risk of precipitation problems by carefully injecting various
incompatible chemicals at different locations in a mainline so that they are sufficiently
diluted and mixed before the next chemical is injected. For example, injection points
for acid and chlorine should be a minimum of 1 m apart.
17.12 SUBSURFACE DRIP IRRIGATION
Subsurface drip irrigation (SDI) uses buried lateral pipelines and emitters to apply
water directly in the plant root zone. Laterals are placed deep enough to avoid damage
by normal tillage operations, but sufficiently shallow so that water is redistributed in
the active crop root zone by capillarity. SDI systems must be compatible with the total
farming and cultural systems being used.
Current commercial and grower interest levels indicate that future use of SDI sys-
tems will continue to increase. SDI requires the highest level of management of all
microirrigation systems to avoid remedial maintenance. A poorly designed SDI system
is much less forgiving than an improperly designed surface drip system. Deficiencies
and water distribution problems are difficult and expensive to remedy. Lamm and
Camp (2007) present an excellent, detailed review of SDI.
These systems require safeguards and special operational procedures to prevent

plugging and facilitate maintenance, but they also have numerous advantages.
Jorgenson and Norum (1992) have presented an overview of the theory behind SDI as
well as varied grower experiences and applications. Camp (1998) has prepared an ex-
cellent summary and analysis of published subsurface drip irrigation research results.
Phene et al. (1987a; 1992) and Phene (1995) listed four operational characteristics
of SDI relative to surface drip irrigation installations with respect to water conserva-
tion and salinity. These were:
The top of the soil surface remains dry, limiting surface evaporation to the rate
of vapor diffusion transport and preventing salt accumulations on the surface.
The use of a very high irrigation frequency (several times per day) that matches
actual crop water use will result in a constant wetted soil volume and a net up-
ward hydraulic gradient, which minimizes leaching.
Supplying water and nutrients directly to the root zone allows root uptake to be
more efficient if irrigation and fertilization schedules are appropriate.
Soil crusts, which may impede infiltration and cause ponding and runoff, are
bypassed so that surface infiltration variability becomes insignificant.
Also, Camp et al. (1987) and Grimes et al. (1990) found more uniform soil water dis-
tribution under subsurface than surface drip systems.
Under proper management, properly designed and managed SDI irrigation systems
offer several other advantages to growers because of their potential for:
maintaining access to fields with tillage, planting, spray and harvest equipment
that is not restricted by irrigation;
obtaining better weed suppression with minimal chemicals because there is less
seed germination with dry soil surfaces;
efficiently and safely applying labeled plant-systemic pesticides and soil fumi-
gants for improved disease and pest control;
reducing surface wetting often reduces fungal disease incidence (e.g., molds,
mildews) by maintaining dryer plant surfaces and lower air humidity within the
plant canopy;
reducing pesticide exposures for workers when chemicals are applied below the
soil surface;
implementing minimum tillage, permanent beds, and multiple cropping systems
(Bucks et al., 1981), although much of the necessary equipment modifications
and farming techniques remain to be developed; and
minimizing flow-rate sensitivity to temperature fluctuations because emitters are
buffered by the soil.
Phene et al. (1987a; 1992) and Phene (1995) also listed several disadvantages, in-
cluding:
initial system cost may be high;
potential for rodent damage;
salt may accumulate between drip lines and soil surface;
low upward water movement in coarse-textured soils;
high potential for emitter plugging; and
insufficient technical knowledge, dissemination, and hands-on experience by
growers and researchers.
Specific problems that have been observed include plugging by root intrusion (Tollef-
son, 1988; Bui, 1990); pinching of hoses due to compaction or squeezing by large
roots (Bui, 1990); and rodent and insect damage (Bui, 1990). In addition, fertility man-
agement becomes more critical with SDI because roots tend to grow deeper than with
surface drip systems and some surface applied nutrients may not be sufficiently avail-
able (Phene, 1995).
A broad range of yield increases have been observed under SDI when compared to
surface, sprinkler, and even surface drip irrigation systems from small to up to more
than a 100% difference. SDI research has been reported on crops including cotton
(Plaut et al., 1985; Tollefson, 1988; Constable and Hodson, 1990; Hutmacher et al.,
1995), field corn (Lamm et al., 1995), sweet corn (Bar-Yosef et al., 1989), tomatoes
(Davis et al., 1985; Phene et al., 1987a; Grattan et al., 1988; Bogle et al., 1989), canta-
loupe (Phene et al., 1987b), potatoes (Bisconer, 1987), asparagus (Sterret et al., 1990),
alfalfa (Oron et al., 1989; Bui and Osgood, 1990), cabbage and zucchini (Rubeiz et al.,
1989). Most yield increases have been attributed to better fertilization, better water
management, improved water distribution uniformities, and improved disease and pest
control. Grattan et al. (1988) cited better weed control as the major factor in their ob-
served yield increases. Moore and Fitschen (1990) reported that the conversion of
5900 hectares of sugarcane in Hawaii from furrow to SDI over a period of 12 yr re-
sulted in an average 27% net yield increase.
17.12.1 Design of SDI Systems
Designs should follow the same general requirements as for all microirrigation sys-
tems. However, extra attention to filtration, water treatment, pressure regulation,
proper location of check and air-vacuum relief valves, and flushing are crucial to the
success of SDI systems. As with surface microirrigation systems, injected chemicals
and fertilizers must be compatible with all other injected chemicals, and control of
water pH is critical.
Both tapes and tubing have been used successfully for SDI. Solomon (1992) dis-
cussed trade-offs and selection criteria of emission devices for SDI. A major factor in
the life of these systems is tape wall thickness with thicker-walled tapes generally last-
ing longer. However, even thicker tapes require special considerations and must be
buried deep enough to avoid being disturbed by tillage or harvesting equipment, but
shallow enough to prevent permanent collapsing of the tape by soil weight or cultural
operations. SDI tapes are often used with high-value, shallow-rooted (e.g., strawber-
ries) and annual or biannual crops (e.g., various vegetables, melons, sugar cane, cot-
ton). They are typically placed at shallower depths in semipermanent minitill beds on
land with little slope or short runs because tapes are not pressure-compensating. As a
general rule, there should never be any farm equipment wheel traffic or other activities
directly over the tape that would compact the soil and flatten the tape, permanently
destroying its utility.
Tubing is used more often than tape for SDI on perennial crops, and may be placed
at deeper depths than for annual crops. Pressure-compensating emitters on tubing al-
low SDI systems to be used on diverse topography with steep slopes. Tubing is more
rigid and thus more resistant to pinching and compaction, but the same general opera-
tional considerations for tapes should also be followed.
Individual SDI laterals are often connected with manifolds at both the top and distal
end of the tubing. The bottom manifold typically provides greater flexibility and saves
time with flushing operations. The additional expense also creates a hydraulically
looped system that can supply water from both ends of a lateral. This can be beneficial
over the short term if an individual lateral is pinched by roots or soil compaction
caused by farm equipment. Provisions for adequate air-vacuum relief on both the head
and flushing manifolds is critical to reduce emitter plugging due to soil particles being
pulled into the orifices by vacuum conditions at shutdown.
Wetting the soil surface, or “surfacing,” negates many of the benefits of SDI and is
usually undesirable, especially on permanent crops. It can increase weed growth,
move salts to the surface, increase soil evaporation, and increase fungal disease inci-
dence. Reducing the negative effects of surfacing may require plastic film mulches
and more aggressive water and soil management practices. On the other hand, surfac-
ing is sometimes appropriate to ensure adequate wetting across the full width of a bed
for germination, fumigation, and/or proper growth of shallow-rooted crops (e.g., on-
ions).
Surfacing is primarily the result of water application rates exceeding the saturated
hydraulic conductivity. However, it may also occur if the tubing is placed too shallow
and the surface is wetted by capillarity, or if water from several emitters runs along the
lateral and collects at a point and the low gravity head forces water to the soil surface.
In addition, surfacing may be the end result of chemical processes that effectively re-
duce saturated hydraulic conductivity, including high-bicarbonate waters that precipi-
tate calcium and plug soil pores, application of irrigation water with low electrical
conductivity on saline or sodic soils, or application of water with high sodium levels
(see Chapter 7). Some growers in California have reportedly tried using narrow shanks
ahead of the buried tubing installation machine to till deeper than tubing placement as
a way to discourage surfacing, with mixed results.
Nevertheless, exceeding the effective saturated hydraulic conductivity with SDI
systems causes water to be discharged against higher-than-atmospheric pressures,
forcing the water to the soil surface. Because manufacturers design drippers to apply
water at atmospheric pressure, this back pressure reduces emitter flow rates as much
as 50% (Sadler et al., 1995; Shani et al., 1996; Warrick and Shani, 1996). Thus, sur-
facing negatively affects uniformity, the ability to accurately schedule water applica-
tion amounts, and presents significant design as well as management implications.
The surfacing process often creates light-textured “chimneys” where the fine soil
particles are flushed from around the emitter resulting in a direct, low-pressure path-
way to the soil surface that tends to make the problem worse. These chimneys will
sometimes disappear after a year or two as the soil structure adjusts or pathways along
the tubes disperse, but they usually require tillage for remediation. Short, frequent
pulsing of water applications throughout the day may help reduce surfacing problems.
Low-EC waters may require injection of finely powdered gypsum (e.g., >2.5 meq L-1)
or other amendments into the water to improve infiltration on sodic soils.
To avoid surfacing problems, emitters on SDI tend to have smaller flow rates and
closer spacing than surface drip systems, thus creating a higher potential for plugging.
Consequently, filters should be designed to remove particles as small as 150 to 200
microns. Check valves, vacuum breakers, and air vents must be installed to prevent
backsiphoning and aspiration of soil particles into the emitter orifices at system shut-
down. Designs must ensure that each lateral can be flushed properly.
17.12.1.1 Depth considerations. The ideal depth of buried SDI laterals depends on
discharge rate, emitter spacing, soil type, root distributions of crops in a rotation, water
movement for seed germination (also affected by seed depth and soil salinity), chemi-
gation programs, and the desirability of surface wetting. The tubing must be buried
deep enough so that it is not damaged by equipment, but shallow enough to effectively
move water into the crop root zone. Tubing and tape depths may be adjusted on bed
systems by use of equipment to remove or replace soil.
It is generally recommended that the tubing be placed as shallow as possible with-
out surface wetting because most of the roots, air exchange, and biological activity is
located in the upper soil layers. As a general rule, tubing is placed at shallower depths
on coarse-textured soils and slightly deeper with finer-textured soils. Most SDI sys-
tems are installed at 10 to 50 cm depths. SDI can be installed on shallow-rooted row
crops such as strawberries or onions at 2- to 8-cm depths up to 40 to 50 cm deep for
crops such as cotton, maize, potatoes, or sugar beets. Many vegetable crops use SDI
laterals at 15- to 20-cm depths near the plants. Thin-walled tape products are rarely
buried more than 15 cm below the surface in beds.
There are trade-offs with tube placement depth. Philip (1991) showed for a steady
state subsurface irrigation source that the deeper the source relative to the roots, the
larger the deep percolation and the smaller the soil evaporation. Barth (1995) dis-
cussed placing V-shaped impermeable foil barriers under the SDI lateral to control
downward movement of water and reported good results on sandy soils. Deeper instal-
lations reduce surface weed germination and allow more tillage operations. However,
if laterals are too deep, much of the water is applied below the crop root zone, reduc-
ing biotic activity (critical to nutrient uptake) in the shallow soil layers. In addition,
deep installations may limit crop germination potential and limit use of some surface-
applied chemicals. It may be possible to inject air through deeper lines to encourage
more root development and soil biota as well as aerate waterlogged soils, but this prac-
tice has not been thoroughly investigated. Shallow depths may result in excessive sur-
face soil salinity when using saline waters and may also be more subject to damage
from burrowing rodents and insects. Deeper tubing placements will require more trac-
tor power for installation.
Shallow tube placement may be necessary to supply water near the soil surface on
plantings with limited root zones. For some soils it is possible to wet a seed bed from a
depth of 30 to 50 cm. However, on coarse-textured soils, emitter lines may have to be
within 5 to 20 cm of the surface.
17.12.1.2 SDI on annual row crops. Hanson et al. (1994) discussed installation,
operation, and maintenance of SDI for row crops. Annual row crops can be grown on
temporary or permanent beds with or without plastic mulches However, it is some-
times necessary to use sprinkler irrigation for activating herbicides or germination of
small-seeded crops such as lettuce or onions, particularly under saline soil conditions.
Emitters with high flow rate or tapes with closely spaced emission points (every 20 to
45 cm) that wet the soil surface are common on annual crops. Shallow tapes are gen-
erally expected to last 1 to 2 yr, although some installations of thicker-walled tapes on
permanent beds may last 5 to 15 yr. These require special tillage and cultural practices
(such as controlled traffic and permanent wheel tracks) to avoid damage. To ensure
adequate wetting of the root zone of annual crops and to avoid dry areas due to emitter
plugging, emitters should be spaced close enough together to produce an overlapped,
line-source wetting pattern. Spacing between emitter lines or tapes are often 1 to 1.5 m
but will vary depending on soil type, crop characteristics, and cultural practices.
Use of semi-permanent SDI on annual or biannual row crops requires the estab-
lishment of consistent row spacing for all crops used in the rotation and replanting at
the same locations. If these practices are not followed, the drip lateral position may
vary from being centered under the bed and row to being located under the furrow.
However, research results have demonstrated that yields were not reduced when the
drip tubing was not exactly centered under the bed (Ayars et al., 1995) on cotton. But,
the incidence of mechanical damage increased as the lateral line was moved from the
center towards the edge of the bed. Emerging technologies such as auto-steer on trac-
tors and harvesters will help ensure proper alignment between beds and drip lines year
after year..
17.12.1.3 SDI on perennial crops. SDI systems on perennial tree and vine crops
should have a life expectancy of 7 to 20 yr with appropriate design and maintenance.
Depths of 20 to 40 cm are common with some as deep as 1 m depending on crop root
development patterns and soil type. Generally, low-flow emitters (<4 L h-1) are used to
keep the soil surface dry and to reduce annual weed growth. Schwankl et al. (1995)
presents a detailed discussion on microirrigated tree and vine crops.
17.12.2 Installation
Most SDI systems are installed by tractor-mounted, parabolic-shaped injector
shanks with mole- or bullet-shaped tips to create a cavity for the tape or tubing. Vi-
brating shanks are recommended for SDI installations deeper than 20 cm because less
horsepower is needed, cutting through roots and around rocks is easier, and the cavity
around the tubing tends to quickly backfill. As mentioned earlier, emitters should face
upward to minimize plugging due to particulate accumulations in the bottom of the
lines between flushing events. Kinking as well as excessive stretching of the dripper
lines must be avoided during installation. Lamm et al. (1995) discusses some problems
associated with installation of SDI systems.
Laterals are normally installed first. Headers or manifolds are cross-trenched to a
depth of about 0.3 to 0.5 m deeper than the laterals, cutting through the pre-installed
laterals. The trenches should be as narrow as possible to provide lateral support for the
low-pressure PVC pipes. The mains and submains are often assembled on the surface
and lowered into the trenches. Open trenches provide for easier connections and facili-
tates hydraulic testing of all fittings and laterals by allowing visual inspection to en-
sure proper operation. In addition, leaks can be easily repaired. Any metal fittings or
wire used to make connections should be stainless steel.
Laterals are connected to the manifolds with 6- to 13-mm flexible PVC or polyeth-
ylene tubing. Connections to the dripper tubing or tape must avoid any kinking or
sharp bends. Installers should leave adequate amounts of tubing at the distal ends so
that flushing connections can be made correctly. Laterals should be kept as uniformly
deep and as straight as possible. In addition, submains and headers should be deeper
than the laterals to prevent them from draining, and to facilitate rapid startup and pres-
surizing of the system. Likewise, control valves should be located aboveground for
easy access and to help keep mains and submains full of water. If laterals are con-
nected by manifolds for flushing, properly sized flushing manifolds should be at the
same depth as the laterals.
17.12.3 Root Intrusion
Root intrusion is a major problem with SDI. It can be minimized by chemicals,
emitter design, and water management. Chemical controls include the use of slow-
release herbicides (e.g., triflurilan) embedded into emitters or injected at low rates
(where permitted by the pesticide label) or careful, periodic injections of soil fumi-
gants (e.g., metam sodium) through the water. Other chemical measures include fre-
quent injection of chlorine (e.g., 7 to 8 mg L-1) to discourage root growth near emit-
ters. Injection of high quality phosphoric acid at about 15 mg L-1 on a continuous basis
will reduce root growth by acidifying the soil and also provides some control for al-
gae, but may not affect certain bacteria. Root-intrusion control programs often require
use of several measures.
Emitters that are plugged by roots may sometimes be cleared by increasing pres-
sures coupled with injections of acids that drop the pH to very low levels (e.g., pH <4)
combined with very high doses of chlorine (e.g., 300 to 500 mg L-1) that are left in the
lines for at least 24 hours before flushing to “burn” the roots out. Injection of copper
sulfate (about 15 mg L-1) is sometimes beneficial. But, these practices can damage
certain types of emitters and tubing. In addition, long-term, low-pH water applications
may make some nutrients unavailable and create toxic soil conditions.
Emitter design appears to be a factor in root intrusion. It has been observed that
roots tend to follow the seam in a tape and emitters located on or near the seam are
more subject to root intrusion problems. Some emitters and tapes are constructed with
special physical barriers to root intrusion, but most of these are less successful than
chemical controls. SDI emitters, porous tubes, or tapes that are not chemically treated
and/or have a physical barrier will suffer from root intrusion problems. Smaller ori-
fices tend to discourage root intrusion, but are more easily plugged by other means
(Tollefson, 1988; Bui, 1990).
Deficit irrigation is sometimes desirable to increase quality, yields, or to control
vegetative growth, but its use will increase chances for root intrusion due to increased
root concentrations in the wetted area around the emitters. Very high frequency puls-
ing to frequently saturate the soil volume will discourage root growth of many plants
around the emitter, but roots of some water-loving plants such as celery will actively
grow in saturated soils and can plug emitters.
17.12.4 Flushing
Because SDI systems are buried, it is often convenient to install flushing manifolds
at the distal end of the laterals, especially with closely spaced (e.g., every 30 to 120
cm) lines. It is usually more economical and effective to individually flush laterals
with widely spaced laterals. Flushing design requirements are the same as for surface
systems, but may be compromised by pinching of tubing either by roots or soil com-
paction caused by farm equipment operations.
17.12.5 Salinity Management
Salinity may be a problem with SDI in arid and semiarid areas because any leach-
ing above the tubing occurs only as the result of rain. Thus, salts tend to accumulate in
this area during the season. High salt concentrations exceeding 10 dS m-1 have been
found in the top 6 to 10 cm of the soil profile (Ayars et al., 1995). Salinity distribution
measurements have showed that salts are moved to below the plant row when the lat-
erals are placed under the furrows rather than under the beds (Ayars et al., 1995).
Similar results were reported by Hanson (1993).
17.13 SUBIRRIGATION
A nontraditional use of microirrigation is to improve the efficiency of subirrigation.
With subirrigation, the water table in the field is controlled at a level just below the
crop root zone so that water flows upward into the root zone by capillarity. Because
shallow water tables must be established and maintained, subirrigation can only be
used on relatively flat lands where large water supplies are available, the lateral per-
meability of surface soil layers is high, and uniform restrictive soil layers or perma-
nent shallow water tables already exist. The requirement for large, inexpensive water
supplies generally limits the use of subirrigation to humid regions without salinity
problems.
The water table may be controlled at a constant depth or at a depth that is allowed
to fluctuate as the root zone changes with crop growth. In either event, diurnal water
table fluctuations occur as the water table drops due to peak ET rates, then recovers
overnight or during low ET demand. With conventional subirrigation systems, the
water table in the field is controlled by physically regulating water levels in open field
drainage ditches or controlling the head on subsurface drainage tubing. Open field
ditches are most often used because of their low cost, and, in some areas, subsurface
drain tubes can become plugged by iron ochre or by other chemical precipitates.
In recent years both surface and subsurface drip irrigation systems have been used
for subirrigation water applications in place of the drainage system. Water applied at
high rates by the drip system rapidly percolates to the water table. This excess water
raises the water table across the field somewhat above the bottom of plant root zone,
thereby creating a temporarily high water table that slowly drains away with time.
Advantages of drip systems for water table control are increased uniformity of wa-
ter application distributions and reduced irrigation water requirements. Uniformity is
increased because water is distributed by lateral pipelines which are controlled by the
system hydraulics. Water requirements are reduced because water is placed directly
into the plant beds, some water is supplied by the water table, and the potential for
surface runoff is greatly reduced. However, some runoff may occur because a mini-
mum soil surface gradient is needed to provide adequate drainage of excess rainfall.
Stanley and Clark (1991) reported that irrigation water requirements for tomato
production in south Florida were reduced 33% to 40% with subsurface drip irrigation
as compared to conventional open-ditch subirrigation systems because of reduced run-
off rates. Smajstrla et al. (1995) reported that potato yields were statistically similar
but 33% less water was applied for water table control with a subsurface drip irriga-
tion system compared to a conventional open-ditch subirrigation system.
17.14 MICROIRRIGATION IN NURSERIES
AND GREENHOUSES
Many greenhouse and nursery plants can be effectively and economically irrigated
with several conventional irrigation methods including drip, microsprinklers, misters,
impact and other rotating sprinklers, ebb-flow systems, traveling boom sprays, and
hand watering systems. Ebb-flow systems are used with potted plants in an impervious
basin that is periodically flooded and then drained. Excess water is used for irrigation
of the next basin, thus water is used without waste. Traveling boom spray systems
apply water with spray nozzles on a lateral pipeline that travels on rails mounted
above the plants. The travel speed is adjusted to apply the required amount of water.
This system is primarily applicable to greenhouse production systems and is most effi-
cient on closely spaced containers so that water loss between containers is minimized.
Hand watering, which has low maintenance costs, can be an economical alternative to
microirrigation in some production systems and areas where labor costs are low.
Microirrigation offers a number of advantages for many nursery and greenhouse
production systems, especially reducing water and nutrient requirements as well as
lowering labor costs (Ross, et al., 1990; Ross, 1994). However, microirrigation sys-
tems often have a higher capital cost than conventional irrigation methods, especially
for irrigation of small, closely spaced plants, because of the extensive network of pipe-
lines and large numbers of emitters that are required. These initial high capital costs
can often be offset by savings in operational costs (labor, water, fertilizer, etc.). The
useful life of the system emitters can be 4 to 5 yr with proper maintenance. Except for
accidental damage, the distribution system can last for 10 to 15 yr.
Microirrigation is often the most cost effective and adaptable to large or widely
spaced plants which minimize the number of laterals and emitters required, thus reduc-
ing system cost. Drip systems are regularly used for high-valued plants in containers
as small as 15 cm in diameter. Over-crop sprinkler irrigation cannot be used in some
ornamental plant production systems because of plant disease problems resulting from
frequent wetting of the foliage, or the water quality is such that the plant would be
damaged by direct wetting of leaves, fruits, or stems.
The potential exists for extensive use of drip irrigation for most container-grown
crops, ranging from plants in larger containers (e.g., larger than 4 L) to as small as one
liter, depending on the plant density (pot-to-pot spacing) and the economic value of
the crop. Microsprinkler systems are often preferred for applications under the plant
canopy to crops such as larger ornamental tree and shrubs that are in large beds (e.g.,
balled-and-burlapped plants). Water-soluble fertilizers can be effectively applied
through drip systems directly to the plant and irrigations can be scheduled as necessary
without interfering with other nursery operations. Foliar-applied chemicals with mi-
crosprinklers tend to be more effective than sprinkler applications because the low-
water applications minimize washing the material off the foliage.
Bedding plants and many of the smaller potted flowering plants that are grown in
trays are typically sprinkler irrigated (fixed or moveable boom) or hand watered. The
containers are typically small (less than 2 L) and closely spaced. Microirrigation sys-
tems are not used for bedding plants grown in flats or trays. Potted foliage plants are
often hand watered or irrigated with traveling boom spray systems.
Most woody ornamental field nurseries use drip irrigation, while some use sprin-
kler or microsprinkler systems. Woody container plants are often sprinkler irrigated,
although microirrigation systems are commonly used for plants in containers larger
than 4 L.
Drip irrigation systems are not readily adaptable to plantings of sod, ornamental
fern, or similar crops with continuous lateral root development and canopies that cover
the entire soil surface. In these cases, drip and microsprinkler systems tend to perform
poorly because they are most adaptable to partial root zone or concentrated water ap-
plications rather than the uniform, shallow wetting patterns required for these crops.
Lateral water movement from drip irrigation systems is limited by the hydraulic char-
acteristics of the potting growth medium because of its high porosity and drainage
capability.
Crops with broad, shallow lateral rooting systems are most effectively irrigated
with sprinkler, surface, or subirrigation systems that distribute water uniformly over
the entire surface rather than only partially cover the root zone as is characteristic of
microirrigation. For example, sod is grown using irrigation methods that encourage
shallow root development to facilitate harvest. This is accomplished with sprinkler or
subsurface systems, using frequent, small applications with sprinklers or otherwise
artificially maintaining a high water table. The irrigation method will also affect how
effectively specific crops are harvested and transplanted by various methods and
available equipment.
Some crops, such as ornamental fern, may use over-crop sprinkler irrigation be-
cause it may be necessary for freeze protection. In these cases, sprinklers are also used
for irrigation; however, water use efficiency is poor.
Microirrigation systems in nurseries and greenhouses also require special filtration
and chemical water treatment systems to prevent emitter plugging, like those required
for field crops. Also, the aisles between plant beds may become dry when microirriga-
tion systems are used, and quality problems may occur due to blowing sand or dirt,
which may require the extra expense of a sprinkler system or periodic hand watering
for dust control. However, a dry aisle may be desirable to reduce plant diseases, in-
sects, weeds, and to facilitate human traffic. These higher costs can usually be eco-
nomically justified by the high cash value for many (but not all) ornamental and land-
scape plant applications.
Water collection systems to recycle runoff and leachate from pots and containers
are commonly used with all greenhouse and nursery irrigation techniques. Some reuse
systems utilize impervious surfaces in greenhouses and nurseries to route and collect
runoff and drainage to a collection tank. This method is highly efficient and environ-
mentally friendly because excess water and nutrients are reused on subsequent irriga-
tions. However, the potential for increased plant disease and enhancement of insect
propagation is ever present with collection and reuse systems. Sanitation and poten-
tially extraordinary chemical water treatment procedures may be required to prevent
the spread of plant disease and other pests from the recycled water.
REFERENCES
Anyoji, H., and I. P. Wu. 1987. Statistical approach for drip lateral design. Trans.
ASAE 30(1): 187-192.
ASAE. 2005a. S-376.1: Design, installation and performance of underground,
thermoplastic irrigation pipelines. St. Joseph, Mich.: ASAE.
ASAE. 2005b. EP-405.1: Design and installation of microirrigation systems. St.
ASAE. 2005c. EP-435: Polyethylene pipe used for microirrigation laterals. St. Joseph,
Mich.: ASAE.
ASAE. 2005d. EP-458: Field evaluation of microirrigation systems. St. Joseph, Mich.:
ASAE.
Ayars, J. E., C. J. Phene, R. A. Schoneman, B. Meso, F. Dale, and J. Penland. 1995.
Impact of bed location on the operation of subsurface drip irrigation systems. In
Proc. 5th Int’l Microirrigation Congress, 141-146. St. Joseph, Mich.: ASAE.
Bar-Yosef, G., B. Sagiv, and T. Markovitch. 1989. Sweet corn response to surface and
subsurface trickle phosphorous fertigation. Agron. J. 81: 443-447.
Barth, H. K. 1995. Resource conservation and preservation through a new subsurface
irrigation system. In Proc. 5th Int’l Microirrigation Congress, 168-174. St. Joseph,
Mich.: ASAE.
Bisconer, I. 1987. Subsurface microirrigation of potatoes in Colorado. ASAE Paper
Bogle, C. R. T. K. Hartz, and C. Nuñez. 1989. Comparison of subsurface trickle and
furrow irrigation on platic-mulched and bare-soil for tomato production. J. Am.
Soc. Hort. Sci. 114(1): 40-43.
Boman, B. J., and L. R. Parsons. 1998. Microsprinkler selections considerations for
humid-region tree crops. ASAE Paper No. 98-2044. St. Joseph, Mich.: ASAE.
Bralts, V. F., and L. J. Sergerlind. 1985. Finite elements analysis of drip irrigation
submain unit. Trans. ASAE 28(3): 809-814.
Bralts, V. F., I. P. Wu, and H. M. Gitlin. 1981. Manufacturing variation and drip
irrigation uniformity. Trans. ASAE 24: 113-119.
Bralts V. F., D. M. Edwards, and I. P. Wu. 1987a. Drip irrigation design and
evaluation based on the statistical uniformity concept. Adv. Irrig. 4: 67-117.
Bralts V. F., I. P. Wu, and H. M. Gitlin. 1987b. Drip irrigation uniformity considering
emitter plugging. Trans. ASAE 24(5): 1234-1240.
Bucks, D. A., L. J. Erie, O. F. French, F. S. Nakayama, and W. D. Pew. 1981.
Subsurface trickle irrigation managment with multiple cropping. Trans. ASAE. 24:
1482-89.
Bucks, D. A. 1995. Historical developments in microirrigation. In Proc. 5th Int’l Mi-
croirrigation Congress, 1-5. St. Joseph, Mich.: ASAE.
Bucks, D. A., F. S. Nakayama, and R. G. Gilbert. 1979. Trickle irrigation water
quality and preventive maintenance. Agric. Water Mgmt. 2: 149-62.
Bui, W. 1990. Performance of “turbo model” drip irrigation tubes. In Proc. Third
Nat’l. Irrigation Symp., 198-203. St. Joseph, Mich.: ASAE.
Bui, W., and R. V. Osgood, 1990. Subsurface irrigation trial for alfalfa in Hawaii. In Proc.
Burt, C. M., K. O’Connor, and T. Ruehr. 1995. Fertigation. San Luis Obispo, Calif.:
Irrigation Training and Research Center, California Polytechnic State Univ.
Burt, C. M., and S. W. Styles. 1994. Drip and Microirrigation for Trees, Vines and
Row Crops. San Luis Obispo, Calif.: Irrigation Training and Research Center,
California Polytechnic State Univ.
Camp, C. R. 1998. Subsurface drip irrigation: A review. Trans. ASAE 41(5): 1353-
1367.
Camp, C. R., W. J. Bussher, and E.J. Sadler. 1987. Wetting patterns for line-source
trickle emitters. ASAE Paper No. 87-2524. St. Joseph, Mich.: ASAE.
Camp, C. R., E. J. Sadler, and W. J. Busscher. 1997. A comparison of uniformity
measures for drip irrigation systems. Trans. ASAE. 40(4): 993-999.
Chalmers, D. J., P. D. Mitchell, and L. van Heek. 1981. Control of peach tree growth
and productivity by regulated water supply, tree density and summer pruning. J.
Clark, G. A. 1992. Drip irrigation management and scheduling for vegetable
production. HortTech. Jan./Mar. 2(1).
Clothier, B. E., D. R. Scotter, and E. R. Harper. 1985. Three dimensional infiltration
and trickle irrigation. Trans. ASAE 28(2) :497-501.
Constable, G. A., and A. S. Hodson. 1990. A compariosn of drip and furrow irrigated cot-
ton on a cracking clay soil: Yield and quality of four cultivars. Irrig. Sci. 11: 149-53.
Davis, K. R., C. J. Phene, R. L. McCormack, R. B. Hutmacher, and D. W. Meeks.
1985. Trickle frequency and installation depth effects on tomatoes. In Proc. of the
Third Int’l Drip/Trickle Irrigation Congress, 896-902. St. Joseph, Mich.: ASAE.
Doorenbos, J., and A. H. Kassam. 1979. Yield response to water. FAO Irrigation and
Drainage Paper No. 33. Rome, Italy: Food and Agriculture Organization.
Drake, S. R., and R. G. Evans. 1997. Irrigation management influence on fruit quality and
storage life of ‘Redspur’ and ‘Golden Delicious’ apples. Fruit Varieties J. 51: 7-12.
Ebel, R. C., E. L. Proebsting, and R. G. Evans. 1995. Deficit irrigation to control
vegetative growth in apple and monitoring fruit growth to schedule irrigation.
HortSci. 30(6): 1229-1232.
Evans, R. G. 1994. Chapt. 16: Designing multipurpose water application systems. In
Proc. 1994 Pacific Northwest Fruit School: Tree Fruit Irrigation, 171-192.
Yakima, Wash.: The Good Fruit Grower.
Evans, R. G., T. W. Ley, M. W. Kroeger, and M. O. Mahan. 1988. Evaluation of
microsprinklers for undertree frost protection. Paper 2C-3. In Proc. 4th Int’l.
Micro-Irrigation Cong. Albury-Wodonga.
Evans, R. G., S. E. Spayd, R. L. Wample, and M. W. Kroeger. 1990. Water
requirements and irrigation management of Vitis vinifera grapes. In Proc. Third
Nat’l. Irrigation Symp., 154-161. St. Joseph, Mich.: ASAE.
Evans, R. G., S. E. Spayd, R. L. Wample, M. W. Kroeger, and M. O. Mahan. 1993. Water
use of Vitus vinifera grapes in Washington. Agric. Water Mgmt. 23(1993): 109-124.
Feng, J., and I. P. Wu. 1990. A simple computerized drip irrigation design. In Proc.
Gilad, Y., D. Peleg, and G. Tirosh. 1968. Irrigation equipment tests. Report No.
81268. Israel: ICWA.
Grattan, S. R., L. J. Schwankl, and W. T. Lanini. 1988. Weed control by subsurface
drip irrigation. Calif. Agric. 42(3): 22-24.
Grimes, D. W., D. S. Munk, and D. A. Goldhamer. 1990. Drip irrigation emitter depth
placement is a slowly permeable soil. In Proc. Third Nat’l. Irrigation Symp., 248-
254. St. Joseph, Mich.: ASAE.
Hanson, B. R. 1993. Salinity under drip irrigation of row crops. In Proc. 1993
Irrigation Association Exposition and Technical Conference, 196-202.
Hanson, B. R., L. Schwankl, S. R. Graham, and T. Pritchard. 1994. Drip irrigation of
row crops. Water Management Series Publication No. 93-05. Davis, Calif.:
Univ.California-Davis, Cooperative Extension.
Hills, D. J., F. M. Nawar, and P. M. Waller. 1989. Effects of chemical clogging on
drip tape irrigation uniformity. Trans. ASAE 32(4): 1202-1206.
Howell, T. A., and E. A. Hiler. 1974. Trickle irrigation lateral design. Trans. ASAE 17:
902-908.
Hutmacher, R. B., C. J. Phene, K. R. Davis, S. S. Vail, T. A. Kerby, M. Peters, C. A.
Hawk, M. Keeley, D. A. Clark, D. Ballard, and N. Hudson. 1995. Evapotranspira-
tion, fertility management for subsurface drip Acala and Pima cotton. In Proc. 5th
Int’l Microirrigation Congress, 147-154. St. Joseph, Mich.: ASAE.
Jensen, M. E., R. D. Burman, and R. G. Allen. 1990. Evapotranspiration and
Irrigation Water Requirements. ASCE Manual and Report on Engineering Practice
No. 70. New York, N.Y.: American Soc. Civil Engineers.
Jorgenson, G. S., and K. N. Norum, eds. 1992. Subsurface Drip Irrigation: Theory,
Practices and Application. CATI Publication No. 92-1001. Fresno, Calif.: CSRF.
Karmeli, D., L. J. Salazar, and W. R. Walker. 1978. Assessing the Spatial Variability
of Irrigation Water Application. Environmental Protection Technology Series.
Cincinnati, Ohio: U.S. Environmental Protection Agency.
Keller, J., and D. Karmeli. 1974. Trickle irrigation design parameters. Trans. ASAE
17(4): 678-684.
Kopec, A. R., M. N. Langley, and M. G. Bos. 1984. Major variables which influence
effective precipitation. Int. Comm. Irrig. Drain. Bull. 33(2): 65-70, 84.
Lamm, F. R., J. E. Ayers, and F. S. Nakayama, eds. 2006. Microirrigation for Crop
Production. Elsevier. 642 pp.
Lamm. F. R., and C. R. Camp. 2007. Subsurface drip irrigation. Chapter 13 in
Microirrigation for Crop Production, 473-551. F. R. Lamm, J. E. Ayars, and F. S.
Nakayama, eds. New York, N.Y.: Elsiever.
Lamm, F. R., G. A. Clark, M. Yitayew, R. A. Schoneman, R. M. Mead, and A. D.
Schnieder. 1995. Installation issures for SDI systems. In Proc. of 1995 Annual Irri-
gation Association Exposition and Technical Conference. St. Joseph, Mich.: ASAE.
Leib, B. G., H. W. Caspari, C. A. Redulla, P. K. Andrews, and J. D. Jabro. 2006.
Partial rootzone drying and deficit irrigation of ‘Fuji’ apples in a semi-arid climate.
Irrig. Sci. 24(2): 85-99.
Meshkat, M., and R. C. Warner. 1985. A user friendly interactive trickle irrigation
design model. In Proc. of the Third Int’l Drip/Trickle Irrigation Congress, I: 439-
451. St. Joseph, Mich.: ASAE.
Middleton, J. E., E. L. Proebsting, and S. Roberts. 1981. A comparison of trickle and
sprinkler irrigation for apple orchards. Wash. Agric. Exp. Sta. Bulletin 0895.
Mitchell, P. D., P. H. Jerie, and D. J. Chalmers. 1984. The effects of regulated water
deficits on pear tree growth, flowering, fruit growth and yield. J. American Soc.
Hort. Sci. 109(5): 604-606.
Moore, R., and J. Fitoschen. 1990. The drip irrigation revolution in the Hawaiian
sugarcane industry. In Proc. Third Nat’l. Irrigation Symp., 223-227. St. Joseph,
Mich.: ASAE.
Moser, E. 1979. Technically-oriented research on drip irrigation equipment for special
crops. Int. Soc. Hort. Sci. (ISHS) 79: 37-45.
Nakayama, F. S., B. J. Boman, and D. J. Pitts. 2007. Maintenance. Chapter 11 in
Microirrigation for Crop Production, 389-430. F. R. Lamm, J. E. Ayars, and F. S.
Nakayama, eds. New York, N.Y.: Elsiever.
Nakayama, F. S., and D. A. Bucks. 1991. Water quality in irrigation: A Review. Irrig.
Sci. 12: 187-92.
Nakayama, F. S., and D. A. Bucks, eds. 1986. Trickle Irrigation for Crop Production:
Design, Operation and Management. New York, N.Y.: Elsevier and St. Joseph,
Mich.: ASAE.
Or, D., and F. E. Coelho. 1996. Soil water dynamics under drip irrigation: Transient
flow and uptake models. Trans. ASAE 39(6): 2017-2025.
Oron, G., Y. DeMalach, and Z. Hoffman. 1989. Subsurface trickle irrigation of alfalfa
with treated wastewater. Progress Report. Israel: Ben Gurion Univ., Institute of
Desert Research.
Parchomchuk, P. 1976. Temperature effects on emitter discharge rates. Trans. ASAE
19(4): 690-692.
Peng, G. F., I. P. Wu, and C. J. Phene. 1986. Temperature effects on drip line
hydraulics. Trans. ASAE 29(1): 211-215.
Peretz, J., R. G. Evans, and E. L. Proebsting. 1984. Leaf water potentials for
management of high frequency irrigation of apples. Trans. ASAE. 27(2): 437-442.
Phene, C. J. 1995. The sustainability and potential of subsurface drip irrigation. In
Proc. 5th Int’l Microirrigation Congress, 359-367. St. Joseph, Mich.: ASAE.
Phene, C. J., K. R. Davis, R. B. Hutmacher, and R. L. McCormick. 1987a. Advantages of
subsurface drip irrigation for processing tomatoes. Acta Horticulturae 200: 101-113.
Phene, C. J., K. R. Davis, and R. L. McCormick. 1987b. Evapotranspiration and
irrigation scheduling of drip irrigated cantaloupes. ASAE Paper No. 87-2526. St.
Phene, C. J., R. B. Hutmacher, and J. E. Ayars. 1992. Subsurface drip irrigation:
Realizing the full potential. In Proc. of Conference on Subsurface Drip Irrigation,
137-158. CATI Publication 921001. Fresno, Calif.: California State Univ.
Philip, J. R. 1991. Effects of root and subirrigatin depth on evaporation and
percolation losses. J. American Soil Sci. Soc. 55(6): 1520-1523.
Pitts, D. J., F. S. Zajueta, and A. G. Smajstrla. 1986. Micro Irrigation System Submain
Design Evaluation. Florida Cooperative Extension Service, Univ. Florida.
Pitts, D. J., K. Petersen, G. Gilbert, and R. Fastenau. 1996. Field assessment of
irrigation system performance. Appl. Eng. in Agric. 12(3): 307-313.
Plaut, S., M. Rom, and A. Meiri. 1985. Cotton response to subsurface trickle
irrigation. In Proc. of the Third Int’l Drip/Trickle Irrigation Congress, 916-20. St.
Proebsting, E. L., J. E. Middleton, and S. Roberts. 1977. Altered fruiting and growth
characteristics of ‘Delicious’ apples associated with irrigation method. HortSci. 12:
349-350.
Rawlins, S. L. 1977. Uniform irrigation with a low head bubbler system. Agric. Water
Mgmt. 1: 166-178.
Rodriguez-Sinobas, L., L Juana, and A, Losada. 1999. Effects of temperature changes
on emitter discharge. J. Irrig. Drain. Eng. 125(2): 64-73.
Ross, D. S. 1994. Reducing water use under nursery and landscape conditions. In
Recycling and Resource Conservation, A Reference Guide for Nursery and
Landscape Industries. C. W. Heuser, Jr., and P. E. Heuser, eds. Harrisberg, Pa.:
Pennsylvannia Nurserymens Assoc., Inc.
Ross, D. S., R. A. Parsons, and H. E. Carpenter. 1990. Trickle irrigation in the eastern
United States. Publication NRAES-4. Ithaca, N.Y.: Northwest Regional
Agricultural Engineering Service, Cornell Univ.
Rubiez, I. G., N. F. Oebker, and J. L. Stroehlein. 1989. Subsurface drip irrigation and
urea phosphate fertigation for begetables on calcareous soils. J. Plant Nutrition
12(12): 1457-1465.
Sadler, E. J., C. R. Camp, and W. J. Busscher. 1995 Emitter flow rate changes caused
by excavating subsurface microirrigation tubing. In Proc. 5th Int’l Microirrigation
Congress, 763-768. St. Joseph, Mich.: ASAE.
Sammis, T. W. 1980. Comparison of sprinkler, trickle, subsurface and furrow
irrigation methods for row crops. Agron. J. 72: 701-04.
Sammis, T. W., and I. P. Wu. 1985. Effect of drip irrigation design and management
on crop yield. Trans. ASAE 28(3): 832-838.

Scherm, H., and A. H. C. van Bruggen. 1995. Comparative study of microclimate and
downy mildew development in subsurface drip- and furrow-Irrigated lettuce fields
in California. Plant Disease 79(6): 620-25.
Schwankl, L., B. Hanson, and T. Pritchard. 1995. Micro-irrigation of Trees and Vines:
A Handbook for Water Managers. Water Managment Series Publication No. 94-01.
Davis, Calif.: Univ. California-Davis, Cooperative Extension.
Seginer, I. 1978. A note on the economic significance of uniform water application.
Irrig. Sci. 1: 19-25.
Shani, U., X. Xue, R. Gordin-Katz, and A. W. Warrick. 1996. Soil-limiting flow from
subsurface emitters. I: Pressure measurements. J. Irrig. Drain. Eng. 122(5): 291-295.
Smajstrla, A. G., B. J. Boman, D. Z. Haman, D. J. Pitts, and F. S. Zazueta. 1997. Field
evaluation of microirrigation water application uniformity. Bulletin 265.
Gainesville, Fla.: Florida Cooperative Extension Service, Univ. Florida.
Smajstrla, A. G., and S. J. Locascio. 1996. Tensiometer controlled, drip irrigation
scheduling of tomatoes. Appl. Eng. Agric. 12(3): 315-319.
Smajstrla, A. G., S. J. Locascio, and D. R. Hensel. 1995. Subsurface drip irrigation of
potatoes. Proc. Fla. State Hort. Soc. 108 :193-195.
Soloman, K. H. 1979. Manufacturing variation of emitters in trickle irrigation systems.
Trans. ASAE 22: 1034-1038, 1043.
Solomon, K. H. 1984. Global uniformity of trickle irrigation systems. ASAE Paper
Solomon, K. H. 1992. Subsurface drip irrigation: Product selection and performance.
In Subsurface Drip Irrigation: Theory, Practices and Application, 17-39. CATI
Publication No. 92 1001. Fresno, Calif.: California State Univ.
Stanley, C. D., and G. A. Clark. 1991. Water table management using microirrigation
tubing. Soil and Crop Sci. Soc. Fla. Proc. 50: 6-8.
Sterret, S. B., B. B. Ross, and C. P. Savage. 1990. Establishment and yield of
asparagus as influenced by planting and irrigation method. J. American Soc. Hort.
Sci. 115(1): 29-33.
Thorton, J. R., and D. Behoteguy. 1980. Operation and Installation of a bubbler
system. ASAE Paper No 80-2059. St. Joseph, Mich.: ASAE.
Tollefson, S. 1988. Commercial production of field and vegetable crops with
subsurface drip irrigation. In 1988 Technical Conference Proceedings, 144-153.
Falls Church, Va.: Irrigation Association.
Wample, R. L. 1996. Issues in vineyard irrigation. Wine East 24(2): 8-21.
Wample, R. L. 1997. Important issues in vineyard irrigation. Good Fruit Grower
48(14): 15-22, 39.
Warrick, A. W., and D. O. Lomen. 1983. Linearized moisture flow with root
extraction over two-dimensional zones. J. Soil Sci. Soc. America 44(5): 869-872.
Warrick, A. W., and U. Shani. 1996. Soil-limiting flow from subsurface emitters. II:
Effect on uniformity. J. Irrig. Drain. Eng. 122(5): 296-300.
Wu, I. P. 1988. Linearized water application function for drip irrigation schedules.
Trans. ASAE 31(6): 1743-1749.
Wu, I. P. 1992a. Energy gradient line approach for direct hydraulic calculation in drip
irrigation design. Irrig. Sci. 13: 21-29.
Wu, I. P. 1992b. A simple graphic solution for drip lateral design. Paper No. 9209
103. In Proc. Int’l Conference on Agricultural Engineering.

Wu, I. P. 1993a. Microirrigation design for trees. ASAE Paper No. 93-2128. St.
Wu, I. P. 1993b. Design considerations of drip irrigation systems. In Transactions,
R53, Q44,15th Congress on Irrigation and Drainage, 693-711. The Hague, The
Netherlands: ICID.
Wu, I. P. 1995. A simple optimal microirrigation scheduling, In Proc. 5th Int’l
Microirrigation Congress, 781-786. St. Joseph, Mich.: ASAE.
Wu, I. P., J. S. Feng, and K. Y. Yabusaki. 1989. Emitter spacing and uniformity of
irrigation application. ASAE Paper No. 89-2080. St. Joseph, Mich.: ASAE.
Wu, I. P., and H. M. Gitlin. 1974. Drip irrigation design based on uniformity. Trans.
ASAE 17: 429-432.
Wu, I. P., and H. M. Gitlin. 1975. Energy gradient line for drip irrigation laterals. J.
Irrig. Drain. Eng. 101(IR4): 323-326.
Wu, I. P., T. A. Howell, and E. A. Hiler. 1979. Hydraulic design of drip irrigation
systems. Technical Bulletin No. 105. Manoa, Hawaii: Hawaii Agricultural
Experiment Station, Univ. Hawaii.
Wu, I. P., and J. M. Irudayaraj. 1987. Evaluation of uniformity parameters for drip
irrigation design. ASAE Paper No. 87-2522. St. Joseph, Mich.: ASAE.
Wu, I. P., and J. M. Irudayaraj. 1992. Hydraulic variation of a drip irrigation submain
unit. ASAE Paper No. 92-2056. St. Joseph, Mich.: ASAE.
Wu, I. P., J. M. Irudayaraj, and J. S. Feng. 1988a. Plugging evaluation for a drip
irrigation submain unit. ASAE Paper No. 88-2062. St. Joseph, Mich.: ASAE.
Wu, I. P., J. M Irudayaraj, and J. S. Feng. 1988b. Grouping effect on the uniformity of
drip irrigation. Paper No. 10-4. In Congress Proc. Fourth Int’l Microirrigation
Congress, V1.
Wu, I. P., G. Y. Lin, and L. S. Lau. 1991. Plugging evaluation in the reuse of sewage
effluent by drip irrigation. In Proc. 1991 Nat’l. Conf. ASCE, 780-786. Reston, Va.:
ASCE.
Wu, I. P., and C. J. Phene. 1984. Temperature effect on drip emitters and lateral line.
Wu, I. P., K. Y. Yabusaki, and J. M. Irudayaraj. 1985. Computer simulation of total
emitter flow variation: Drip/trickle irrigation in action. Proc. Third Int’l Drip/
Trickle Irrigation Congress 2: 873-886.
Wu, I. P., and R. Yue. 1993. Drip lateral design using energy gradient line approach.
Trans. ASAE 36(2): 389-394.
Wuertz, H. 1992. Subsurface drip irrigation on Sundance Farms Ltd. In Proc. of
Conference on Subsurface Drip Irrigation, 73-83. CATI Publication 921001.
Fresno, Calif.: California State Univ.
Yitayew, M., K. Didan, and C. Reynolds. 1999. Microcomputer based low-head
gravity-flow bubbler irrigation system design. Computers and Electronics in Agric.
22: 29-39.
Yitayew, M., C. A. Reynolds, and A. E. Sheta. 1995. Bubbler irrigation system design
and management. In Proc. 5th Int’l Microirrigation Congress, 402-413. St. Joseph,
Mich.: ASAE.
Zur, B., and S. Tal. 1981. Emitter discharge sensitivity to pressure and temperature. J.
Irrig. Drain. Eng. 107(IR1): 1-9.
CHAPTER 18
WATER TABLE
CONTROL SYSTEMS
James L. Fouss (USDA-ARS, Baton
Rouge, Louisiana)
Robert O. Evans (North Carolina State University,
Raleigh, North Carolina)
James E. Ayars (USDA-ARS,Parlier, California)
Evan W. Christen (CSIRO Land and Water,
Griffith, New South Wales, Australia)
Abstract. Agricultural water management systems are designed and installed to (1)
improve crop production by controlling the durations of excessive and deficient soil-
water conditions in the root zone, and (2) improve the water quality of drainage dis-
charge by controlling drainage flows to reduce agrochemical losses from farmland.
Controlling subsurface drainage discharge to maintain a shallower water table and
limit drainage outflow has been shown to reduce annual nitrate loss from cropland as
much as 50%. The development of an integrated water table control system includes
determination of site suitability, drain depth and spacing, preparation of a field instal-
lation plan, and incorporation of components for operating the system in the con-
trolled drainage and subirrigation modes. The system design should permit control of
the water table depth in the soil profile over the range needed for the cultural prac-
tices to be followed and the crops to be grown, and the operational requirements to
reduce agrochemical losses. These objectives should also involve the efficient utiliza-
tion of shallow groundwater supplied by rainfall or irrigation.
Several methods are available for determining optimum design drain depth and
spacing for water table management. The operational mode of the system, that is,
whether it is in the conventional subsurface drainage, controlled drainage, or subirri-
gation mode, may vary from day to day, month to month, and from year to year. In
most locations, it is not clear whether the greatest demands on the system design are
to provide good drainage management under shallow water table conditions, or to
provide sufficient subirrigation during the driest periods. Because of complicating
design factors, a simulation modeling approach should be used to conduct a complete
analysis and final design of the drainage water management or water table control
system, and to predict its performance over a period of 20 to 30 years for the clima-
tologic conditions at the specific site.
Computer simulation models such as DRAINMOD can be used to evaluate various

system design options for a specific site. A long-term simulation (20 to 30 years) can
provide a good evaluation of the expected performance of the system. DRAINMOD-
NII includes a routine to comprehensively evaluate the impact of system design and
operational parameters on transport of various forms of nitrogen within the soil pro-
file and losses in runoff and subsurface flow. This version of DRAINMOD is an impor-
tant tool for designing drainage water management systems, along with a seasonal
operation plan, to meet emerging water quality requirements for surface receiving
streams and water bodies.
The design for a water table management system that results in the optimum net
profit, while minimizing the environmental impacts offsite, should be the best design
and management strategy. The system may be technically feasible, but the final deci-
sion should be based on the feasibility of the system not only to pay for itself but to
return a profit to the farmer for his investment, while minimizing offsite environmental
impacts. Therefore, the final decision for a given or recommended design should be
based on a thorough evaluation of economic and environmental impacts.
Keywords. Agrochemical, Arid, Controlled drainage, Drainage management,
DRAINMOD, Environmental impacts, Groundwater, Humid, Runoff, Salinity, Subirri-
gation, Subsurface drainage, Surface drainage, Water management, Water quality,
Water table, Water table control.
18.1 INTRODUCTION
Agricultural drainage and related water management systems to control water table
depth have historically been installed to eliminate water-related factors that limit crop
production, or to reduce those factors to acceptable levels. In recent years, research
has shown that management of drainage water provides an opportunity to control the
loss of agrochemicals (fertilizer nutrients and pesticides) from cropland, as these are
carried in surface runoff and subsurface drainage discharge. For example, controlling
subsurface drainage discharge to maintain a shallower water table and control (limit)
drainage outflow can reduce annual nitrate loss in discharge by as much as 50%.
Likewise, management of surface darainage systems has been shown to reduce phos-
phorus loss in runoff.
Recent attention to agricultural drainage management systems1 for water table con-
trol resulted from widely published nutrient (nitrate and phosphorus) contamination
causing hypoxia in the northern Gulf of Mexico (see, for example, Rabalais et al.,
1999). The primary source of the contamination is from drained agricultural lands in
the Midwest and is transported through the Mississippi River system to the Gulf. In
the irrigated western U.S., toxic elements such as selenium are transported by drainage
flows to wildlife habitats, where these elements can bio-accumulate to toxic levels and
eventually cause wildlife deaths. Thus, surface and subsurface drainage water man-
1
An interagency Agricultural Drainage Management Systems Task Force was formed in 2002 to promote
and implement drainage water management (controlled drainage) practice in eight Midwest states to reduce
nitrate loss from subsurface-drained cropland contributing to nutrient load in the Mississippi River system
and hypoxia in the northern Gulf of Mexico; see Fouss et al. (2004) and web site: http://extension.osu.edu/
~usdasdru/ADMS/ADMSindex.htm
686 Chapter 18 Water Table Control Systems
agement systems are crucial for mitigating environmental impacts for both rainfed and
irrigated croplands.
In the irrigated West, water table control objectives are to improve the quality of
drainage discharge waters (in terms of both agrochemicals and salinity) and improve
irrigation water use efficiency. This is achieved by encouraging crop water use from
the water table and reduction of bypass flows to drains, where these do not contribute
to salt leaching and direct loss of irrigation water. Another important function of water
table control is shifting a subsurface drainage system from a reclamation phase, when
salt leaching from the root zone is the highest priority, to a maintenance phase, when a
lesser level of drainage can maintain appropriate root-zone salinity.
System design should permit control of the water table depth in the soil profile over
the range needed for the cultural practices to be followed, the crops to be grown, and
the operational requirements to reduce agrochemical losses. An integrated design for a
water table control system includes determination of site suitability (including an ade-
quate outlet); drain depth and spacing; drain conduit diameter; a drain layout and in-
stallation plan; an adequate water supply (if subirrigation is used); a method or struc-
tures for operating the system; and a monitoring and operational plan. An acceptable
application of water table control depends on the costs of the required water manage-
ment system in relation to the economic and environmental benefits expected. Such
benefits vary from year to year with both weather and economic conditions and are
difficult to quantify because of the complex interrelationships of crop production and
environmental processes.
As more is learned about plant growth and yield, machinery-soil interactions (e.g.,
trafficability), soil salination and reclamation processes, and agrochemical losses in
drainage flows, it will be possible to simulate entire crop production processes, includ-
ing crop selection, rotations, and fertilization. This will permit optimizing the water
table management system design based on profit potential and environmental benefits
or impacts. Lacking this knowledge at the present time, more intermediate or tradi-
tional objectives of water table management systems must be used. Such objectives
are easier to quantify and generally form the basis for system selection and design.
18.2 MANAGEMENT OF SOIL WATER
BY WATER TABLE CONTROL
18.2.1 Application of Water Table Control Systems
The performance criteria for water table management systems can be described in
terms of several individual parameters, such as trafficability, control of excess and
deficit soil-water conditions, salinity control, subirrigation efficiency, and off-site en-
vironmental impacts.
1. Traditional drainage parameters—Parameters for trafficability and control of
excess soil water were traditional parameters in rainfed areas. The recent focus on
drainage water management (controlled drainage) to reduce nutrient (particularly ni-
trate) loss in subsurface drainage discharge has caused a shift in the performance pa-
rameters to include environmental or water quality impacts of subsurface drainage. In
arid regions, salinity control has been a traditional primary parameter; in recent years
parameters concerning environmental impacts have become very important. The major
difference between current and traditional parameters is that operation in the con-
trolled drainage and subirrigation modes may complicate or reduce the ability of the
system to satisfy traditional drainage requirements and still meet current requirements
for water quality protection in both rainfed and arid regions.
2. Deficit soil-water conditions—While wetness is the primary concern to most
landowners, yields on poorly drained soils with shallow water tables are sometimes
substantially reduced by deficit soil-water conditions or drought stresses. Shallow
groundwater conditions in the crop root zone often promote plant stresses caused by
excessive soil-water conditions. However, the temporal and spatial variability of rain-
fall in many humid areas frequently results in excessive soil water and related reduced
crop growth early in the growing season followed by deficit soil water conditions later
in the growing season. Early excess water stress often aggravates deficit rainfall condi-
tions later in the season because shallow root depths caused by high water tables early
in the growing season may not be sufficiently deep to access deeper soil water needed
by the crop later in the season. Simulation studies by Skaggs and Tabrizi (1983) found
that drought stresses reduced corn yields on a poorly drained North Carolina soil by an
average of 22% even though average annual rainfall exceeded 1300 mm. Drought
stresses may be increased by the drainage systems that are required to farm these shal-
low water table soils (Doty et al., 1984, 1987; Skaggs and Tabrizi, 1983). The effects
of drainage and associated water management systems on drought stresses and yields
should be considered in the design and operation of those systems.
3. Controlled drainage—The performance criteria for the controlled drainage mode
of operation depends upon whether capabilities also exists to provide the subirrigation
mode. Some modest benefits in terms of reducing deficit soil-water conditions can be
achieved in some soils by maintaining, “controlling”, the water table level near the
root-zone depth to insure that rainfall is retained in the soil profile for crop use. This
controlled drainage mode may result in short-term excess soil-water conditions in the
root zone immediately following rainfall for the benefit of less deficit soil water stress
later. If the water table remains elevated for extended periods during early crop devel-
opment, crop roots may not develop deep enough to prevent subsequent drought stress
later in the growing season during extended periods with below normal rainfall. Thus,
an operational plan must be developed that considers seasonal rainfall amount and
frequenty to achieve a balance between excessive and deficit soil water conditions
throughout the growing season.
4. Subirrigation—The performance criteria for operation in the subirrigation mode
includes the control of both deficit and excess soil-water conditions. That is, systems
designed for subirrigation must also be designed to operate in the controlled drainage
and conventional drainage modes to insure that rainfall events do not cause extreme
fluctuations of water table depth into the root zone. Subirrigation may be the govern-
ing design performance parameter since a closer subsurface conduit spacing is typi-
cally required to supply adequate subirrigation water to the soil profile than the con-
duit spacing required to remove excess soil-water from the root zone and soil profile.
In more humid regions, subsurface drainage following rainfall events that occur while
subirrigating is often the governing design performance parameter. Even where subir-
rigation is economically feasible, not all farmers will install the drainage conduits at a
close enough spacing for the future retrofit for subirrigation because of the higher ini-
tial cost for the installed system. Subirrigation design and operation should consider
balancing the timing and magnitude of excessive and deficit soils water conditions
similarly to controlled drainage.
5. Off-site environmental impacts—Drainage discharge water from agricultural

cropland containing nitrate-Nitrogen has caused undesirable off-site environmental
impacts to receiving streams and water bodies. Rabalais et al. (2002) identified nitrate
discharged in drainage waters from cropland in the Midwest and Upper Midwest states
as the major source of nitrate carried down the Mississippi River System to the hy-
poxic zone (Rabalais et al., 1999) of the Northern Gulf of Mexico . To address these
problems on a broad scale and to promote and implement solutions involving drainage
water management, the Agricultural Drainage Management Systems Task Force was
formed in 2002 (Fouss et al., 2004). Controlling subsurface drainage discharge to
maintain a shallower water table and limit drainage outflow has been shown to reduce
annual nitrate loss from cropland as much as 50% (Gilliam et al., 1979; Evans et al.,
1995; Fausey aet al., 2004). The Task Force concluded that a suite of Best Manage-
ment Practices, not just drainage water management, would be required to solve the
water quality problems throughout the Mississippi River Basin and the Gulf (Fouss
and Appelboom, 2006).
18.2.2 Site Evaluation and Suitability
Several soil properties and site parameters influence the design of a water table
management system. Important properties include: lateral hydraulic conductivity,
depth to a restrictive layer, soil-water characteristics, upward flux and drainable poros-
ity as functions of water table depth, soil layering, infiltration, crop rooting depth, to-
pography, and suitable drainage outlet. The design of the system is more sensitive to
some parameters than others; some are more difficult to measure in the field than oth-
ers; and some properties are more spatially varied than others. It is generally not prac-
tical to measure all of the important soil and site parameters in the field for every po-
tential site. Properties that are generally practical to measure are hydraulic conductiv-
ity (field effective), soil-water characteristic, depth and thickness of soil horizons, and
depth to the restrictive layer. Saturated lateral hydraulic conductivity is one of the
most important factors influencing the design of the system (Skaggs, 1980, 1981) and
is also one of the most spatially varied properties in the field (Tabrizi and Skaggs,
1983; Rogers and Fouss, 1989; and Rogers et al., 1991). Therefore, effort should be
made to determine representative, field effective hydraulic conductivity values for
each site. Drainable porosity can be calculated from soil-water characteristic data,
when available. System design is also sensitive to upward flux as a function of water
table depth (Skaggs, 1980); this relationship is usually estimated from steady-state
solutions, which require the unsaturated hydraulic conductivity function. Drainable
porosity and infiltration may also be estimated from the soil-water characteristic and
unsaturated hydraulic conductivity function.
The feasibility of subirrigation at a specific site depends on the source, cost, and re-
liability of a water supply. Evans et al. (1988a) discussed many types of water supplies
and their associated reliability and cost for subirrigation. Potential seepage losses
should also be identified. In poorly drained soils with a restricting barrier (layer) in the
profile, vertical losses are normally small. Lateral seepage losses, on the other hand,
may consume more than 25% of the pumping capacity in some cases. Lateral seepage
losses can be minimized with good planning, system layout, and management. When-
ever possible, supply canals should be located near the center of subirrigated fields
rather than along field boundaries. Perimeter ditches and outlet canals should also be
equipped with control structures. When seepage losses cannot be controlled, it is nec-
essary to determine the length of the seepage boundary, the hydraulic gradient, and
hydraulic conductivity along the boundary. Skaggs (1980) described methods for es-
timating seepage losses under steady state conditions, with specific examples on how
to approximate the seepage losses to nearby drains or canals, adjacent undrained
fields, and vertical or deep seepage.
Often, an experienced engineer can determine the suitability of a potential site for
water table management by a qualitative site investigation. The presence of six general
site conditions (discussed below) will usually indicate whether or not water table man-
agement is feasible or practical.
1. Site drainage benefits—For soils in the humid regions of the eastern U.S., water
table management is practical only on those sites that will benefit from improved sub-
surface drainage. Where subsurface drainage is not needed under natural conditions,
such as in soils lacking a shallow water table (hydric soil), irrigation needs are usually
most efficiently satisfied by surface application systems (e.g., furrow, sprinkler, or
drip irrigation). A soil survey report will indicate the natural drainage conditions for a
given soil series. Soils that are classified as either somewhat poorly drained, poorly
drained, or very poorly drained will usually benefit from artificial drainage and are
candidates for water table management.
2. Topography—Soils that support water table management systems are usually
relatively flat. For example, poorly drained soils in the Southeastern Coastal Plains,
the Midwest Great Lakes Region, the Florida Flatwoods, and the Mississippi Delta
rarely occupy landscape positions on slopes greater than 2%. In fact, very few systems
have been installed on slopes greater than 0.5%. As the slope approaches 1%, the
number and cost of control structures to maintain a water table depth within a desired
range for a specific crop may become economically prohibitive. Grain crops can usu-
ally tolerate a range in the water table depth of 0.30 m to 0.45 m, whereas shallow-
rooted vegetable crops can only tolerate a water table fluctuation of 0.15 m to 0.20 m
without showing signs of water stress during dry periods. A control structure is then
needed for each change in ground surface elevation corresponding to the desired water
table range for that zone. From a physical standpoint, the maximum slope that can be
tolerated tends to be site specific and related to hydraulic conductivity. At slopes
above 2%, a uniform water table depth is difficult to maintain because of lateral seep-
age when the conductivity exceeds 0.5 m/day (0.7 in/h). By contrast, at lower conduc-
tivity, the cost for the closer drain spacing often becomes prohibitive. In general, the
limiting factor with respect to slope will be economics rather than physical slope con-
ditions.
3. Hydraulic conductivity—Hydraulic conductivity is the single most important soil
factor affecting the economic feasibility of a water table management system. For pre-
liminary planning purposes, hydraulic conductivity can be estimated from values re-
ported in soil survey reports. As with slope, prohibitive hydraulic conductivity values
are a function of economics rather than physical limitations in the design and opera-
tion of the system. The limiting conductivity for a specific site will be dictated largely
by the value of the increased crop obtained, or the potential value of obtaining consis-
tently high yields year after year.
4. Restrictive layer or seasonal shallow water table—Potential sites typically have
a barrier at some depth in the soil profile, which prevents excessive vertical seepage
losses. A restrictive layer is often encountered between 1.5 m and 10 m. The existence
of the barrier becomes increasing difficult to locate with a hand auger when its depth
exceeds 3 m. The presence of a seasonal shallow water table is usually sufficient evi-
dence to indicate that a water table can be maintained on the site at an elevation suit-
able for subirrigation, whether or not the restrictive barrier can be located. In addition,
the depth of the seasonal shallow groundwater is a good indicator of the natural drain-
age of the site.
The depth of the seasonal shallow water table can be estimated from a soil survey
or by inspection in the field. Gray mottles in the soil profile indicate the position of the
seasonal water table under natural drainage conditions. When the gray mottles occur
within 0.5 m of the soil surface, the site is a candidate for water table management. As
the depth to the gray mottles increases to 1.0 m, the site is marginally suited and ex-
cessive seepage may be a problem. In this case, the actual depth to the restrictive bar-
rier should be determined. Soils with gray mottles more than 1.0 m from the surface
are naturally well drained. In these cases, benefits from subsurface drainage will be
minimal and the cost of water table control would have to be justified on the basis of
subirrigation benefits alone. Unless the natural drainage condition can be managed,
excessive seepage will occur and the site is not well suited for water table manage-
ment.
5. Drainage outlet requirements—When evaluating the potential of any site for a
water table management system, drainage is a primary consideration. A drainage out-
let that will remove excessive surface and subsurface water within an acceptable inter-
val of time (typically 24 h) must be available. When a gravity outlet to an existing
stream or canal is available, the drainage outlet should be at least 1.2 m lower than the
lowest land surface for the system. Where a gravity outlet is not available or possible,
an outlet structure (e.g., a sump or pump station) can be constructed and drainage flow
pumped to a higher elevation for discharge into an existing or constructed channel or
for storage and later use during subirrigation. In some cases the sump outlet structure
may be needed only for subsurface drainage, as a suitable gravity flow outlet is avail-
able for surface runoff. An example of a pumped subsurface drainage outlet sump is
illustrated in Figure 18.1; this system is shown in the controlled drainage mode of op-
eration. In cases where drainage water must be pumped, it is desirable to store as much
of the drainage water as possible in a nearby on-farm pond or wetland reservoir. If
stored, this water can be used later during the growing season to satisfy subirrigation
requirements, often without the need to be pumped again. That is, a subsurface drain-
age-subirrigation system usually requires pumping in only one direction. Such a sys-
tem can usually utilize gravity flow in the opposite mode.
6. Water supply requirements—The reliability, location, quantity, and quality of ir-
rigation water are critical considerations in evaluating a site for a subirrigation. The
water source should be located as close as possible to minimize conveyance losses and
costs. The quantity of water needed for subirrigation will vary depending upon the
weather (rainfall and evapotranspiration), crop being irrigated, and water loss rate
from the field by deep and lateral seepage. Water source requirements may range from
70 L/min/ha irrigated in the southeastern U.S. to 40 L/min/ha irrigated in the cooler
Midwest and Canada.
Once the physical suitability of a site for water table management has been deter-
mined, the cost of the system should be estimated and discussed with the landowner
before any additional time is spent on design. Average cost for many of the compo-
Figure 18.1. Controlled-drainage and subirrigation modes of operation for water table
management; outlet (sump) water level control based on feedback of field water table depth.
nents of the system are usually available from the local Natural Resources Conserva-
tion Service (NRCS), Cooperative Extension Service, drainage pipe manufacturers, or
drainage contractors in the area. The dominant costs of the system are usually drain
tubing, system installation, outlet construction (if gravity outlet is not available), water
supply costs, and system controller apparatus. The drain tube diameters for laterals
and collector mains and lateral drain spacing must be estimated before the tubing and
installation costs can be determined. At this stage of the design process, it is adequate
to estimate the drain spacing from local rules of thumb or shortcut methods for either
subsurface drainage (including controlled drainage) or subirrigation, depending upon
which alternative is desired or deemed more important.
18.3 SYSTEM DESIGN AND OPERATION
IN HUMID REGIONS
18.3.1 System Design Objectives
The design objectives of a water table control system include all the traditional ob-
jectives of drainage systems, plus objectives of conserving soil water and reducing
deficit soil-water conditions, plus reducing agrochemical losses in drainage discharge.
These objectives include:
provide trafficable or workable soil conditions so that farming operations, such
as seedbed preparation, tillage, and harvesting, can be conducted in a timely
manner, and without damage to the soil structure;
reduce plant stresses caused by excessive soil-water conditions;
control soil salinity and alkalinity;
reduce or eliminate plant stresses caused by deficit soil-water conditions;
minimize harmful offsite environmental impacts from agrochemical losses;
conserve and efficiently utilize water supplied by precipitation, thus minimizing

irrigation water requirements; and
maintain a soil-water environment so that other practices, such as conservation
tillage, post-harvest residue, or cover crops, are more effective and beneficial.
The first three objectives are traditional ones for drainage system design. To incor-
porate controlled drainage or subirrigation, the others may be addressed. The relative
importance of these objectives is case dependent, varying seasonally and from year to
year with soil, crop, site, and climatological factors. For example, the drainage inten-
sity required to provide workable conditions on the soil for spring planting may cause
overdrainage and increase deficit soil-water conditions later in the growing season.
This situation can be avoided by appropriate design and timely management, provided
the system designer properly considers the objectives and factors controlling system
performance, including the variable nature of rainfall. Weather forecast information
(e.g. rainfall probability) may become an important input for real-time operation of
water management systems, particularly the integrated management of water and ag-
rochemicals to improve water quality.
There may be several water management system design alternatives that satisfy the
design objectives and environmental requirements. Whether or not a given system
design will satisfy the objectives depends upon the location, crop, and soil properties.
Of course, the objectives themselves may depend upon the individual farmer’s man-
agement capabilities, equipment, and manpower available. For example, some farmers
may be interested only in controlled drainage and do not feel the additional cost of
subirrigation facilities can be justified for their farm enterprises. In system design,
however, the future conversion of a controlled drainage system for operation in the
subirrigation mode should be considered.
18.3.2 Design Methods
Drain depth is usually dictated by the relative permeability of soil layers, depth of
slow-draining layers, outlet depth, limitations of installation equipment, and local
standards for drain grade and depth. When site and soil characteristics allow drain
depth flexibility, a depth between 0.75 m and 1.5 m will usually be optimal. However,
with the recent focus on water quality of subsurface drainage effluent, it has been rec-
ommended (ADMS-TF, 2005) that drain depths greater than 1.0 m (without outlet
controls) be avoided, if possible. The shallower drains, even without outlet controls,
remove less water from the soil profile and thus reduce the potential loss of applied
agrochemicals.
Several methods are available for determining optimum design spacing. On a field-
to-field basis, all of the methods will provide a better estimate of the required drain
spacing if the saturated hydraulic conductivity and depth to the restrictive layer have
been measured on the specific site rather than estimated from the soil survey or local
drainage guide. The operation of the system, whether it is in the conventional subsur-
face drainage, controlled drainage, or subirrigation mode, varies from day to day and
from year to year. This increases the complexity of designing controlled drainage and
subirrigation systems. For most locations, it is not clear whether the greatest demands
on the system design are to provide good drainage under shallow water table condi-
tions, or to provide sufficient subirrigation to meet evapotranspiration (ET) demands
during the driest periods. For these reasons, we recommend that a simulation model
approach be used to conduct a complete analysis and final design of the water table
management system, and to predict its performance over 20 to 30 years for the clima-
tological conditions at the site. Steady-state solution approaches, such as the Hoog-
houdgt solution with the use of locally determined design drainage rates (Skaggs and
Tabrizi, 1986), can be used to obtain a relatively good first estimate of the required
drain spacing needed in the conventional subsurface drainage mode. When operated in
the controlled drainage or subirrigation mode, steady-state solutions presented by
Ernst (1975) can be used. The computer program SI-DESIGN (Belcher et al., 1993)
also provides a rapid estimate of the required drain spacing for a design storm. These
estimates will reduce the number of simulation trials needed to determine a optimum
drain spacing. Otherwise, a greater range of drain spacings, from too narrow to too
wide, will need to be tried in subsequent simulations to determine an optimum drain
spacing.
18.3.2.1 DRAINMOD. The computer simulation model DRAINMOD (Skaggs,
1978, 1980; NRCS, 1994), provides one of the more comprehensive methods of relat-
ing water management system design to soil properties, climatological conditions,
crop requirements (particularly corn and soybeans), and management alternatives
(Fouss et al., 1987a). The model was developed specifically for subsurface drainage,
controlled drainage and subirrigation system analysis, and considers the influence of
rainfall infiltration, evapotranspiration, surface runoff, subsurface drainage or irriga-
tion, and seepage on water table depth. The model avoids complex and time-
consuming numerical computational methods by using simplifying assumptions such
as “drained to equilibrium” conditions directly above the water table. This allows per-
sonal or workstation computers to run simulations to evaluate various system design
options. DRAINMOD was not intended, however, to predict or simulate transient
changes in water table depth (or shape) that may occur with large or frequent changes
in the drainage outlet water level as with controlled water table management systems
(Fouss, 1985). The model can be satisfactorily used to design controlled systems pro-
vided outlet water level changes are infrequent, such as for manually controlled sys-
tems; or are small (e.g., +0.15 m) where automated changes in outlet water level occur
no more frequently than 3-day intervals. More complex simulation models are avail-
able for the designer to use in evaluating the performance of selected facilities or de-
sign options for fully automatic control of the outlet water level (particularly pumped
drainage from sump outlet structures); see Fouss and Rogers (1992).
The procedures for using DRAINMOD to conduct a series of simulations in a sys-
tematic manner to arrive at an optimum drain spacing and the best weir setting(s) for
the outlet water level control during the growing season are documented by Nolte
(1986), Evans and Skaggs (1989), and NRCS (1994). Thus, a detailed design example
is not given here. This chapter includes a summary of the key points and additional
simulation approaches specific for controlled water table systems. Evans and Skaggs
(1989) suggest that the drain spacing calculated from one of the shortcut methods,
such as the Ernst (1975) equation, can be used as a starting point. An approximate
weir depth is chosen and simulations are conducted for a range of spacings, at least two
less than and two greater than the first estimate. The relative yield is plotted to deter-
mine the spacing giving maximum yield. Additional simulations can be conducted for
three or more weir settings for the controlled drainage and subirrigation period of op-
eration. These simulations provide data to perform an economic evaluation (Evans et
al., 1988b), and the drain spacing and weir setting that produces the highest projected
net profit is selected as the final design. For the final design selection, an additional
evaluation step is suggested. This involves conducting simulations where the start-up
time for subirrigation (i.e., raising the weir) is changed, e.g., starting irrigation pump-
ing 2 to 3 weeks after planting for corn and varying the start time for 7 to 10 days until
tasseling. The final design evaluation involves water usage and pumping cost for the
different start-up times. DRAINMOD does not include economic or environmental im-
pact analyses, so these must be performed externally (see Section 18.3.3).
The additional simulations to optimize drain spacing and outlet water level control
depths (elevations) can be performed somewhat easier with the revised model,
DRAINMOD with Feedback Control (Fouss, 1985; Fouss et al., 1989). This version of
the model has a subroutine to simulate adjustments of the outlet water level (weir) at
pre-selected (input) dates and levels, at different levels based on rainfall amount and
monitored field water table depth (i.e., water table depth at the midpoint between
drains), and for periodically adjusted outlet water levels based on feedback signals of
monitored water table depth. This model has direct application in designing and select-
ing the final operational method for the water table management system; and, it is
noted here because of its potential application in arriving at the final design drain
spacing and outlet water level control guidelines. Fouss (1985) and Fouss et al. (1990)
suggest conducting complementary simulations to evaluate performance of the se-
lected system design (depth and spacing) for the wettest and the driest years during a
30-year period, and to compare the performances of different methods or systems for
control of the outlet water level. For most systems, the outlet water level should be
held relatively constant at some optimum elevation if the system performance and
crop yields are acceptable. Such a simplified operational procedure minimizes the
need for frequent adjustments or for more complex control (i.e., feedback automatic
control) of the outlet water level, thus reducing the cost of the total system.
DRAINMOD-NII (Youssef et al., 2005) allows the drainage design evaluation to
include a comprehensive analysis of the impact of system design and operational pa-
rameters on transport of various forms of nitrogen within the soil profile and losses in
runoff and subsurface drainage flow. This version of DRAINMOD is an important
tool for designing drainage water management systems, along with their operational
plans, to meet emerging water quality requirements. It does not have automated con-
trol or operational management options incorporated into the code, other than those
operational changes often made manually (e.g., outlet weir depth) on specific calendar
dates that could be input to the simulation model. A revision of DRAINMOD-NII is
planned to include routines for operational management options, including automated
control.
18.3.3 System Layout and Sizing Components
For both controlled drainage and subirrigation, a relatively constant depth to the
water table is needed during much of the growing season. To accomplish this, the field
should be divided into zones where the surface elevation does not vary more than 0.3
to 0.45 m. Within each zone, regularly spaced lateral pipes carry water to collector
pipes which deliver the water to the system outlet. For subirrigation, those same “col-
lector” pipes deliver irrigation water to each zone.
If full pipe flow without pressure is assumed, Manning’s equation relates pipe
roughness, flow area, hydraulic radius, and hydraulic grade line (HGL), and may be
used as the basic equation for determining pipe diameter. For controlled drainage sys-
tems, the collector pipes are sized for the maximum drainage rate needed for optimum
performance of the system with the HGL taken as the bottom slope of the collector
pipe. For subirrigation systems, the pipe often must be sized for both the maximum
drainage rate with the HGL set equal to the bottom slope as well as the subirrigation
rate with the HGL a function of the weir settings while subirrigating.
A number of design aids are available to assist with sizing the collector pipes.
These aids take the form of tables, nomographs, slide rules, computer spreadsheets,
and computer programs. For examples, the MAIN module of SI-DESIGN (Belcher et
al., 1993) can be used for determining the needed diameters of the submain and main
collector pipes. The user describes the system layout by appropriate input data. The
module calculates collector pipe design parameters such as drainage coefficient, subir-
rigation rate, subirrigation water table depth, and pipe depth. Results from the
LSPACE and MAIN modules can be used in a COST module to estimate the cost for
installing the system.
An accurate and detailed topographic map is essential for properly designing a wa-
ter table management system. During operation of the system in the subirrigation
mode, it is important that the shallow groundwater be maintained at a relatively uni-
form depth below the soil surface. Unless the land is nearly flat, this requires dividing
the field into control zones of nearly uniform surface elevations.
Burnham and Belcher (1985) described a laser-plane type surveying system that
aids in the design and layout of water management systems. The laser equipment
components can be used for land surveying as well as automatic grade and depth con-
trol on machines installing subsurface drains or doing land grading (Kendrick-
Peabody, 2004). For land surveying the laser-plane receiver unit is mounted onto a
four-wheel drive vehicle and the receiver elevation outputs are automatically read and
stored directly by an interfaced portable PC as each preselected grid point in the field
is passed by the vehicle. Each grid crossing is determined by a ground travel meas-
urement unit connected to the computer logic circuit as the vehicle travels along prese-
lected and marked lines in the field, e.g., about every 30 m. Typically 32 hectares (80
acres) can be covered in two hours. After the fieldwork, data are downloaded from the
portable PC and commercially available software is used to plot topographic-contour
maps and three-dimensional views of the field. An electronic interface between design
and installation is in the development phase. It is anticipated that a laser control sys-
tem mounted on installation equipment will be capable of coordinating and guiding
the machine during installation, controlling depth and grade by using data downloaded
from the design computer. Currently, commercially available global position systems
(GPS) are utilized to obtain field survey information and during drain pipe installation.
There are other specialized software packages available to aid the engineer in de-
veloping topographic maps, laying out subsurface drainage and water table control
systems, compiling a list of materials needed, and estimating materials and installation
costs. Examples include SUBDRAIN (Bottcher et al., 1984), from Cornell University,
for subsurface drainage design, and LANDRAIN (Sands and Gaddis, 1985), a com-
puter-aided design (CAD) program for subsurface drainage systems, with automatic
drainpipe sizing routines. In both cases, the drain spacing must be predetermined by
other methods. A companion program, LANDIMPROVE (Sands and Gaddis, 1985), is
available for land leveling and grading design. And finally, there are now commercial
laser-plane and GPS-based systems available to assist the designer or contractor in this
regard (Grandia, 2002; Welch, 2002).2
18.3.4 System Operational Objectives
The overall design and operational objectives for water table management systems
were outlined in the previous sections. The specific operational objectives for a water
table management system can vary with crop, soil, climate, and topographic condi-
tions, but the following objectives apply to most systems designed for humid regions:
provide trafficable soil conditions required for conducting timely farm field op-
erations, without damaging soil structure (note that drainage of the soil profile to
a greater depth may be necessary during the time required for the field opera-
tions).
minimize the frequency and duration of excess soil water in the root zone caused
by rainfall as well as deficit soil-water conditions during droughty periods;
prevent over-drainage of the soil profile, thus maintaining the water table at a
depth shallow enough to provide a soil-water supply for crop roots (and also re-
ducing agrochemical losses in any drainage discharge); and
minimize the need for pumping subirrigation water by efficient use of infiltrated
rainfall.
For topographic conditions that dictate a pumped subsurface drainage outlet (e.g., a
gravity outlet is not available or possible), minimizing pumping (energy) requirements
for the subsurface drain effluent may also be an important objective. As noted earlier,
the initiative to promote and implement drainage water management technology (con-
trolled drainage) to improve drainage effluent water quality may be only the beginning
of changes in drainage and water management design objectives. In the future, more
precise integrated management of water, fertilizers, and pesticides may be required to
achieve and sustain acceptable surfacewater and groundwater quality.
18.3.5 Selection and Design of Operational Methods
This section outlines several methods that have been developed to operate dual-
purpose controlled-drainage and subirrigation systems for water table management.
For both manual and automatic operation of water table management systems, basic
principals of operation must be understood and followed in order to meet the opera-
tional objectives.
18.3.5.1 Manual operation. Many dual-purpose systems in humid regions are op-
erated manually. Proper manual operation requires skill of the farmer or manager and
attention to system responses and performance. Frequent visits to the field are required
during the growing season to monitor water table depth.3
Drainage rate is commonly controlled by a manually adjusted weir at the drain out-
let for each zone or in the main outlet ditch, in conduit mains or submains, or perhaps
in individual lateral drainlines. Depending upon the size and length of an individual
outlet ditch controlling a zone, the response of the water table depth in the field may
2
Trade and company names are included in this monograph for the benefit of the reader and do not imply
endorsement or preferential treatment of the product listed by USDA or cooperators.
3
Manual water table measurements are commonly made in observation wells fabricated from small-
diameter, perforated, plastic pipe installed vertically in the ground midway between drainlines. In some
soils a filter sock and sand backfill around the well are required. A blowtube is typically used to measure
the water table depth.
be relatively slow (several hours) because of the large amount of water that must to be
removed (drainage) or added (subirrigation) to change the water level in the ditch,
Figure 18.2. Examples of flow-limiting devices for subsurface submains or mains are
illustrated in Figures 18.3 (a) & (b); the field water table response with this type of
outlet water level control is more rapid than where an outlet ditch is used.
When excessive rainfall occurs, the water level at the weir or outlet may be lowered
to drain depth to allow the system to function in a conventional subsurface drainage
mode to lower the water table more quickly. For manually operated systems, the low-
ering of the weir or adjusting of a flow-limiting device to “full open” is often delayed
until after the storm, when excessive soil-water conditions already exist. A major deci-
sion for the farmer is when to raise or reset the outlet water level. Delaying too long
may lead to overdrainage of the soil profile, thereby reducing potential agrochemical
losses in soil-water discharges and accelerating the need to irrigate if subsequent rain-
fall does not reestablish the desired soil water in the root-zone. If the farmer waits un-
til the observed (measured) water table midway between drains recedes to pre-storm
depth, overdrainage of the soil profile may occur, especially for fine-textured soils.
During periods when subirrigation is needed, the outlet weir is raised (manually) and
irrigation water may be pumped into the outlet weir structure periodically as needed
during selected hours each day. A float-activated valve on the irrigation water line
may be used to hold the water level nearly constant just below the weir’s overflow
elevation. If an irrigation water supply is not available, the outlet control weir is raised
to control subsurface drainage and to capture rainfall in the soil profile up to the eleva-
tion of the weir. Any excess soil water from rainfall overflows the preset weir and the
field water table slowly recedes to the elevation of the weir.
18.3.5.2 Automated operation. Dual-purpose system automation provides an al-
ternative to labor-intensive system management methods. Automatic control of the
system can take on many different options or modes. Fully automatic or semi- auto-
matic control of the outlet water level, with or without feedback of the monitored
Figure 18.2. Flashboard riser type structure used in open ditch. Considerable time
may be required to raise or lower the water level in the ditch due to the ditch volume.
(a)
(b)
Figure 18.3. (a) Weir-type flow-limiting device within a drainline or submain riser
pipe; (b) Float-activated flow-limiting valve within a drainline or submain riser pipe.
water table depth in the field, allows the field water table to be maintained within pre-
determined depth limits. For purposes of discussion here, it is assumed that the
drainlines are connected into a sump structure for control of the outlet water level. The
water level in the sumps during drainage cycles can be controlled by pumping from
the sump structure into a surface drainage channel (see Figure 18.1). During subirriga-
tion, the water level in the sump outlet control structure is maintained within a prede-
termined range by pumping from an external source such as a well. Automatic control
of the sump water level (SWL) to regulate subsurface drainage discharge (i.e., con-
trolled drainage) and subirrigation flow into the soil profile may additionally include
options with or without feedback of the field-monitored water table depth (WTD) be-
tween drainlines. Automatic control with feedback is much preferred. A cross-
sectional schematic of the sump structure and WTD sensor4 in the controlled-drainage
4
Fouss and Rogers (1992, 1998) and Fouss et al. (1999) provide detailed discussions on the various types of
water table depth (WTD) sensors that can be used in automated control systems.
mode of operation is shown in Figure 18.4. The mode of operation is automatically

switched by the system controller from controlled-drainage to subirrigation, and vice
versa, as needed to maintain the SWL between the maximum and minimum water
level elevations specified. Brief descriptions of various methods of control are given
below. It should be emphasized here that over-aggressive drainage to maintain or
lower a water table that fluctuates quickly to a shallow depth during rainfall events
may result in excessive loss of agrochemicals in the drainage discharge. Thus, an op-
erational balance needs to be followed between controlled subsurface drainage and
subirrigation modes to prevent excessive short-term agrochemical losses in drainage
discharge (ADMS-TF, 2005).
Float-switch activated control. A combination of four float-activated electrical
switches can be configured to operate the drainage and irrigation pumps in an outlet
sump structure. A single-float activated switch can be used at the maximum SWL ele-
vation to change the system operation from subirrigation to controlled drainage. An-
other single-float switch can be used at the minimum SWL elevation to change opera-
tion from controlled drainage to subirrigation. A double-float activated switch can be
used to operate the drainage pump between high-drainage (HD) and low-drainage
(LD) sump water levels. Another double-float switch is used to operate the irrigation
pump (or valve) between the low-irrigation (LI) and high-irrigation (HI) sump water
levels. This type of control mode does not involve feedback of the WTD in the field.
The elevations of the float switches typically require manual repositioning to adjust
the SWL control threshold elevations, that is, the MIN, LD, HD, LI, HI, and MAX
water level elevations in the sump. The subsurface drainage and subirrigation pumps
(or valves) can also be operated by manual electrical switches, if desired; this is com-
monly done to periodically check the functioning of the drainage pumps and irrigation
pumps or valves.
Microprocessor-controlled system. This is a “fully” automated control system. In
the controlled-drainage mode, without feedback of the field WTD, the SWL is main-
Figure 18.4. Controlled-drainage mode of sump-controlled water table management.

WTD = water table depth, HI = high-irrigation, LI = low-irrigation, HD = high-drainage,
and LD = low-drainage sump water elevations.
tained between preset HD and LD elevations in the sump (Figure 18.4) by pumping
water from the sump. These desired HD and LD elevations (on/off electrical switch
levels for the drainage pump) may be stored as software values in a microprocessor
controller system. When subsurface drainage flow ceases and the SWL recedes to the
minimum elevation (via subirrigation flow from the sump), the system operation is
switched to the subirrigation mode. With feedback of the field WTD, the microproces-
sor controller system will operate the drainage pump as described above for controlled
drainage without feedback, except if the monitored field WTD is outside of a desired
depth range (WTDmin ≤ WTD ≤ WTDmax) during the previous 24-h period (midnight
to midnight). If the WTD is out of range, the SWL control threshold elevations (MAX,
HI, LI, HD, LD, and MIN) are automatically adjusted upward or downward, as appro-
priate, by an amount Y stored in the software in the microprocessor controller pro-
gram, for the next 24-h period. Once the field WTD returns to the desired range, the
SWL control thresholds are returned to the predetermined standard elevations (MAXS,
HIS, LIS, HDS, LDS, MINS) by the controller unit the following midnight. The sys-
tem operation is switched to subirrigation anytime the SWL falls below the minimum
elevation.
In the subirrigation mode, without feedback control, the SWL is maintained be-
tween the low-irrigation and high-irrigation elevations by pumping into the sump from
a nearby well. A microprocessor controller system, as described above for the con-
trolled-drainage mode, can be used to operate the irrigation well pump. As gravity-
flow subirrigation from the sump lowers the SWL to the low-irrigation elevation, the
irrigation pump is operated to raise the SWL to the high-irrigation elevation. The field
WTD is not monitored in this mode. With feedback control, the sump operation is the
same as subirrigation without feedback, except the SWL control threshold elevations
are adjusted as described above for controlled drainage with feedback, whenever the
monitored field WTD is outside the desired range for a midnight-to-midnight 24-h
period. The same feedback adjustment parameter Y as used for the controlled-drainage
mode may be used for the subirrigation mode, or a different adjustment parameter Z
can be selected, if desired. The value of Y or Z will depend on the system design, soil
characteristics, water table response, etc., but typically is set to a value of about 10%
to 15% of the drain depth. If rainfall occurs when the system is operating in the subir-
rigation mode, and the cumulative amount exceeds a threshold in a given period of
time, the system operation may be switched to the controlled-drainage mode. If the
rainfall does not exceed this threshold (which is based on experience at the site), but
infiltration is sufficient to cause subsurface drainage into the sump that raises the SWL
to the maximum elevation, the system operation will be switched to the controlled-
drainage mode.
An example of fully automated operation of a controlled drainage and subirrigation
system for water table control was reported by Fouss and Rogers (1998) and Fouss et
al. (1995, 1999), and full details on the design and objectives of the research project
were reported by Willis et al. (1991).
Semi-automatic float valve for gravity-flow systems. Where a gravity-flow outlet is
possible, an alternative method of SWL control in the subsurface drainage mode can
involve a dual-float activated valve as illustrated in Figure 18.3 (b). The larger float
provides the force to lift the outlet pipe-plug at a preset water level, and the smaller
float provides sufficient power to hold the pipe-plug and prevent it from closing again
until the water level in the sump recedes to some preset lower level. Such a valve
mechanism allows fast subsurface drainage during peak flow events, functioning simi-
lar to a two-stage weir described by Fouss et al. (1987b). The overflow pipe in the
sump allows sufficient drainage to control the SWL during smaller rainfall events
when rapid drainage is not required because excess soil-water conditions last only for
short periods of time.
Automated operation for gravity-flow systems. In humid areas, dual-purpose sys-
tems often have multiple field zones with each zone outletting to a main and the main
outletting by gravity into an open channel. Automation of such systems requires slight
modification of the methods described above for pumped-outlet systems. A computer-
based prototype automation system has been developed at Michigan State University
(Belcher and Fehr, 1990). This system is capable of minimizing the fluctuation of the
water table above or below the desired elevation, alerting the operator to problems
(such as a pump not working), providing a visual display in the office (via modem and
PC) of what is happening in the field, providing the user the ability to input the desired
water table elevation in each zone as a function of time, and storing useful operation
data throughout the growing season (water table depths, pumping duration, etc.).
The system uses rainfall, discharge of underground pipes, and feedback of moni-
tored water table level to control the on/off cycles of the irrigation pump. When it
rains, the system shuts off the irrigation pump and adjusts the flow restriction device
in each zone as needed to hold the water table near the desired elevation. When the
water table falls below the desired elevation, the system restricts drainage flow and
activates the irrigation pump. The system allows the subirrigation system operator to
monitor and modify the subirrigation operation parameters from a remote office via
PC and modem. Figure 18.5 is a schematic of the automation system for a single water
Sensor/Controller
Rainfall
Irrigation Sensor
Pump
Water Water Table

Table Depth Sensor
Submain Flow Limiting

Device
Figure 18.5. Schematic of gravity discharge subirrigation automation feedback control.
table management zone. A prototype system tested in Michigan has the capability to
handle up to 16 zones (Belcher and Fehr, 1990).
Two types of flow restriction devices have been used in this automated control sys-
tem. The first consists of a diaphragm valve placed in-line in the submain. The con-
troller restricts flow by inflating the valve and allows flow by deflating the valve. The
second device is a modification of the weir control structure shown in Figure 18.3 (a),
in which a small DC motor is added to raise or lower the bottom weir plate to block or
allow subsurface drainage flow. This type of weir control structure is commercially
available; e.g., the “Smart Drainage System.”5
Remote control. Three general types of remote control of water table management
systems are currently available: one- or two-way radio control, electronic communica-
tions from a cell phone, and digital communications from PCs and modem equipment
via telephone lines. Communications via satellite are economically feasible only for
larger projects at great distances from the control decision center. A typical application
of remote control would be to change the mode of operation at one or more outlet
(sump) structures; for example, to stop pumping subirrigation water and/or to permit
drainage of the system in advance of predicted heavy rainfall (see below on use of
weather forecasts). Two-way radio communication permits confirmation that the op-
erational change was actually made at the remote site. Electronic communications via
a PC and modem to a microprocessor-controller system for sump structures permits
more sophisticated remote control, or provides a means to override an automated con-
trol system. Such remote control or control override can permit the manager to change
the sump water trigger levels for controlled-drainage, subirrigation, or other mode
switch levels, by merely changing stored values in the microprocessor program. The
microprocessor system also permits monitoring of field site parameters, such as water
table depth, rainfall amounts, or mode of operation from the PC in the control center.
Seasonal modes of operational control. In the humid region of the U.S. and Can-
ada, removing excess soil water from the field is the system’s most important role.
Drought conditions in soils of this region are generally temporary in nature, and thus
the system functions in the controlled drainage or conventional subsurface drainage
mode the majority of the time. When rapid drainage is required, the outlet water-level
control structure should be set to the depth of the subsurface drainage conduit to pro-
vide the maximum drainage rate; it needs to be recognized that this may cause exces-
sive losses of applied agrochemicals from the soil profile while rapid drainage is per-
mitted. When trafficable field operations are required, the outlet water level should be
maintained near or slightly above the drain depth until after the crop has been planted.
The recommended operational practice during the winter (non-cropping) months is to
keep the system in the controlled drainage mode to maintain the water table shallow in
the soil profile, which reduces nitrate losses because drain outflow is decreased and a
denitrification zone can develop in the upper soil layers (Gilliam et al., 1979, 1999).
During the growing season, the outlet water level should be controlled (by one of the
methods described above) to maintain the field water table depth within the desired
range relative to the crop root zone to provide a steady supply of water to plant roots
by upward flux. A week to ten days prior to crop harvest, the outlet water level should
5
Agri Drain, Inc., 1462 340th St., Adair, IA 5002: info@agridrain.com.
again be lowered to near or slightly above drain depth to provide good trafficable con-
ditions for harvesting operations, and to reduce damage to the soil structure by equip-
ment traffic. Following harvest, the operational-control annual cycle described above
is repeated.
Use of weather forecasts to aid in operational management. In many areas of the
humid region of the U.S., the predicted probability of rainfall and the estimated daily
amount of rainfall in National Weather Service forecasts (e.g., the daily, three- and
seven-day forecasts) are becoming sufficiently accurate to permit adjustment of
day-to-day operation of water table management systems, and to aid in the timing of
fertilizer and pesticide applications (Fouss and Willis, 1994; Schneider and Garbrecht,
2006). Principal management objectives may be to reduce the occurrences of severe
excess soil-water events, and the duration of deficit soil-water conditions, improve the
efficiency of utilizing rainfall received, thus minimizing the need for pumping irriga-
tion water, and increase the effectiveness of fertilizers and pesticides applied and re-
duce the potential of losses.
For example, if large rainfall amounts are predicted, the system operation may be
switched from subirrigation to controlled-drainage or conventional subsurface drain-
age several hours to a day in advance of the predicted rain. Also, following the rainfall
and the subsequent recession of the water table to the desired depth, the restarting of
subirrigation may be delayed if significant rainfall is predicted to occur again within
the next two to three days. The minimum percent probability of rain for which these
actions may be taken can differ in various geographic regions because of the regional
accuracy of the forecasts (Schneider and Garbrecht, 2006). Expert advice or experi-
ence will be needed to determine the minimum percent probability that best applies in
a given area. Use of weather forecasts to aid in operational management is best justi-
fied for high-value crops, such as vegetables, and crops that are most affected by wet
soil conditions.
18.4 SYSTEM DESIGN AND OPERATION IN ARID REGIONS
18.4.1 Characterizing Climate Conditions
In humid regions the source of water used in water table control systems is either
rainfall percolating through the soil, which is captured by controls on the drainage
system, or water pumped into the drainage system. The random nature of rainfall
makes it difficult, if not impossible, to determine the availability of rainwater in a wa-
ter table control system. When the drainage system is being used for subirrigation and
a water source is available, it is possible to pump water into the drainage system, as
needed, to maintain the water level. In this case the water supply is a stream or river,
or ponded water.
In arid and semi-arid areas the source of excess water being used in a water table
control system is generally deep percolation from irrigation on the subject field or
lateral flow from an adjacent field. Rainfall amounts in these areas are generally lim-
ited and add little water to shallow groundwater. For example, the west side of the San
Joaquin Valley is semi-arid, averaging 150 mm of rainfall each year, generally during
the winter months. The average rainfall event is 5 mm or less and is not effective in
providing water for either winter crop use or deep percolation. Rainfall can be more
important in other regions, such as southeast Australia, leading to shallow water tables
after winter or sporadic heavy rainfall. These make irrigation management more diffi-
cult; water table control can assist in making better use of these rainfall contributions.
Generally, as rainfall is not a significant contributor to the total water supply, it is pos-
sible to estimate the shallow groundwater availability based on the water requirements
of the crop rotation and the irrigation system efficiency and practices.
18.4.2 Characterizing Shallow Groundwater Conditions
Shallow groundwater is affected by the geology of the area and the irrigation prac-
tices on adjacent fields, in addition to the irrigation practices on the field itself. In ar-
eas with soils developed from alluvial deposits, there are often sand stringers or zones
of soil with high hydraulic conductivities passing from one field to an adjacent field at
depths less than 2 m. In these instances, deep percolation in adjacent fields is transmit-
ted to the field of interest and water table control becomes significantly more difficult.
In humid areas, shallow groundwater quality conditions are characterized by the pres-
ence of manmade chemicals such as nitrate, other fertilizer elements, and pesticides. In
arid areas shallow groundwater may contain salt and trace elements, in addition to
man-made contaminants, making it more difficult to characterize its quality.
The remainder of this section will discuss the impact of salt and trace elements in
the groundwater on shallow groundwater management. Many of these topics are dis-
cussed in more detail elsewhere in this monograph. However, basics concepts are re-
viewed here to provide a basis for the discussion of design and management of drain-
age systems used for water table management in arid areas.
18.4.2.1 Soil salinity. Sources of salinity include irrigation water, fertilizers, and
natural sources such as salt deposited from the ocean during the formation of the soil.
Regardless of its purity, all irrigation water contains salt. Irrigation water with an elec-
trical conductivity of 0.3 dS m-1 is considered suitable for use on all crops; however,
even this contains approximately 200 mg L-1 of salt, which accumulates in the soil as
crops remove water.
In arid and semi-arid areas, it is quite common for soils to contain salt of geologic
origin. The Mancos shale, the parent material of the soils in the Grand Valley of Colo-
rado, contains lenses of salt deposited while this area was being formed under the
ocean (Walker et al., 1979). The soils on the west side of the San Joaquin Valley of
California were derived from the Coast Range, also marine sediment, and as a result
contain high concentrations of salt and elements such as boron, selenium, arsenic, and
other trace elements (Deverel and Fio, 1990). In southeast Australia the huge stores of
salt found in the landscape were deposited over geologic time by dust and rainfall
transport (Herczeg et al., 2001; Ladaney-Bell and Acworth, 2002).
The salt concentration in the soil profile generally increases with depth in arid and
semi-arid areas of irrigated agriculture. Soils in the irrigated regions of semi-arid
southeast Australia often have a 10-fold increase in salinity between 0.1 and 2 m
(Hornbuckle and Christen, 1999). The surface soils are leached and the salts are
moved deeper into the profile, so the salt concentration in the surface soil is low
enough that neither seed germination nor early plant growth is adversely affected.
If the water table is close to the soil surface, it is possible for evaporation to move
water and salt up from shallow saline groundwater and accumulate salt on the soil
surface. When a crop uses significant quantities of water from shallow saline ground-
water the salt moves with the water and is deposited in the root zone. Either situation
creates an inverse salinity profile in which the soil salinity decreases with depth. This
type of profile should be managed to prevent deleterious effects on seed germination
and plant growth. Pre-plant irrigation during a fallow period is one method routinely
employed in the San Joaquin Valley of California to leach salts and restore the salinity
profile prior to planting.
18.4.2.2 Leaching. Leaching is required to both maintain a salt balance in the root
zone and to prevent inverted salinity profiles. Assuming piston displacement and no
contribution of salinity by groundwater, the minimum fraction of applied water needed
to maintain a salt balance is referred to as the leaching requirement (Lr ). This is estab-
lished based on the salt input and the crop salinity tolerance and is the minimum vol-
ume of water that must pass through the soil profile to maintain the salt balance. The
leaching requirement is in addition to the crop water demand. The requirement can be
estimated from
Lr = Dd*/Da = ca/cd* (18.1)
where Dd is the equivalent depth passing below the root zone, Da is the equivalent
depth of application (irrigation plus rain) (Hoffman, 1990), ca is the weighted mean
concentration of the applied water and cd* is the salt concentration in the water passing
below the root zone. The required values versus the actual are designated by asterisks.
When the water table is being managed to provide part of the crop water requirement,
ca needs to be modified to include the salinity of the groundwater and the percentage
of water used by the crop (Fouss et al., 1990).
The leaching fraction (Lf) is the actual amount of water passing through the soil
rather than the required amount. The leaching requirement is often met through irriga-
tion inefficiency and/or of the irrigation system distribution uniformity, or even in
some cases by winter rainfall, thus additional water is not often required for leaching.
This is particularly true when a surface irrigation method, such as flood/border or fur-
row, is used.
Water passing through the root zone moves to the shallow groundwater where it is
either collected and discharged for disposal through a subsurface drainage system or
percolates to the regional groundwater, or if the water table remains shallow the
groundwater can be evaporated leading to salt accumulation in the surface layers.
18.4.2.3 Crop salt tolerance. The crop salt tolerance, soil salinity, applied water
quality, groundwater quality, and stage of growth will establish the potential of shal-
low groundwater for meeting crop water requirements. Crop salt tolerance has been
evaluated through experimentation as ranging from sensitive to tolerant. The tolerance
is characterized mathematically using the Maas-Hoffman equation (Maas and Hoff-
man, 1977), which sets a salinity threshold at which yields begin to decline. The rate
of decline varies with a crop and its salt tolerance.
The studies used to characterize plant salt tolerance were done under conditions of
uniform root-zone salinity, a condition not generally found in the field. Other research
has demonstrated that the average root-zone salinity (Shalhevet, 1994) is the most
important factor to consider when characterizing the tolerance and the distribution of
salt in the soil profile. Crops will selectively extract water from the least saline por-
tions of the root zone. As the soil profile becomes more salinized, there will be less
water available for crop use and potential yield losses will be realized.
Where crops use water from a shallow water table, it has been found that the crops
apparently use water at higher salinity levels than predicted with the Maas and Hoff-
man parameters without yield decline. The reasons for this are not well understood;
since crops generally only receive part of their water requirement from a water table
(saline) and the bulk of their water requirement from irrigation (non-saline) and the
overall volume-weighted salinity of the water used by the crop may be tolerable. This
can be demonstrated using data from Ayars and Schoneman (1984) for a cotton crop.
The estimated crop evapotranspiration was 647 mm with 429 mm applied low salinity
(2 dS m-1) and 167 mm of upflow of saline (7 dS m-1) water. The weighted average of
the water supplied by these two sources is 3.13 dS m-1. This value is less than the
threshold value of soil salinity that would result in a yield reduction. While this value
is less than the threshold, the time of uptake of water from shallow groundwater also
has an effect since much of the extraction occurs late in the growth cycle when the
plant is most salt tolerant.
There may also be some uncertainty in the actual salinity of the groundwater and
hence the water that the crop is taking up. This can be due to the spatial variation in
groundwater salinity across a field and also the temporal variation as water tables rise
and fall. There is also variability introduced by sampling groundwater at depths below
that at which the groundwater is interacting with the root zone (Northey et al., 2006).
18.4.2.4 Drainwater quality. Drainwater quality is generally not an accurate de-
scription of the shallow groundwater quality being used in a water table control sys-
tem. The shallow groundwater below the root zone is a mixture of the salts in equilib-
rium with the soil salinity at that depth and the deep percolate from the applied irriga-
tion water. Drainwater from a subsurface drain is a mixture of the shallow groundwa-
ter and deeper groundwater. In many arid areas the water quality becomes progres-
sively poorer as the depth in the soil profile increases. Grismer (1990) demonstrated
that as a drain spacing gets wider or the drain depth gets deeper, proportionately more
water is taken from deeper in the profile, which results in a poorer water quality than
expected from sampling shallow groundwater. Christen and Skehan (1999) also dem-
onstrated that drainwater salinity increases as the depth to the water table increases
and hence flow paths to the drains become deeper. They found a 50% increase in drain
flow salinity from when the mid-drain water table was at 1 m (8 dS m-1) and at 1.6 m
(11.5 dS m-1). Water quality sampling of shallow groundwater using wells installed at
the appropriate depths will provide a better characterization of the shallow groundwa-
ter quality than will drainwater samples.
18.4.2.5 Depth to groundwater. In irrigated areas the source of shallow ground-
water is oftentimes deep percolation from inefficient irrigation and, as a result, the
groundwater fluctuation responds to the irrigation management in the area. Early in
the season irrigations are difficult to schedule and match to the restricted root zone of
young plants and often even pre-plant irrigation is practiced. These early-season irri-
gations lead to large volumes of drainage past the root zone so that groundwater is
often closest to the soil surface at the beginning of the irrigation season. As the irriga-
tion season progresses, the depth to water table increases through a combination of
lateral and vertical flow in the groundwater system and plant uptake. If the irrigation
system is inefficient, it is possible to have a condition where the depth to groundwater
decreases until the end of the irrigation season at which time it gradually increases.
For example, Figure 18.6 illustrates groundwater depth data showing an increase in
water table elevation under furrow-irrigated tomato plots in a field without subsurface
drains. The depth to groundwater affects the total uptake by the plants and the time the
uptake begins (Ayars et al., 2006).
Figure 18.6. Water table response under two furrow irrigated tomato plots
(NCFA, NCFB) in a field without subsurface drainage.
18.4.3 Irrigation System Selection, Management, and Operation

Irrigation system selection will depend on the soil, crop, and financial conditions of
the enterprise and the preferences of the manager. The available systems can be sub-
divided into either pressurized or non-pressurized and within each category there are
many options. Pressurized systems include sprinklers (hand-move), center-pivot, lin-
ear-move, drip systems, and microsprays. Non-pressurized systems (surface irrigation
methods) include level basins, furrows with surges, and/or gated pipe systems. The
design of these systems is discussed elsewhere in this monograph. The important as-
pects of irrigation system design for this section are the potential irrigation efficiency,
uniformity, and management.
Non-pressurized irrigation systems are characterized as the least efficient of sys-
tems with irrigation efficiency and distribution uniformity in the range of 50% to 80%,
resulting in excess deep percolation. The infiltration rate of the soil surface limits the
potential irrigation efficiency and distribution uniformity because of the variability of
the infiltration with time and space. It is very difficult to apply small depths of water
with surface irrigation with the exception being level-basin systems that apply as little
as 55 mm in an application (Dedrick et al., 1982). Surface systems work best on soils,
such as clays and clay loams, because of the low infiltration rates compared to sands.
Because pressurized systems have the potential for high uniformity of application,
high distribution uniformities, and small depths of application, they enable better con-
trol of deep-percolation losses. The advantage of pressurized over non-pressurized
systems is that infiltration is controlled by the application rate of the sprinkler rather
than the soil surface. With pressurized systems it is possible to apply irrigations as
small as 2 to 4 mm, three to four times daily, as compared to surface systems that are
limited to 50 mm as the minimum application.
Irrigation scheduling determines the time and depth of application. In shallow
groundwater conditions the timing and the depth of application can be altered by the
crop water use from shallow groundwater. The timing is altered because a portion of
the water requirement is met from the groundwater, not the stored soil water, and tra-
ditional volume-balance calculations assume all water use is from stored soil water
(Ayars and Hutmacher, 1994). Thus, groundwater contribution extends the interval
between irrigations when timing is based on soil water depletion. If saline groundwa-
ter is being used by plants, salts are being moved into the root zone and the potential
exists for an increase in soil water osmotic potential stress, which is added to the ma-
tric potential. The plant will respond to the combined stresses and indicate an earlier
irrigation than would matric stress alone. In this case less water will be removed from
stored soil water than indicated by matric stress alone.
Alternative methods have been developed to determine plant stress levels that can
be used as indicators for irrigation timing in shallow groundwater areas. Leaf water
potential, which reflects osmotic and matric stress, has been used successfully to
schedule irrigation in cotton (Ayars and Schoneman, 1984; Kite and Hanson, 1984).
This is a valuable technique and is applicable for crops having the stress values needed
for irrigation scheduling.
The crop water stress index has also been shown to reflect both matric and osmotic
stress using infrared thermometry to determine plant temperature. The temperature
data are used with vapor pressure data to determine the index (Howell et al., 1984).
After reaching a critical value, irrigation is indicated. This technique is limited to the
time after full canopy cover is reached to prevent background soil temperature read-
ings from affecting the results. At present its use is limited because of the lack of data
needed to establish the critical index values needed for irrigation scheduling.
After the time to irrigate is established, the depth of application is determined. Cur-
rently this is done by gravimetric analysis or using neutron attenuation to establish the
water depletion in the interval between irrigations. New devices to determine soil wa-
ter content which offer promise include time-domain reflectometry and capacitance
systems. Ayars and Hutmacher (1994) demonstrated a modified cotton crop coeffi-
cient that accounted for the groundwater contribution to crop water use (Figure 18.7)
as a function of groundwater salinity and depth enabling a volume-balance calculation
for soil water depletion by the crop.
Figure 18.7. Basal cotton crop coefficient and regressions of modified crop coefficients
derived from lysimeter data for five groundwater qualities and at various depths.
18.4.4 Subsurface Drainage System Design

for Shallow Groundwater Management
Design possibilities for both existing and new systems will be considered in this
section. There are millions of hectares of drained irrigated land which might be con-
sidered for shallow groundwater management. The ability to control flow from the
drains and to maintain the water table at a desirable level over a significant portion of
the field is the major factor in determining the suitability for the field for modifica-
tions to control shallow groundwater. Also, the cost to modify the system and the po-
tential disruption of cultural operations in the field should be considered in the design.
Previous drainage system modifications have included use of weir type flow restrict-
ing valves (Fig. 18.3 a, b) installed on drainage laterals (Lord, 1987) to control the
water table.
For example, an existing drainage system on a 65-ha field on the west side of the
semi-arid San Joaquin Valley was modified by installing butterfly valves on each of
the seven drainage laterals (Ayars, 1996) to test possible control structures. In addi-
tion, weir structures were installed along the submain to provide regional control of
the water table. The schematic of the system is given in Figure 18.8. In this system the
drainage laterals were installed perpendicular to the direction of cultivation in the
field. Examples of alternative water table control structures are shown in Figures 18.9
and 18.10. The structure in Figure 18.9 was used on an individual lateral in a vineyard
Figure 18.8. Schematic layout of drain system and control structures on the
shallow groundwater management study site on a field located on the west side
of the San Joaquin Valley of California.
in Australia. The weir structure in Figure 18.10 was used on the submain of the drain-
age system depicted in Figure 18.8. Figures 18.11a,b give the water table depth be-
tween a set of three subsurface drainage laterals in the experimental field for two dates
during the production of a tomato crop. The water table control system resulted in less
applied irrigation water and improved crop quality in the areas with the water table
closest to the soil surface (Ayars, 1996; Ayars et al., 2000).
Figure 18.9. Individual lateral control structure used to control the water table
in a vineyard in Australia.
Figure 18.10. Control structure used to measure drainage flow and control the
water table in California.
Figure 18.11. Depth to groundwater after closing (A) or opening (B)

lateral control valves on the field diagrammed in 18.8.
A study by Christen and Skehan (2001) in a vineyard in semi-arid southeast Austra-

lia compared no management (continuous flow) with active management of subsur-
face drains, 2 m deep and 20 m apart. The management of the system prevented drain-
age once the water table reached 1.2 m deep or when an irrigation event was occur-
ring. Without management, the drains flowed continuously during the two irrigation
seasons with a salinity of around 11 dS m-1. This resulted in a drainage salt load over
-1
the two seasons of 5867 kg ha . The management measures were able to reduce the
volume of drainage and its salinity, resulting in a 50% reduction in the salt load. The
unmanaged drains removed 11 times more salt than was applied in the irrigation water
(i.e., there was mining of geologic salt), whereas drainage management reduced this to
five times the salt applied (Table 18.1). This study also found that root-zone salinity
was successfully controlled in the managed treatment and that there was no difference
in grapevine yields between the treatments.
These experimental results indicate that drainage management in irrigated semi-
arid areas can have both agronomic and environmental benefits.
Design of drainage systems to incorporate shallow groundwater management will
require the adoption of new design criteria for the depth and placement of the drains
and the depth to water table at the midpoint between the drains (Doering et al., 1982;
Ayars, 1996). Both the drain depth and the allowable midpoint depth need to be re-
duced from the current recommendation of 2.4 m for drain depth and 1.2 m for mid-
point water table depth (U.S. Department of Interior, 1993). Changes in the design that
relax current criteria will require additional management criteria to prevent salination
of the soil profile.
The first proposed subsurface drainage design change is to set the recommended
midpoint water table depth to approximately 0.9 m for all situations. The value of 0.9
m was selected as a compromise to permit use of shallower drain depth installation
while maintaining a reasonably wide lateral spacing. It was observed in a previous
study (Ayars and McWhorter, 1985), that when crop water use of shallow groundwa-
ter is included in the drainage system design, the minimum depth to the water table
occurs early in the season when the rooting depth is shallow.
The second change in drain design criteria is to reduce the drain depth in order to
reduce the effective depth of the groundwater collection by the drainage system. How-
ever, reducing the drain depth also results in a reduction in the lateral spacing in order
to adequately control the water table position. Relaxing the midpoint water table depth
requirement will compensate to maintain a reasonable drain spacing for irrigated con-
ditions.
By reducing the drain depth and spacing, less groundwater is collected from deep
in the soil profile, and in cases where the water quality declines with increased depth
in the soil profile, less poor quality water will be extracted (Grismer, 1990). The re-
duction in drain depth will also lead to smaller volumes of water being discharged
from the drains and more water being used by the crop (Doering et al., 1982). Irriga-
tion scheduling cognizant of salinity stresses at seed germination, and later upward
flow from the water table for meeting consumptive use needs of the crop, will become
part of salinity management in the root zone required in the overall irrigation/drainage
management system.
Table 18.1. Effects of drainage management over two irrigation seasons in
semi-arid southeast Australia.
Unmanaged Drains Managed Drains
Drainage volume (mm) 70 47
Drainage salinity first irrigation (dS/m) 12 11
Drainage salinity last irrigation (dS/m) 11 7
Salt load (kg/ha) 5867 2978
Ratio of salt removed to salt applied 11 5
The proposed steps for the design of an integrated water management system in
arid areas are as follows. Develop the midpoint water table depth criterion based on
the unrestricted growth of plant roots as a function of time such that the rooting depth
does not exceed the depth to water table during the growing season. The root devel-
opment can be approximated using field data or the relationship of Borg and Grimes
(1986). The next step is to develop the irrigation schedule for the crop using a modi-
fied crop coefficient (Kc ) similar to that developed by Ayars and Hutmacher (1994) to
establish the timing and depth of application. The irrigation system efficiency will
then be used to establish the deep percolation losses to the system, which are the input
to the design program. The method of Ayars and McWhorter (1985) can be used to
calculate the midpoint water table depth using the design deep percolation and irriga-
tion schedule. The depth and spacing will be varied until the root-extension criterion is
not violated. A simple water balance program can then be used to determine the salt
accumulation in the root zone. This will provide the information needed to manage the
salt balance.
18.4.4.1 Example of drain system design using proposed changes. Drain spacing
and depth were calculated for two soils, clay loam and sandy loam, based on one year
of climate data from the west side of the San Joaquin Valley, and a common irrigation
schedule adapted for cotton grown on these soils. Two different irrigation schedules
are considered: one assuming that there is no groundwater contribution to the crop,
and one assuming a groundwater contribution. Both schedules were operated assuming
irrigation efficiencies of 60% and 80%. A program developed at the USDA-ARS Wa-
ter Management Research Laboratory that implemented the design method of Ayars
and McWhorter (1985) was used in the drain spacing and water table position calcula-
tion. The results of the designs are summarized in Table 18.2.
The results summarized in Table 18.4.2 demonstrate that improving irrigation effi-
ciency significantly affects the computed drain spacing resulting in reduced drain
flows and disposal volumes. Reduced drain flows from improved irrigation efficiency
are expected since there is less deep percolation. Also, larger drain spacing is possible
for the sandy loam soil, as compared to the spacing for the clay loam, as would be
expected. Groundwater contribution to the crop water use also increases the drain
spacing relative to the case when the groundwater contribution is not included in the
water management of the crop.
Table 18.2. Summary of drain spacing calculated using traditional and
proposed changes in drainage design criteria to account for water quality.
Water Irrigation Efficiency
Drain Table Drain Spacing (m) Drain Spacing (m)
Depth Depth (groundwater contribution) (no groundwater contribution)
(m) (m) Soil Type 60% 80% 60% 80%
1.5 0.9 Clay loam 317 341 160 320
1.8 0.9 Clay loam 445 573 228 447
2.4 0.9 Clay loam 630 833 380 625
2.4 1.2 Clay loam 543 707 299 542
1.5 0.9 Sandy loam 495 638 285 468
1.8 0.9 Sandy loam 709 926 401 642
2.4 0.9 Sandy loam 994 1381 600 890
2.4 1.2 Sandy loam 861 1168 499 773
The example design that reflects the use of the USBR design criteria is for a drain
depth of 2.4 m with a design depth to water table of 1.2 m and an irrigation efficiency
of 60%. This situation is typical of furrow irrigation in many parts of the world and is
used here as a basis for comparison with the other designs.
What was unexpected is that the design drain spacing for the drain depth of 1.5 m,
a design water table depth of 0.9 m, and 80% irrigation efficiency with no groundwa-
ter contribution to the crop water use, was 320 m for clay loam and 468 m for sandy
loam, which is similar to the results for 60% irrigation efficiency with contribution to
the crop water use, where the spacing was 317 m for clay loam and 495 m for sandy
loam. These are approximately equal to the USBR spacing in both soils—299 m in the
clay loam and 499 m in the sandy loam. From an installation cost perspective, the pro-
posed changes are not expected to be any more expensive than the USBR designs and
probably less. The alternative designs meet the proposed new criteria of shallower
installation and shallower depth to midpoint water table.
The results show that there are two ways to achieve the new design depths. The
first way is to improve the irrigation efficiency from 60% to 80%, which is possible by
changing from furrow to pressurized systems or aggressively managing the furrow
system. The second way is to include the crop water use from shallow groundwater to
offset the inefficiency in the irrigation system, yielding approximately the same calcu-
lated drain spacing.
Similar conceptual development of drainage design and management practices to
reduce irrigation water losses, improve crop water use from water tables and reduce
drainage salt loads have been developed for subsurface drainage in irrigated land uses
in Australia (Christen and Ayars, 2001). This has come about as a response to the
ever-increasing restrictions upon subsurface drainage water disposal into waterways
due to the high levels of salts and, to a lesser extent, agrochemicals.
18.4.5 Subsurface Drainage System Management
The above section deals with new design criteria for water table management. To-
gether with these design criteria an operational management plan is required that sets
out how the drainage system will be managed. This is also of great importance to ex-
isting subsurface drainage systems, which have not had the benefit of the improved
drainage design criteria and so will benefit most from astute management. A review of
12 subsurface drainage systems in the irrigated areas of Australia by Christen et al.
(2001) found that most systems were draining greater volumes of water than designed
for, leading to excessively high leaching fractions and reduced irrigation water-use
efficiency. The salt load removed by these systems was also often found to be far
greater than the salt applied by irrigation (Figure 18.12), indicating a mining of geo-
logic salt.
Management of subsurface drainage systems in irrigated semi-arid areas is focused
on restricting the drainflow at certain periods, resulting in raising of the water table
above where it would normally occur if the drains were allowed to flow freely. The
method of drain flow restriction will depend upon the drainage system design, man-
agement ease, topography, and cost. Flow restriction can be achieved by placing weirs
within sumps, risers on the end of laterals (Fig 18.9), or reducing the depth of the float
switch at a pumping point. These methods will allow drain flow to occur when the
water table height increases above the desired level (Ayars et al., 2000).
50
40
Salt drained (t/ha/yr)

30
20
10
1:1 line
0
0 1 2 3 4 5
Salt applied (t/ha/yr)
Figure 18.12. Irrigation and drainage water salt loads in 12 irrigated areas of Australia.
Methods that completely restrict flow include placing valves on drainage lines or
turning pumps off. When this is done there needs to be monitoring of the water table
position to ensure that it does not become harmfully shallow. New developments in
this area are introducing methods of linking electronic sensing of water table height
with pump or valve operation to provide an automated system. The automation of such
systems can also include sensing of the receiving waters and control of the drainage
flows in keeping with any restrictions or licensing conditions. Where drainage dis-
posal is restricted to evaporation basins the system can be developed to include moni-
toring of the basin water level and modifying the drainage flow as required (Christen
et al., 2004).
In regards to drainage system management, it is important to consider the situation
where a drainage system is designed and installed as a reclamation practice for a saline
or waterlogged condition (as most drainage systems are) and then the subsequent
management of the drainage system once the reclamation phase has been completed,
when the leaching of stored salts from the root zone has been accomplished and a
lesser level of drainage is required to maintain the root-zone salinity. During the rec-
lamation phase a high degree of leaching is required; however, direct flow to the
drains by preferential flow through the trench area should be avoided as this contrib-
utes little to the leaching process and wastes irrigation water. Grismer (1990) found
that in a heavy clay soil trench flow accounted for almost all of the drain flow for 40 h
after irrigation, resulting in low salinity of the drainage water. Christen and Skehan
(1999) also found that the drain flow salinity dropped dramatically during irrigation
indicating preferential flow of the non-saline irrigation water (Figure 18.13). This in-
dicated that for about 24 h after irrigation about 50% of the drainage water was irriga-
tion water. Preventing this trench flow during irrigation can be done by turning off a
pump or blocking (or closing) a valve on the drains. When this is undertaken consider-
able water and salt load savings can accrue. This form of management is also applica-
ble after reclamation has occurred, to conserve irrigation water.
Arrows indicate irrigations
13
12
Drain water salinity (dS/m)

11
10
6
11-Nov 1-Dec 21-Dec 10-Jan 30-Jan 19-Feb
Figure 18.13. Effect of irrigation on drainage water salinity.
Determining when reduced drainage should be implemented can be assessed by soil

salinity measurements using combined soil sampling and electromagnetic surveys.
Drainage reclamation is rarely uniform and a spatial survey can provide indications as
to where further drainage work is required. This can be seen in an example from an
irrigated pasture farm. Figure 18.14 shows the distribution of apparent soil electrical
conductivity (ECa) measured with an EM38 survey three years after subsurface drain-
age was installed. The dark area in the left of the survey is fully reclaimed and drain-
age can be reduced to a management level. In the rest of the area where the drains
were installed the soil salinity is low, but there are some areas with high soil salinity
“hot spots” remaining. These hot spots can be addressed by a range of actions such as
installing more drainage, deep ripping, or addition of gypsum. These results can also
be used to indicate the type of management that should be applied to different sections
of drainage depending upon their salinity and water table status (Christen et al., 2002).
300
290
280
270
260
250
240
230
220
210
200
190
180
170
160
150
140
130
120
110
100
90
80
70
60
Figure 18.14. EM38 survey of irrigated pasture three years after subsurface drain-
age installation showing apparent soil electrical conductivity (ECa ms/cm). White
lines are subsurface drainage pipes. (Christen et al., 2004)
18.5 DOCUMENTATION OF SYSTEM

DESIGN AND INSTALLATION
Engineering practice standards have been developed for the design, installation, and
operation of shallow groundwater management systems for agricultural crop produc-
tion. The American Society of Agricultural and Biological Engineers (ASABE, for-
merly ASAE) has issued Engineering Practice (EP) 479 covering the design, installa-
tion, and operation of a subsurface drain tube system for both subirrigation and con-
trolled drainage in crop production systems (ASABE Standards, 2006). The American
Society for Testing and Materials (ASTM) has issued Standard Practice for the Sub-
surface Installation of Corrugated Thermoplastic Tubing for Agricultural Drainage and
Water Table Control, ASTM Designation F-449 (ASTM, 2006). A Recommended
Practice for the design and installation of subsurface drains for water table control has
also been developed by the Natural Resources Conservation Service (NRCS, 2006).
The final design of the water table management system needs to be adequately
documented by the engineer or designer for reference and use by the farmer and con-
tractor to properly prepare the site and install the system components. The documenta-
tion should include a complete layout plan showing locations and depths of all drain
pipes and control structures; a list of all materials and equipment components needed
to install the system; recommended or acceptable installation methods and equipment;
a list of all applicable installation practices and material specifications or standards;
and detailed guidelines and plan for system operation. If the water table management
system is a retrofit of an existing subsurface drainage system, the documentation
should identify both previously existing and new components. Further, it is recom-
mended that the documentation include the final economic evaluation of the recom-
mended system design. The documentation should be modified as needed (by the en-
gineer, farmer, or contractor) to reflect the system design as actually installed.
It is strongly recommended that documentation of the newly installed system be re-
tained for a permanent record with other property documents or deeds. Contractors
may wish to maintain a copy of installation documentation for possible future refer-
ence, e.g., for maintenance or repairs. If the land is sold or leased, or use changed,
such documentation should be provided to the new landowner or tenant.
18.6 SUMMARY
Agricultural water management systems are designed and installed to (1) improve
crop production by controlling the durations of excessive and deficient soil-water con-
ditions in the root zone, and (2) improve the water quality of drainage discharge by
controlling drainage flows to reduce agrochemical losses from farmland. The devel-
opment of an integrated design for a water table control system includes the determi-
nation of the suitability of the site, the required drain depth and spacing, the prepara-
tion of a field installation plan for the system, and the selection and/or design for an
operating system in the controlled drainage and subirrigation modes. The system de-
sign should permit control of the water table depth in the soil profile over the range
needed for the cultural practices to be followed and the crops to be grown, and the
operational requirements to reduce agrochemical losses. These objectives should also
involve the efficient utilization of shallow groundwater supplied by natural rainfall or
irrigation.
Several methods are now available for determining optimum design drain depth
and spacing for water table management. The operation of the system, that is, whether
it is in the conventional subsurface drainage, controlled drainage, or subirrigation
mode, varies from day to day and from year to year. In most locations, it is not clear
whether the greatest demands on the system design are to provide good drainage man-
agement under shallow water table conditions, or to provide sufficient subirrigation
during the driest periods. Because of these design complication factors, it is recom-
mended that a simulation model approach be used to conduct a complete analysis and
final design of the drainage water management or water table control system, and to
predict its performance over a period of 20 to 30 years for the climatologic conditions
at the specific site.
Computer simulation models such as DRAINMOD can be used to evaluate various
system design options for a specific site. A long-term simulation (20 to 30 years) can
provide a good evaluation of the expected performance of a given water table man-
agement system. DRAINMOD-NII includes a routine to comprehensively evaluate the
impact of system design and operational parameters on transport of various forms of
nitrogen within the soil profile and losses in runoff and subsurface flow. This new
version of DRAINMOD is an important tool for designing drainage water manage-
ment systems, along with a seasonal operation plan, to meet emerging water quality
requirements. The computer program SI-DESIGN allows designers to calculate a de-
sign rainfall and evaluate the subsurface drain lateral spacing and collector main size
alternatives. This program also provides a means of estimating system cost, conduct-
ing an economic analysis of profit potential from operating the system, and estimate
the biomass production efficiency of the water management system.
The design for a water table management system that results in the optimum net
profit, while minimizing the environmental impacts offsite, should be the best design
and management strategy to be recommended to the farmer. The system may be tech-
nically feasible, but the final decision should be based on the feasibility of the system
not only to pay for itself but to return a profit to the farmer for his investment, while
minimizing offsite environmental impacts. Therefore, it is very important that the final
decision for a given or recommended design be based on a thorough evaluation of
economic and environmental impacts.
REFERENCES
ADMS-TF. 2005. Agricultural Drainage Management Systems Task Force web site,
http://extension.osu.edu/~usdasdru/ADMS/ADMSindex.htm.
ASAE. 1990. EP479: Design, installation and operation of water table management
systems for subirrigation/controlled drainage in humid regions. St. Joseph, Mich.:
ASAE.
ASTM. 2006. F-449: Subsurface installation of corrugated polyethylene pipe for
agricultural drainage or water table control. West Conshohocken, Pa.: American
Society for Testing Materials.
Ayars, J. E. 1996. Managing irrigation and drainage systems in arid areas in the
presence of shallow groundwater:case studies. Irrig. Drain. Systems 10: 227-244.
Ayars, J. E., and R. A. Schoneman. 1984. Managing irrigation in areas with a water
table. In Water Today and Tomorrow, Proc. Spec. Conf. Irrig. and Drain. Div.,
ASCE, 528-536. J. A. Replogle, and K. G. Renard, eds. New York, N.Y.: American
Ayars, J. E., and D. B. McWhorter. 1985. Incorporating crop water use in drainage
design in arid areas. In Proc. Spec. Conf. Irrig. and Drain. Div., ASCE, 380-389. C.
G. Keyes, and T. J. Ward, eds. New York, N.Y.: American Soc. Civil Engineers.
Ayars, J. E., and R. B. Hutmacher. 1994. Crop coefficients for irrigation cotton in the
presence of groundwater. Irrig. Sci. 15(1): 45-52.
Ayars, J. E., R. B. Hutmcaher, R. A. Schoneman, R. W. O. Soppe, S. S. Vail, and F.
Dale. 2000. Realizing the potential of intetgrated irrigation and drainage water
management for meeting crop water requirements in semi-arid and arid areas. Irrig.
Drain. Systems 13: 321-347.
Ayars, J. E., E. W. Christen, R. W. Soppe, and W. Meyer. 2006. Resource potential of
shallow ground for crop water use: A review. Irrig. Sci. 24: 147-160.
Belcher, H. W., and B. W. Fehr. 1990. Performance of a subirrigation automation
system. ASAE Paper No. 90-2607. St. Joseph, Mich.: ASAE.
Belcher, H. W., and F. M. D’Itri, eds. 1995. Subirrigation and Controlled Drainage.
Boca Raton, Fla.: Lewis Publishers.
Belcher, H. W., G. E. Merva, and W. H. Shayya. 1993. SI-DESIGN: A simulation
model to assist with the design of subirrigation systems. In 15th Int’l. Cong. of
ICID Workshop on Subsurface Drainage Models, 295-308. The Hague,
Netherlands: ICID-CIID, CEMAGREF.
Bengtson, R. L., C. E. Carter, H. F. Morris, and S. A. Bartkiewicz. 1988. The
influence of subsurface drainage practices on nitrogen and phosphorus losses in a
warm, humid climate. Trans. ASAE 31: 729-733.
Bengtson, R. L., C. E. Carter, J. L. Fouss, L. M. Southwick, and G. H. Willis. 1995.
Special Issue: Water Quality in Humid Regions. Agricultural drainage and water
quality in Mississippi Delta. J. Irrig. Drain. Eng. 121(4): 292-295.
Borg, H., and D. W. Grimes. 1986. Depth development of roots with time: An
empirical description. Trans. ASAE 29: 194-197.
Bottcher, R., T. Steenhuis, and M. Walter. 1984. SUBDRAIN: An interactive,
colorgraphics subsurface tile drainage design program. Northeast Regional
Agricultural Engineering Service, Cooperative Extension Report. Ithaca, N.Y.:
Cornell Univ.
Bucks, D. A. 2004. A new agenda for management of agriculturally drained lands. In
Proc. Eighth Int’l. ASAE Drainage Symp. R. A. Cooke, ed. St. Joseph, Mich.:
ASAE.
Burnham, J., and H. W. Belcher, Jr. 1985. Laser surveying for water management
system design. ASAE Paper No. 85-2559. St. Joseph, Mich.: ASAE.
Carter, C. E., J. L. Fouss, and V. McDaniel. 1988. Water management increases
sugarcane yields. Trans. ASAE 31(2): 503-507.
Christen, E. W., and D. Skehan. 1999. Design and management of subsurface drainage
for improved water quality: A field trial. CSIRO Land and Water Technical Report
6/99. Griffith, NSW, Australia: CSIRO Land and Water, Griffith.
Christen, E. W., J. E. Ayars, and J. W. Hornbuckle. 2001. Subsurface drainage design
and management in irrigated areas of Australia. J. Irrig. Sci. 21: 35-43.
Christen, E. W., and D. Skehan. 2001. Design and management of subsurface
horizontal drainage to reduce salt loads. J. Irrig. Drain. Eng. 127(3): 148-155.
Christen, E. W., and J. E. Ayars. 2001. Subsurface drainage system design and
management in irrigated agriculture: Best management practices for reducing
drainage volume and salt load. Tech. Report 38-01. Griffith, NSW, Australia.
CSIRO Land and Water. Available at: www.clw.csiro.au/publications/technical
2001/tr38-01.pdf.
Christen, E. W., J. W. Hornbuckle, and S. Herath. 2002. Management options to
reduce salt loads from tile drainage in the Campaspe West District. Tech. Report
XX/02. Griffith, NSW, Australia: CSIRO Land and Water.
Christen, E. W., Hornbuckle, J.W., and R. Zandonna. 2003. Automated subsurface
drainage management system to reduce costs and downstream environmental
impact. In Engineering Salinity Solutions 2004, Proc. 1st National Salinity
Engineering Conference, 194-199. S. Dogramaci and A. Waterhouse, eds.
Engineers Australia.
Christen, E.W., Hornbuckle, J.W., and J.E. Ayars. 2004. A methodology to assess the
performance of subsurface drainage salinity control. In Engineering Salinity
Solutions 2004, S. Dogramaci and A. Waterhouse, eds. Proc. 1st National Salinity
Engineering Conference, 21-25. Engineers Australia.
Dedrick, A. R., L. J. Erie, and A. J. Clemmens. 1982. Level-basin irrigation. In
Advances in Irrigation, 1: 105-145. D. Hillel, ed. New York, N.Y.: Academic
Press.
Deverel, S. J., and J. L. Fio. 1990. Ground-water flow and solute movement to drain
laterals, western San Joaquin Valley, California I: Geochemical assessment. Open-
file Report 90-136. Sacramento, Calif.: U.S. Geological Survey.
Doering, E. J., L. C. Benz, and G. A. Reichman. 1982. Shallow-water-table concept
for drainage design in semiarid and subhumid regions. In Advances in Drainage,
Proc. of the Fourth National Drainage Symposium, 34-41 St Joseph, Mich.: ASAE.
Doty, C. W., J. E. Parsons, A. Nassehzadeh-Tabrizi, R. W. Skaggs, and A. W. Badr.
1984. Stream water levels affect field water tables and corn yields. Trans. ASAE
27(5): 1300-1306.
Doty, C. W., J. E. Parsons, and R. W. Skaggs. 1987. Irrigation water supplied by
stream water level control. Trans. ASAE 30(4): 1065-1070.
Ernst, L. F. 1975. Formulae for groundwater flow in areas with subirrigation by means
of open conduits with a raised water level. Misc. Reprint 178, 55-84. Rome, Italy:
Institute for Land and Water Development Division, FAO.
Evans, R. O., and R. W. Skaggs. 1989. Design guidelines for water table management
systems on coastal plain soils. Applied Eng. in Agric. 5(4): 539-548.
Evans, R. O., R. E. Sneed, and R. W. Skaggs. 1988a. Water supplies for subirrigation.
AG-389. Raleigh, N.C.: North Carolina Agricultural Extension Service.
Evans, R. O., R. W. Skaggs, and R. E. Sneed. 1988b. Economic evaluation of
controlled drainage and subirrigation systems. Ag-397. Raleigh, N.C.: North
Carolina Agricultural Extension Service.
Evans, R. O., R. W. Skaggs, and R. E. Sneed. 1990a. Normalized crop susceptibility
factors for corn and soybeans to excess water stress. Trans. ASAE 33(4): 1153-
1161.
Evans, R. O., J. W. Gilliam, and R. W. Skaggs. 1990b. Controlled drainage and
subirrigation effects on drainage water quality. Proc. Fourteenth Int’l.l Cong. on
Irrigation and Drainage, ICID. 1A(Q42): 13-20.
Evans, R. O., J. W. Gilliam, and R. W. Skaggs. 1995. Controlled versus conventional
drainage effects on water quality. Irrig. Drain. Division, ASCE 121(4):271-276.
Feddes, R. A., P. J. Kowalik, and H. Zaradny. 1978. Simulation of field water use and
crop yield. PUDOC, Wageningen: Simulation Monographs.
Fausey, N. R., K. W. King, B. J. Baker, and R. L. Cooper. 2004. Controlled drainge
performance on Hoytville soil in Ohio. In Drainage VIII: Proc. Eighth Int’l.
Drainage Symp., 84-88. R. Cook, ed. St. Joseph, Mich.: ASAE.
Fouss, J. L. 1985. Simulated feedback-operation of controlled-drainage/subirrigation
systems. Trans. ASAE 28(3): 839-847.
Fouss, J. L., and G. H. Willis. 1990. Research need on integrated system for water and
pest management to protect groundwater quality. In Proc. of the 1990 Conference
IR Div., 288-296. New York, N.Y.: American Soc. Civil Engineers.
Fouss, J. L., and J. S. Rogers. 1992. Drain outlet water level control: A simulation
model. In Proc. Sixth Int’l. Drainage Symp, Drainage and Water Table Control,
Fouss, J. L., and G. H. Willis. 1994. Integrated water-fertilizer-pesticide management
for environmentally sound crop production. In Proc. Univ. Florida Second Conf.
Environmentally Sound Agriculture, 53-61. St. Joseph, Mich.: ASAE.
Fouss, J. L., and J. S. Rogers. 1998. Justification of automated water table control
systems. In Proc. 7th Int’l. Drainage Symposium, Drainage in the 21st Century:
Food Production and the Environment, 400-412. St. Joseph, Mich.: ASAE.
Fouss, J. L., and T. W. Appelboom. 2006. Combination of drainage water manage-
ment, cover cropping, and wetland diversion, as a suite of BMPs to reduce nitrogen
loss from cropland. In Examining the Confluence of Environmental and Water
Concerns. Proceedings of the ASCE-EWRI 2006 World Environmental and Water
Resources Congress, Session on Drainage BMPs for Nitrogen Management. Ran-
dall Graham, ed. CD-ROM.
Fouss, J. L., R. L. Bengtson, and C. E. Carter. 1987a. Simulating subsurface drainage
in the lower Mississippi valley with DRAINMOD. Trans. ASAE 30(6): 1679-1688.
Fouss, J. L., R. W. Skaggs, and J. S. Rogers. 1987b. Two-stage weir control of
subsurface drainage for water table management. Trans. ASAE 30(6): 1713-1719.
Fouss, J. L., J. S. Rogers, C. E. Carter. 1989. Sump-controlled water table
management predicted with DRAINMOD. Trans. ASAE 32(4): 1303-1308.
Fouss, J. L., R. W. Skaggs, J. E. Ayars, and H. W. Belcher. 1990. Water table control
and shallow groundwater utilization. In Management of Farm Irrigation Systems,
Fouss, J. L., J. S. Rogers, G. H. Willis, L. M. Southwick, and C. E. Carter. 1995.
Automated water table control/data acquisition for water quality research. In Clean
Water—Clean Environment—21st Century, 101-104. St. Joseph, Mich.: ASAE.
Fouss, J. L., R. O. Evans, D. L. Thomas, and H. W. Belcher. 1999. Operation of
controlled-drainage and subirrigation facilities for water table management. In
Agricultural Drainage, 743-766. R. W. Skaggs, and J. van Schilfgaarde, eds.
Agron. Monog. 38. Madison, Wis.: ASA, CSSA, and SSSA.
Fouss, J. L., D. A. Bucks, and B. C. Grigg. 2004. The agricultural drainage
management systems task force: Decreasing nutrient export through the Mississippi
River drainage basin. In Total Maximum Daily Load (TMDL) Environmental
Regulations II, Conference Proc. A. Saleh, ed. St. Joseph, Mich.: ASAE.
Gayle, G., R. W. Skaggs, and C. E. Carter. 1985. Evaluation of a water management

model for a Louisiana sugar cane field. J. American Soc. of Sugar Cane
Technologists 4: 18-28.
Gayle, G. A., R. W. Skaggs, and C. E. Carter. 1987. Effects of excess soil water
conditions on sugarcane yields. Trans. ASAE 30(4): 993-997.
Gilliam, J. W. 1987. Drainage water quality and the environment. Proc. Fifth Nat’l.
Drainage Symp. ASAE, St. Joseph, Mich. pp. 19-28.
Gilliam, J. W., J. L. Baker, and K. R. Reddy. 1999. Chapter 24: Water quality effects
of drainage in humid regions. In Agricultural Drainage, 801-830. R.W. Skaggs and
J. van Schilfgaarde, eds. Madison, Wisc,: Crop Sci. Soc. America, Soil Sci. Soc.
America, and American Soc. Agronomy.
Gilliam, J. W., and R. W. Skaggs. 1985. Use of drainage control to minimize potential
detrimental effects of improved drainage systems. In Development and
Management Aspects of Irrigation and Drainage Systems, 352-362. New York,
Gilliam, J. W., and R. W. Skaggs. 1986. Controlled agricultural drainage to maintain
water quality. J. Irrig. Drain. Eng. 112: 254-263.
Gilliam, J. W., R. W. Skaggs, and S. B. Weed. 1979. Drainage control to reduce
nitrate losses from agricultural fields. J. Environ. Qual. 8(1): 137-142.
Grandia, C. 2002. Pioneering GPS for site prep (Trimble). Midwest Contractor, Au-
gust 26, 2002, 6 pp.
Grismer, M. E. 1990. Subsurface drainage system design and drain water quality. J.
Irrig. Drain. Eng. 119: 537-543.
Hardjoamidjojo, S., R. W. Skaggs, and G. O. Schwab. 1982. Predicting corn yield
response to excessive soil water conditions. Trans ASAE 25(4): 922-927, 934.
Herczeg, A. L., S. S. Dogramaci, and F. W. Leaney. 2001. Origin and evolution of
solutes in a large semi-arid, regional multi-aquifer system: Murray basin, for
Australia. Marine Freshwater Res. 52: 41-52.
Hoffman, G. J. 1990. Leaching fraction and root zone salinity control. In Agricultural
Salinity Assessment and Management, 237-261. K. K. Tanji, ed. New York, N.Y.:
Hornbuckle, J. W., and E. W. Christen. 1999. Physical properties of soils in the
Murrumbidgee and Coleambally irrigation areas. CSIRO Land and Water
Technical report 17/99. Griffith, Australia: CSIRO Land and Water.
Howell, T. A., J. L. Hatfield, J. D. Rhoades, and M. Meron. 1984. Response of cotton
water stress indicators to soil salinity. Irrig. Sci. 5: 25-36.
Kendrick-Peabody, E. 2004. Success stories in agriculture: Researchers Fouss and
Fausey develop laser grade-control system that transforms drainage and irrigation
technology; NewsCAST Vol. 31(1): 15-18. Washington, D.C.: Council for
Agricultural Science and Technology, www.cast-science.org/cast/src/cast_top.htm.
Kite, S. W., and B. R. Hanson. 1984. Irrigation scheduling under saline high water
tables. California Agric. 38: 12-14.
Ladaney-Bell, J. R. W., and R. L. Acworth 2002. Salinisation processes in the
irrigation environment riverine plain, Murray Darling Basin. In Irrigation Australia
2002 conference, Irrigation:Conservation or Conflict?. Sydney, Australia:
Irrigation Association of Australia, Sydney.
Logan, T. J., G. W. Randall, and D. R. Timmons. 1980. Nutrient content of tile

drainage from cropland in the North Central Region. North Central Res. Bull. No.
268. Wooster, Ohio.
Lord, J. M. 1987. Phase II: Study of innovative techniques to reduce subsurface
drainage flows. Sacramento, Calif.: San Joaquin Valley Drainage Program, U.S.
Bureau of Reclamation.
Maas, E. V., and G. J. Hoffman. 1977. Crop salt tolerance-current assessment. Irrig.
Drain. Division, ASCE 103: 115-134.
Munster, C. L., R. W. Skaggs, J. E. Parsons, R. O. Evans, J. W. Gilliam, and M. A.
Breve. 1994. Simulating aldicarb transport in a drained field. Trans. ASAE 37(6):
1817-1824.
Nolte, B. L., R. L. Burris, C. J. W. Drablos, N. R. Fausey, L. R. Massie, S. W. Melvin,
G. E. Merva, T. J. Olscheske, J. F. Rice, G. O. Schwab, R. D. Wenberg, and R. Z.
Wheaton. 1986. DRAINMOD: Documentation for the water management
simulation model. NCCI Software J. 2(1): June 1986.
Northey, J., E. W. Christen, J. E. Ayars, and J. Jankowski. 2006. Occurrence and
measurement of salinity stratification in the shallow groundwater in the
Murrumbidgee Irrigation Area, south-eastern Australia. Ag. Water Mgmt. 81: 23-
40.
NRCS (Natural Resources Conservation Service, USDA). 2006. Recommended
practice for the design and installation of subsurface drains for water table control.
NRCS, USDA; Section IV, FOTG, Code 606, Subsurface Drain.
NRCS (Natural Resources Conservation Service, USDA). 1994. DRAINMOD Users
Guide. Washington, D.C.: USDA, NRCS.
Rabalais, N. N., R. E. Turner, D. Justic, Q. Dortch, and W. J. Wiseman, Jr. 1999.
Characterization of hypoxia: Topic 1 report for the integrated assessment on hy-
poxia in the Gulf of Mexico. Coastal Ocean Program Decision Analysis Series No.
15. Silver Spring, Md.: NOAA Coastal Ocean Program.
Rabalais, N. N., R. E. Turner, and D. Scavia. 2002. Beyond science into policy: Gulf
of Mexico hypoxia and the Mississippi River. BioScience 52(2):129-142.
Rogers, J. S. 1985. Water management model evaluation for shallow sandy soils.
Trans. ASAE 28(3): 785-790, 794.
Rogers, J. S., and J. L. Fouss. 1989. Hydraulic conductivity determination from
vertical and horizontal drains in layered soil profiles. Trans. ASAE 32(2): 589-595.
Rogers, J. S., H. M. Selim, C. E. Carter, and J. L. Fouss. 1991. Variability of auger
hole hydraulic conductivity values for a Commerce silt loam. Trans. ASAE 34(4):
876-882.
Sands, G. R., and R. J. Gaddis. 1985. LANDRAIN: A computer-aided-design (CAD)
program for subsurface drainage systems. ASAE Paper No. 85-2556. St. Joseph,
Mich.: ASAE..
Schneider, J. D., and J. M. Garbrecht. 2006. Dependability and effectiveness of sea-
sonal forecasts for agricultural applications. Trans.ASABE 49(6): 1737-1753.
Shalhevet, J. 1994. Using water of marginal quality for crop production: Major issues.
Agric. Water Mgmt. 25: 233-269.
Skaggs, R. W. 1978. A water management model for shallow water table soils. Report
No. 134. Chapel Hill, N.C.: Water Resources Research Institute, Univ. North
Carolina.
Skaggs, R. W. 1980. Drainmod reference report: Methods for design and evaluation of
drainage-water management systems for soils with high water tables. Ft. Worth,
Tex.; USDA-SCS, Nat’l. Technical Ctr.
Skaggs, R. W. 1981. Water movement factors important to the design and operation of
subirrigation systems. Trans. ASAE 24(6): 1553-1561.
Skaggs, R. W. 1982. Field evaluation of a water management simulation model.
Trans. ASAE 25(3): 666-674.
Skaggs, R. W., and J. W. Gilliam. 1981. Effect of drainage system design and
operation on nitrate transport. Trans. ASAE 24(4): 929-934.
Skaggs, R. W., and A. Tabrizi. 1983. Optimum drainage for corn production. Tech.
Bulletin 274. Raleigh, N.C.: North Carolina Agricultural Research Service.
Skaggs, R. W., and A. N. Tabrizi. 1986. Design drainage rates for estimating drain
spacings in North Carolina. Trans. ASAE 29(6): 1631-1640.
Skaggs, R. W., N. R. Fausey, and B. H. Nolte. 1981. Water management model
evaluation for North Central Ohio. Trans. ASAE 24(4): 922-928.
Tabrizi, A. N., and R. W. Skaggs. 1983. Variation of saturated hydraulic conductivity
within a soil series. ASAE Paper No. 83-2044. St. Joseph, Mich.: ASAE.
Thomas, D. L., R. R. Lowrance, and M. C. Smith. 1991. Drainage-subirrigation effect
on water quality in Georgia Flatwoods. J. Irrig. Drain. Eng. 117(1): 126-137.
Thomas, D. L., P. G. Hunt, and J. W. Gilliam. 1992. Water table management for
water quality improvement. J. Soil Water Cons. 47(1): 65-70.
U.S. Department of Interior. 1993. Drainage manual. Denver, Colo.: U.S. Department
of Interior.
Walker, W. R., G. V. Skogerboe, and R. G. Evans. 1979. Reducing salt pickup from
irrigated lands. J. Irrig. Drain. Eng. 105: 1-14.
Welch, B. 2002. 3-D systems save prep time (Trimble). Construction News, August
19, 2002: 4 pp.
Willis, G. H., J. L. Fouss, J. S. Rogers, C. E. Carter, and L. M. Southwick. 1991.
Chapt. 11: System design for evaluation and control of agrochemical movement in
soils above shallow water tables. In Groundwater Residue Sampling Design, 195-
211. ACS Symposium Series 465. Washington, D.C.: ACS.
Youssef, M. A., R. W. Skaggs, G. M. Chescheir, and J. W. Gilliam. 2005. The
nitrogen simulation model, DRAINMOD-NII. Trans. ASAE 48(2): 611-626.
CHAPTER 19
CHEMIGATION
Ted W. van der Gulik (BC Ministry of Agriculture
and Lands, Abbotsford, British Columbia)
Robert G. Evans (USDA-ARS, Sidney, Montana)
Dean E. Eisenhauer (University of Nebraska,
Lincoln, Nebraska)
Abstract. Chemigation is the application of water-compatible chemicals by the use
of an irrigation system for fertilizing or control of crop pests, or for maintenance of a
microirrigation system. Chemigation, in all its forms, is commonly used in irrigated
production across the world, and its use can greatly influence the design and opera-
tion of farm irrigation systems. Its use must follow regulatory requirements and be
done according to the chemical label. A major concern with chemigation is safety, in
terms of protecting personnel, the water supply, and the environment. There are nu-
merous ways to apply the various chemicals depending on the volumes and required
precision of the injection. These systems must be carefully calibrated and used with
great care.
Keywords. Chemicals, Crop protection, Fertilizers, Irrigation, Safety, Systems.
19.1 INTRODUCTION
19.1.1 Definition
Chemigation is the application of various chemicals to a crop through an irrigation
system. In agricultural operations the type of chemicals applied may be fertilizers, pesti-
cides, insecticides, fungicides, herbicides, nematicides, and growth regulators (Rolston
et al., 1986; Bar-Yosef, 1999). However, injection of water-soluble fertilizers is the most
common form of chemigation (Wright et al., 1992; van der Gulik, 1993; Burt et al.,
1995; Scherer et al., 1999). Depending on the use and the user, various forms of chemi-
gation are also referred to as fertigation, pestigation, insectigation, fungigation, herbiga-
tion, and nematigation, to name just a few.
Chemicals may also be injected into an irrigation system for maintenance purposes,
such as algaecides and chlorine in microirrigation systems (Evans and Waller, 2007).
Specific details on microirrigation system maintenance with chemical treatments are
provided in Chapter 17.
Chemigation has been considered and probably practiced since sprinklers were first
used on farms in the first part of the 20th century. One of the first discussions on
chemigation in the literature was by Bryan and Thomas in 1958. Threadgill (1985) con-
ducted a survey of chemigation use in 35 states in 1983 and found that about 4.3 mil-
lion hectares in the U.S. were utilizing chemigation at least once during the season.
726 Chapter 19 Chemigation
The 1998 Census of Agriculture (NASS, 1998) reported that about 4 million hectares
were chemigated on more than 35,000 farms across the U.S. at that time. The use of
chemigation is expected to increase as producers continue to convert to pressurized
irrigation systems because of the convenience and reduction in application costs over
more conventional chemical application methods. In addition, it is expected that the
current emphasis on precision farming techniques will accelerate the growth of chemi-
gation in the near future.
The application of chemicals is almost universal with most drip systems and a large
percentage of sprinkler systems are using this technology. Chemigation is used to a
lesser extent on surface-irrigated land. Microirrigation and center pivot systems are
well-suited to the use of chemigation because of their high distribution uniformities
and ease of operation (Rolston et al., 1986; Threadgill et al., 1990; van der Gulik and
Evans, 2006; Evans and Waller, 2007).
Prior to using chemigation an operator must understand federal and local regula-
tions that apply; understand safety issues and environmental protection measures that
must be followed; have knowledge of the irrigation system operation and chemical to
be applied; be able to determine an appropriate injection rate and timing of application
for the area treated and crop grown; and ensure the irrigation system is properly
flushed after chemigation has been completed (van der Gulik 1993; Burt et al., 1995;
Evans and Waller, 2007).
19.1.2 Benefits
The use of chemigation with irrigation systems has shown steady growth. Benefits
include:
reduced labor and chemical application cost,
incorporation and activation of fertilizers for crops grown under drip systems in
drier climates,
timely application of chemicals,
reduction of soil compaction and mechanical damage to the crop,
improved operator safety when applying pesticides,
reduction in the amount of chemical use,
potential reduction of environmental contamination, and
improved crop production.
The benefits of chemigation do come with extra cost and precautions. Backflow pre-
vention, other safety equipment, and injectors will increase system cost. Irrigation
system distribution uniformity must be maximized to ensure best possible chemical
distribution uniformity. To ensure proper operation of the system, injector calibration
is necessary. Irrigation drift must be minimized and all irrigation systems must be
managed to control over- or under-application of chemicals..
The use of microirrigation irrigation systems and fertigation will continue to increase
due to the many advantages offered by this technology. Better application efficiencies
offer a reduction in the amount of fertilizer used. Fertigation offers the possibility of
timely applications of the appropriate amount of fertilizers which will result in less im-
pact on the environment
The proper application of fertilizers imposes minimal health concerns and risks to
water purveyors. However, the potential to apply other chemicals is seen as a greater
risk. Potential contamination of the water source due to backsiphonage and backpressure
are possible if an unexpected shutdown of the irrigation system should occur while injec-
tion is taking place. This risk can be minimized by following good chemigation practices
and if proper backflow prevention devices are installed and inspected regularly.
19.2 BACKFLOW PREVENTION AND SAFETY
19.2.1 Why Chemigation Is a Potential Hazard
Chemigation is considered a hazard if there is a potential cross-connection between
a chemical tank and a potable water source. A cross-connection is any connection or
structural arrangement between a potable water system and any nonpotable water sys-
tem or chemical source through which backflow can occur. The cross-connection can
be to a purveyor’s main line, stream, lake, or groundwater used as a source of potable
water. Injecting chemicals into an irrigation system therefore presents a potential haz-
ard to public health and safety, requiring approved prevention measures.
The safety devices that are required will depend on federal or local laws and the in-
formation provided on the chemical label. Some chemical labels require specific back-
flow prevention devices to be installed before chemigation can be used. Specifications
and management of backflow devices may vary between jurisdictions. Applicators
should always consult with local officials or experts for information on the required
backflow prevention devices for their area.
19.2.2 Regulations and Safety
The first rule of chemigation is always safety. Special chemigation safety devices,
check valves, and air relief valves are required for all chemical injection systems un-
der federal and local regulations. Well heads must be protected from reverse flows,
system drainage, or backsiphoning. Electric and hydraulic interlocks with time delays
must be installed between the injectors and irrigation pumps to prevent chemical injec-
tion when the irrigation system is not operating. Chemical injection areas should al-
ways be securely fenced with appropriate containment facilities in the event of a spill
(Shulze and Buttermore, 1994). Special protective equipment, safety showers, and any
required chemical neutralizing agents should be readily accessible and clearly marked.
Personnel must be specifically trained and, in many areas, licensed for chemical appli-
cations. Injection of any pesticide into an irrigation system must be specifically per-
mitted by the pesticide label and may also be subject to additional state regulations.
Detailed records of all chemical applications need to be maintained for safety, evalua-
tions, legal, and regulatory requirements.
All chemicals and chemical-water mixtures must be checked to avoid phytotoxic
effects before any injection occurs. In addition, it is critical that all the chemicals be-
ing injected at one time are compatible with each other and the water chemistry and
concentration limits are not exceeded so that precipitates do not form and plug emit-
ters. Some chemical combinations, such as calcium nitrate and phosphoric acid, will
immediately form a precipitate creating a severe plugging situation. Emulsifiable pes-
ticide concentrates and wettable powders may require special design and management
considerations (e.g., mechanical supply tank agitation) to help ensure uniform applica-
tions and reduce plugging. Acidification to lower water pH may sometimes be re-
quired prior to injection of a chemical. Precipitation tests should be conducted at the
same dilutions, pH, and other conditions of chemical application. Additional informa-
tion and procedures can be found in Ptacek (1986), Smajstrala et al. (1986), van der
Gulik (1993), Burt et al. (1995), Clark et al. (1998), Scherer et al. (1999), and Gran-
berry et al. (2001).
19.2.3 Backflow Prevention Devices

Backflow can occur in a chemigation system by either backsiphonage or backpres-
sure (Smajstrala et al., 1985; Wright et al., 1992; van der Gulik, 1993; Solomon and
Zoldoske, 1998; ASABE, 2006). Backsiphonage is caused by low pressure or a re-
duced pressure in the water supply piping. Principal causes of backsiphonage are:
failure of the irrigation pipeline main line check valve upon pump shutdown or
power failure;
creation of a severe hydraulic gradient by undersized piping in the supply line;
pipeline breakages in the water supply main line, which is lower than the cus-
tomer service point;
reduced main line pressure due to a high water withdrawal rate, such as for fire
fighting or main line flushing; and
reduced main line supply pressure due to pump or power failure.
Backpressure occurs when the user system is operating at a higher pressure than the
water supply system. Major sources of backpressure are:
booster pumps on the user system used to increase flows and satisfy pressure re-
quirements;
interconnection with other piping systems operating at higher pressures;
electric or gas-driven injector systems;
connections to pressurized systems, such as boilers; and
elevation differences between the irrigation system and the water supply system
or water source.
The following devices can be considered for use on a chemigation system. The de-
vice selected will depend the degree of hazard, injector type, and irrigation system
setup. All backflow prevention devices should be tested by a certified tester before
every irrigation season, and inspected prior to each chemigation event. The backflow
prevention device should be installed between the pump discharge and the point of
chemical injection.
19.2.3.1 Air gap. A 0.25-m air gap can be an effective backflow prevention device.
Air gaps are most commonly used where the water source for the chemigation system
is a self-contained pond or reservoir. The air gap must be maintained between the
maximum surface elevation of the reservoir and the pipeline used to fill the reservoir.
Crops such as cranberries often use reservoirs for irrigation supply and frost protection
and can utilize an air gap as a backflow prevention device.
19.2.3.2 Atmospheric and pressure vacuum breaker. Vacuum breakers are ef-
fective for backsiphonage situations only. An atmospheric vacuum breaker (AVB) can
only be used in situations where the unit is not subject to continuous pressure and
therefore does not have shutoff valves downstream of the unit. Unlike an atmospheric
vacuum breaker the pressure vacuum breaker (PVB) has an atmospheric vent valve
which is internally loaded by a spring. A spring helps open the valve and the PVB can
therefore be installed on the pressure side of a shutoff valve and used in situations that
are operating under continuous pressure. An AVB and a PVB must be installed 30 cm
(12 in.) above the highest sprinkler or dripper on the chemigation system.
Acceptable use of atmospheric and pressure vacuum breakers include situations
where nonpotable water is pumped into an irrigation system that is cross-connected to an
irrigation district or municipal pipeline where only backsiphonage is likely. Vacuum
breakers are not approved for use on irrigation systems that are applying chemigation.
19.2.3.3 Check valve with low-pressure drain and vacuum relief valve. Typi-
cally, the check valve low-pressure drain and vacuum relief valve are combined with
an inspection port, whether manufactured as a single unit or assembled as separate
components. When combined, these devices constitute an approved backflow preven-
tion device for chemigation systems. The check valve and appurtenant devices should
be installed in accordance with the manufacturer’s specifications and maintained in a
working condition.
Check valve. A check valve consists of a single internally loaded flapper capable of
closing and preventing backflow of the irrigation back into the water source. The term
is used generally to include all types of elements regardless of shape and method of
function. The check valve should provide for a watertight seal against reverse flow.
The check valve should contain a quick-closing and tight sealing mechanism that
will close the moment water ceases to flow in the downstream direction. The check
valve construction should allow for easy access for internal and external inspection
and maintenance. The preferred installation method for the check valve is to be in-
stalled horizontally above ground with adequate space to provide ease of maintenance,
testing, and inspection. The check valve should be inspected and tested after installa-
tion to ensure it is installed correctly and operating satisfactorily.
The irrigation line and check valve should be drained in the fall and protected from
freezing. The manufacturer can provide recommendations on how to drain each water-
trapping cavity of the device.
Low-pressure drain. The automatic low-pressure drain is used with the check valve
and consists of a spring-loaded or hydraulically actuated valve located on the bottom
of the irrigation line between the valve and the water source. The automatic low-
pressure drain is designed such that if the check valve leaks after system shutdown, the
automatic low-pressure drain will allow the water-chemical mixture to be drained
away from the water source rather than into it.
The automatic low-pressure drain must be installed between the water source and
the check valve so that any fluid which seeps past the check valve back toward the
water source will automatically drain out of the irrigation pipe. The drain should be at
least 2 cm (3/4 inch) in diameter and should be located on the bottom of the horizontal
irrigation pipe between the water source and the check valve. The outside opening of
the drain should be at least 5 cm (2 inches) above grade. The flow from the drain
should be controlled by a pipe, trough, ditch, and slope of soil surface or other means
so that it will drain away from the water source.
Vacuum relief valve. The vacuum relief valve is combined with the check valve and
low-pressure drain. The device consists of a spring-loaded or hydraulically actuated
atmospheric vacuum breaker valve. An atmospheric vacuum breaker allows air to en-
ter the irrigation pipeline when the line pressure is reduced to a gauge pressure of zero
or less. The vacuum relief device is typically located on the top of the horizontal irri-
gation pipeline between the check valve and the water source.
The vacuum relief device should be installed in such a position and in such a man-
ner that insects, animals, floodwater, or other pollutants cannot enter the irrigation
pipeline through the vacuum relief device. The vacuum relief device may be mounted
on the inspection port as long as it does not interfere with the inspection of the other
antipollution devices including the check valve and low-pressure drain.
Inspection port. An inspection port allows for easy access to the internal compo-
nents of the check valve, automatic low-pressure drain, and vacuum relief device for
the purposes of testing, inspection, and maintenance. The inspection port should allow
for visual inspection to determine if leakage occurs past the check valve, seal, seat,
and any other components of the backflow prevention device. The port should have a
minimum 10-cm (4-in.) diameter orifice or viewing area. For diversion works with
irrigation pipelines too small to install a 4-in, diameter inspection port, the check valve
and other appurtenant devices should be mounted with quick disconnects, flange fit-
tings, dresser couplings, or other fittings that allow for easy testing, inspection, main-
tenance, and removal.
19.2.3.4 Double check valve assembly. A double check valve assembly (DCVA)
consists of two approved check valves, internally loaded either by a spring or weight,
installed as a unit between two tightly closing shutoff valves. The DCVA is an ap-
proved backflow prevention device effective against backflow caused by backpressure
or backsiphonage. The DCVA must be installed upstream of the chemical injection
system at a location that is readily accessible for testing.
The irrigation lines should be thoroughly flushed before installation of the DCVA.
Most failures during testing are due to debris fouling either the first or second check
valve seats. The DCVA should be installed above ground with adequate space to sim-
plify maintenance and testing. It shall be inspected and tested after installation to en-
sure it is installed correctly and operating satisfactorily. A strainer with a blow-out
tapping should be installed ahead of the DCVA. The DCVA must be drained in the
fall and protected from freezing. The manufacturer can provide recommendations on
how to drain all water-trapping cavities of the device. The DCVA must be tested by a
Certified Tester before every irrigation season.
If possible, a DCVA should not be installed in a pit, because any leaky test cocks
would then become cross-connections when the pit is flooded. If the unit must be in-
stalled in a pit, provisions for pit drainage must be provided. Test-cock taps should
also be plugged to reduce the danger of leaks if the device does become submerged.
The vault should be large enough to provide free access for testing or repairing the
device. DCVAs larger than 6.4 cm (2.5 in.) shall have support blocks to prevent damage.
19.2.3.5 Reduced-pressure device. A reduced-pressure backflow device (RPBD)
consists of two independently acting, internally loaded check valves separated by a
reduced-pressure zone. The device should be installed as a unit between two tightly
closing shutoff valves. The RPBD is effective against backflow caused by backpres-
sure and backsiphonage and is designed to be used in situations that are considered
very hazardous. The RPDB, while slightly more costly, is considered the best protec-
tion for backflow prevention. The main reason is that the unit will leak water when it
is not operating properly, allowing a quick visual inspection to inform the operator if
the unit is malfunctioning. The unit can then be fixed prior to chemigation proceeding.
An RPBD must be installed upstream of the chemical injection system and preferably
above ground with adequate space to ease maintenance and testing. A strainer with a
blow-out tapping should be installed ahead of the RPBD. The lines should be thor-
oughly flushed before installation of the RPBD. Most failures during testing are due to
debris fouling either the first or second check valve seats. If possible, the RPBD
should not be installed in a pit below ground level. Flooding of the pit could cause a
direct cross-connection through the relief valve. If installation in a pit is absolutely
necessary, adequate drainage must be provided. Devices that are larger than 6.4 cm
(2.5 in.) shall have support blocks to prevent damage. A RPDB is susceptible to fluc-
tuating supply pressures on an extreme low flow or static flow condition, which may
cause nuisance dripping and eventual fouling of the device. The RPBD shall be in-
spected and tested after installation to ensure it is installed correctly and operating
satisfactorily, and it must be tested by a Certified Tester before every irrigation season.
The RPBD must be drained in the fall and protected from freezing. The manufacturer
can provide recommendations on how to drain each water-trapping cavity of the device.
19.2.4 Safety Equipment
A chemigation system should also contain additional antipollution and safety de-
vices on the irrigation system and the chemical injection system (Smajstrala et al.,
1985; Wright et al., 1992; Burt et al., 1995; Solomon and Zoldoske, 1998; Evans and
Waller, 2007). These devices are used to reduce hazard to the environment and the
operator during the chemigation application.
19.2.4.1 Injection-line equipment. Injection-line check valve. An injection-line
check valve should be installed to prevent the flow of water from the irrigation system
into the chemical supply tank, possibly overflowing the supply tank. A check valve is
also used to prevent gravity flow from the chemical supply tank into the irrigation
system if the opening pressure of the check valve is sufficient. The opening or crack-
ing pressure of the chemical injection-line check valve should be at least 70 kPa or
greater to prevent gravity flow. The injection-line check valve should be located
downstream from any backflow prevention equipment and fresh water supply valves.
When physically possible, the point of chemical injection should be higher than the
chemical supply tank and lower than the lowest sprinkler or outlet on the irrigation
system. This will prevent siphoning from the chemical supply tank. The point of
chemical injection should also be located as far as possible from the water source to
protect the water source in the event of a chemical leak or spill to the ground.
Flow sensor. An injection-line flow sensor installed on a high-pressure injection-
line upstream of the chemical injection-line check valve can assure system shutdown
if flow in the injection line ceases. This device guards against continued operation
after a rupture or disconnection of injection line, injection pump failure, loss of prime,
empty chemical supply tank, or plugging of the injection port.
Manual valve. A manually operated valve should be installed on the chemical sup-
ply tank. Installation of a manual valve will allow the operator to manually stop the
flow of chemical from the supply tank during equipment maintenance, or in case of
accidents.
Strainer. A strainer should be installed on the suction side of the chemical injection
pump. A strainer located upstream of the calibration tube, injection pump, air bleeder
valve, and injection-line check valve is essential to prevent foreign materials from
clogging or fouling these devices or other safety equipment.
Injector calibration device. A calibration device of sufficient volume should be in-
stalled on the suction side of the injection pump to accurately calibrate the injection
pump. The installation of a calibration tube provides an easy way to check and fine tune
the injection pump output. The calibration tube volume and graduation markings should
be sufficient to provide for a minimum 5-min calibration period. Calibration tube mark-
ings should be large enough for the user to easily read the scale. Chemical compatibility
is a key to preventing the calibration tube from being discolored or degraded.
Air-bleeder valve. An air-bleeder valve should be installed on the high-pressure

side of the chemical injection pump immediately upstream of the chemical injection-
line check valve. The air-bleeder valve can be used to relieve trapped air and pressure
in the high-pressure injection line, which might otherwise affect the calibrated injec-
tion rate. Pressure within the line should be relieved any time the injection line is to be
disconnected. This prevents the operator from being sprayed with the chemical in the
line. The valve is especially helpful while making equipment inspections.
Supply tanks. Chemical supply tanks should be constructed of chemically resistant
materials. Supply tanks that remain in the field throughout the year should also be
constructed of sunlight-resistant materials. The supply tanks should be designed so
that they can be easily drained after each use. The capacity of the tanks will depend
upon the type of chemical injected. A containment system helps guard against site
contamination in the event that a supply tank leaks or ruptures. Containment structures
can be made of chemically resistant plastics, painted metal, or concrete.
Chemical supply tanks should be located as far away from the water source as is
possible. The slope of soil surface should be graded to force drainage away from the
water source in the event of a chemical leak or spill.
Solenoid valve. For further safety, a normally closed solenoid valve on the chemi-
cal suction line can be electrically interlocked with the engine or motor driving the
injection pump. This valve, which is located on the suction side of the chemical injec-
tion pump, provides for a positive shutoff of the chemical injection line. The chemical
cannot flow if the injection pump is stopped. The solenoid valve should be constructed
of chemically resistant materials since it will be in contact with the chemical concentrate.
19.2.2. Injector pump interlocks. An interlock system should be used between the
power system of the injection unit, the irrigation pumping plant, and the irrigation
system, if it is electrically controlled. Interlocks can be accomplished electrically, hy-
draulically, and mechanically. The interlock should function so that if the irrigation
pump stops, the injection pump will also stop. This type of interlock is referred to as a
one-way interlock. A disadvantage to the one-way interlock is that some motorized
irrigation systems may continue to run if the injection pump stops. In this situation, it
may be difficult for the operator to determine where the chemical treatment ended in
the field.
A two-way interlock ensures that both systems shut down simultaneously and, for
non-electrical systems, can be accomplished by installing a flow or pressure sensor on
the high-pressure injection line. A loss of pressure in the chemical injection line due to
injection-line breakage, an empty chemical tank, a plugged strainer, or injection pump
failure would then trigger a simultaneous shutoff of both the irrigation system and the
chemical injection system.
19.3 INJECTION SYSTEMS
There are four basic types of injection systems used for chemigation. These are
centrifugal pumps, positive displacement pumps (piston, diaphragm, gear, lobe, peri-
staltic, and others), pressure differential injectors, and Venturi injectors. Some injec-
tors may be a combination of these types. Pressure differential systems are often the
least expensive and are also the least accurate. The basic costs of Venturi injectors and
water-drive systems are lower than the other alternatives; however, the cost of these
injectors with feedback and control systems is in the range of the cost of positive dis-
placement pumps. Positive displacement pumps are generally preferred for most
chemigation applications due to their accuracy in metering chemicals into the system,
and backflow prevention.
Chemicals injected can be gases or liquid; however, this discussion will focus on
water-based injection systems. Injected gases such as chlorine and anhydrous ammo-
nia have their own special requirements, and producers should work with their chemi-
cal supplier for these unique applications. Several companies manufacture liquid injec-
tion pumps for chemigation systems. These devices are used to apply water-soluble
fertilizers, pesticides, plant growth regulators, wetting agents, soil amendments, min-
eral acids, and various other chemicals.
19.3.1 Injector Selection
The primary selection criteria for injectors are durability, accuracy, ease of opera-
tion and repair, service life, flow-rate range, and resistance to corrosion by the chemi-
cals being used. Other important considerations are the cost, available power source,
chemicals to be injected, and the number of chemicals to be injected simultaneously.
Injection systems should be able to inject any water-soluble chemical, acid, emulsi-
fiable concentrate, or wettable powder at low concentration levels (e.g., 1 to 100 mg/L).
Solutions may have a wide pH range or other corrosive index that may require multi-
ple injectors made from different materials. Separate injectors may also be required for
fertilizers and pesticides as fertilizers typically require a higher injection rate than the
other chemicals. It is often not possible to accurately adjust an injector of the type re-
quired for fertilizer to inject at the low flow rate required for pesticides or acids.
The choice of an injector will depend on consideration of the following factors:
What is the size of field, pesticide and fertility programs, and types of crop(s) to
be covered?
What are the irrigation system water flow rates?
What chemicals will be injected, at what desired concentrations (particularly if
acid injection is desired)?
Are multiple injection heads needed (for incompatible chemicals)?
What are the pressure and flow requirements for proper operation of the injec-
tor?
Is a portable or stationary injector desirable?
Once the size and type of injector are determined, consider the ease of repair and
reported longevity of the unit. Growers should also consider the manufacturer’s reli-
ability, technical support, service, and other qualifications. Much of this information is
available on the Internet, and by talking to other growers and extension personnel.
Injection of multiple chemicals at the same time requires careful planning to ensure
compatibility, separation distances, and flexible pump operations over a range of con-
ditions. Some positive displacement pumps are available that can drive multiple, sepa-
rate heads for each different chemical.
Feedback and control systems are available that measure irrigation pipeline flow
rate and automatically adjust Venturi or positive displacement pump injection flow
rates. For example, drip irrigation systems may have several different sizes of blocks
or blocks with different water requirements. A feedback and control system could ad-
just the rate of injection as the irrigation system is automatically changed from block
to block. Controllers can also manage variable speed drives to provide a broad range
of injection rates, which is advantageous with large (e.g., center pivots) or multi-zone
systems with either constant or varying flow rates.
19.3.2 Centrifugal Pump Systems

Small radial-flow centrifugal pumps (booster pumps) are often used to inject
chemicals into irrigation systems. The pump draws water directly from the chemical
reservoir and delivers a pressure that is higher than the pressure in the irrigation line to
inject the chemical. Consequently, the flow rate of the chemical from the pump de-
pends on the pressure in the irrigation line at the time of injection. To accurately de-
termine the amount of chemical being injected requires calibration while the system is
operating. Because these systems are sensitive to irrigation system pressure fluctua-
tions and changes, they are not recommended for the injection of toxic chemicals
where the injection rates must be precisely controlled.
19.3.3 Positive Displacement Pumps
Positive displacement pumps are the most recommended injection system. They are
generally classified as reciprocating (piston and diaphragm), rotary (gear and lobe), or
a miscellaneous type (e.g., peristaltic, progressive cavity), depending on the mecha-
nism used to transfer energy to the fluid. Few rotary and other types of positive dis-
placement pumps are used for chemical injections into irrigation systems. Gear-and-
lobe rotary pumps and peristaltic pumps can be used when only small injection rates
are required. This discussion only covers piston pumps, diaphragm pumps, and com-
bination piston/diaphragm pumps, which are the most commonly used types for
chemical injection into irrigation systems.
Positive displacement pumps are recommended where precise control of injection
flow rate of chemicals is required, as with pesticides. They are also the pump of choice
for highly viscous materials. They are easy to monitor and calibrate, and most can be
adjusted while in operation (preferred).
Piston, fluid-filled diaphragm, and piston/diaphragm pumps come closest to being
ideal positive displacement pumps. These systems can typically control injection flow
rates with a range of error of ±1% to ±2%. Injected flow rates remain constant over a
range of chemical viscosities and irrigation pipeline pressures as long as there is
minimal pressure in the irrigation pipeline. Both piston pumps and diaphragm pumps
are adjustable; however, some pumps cannot be adjusted while the system is running.
A piston or a diaphragm displaces a given amount of chemical with each stroke.
Both piston and diaphragm pumps consist of a pumping device and two check valves
(Figure 19.1). As the diaphragm or piston is retracted, chemical is drawn into the
chamber from the tank through the first check valve. As the diaphragm or piston is
pushed out, the chemical is forced out the second check valve into the irrigation pipe-
line. Diaphragm pumps have a small flexible Teflon or rubber diaphragm that is
moved in and out of a small chamber. Diaphragm pumps are thus less prone to corro-
sion than piston pumps, because they have less contact with the chemical. For piston
pumps, which draw chemical into a long metal cylinder, care must be taken to select a
piston and cylinder that will not be corroded by the chemicals.
Positive displacement pumps can be purchased that cover a wide range in flow
rates and chemical properties. For example, piston pumps can be purchased with two
different sized cylinders on either side of the pump in order to provide a range of in-
jection flow rates. Diaphragm pumps generally have a fixed operating range, but dif-
ferent injector heads can be purchased to provide a broader range of flow rates as well
as chemicals.
Figure 19.1. Schematic showing a typical layout for positive displacement

injection pump systems (from van der Gulik, 1993).
Positive displacement pump systems are especially adaptable to feedback and con-
trol systems, which provide additional management flexibility. Variable frequency,
electric motor drives are often used with control systems.
Since positive displacement pumps displace a fixed volume of fluid with each
stroke, they can produce very high injection pressures. This situation should be
avoided (e.g., trying to operate against a closed valve in a discharge line) since this
will often result in pump or injection line damage.
Three-phase electric motors are the most common source of power for piston and
diaphragm pumps. Some injection pumps are driven by belt power, compressed gases,
or a water motor. Another problem with electric injector pumps is that they can con-
tinue to inject chemical once the irrigation system is shut down. Preventing independ-
ent operation requires electrical and mechanical interlocks and careful monitoring.
A variation of positive displacement pumps are water-pressure driven injectors. As
is the case with Venturi systems, the water-driven systems are installed on a parallel
bypass line that draws water from the main line above a flow constriction through the
device to inject the material. However, in this case, the water powers a small turbine
or piston device. The injection rate is controlled by regulating the amount of water
going to the drive unit. Water-driven injection devices generally operate at injection
ratios from 0.2% to 2% of the total flow. The turbine uses system pressure to drive a
small cam and piston rod unit to move a diaphragm or piston injector. Piston water
drives use as small amount of water (about 3 times the amount injected), but do not
lower system pressure due to their operation. A drain needs to be provided for disposal
of water from the water-powered piston drives since the water is not returned to the
system.
Water-powered injectors are often referred to as ratio or proportional feeders be-
cause the quantity of material injected depends on the flow rate through the injector,
which is a function of the pressure in the main line. Thus, the concentration of chemi-
cal in the irrigation water will remain the same because it is always proportional to the
system flow rate. This characteristic can be an advantage in situations where the sys-
tem automatically changes from one zone to another with variable flow requirements,
as long as the concentration in each zone remains the same.
19.3.4 Venturi Injection Systems
Venturi injectors rely on the Venturi pressure drop principle to draw chemical from
the tank into the irrigation pipeline. A small Venturi (Figure 19.2) can be used to inject
chemicals into a relatively large mainline by shunting a portion of the flow through the
injector. To assure that the water will flow through the shunt and injector, a pressure
drop must occur in the main line. This is created by a partially open valve, an orifice,
or other obstruction. These obstructions are located between the mixing water supply
fitting and the injection point. It serves to increase water velocity and decrease pres-
sures to values below atmospheric pressure in the throat of the Venturi device.
Most chemical stock tanks in Venturi systems are vented to atmospheric pressure
and the pressure differential between the atmosphere and the throat of the Venturi de-
vice forces the chemical solution into the irrigation pipes at this reduced-pressure
zone. Mixing of chemical with irrigation water is facilitated by the flow velocity in the
Venturi device. Because chemical is sucked into the irrigation system after the main or
Figure 19.2. Schematic of the basic Venturi injector system.

Courtesy of Mazzei Injector Corporation (MIC).
booster pumps, there is no contact between the chemical and the pump; thus, Venturi
injection systems are less susceptible to corrosion than some other types of injectors.
Venturi injectors come in various sizes and can be operated under different pressure
conditions. Suction capacity (injection rate), head loss required, and working pressure
range will depend on the model. Venturi injection systems, with or without feedback
and control, can be purchased in units with injectors of different sizes to provide a
broad range of injection flow rates. A large chemical stock tank may be required be-
cause the proportion ratios are low (typically in the range of 1:15). This tends to limit
the use of Venturi-type injectors to relatively small areas.
A Venturi injector does not require external power to operate, although ancillary
equipment (e.g., control systems, booster pumps) may require power. It does not have
any moving parts, which increases its life and decreases probability of failure. The
injector is usually constructed of plastic and it is resistant to most chemicals. It re-
quires minimal operator attention and maintenance. Since the device is very simple, its
cost is low as compared to other equipment of similar function and capability.
The flow rate of a Venturi injector can vary dramatically due to changes in irriga-
tion pipeline pressure. Most Venturi injectors require at least a 20% differential pres-
sure to initiate sufficient vacuum for proper operation. However, as long as the pres-
sure differential across the upstream and downstream side of the Venturi is greater
than 200 kPa, the injection flow rate is relatively insensitive to irrigation pipeline pres-
sure. This situation creates a minimum pressure within the Venturi throat that is
slightly above the limit of zero absolute pressure. The pressure differential at which
each Venturi injection system reaches a minimum can be observed on the manufac-
turer’s flow rate vs. pressure differential curves. If a constant chemical injection flow
rate is required throughout the event, then a high pressure differential across the Ven-
turi must be maintained or a centrifugal pump is used to boost pressure in the Venturi
bypass line. The piping system shown in Figure 19.1 is not suitable for accurate cali-
bration and constant injection rate. Thus, Venturi systems may also feature a pressure
regulator for more precise control of injection rates.
The flow rate of a Venturi injector can be quite sensitive to temperature changes
because the viscosity of some chemicals can vary substantially with temperature. For
example, the change in injection flow rate can be in the range of 5% to 10% over a
20°C temperature range for relatively viscous fertilizers such as UAN32 or CAN17.
However, there is basically no change in flow rate due to temperature for chemical
solutions with viscosities in the range of water because these viscosity changes with
temperature are minimal. When chemical viscosity is an issue, a good method to keep
flow constant over a temperature change is a control system with feedback from a
flow sensor.
It is also important to realize that the suction capacity depends on the liquid level in
the supply tank. As the liquid level drops, the suction head increases, resulting in a
decreased injection rate. To avoid this problem some manufacturers provide an addi-
tional small tank on the side of the supply tank, where the float valve maintains the
fluid level relatively constant. The fluid is injected from this smaller auxiliary tank.
Often the injection criterion is that a known volume of chemical must be pumped
during a specified period for each irrigation zone, but the concentration does not have
to remain constant. One commonly used alternative for pressurized irrigation systems
is the combination of a Venturi device with a pressurized chemical tank. Since the
water flowing through the tank is under pressure, a sealed airtight pressure-supply tank
(constructed to withstand the maximum operating pressure) is required. However, the
injected concentrations will change gradually due to the dilution of the chemical in the
tank as the water enters the tank during injection.
19.3.5 Pressure Differential Systems
Mixing-tank injectors and proportional mixers located on the discharge side of the
irrigation pump are two approved pressure differential injection systems. Various me-
tering or proportioning valves are used with mixing and proportioning tanks that oper-
ate on pressure or flow changes in the irrigation system. Frequently, these devices are
an application of a Venturi meter or an orifice with changing diameter.
Using the irrigation pump’s suction line is not an approved method of injection in
most jurisdictions. The reason is that it is difficult to prevent contamination of surface
water sources in the event of an unexpected pump shutdown. Chemical solution can
then easily be siphoned out of the tank and into the water supply. It is also difficult to
monitor or determine the chemical flow rate as it depends on the amount of the
pump’s suction at the injection point, the length and size of the suction line, and the
level of chemical in the supply tank. It is also difficult to adjust the injection rate in
these systems. In addition, the chemical solution is drawn through the pump and could
result in significant corrosion and breakdown of seals and bearings. This method of
injection is not recommended.
A mixing-tank injector operates on the discharge side of the main pump and oper-
ates on the same pressure differential concept as Venturi systems. They are installed
on a similar bypass shunt that runs parallel to the main irrigation line. However, in
these simple, low-precision systems the chemical reservoir is sealed instead of being
vented to the atmosphere. Water from the main line enters the reservoir and displaces
the same volume of the chemical mixture into the irrigation system. This results in a
constant rate decrease in the chemical concentration in the reservoir, requiring that
specific chemical batches be prepared for each and every zone to ensure that each
zone receives approximately the right amount of material. The injection into the main
line is often controlled by a metering device installed on the inlet side of the injector.
Proportional mixers are a modification of pressurized mixing tanks and operate on
a volume water displacement principle. It is commonly used in irrigation systems
where flow fluctuations are expected, perhaps due to varying set sizes or sharing water
with other users. Sufficient concentrated chemical for one injection event is contained
in a collapsible, impermeable bag that is placed into the pressure tank. The chemical
remains separated from the water and feeds chemical solution into a proportioning
valve that injects it into the irrigation system. Chemicals in the bag are forced into a
proportioning valve by the pressure differential and injected into the irrigation system.
The volume of chemical is displaced by water entering the tank at the same rate the
chemical is injected. This makeup water does not mix with the chemical and is not
returned into the system. Injection is completed when the bag is totally collapsed and
the tank is full of water. The proportioning valve responds to the changes of flow, but
not pressure changes in the irrigation system. Thus, as long as the pressure and the
flow rate in the system do not vary significantly, the injection rate will remain rela-
tively constant. To insert a new bag of chemical the tank is isolated with valves and
drained.
19.3.6. Operation and Maintenance of Injector Systems

It is highly recommended that injector systems utilize some method (mechanical or
liquid) to agitate the injector supply tank to keep the chemicals in solution, especially
when working with many pesticides and dry chemicals. In addition, if large fertilizer
tanks are not used for extended periods, water-soluble fertilizers can accumulate in the
bottom of stock tanks due to density differences, resulting in large differences in fertil-
izer concentrations during the injection event. Also, dilution ratios higher than 1:200
require agitation as fertilizer may not dissolve completely due to exceeding the solu-
bility limits of the chemical.
The following practices are recommended in most installations:
Chemical injections should be made in the center of a water stream (e.g., center
of a pipe diameter), whenever possible, for better mixing.
Keep the injector’s suction line as short as possible (1.5 m or less),
The suction hose on the injector should also have a strainer to prevent precipi-
tated and non-water soluble materials from entering the injectors and the irriga-
tion system.
Stock tanks should be covered to prevent algae and/or debris buildup, contami-
nation, or evaporation of stock solution.
Stock tanks need to be opaque since the chelating agents in some micronutrient
fertilizers tend to break down if they are exposed to light.
Injectors should not be exposed to freezing temperatures as cracking and/or
warping may result.
Suction and discharge tubing needs to be replaced regularly (e.g., every couple
of years).
Intake strainers should be suspended 7 to 10 cm (3 to 4 in.) from the bottom of
the solution tank to avoid pulling up undiluted concentrate.
Inject clean water after use to flush the system.
Regularly clean the solution tanks (weekly or biweekly, depending on frequency
of use) to prevent dirt and scale buildup that might plug or abrade the injectors.
Suction tube strainers should be cleaned using clean, clear water and inspected
regularly for clogs and/or cracks.
Inspect and service O-rings. Petroleum-based lubricants such as Vaseline, lano-
line, WD-40, or motor oil should never be used on dosage pistons or seals.
19.4 INJECTION SYSTEM CALIBRATION
The balance between chemical injected and irrigation water flow rate are critical for
proper chemigation. If there is too little water in the chemical and water mixture there
will be uneven chemical distribution, high volatilization (chemical loss), or chemical
buildup in irrigation lines. On the other hand, too much water in the chemical mixture
will result in possible dilution of chemical below effective concentrations or loss of
the applied chemical to groundwater (Smajstrala et al., 1986; Clark et al., 1998;
Scherer et al., 1999; Werner, 2002; Evans and Waller, 2007).
Consequently, calibration of the injector flow rate is very important. In this process,
how much chemical is being injected using different settings is determined. Periodic
calibration during the irrigation season is needed to ensure that an injector is operating
properly. Calibration methods are focused on either mass or volume determinations of
injection rates, and are independent of concentrations that must be determined sepa-
rately. Adjustments may be done by adjusting the injection rate (e.g., in a Venturi), or
by adjusting the concentration of the chemical stock solution.
There are numerous ways to calibrate injection systems depending on the type of
system and chemical involved. Many chemicals are supplied as either a percentage by
weight of a dry or liquid mixture. Therefore, the mass of chemical mixture required
will depend on the concentration of raw chemical in the mixture. When chemicals are
supplied in liquid form, as is often the case with chemigation, it is more convenient to
measure volumes rather than masses or weights, so volumetric calibrations are the
most common. The owner’s manual for the injector system should recommend ways
to test a specific injector.
For supply tanks that are vented to the atmosphere, it is possible to calibrate injec-
tors by weighing tanks with scales or load cells. This is also the recommended proce-
dure for gaseous chemical injector systems. Where other techniques are impractical,
fluorescent tracer techniques can also be an accurate means to calibrate injectors.
The injection rate of the chemical injection pump should be determined for a par-
ticular setting of the injection rate control knob, with the irrigation system operating,
so the injection pump is working against the water pipeline pressure. Otherwise, the
tested injection rate will be higher than when the injection pump is injecting into a
pressurized irrigation pipeline. Calibrations should be done at constant pressure with
clean water and no chemicals. The injector should also be operated prior to the test to
remove any air bubbles in the lines.
Volumetric calibrations are based on letting the injection pump draw from a cali-
brated container on the suction side of the injector pump. Usually, a calibrated sight
tube is used to determine the volume extracted from the chemical supply tank. Flow
meters on the injection lines may also be used to indicate volumetric injector flow
rates.
The water volume being injected is measured over a given time at each particular
injector setting (usually expressed on a dial as a number) for a predetermined amount
of water (making sure that the volume is appropriate for the container). The container
should be filled to a known volume. The injector and a stop watch or timer are turned
on simultaneously. Record the time, line pressure, and volume when the volume in the
container is reduced by the predetermined amount. This process is repeated for differ-
ent dial settings. There are some special considerations for stationary, continuous
move and surface irrigation systems.
19.4.1 Stationary Irrigation Systems
Stationary irrigation systems include hand move and wheel move sprinklers, per-
manent solid set sprinklers, and microirrigation systems (includes drippers and mi-
crosprinklers.) Chemigation with stationary systems is a relatively simple procedure if
the chemigation begins and ends within the duration of a single set as the chemical is
applied to a given area. Application through sprinkler systems will only be as spatially
uniform as the overlap water application pattern. The injection is repeated at the same
rate and duration for each set until the entire field has been covered
Microirrigation applications generally do not have overlap from adjacent water ap-
plication devices, and uniformities are primarily a function of time and the coefficient
of uniformity of the individual devices. Under good management, this is usually
highly efficient with few losses because the plants’ roots have developed over time to
fully utilize the applied water and chemicals in the wetted soil volume. Some new
systems have long microirrigation laterals with large diameters to prevent excessive
friction loss. Chemigation uniformity and duration can then be especially difficult be-
cause of the length of time it takes for water and chemical to reach the last emitter.
19.4.2 Continuous Move Irrigation Systems
Continuous move irrigation systems include center pivots, linear moves, and trav-
elers. For these systems injections must be at a constant concentration over the entire
irrigation to ensure the same amount of chemical is applied to the whole field. If the
rate varies as the machine moves, variations in chemical applications will be exacer-
bated. The injection rate must be calibrated so that the full application continues for
the entire duration of the irrigation event (e.g., one complete revolution). The grower
does not want to run out of material before the end of the irrigation nor does he or she
want to apply an additional application to lands that were previously covered in that
pass.
End guns and corner systems present some unique chemigation challenges. Inter-
mittent end gun operation and the variable irrigation system flow rates as a corner sys-
tem swings out and back in again changes injection flow rates and may also affect
pressure and application uniformities. In these situations, chemigation most commonly
utilizes a flow-meter feedback controller system tied to the angle resolver at the pivot
(through the control panel) and/or a GPS system to regulate injection rates as irrigation
system flow rates change. Calibration of the injection system must consider the full
range of flow rates and the pump’s response characteristics as both flow and pressure
vary. End guns may be turned off in certain areas of the field to prevent potential
safety or environmental problems.
19.4.3 Injection Flow Verification
Once the actual chemigation operation is started, the injection rate should be re-
checked and the meter readjusted (if necessary) at least once during the chemigation
process. A flow indicator installed on either the suction or outlet tubing of an injector
is an outstanding diagnostic tool. Some will allow the grower to tell with a glance if
the pump is working properly.
There are many types of flow meters to measure chemical injection rates, including
variable-area (called tapered-tube or rotometer), positive displacement, paddlewheel,
pitot tube, ultrasonic, and mass flow meters. Each type has its special features and
there is a wide range in cost. The variable-area flow meter offers many advantages
over some more costly and technical types, making it the meter of choice for many
chemical flow measurement applications. Some manufacturers offer various flow in-
dicators as standard equipment.
Another indicator of the fertilizer injector output is by measuring the electrical con-
ductivity (EC) of the dilute fertilizer solution. Good estimates of fertilizer concentra-
tions can be obtained using a portable EC meter where the water is applied in the field.
Permanent EC monitors, which may also measure pH, can be installed a suitable dis-
tance downstream of the injection point. Alternatively, the concentration of a sample
of the water-fertilizer mixture can be analyzed by a reputable testing laboratory.
19.5 IRRIGATION SYSTEM CONSIDERATIONS
The design and operation of irrigation systems will impact the effectiveness of a
chemigation system. There are many operation characteristics that need to be consid-
ered, but the main concern is the distribution uniformity of the system. It is virtually
impossible to achieve good chemical application uniformity if the irrigation system is

not designed to operate at a high uniformity. Center pivot systems with drop tubes and
trickle/drip systems provide the best uniformity. Sprinkler and gun systems are sus-
ceptible to wind drift, pressure variations, and crop interference, which often reduces
the achievable uniformity.
19.5.1 Irrigation System Characteristics
The physical characteristics of the irrigation system will determine the type of in-
jection system, selection of chemicals that can be applied, application rate, and dura-
tion of application.
Irrigation systems are either stationary and continuous move systems. Stationary
systems include handlines, wheel move, solid set sprinklers, trickle, and microsprin-
kler systems. These systems irrigate a block of land at a constant application rate over
time. A batch of chemicals can therefore be mixed and applied during the irrigation
interval. The duration of application will be determined by the type of chemical being
applied. Some chemicals must be incorporated to be effective, requiring application of
enough water to move the chemical into the soil. Other chemicals, such as nitrate ni-
trogen, are very mobile and should be left near the soil surface to avoid potential
groundwater contamination. Chemicals intended for foliar applications should only be
applied by overhead sprinkler systems during the end of an irrigation set with a mini-
mum amount of flushing.
Continuous move systems include center pivots, lateral move, traveling guns, and
overhead boom systems used in nursery and greenhouse operations. These systems
irrigate a block of land with a predetermined amount of water but at varying applica-
tion rates (the application rate will be greater at the last tower, especially on those sys-
tems using an end gun, than the application rate at the pivot point). The rate of chemi-
cal injection must therefore be matched with the rate of travel. Batch application of
chemicals cannot be used with these types of irrigation system.
Table 19.1 provides some general guidance on the uniformities that can be
achieved. Depending on design, installation, and operation some systems may be rated
higher or lower than what is shown. Chapter 5 provides more detailed information on
irrigation system uniformities.
All system components must be able to withstand the effects of corrosion from in-
jected basic and acidic chemicals at the expected concentrations at each location. Con-
sult material compatibility tables and match materials with the chemicals. For exam-
ple, concentrated sulfuric acid is not compatible with PVC tubing and will quickly
Table 19.1. Irrigation system uniformities.

Irrigation Method Uniformity
Wheel move system fair
Handline fair
Solid set over-tree sprinkler fair
Solid set under-tree sprinkler poor
Solid set gun poor
Microsprinkler fair
Traveling gun poor
Center pivot good
Surface systems fair
Trickle/drip good
create a hazardous situation, but PVC is acceptable when the acid is greatly diluted.
As a general rule, uncoated metallic components (except specific stainless steels)
should be used as little as possible.
19.5.2 Site Considerations
Proximity of the irrigation system to dwellings, neighboring crops, roadways, play-
grounds, and residential areas, and surface water sources such as ditches, streams and
lakes, must be carefully considered. The safety of people, wildlife, domestic animals,
and other non-target areas must be considered.
19.5.2.1 Topography. Field topography can cause pressure differences along an ir-
rigation lateral, which affects application uniformities. Low-pressure center pivot and
trickle systems are most susceptible to pressure differences caused by elevation or
friction loss. Sprinkler irrigation systems can also be affected if the elevation changes
are in excess of 5 m. Pressure or flow regulators on each individual sprinkler head
may be required to ensure uniform discharge for each sprinkler along the lateral. Pres-
sure-compensating emitters are recommended to maintain system uniformity for
trickle systems.
19.5.2.2 Crop. The crop grown often determines the type of irrigation system that
will be used. Trickle irrigation systems are often used for tree fruits, grapes, strawber-
ries, and other horticultural crops because of water application efficiency and the abil-
ity to control fertilizer application directly to the plant roots (Doerge et al., 1991; Burt
et al., 1995; California Fertilizer Association, 1995; Rosen and Eliason, 1996). High-
density orchard plantings require quick tree response after planting and early fruit de-
velopment to realize a return on capital investment. However, the injection of some
pesticides through a trickle system may not be effective.
Cranberries and forage crops are well suited for sprinkler irrigation systems. Over-
head and under-tree sprinkler systems are also popular on orchards and vineyards for a
variety of reasons. Overhead sprinkler systems used for irrigation, frost protection, or
crop cooling on tree fruits or cranberries can be adapted to chemigation. A minimum
application uniformity of 80% should be achieved.
Center pivot and lateral move overhead sprinkler irrigation systems are commonly
used for corn and forages, but can be used on any crop providing that the fields are
large enough to accommodate a pivot.
19.5.2.3 Soil type. The maximum infiltration rate of water into the soil and avail-
able water storage capacity of the soil differ with soil type. Soil types can vary signifi-
cantly over an entire field requiring a change in the operation of the irrigation system.
Coarse-textured soils can have high infiltration rates but can store very little water
within the root zone. Conversely, fine-textured soils can store large amounts of water
but have low infiltration rates. High application rates on fine-textured soils increase
runoff potential, while excessive amounts of irrigation on coarse-textured soils in-
crease the potential for leaching of chemicals below the crop root zone and into
groundwater.
Chemigation systems must be operated within the limits of the soil types present to
reduce the potential for runoff and for leaching. An understanding of soil water-
holding capacity is also important to ensure that the chemical added is moved into the
soil to an appropriate depth with respect to the plant rooting volume.
19.5.2.4 Drift and runoff potential. Drift and runoff are two leading causes of
chemical losses from chemigation systems. Leaching can also be a cause of chemical
losses. Environmental conditions during application, sprinkler types, type of chemical

being applied, and climatic conditions after application all affect the magnitude of
chemical losses.
Water discharged from some sprinkler nozzles under pressure emerges as a fine
spray. The amount of physical drift will depend upon the sprinkler type, system pres-
sure, wind speed, and crop height and density. Part of the spray is evaporated within
the wetted area, intercepted by vegetation and soil, or carried away by wind outside
the intended target area. Wind drift can cause a potentially hazardous situation.
Chemigation should not be carried out if wind conditions are strong enough to cause
significant drift to non-target areas. Some pesticide labels carry statements prohibiting
application when wind speeds are sustained above a certain limit.
Many center pivot systems will turn off high-pressure end gun sprinklers as they
move around the field to avoid water applications to roads, buildings, or waterways. In
these cases, the end guns are often turned off during the whole chemigation event to
avoid changes in system pressure as these sprinklers cycle on or off in various sectors
of the field, as that could cause poor uniformities along the entire lateral length. Com-
bining the high pressures, small droplets produced, and the end guns being mounted 3
to 4 m high directly on the lateral piping creates a greater chance of chemical drift
beyond the intended application area.
Runoff depends not only on the irrigation system application rate and soil infiltra-
tion rate but is also influenced by factors such as field slope, surface vegetation, crop
cover, and soil surface residue. Irrigation systems applying chemicals should be de-
signed and operated to prevent any runoff from occurring. Some regions will require
runoff to be contained before it leaves property boundaries or the place of use. Treated
water may be required to be reused on the crop or site treated. Chemical label restric-
tions may prohibit the use of treated water on crops or sites other than those listed on
the chemical label.
19.5.3 Design Considerations for Sprinkler Systems
Sprinkler irrigation systems used to apply chemicals should be designed with a co-
efficient of uniformity of at least 80% and preferably 90% whenever possible. Even
for uniformities of 90% the ratio of the depth of irrigation applied on one part of the
field compared to another can be as high as 3:1. A coefficient of uniformity of 80%
can only be obtained by designing sprinkler systems to the following minimum stan-
dards:
The maximum pressure variation along the lateral must not exceed 20% of the
sprinkler operating pressure and 10% (or less) pressure variation is desirable.
Flow-control nozzles must be used if pressure fluctuations exceed the 20% al-
lowance. Another option is to use pressure regulators at sprinkler heads operat-
ing at pressures exceeding the normal operating range.
For stationary sprinkler systems the sprinkler spacing along the lateral and the
lateral spacing should not exceed 50% of the sprinkler wetted diameter. If the
predominant wind speed exceeds 5 km/hour then the spacing should be reduced
to 40% of the wetted diameter.
The sprinkler should be operated within the manufacturer’s recommended pres-
sure range to provide adequate stream breakup for proper dispersal.
The sprinkler should have a rotation time of less than 1 min and the system op-
erated for at least 15 min when chemigating to improve uniformity.
19.5.4 Microirrigation System Design Considerations

Microirrigation systems include drip or trickle systems that apply irrigation directly
to the soil surrounding plant roots rather than the entire field. All microirrigation sys-
tems should be designed so that biocides, fertilizers, and other chemicals can be in-
jected and uniformly applied through the irrigation system. Microirrigation inherently
offers tremendous benefits for chemical injection and applications. Consistent soil
water contents and wetted soil volumes tend to increase plant uptake efficacy of many
chemicals. Water-soluble nutrients can be injected to closely match crop requirements,
increase nutrient use efficiencies, and reduce costs.
Microirrigation systems can be operated to achieve an emission uniformity of 90%
if care is taken in the layout and design of the system. The system should be designed
and operated to achieve these high application uniformities to avoid undesirable leach-
ing and ensure good application uniformity since the chemical application uniformity
cannot exceed the water application uniformity.
Microirrigation systems operate at efficiencies in the 85% to 95% range compared
to sprinkler systems that are only 60% to 80% efficient. Microirrigation systems are
therefore much superior chemigation systems compared to most sprinkler systems.
Low-pressure center pivots systems with drop tubes are the only systems that can
come close to matching the performance of drip/trickle systems for fertigation. How-
ever, these systems are limited in their ability to apply herbicides and insecticides ef-
fectively.
Factors that should be considered in the design of microirrigation systems for
chemigation are:
Emitter spacing should ensure that at least 60% of the plant root area is irrigated
during drier seasons.
The emitter selected should provide good uniformity and match the terrain, crop
type, and available water quality. Emitter flow characteristics and product dura-
bility should be considered. The manufacturer’s variance coefficient should be
less than 0.05 and preferably 0.03. Testing of emitter flow rates at the beginning
and end of the zone will confirm that flow rate uniformity is within acceptable
limits.
If pressure compensating emitters are not used the emitter operating pressure
range should be kept within ±10% of the emitter operating pressure.
All chemical injections should be filtered. Injection should occur after the pump
and before the media or final screen filters to trap any undissolved material. In-
jection installations should always provide for complete mixing and uniform
concentrations before the chemicals reach the field. Materials should be injected
into the center of the water flow to ensure quick dilution to reduce deterioration
of the filter tanks, piping, valves, or other components. Generally, injection rates
should not exceed 0.1% of the system water flow rate although concentration
limits (e.g., for chlorine) and label requirements for pesticides are usually less.
The injection of chemicals often increases the susceptibility for emitter plug-
ging. There are numerous products being promoted as universal line or emitter
cleaners, biocides, and fertilizers. These products should be used only if these
claims can be proven with unbiased, high-quality research. Many of these mate-
rials are costly, only treat symptoms without addressing the underlying prob-
lems, and eventually fail.
19.5.5 Design Considerations for Surface Systems

Chemigation with surface irrigation systems usually involves only fertilizers or
other soil amendments (e.g., PAM). Pesticides are rarely injected into surface systems
due to environmental concerns due to runoff.
Irrigations should be designed and managed to ensure distribution uniformities as
high as possible during chemigation. Options for high uniformity include surge flow
and reducing the length of runs, especially on cracking soils. There should be provi-
sions for the collection and reuse of all tailwater. The tailwater should be reused on the
same set as much as possible. Injections should not exceed the time needed for each set
so that each set receives the correct amount of chemical. Chemicals are typically injected
only for about half of the total set time, and there are various alternatives available.
One option is to start injection after the water has advanced half the length of the
furrow or border to reduce excess leaching at the head of the field. However, tailwater
management and ensuring uniform applications is quite difficult. More commonly,
each set is irrigated for the same duration at the same flow rates and concentrations.
Chemicals must be injected continuously into the incoming water supply at a con-
stant rate at sufficient distance upstream to ensure adequate mixing for the duration of
each injection period. Liquid or powdered fertilizers and soil amendments can be
added to the incoming water supply by gravity or battery-powered devices. Gases such
as anhydrous ammonia are often injected by a hose inserted in the ditch or pipe using
the tank pressure to power the process.
19.5.6 Operation Considerations
In addition to system selection and design there are many operational considera-
tions to be taken into account to ensure an effective chemigation program is carried
out in a safe manner.
19.5.6.1 System inspection. Injection and safety equipment should be flushed to
prevent the accumulation of precipitates in the injection equipment. The injection
pump, injection lines, and the injection-line check valve should be flushed after each
use. This equipment should be flushed with clean water or other solvent as specified
by the chemical label.
All injection and safety equipment should be inspected prior to each chemigation
event. The operator should follow the manufacturer’s recommendations for cleaning
and maintenance of this equipment. Inspection of these devices will minimize the po-
tential for failure. Moving parts should be lubricated as necessary before each chemi-
gation event and before storage during the off season to preserve the equipment.
The irrigation-pipeline check valve should be repaired or replaced if leakage is
found. Operators should not chemigate if the check valve leaks. Remember, the low-
pressure drain is for backup only.
The low-pressure drain should be inspected before each chemigation event. If the
drain is functioning properly, some water should discharge from the outlet immedi-
ately after start-up. The drain valve should eventually close as the system pressure
increases.
The injection-line check valve should be inspected for backflow prevention. Re-
move the injection line from the inlet side of the chemical line check valve and ob-
serve whether back leakage occurs when the irrigation system is pressurized. Remem-
ber to bleed any trapped air or pressure from the injection line prior to removal. You
may also inspect the chemical injection-line check valve for leakage in the normal
direction of flow by removing the check valve from the injection port. To inspect the
valve, insert the discharge end of the injection-line check valve into a bucket and start
the injection pump. Pump some chemical or water through the system, then shut off
the injection pump and observe whether leakage occurs through the chemical injec-
tion-line check valve.
The interlock system should also be inspected before each use as switches and
other items could fail after weathering and wear.
Finally, inspect the irrigation system joints, fittings, and nozzles. Check system
nozzles and outlets for wear and function ability. Check all fittings and connections on
the system to avoid leaks and possible over application.
Operators should chemigate only with reliable, well-maintained injection and
safety equipment. Chemigation can be a relatively safe practice provided proper pre-
cautions are taken.
19.5.6.2 Irrigation operation characteristics. Physical characteristics of irrigation
systems that affect the uniformity of chemical injection include the following:
Solute dispersion occurs as the chemical travels along the irrigation pipeline.
The friction affect of the pipe walls on the fluid motion causes this dispersion. A
slug of chemical injected into an irrigation system becomes diffuse as the
chemical travels along with the irrigation water.
The irrigation main line contains a significant amount of water. The travel time
for the chemical to reach the discharge point and the time required to flush the
system must be considered. The irrigation system should be flushed after the in-
jection is complete. The operator should flush the irrigation system for at least
10 to 15 min after each chemical injection period.
The operating flow rate for each zone will be different.
For stationary systems, uniformity of application generally increases with the
length of set time. Chemicals that require a short application duration may be
difficult to apply uniformly.
For continuously moving irrigation systems such as center pivots the amount of
chemical that can be applied is dependent on the travel speed and concentration
of the chemical solution. The available chemical solution may dictate the irriga-
tion travel speed. Check to ensure that the speed chosen will provide good appli-
cation uniformity.
19.5.6.3 Operation guidelines. The following steps can be used to ensure that a
chemigation system is operating properly.
1. Prepare a worksheet showing zones, flow rate per zone, area covered or plants
per zone, injection rate, and injection time. This is useful for future reference.
2. The irrigation lines should be completely filled and pressurized before starting
chemigation.
3. Solid set sprinkler systems should, preferably, be operated for 1 h to achieve
good application uniformity. This may not be possible for all chemigation appli-
cations, but the minimum application duration suggested is 15 min.
4. Fertilizers and other agrichemicals (except chlorine) should never be left in the
pipeline when the system is not operating. The general “one-fourth” rule of
thumb is that chemigation should start after one-fourth of the irrigation set time,
injection should occur during the middle two-fourths, and the lines flushed with
clean water during the last one-fourth of an irrigation event. (See the following
section for additional information.) Pesticide and fertilizer injections should be

made in small, frequent doses that fit within scheduled irrigation intervals that
match plant water use to avoid unnecessary leaching. Likewise, excess water
applications for leaching soil salts should never be done when chemicals (except
chlorine) are being injected.
5. The system should be flushed after chemigation has been completed. The irriga-
tion system must be operated long enough to clear all lines of the chemical being
applied. If the irrigation system is shut down before all the chemicals have ex-
ited the lateral lines, extra chemical will be applied at low spots where water
drains through emitters, sprinklers, or system drains. Chemical that was intended
for the end of the lines will then not reach the target area. Some chemicals may
plug nozzles or emitters if not thoroughly flushed from the system. A flushing
time of 30 min should be sufficient for most systems, depending upon system
design, although systems with lengthy and large main lines may require longer
durations for system flushing.
6. A dye test can be conducted to determine the length of time required for the last
of the chemical to exit the final sprinkler or emitter. The amount of flush time
can be reduced by injecting the chemical at the zone control on some systems.
7. Mixing a solution separately for each zone reduces the likelihood of error during
the application process and allows for proper flushing of the irrigation system to
increase application uniformity. If a controller with the capability of program-
ming injections during scheduled irrigations is used, a large batch tank of
chemical can be mixed for all zones. The amount of chemical applied to each
zone will then be controlled by adjusting injection times.
8. If applying chemicals that may damage the crop foliage, the irrigation system
should be operated long enough after chemigation to ensure that the chemical is
washed off.
9. Post-injection treatments may be required to prevent the accumulation of algae,
slimes, or precipitates that may plug microirrigation systems. High carbonate
and/or iron concentrations in some irrigation waters may react with fertilizers
and cause insoluble calcium or iron compounds. Certain bacteria can also fix
iron as a byproduct of metabolism and produce slime or jelly-like material inside
the irrigation lines. Algae growth may also be enhanced by the addition of nutri-
ents in the water. Special maintenance procedures such as chlorination, adding
algaecides and bactericides, and pre-treating water with chelating agents may be
required when performing fertigation with irrigation systems.
10. The acidity of the soil should be monitored, especially when applying ammo-
nium fertilizers through a trickle irrigation system. Acidity will be dependent on
the buffering capacity of the soil. Selection of an appropriate fertilizer source
will reduce acidity problems. Treatment with lime may also be an alternative.
19.5.6.4 Determining depth of chemical application. To be effective, fertilizers
applied through the irrigation system must be stored within the plant’s root zone. Irri-
gation applications that exceed the holding capacity of the soil will cause leaching
beyond the plant’s rooting depth. The specific depth in the soil to which fertilizers or
chemicals are applied can be determined from the application rate of the irrigation
system, the duration of irrigation, soil texture, and soil moisture content before the
chemigation is applied. Table 19.2 presents a guide to determining the approximate
soil moisture content of the soil using the hand feel method. Many other soil monitor-
ing methods and products are available to determine soil moisture content, some of
which are discussed elsewhere in this monograph.
Table 19.2. Soil moisture, appearance, and description
(B.C. Ministry of Agriculture and Lands, 2006).
Available Feel or Appearance of Soil
Water[a] Sand Sandy Loam Loam/Silt Loam Clay Loam/Clay
> 100% Free water ap- Free water is Free water can be Puddles; free
(approach- pears when soil is released with squeezed out. water forms on
ing satura- bounced in hand. kneading. surface.
tion)
100% Upon squeezing, Appears very Appears very Appears very
no free water dark. Upon dark. Upon dark. Upon
appears on soil, squeezing, no squeezing, free squeezing, no
but wet outline of free water ap- water appears on free water ap-
ball is left on pears on soil, but soil, but wet out- pears on soil, but
hand. wet outline of line of ball is left wet outline of
(80)[b] ball is left on on hand. Will ball is left on
hand. Makes ribbon about 1 in. hand. Will rib-
short ribbon. (170)[b] bon about 2 in.
(125)[b] (210)[b]
75% to Tends to stick Quite dark. Dark colored. Dark colored.
100% together slightly, Forms weak ball, Forms a ball, is Easily ribbons
sometimes forms breaks easily. very pliable, out between
a weak ball with Will not slick. slicks readily if fingers, has slick
pressure. (100 to 125)[b] high in clay. feeling.
(70 to 80)[b] (125 to 170)[b] (160 to 210)[b]
50% to Appears to be Fairly dark. Fairly dark. Fairly dark.
75% dry, will not form Tends to form a Forms a ball, Forms a ball,
a ball with pres- ball with pressure somewhat plastic, ribbons out be-
sure. but seldom holds will sometimes tween thumb and
(40 to 70)[b] together. slick slightly with forefinger. (100
(70 to 100)[b] pressure. to 160)[b]
(80 to 125)[b]
25% to Appears to be Light colored. Lightly colored. Slightly dark.
50% dry, will not form Appears to be Somewhat crum- Somewhat pli-
a ball with pres- dry, will not form bly, but holds able, will ball
sure. a ball. together with under pressure.
(20 to 40)[b] (30 to 70)[b] pressure. (40 to (50 to 100)[b]
80)[b]
0% to Dry, loose, sin- Very slightly Slightly colored. Slightly colored.
25% gle-grained, colored. Dry Powdery, dry Hard, baked,
flows through loose, flows sometimes cracked, some-
fingers. through fingers. slightly crusted, times has loose
(0 to 20)[b] (0 to 30)[b] but easily broken crumbs on sur-
down into pow- face.
dery condition. (0 to 50)[b]
(0 to 40)[b]
[a]
Available water is the difference between field capacity and permanent wilting point.
[b]
Numbers in parentheses are available water contents expressed as millimeters of water per meter of soil
depth.
19.6 CALCULATING INJECTION RATES

Calculating the injection rate will help to ensure that the operator can complete the
chemigation in the time frame that has been allotted, determine that all of the chemical
can be applied, and confirm whether the injector will be able to apply the dosage that
is desired. To ensure the best performance the injection rate used should not be at the
low end or high end of the injection system.
The injection time will often dictate the injection rate. In the case of the center
pivot example discussed in Section 19.6.3 the injection time is the time it takes for the
center pivot to make one complete revolution. For traveling gun systems the time it
takes for the gun to cover one set would be the injection time. Drip systems that
change from one zone to another while chemigating will have the injection time set by
the zone operating time.
The following three methods—weight, volume, and injection rate formula—can be
used to determine an injection rate. The type of irrigation and injection system and
type of chemical being applied will often determine the method that is used.
19.6.1 Weight Injection Method
The weight injection method is often used for granular fertilizers and pressure dif-
ferential or Venturi injectors. This method works well for handline, wheel move, and
other stationary sprinkler systems being used to apply fertilizers. The following steps
are required to determine the injection rate for fertilizers using the weight method:
1. Determine the amount of nutrient that has to be applied to a hectare.
2. Using the percent of available nutrient available in the fertilizer calculate the to-
tal amount of fertilizer that has to be applied per hectare. For example, if 50 kg
of N is required per hectare and the fertilizer has a concentration of 15% N then
333 kg of fertilizer must be applied per hectare.
3. Calculate the area that is being covered by each set of the irrigation system. This
calculation can be done by using the sprinkler spacing and determining the
number of sprinklers operating on the set to be treated. It is important to remem-
ber that the same amount of chemical must be applied during each set of the ir-
rigation system in order for the entire field to obtain uniform coverage.
4. Determine the total amount of fertilizer to be applied by multiplying the amount
to be applied per hectare by the area covered during one set of the irrigation sys-
tem. Dissolve this amount of fertilizer into a tank ensuring that the solution will
stay in suspension. All of this solution will be applied during the irrigation set.
5. Determine the length of injection time desired, but make sure that there is suffi-
cient time built in for proper system start up and system flushing. If the fertilizer
is to be applied to a certain depth in the soil check the information above, in
Section 19.5.6.4.
6. The injection rate will be the amount of solution that is in the tank divided by
the injection time selected.
If using a pressure differential injector without a diaphragm, the concentration of fertil-
izer inside the tank will diminish as the solution is displaced by incoming water. Care
should be taken to ensure there is sufficient time to apply all of the fertilizer solution.
19.6.2 Volume Injection Method
The volume injection method is similar to the weight method but is used for a nu-
trient solution rather than granular fertilizer. This method is often used for stationary
sprinkler and drip/trickle irrigation systems. All types of injectors can be used with
this method, but pump and water driven injectors as well as Venturi injectors are more
common. The following steps are required to determine the injection rate for fertilizers
using the weight method:
1. Similar to the steps in the weight method, calculate the amount of nutrient to be
applied to every hectare and the area covered by the irrigation.
2. The density of the solution must be known to determine the amount of fertilizer
that is contained in the solution. For example, a solution with a density of
1.3 kg/L will have 1.3 kg of fertilizer dissolved in each liter of water.
3. The concentration of nutrient in the fertilizer must also be known to determine
the amount of solution that must be applied.
4. The amount of solution that must be applied per hectare can be calculated using
the fertilizer concentration and the solution density. For example, if 50 kg of N
is required per hectare and the solution density is 1.3 kg/L with a nutrient con-
centration of 25%, then the amount of solution will be (50 kg/ha)/1.3 kg/L =
37.6 L /ha. Applying a concentration of N of 0.25 will result in 37.6 L/0.25 =
150 L to be applied per ha.
5. The total amount of solution to be applied can be calculated by using the actual
area covered by the irrigation system multiplied by the amount of solution re-
quired per hectare.
6. The amount of solution to be applied during the selected time will determine the
injection rate.
19.6.3 Injection Rate Formula
For continuously moving irrigation systems, such as a center pivot or a drip system
that will automatically change from one zone to another during chemigation, an injec-
tion rate formula can be used. A well-calibrated injection system will be required for
these types of systems. An electrically driven injection pump or pump-assisted Venturi
system is usually used. Since this method usually results in the entire field being
treated, a large storage tank is required for sufficient chemical.
If all of the parameters are known the injection rate can be determined by:
Qc × A
Ic = (19.1)
C ×T
where Ic = chemical injection rate (L/min)

Qc = quantity of nutrient to be applied to the target area (kg/ha)
A = area that is to be treated and covered by the irrigation system (ha)
C = concentration of injected solution (kg/L) (this is the actual concentration
of the nutrient or chemical in solution to be applied)
T = length of time the injector is operating (min).
For example, a pivot with a wetted radius of 410 m (1350 ft) is to apply 50 kg of
nitrogen per hectare. The travel speed of the end tower is 1.67 m/min (5.5 ft /min). A
urea solution of 23% N and a density of 1.14 kg/L is to be used as the fertilizer source.
Calculate an injection rate as:
1. The quantity of nutrient to be applied (Qc) is 50 kg/ ha of N.
2. The area (A) covered by the pivot can be determined from the wetted radius and
is 53 hectares.
3. The concentration of N in the solution (C) is 1.14 kg/L × 0.23 = 0.26 kg /L of N.

4. The length of time is the time it takes the pivot to make one complete revolution.
This will be the circumference of the pivot divided by the travel speed and there-
fore the injection time is 1336 minutes. Thus, the injection rate is:
50 kg/ha × 53 ha
Ic = = 7.63 L/min
0.26 kg/L × 1336 minutes
REFERENCES
ASABE. 2006. EP409.1: Safety devices for chemigation. St. Joseph, Mich.: ASABE.
Bar-Yosef, B. 1999. Advances in fertigation. Advances in Agronomy 65: 2-77.
Bausch, W. C. 1985. Chemigation: Recommended Safety Devices. Colorado State
Univ. Coop. Ext. Pub. #2.801. Fort Collins, Colo.: Colorado State Univ.
B.C. Ministry of Agriculture and Lands. 2006. Irrigation Scheduling Techniques
Factsheet. British Columbia Ministry of Agriculture and Lands.
Bryan, B. B., and E. L. Thomas. 1958. Distribution of fertilizer materials applied with
sprinkler irrigation systems. Univ. Arkansas Exp. Sta. Res. Bull. 598. Fayetteville,
Ark.: Univ. Arkansas.
Burt, C. M., K. O'Connor, and T. Ruehr. 1995. Fertigation. San Luis Obispo, Calif.:
Irrigation Training and Research Ctr., California Polytechnic State Univ.
California Fertilizer Association, 1995. Methods of applying fertilizer. Chapter 7 in
Western Fertilizer Handbook, 161-188. 8th ed. Sacramento, Calif.: California
Fertilizer Assoc.
Clark, G. A., D. Z. Haman, and F. S. Zazueta. 1998. Injection of chemicals into
irrigation systems: Rates, volumes, and injection periods. IFAS Ext. Bull. 250.
Gainesville, Fla.: Univ. Florida.
Doerge, T. A., R. L. Roth, and B. R. Gardner. 1991. Nitrogen fertilizer management in
Arizona. College of Agriculture Publication 191025. Tucson, Ariz.: Univ. Arizona.
Evans, R. G., and P. M. Waller. 2007. Application of chemical materials. Chapter 8 in
Microirrigation for Crop Production. F. R. Lamm, J. E. Ayers, F. S. Nakayama,
eds. New York, N.Y.: Elsiever.
Granberry, D. M., K. A. Harrison, and W. T. Kelley. 2001. Drip chemigation:
Injecting fertilizer, acid and chlorine. Bull. 1130. Athens, Ga.: Univ. of Georgia
College of Agric. and Environmental Sci. Coop. Ext. Service.
Haman, D. Z., A. G. Smajstrla, and F. S. Zazueta. 1990. Chemical Injection Methods
for Irrigation. Gainesville, Fla.: Univ.Florida, IFAS.
Irrigation Association. 2000. Chemigation. 5th ed. Falls Church, Va.: Irrigation Assoc.
Nakayama, F. S., and D. A. Bucks. 1986. Trickle Irrigation for Crop Production:
Design, Operation, and Management. Amsterdam, The Netherlands: Elsevier.
NASS. 1998. 1998 Farm and Ranch Irrigation Survey. Washington D.C.: National
Agricultural Statistics Service.
Ptacek, L. R. 1986. Subsurface irrigation and the use of chemicals. In Proc.1986
Irrigation Association Annual Conference, 225-234. Falls Church, Va.: Irrigation
Assoc.
Rosen, C. J., and R. Eliason. 1996. Nutrient management for commercial fruit and
vegetable crops in Minnesota. Univ. of Minnesota Ext.Pub. BU-05886. Available
at: www.extension.umn.edu/distribution/cropsystems/DC5886.html.
Rolston, D. E., R. J. Miller, and H. Schulbach. 1986. Chapt. 4.3: Management

principles: Fertigation. In Trickle Irrigation for Crop Production, 317-344. F. S.
Makayama and D. A. Bucks, eds. Elsevier.
Scherer, T., W. Kranz, D. Pfost, H. Werner, J. A Wright, and C. D Yonts. 1999.
Chemigation. Chapter 8 in Sprinkler Irrigation Systems Handbook, 145-165.
MWPS#30. Ames, Iowa: Midwest Plan Service.
Schulze, L. D., and G. Buttermore. 1994. Fertilizer and pesticide containment
guidelines. Publication # G94-1185-A. Lincoln, Nebr.: Univ. of Nebraska Coop.
Ext., and Nebraska Dept. of Environmental Quality. Available at: www.ianr.unl.
edu/pubs/water/g1185.htm.
Smajstrla, A. G., D. S. Harrison, W. J. Becker, F. S. Zazueta, and D. Z. Haman. 1985.
Backflow prevention requirements for Florida irrigation systems. Bull. 217.
Gainesville, Fla.: IFAS, Univ. of Florida.
Smajstrla, A. G., D. Z. Haman, and F. S. Zazueta. 1986. Chemical injection
(chemigation): Methods and calibration. Agric. Eng. Ext. Report 85-22 (revised).
Gainesville, Fla.: IFAS, Univ. Florida.
Solomon, K. H., and D. F. Zoldoske. 1998. Backflow prevention and safety devices
for chemigation. CATI Pub. #981201. Fresno, Calif.: Univ. California.
Threadgill, E. D. 1985. Chemigation via sprinkler irrigation: Current status and future
development. Applied Eng.in Agric. 1: 16-23.
Threadgill, E. D., D. E. Eisenhauer, J. R. Young, and B. Bar-Yosef. 1990.
Chemigation. Chapt. 20 in Management of Farm Irrigation Systems, 749-780. G. J.
Hoffman, T. A. Howell, and K. H. Solomon, eds. St. Joseph, Mich.: ASAE.
van der Gulik, T. 1993. Chemigation guidelines for British Columbia. Abbotsford,
B.C., Canada: British Columbia Ministry of Agriculture and Food.
van der Gulik, T. W., and R. G. Evans. 2006. Chemigation. Chapter 30 in Irrigation.
6th Edition. R. Ryck, ed. Silver Springs, Md.: Irrigation Assoc.
Werner, H. 2002. Chemigation: Calibrating systems for center pivot irrigation. Ext.
Bull. FS 863. Brookings, S.D.: Cooperative Ext. Service, South Dakota State Univ.
Wright, J., F. Bergsrud, and J. Peckham. 1993. Chemigation safety measures. Coop-
erative Extension Pub. FO-06122-GO. Minneapolis, Minn.: Univ. Minnesota.
Wright, J., F. Bergsrud, G. Rehm, C. Rosen, G. Malzer, and B. Montgomery. 1992.
Nitrogen application in irrigation water: Chemigation. Exte. Service Bull. BU-
6118-GO. Minneapolis, Minn.: Univ. Minnesota. Available at: www.extension.
umn.edu/distribution/cropsystems/DC6118.html.
CHAPTER 20
WASTEWATER AND
RECLAIMED WATER
IRRIGATION
Marshall J. McFarland (Agricultural Research and
Extension Center, Stephenville, Texas)
Matt A. Sanderson (USDA-ARS, University Park,
Pennsylvania)
Anne M. S. McFarland (Texas Institute for Applied
Environmental Research, Tarleton State
University, Stephenville, Texas)
Abstract. Irrigation with reclaimed water from municipal wastewater treatment
plants and wastewater from industrial sources is very widespread and socially ac-
ceptable. Increasingly, reclaimed water and wastewater are seen as important re-
sources for water conservation. The design and operation of irrigation systems using
these sources of water should take the constituents of the water into account. These
constituents include nutrients and potential threats to human health and the environ-
ment. Nutrient balances to take the nitrogen and phosphorus content into account may
be advisable. Irrigation with reclaimed water requires the use of separate distribution
systems in urban environments.
Keywords. Human health, Irrigation systems, Nitrogen, Nutrient balance, Re-
claimed water, Phosphorus, Wastewater, Water quality.
20.1 INTRODUCTION
Irrigation with reclaimed water from municipal wastewater treatment plants and
wastewater from municipal and agricultural sources is becoming more widespread and
accepted. The design and operation of irrigation systems utilizing wastewater and re-
claimed water has some very important differences from irrigation systems designed
and operated to apply water from potable or first-use water sources. The differences
arise due to the constituents of the wastewater and reclaimed water, primarily as these
constituents degrade human health and the environment. The design and operation
principles also apply when the irrigation source is a surface water supply with de-
graded quality.
The terms reclaimed water and wastewater are occasionally used interchangeably,
but there are important differences. Reclaimed water has received tertiary treatment,
including filtration and disinfection, from municipal wastewater treatment systems.
The water is free of coliform bacteria and has biochemical or chemical oxygen de-
mand, turbidity, and odor properties comparable to those of potable water. Wastewater
has undergone primary treatment (removal of settleable and floating materials) and
some secondary treatment (stabilization of most organic materials and some reduction
of nutrient levels). The amount of secondary treatment will depend on the design and
operation of the waste treatment system. The range is from very little secondary treat-
ment in some single-stage lagoon systems to near-complete secondary treatment in
facilities with multiple stages, which may include constructed wetlands as a final pol-
ishing stage. Wastewater may be available from agricultural waste treatment systems,
municipal wastewater treatment plants, and food processing plants. Wastewater from
industrial sources is not normally used for irrigation or production of reclaimed water.
Wastewater will have significant organic material, characterized by biochemical or
chemical oxygen demand (BOD/COD), suspended solids (SS), possibly pathogens,
nutrients, and may contain other constituents of concern. Reclaimed water, in general,
may be applied to areas with public access, such as parks, playgrounds, ornamental
plantings, and golf courses. Wastewater, however, may not be applied to areas with
public access and requires land treatment before the runoff or percolated water will
meet water quality standards for primary contact and guidelines for nutrient content.
There are two general purposes for using wastewater and reclaimed water for irri-
gation. The first is to use the water with its nutrient content as a resource. The second
is to use irrigation to treat and/or dispose of wastewater or reclaimed water in a plant-
soil system. Irrigation (slow rate land application) is one of the three recommended
methods of land treatment of municipal wastewater; the other two are rapid infiltration
and overland flow (USEPA, 1981).
Irrigation with reclaimed water and wastewater has many advantages and disadvan-
tages (Heaton, 1981; Lejano et al., 1992) when compared with irrigation with water
from nondegraded sources. Reclamation of wastewater constitutes a new, reliable
source of irrigation water. The nutrients have a positive value for crop, landscape, and
turf production. Irrigation with wastewater and reclaimed water is an economical, cost
effective, environmentally acceptable, and increasingly socially acceptable form of
land treatment.
There are also disadvantages to using wastewater and reclaimed water for irriga-
tion. Irrigation with wastewater and reclaimed water is regulated by national and state
agencies for the protection of public health and the environment. Record keeping and
monitoring may be required. Even the highest-quality reclaimed water has lower qual-
ity than potable water. The water may have elevated levels of dissolved solids, so-
dium, nutrients, heavy metals, organics (BOD), pathogens (bacteria, viruses, protozoa,
and nematodes), trace organics, and toxic anions that require additional considerations
in design, operation, and management. Design and operation of such irrigation sys-
tems are more complex than systems using other water supplies. Irrigation with re-
claimed water requires the use of separate distribution and application systems, with
special precautions to prevent mixing of reclaimed water and potable water supplies.
There are many examples of successful reclaimed water and wastewater irrigation
systems documented in the literature (e.g., Mantovani et al., 2001, surveyed 40 reuse
756 Chapter 20 Wastewater and Reclaimed Water Irrigation
projects in the U.S. and 25 more in ten other countries). The Lubbock, Texas, system
is an excellent example of wastewater irrigation illustrating its advantages and disad-
vantages. Since 1938, the Gray farm east of Lubbock has been irrigated with secon-
dary treated municipal wastewater (George et al., 1987). Before 1982, cotton and
wheat were flood or furrow irrigated with 2 to 4.5 m of wastewater each year. This
overirrigation resulted in groundwater mounding beneath the farm, with a pronounced
degradation of groundwater quality. Groundwater samples collected between 1980 and
1982 showed nitrate-nitrogen concentrations between 5 and 36 mg/L, COD between
27 and 125 mg/L, and total phosphorus between 0.1 and 3.5 mg/L. In 1982, the land
treatment system was expanded in area with the addition of the nearby Hancock farm
for wastewater irrigation, additional storage reservoirs, and center pivot irrigation sys-
tems. After system expansion, the groundwater quality improved between 1982 and
1987 in most of the 27 monitoring wells.
The Water Conserv II project in Florida is another excellent example of the use of
reclaimed water in agriculture (Cross and Jackson, 1993; Parsons et al., 1995). Re-
claimed municipal wastewater from Orlando and Orange County is used to irrigate
over 3000 ha of citrus. In 1995, the system delivered an average of 95,000 m3/d. The
municipal wastewater receives tertiary treatment and disinfection, so it has a quality
equivalent to potable water except for an unknown protozoan content and a low nutri-
ent content. Reclaimed water is delivered at operating pressure to the property bounda-
ries of the citrus at no cost. Reclaimed water in excess of citrus ET is disposed of in
rapid-infiltration basins to recharge the Floridan aquifer. Benefits of the system are
significant. Orlando and Orange County are meeting the mandate of zero discharge of
municipal wastewater into surface waters, the citrus growers benefit from the pressur-
ized free water with its nutrients, the net pumpage from the Floridan aquifer is re-
duced, and public attitude has shifted. The public and the growers now regard re-
claimed water as a resource, rather than as a waste disposal problem (Parsons et al.,
1995). An important aspect of Water Conserv II is that reclaimed water is used for
sprinkler and microirrigation of a human food crop, although the crop is peeled before
consumption.
20.2 CONSTITUENTS AND CHARACTERISTICS
OF WASTEWATER AND RECLAIMED WATER
The quality of reclaimed water and wastewater is impacted by a wide variety of
chemical, biological, and physical constituents, with a broad range of concentrations.
The constituents and concentrations depend on the source waters, the nature of the
industry or activity that generates the wastewater, and the type and degree of treat-
ment, including storage, before use. Examples of concentrations for major constituents
are presented in Table 20.1.
The chemical constituents of concern also include total dissolved solids (salts), dis-
solved gases (e.g., ammonia), ions of elements and compounds (e.g., sodium, chloride,
and nitrate), heavy metals (trace elements), alkalinity, trace organic compounds, and
oil and grease. Dissolved solids concentration is characterized by electrical conductiv-
ity. Sodium concentration may be characterized by the sodium adsorption ratio. Hy-
drogen ion activity (pH) and reduction-oxidation potential (Eh) are important chemical
characterizations. Biological constituents of concern include bacteria, viruses, proto-
zoa, helminths (nematodes, tapeworms, and roundworms), phytoplankton (primarily
Table 20.1. Examples of concentrations of constituents in wastewater and reclaimed water.

Concentrations[b]
Example Type, Location, and Source BOD5 COD NH3-N Total N P TSS
No.[a] of Degraded Water (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L)
Medium-strength,
1 untreated domestic 220 500 25 0 8 220
wastewater
High-strength,
2 untreated domestic 400 1000 50 85 15 350
wastewater
Lubbock, Texas,
3 treatment plant 175 17 24 9
effluent
Tallahassee, Florida,
4 municipal holding 12 65 2 6 7 30
pond effluent
Water Conserv II,
5 <5 0 5 <5
Orange County, Florida
Design dairy concentrated
6 animal feeding operation 1500 350 120 200 58
(CAFO) lagoon
Design beef CAFO
7 1400 180 200
runoff lagoon
Design swine
8 1200 400 220 349 76
CAFO lagoon
Design poultry
9 550 750 100
CAFO lagoon
Erath Co., Texas,
10 6397 248 260 85 2884
Dairy A milking parlor
Erath Co., Texas,
11 1480 161 172 53 839
Dairy A primary lagoon
Erath Co., Texas,
12 Dairy A second 650 117 117 39 480
stage lagoon
Erath Co,. Texas,
13 5467 267 282 55 2333
Dairy B primary lagoon
Piedmont Region, South
14 Carolina, swine 1500 80 175 80
lagoon effluent
Paris, Texas, soup
15 processing wastewater 550 1190 0.7 28 6 795
overland flow system
California
16 950 33 288
wine processing
Ohio
17 1145 3148 91 186 12 960
coffee processing
Washington
18 1440 2190 30 9 690
vegetable processing
United Kingdom
19 1900 30 115 15 640
meat processing
(continued)

[a]
Sources of values for the examples 1 - 19.
1.-2. USEPA, 1981 and Tchobanoglous and Burton, 1991
3. George et al., 1987
4. Overman and Schanze, 1985
5. Cross and Jackson, 1993
6.-9. SCS, 1992
10.-13. Sweeten and Wolfe, 1994
14. Hegg et al., 1984
15. Tedaldi and Loehr, 1992
16. Crites, 1987
17. Loehr et al., 1988
18. Jones et al., 1993
19. Russell et al., 1993
[b]
BOD5 is biological oxygen demand, five-day method; COD is chemical oxygen demand; NH3-N is
ammonia nitrogen; total N is total nitrogen; P is phosphorus, and TSS is total suspended solids.
algae), zooplankton, and organic matter. Organic compounds that require oxygen for
microbial degradation are characterized by biochemical oxygen demand (BOD) or
chemical oxygen demand (COD). The physical characteristics of wastewater are typi-
cally described by suspended solids, settleable solids, color, clarity, odor, and tem-
perature. This section briefly discusses the constituents of interest in wastewater irri-
gation and slow rate land treatment of wastewater. More complete descriptions are
contained in Tchobanoglous and Burton, 1991; Greenberg et al., 1992; USEPA, 1981;
Reed et al., 1988; and USGA, 1994. The standard methods for determination of the
presence, characterizations, and concentration of all constituents are published in
Standard Methods for the Examination of Water and Wastewater (Greenberg et al.,
1992).
The chemical constituents of concern also include total dissolved solids (salts), dis-
solved gases (e.g., ammonia), ions of elements and compounds (e.g., sodium, chloride,
and nitrate), heavy metals (trace elements), alkalinity, trace organic compounds, and
oil and grease. Dissolved solids concentration is characterized by electrical conductiv-
ity. Sodium concentration may be characterized by the sodium adsorption ratio. Hy-
drogen ion activity (pH) and reduction-oxidation potential (Eh) are important chemical
characterizations. Biological constituents of concern include bacteria, viruses, proto-
zoa, helminths (nematodes, tapeworms, and roundworms), phytoplankton (primarily
algae), zooplankton, and organic matter. Organic compounds that require oxygen for
microbial degradation are characterized by biochemical oxygen demand (BOD) or
chemical oxygen demand (COD). The physical characteristics of wastewater are typi-
cally described by suspended solids, settleable solids, color, clarity, odor, and tem-
perature. This section briefly discusses the constituents of interest in wastewater irri-
gation and slow rate land treatment of wastewater. More complete descriptions are
contained in Tchobanoglous and Burton, 1991; Greenberg et al., 1992; USEPA, 1981;
Reed et al., 1988; and USGA, 1994. The standard methods for determination of the
presence, characterizations, and concentration of all constituents are published in
Standard Methods for the Examination of Water and Wastewater (Greenberg et al.,
1992).
20.2.1 Chemical Constituents and Characteristics
Nitrate-nitrogen (NO3-N) and phosphates (PO4 species) are the primary chemical
anions of concern in wastewater and reclaimed water. Nitrate in water consumed by
very young infant humans interferes with the ability of hemoglobin in blood to trans-
port oxygen (NAS, 1978). The name of the disease is methemoglobinemia (blue baby
syndrome). To avoid this disease, the USEPA has established a drinking water stan-
dard of 10 mg NO3-N/L. Nitrates are also toxic to animals when there are high levels
in forage. Nitrates contribute to the eutrophication of surface water. Ammonia and
nitrite are toxic in certain circumstances to plants and animals, including humans.
Phosphorus is usually the limiting nutrient in surface water bodies, so is frequently
associated with eutrophication. When the P levels become high, especially relative to
organic carbon and N, algal blooms will significantly change the biological and
chemical composition of a surface water body.
20.2.1.1 Nitrogen. Nitrogen forms in wastewater may be reported as NO3-N, am-
monia, organic N, and total Kjeldahl nitrogen (TKN). Ammonia (NH3) and the ammo-
nium ion (NH4+) exist in aqueous solutions in an equilibrium that is highly pH de-
pendent. As the pH increases above 7, the relative proportion of ammonia increases
exponentially. Organic N is defined functionally as the organically bound N in the tri-
negative oxidation state, so does not include all organic N compounds. Organic N in-
cludes proteins, peptides, nucleic acids and urea, and various synthetic organic materi-
als. TKN is a term that reflects the method for the determination of organic N and
ammonia in one analytical method. Nitrite, NO2-, is an intermediate oxidation state of
N, both in the oxidation of ammonium to nitrate and in the reduction of nitrate.
20.2.1.2 Phosphorus. Phosphorus (P) compounds are classified as orthophos-
phates, condensed phosphates (the polyphosphates), and organically bound phos-
phates. Phosphates that respond to colorimetric tests without preliminary hydrolysis or
oxidative digestion of the sample are termed reactive P, which is largely a measure of
orthophosphate concentration. Condensed phosphate or acid-hydrolyzable P is the
fraction produced by acid hydrolysis at boiling water temperature. Organically bound
P is the fraction that is converted to orthophosphate only by oxidative destruction of
the organic matter in the sample. Total P is the sum of the reactive, acid-hydrolyzable,
and organic P. The various forms of P occur in solution, in particles or detritus, or in
the bodies of aquatic organisms. Some phosphates are still used in commercial clean-
ing preparations and in treatment of boiler waters. Organic phosphates are formed by
biological processes and are contributed to wastewater in fecal material, plant resi-
dues, and food residues.
20.2.1.3 Carbon. Total carbon in wastewater is composed of inorganic carbon
[carbonate (CO32-), bicarbonate (HCO3-), and dissolved carbon dioxide] and organic
carbon. Total organic carbon (TOC) is used to describe the degradable organic mate-
rial in wastewater. The organic carbon in water and wastewater is composed of a vari-
ety of organic compounds in various oxidation states. BOD and COD are used to char-
acterize the fractions that can be oxidized by biological or chemical processes. The
presence of organic carbon that does not respond to either of these tests makes these
tests unsuitable for the measurement of the total organic carbon. Total organic carbon
is independent of the oxidation state of the organic matter and does not measure other
organically bound elements, such as nitrogen and hydrogen, nor the inorganics that
can contribute to the oxygen demand measured in the BOD and COD tests. The tests
for TOC are based on the breakdown of the organic molecules into single-carbon units
that are converted to a single molecular form (carbon dioxide) that can be measured
quantitatively.
20.2.1.4 Oil and grease. Oil and grease concentrations in wastewater are defined
by the amount extracted by a given solvent (usually trichlorotrifluorethane), as op-
posed to an absolute quantity of a specific substance. Most substances in the oil and
grease group are mineral hydrocarbons and biological lipids. This includes fatty matter
from animal and vegetable sources and hydrocarbons of petroleum origin. The oil and
grease group also includes non-volatilized substances extracted by the solvent from an
acidified sample. These substances include chlorophyll, sulfur compounds, and certain
organic dyes. Oil can reduce germination of seeds and reduce crop yields when ap-
plied at high rates. Grease will reduce soil permeability and have an adverse impact on
soil microorganisms (Ritter, 1987). Oil and grease are most likely to occur in waste-
water from food processing operations, especially meat packing and poultry. Poultry
processing wastewater may have 170 to 230 mg/L oil and grease (Ritter, 1987).
20.2.1.5. Boron. Boron concentrations in municipal wastewater will be higher than
the source water concentrations as a result of the boron content of household and
commercial detergents. Typical boron levels will be from less than 0.1 mg/L to over
4 mg/L. The maximum recommended application level of boron in irrigation is 2.0
mg/L (see Chapter 7). Boron is an anion, so is not held by the soil exchange sites, but
borates are moderately held by soil cation exchange sites. Boron is reported to be more
difficult to leach than other anions, requiring twice as much leaching water (Peacock,
1994). Boron toxicity is most pronounced in tree crops and is manifested by tip and
marginal burning of the leaves. Boron toxicity will not ordinarily show in turfgrasses
because the boron that accumulates is removed by frequent mowing.
20.2.1.6. Other ions. Determination of the common anions such as bromide, chlo-
ride, fluoride, nitrate, phosphate, and sulfate are desirable to characterize a water and
to assess the need for a specific treatment. Ion chromatography is typically used to
determine the anions present in the water sample. The concentrations of sodium, cal-
cium, and magnesium ions are needed for calculation of the sodium adsorption ratio
(SAR). Knowledge of the salinity and SAR of irrigation water from any source is nec-
essary due to the adverse impacts of high SAR on soil properties. Chloride may be
phytotoxic when applied to the foliage of certain plants. Corrosion of dissimilar metals
in irrigation systems is increased when the wastewater has a high salinity, because of
the increased electrical conductivity of the water. Anodized pipe, avoidance of dis-
similar metals, and impressed voltage systems may be needed (Tchobanoglous and
Burton, 1991; Fox and Nuss, 1987).
Detailed information on salinity, SAR, specific anion, and specific cation effects on
plants and soils is contained in other sections of this book (particularly in Chapter 7)
and in Tanji (1990). Salinity in wastewater will originate from source water salinity,
concentrations through recycling (e.g., use of effluent as flush water), concentration
from evaporation of wastewater, and from concentrated animal feeding operations
(CAFOs). Specific ions originate from certain food processing operations. Caustic
peeling processes for peaches and potatoes, for example, may result in high sodium in
the wastewater; sugar processing wastewater may contain high levels of sulfur; sea-
food processing wastewater may contain high nitrate and sodium levels (Ritter 1987);
and high-strength CAFO wastewater will have high salinity, chloride, sodium, ammo-
nium, and orthophosphate concentrations (Sweeten and Wolfe, 1994).
20.2.1.7 Heavy metals. Heavy metals (also known as trace metals or trace ele-
ments) are of concern in wastewaters because they may accumulate in the food chain
to a toxic concentration and may move through the soil into groundwater (USEPA,
1981). Heavy metals are present in low concentrations in municipal wastes that have
minimal industrial components and in food processing wastes. Wastewater from ani-
mal industries will have low concentrations of copper and iron and perhaps other met-
als from feed additives or antibiotics, but these should be of relatively low concentra-
tions and not normally a concern in land application. In wastewater, heavy metals tend
to form metal hydroxides, phosphates, carbonates, and other precipitates that adsorb to
organic solids or precipitate with other constituents (Page and Chang, 1981). These
have low solubility and will be effectively filtered by the soil particles. As cations,
heavy metals are held on the soil cation exchange sites, so do not move readily in the
soil profile (Ritter, 1987; McBride, 1994).
Heavy metal cations, unlike sodium, calcium, magnesium, and potassium cations,
do not readily participate in cation exchange due to their typically low concentrations.
Plant uptake will occur, however, only when the metals are present in soluble or ex-
changeable form. Generally this occurs when the soil cation exchange sites are occu-
pied and when the soil pH is less than 6.5 (USEPA, 1981). At the application rates
usually encountered, soils have an adsorption capacity for heavy metals at least
equivalent to that for phosphorus sorption. When heavy metals are taken up by plants,
most metals will cause visible plant symptoms or plant death before the metal concen-
tration in the foliage is toxic to animals or humans. Cadmium, copper, and molybde-
num are exceptions in that toxic concentrations may build up in plant tissues before
plant symptoms are present.
Cadmium is toxic to animals, humans, and aquatic life forms at concentrations as
low as 15 mg/L and is a carcinogen. Cadmium may enter water through industrial dis-
charges or the deterioration of galvanized pipe, where cadmium is an impurity of the
zinc. Copper, molybdenum, nickel, lead, and zinc are other heavy metals that may be
present in wastewater. Copper is toxic to plants at concentrations of 0.1 to 1.0 mg/L
and can be toxic to ruminants, especially sheep, but is not a health hazard to humans.
Molybdenum is not toxic in soils and water at normal concentrations, but may be toxic
to animals at 10 to 20 mg/L in forage that is low in copper. Nickel is toxic to a number
of plants at concentrations of 0.5 to 1.0 mg/L. Zinc is an essential nutrient, but is also
toxic to many plants in varying concentrations. Phytotoxicities are more common in
acid than in alkaline soils. Lead is a serious toxin that accumulates in human body
tissues. Lead in water may come from dissolution of old lead plumbing or from sol-
dered pipe joints in contact with softened, acidic water. Lead is highly immobile in the
soil, however, and is unlikely to be absorbed by plants (Ritter, 1987). Other heavy
metals may be of concern if industrial wastewater is used for land treatment.
Metal presence and concentrations are determined by atomic absorption and other
methods.
20.2.1.8. Other chemical characteristics. Alkalinity of water refers to its acid-
neutralizing capacity. It is the sum of all titratable bases, so is an aggregate property of
water. Alkalinity is primarily a function of the carbonate, bicarbonate, and hydroxide
content, although it may also include contributions from borates, phosphates, silicates,
or other bases.
The acidity of water refers to its quantitative capacity to react with a strong base to
a designated pH. Acidity is a measure of an aggregate property of water and can be
interpreted only when the chemical composition of the sample is known. Acidity con-
tributes to the corrosiveness and influences chemical reaction rates, chemical speci-
ation, and biological processes. Acidity is determined by titration to an end-point pH.
Oxidation and reduction reactions (redox) mediate the behavior of many chemical
constituents in wastewaters as well as most aquatic compartments of the environment.
The reactivities and mobilities of important elements in biological systems depend
strongly on redox conditions. Reactions of both electrons and protons are pH and Eh
dependent. Like pH, Eh represents an intensity factor, but does not characterize the
capacity for oxidation or reduction.
20.2.2 Biological Constituents and Characteristics
20.2.2.1 Pathogens. Pathogenic organisms are present in all wastewater, and
should be regarded as present in all surface waters. Pathogenic organisms in wastewa-
ter are described in Section 20.4.1.
20.2.2.2 Organic matter. Wastewater that has not received full tertiary treatment
will contain organic matter. The standard indicators of organic matter content are
BOD, COD, and TOC, which have been mentioned previously.
20.2.2.3 Oxygen demand. The BOD and COD determinations are empirical tests
to measure the relative oxygen requirements of wastewaters, effluent, and polluted
waters. Biochemical oxygen demand is the amount of dissolved oxygen required to
meet the metabolic needs of microorganisms that degrade carbonaceous organic mate-
rial in a water sample (Loehr et al., 1979) and is used as an indirect measure of the
biodegradable organic concentration of the sample. The BOD test also includes the
oxygen used to oxidize inorganic matter such as sulfides and ferrous iron. Tcho-
banoglous and Burton (1991) provide details of the laboratory tests, limitations, and
reaction kinetics of BOD determination.
COD is a quantitative measure of the amount of oxygen required for the strong
chemical oxidation of organic carbonaceous material. Ammonia is not oxidized unless
there is a significant concentration of chloride ions. The COD of a waste is generally
higher than the BOD because more compounds can be chemically oxidized than bio-
logically oxidized (Tchobanoglous and Burton, 1991).
20.2.2.4 Trace organics. Trace organic compounds such as pesticides and other
agricultural chemicals may be toxic and may be significant contaminants of wastewa-
ters. Most trace organics are effectively removed by soil treatment. Trace organics
such as herbicides are adsorbed, followed by biodegradation and volatilization. Plant
uptake may occur as a function of solubility, size, concentration, and polarity of the
organic molecule; the organic matter content, pH, and microbial activity of the soil;
and the climate (USEPA, 1981; Hutchins et al., 1985). It is unlikely that crop uptake
of trace organics is sufficient to pose a threat to animals or humans (USEPA, 1981).
20.2.2.5 Microorganisms. As a consequence of the nutrient content of wastewater,
any storage of wastewater and reclaimed water prior to irrigation will result in active
growth of phytoplankton and other microorganisms. The major microorganisms of
concern in irrigation water supplies are those, such as algae, slime bacteria, and colo-
nial protozoa, that affect physical and chemical water quality.
20.2.3 Physical Constituents and Characteristics
Physical properties of water include odor, dissolved oxygen, solids, color, and tur-
bidity. Turbidity is caused by suspended mineral matter, finely divided organic and
inorganic matter, soluble colored organic compounds, phytoplankton, and zooplank-
ton. Turbidity is generally reported in nephelometric turbidity units (NTU). Color is

the true color of the water, after turbidity has been removed.
20.2.3.1 Odor. Odor is characterized by its intensity and hedonic tone (the pleas-
antness or unpleasantness), and character or quality, which are usually described with
words that indicate a likely source (Dravnieks, 1979). Offensiveness is a combination
of intensity and hedonic tone as well as duration and frequency. Odor is a common
problem with wastewater treatment and application. In general, the more complete the
wastewater treatment, the fewer problems there will be with odor. Irrigation of waste-
water in land application systems requires application with only minimal offensive
odor.
20.2.3.2 Dissolved oxygen. Dissolved oxygen (DO) is essential to the biological
and chemical degradation of the carbonaceous and nitrogenous organic material in
wastewater. If the DO is insufficient to meet the oxygen needs of the organic matter,
the receiving water body will be anaerobic except perhaps for a very thin surface
layer. Nearly all aquatic organisms, with the exception of anaerobic bacteria, must
have free oxygen for respiration to survive (Wheaton, 1977).
20.2.3.3 Solids. Solids refer to matter suspended or dissolved in water or wastewa-
ter. Solids may adversely affect wastewater quality in a number of ways. Total solids
are quantified by the amount of material residue left in a vessel after evaporation and
subsequent drying in an oven at a defined temperature. Total suspended solids (TSS)
are the solids removed by a filter (2.0 µm or smaller nominal pore size). Fixed solids
is the term applied to the residue of total suspended and dissolved solids after heating
for a specified time and temperature. The weight loss is a measure of the volatile sol-
ids.
20.2.4 Quality of Selected Wastewater and Reclaimed Water
20.2.4.1 Municipal. Municipal wastewater may be characterized as low, medium,
and high strength (Tchobanoglous and Burton, 1991) based on the concentrations of
the constituents of greatest concern (Table 20.1).
20.2.4.2 Concentrated animal feeding operations (CAFOs). Dairy lagoon efflu-
ent will have a wide variation in strength as a consequence of the source of the wastes,
the dilution by milking parlor washwater and rainfall, and the concentration by evapo-
ration in the open lots and in the waste treatment system. Sources of wastes may in-
clude any combination of milking center wastes, holding shed wastes, scraped feed
lane wastes, and open lot runoff (SCS, 1992; Sweeten and Wolfe, 1994). Water use
per cow per day in the milking center may range from about 20 L to over 550 L if
fresh water is used for flushing of manure (SCS, 1992). Where fresh water is limited,
dairy producers use lagoon effluent for flushing manure from feed lanes. Wastewater
from beef, poultry, and swine CAFOs will have similar variations in strength.
20.2.4.3 Food processing wastes. Wastewater from the wine industry (Table 20.1)
is characterized by low pH, low nutrients, and a relatively high BOD (Crites, 1987).
Raw wastewater from the Campbell Soup Company in Paris, Texas, is characterized
by low nitrogen and phosphorus, but relatively high oil and grease (125 mg/L) and
oxygen demand. Food processing wastewaters are extremely variable in content and
level of treatment.
20.3 NUTRIENTS IN WASTEWATER

AND RECLAIMED WATER
Nitrogen (N) and phosphorus (P) are plant nutrients that may be managed as re-
sources or as potential contaminants that must be removed by soil-plant systems in the
land treatment process. If a reservoir is to be used with the irrigation system, the major
concern with nitrogen and phosphorus as nutrients is their effect on the trophic level of
surface water. Concentrations of 0.3 mg/L inorganic N and 0.02 mg/L inorganic P
have been identified as the critical thresholds for promotion of algae and aquatic
macrophytes in lakes (Loehr et al., 1980; Daniel et al., 1993).
Phosphorus is the single most limiting nutrient for eutrophication, although N may
be the limiting nutrient in waters containing more than 0.03 mg/L dissolved P or when
the total N to total P ratio is less than 15:1 (Daniel et al., 1993). Municipal wastewater
and animal agricultural wastewater normally have total N to total P ratios of much less
than 15:1, typically less than 5:1 (see Table 20.1). In soils, a N:P ratio of about 7:1 is
considered ideal for uptake by plants (Bowmer and Laut, 1992).
The irrigation or environmental engineer should be cognizant of basic nitrogen and
phosphorus forms in wastewater and reclaimed water. The nitrogen form and content
will vary considerably, depending on the treatment before irrigation. Nitrogen and
phosphorus losses will occur in the treatment ponds and storage reservoirs as a func-
tion of water chemistry, biology, and environmental conditions.
Nitrogen in wastewaters is normally in the organic nitrogen and ammonium-
ammonia forms. The nitrate content of anaerobically treated wastewaters will ordinar-
ily be very low. In decomposition of organic wastes, the oxygen demand for the mi-
crobial metabolism is satisfied first from the available dissolved oxygen. Nitrates are
the next source of oxygen, so any nitrate in the system is then reduced to ammonium.
When the wastewater is applied to an aerobic plant-soil environment, bacteria nitrify
the ammonium to nitrite as an intermediate state, then to nitrate. Nitrate is highly mo-
bile in soil water, but ammonium is not. Ammonium is adsorbed to negative charge
sites on clay minerals, with an adsorption capacity sufficient to retain all ammonium
from slow rate (irrigation) applications (Broadbent and Reisenaur, 1985).
Biological transformations of nitrogen are immobilization, mineralization, nitrifica-
tion, denitrification, and nitrogen fixation by plants (Keeney, 1983). Immobilization
refers to the conversion of inorganic forms to organic compounds. Mineralization is
the conversion of organic nitrogen to ammonium. Plant uptake of inorganic nitrogen
(nitrate and ammonium) and incorporation into organic plant components is the pri-
mary immobilization process.
The phosphorus content of wastewater, like nitrogen, will also vary significantly as
a function of the source and type of treatment. The phosphorus content of the waste-
water to be used for irrigation should be determined.
Phosphorus exists in several forms in water and the soil-plant environment. Particu-
late P is composed of organic material, crystalline minerals, amorphous precipitates,
and reactive phosphorus (primarily orthophosphates) sorbed to soil particles, primarily
clays. It should be noted that the orthophosphates are anions, so they do not participate
in cation exchange and are not adsorbed to the clay mineral sites with negative charge
potentials. The phosphorus minerals and precipitates have varying solubility and are
generally reaction products with calcium, iron, and aluminum. Soluble P is composed
of orthophosphates (PO43-, HPO42-, and H2PO4-), polyphosphate (P2O7), and soluble
organic compounds. The distinction between particulate and soluble is somewhat arbi-
trary, because the separation is based on the diameter of the filter pores (0.45 µm ).
Phosphorus soil and aquatic chemistry is complex; additional information is contained
in Diaz et al. (1994), McBride (1994), and Stumm and Morgan (1996).
In a soil solution, an equilibrium is established between the phosphorus in solution
and the labile phosphorus on soil particle surfaces. As plants remove phosphate from
the soil water, the equilibrium is restored. The soluble P, especially orthophosphate,
will move with soil water and will equilibrate with labile phosphorus. Conventional
wisdom is that a soil will not have a leachate with a significant P concentration. In
general, this is true, especially with a tilled, row-crop soil with a moderate to high P-
sorption capacity. Phosphorus profiles in soils receiving animal wastewater or solid
wastes typically show an exponential decrease in concentration with depth (Sharpley
et al., 1984; Allhands et al., 1995). In addition to the conventional soil fertility tests,
bioavailability tests may be used to provide additional management information
(Sharpley and Withers, 1994).
20.4 HEALTH CONCERNS
Pathogenic microorganisms, including bacteria, viruses, and protozoa, are present
in human and animal wastes and the waters containing these wastes (NRC, 1998).
There is no doubt that irrigation of food crops with sewage or treated wastewater con-
taminated with sewage results in increased disease occurrence (Rosas et al., 1984;
Rose, 1986; Ali, 1987) or that the potential exists for disease transmission from irriga-
tion of public use areas and food crops with wastewater. The National Research Coun-
cil (NRC, 1996) concluded that food crops irrigated with reclaimed water do not pre-
sent a greater risk to consumers than do crops irrigated from conventional sources.
The essential elements to protect human health are adherence to state standards for
reclaimed water, site restrictions, and good system reliability.
Pathogens could be transmitted to humans by ingestion of the wastewater, inhala-
tion of aerosols from the wastewater, self-inoculation or contamination of foods or
liquids after direct contact, contamination of food or animal feed, and contamination
of potable water supplies. For a disease to occur from wastewater irrigation, the
pathogen would have to survive any treatment process, persist in the environment, and
be present in sufficient numbers to cause the disease in a susceptible individual (Rose,
1986). Health and safety are the overriding concerns when treated wastewater is ap-
plied to public areas, food crops, and forage systems. Although disease transmission
by irrigation of treated wastewater (not contaminated with raw sewage) has not been
documented as a problem (Rose, 1986), contaminated water from wastewater treat-
ment plants and/or animal operations has been implicated in disease outbreaks (e.g.,
MacKenzie et al., 1994).
For irrigation of public areas with municipal wastewater, state departments of
health generally do not accept any risk above that of irrigation with potable water. The
biological quality standard is water that is essentially bacteria and virus free (Asano et
al., 1992). Most major municipal reclaimed wastewater projects produce water with
residual chlorine of 1 mg/L. Examples are Colorado Springs, Colorado (Schwebach et
al., 1988), Orlando, Florida (Cross and Jackson, 1993), Monterey, California (Sheikh
et al., 1990), and San Diego, California (Shamloufard et al., 1995). Irrigation of agri-
cultural crops with municipal wastewater is done with water that has at least secondary
treatment. The Muskegon County, Michigan (Brenner et al., 1988), and Lubbock,
Texas (Moore et al., 1988), projects are examples.
Wastewater irrigation should be considered to have the potential for pathogen
transmission. The information on biological quality of the water or associated epide-
miology is not well understood enough to develop standards. There is no evidence that
regulations or standards are needed, but there is sufficient information to indicate that
considerable caution and adherence to recommended practices are essential to protect
human health.
Quality of wastewater applied to forage crops may be a concern in the food proc-
essing industry. Applying animal wastes to forage crops to be ensiled may result in the
introduction of several bacteria species such as Bacillus and Clostridium (Ostling and
Lindgren, 1991). The spores of Clostridium will survive in the silage. Therefore, Clos-
tridium may have an adverse effect on the ensiling process and result in poor silage
and reduced animal performance, whereas Bacillus may encourage aerobic deteriora-
tion of silage when removed from the silo and fed. If the spores enter the milk through
contamination and if the milk is used in making cheese, “blowing” (production of bu-
tyric acid and gas) of hard cheeses may result (Jonsson, 1991).
E. coli is the most frequent species found in animal manure, but populations decline
drastically one week after waste application to growing forage (Rammer et al., 1994).
Animal wastes may also be a source of Listeria monocytogenes (Husu et al., 1990) and
consumption of raw vegetables contaminated with Listeria from untreated animal
wastes may cause listeriosis, a potentially fatal disease.
20.4.1 Biological Organisms Found in Wastewater
The important biological constituents of wastewater are algae, bacteria, viruses,
protozoa, and helminths. Virtually all wastewater without tertiary treatment contains
pathogenic organisms that can produce disease in humans. Pathogenic protozoan oo-
cysts and cysts may be present even in treated wastewater that has 1 mg/L residual
chlorine. Additional information is contained in Rose (1986) and Shuval et al. (1986).
Information on algae is contained in Goldman et al. (1972), Reddy (1981), Vymazal
(1995), and Stumm and Morgan (1996).
20.4.1.1 Bacteria. The most significant pathogenic bacteria cause diseases such as
dysentery (Shigella), typhoid, and gastroenteritis, which is generally caused by Salmo-
nella, Campylobacter, and certain types of E. coli (Rose, 1986). These species inhabit
the intestinal tracts of humans and other mammals. Salmonella species are especially
common in cattle feces (Thelin and Gifford, 1983) and human waste (Rose, 1986).
The bacterial quality of wastewater is assessed with indicator bacteria that may not
be pathogenic. Indicator bacteria are used because they are present in wastewater in
greater numbers than pathogenic bacteria, they are easier to isolate, and they are safer
to handle (Thelin and Gifford, 1983). If indicator bacteria are present in a water sam-
ple, the presence of pathogenic bacteria is assumed. Indicators of bacterial quality of
water are total coliform (TC), fecal coliform (FC), and fecal streptococci (FS). Both
FC and FS are present in the feces of humans, and domestic livestock and poultry.
Several procedures for determination of the TC, FC, and FS bacteria concentrations
have been published (e.g., Niemi and Niemi, 1991; Sherer et al., 1992).
20.4.1.2 Viruses. Enteroviruses that can be transmitted in water include poliovirus,
echo, hepatitis A, and coxsackie (Rose, 1986). Norwalk, rotavirus, and other viruses
capable of producing infection may also be present in wastewater (Rose, 1986; Ward
et al., 1989). Rose (1986) states that the presence of viruses in wastewater causes con-
cern because there is a dearth of information on the occurrence and significance of
viruses; detection methods may recover less than 50% of the viruses in a sample; the
survival of enteric viruses in the environment is not well known and viruses may not
be removed efficiently in treatment systems; small numbers of viruses could cause an
infection; and the bacterial indicator system is not always reflective of the presence of
viruses.
Secondary and tertiary treatment of wastewater removes viruses with an efficiency
of 90% to 99%, but some viruses will be present following disinfection (Rose, 1986).
It is suspected that viruses, when aerosolized from spray irrigation, may be more resis-
tant to environmental conditions than bacteria (Rose, 1986). Wastewater treatment
ponds and storage reservoirs have a marked impact on virus survivability (Ward and
Irving, 1987). Reductions of one to two logs in virus counts were reported in the Lub-
bock study (Moore et al., 1988; Ward et al., 1989). Under favorable conditions, vi-
ruses can survive on vegetable surfaces for over two months. Low temperatures, high
humidity, presence of organic matter, and shading from direct sunlight protect against
desiccation and favor virus persistence (Ward and Irving, 1987). Viruses die rapidly in
a warm and dry environment. Virus survivability in the process of aerosolization and
in aerosol form is thought to be low (Schwebach et al., 1988).
20.4.1.3 Helminths. Parasitic nematodes, roundworms, hookworms, tapeworms,
and whipworms are endemic in wastewater in areas of the world with poor hygiene
(Rose, 1986). The ova from these parasitic worms aid in their survival in wastewater.
The ova are generally resistant to chlorine, but settle out during sedimentation. High
rate algal ponds (shallow treatment ponds specifically designed to have a high algal
population for daytime oxygen production to reduce oxygen demand of the wastewa-
ter) and waste stabilization ponds are fairly effective in the removal of helminth ova
(Ayres et al., 1992). Ascaris lumbricoides (roundworm) ova are the most resistant of
these enteric pathogens, so are often used as indicators of parasites (Rose, 1986; El
Hamouri et al., 1994). The World Health Organization (1989, cited in Ayres et al.,
1992) recommends a maximum of one human intestinal nematode egg per liter for
irrigation.
20.4.1.4 Protozoa. Giardia lamblia and G. muris are parasitic protozoa that cause
diarrhea in humans who ingest their cysts. Giardia cysts should be presumed to be in
all wastewater and surface waters used for irrigation (LeChavallier et al., 1991). Pre-
cautions to avoid infection are the same as for Cryptosporidium, which is a more sig-
nificant threat in wastewater irrigation. C. parvum is the parasitic protozoa that was
responsible for the outbreak of severe diarrhea and related symptoms (cryptosporidio-
sis) in Milwaukee, Wisconsin, in the spring of 1993 (MacKenzie et al., 1994).
The prevalence of C. parvum oocysts in wastewaters from all sources, especially
those from dairy and dairy calf industries, indicates the need for caution and protective
measures when using wastewater for irrigation. A need for reasonable precautions is
also indicated for irrigation with surface water from agricultural and urban watersheds
(Ongerth and Stibbs, 1987; Hansen and Ongerth, 1991; LeChevallier et al., 1991).
20.4.2. Health Aspects of Irrigation with Wastewater
A five-year study of the Monterey, California, wastewater reclamation project indi-
cated that the use of tertiary treated wastewater (reclaimed water) for food crop irriga-
tion is safe and acceptable (Sheikh et al., 1990). The food crops were artichokes, cel-
ery, broccoli, lettuce, and cauliflower.
In Colorado Springs, Colorado, tertiary treated municipal wastewater is used for ir-
rigation of public park areas (Schwebach et al., 1988; Durand and Schwebach, 1989).
The tertiary treatment consists of filtration and chlorination of effluent from the
wastewater treatment plant. The city also uses urban mountain runoff water, without
treatment, for irrigation of public park areas. An epidemiological study was conducted
to examine the gastrointestinal illness levels associated with irrigation with different
water qualities. The conclusion of the Colorado Springs epidemiological study was
that there was no difference in self-reported gastrointestinal illness rates between park
visitors in parks irrigated with potable water compared with those in parks irrigated
with treated wastewater. “Wet grass conditions” from any source of water were asso-
ciated with a statistically significant increase in self-reported gastrointestinal illness.
The correlation was even stronger when the irrigation water FC or FS counts were
above 500/100 mL. The authors hypothesized that endotoxins from living or dead en-
terobacteria are ubiquitous to the grass and are transferred to people from the wet
grass.
Several epidemiological studies of health risk as a result of wastewater irrigation
have been conducted in Israel, where 70% of the urban wastewater flow of the nation
is used as a water resource for irrigation (Shuval et al., 1989). Most of the wastewater
receives primary treatment and secondary treatment in stabilization ponds. Shuval et
al. (1989) conducted a study of irrigation workers, their families, and the general
population in 20 agricultural settlements. The quality of the irrigation wastewater was
poor, with coliform counts of 104 to 105/100 mL. The authors concluded that enteric
disease levels were no higher in wastewater workers, their families, and people who
may have been in contact with irrigation aerosols than in the general population. The
studies did note that low levels of bacteria and viruses of wastewater origin were
found at distances up to 730 m downwind from an irrigated field.
The Lubbock infection surveillance survey (LISS), an epidemiological study of the
health risk associated with wastewater irrigation, was conducted from 1980 to 1983
(Moore et al., 1988; Ward et al., 1989). Municipal wastewater from two trickling-filter
plants from the city of Lubbock, Texas, was applied to agricultural crops. The waste-
water was piped to a 1153-ha farm for irrigation, primarily through 22 center pivot
systems. In 1982, the wastewater was pumped directly to the pivots. In 1983, playas
(shallow, natural water-filled depressions, usually with no external drainage) were
used as wastewater storage reservoirs prior to irrigation. In the spring of 1982, the
average concentrations for FC were 4.3 × 106 cfu/100 mL (cfu is colony-forming
units), whereas the 1983 spring concentrations were 5.2 × 103 cfu/100 mL. All levels
exceeded the EPA guidelines of 1000 FC/100 mL for this irrigation. The enterovirus
concentrations were 40 pfu/100 mL for spring 1982 and 2 pfu/100 mL for spring 1983
(pfu is plaque-forming units). Ward et al. (1989) concluded that exposure to wastewa-
ter aerosols in the LISS was not correlated with an increase in disease incidence, but
there was an indication that new viral infections occurred more frequently in people
with high exposure than in people with moderate or low exposure. There was no evi-
dence that wastewater spray irrigation caused increased rotavirus infection, but spread
of other viral infections may have occurred (Ward et al., 1989).
20.5 CROPS SUITABLE FOR IRRIGATION

WITH WASTEWATER AND RECLAIMED WATER
Consideration must be given to the crop when irrigating with wastewater or re-
claimed water. If disposal of the water and/or nitrogen content is the primary consid-
eration, crop selection for land application sites may be based either on water disposal
or utilization or nitrogen removal. For water use, maximum ET rates throughout the
growing season are desired. The crop may or may not be harvested and removed. For
N removal, crop selection should be based on the total N removed, which is a function
of the crop yield removed from the field and the N content of that yield. Land applica-
tion systems are not usually designed based on phosphorus loading as a limiting fac-
tor. The P uptake by most cropping systems will be low relative to the P input from
wastewater, especially if the system is designed to remove N from CAFO systems. If
P is applied at rates in excess of the crop uptake and removal, the P sorption capacity
of the soil profile should be determined and used to calculate the number of years the
application site may be used for wastewater application.
Characteristics of crops selected for land treatment sites are (USEPA, 1981; George
et al., 1985): high nutrient uptake and removal; high crop ET; tolerance to high soil
moisture conditions; revenue or local crop use potential; suitable for equipment, labor,
and management system of producer; adaptation to climate and soil; satisfaction of
irrigation water quality requirements; restrictions based on use of crops for human
consumption; and goodness of fit into overall operation of producer.
Virtually any locally adapted crop, including horticultural crops, may be produced
with reclaimed water or wastewater without adversely affecting the quality or quantity
of the crop. Kirkham (1986) summarized problems associated with the use of waste-
water for vegetable crops. She concluded that apparently few problems are associated
with secondary effluent applied to vegetables. If nutrient removal is a primary objec-
tive of the land treatment system, forage crops are normally selected. Most forage
crops have a high ET throughout the growing season, have high nutrient requirements,
are tolerant of poor drainage and/or periodic flooding, are tolerant of high salinity, and
can be interseeded or interplanted with a mix of warm-season and cool-season species
to extend the growing season. Perennial forage crops have a contiguous root system,
whereas an annual row crop will have areas of soil without plant root systems for ni-
trogen uptake for significant periods in the spring. Forages also have low cultural re-
quirements (e.g., pest management, cultivation and tillage, and critical timing of
phenologically based cultural practices). In addition, forages normally have a local
market or can be used in the animal agriculture operation that produces the wastewa-
ter. Barnes et al. (1995) present climatic adaptability, yields, response to fertilization,
and other information for various forages.
Row crops may have high yields, but may have two disadvantages. The period in
the spring before the crop begins to grow and the period after harvest are known to be
times with maximum potential for nitrate leaching. Secondly, if only a portion of the
crop is removed at harvest (e.g., grain), most nutrients remain in the field.
Turfgrasses that will have direct human contact may be sprinkler irrigated with re-
claimed water, but not wastewater, due to wastewater quality not meeting primary
contact standards. Consequently, turfgrass systems in urban and residential settings are
used more for beneficial reuse of reclaimed water than for water treatment and water
disposal.
Woodlands and forests are acceptable for wastewater land treatment. The advan-
tages are minimal site maintenance, high organic content of surface litter and soil, and
a stable soil profile. Older forests without understory are not suitable for nutrient re-
moval, however (Burton and King, 1981).
20.6 DESIGN OF NITROGEN LOADING RATE
For almost all systems for application of reclaimed water or wastewater, either the
hydraulic loading rate or the N loading rate will be the limiting factor. If the wastewa-
ter or reclaimed water has a relatively low N content, the determination of the total
land area to irrigate and the design of the system are in accordance with the other
chapters of this book. The design procedure will be altered slightly, however, because
the starting point will be the volume or flow rate of water available for irrigation. If
the total N content of the wastewater is high—more than about 10 mg/L, which is the
case for most agricultural wastewater—the N loading rate may be the limiting factor in
the system design. Both the hydraulic loading rate and the N loading rate should be
compared to determine which is limiting. The hydraulic loading rate will be ET based,
in accordance with other chapters. The N loading rate may be calculated with the use
of a simplified annual N balance.
20.6.1 Design Procedure
The design procedure is to start with the N content and the volume or flow rate of
the wastewater. Calculation of the total N balance is required to determine the area of
land required for irrigation for land treatment. Sufficient land must be available for
irrigation and/or land treatment without surface runoff or percolation containing more
than 10 mg/L NO3-N. The design procedure is:
1. Estimate the mean annual total N content (mg/L) of the wastewater, including
captured runoff, as a function of the characteristics of the CAFO or industry and
the pretreatment of the wastewater.
2. Estimate the annual total volume of the wastewater. This presumes that the
wastewater storage reservoirs are properly designed for storm water runoff and
winter storage requirements.
3. Determine the annual N (product of the values of steps 1 and 2, in kg) that will
be applied in the irrigations.
4. Determine the grazing or cropping systems that will be used and the preliminary
estimate of the total land area to be irrigated. Estimate the crop dry matter
(Mg/ha) that will be removed in a “normal” year as a function of water and nu-
trient availability. Estimate the total N (kg/ha) removed in the crop, as a per-
centage of the crop dry matter. Convert the annual N available in the irrigation
water to kg/ha. Convert the volume of wastewater to mm depth for a unit land
area.
5. From other chapters (Chapters 5, 7, and 8) in this book, determine the irrigation
water requirements from the consumptive water use, the precipitation, the irriga-
tion system efficiency and uniformity, and any leaching requirements. Express
the irrigation water requirements as depth (mm).
6. Compare the irrigation wastewater available (mm depth for a unit land area)
with the consumptive irrigation water requirement (mm). If the available water
is much less than the consumptive requirement, either another water source must
be developed for irrigation or the crop yields (and N removal) may need to be
revised downward to take decreased water availability into account. If the avail-
able water is much more than the consumptive requirement, then N is not the
limiting factor and the system should be designed for hydraulic loading instead
of nutrient loading. Additional fertilizer N may be needed.
7. Use a N balance method, such as the method presented in the next section, to
compare the kg N/ha of the applied wastewater with the kg N/ha removed dur-
ing the land treatment. If the N removed in the land treatment system is less than
the N to be applied, either the N to be applied has to be reduced (additional pre-
treatment) or the N removed has to be increased (e.g., increase denitrification
losses or add more land).
8. Finalize the land area to be irrigated and proceed with system hardware selection
and design.
20.6.2. Nitrogen Balance
A nitrogen balance calculation is needed to determine the total land area required
for irrigation with the wastewater without exceeding the agronomic rates for N re-
moval. The total N balance is normally calculated for a unit land area on an annual
basis. Procedures have been reported for the calculation of the water and nutrient bal-
ances of the application area (e.g., Loehr et al., 1979). Thompson et al. (1997) provide
a listing of the computer programs that calculate manure application rates. Some of
these will be adaptable to liquid application. The N input portion, Ni (kg/ha), of the N
balance is:
Ni = 0.01 × [(I × Ic) + (P × Pc)] + F + O + S (20.1)
where 0.01 = constant of proportionality, from 104 L/(ha mm) × 10-6 kg/mg
I = effective irrigation amount reaching the plant and land surfaces, mm
P = effective precipitation amount reaching the plant and land surfaces, mm
Ic = total N concentration in the irrigation, mg/L
Pc = total N concentration in the precipitation, mg/L
F = total N content of fertilizer added, kg/ha
O = N from other sources, such as upward movement from groundwater to root
zone and contribution from legumes, kg/ha
S = mineralization of immobilized N from previous years, kg/ha.
The nitrogen loss or removal portion, No (kg/ha), is:
No = (0.01 × I × Ic)(Vf + If + Df) + 0.01 × [(G × Gc) + (R × Rc)] + Cu (20.2)
where Vf = fractional loss of the input N losses due to volatilization
If = fractional loss of the input N losses due to immobilization
Df = fractional loss of the input N losses due to denitrification
G = percolation from the root zone to groundwater, mm
Gc = concentration of N in the percolate, mg/L
R = runoff from the application site, mm
Rc = concentration of N in the precipitation runoff, mg/L
Cu = crop N removal, kg/ha.
If No is equal to or less than Ni, N is the limiting factor for the loading rate. Equat-
ing No and Ni, and solving for I results in:
I = [Cu − F − O − S + 0.01 × [(G × Gc) + (R × Rc)
− (P × Pc)]/[0.01 × I c × (1 − Vf − If − Df)] (20.3)
The maximum allowable value of NO3-N in the percolation to groundwater, Gc, and
the surface runoff of precipitation, Rc, is 10 mg NO3-N/L.
The percolation to groundwater will not be zero for most land application systems,
but is assumed to be zero in design to be on the conservative side. In some areas, the
spring thaw will result in a net discharge of soil water to groundwater. In some cli-
mates, excessive precipitation will saturate the soil profile and result in a net discharge
of soil water to groundwater. Site-specific evaluation of G and Gc should be based on
measurements or calibrated simulation model results.
Surface runoff of precipitation may be estimated from models such as the rational
method and the SCS Curve Number, among others. Nitrogen losses of 3% to 7% of
the total N application were reported from Georgia for dairy wastes applied with a
center pivot (Hubbard et al., 1987). At these rates, 25 to 30 kg N/ha would be re-
moved. For a conservative design, N removal in surface runoff may be neglected. This
figure is usually less than 10% of the N removal in crop harvest.
If nitrate is the limiting factor for land application of wastewater, N fertilizer will
not be added. However, if P loading is a concern, additional N may be required to bal-
ance the fertility and increase yields. The contribution of N from legumes should be
included in the calculation.
The precipitation climatology for the site can be obtained from a variety of sources.
USEPA (1981) recommends that the design monthly precipitation be set to the 5-yr
return period frequency. Alternatively, if the annual precipitation is used, the 10-yr
return period should be used, distributed by month based on the ratios of average
monthly to average annual precipitation. The nitrogen content of precipitation may be
assumed to be 0.5 mg/L (Loehr et al., 1979).
The fractions of applied total N that are volatilized, immobilized, and denitrified
are functions of the pretreatment processes, irrigation system, soil-plant environment,
and land treatment system management. In the absence of site-specific values, some
approximate values may be used in design. For high-strength, high-pH wastewater
from CAFOs or processing plants, volatilization losses in sprinkler irrigation may ap-
proach 20% (Broadbent and Reisenaur, 1985). Volatilization losses will increase with
increasing concentrations of ammonia in the wastewater. For low-strength, neutral-pH
reclaimed wastewater, the volatilization losses will be near zero, but for reclaimed
wastewater with low N content, hydraulic loading will normally be the limiting factor
in design. For well-aerated, coarse-textured soils, denitrification may be neglected.
Periodic waterlogging, fine-textured soils, and frequent irrigation promote the condi-
tions that favor denitrification. For these conditions, denitrification losses may be as
high as 40% (Broadbent and Reisenaur, 1985). Denitrification losses in the range of
15% to 25% should be used in design. Microbial immobilization losses of nitrogen
will be typically in the 22 to 45 kg/ha range, but may be as high as 55 kg/ha (Broad-
bent and Reisenaur, 1985). For an annual irrigation application of 333 mm irrigation
with a total N content of 300 mg/L, the microbial immobilization is only about 5%.
Consequently, the immobilization fraction, If, may be neglected. The USEPA (1981)
guidelines are 15% to 25% for denitrification and volatilization combined and 0% net
storage in soil (immobilization and mineralization).
Nitrogen that is applied in organic form (slurry, dry manure, or suspended solids)
should be taken into account in the design loading rate. About 25% to 35% of the or-
ganic N will be released in the first cropping season. For long-term application of sol-
ids, the total N content of the solids should be added to the N balance.
Net mineralization of the soil organic N is about 2% to 4% per year of the total soil
N, which may be 1000 to 5000 kg/ha (Keeney, 1983). The mineralization term for a
high-fertility, high-productivity land application site may be as much as 200 kg/ha.
Ideally, the mineralization term could be indexed to the crop uptake term. If a har-
vested forage crop yield of 10 Mg/ha has a total N content of 2%, 200 kg total N will
be removed in the aboveground biomass that is harvested. As a generality, there will
be a corresponding 200 kg total N immobilized in organic form in the belowground
biomass. Most of this will be subsequently mineralized to inorganic N in succeeding
years, and could be incorporated into the design N loading rate calculation. If the net
storage is assumed to be 0% of the applied N for a steady state situation, the design
will be conservative (no net sink to soil organic material).
With proper application rates, annual crop uptake and harvest removal of N will be
from 50% to above 90% of applied N in the year of application. The remainder is typi-
cally in non-harvested, organically bound N in plant roots and other plant residues,
with contributions from other N losses. For design of loading rates at high N applica-
tions, a value between 55% and 75% N crop uptake and harvest removal should be
used. The value would be a function of the crop and the soil. For forage crops, crop N
uptake may be generalized as a percentage of the harvested dry matter, especially at
higher N application rates. The range for design is 1% to 3% total N in harvested dry
matter for forages.
For most pure perennial grass pastures on the acid soils of the southeastern U.S.,
the desired fertility N:P:K ratios are 4:1:2 (Chamblee and Spooner, 1985). If nitrogen
is lost from the wastewater treatment and application systems, additional N may be
needed to increase the uptake of P in the forages. This may be provided in supplemen-
tal fertilization or from the use of interplanted or double-cropped legumes.
The nitrogen balance equation simplifies, with the assumptions of the preceding
paragraphs including no fertilizer nitrogen applied, to:
I = [Cu − S − 0.01 × (P × 0.5)]/0.01 × Ic × 0.75 (20.4)
For a sample calculation, assume the input data for irrigation from a dairy CAFO la-
goon, on an annual basis: The wastewater has a total N content of 300 mg/L. The ef-
fective precipitation is 1270 mm. The crop in the southeast U.S. is Coastal bermuda-
grass, with interplanted winter wheat, harvested as forage and removed. The total dry
matter yield is 15 Mg/ha, with an N content of 1.5%, so the N removed is 225 kg/ha.
The annual mineralized N (S) is 50 kg/ha. Solving for I shows the effective irrigation
amount should not exceed 75 mm. If more than 75 mm irrigation water is applied,
more N will be applied than can be removed by the crops. The land area required for
the wastewater irrigation should be determined by the total volume of the wastewater
that will be available on an annual basis.
20.6.3 Phosphorus Balance
Due to the complexity of the P chemistry of soils, similar models for P have not
been developed. If the P loading is the limiting factor, the general requirement is that
the total P loading should not exceed the P agronomic uptake and removal rate. The P
content of the removed crops is used to determine the total land area required for the
application of the wastewater.
20.7 IRRIGATION SYSTEMS USING WASTEWATER

Surface irrigation was initially used for land treatment of wastewater and irrigation
with reclaimed water, because it was a method with low cost, low technology, and low
capital investment. The major problems with surface irrigation are deep percolation of
water containing nitrates at the head of the field and the necessity to contain and recy-
cle the tailwater runoff from the land irrigation site (Reed and Crites, 1984). When the
local conditions are favorable for gravity flow, the system may be designed as an over-
land flow treatment system (USEPA, 1981; Tedaldi and Loehr, 1992). Sprinkler irri-
gation systems, however, are the dominant systems for applying wastewater and re-
claimed water, with microirrigation systems gaining in popularity.
20.7.1 Sprinkler and Microirrigation Systems for Reclaimed Water
Reclaimed water may be used for any type of sprinkler or microirrigation. The use
of reclaimed water does not change the design, except for the additional safety fea-
tures that must be added to protect a potable water supply. These are discussed sepa-
rately in this chapter. For urban and turf uses, solid set systems with underground
mains and laterals are standard. For tree, forage, and crop irrigation, the quality of the
reclaimed water may exceed the quality of the nearby surface and groundwater as a
result of the filtration and disinfection. Again, design of the system is the same as for
other water sources.
Sprinkler and microirrigation systems have the following advantages when used
for reclaimed and wastewater irrigation:
The applications of water and nutrients are relatively uniform.
Small irrigation applications are practical.
With good management, there will be minimal runoff, which precludes the ne-
cessity for a runoff collection and reuse system.
The system may be designed to minimize worker contact with the wastewater
and any pathogens it may contain.
The system may be automated to apply water to minimize public exposure to the
water or irrigation water aerosols.
The system may be automated and simplified to facilitate irrigation management
by operators whose primary business is not crop production or irrigation. Exam-
ples are center pivots using dairy CAFO wastewater and subsurface drip systems
for application of domestic wastewater from subsurface flow wetland systems.
Sprinkler application of pathogens in the wastewater is effective in promoting
rapid die-off, especially of viruses.
Sprinkler irrigation will result in N losses due to ammonia volatilization, which
will reduce the N loading and the potential for nitrate leaching.
Record keeping is facilitated with the use of center pivots and other automated
systems.
With microirrigation, reclaimed water and wastewater may be applied below the
ground surface to minimize human contact and runoff potential of applied con-
stituents.
Disadvantages of using wastewater, but not reclaimed water, with sprinkler and mi-
croirrigation systems are:
Particulate matter in the water may accumulate in mains, laterals, nozzles, emit-
ters, and other system components. Anaerobic conditions, corrosive conditions,
odors, and particulate aggregates that reduce pipe capacity and plugged nozzles
and orifices may result.
Aerosolization of the wastewater in sprinkler or sprayer irrigation may transport
viruses and bacteria downwind to a distance of several hundred meters unless
controlled by operating pressure and nozzle placement and design.
Disinfection of wastewater is not an option for control of algae, iron bacteria.
Drying periods between an irrigation and harvest are required when the waste-
water contacts the plant surfaces.
Not all crops, especially those used for human food without cooking, may be ir-
rigated with wastewater and reclaimed water. Restrictions depend on the nature
of the crop and the quality of the wastewater or reclaimed water.
Wastewater lines and potable water lines must be separated to avoid contamina-
tion of potable water. This is also true for reclaimed water.
20.7.2 Center Pivot Systems
Center pivot systems are gaining in popularity to apply reclaimed water and
wastewater irrigation due primarily to their fit into the management system, especially
for managers whose primary operation is not crop production or irrigation. For a proc-
essing plant operator or a dairy manager, the center pivot is easier to manage, operate,
and maintain than other sprinkler systems. When compared to a side roll (wheel line)
or a big gun (gun sprinklers), the system has less reliance on skilled labor for schedul-
ing the irrigation sets and moving the system. Record keeping is facilitated. The uni-
formity of water and nutrient application is also greater with the center pivot than with
a big gun.
The experiences of an irrigation equipment company in the major dairy region of
Texas are worthy of mention. As of 1995, the company installed 33 center pivots for
irrigation of wastewater from dairies and a cheese processing plant in the Stephenville
area (M. Stewart, 1995 personal communication. DeLeon Irrigation Co., DeLeon,
Texas). The company also installs center pivots for row crops and forages, so has a
good basis for comparison of system operation with different qualities of water. Plug-
ging of nozzles and other system components and corrosion were the two major poten-
tial problems associated with center pivot irrigation of wastewater.
Several features of design and operation significantly decrease the nozzle plugging
potential. When compared to a center pivot for row crops, the nozzle drops have a
greater spacing and the nozzle diameters are larger. As an example, a recent installa-
tion has drops on a 4.5 m (17 ft) spacing, with a smallest nozzle diameter of 7.1 mm
(9/32 in). The design operating pressure ranged from 135 to 170 kPa (20 to 25 psi),
which is higher than a typical low-pressure center pivot system. This combination of
larger nozzle diameters and higher pressure results in a system that expels most of the
solids in the water. An end gun on the pivot has proven indispensable for a relatively
plugging-free system. With an end gun, the velocity of flow in the entire length of
lateral keeps solids from settling and accumulating or aggregating. The booster pump
for the end gun occasionally picks up some trash, but the outage rates have not been
excessive. Self-draining valves, one for each span, provide drainage at the end of the
irrigation period. This draining and drying reduces solids accumulation and resulting
pipe corrosion, development of odors, and continuous growth of microorganisms.
The source of wastewater for dairy systems should be the second stage of the la-
goon system. The primary lagoon should not be used as a source of water for center
pivot irrigation because of high suspended solids, greater potential for odor, and
higher levels of nutrients. The pump and intake are mounted on a floating platform
some distance from the shore, with an intake depth of about 0.3 m. This avoids float-
ing and bottom solids. As a consequence of high BOD in the wastewater, the water is
normally anaerobic, so algae are not a problem. The inlet pipe has a trash screen, with
openings at least 12.5 mm in diameter. A separate cage of fine mesh screen for the
intake is not required, nor is a filter for the system. At the end of the irrigation cycle,
the inlet pipe is backflushed as water drains from the line into the lagoon. An air relief
valve at the high point of the irrigation line, normally the lagoon embankment, is nec-
essary. The design capacity of the center pivot is normally based on crop irrigation
needs for the area covered, rather than on the average flow rate of wastewater. This
overdesign of system capacity results in higher flow rates and greater flexibility of
system operation, especially for lagoon pump-down following runoff from open lots
due to heavy rains.
20.7.3 Big Gun Systems
Big gun systems are widely used in wastewater irrigation, especially when the
wastewater application fields are nonuniform in shape or when tree crops or wooded
areas are used for the land treatment (Yoder, 1994). Big gun systems have the advan-
tage of large nozzle diameters, which enables pumping of slurries and wastewater with
suspended solids. The high pressure of big gun systems also facilitates pumping with
solids without system plugging. Filtration requirements are eliminated, but the intake
pipe inlet should be screened to keep large objects out of the pump and lines. Rapid
emergency pump-down of lagoons with high suspended solids is facilitated with big
gun systems.
20.7.4 Microirrigation Systems
The use of microirrigation for reclaimed water and wastewater is increasing despite
problems with clogging due to physical, biological, and chemical constituents in the
wastewater. The potential for emitter plugging due to degraded water quality is the
major concern for microirrigation with reclaimed water (Hills and Tajrishy, 1995).
The high quality of reclaimed water that has had tertiary treatment and disinfection is
very compatible with microirrigation systems. However, virtually all agricultural
wastewater and most municipal wastewater with only secondary treatment have too
high an organic matter content (high BOD and TSS) for chlorination or other disinfec-
tion to be practical. Chlorination of high-strength organic material in water is known
to produce carcinogenic substances, so high doses of chlorine are not a feasible solu-
tion. Consequently, microirrigation with emitters is generally limited to reclaimed
water or low-strength secondary wastewater from reservoir storage. Microsprinklers,
microsprayers, bubblers, and other large-orifice forms of microirrigation could be used
if the water is adequately filtered. Hills and Brenes (2001) concluded that drip tape
appears to be suitable for use with effluent from a secondary clarifier after chlorination
and filtration with a silica sand no. 20 filter and 105-µm screen filter at each manifold.
Periodic chlorination, ideally with a clean water source and flushing, would still be
required to control growth of biological films and slimes.
20.7.4.1 Subsurface drip systems. Subsurface drip irrigation will pose less threat
of pathogen contamination (Phene and Ruskin, 1995; Trooien et al., 2000), but may be
more difficult to manage. Gushiken (1995) lists the following advantages of subsur-
face drip irrigation with reclaimed water: (1) minimizes health risks; (2) minimizes
liability exposure associated with overspray and aerosol drift near residential proper-
ties; and (3) eliminates odor, ponding, and runoff, in addition to the other advantages
of subsurface drip irrigation over other forms of irrigation.
When subsurface drip irrigation was used in a high-frequency mode in a California
study, minimum leaching of nitrate occurred (Phene and Ruskin, 1995). The four rec-
ommended conditions are: (1) irrigation events are short and frequent and designed to
replace crop water uptake as closely as possible (no leaching); (2) nitrogen is applied
through the system at a rate equivalent to the crop uptake rate less the amount mineral-
ized from the soil; (3) the crop is deep rooted; and (4) the water table is at least 2 m
from the soil surface (Phene and Ruskin, 1995).
Oron et al. (1992) studied drip irrigation (surface and subsurface) of cotton, alfalfa,
sweet corn, and wheat with secondary wastewater in Israel. They noted no technical
failures if proper filtration was done. No contamination of agricultural products with
microbes was detected. Subsurface drip irrigation for home landscapes is increasingly
used for effluent from household septic systems when subsurface constructed wetlands
are incorporated into the design. The wetlands remove the suspended solids that oth-
erwise rule out microirrigation systems for wastewater.
20.7.4.2 Emitter plugging. Two approaches are used to reduce clogging: improved
emitter design and pretreatment of water used for microirrigation (Hills and Tajrishy,
1995). The improved emitter designs include turbulent-flow labyrinth or self-cleaning
types. Treatment for reclaimed water is primarily filtration, followed by disinfection
with chlorine or an equivalent oxidant, or exposure to ultraviolet light. Hills and Ta-
jrishy (1995) state that filtration followed by UV disinfection and chlorination may be
necessary to produce water with an acceptable bacterial count for microirrigation of
fresh market produce. Filtration alone is not sufficient to eliminate the clogging poten-
tial (Adin and Sacks, 1991; Hills and Tajrishy, 1995). In addition, filtration of waste-
water from storage reservoirs results in the need for very frequent backflushings (Rav-
ina et al., 1995), which interrupts the irrigation and creates a disposal problem.
Municipal effluent from a secondary clarifier in California was examined for plug-
ging potential in a microirrigation system by Hills and Tajrishy (1995). Combinations
of filtration (silica sand no. 20 or 150-mesh (100-µm) screen) and disinfection were
used with turbulent labyrinth and self-flushing in-line emitters. Their conclusions
were: (1) Chlorine (or equivalent disinfection) is necessary to prevent growth of bacte-
rial slimes and algae within microirrigation systems using reclaimed water. (2) Filtra-
tion alone with either media or screen filters does not prevent clogging. (3) Adequate
filtration with granular media filters will reduce the chlorination requirement and fre-
quency of line flushing. (4) Intermittent chlorination of 2 mg/L free residual chlorine
during the last hour of an irrigation cycle is as effective as continuous chlorination of
0.4 mg/L for preventing biological film formation and emitter clogging. (5) UV disin-
fection alone does not prevent clogging of emitters. The self-flushing emitters required
granular media filtration and chlorination to maintain high uniformity, but either me-
dia or screen filtration with chlorination was sufficient for the labyrinth turbulent flow
emitters.
In a study of microirrigation with filtered wastewater from a reservoir in Israel,
Adin and Sacks (1991) reached the following conclusions: (1) Clogging of inline emit-
ters is caused primarily by suspended solids, but they do not necessarily initiate the
clogging process. (2) Sediment buildup begins with deposition of amorphous biologi-
cal films, to which other particles adhere. (3) Algae cause emitter clogging only when
attached to other particles. (4) Filtration prevents immediate clogging by large or ir-
regular-shaped particles. (5) Clogging potential may be reduced by chemical pretreat-
ment with oxidants and flocculants or by modifying the internal emitter design.
Labyrinth long-path emitters were much more prone to clogging than a self-
regulating emitter. In an earlier study, Adin and Elimelech (1989) reported that screen
filters (80- and 130-µm polyester) performed very poorly, with only a few percent
particle removal, with wastewater from a storage reservoir. Deep-bed granular media
filters (effective grain sizes of 0.7, 0.9, and 1.2 mm) removed 30% to 70% of the sus-
pended particles.
Juanico et al. (1995) examined effects of adding municipal wastewater effluent to a
shallow, freshwater reservoir in Israel. They noted that the addition of effluent very
quickly (within 2 wks) and very dramatically (by a factor of 2 to 4) increased the
clogging capacity of the water on an 80-µm micron mesh screen filter. This was pri-
marily due to an increase in large plankton species (cyanophytes, rotifers, and crusta-
ceans). Sagi et al. (1995) identified a colonial protozoa (Epystilys balanarum) and
sulfur bacteria (Beggiotoa alba) as the major sources of clogging of emitters in Israel.
In addition to accumulation on all components of the system, these organisms pro-
duced a film or slime that enhanced the accumulation of other particles. The sulfur
bacteria were present only in water containing hydrogen sulfide (H2S) and oxygen.
Both are controlled by chlorination, but not by filtration.
20.7.5 System Components
Pumps selected for wastewater irrigation systems are generally centrifugal, pow-
ered by electric motors (Sneed, 1991; MWPS, 1993). The procedures for determina-
tion of the total dynamic head are the same as for pumping of first-use water. Pump
flow capacity is normally determined by the volume or flow rate of the wastewater.
An exception will occur when dewatering of a lagoon is specified by regulation. For
example, Texas regulations require the CAFO lagoon be pumped to the permanent
marker (top of the design operating volume) within 21 days whenever the lagoon level
reaches or exceeds a level corresponding to one-half of the design runoff volume
(TWC, 1990; TNRCC, 1994). Natural Resources Conservation Service (NRCS) guide-
lines for the design capacity of the pump are based on three limitations: no pumping at
night, no pumping during rainfall, and no pumping to saturated ground except when
necessary to protect the lagoon from overtopping. This simplifies, with assumptions,
to an informal NRCS guideline that the pump capacity should be sufficient to pump
the storm runoff from a 24-hour, 25-year precipitation event on the area of the CAFO
within 10 days, with the pump operating 10 h/day (K. Schrunk, 1994 personal com-
munication, NRCS Area Office, Stephenville, Texas). Pumps sized according to this
guideline will determine the desired capacity of the irrigation system. The pump will
be overdesigned for normal wastewater generation rates. In practice, when N loading
is the limiting design factor, the irrigation system may be in operation only a few days
per month under normal or below-normal precipitation conditions.
The solids content of the wastewater has to be taken into account in the component
selection. In general, sprinkler irrigation systems can be used with wastewater solids
content up to 4% without affecting system design (MWPS, 1993). For higher solids
content, specialized equipment is needed. Open-impeller centrifugal pumps can handle
liquids with up to 15% solids, especially when the pump has a cutter blade at the inlet
(MWPS, 1993). The pumps with cutter blades (trash pumps) are compatible with big
gun sprinkler systems, but care should be taken to ensure that the system has sufficient
pressure. If the wastewater has more than about 1% solids, an experienced design en-
gineer should be consulted (Sneed, 1991).
Filters will be needed for microirrigation with wastewater. An experienced design
engineer should be consulted for filtration system selection.
20.7.6 System Pipeline Operation
Several general guidelines to minimize problems with system operation are appar-
ent. Thermoplastic (e.g., PVC) system components will minimize corrosion problems,
but may increase problems associated with algal growth in the lines. PVC pipes for
aboveground operation would need to have ultraviolet light protection. The irrigation
lines should be drained or flushed with high-quality water after irrigation to avoid cor-
rosion in metal lines and development of odors in any lines. The major problems occur
with solids that settle and develop anaerobic conditions. The nutrient content of the
wastewater promotes the growth of algae, bacteria, and protozoa that may cement the
solids and restrict pipe capacity. Hydrogen sulfide gas as a product of anaerobic diges-
tion in pipes will collect at high points in the pipe line. A metal pipe without air re-
lease valves at high points is especially susceptible to corrosion (SCS, 1983; Wescot
and Ayers, 1985).
20.8 DESIGN AND OPERATION OF SYSTEMS
TO PROTECT HUMAN HEALTH
Design and operation of reclaimed water and wastewater systems for areas with
public access require special attention to protect human and animal health. General
guidelines are contained in a U.S. Golf Association guide (USGA, 1994). Although
these guidelines were written for reclaimed water use in areas with public access, most
are also applicable to irrigation with wastewater. The regulations of each state or
country should be reviewed for compliance.
Precautions to take to avoid or decrease the health risk of irrigation with reclaimed
water and wastewater are:
Humans and animals should not be allowed in application fields that are being
sprinkler irrigated with wastewater or reclaimed water. Turf, forage, and other
vegetation should be allowed to dry completely following irrigation before hu-
man entry is allowed into the area. Irrigations should be applied in cycles, with
each irrigation followed by several days of non-irrigation.
Sprinkler systems should not be operated at night or during periods of rain to re-
duce the possibilities of direct runoff. Runoff of wastewater should not be al-
lowed to enter surface water sources.
Irrigation with systems that minimize exposure of humans, animals, or food
crops to the wastewater should be considered for areas with high potential for
traffic or disease. Workers in direct contact with wastewater should change
clothing and wash thoroughly after contact with the wastewater.
To reduce aerosol formation, irrigation systems should have a low operating
pressure and a low height of application. A buffer zone should be established be-
tween the sprinkler irrigation site and areas with public access. Sprinkler or
spray irrigation should be applied during periods with good vertical dispersion,
rather than lateral dispersion, which could carry aerosols downwind.
All agricultural and municipal water of runoff origin should be considered to

have indicator bacteria counts higher than the primary contact standard, unless
water quality analyses indicate otherwise.
To protect against cross connection with potable water lines:
Use a minimum of a reduced-pressure principle backflow device at the service
connection of the potable water system.
Include a double check-valve assembly for any dedicated potable water supply
for fire protection.
Include an air-gap separation assembly for potable water lines connected to lake
or pond fill lines.
Remove all physical connections between the reclaimed water system and the
potable water system.
Do not use common trenches for potable water and reclaimed water lines and do
not allow unknown lines in a retrofit.
The reclaimed water system should have lower pressure than the potable water
system.
Maintain a separation between reclaimed water lines and potable water lines.
These requirements may vary from state to state. The minimum separation for
parallel lines should be 3.0 m, with 1.2 m for special circumstances. The re-
claimed water line should be 0.3 m deeper than the potable water line. When
lines cross, the reclaimed water line should be at least 0.3 m below the potable
water line.
Potable and wastewater lines should not be connected or have a possibility of
being connected.
Reclaimed water lines should be protected against unanticipated uses by the
public. Special quick-coupler valves or special tools for access to the reclaimed
water line are recommended. No reclaimed water lines should have hose bibbs.
The clean water source for flushing of wastewater lines after use should be pro-
tected against backflow or contamination.
All onsite irrigation and potable water wells should be protected. A concrete slab
extending 3.0 m in all directions is recommended. Drinking fountains should have
special, self-closing covers to protect them from spray from sprinkler irrigation sys-
tems using reclaimed water.
All reclaimed water lines, valves, and other equipment should be clearly marked.
Reclaimed water lines should be a different color or color pattern (e.g., stripes) than
potable water lines. Purple is the preferred color. Warning signs should be numerous
and conspicuous, preferably with a purple background. For example, golf score cards
and other printed material should contain appropriate warnings.
Install low-pressure sensors with the irrigation system controller to shut the system
down in the event of uncontrolled discharge from a pipe or nozzle break. Incorporate
valves for flushing at low elevations and at the end of lines. Incorporate valves for air
relief at high locations on the lines.
Depending on the concentration of suspended solids in the reclaimed water lines,
use dirty-water valves and nozzles that are less prone to clogging (e.g., impact nozzles
instead of gear-driven or valve-in-head nozzles). If the reclaimed water is high in sa-
linity, avoid the use of dissimilar metals in the system components.
If storage of reclaimed water is necessary, use storage tanks instead of reservoirs. If

the reclaimed water is stored in a reservoir, filtration at the intake will likely be
needed. Consider a primary screen of 10- to 30-mesh at the source inlet, a media filter
for algae and other suspended solids, and a stainless steel screen filter if a microirriga-
tion system is used. Design on-site storage reservoirs for wastewater irrigation to
minimize potential problems with weeds, algae, odors, aesthetics, and health hazards.
When possible, the flow of fresh influent into the irrigation reservoir should be termi-
nated during the irrigation period to prevent short circuiting. Mechanical agitation or
other source of turbulence for the irrigation reservoir should be avoided during the
irrigation, except for the special case of infrequent sediment or sludge removal, which
requires special equipment and precautionary measures regarding crops, timing, rates,
and runoff control practices. Include aeration and oxidation features for reclaimed
water reservoirs and water features that collect runoff from the land application site.
These features include fountains, waterfalls, constructed wetlands for algal oxidation,
and air injectors. On-site reclaimed water storage reservoirs should be protected
against public access unless the water meets USEPA primary contact standards for
bacteria.
Use a dual irrigation system (i.e., two irrigation systems with separate water
sources) if the reclaimed water contains anions or other constituents that could ad-
versely impact growth or appearance of ornamental plants, golf course greens and
tees, and other sensitive plantings. This is especially necessary when the wastewater
treatment plant is down for maintenance or if the delivery times are unreliable. If the
reclaimed water or wastewater is of poor quality, blending with fresh water will help
in controlling total dissolved solids, total suspended solids, and BOD.
20.9 MONITORING
There are two types of monitoring requirements. The first is more or less standard
monitoring procedures to determine if the irrigation system is operating within design
specifications. The second is monitoring to determine if the land treatment system is
operating within design specifications, with an emphasis on protection of human
health (Westcot, 1997). Monitoring for irrigation system performance is adequately
covered in other sections of this monograph, so the focus here will be on monitoring
for land treatment performance. This may be divided into two categories: compliance
monitoring and process monitoring (Reed and Crites, 1984). Monitoring for compli-
ance should include the details of the sampling in a quality management plan (Wescot
and Ayers, 1985). Assistance with development of the plan may be obtained from the
NRCS, the Cooperative Extension Service, a state or federal water quality agency,
and/or a consulting engineering firm.
20.9.1 Compliance Monitoring
Compliance monitoring for nutrient content of surface runoff, nutrient content of
soil water percolating to groundwater, nutrient loading, and hydraulic loading may be
under the purview of appropriate state and federal agency guidelines and regulations.
Records may be required. Considering the dynamic nature of guidelines and regula-
tions and the variations in different jurisdictions, this section is not to be construed as
an authoritative source on monitoring requirements. Consult the NRCS, state exten-
sion waste management specialists, consulting engineers, and state and federal envi-
ronmental protection or natural resources agencies. Record keeping is generally the
responsibility of the operator or manager of the land treatment system.
20.9.2 Process Monitoring

Process monitoring for land treatment of wastewater serves the following purposes
(USEPA, 1981; Westerman and King, 1983): to verify system performance relative to
loading rate, environmental impact, and treatment effectiveness; to determine degrada-
tion of soil, groundwater, and surface water; to assess the vegetation-soil system to
ensure that viable vegetative cover is maintained and that vegetation management is
effective; to determine safety of use of harvested or grazed vegetation; to monitor
wastewater generation rate and composition for significant changes that would impact
system operation; to monitor effectiveness of treatment processes prior to land appli-
cation of reclaimed water or wastewater; and to determine needs for subsurface drain-
age systems, soil amendments, additional nutrients, or other corrective actions.
20.9.3 Soils
Some of the soils data needed for design of the land treatment system will also
serve as baseline information for the monitoring program. If the wastewater has con-
stituents that will be removed by soil treatment, the soil information will also serve as
a baseline. The soil background data that may be needed for baseline include: texture,
structure, and profile; infiltration rate and permeability; soil horizontal and vertical
variations that would impact land application; background phosphorus and nitrogen
content, with special emphasis on the various types of phosphorus; heavy metals con-
tent; calcium, iron, and aluminum content and forms, if phosphorus loading is at a rate
that exceeds crop uptake and removal; chemical properties of soil and soil water, in-
cluding CEC, pH, total salinity or conductivity, specific anions, SAR, COD, Eh, alka-
linity, and phosphorus sorption capacity; organic matter; groundwater levels (which
may be seasonal) and gradient; and groundwater chemical constituents and properties,
if the groundwater is shallow enough to be a factor in land application (these include
nitrate and other forms of nitrogen, anions, pH, Eh, COD, phosphorus, alkalinity, or-
ganic carbon, and metals including calcium, iron, and aluminum).
20.9.4 Groundwater and Vegetation
In some cases, groundwater should be monitored periodically during operation of
the land treatment/irrigation system. Monitoring wells should be located up-gradient,
on site, and down-gradient if the soil profile to groundwater is permeable and the po-
tential exists for groundwater contamination. Careful site selection and the use of well
casings are important to avoid contamination. George et al. (1987) is an excellent ref-
erence for the design of a groundwater monitoring program for a site impacted by ex-
cessive loading.
Sampling of vegetation should be included in a monitoring program. Vegetation
variables of interest are yield and/or total dry matter removed by harvest and nutrient
content, especially total nitrogen, nitrate, and phosphorus. The sampling interval for
monitoring vegetation is normally at harvest. Harvest dates and yields should be re-
corded, with nutrient content recorded if available. Pre-harvest sampling is appropriate
when the vegetation is to be grazed or harvested for forage at times when the nitrate
content may be high enough to be toxic to livestock. These conditions primarily exist
for cereal grain and corn crops with high nitrogen fertility during conditions that favor
low photosynthetic rates (cool, cloudy weather and low soil moisture) (Heath et al.,
1985). An agronomist should be consulted to provide details of the appropriate forage
or crop sampling program.
20.9.5 Water
A suggested protocol for wastewater monitoring will include samples of the follow-
ing constituents and properties at influent and effluent locations: nitrogen forms, such
as organic, total Kjeldahl, ammonium, and nitrate; phosphorus forms, such as ortho-
phosphate, organic, inorganic, bioavailable, and iron oxide strip reactive; sodium, cal-
cium, and magnesium for SAR or ESP calculation; electrical conductivity and total
dissolved solids; Eh, pH, and dissolved oxygen; trace (heavy) metals; alkalinity; BOD
and/or COD; total dissolved and suspended solids, with separation into volatile and
fixed fractions; pathogens or indicator organisms if the wastewater application is in an
area with public access; oil and grease if the wastewater is from a municipal or proc-
essing source; total organic carbon (TOC), depending on wastewater source; and vola-
tile organic compounds (VOC), depending on wastewater source.
20.9.6 Monitoring Frequency
The frequency of the monitoring program depends on the size of the system and the
potential impact of any variations on system operation. The frequency of compliance
monitoring will be established by the regulatory or permitting agencies. Process moni-
toring frequency for wastewater should be at least quarterly, with annual monitoring of
forage, soil, and groundwater variables (Reed and Crites, 1984). Significant variations
in loading or flow may dictate sampling on a more frequent basis. Water levels in la-
goons, ponds and reservoirs used for storage of reclaimed water or wastewater prior to
irrigation withdrawal should be measured at least weekly. Precipitation form and
amount should be measured daily.
Monitoring at the application site may also be required by regulatory or permitting
agencies. Times and amounts of irrigation should be recorded. Ideally, the basis for
irrigation scheduling, such as soil moisture depletion and evapotranspiration rate, will
be monitored and recorded. However, this is not practical for many wastewater appli-
cation locations and operations. Supplemental fertilizer formulations and rates should
be recorded. Solid waste or slurry application and commercial fertilizer application, if
any, on the wastewater irrigation site should be recorded.
REFERENCES
Adin, A., and M. Elimelech. 1989. Particle filtration for wastewater irrigation. J. Irrig.
Drain. Eng. 115: 474-487.
Adin, A., and M. Sacks. 1991. Dripper-clogging factors in wastewater irrigation. J.
Irrig. Drain. Eng. 117: 813-826.
Ali, I. 1987. Wastewater criteria for irrigation in arid regions. J. Irrig. Drain. Eng.
113: 173-183.
Allhands, M. N., S. A. Allick, A. R. Overman, W. G. Leseman, and W. Vidak. 1995.
Municipal water reuse at Tallahassee, Florida. Trans. ASAE 38: 411-418.
Asano, T., D. Richard, R. W. Crites, and G. Tchobanoglous. 1992. Evolution of
tertiary treatment requirements in California. Water Environ. & Tech. 4(2): 36-41.
Ayres, R. M., G. P. Alabaster, D. D. Mara, and D. L. Lee. 1992. A design equation for
human intestinal nematode egg removal in waste stabilization ponds. Water Res.
26: 863-865.
Barnes, R. F., D. A. Miller, and C. J. Nelson, eds. 1995. Vol I: Forages. In An
Introduction to Grassland Agriculture. 5th ed. Ames Iowa: Iowa State Univ. Press.
Bowmer, K. H., and P. Laut. 1992. Wastewater management and resource recovery in
intensive rural industries in Australia. Water Res. 26: 201-208.
Brenner, K. P., P. V. Scarpino, and C. S. Clark. 1988. Animal viruses, coliphages, and
bacteria in aerosols and wastewater at a spray irrigation site. Appl. Environ. Micro.
54: 409-415.
Broadbent, F. E., and H. M. Reisenaur. 1985. Fate of wastewater constituents in soil
and groundwater: nitrogen and phosphorus. In Irrigation with Reclaimed Municipal
Wastewater: A Guidance Manual, 12.1-12.16. G. S. Pettygrove and T. Asano, eds.
Boca Raton, Fla.: Lewis Pub.
Burton, T. M., and D. L. King. 1981. The Michigan State University water quality
management facility: A lake-land system to recycle municipal wastewater. In
Municipal Wastewater in Agriculture, 249-269. F. M. d’Itri, J. Aguirre-Martinez,
and M. Athie-Lambarri, eds. New York, N.Y.: Academic Press.
Chamblee, D. S., and A. E. Spooner. 1985. Hay and pasture seedlings for the humid
south. In Forages: The Science of Grassland Agriculture, 359-370. 4th ed. M. E.
Heath, R. F Barnes, and D. S. Metcalf, eds. Ames Iowa: Iowa State Univ. Press.
Crites, R. W. 1987. Winery wastewater land application. In Proc. Conf Irrigation and
Drainage Division, ASCE, Irrigation Systems for the 21st Century, 529-536. L. G.
James and M. J. English, eds. Reston, Va.: American Soc. Civil Engineers.
Cross, P., and J. L. Jackson, Jr. 1993. Citrus trees blossom with reclaimed water.
Water Environ. & Tech. 5(Feb): 27-28.
Daniel, T. C., D. R. Edwards, and A. N. Sharpley. 1993. Effect of extractable soil
surface phosphorus on runoff water quality. Trans. ASAE 36: 1079-1085.
Diaz, O. A., K. R. Reddy, and P. A. Moore, Jr. 1994. Solubility of inorganic
phosphorus in stream water as influenced by pH and calcium concentration. Water
Res. 28: 1755-1763.
Dravnieks, A. 1979. Measurement methods. In Odors from Stationary and Mobile
Sources, 82-168. National Research Council. Washington, D.C.: National Academy
Press.
Durand, R., and G. Schwebach. 1989. Gastrointestinal effects of water reuse for public
park irrigation. American J. Public Health 79: 1659-1660.
El Hamouri, B., K. Khallayoune, K. Bouzoubaa, N. Rhallabi, and M. Chalabi. 1994.
High-rate algal pond performance in faecal coliforms and helminth egg removals.
Water Res. 28: 171-174.
Fox, D. R., and G. S. Nuss. 1987. Management challenges affecting agricultural reuse
of high-strength food processing wastes. In Proc. Conf. Irrigation and Drainage
Division, ASCE, Irrigation Systems for the 21st Century, 537-544. L. G. James and
M. J. English, eds. Reston, Va.: American Soc. Civil Engineers.
George, D. B., D. B. Leftwhich, N. A. Klein, and B. J. Claiborn. 1987. Redesign of a
land treatment system to protect groundwater. J. WPCF 59: 813-820.
George, M. R., G. S. Pettygrove, and W. B. Davis. 1985. Crop selection and
management. In Irrigation with Reclaimed Municipal Wastewater: A Guidance
Manual, 6.1-6.18. G. S. Pettygrove and T. Asano, eds. Boca Raton Fla.: Lewis
Publishers.
Goldman, J. C., D. Procella, E. J. Middlebrooks, and D. F. Toerien. 1972. The effect
of carbon on algal growth: Its relationship to eutrophication. Water Res. 6: 637-
679.
Greenberg, A. E., L. S. Clesceri, and A. D. Eaton. eds. 1992. Standard Methods for the
Examination of Water and Wastewater. 18th ed. American Public Health Assn.,
American Water Works Assn. and Water Environ. Fed. Washington, D.C.: APHA.
Gushiken, E. C. 1995. Irrigating with reclaimed water through permanent subsurface
drip irrigation systems. In Proc. Fifth Int’l Microirrigation Congress, 269-274. F.
L. Lamm, ed. St Joseph Mich.: ASAE.
Hansen, J. S., and J. E. Ongerth. 1991. Effects of time and watershed characteristics
on the concentration of Cryptosporidium oocysts in river water. Appl. and Environ.
Micro. 57: 2790-2795.
Heath, M. E., R. F Barnes, and D. S. Metcalf, eds. 1985. Forages: The Science of
Grassland Agriculture. 4th ed. Ames Iowa: Iowa State Univ. Press.
Heaton, R. D. 1981. Worldwide aspects of municipal wastewater reclamation and
reuse. In Municipal Wastewater in Agriculture, 43-74. F. M. D’Itri, J. Aguirre-
martinez, and M. Athie-Lambarri, eds. New York, N.Y.: Academic Press.
Hegg, R. O., A. T. Shearin, D. L. Handlin, and L. W. Grimes. 1984. Irrigation of
swine lagoon effluent onto pine and hardwood forests. Trans. ASAE 27: 1311-1418.
Hills, D. J., and M. J. Brenes. 2001. Microirrigation of wastewater effluent using drip
tape. Appl Eng. in Agric. 17: 303-308.
Hills, D. J., and M. Tajrishy. 1995. Treatment requirements of secondary effluent for
microirrigation. In Proc. Fifth Int’l Microirrigation Congress, 887-892. F. L.
Lamm, ed. St Joseph Mich.: ASAE.
Hubbard, R. K., D. L. Thomas, R. A. Leonard, and J. L. Butler. 1987. Surface runoff
and shallow ground water quality as affected by center pivot applied dairy cattle
wastes. Trans. ASAE 30: 430-437.
Husu, J. R., S. K. Sivela, and A. L. Rauramaa. 1990. Prevalence of Listeria species as
related to chemical quality of farm-ensiled grass. Grass and Forage Sci. 45: 309-
314.
Hutchins, S. R., M. B. Tomson, P. B. Bedient, and C. H. Ward. 1985. Fate of trace
organics during land application of municipal wastewater. CRC Critical Reviews in
Environ. Control 15: 355-427.
Jones, C. A., M. E. Harper, B. Horton, and T. J. Smayda. 1993. Wastewater treatment
and disposal facility for a vegetable processing facility. In Nat’l. I Conf. Irrig.
Drain. Eng. ASCE, Management of Irrigation and Drainage Systems: Integrated
Perspectives, 93-100. R. G. Allen, and C. M. U. Neale, eds. Reston, Va.: American
Jonsson, A. 1991. Growth of Clostridium tyrobutyricum during fermentation and
aerobic deterioration of grass silage. J. Sci. Food and Agr. 54: 557-568.
Juanico, M., Y. Azov, B. Teltsch, and G. Shelef. 1995. Effect of effluent addition to a
freshwater reservoir on the filter clogging capacity of irrigation water. Water Res.
29: 1695-1702.
Keeney, D. R. 1983. Transformations and transport of nitrogen. In Agricultural
Management and Water Quality, 48-64. F. W. Schaller, and G. W. Bailey, eds.
Ames Iowa: Iowa State Univ. Press.
Kirkham, M. B. 1986. Problems of using wastewater on vegetable crops. HortSci. 21:
24-27.
LeChevallier, M. W., W. D. Norton, and R. G. Lee. 1991. Occurrence of Giardia and

Cryptosporidium spp. in surface water supplies. Appl. and Environ. Micro. 57:
2610-2616.
Lejano, R. P., F. A. Grant, T. G. Richardson, B. M. Smith, and F. Farhang. 1992.
Assessing the benefits of water reuse. Water Environ. & Tech. 4(8): 44-47.
Loehr, R. C., W. W. Carey, A. Kull, and C. A. Swift. 1988. Full-scale land treatment
of coffee processing wastewater. J. Water Pollution Control Fed. 60: 1948-1952.
Loehr, R. C., W. J. Jewell, J. D. Novak, W. W. Clarkson, and G. S. Friedman. 1979.
Land Application of Wastes. Vol II. New York, N.Y.:Van Nostrand Reinhold
Environmental Engineering Series.
Loehr, R. C., C. S. Martin, and W. Rast, eds. 1980. Phosphorus Management
Strategies for Lakes. Ann Arbor, Mich.: Ann Arbor Science.
MacKenzie, W. R., N. J. Hoxie, M. E. Proctor, M. S. Gradus, K. A. Blair, D. E.
Peterson, J. J. Kazmierczak, D. G. Addiss, K. R. Fox, J. B. Rose, and J. Davis.
1994. A massive outbreak in Milwaukee of Cryptosporidium infection transmitted
through the public water supply. New England J. Med. 331: 161-167.
Mantovani, P., T. Asano, A. Chang, and D. A. Okun. 2001. Management practices for
nonpotable water reuse. Project 97-IRM-6. Alexandria, Va.: Water Environ. Res.
Foundation.
McBride, M. B. 1994. Environmental Chemistry of Soils. New York, N.Y.: Oxford
Univ. Press.
MWPS. 1993. Livestock Waste Facilities Handbook. 3rd ed. Midwest Plan Service
Publication No. MWPS-18. Ames, Iowa: Iowa State Univ. Press.
Moore, B. E., D. E. Cammann, C. A. Turk, and C. A. Sorber. 1988. Microbial
characterization of municipal wastewater at a spray irrigation site: The Lubbock
infection surveillance study J. Water Pollution Control Fed. 60: 1222-1230.
NAS. 1978. Nitrates: An Environmental Assessment. National Academy of Sciences.
Washington, D.C.: National Academy Press.
NRC. 1996. Use of Reclaimed Water and Sludge in Food Crop Production. National
Research Council. Washington, D.C.: National Academy Press.
NRC. 1998. Issues in Potable Reuse: The Viability of Augmenting Drinking Water
Supplies with Reclaimed Water. National Research Council. Washington, D.C.:
National Academy Press.
Niemi, R. M., and J. S. Niemi. 1991. Bacterial pollution of waters in pristine and
agricultural lands. J. Environ. Qual. 20: 620-627.
Ongerth, J. E., and H. H. Stibbs. 1987. Identification of Cryptosporidium oocysts in
river water. Appl. and Environ. Micro. 53: 672-676.
Oron, G., Y. DeMalach, Z. Hoffman, and Y. Manor. 1992. Effect of effluent quality
and application method on agricultural productivity and environmental control.
Water Sci. Tech. 26: 1593-1601.
Ostling, C. E., and S. E. Lindgren. 1991. Bacteria in manure and on manured and
NPK-fertilized silage crops. J. Sci. Food Agric. 55: 579-588.
Overman, A. R., and T. Schanze. 1985. Runoff water quality from wastewater
irrigation. Trans. ASAE 28: 1535-1538.
Page, A. L., and A. C. Chang. 1981. Trace metals in soils and plants receiving
municipal wastewater irrigation. In Municipal Wastewater in Agriculture, 351-372.
F. M. d’Itri, J. Aguirre-Martinez, and M. Athie-Lambarri, eds. New York, N.Y.:

Academic Press.
Parsons, L. R., T. A. Wheaton, and P. Cross. 1995. Reclaimed municipal water for
citrus irrigation in Florida. In Proc. Fifth Int’l Microirrigation Congress, 262-268.
F. L. Lamm, ed. St. Joseph, Mich.: ASAE.
Peacock, C. H. 1994. Wastewater irrigation for golfcourses: Advantages versus
disadvantages. In Wastewater Reuse for Golf Course Irrigation, 204-220. United
States Golf Association. Chelsea, Mich.: Lewis Publ.
Phene, C. J., and R. Ruskin. 1995. Potential of subsurface drip irrigation for
management of nitrate in wastewater. In Proc. Fifth Int’l Microirrigation Congress,
155-167. F. L. Lamm, ed. St. Joseph, Mich.: ASAE.
Rammer, C., C. Ostling, P. Lingvall, and S. Lindgren. 1994. Ensiling of manured
crops: Effects on fermentation. Grass and Forage Sci. 49: 343-351.
Ravina, I, E. Paz, G. Sagi, A. Schischa, A. Marcu, Z. Yechiely, Z. Sofer, and Y. Lev.
1995. Performance evaluation of filters and emitters with secondary effluent. In
Proc. Fifth Int’l Microirrigation Congress, 244-249. F. L. Lamm, ed. St Joseph
Mich.: ASAE.
Reddy, K. R. 1981. Diel variations of certain physico-chemical parameters of water in
selected aquatic systems. Hydrobiologia 85: 201-207.
Reed, S. C., and R. W. Crites. 1984. Handbook of Land Treatment Systems for
Industrial and Municipal Wastes. Park Ridge, N.J.: Noyes Press.
Reed, S. C., E. J. Middlebrooks, and R. W. Crites. 1988. Natural Systems for Waste
Management and Treatment. New York, N.Y.: McGraw-Hill.
Ritter, W. F. 1987. An overview of factors affecting land application of food
processing wastes. In Proc. Conf Irrigation and Drainage Division, ASCE,
Irrigation Systems for the 21st Century, 521-528. L. G. James, and M. J. English,
eds. Reston, Va.: American Soc. Civil Engineers.
Rosas, I., A. Baez, and M. Coutino. 1984. Bacteriological quality of crops irrigated
with wastewater in the Xochimilco plots, Mexico City, Mexico. Appl. and Environ.
Micro. 47: 1074-1079.
Rose, J. B. 1986. Microbial aspects of wastewater reuse for irrigation. CRC Critical
Rev. Environ. Control 16: 231-256.
Russell, J. M., R. N. Cooper, and S. B. Lindsey. 1993. Soil denitrification rates at
wastewater irrigation sites receiving primary-treated and anaerobically treated meat
processing effluent. Bioresource Tech. 43: 41-46.
Sagi, G., E. Paz, I. Ravina, A. Schischa, A. Marcu, and Z. Yechiely. 1995. Clogging of
drip irrigation systems by colonial protozoa and sulfur bacteria. In Proc. Fifth Int’l
Microirrigation Congress, 250-252. F. L. Lamm, ed. St. Joseph, Mich.: ASAE.
Schwebach, G. H., D. Cafaro, J. Egan, M. Grimes, and G. Michael. 1988. Overhauling
health effects perspectives. J. Water Pollution Control Fed. 60: 473-479.
Shamloufard, J., K. Weinberg, and R. Fornelli. 1995. Water water everywhere. Civil
Eng. 65(9): 43-46.
Sharpley, A. N., S. J. Smith, B. A. Stewart, and A. C. Mathers. 1984. Forms of
phosphorus in soil receiving cattle feedlot waste. J. Environ. Qual. 13: 211-215.
Sharpley, A. N., and P. J. A. Withers. 1994. The environmentally sound management
of agricultural phosphorus. Fert. Res. 39: 133-146.
Sheikh, B., R. P. Cort, W. R. Kirkpatrick, R. S. Jaques, and T. Asano. 1990. Monterey

wastewater reclamation study for agriculture. Res. J. WPCF 62(3): 216-226.
Sherer, B. M., J. R. Miner, J. A. Moore, and J. C. Buckhouse. 1992. Indicator bacteria
survival in stream sediments. J. Environ. Qual. 21: 591-595.
Shuval, H. I., A. Adin, B. Fattal, E. Rawitz, and P. Yekutiel. 1986. Wastewater
irrigation in developing countries: Health effects and technical solutions. UNDP
Project Management Report No. 6. World Bank Technical Paper 51. Washington,
D.C.: The World Bank.
Shuval, H. I., Y. Wax, P. Yekutiel, and B. Fattal. 1989. Transmission of enteric
disease associated with wastewater irrigation: A prospective epidemiological study.
American J. Public Health 79: 850-852.
Sneed, R. E. 1991. Land application of wastewater and sludges. Irrig. J. 41(5): 10-
12+.
SCS. 1983. Chapt. 11: Irrigation. Section 15: Sprinkle irrigation. In National
Engineering Handbook. Washington, D.C.: USDA, Soil Conservation Service.
SCS. 1992. Agricultural Waste Management Field Handbook, Part 651. Washington,
D.C.: USDA, Soil Conservation Service.
Stumm, W., and J. J. Morgan. 1996. Aquatic Chemistry: Chemical Equilibria and
Rates in Natural Waters. 3rd ed. New York, N.Y.: Wiley-Interscience, John Wiley
& Sons.
Sweeten, J. M., and M. L. Wolfe. 1994. Manure and wastewater management systems
for open lot dairy operations. Trans. ASAE 37: 1145-1154.
Tanji, K. K., ed. 1990. Agricultural Salinity Assessment and Management. New York,
Tchobanoglous, G., and F. L. Burton. 1991. Wastewater Engineering: Treatment,
Disposal, and Reuse. 3rd ed. New York, N.Y.: McGraw-Hill, Metcalf & Eddy, Inc.
Tedaldi, D. J., and R. C. Loehr. 1992. Effects of waste-water irrigation on aqueous
geochemistry near Paris, Texas. Ground Water 30: 709-719.
TNRCC. 1994. 30 TAC 321.34 Texas Register. December 19: 10281-10394. Austin
Tex.: Texas Natural Resources Conservation Commission.
TWC. 1990. Control of certain activities by rule. Subchapter B. Livestock and poultry
operations. 31 TAC 321.42-321.46. Texas Register. April 27: 2420-2421 and June
22: 3639-3640. Austin, Tex.: Texas Water Commission.
Thelin, R., and G. F. Gifford. 1983. Fecal coliform release patterns from fecal material
of cattle. J. Environ. Qual. 12: 57-63.
Thompson, R. B., D. Morse, K. A. Kelling, and L. E. Lanyon. 1997. Computer
programs that calculate manure application rates. J. Prod. Agric. 10: 58-69.
Trooien, T. P., F. R. Lamm, L. R. Stone, M. Alam, D. H. Rogers, G. A. Clark, and A.
J. Schlegel. 2000. Subsurface drip irrigation using livestock wastewater: Dripline
flow rates. Appl. Eng. Agric. 16: 505-508.
USGA. 1994. Wastewater Reuse for Golf Course Irrigation. United States Golf Assoc.
Boca Raton, Fla.: Lewis Pub.
USEPA. 1981. Process Design Manual: Land Treatment of Municipal Wastewater.
Cincinnati Ohio: USEPA Center for Environmental Research Information.
Vymazal, J. 1995. Algae and Element Cycling in Wetlands. Boca Raton, Fla.: Lewis
Pub.
Ward, B. K., and L. G. Irving. 1987. Virus survival on vegetables spray-irrigated with
wastewater. Water Res. 21: 57-63.
Ward, R. L., D. R. Knowlton, J. Stober, W. Jakubowski, T. Mills, P. Graham, and D.
E. Camann. 1989. Effect of wastewater spray irrigation on rotavirus infection rates
in an exposed population. Water Res. 23: 1503-1509.
Westcot, D. W. 1997. Quality Control of Wastewater for Irrigated Crop Production.
Food and Agriculture Organization of the United Nations. Rome, Italy: FAO.
Wescot, D. W., and R. S. Ayers. 1985. Chapt. 3: Irrigation water quality criteria. In
Irrigation with Reclaimed Municipal Wastewater: A Guidance Manual, 1-37. G. S.
Pettygrove and T. Asano, eds. Boca Raton, Fla.: Lewis Pub.
Westerman, P. W., and L. D. King. 1983. General guidelines for land treatment of
wastewater. Report No. 2. Land Treatment Series. Raleigh N.C.: Water Resources
Research Institute, Univ. of North Carolina.
Wheaton, F. W. 1977. Aquacultural Engineering. New York, N.Y.: John Wiley &
Sons.
Yoder, R. 1994. Traveling sprinklers: Taming Tennessee’s topography. Irrigation J.
44(7): 14-17.
CHAPTER 21
EVALUATING
PERFORMANCE
Joseph M. Lord, Jr. (JMLord, Inc.,
Fresno, California)
James E. Ayars (USDA-ARS,
Parlier, California)
Abstract. Evaluating irrigation system performance requires more than simply
monitoring the hydraulic operation of a given system. It requires understanding where
irrigation fits in the farm management, water requirements of a crop, soil hydraulic
characteristics, field conditions, and economics of the operation. In this chapter we
discuss the background and need for a comprehensive investigation of the system op-
eration in the context of the overall farm management. This requires an investigation
of the management variables that affect the operation, i.e., water supply constraints,
agronomic constraints, and labor constraints. Measures of performance that charac-
terize the system management and operation are discussed. The unique characteristics
of each irrigation system type are discussed to highlight the differences in approach to
evaluation. Finally, suggestions are provided to cover the performance evaluation,
data interpretation, report preparation, and recommendations.
Keywords. Distribution uniformity, Irrigation efficiency, Irrigation system per-
formance, Irrigation system performance evaluation, Microirrigation, Sprinkler irri-
gation, Surface irrigation.
21.1 INTRODUCTION
The Central Valley of California has experienced a rapid escalation in the cost of
agricultural water since the drought period 1989-1993, and, consequently, has seen
heightened competition for water supplies among urban, environmental, fish and wild-
life, and agricultural interests. Water charges have doubled and sometimes tripled. In
1994, a state-controlled water bank developed a market value for agricultural water of
$101/1000 m3. This value generated considerable interest in the agricultural commu-
nity and in some areas of the investment community, and they looked for even the
smallest opportunities to reduce water use in the agricultural sector.
The increased demand for water and the concerns related to disposal of agricultural
drainage water, environmental issues, and water allocations for agricultural use have
motivated government agencies (state and federal) to fund major programs to evaluate
on-farm irrigation system performance. Reasons commonly expressed by water agen-
cies and water districts for performing evaluations of farm irrigation systems include
water conservation, improving “beneficial” water use, and reducing losses to deep
percolation. However, farmers are more interested in improving crop yields, financial
returns, and making a limited supply go further.
The California Mobile Laboratory Program was developed to conduct on-farm
evaluation of irrigation systems by the U.S. Department of Agriculture’s (USDA)
Natural Resources Conservation Service (NRCS) and the State of California’s De-
partment of Water Resources. The expressed objectives of the Mobile Laboratory Pro-
gram are to assess the uniformity and efficiency of an irrigation event; to identify
problems with system design or management; and to identify opportunities for im-
provement. In 1996, approximately 60 water agencies in California were cost-sharing
with California’s Mobile Labs (Department of Water Resources, 1996).
Regardless of who is paying for the service, the evaluation of on-farm irrigation is
usually done for the owner/operator by some technical entity or professional service.
Even though funded or financed by the water district or another interested party, the
evaluation still should be oriented to the operator/owner of the system. It is counter-
productive and does not make economic sense to evaluate something and not try to
suggest design changes or improve the management of a system to address identified
shortcomings.
It is important for the engineer and scientist to recognize that, despite popular be-
lief, most farm irrigation systems were not designed by engineers or scientists, but
were developed from local practices and past successes. Until the early 1980s, the
USBR considered 160 acres to be a farming unit (Reclamation Law) and it is not hard
to believe that many of the major U.S. Bureau of Reclamation (USBR) projects were
conceived as 160-acre replications of identical irrigation units.
The engineer and scientist must recognize that most owners and operators are not
interested in the technical details of the evaluation, but how the results can most effec-
tively help them produce better crops and/or lower their production costs. In this re-
gard, evaluations should be object-oriented—how to improve the performance of the
specific system. A system evaluation should not only measure and record hydraulic
and other data (discharge, pressure, etc.), but should also consider how important per-
formance measures can be improved. Recommendations to the owner/operator for
changes in the irrigation system design, management, and operation are as important
as the data being collected.
There has been considerable material written on how to collect data on irrigation
performance. This has been addressed in other chapters and other books including
Merriam and Keller (1978), Merriam et al. (1983), Burt and Lehmkuhl (1991), and
Burt et al. (1997). Most terminology for defining irrigation system performance has
generally been agreed upon, although some terms are frequently misused. There are
numerous parameters that are used to describe farm irrigation system performance,
e.g., application rate, application efficiency, and distribution uniformity (as summa-
rized in Chapter 5). However, to adequately describe system performance, these meas-
ures need to be viewed in a larger context than simply involving one field and one
irrigation event. The methods and formulas by which performance measures can be
determined are also reasonably well defined and widely accepted, recognizing that
different irrigation methods entail difference procedures for evaluation.
792 Chapter 21 Evaluating Performance
The objective of this chapter is to present processes for evaluating the performance
of on-farm irrigation systems for the benefit of the owner/operator. It incorporates
agronomic as well as engineering factors to determine an irrigation system’s perform-
ance. It also addresses the practical aspects of evaluating a functioning irrigation sys-
tem within a production agriculture environment by understanding and considering the
operator’s labor, water supply, and other operational constraints in the analysis.
21.2 MANAGEMENT VARIABLES
The first step in evaluating an irrigation system is to define the management vari-
ables and the scope of the system. A farm irrigation system can be defined as a set of
physical and social elements employed to: (1) acquire water from a source (well, ditch,
pipeline, or turnout); (2) facilitate and control the movement of this water in and on a
defined area; and (3) disperse this water into the root zone of the crops being grown
(Small and Svendsen, 1992). Major variables used in defining the scope and operation
of the system are discussed in the following sections.
21.2.1 Water Supply
An important step in the evaluation of an irrigation system is to determine and un-
derstand the limitations of the water supply. There are physical, chemical, and tempo-
ral constraints associated with most agricultural water supplies, and in most water dis-
trict operations there are contractual limits to consider. When evaluating the system
performance, it is essential that the evaluator recognize the cause-and-effect relation-
ships between the water supply and operation of the system. All resulting recommen-
dations must consider these relationships.
In system evaluations, ingenuity is commonly needed to describe the system poten-
tial accurately when dealing with water supply constraints. For example, the owner/
operator can adjust the amount being applied in a 24-hour set without a change in
pressure and flow by adopting a spacing change in a hand move sprinkler system,
which results in a loss of distribution uniformity for a single irrigation but enables
recovering this loss through alternate sets. It is important that the operator understands
that the alternate sets must be moved a half set, or between the previous sets. This can
even improve the system’s overall distribution uniformity, as long as the field is not
being deficit irrigated. “Deficit irrigated,” as used here, suggests that the irrigation
event did not refill the root zone reservoir and that soil moisture depletion will remain
on completion of this irrigation.
21.2.2 Agronomic Constraints
Surprisingly, scheduling an irrigation is more often determined by cultural opera-
tions rather than the cultural operations being scheduled by the need for irrigation.
Stand establishment, weed control, maintenance of crop vigor, control of disease, and
seedbed preparation are all agronomic considerations that impact irrigation system
performance and need to be considered in an evaluation.
Westlands Water District was one of the sponsoring agencies for many of the on-
farm irrigation evaluations conducted in Central California (Burt and Katen, 1988).
The district required that all evaluations had to be performed under the direct supervi-
sion of a team consisting of an agronomist (or soil and water scientist), as well as an
engineer specializing in irrigation. This highlights the importance of including the
agronomy of the field in the evaluation.
21.2.3 Operations
Who makes the decision of when and how much to irrigate, and who executes this
decision, are questions that are often overlooked in evaluating an irrigation system.
The decision to irrigate is generally made by the owner/operator and is relayed to an
irrigator who executes the instruction. The irrigator is often a hired laborer who has a
basic understanding of the irrigation process, but no formal training in irrigation man-
agement methods that could be used to improve performance. Irrigation set time is
similarly driven by the availability of labor and water and not well-defined crop water
requirements. Water availability from the provider will also impact performance, i.e.,
whether it is an on-demand system or a rotational delivery.
Labor regulations in the U.S. have greatly reduced the availability of nighttime irri-
gators. It is common in a short-furrow field to irrigate two sets per day: one 10 hours
long and the other 14 hours long. The 10-h set is typically made over a specific num-
ber of furrows; the 14-h set is made over a significantly larger number of furrows.
Both sets commonly have the same flow rate and consequently will differ significantly
in application rate, application efficiency, distribution uniformity, and irrigation effi-
ciency (see Section 21.3) because of the change in area being irrigated and duration of
the set.
The effect of the field operation on the total system performance may be over-
looked if a serious effort is not expended in a pre-evaluation of the farm. Understand-
ing the owner/operator’s situation will pay big dividends in both how well the evalua-
tion is received and the quality of the evaluation.
21.2.4 Environmental Concerns
The relationship between irrigation system management and any underlying
groundwater resource is a significant environmental concern that should be recognized
in evaluating a farm irrigation system. A perched water table within 3 m of the soil
surface is one situation; a regional water table at a depth of 30 m under well-drained
soils is a totally different situation. This latter condition has created opportunities for
conjunctive use and groundwater recharge; what is one field’s loss may be another
field’s supply. Again, in a real-life situation, deep percolation, even greater than the
leaching requirement, can be beneficial when it is considered groundwater recharge
provided the water quality is not significantly degraded. In areas of intense irrigation
development, groundwater is often an integral part of the water resource. In these
situations, surface and groundwater resources should be operated and managed con-
junctively, and all system evaluations should recognize this interrelationship.
21.2.5 Structural Considerations
Two major agricultural suppliers of Colorado River water in Southern California
have opposing environmental restrictions on their water users. One district facilitates
tailwater leaving the farm; the other restricts the water user from discharging any sur-
face water from the farm. An evaluation needs to be conducted accordingly.
Both policies are based on environmental concerns, so it is not readily apparent
why these districts have such differing operating policies, particularly as both districts
ultimately discharge return flows into the same salt sink. The first district was formed
and designed 40 years before the second district and, as a result, may reflect the early
philosophies of the USBR that the value of water is more in having and using it as
opposed to using it beneficially and efficiently. This differing philosophy may also
have been brought about by increased competition for surface water supplies, specifi-
cally Colorado River water.
The water district established first, at the start of the 20th century, constructed a
deep, open drain next to each of its open delivery laterals on a spacing of 800 m to
facilitate removal of farm surface and subsurface drainage water. The other system,
developed in the 1940s, utilized pipelines to deliver water, and its drainage system was
designed to receive only subsurface farm drainage water. It is with some interest to
note that the last major irrigation scheme developed by the USBR, the San Luis Unit
of the Central Valley Project in the 1960s, was authorized without any drainage facili-
ties even though the drainage needs were recognized.
The previous examples emphasized that understanding how on-farm systems were
designed and operated in the context of district facilities and policy is essential to the
interpretation of the system management data. A water user discharging 30% of a field
supply to a salt sink is less than 70% efficient in water use, no matter how productive
and uniformly the remaining water is used. Conversely, to fault or recommend change
to the irrigation management without understanding the constraints on the system and
operating policies is far more questionable.
21.2.6 Economic Considerations
Energy, labor, hardware, and water costs will greatly influence the management of
an irrigation system while most evaluations focus on the hydraulics and hydrology of
an irrigation event. It is important to consider the trade-offs between associated costs:
less labor and more hardware, or vice versa. Total cost per unit of production is the
real definition of economic efficiency, but this is usually beyond the scope of on-farm
irrigation system evaluations. A multi-year study is required to determine the impact
of irrigation operations on agricultural production and is generally not undertaken be-
cause of the cost and time involved.
21.3 MEASURES OF PERFORMANCE
At the core of each evaluation are the system’s performance measures. When dis-
cussing performance, it is valuable to recognize that every irrigation system has a life
cycle and an event performance cycle. Some of the most valuable data produced for
the owner/operator are changes within the system’s life cycle. The performance cycle
can be short term and typically happens within an irrigation event. It can be as simple
as a pressure fluctuation due to back-flushing, or sequencing between different irriga-
tion blocks. The instantaneous performance of a system is almost always affected by
hydraulics, filtration, and elevation. Probably the most difficult performance changes
to evaluate are those associated with seasonal changes, such as the infiltration rate of
soils being furrow irrigated.
There is a longer-term cycle associated with the performance life of the system’s
hardware components, such as emission devices or pumps. These are subtle changes,
typically in hydraulics, which occur over long times due to corrosion and erosion
within high-pressure systems, such as erosion of a pump’s impeller and sprinkler noz-
zles by sand in water. Similarly, without careful and frequent system maintenance, a
microirrigation system using water with high levels of dissolved minerals can have
emitter performance influenced by precipitation of the minerals. Both the short- and
long-term cycles should be recognized and accounted for in the data analysis.
The indirect benefits associated with field losses are also difficult to quantify.
These primarily include the water losses associated with salt management, deep perco-
lation, and/or surface runoff. When these losses contribute to other irrigation supplies
or to groundwater recharge, they need to be recognized for these secondary benefits in
the evaluation of performance. For example, a post-season irrigation for weed germi-
nation may be considered as beneficial use. This irrigation may also contribute to sa-
linity management so the volume actually used for weed germination is not easily
quantified.
Multiple evaluations are required to quantify changes in system performance; un-
fortunately, most evaluation procedures are designed around a single event. The Cali-
fornia Mobile Lab Program originally specified one evaluation per owner/operator.
These single evaluations can develop all of the essential performance measures, but
they produce only a snapshot of the system’s total performance, and more importantly,
have the potential to misrepresent the system’s overall performance. The value of mul-
tiple evaluations cannot be over emphasized, particularly where only a few farm/field
evaluations are used to represent entire irrigation districts and agricultural water use in
general.
The measures most valuable to document a system’s performance are application
rate, application efficiency, distribution uniformity, operational characteristics, and
irrigation efficiency. These measures are listed in order of their value to the owner/
operator.
21.3.1 Application Rate
The system application rate is basic to all irrigation management decisions and it is
the value that all water managers must know, regardless of how they schedule their
irrigations. It is the rate water is applied to a given area and is usually expressed as a
depth per unit time. Knowing the application rate, the farmer can then determine the
required set time to apply a designated amount. The application rate does not involve
efficiency or uniformity.
21.3.2 Distribution Uniformity
Distribution uniformity is used to indicate the system’s distribution problems and is
colloquially often used interchangeably with the terms emission uniformity or Chris-
tensen’s Uniformity Coefficient (see also Chapter 5, Section 5.3.7). One common cal-
culation for distribution uniformity, DU, is
Dlq
DU = (21.1)
D av
where Dlq = the average depth infiltrated in the lowest catches for one-quarter of the
field
Dav = the average depth infiltrated over the entire field.
There are other expressions used to define the evenness of irrigation application, but,
unfortunately, as many as there are, there is not a perfect method of determining uni-
formity. Economy of time and effort limits how many measurements can be made in
any one evaluation.
The real value of the distribution uniformity calculation is that it focuses on the
system as opposed to management. How a system is operated will influence an irriga-
tion’s distribution uniformity, but it is strongly influenced by the system’s design and
hardware.
An irrigation system can have high uniformity and be inefficient through over-
irrigation, or have low uniformity and be 100% efficient through deficit irrigation.
Another consideration is the example of hand move sprinklers having a low measured
distribution uniformity for a single event. By splitting the move on alternate sets and
integrating the system’s application patterns, the distribution uniformity for two se-
quential events can become quite high—but the evaluators have to measure the effects
of two separate events.
21.3.3 Application Efficiency
The application efficiency is the performance measure that has the most confusion
surrounding it. The application efficiency, defined in Chapter 5 as Equation 5.2, in-
corporates management decisions in the evaluation. The definition is:
Vs
ea = (21.2)
Vf
where ea = the water application efficiency
Vs = the volume of irrigation water stored for evapotranspiration by the crop
The term is commonly misused as the system’s irrigation efficiency. A less common
but far greater mistake is to confuse this value with the system’s distribution uniform-
ity.
21.3.4 Irrigation Efficiency
Irrigation efficiency is seemingly the performance measure that is the focus of eve-
ryone except the owners/operators. It is not that they do not care how inefficient their
system might be; it is just that it is very difficult to convince them that some amount,
any amount, of an irrigation is not beneficial to their crop. Irrigation efficiency previ-
ously expressed as Equation 5.4 is defined as:
Vb
ei = (21.3)
Vf
where ei = the irrigation efficiency
Vb = the volume of water beneficially used
Vf = the volume of water delivered to the field.
The realistic aspect to this calculated value is that it is more often misinformation
than information. It really has little value, unless an effort is made to identify what is
represented by a statement of efficiency. Because of these interpretive problems, a
major water district in California has expanded considerable effort and money in
forming and promoting the three definitions listed below (Westlands Water District,
1989a).
Pre-plant irrigation efficiency is the efficiency of a pre-plant irrigation and is the
ratio of the sum of the depth of water used for soil water replacement (SMR) and cul-
tural practices (CP) to the depth of applied water (AW). No leaching requirement is
included, but there will be a leaching benefit derived particularly in the top portion of
the soil profile. Pre-plant irrigation efficiency (PIE) is calculated as:
SMR1 + CP1
PIE = (21.4)
AW 1
Seasonal irrigation efficiency integrates the efficiency of one or more regular sea-
sonal on-farm irrigations. It is the ratio of the sum of soil moisture replacement water
and water used for cultural practices for each irrigation after the pre-plant irrigation to
the sum of water applied during these irrigations. No leaching requirement is included.
Seasonal irrigation efficiency (SIE) can be determined by:
n SMRi + CPi
SIE = ∑ (21.5)
i =2 AWi
Annual irrigation efficiency is used to calculate the efficiency of all on-farm irriga-
tions and is the ratio of the sum of the soil water replacement water and water used for
cultural practices for all irrigations, plus the water to satisfy the seasonal leaching re-
quirement (Lr), to the sum of the depths of water applied during all irrigations, includ-
ing the pre-plant irrigation. The annual irrigation efficiency (AIE) is:
n
SMRi + CPi + Lr
AIE = ∑ (21.6)
i =1 AWi
Another critical aspect of irrigation efficiency calculations is that the interpretation

can easily become flawed because of the mathematics. When early-season irrigations
are excessive, as they commonly are, and late-season irrigations do not refill the soil
reservoir and some crop water stress commonly occurs, their sums can be misleading.
Arithmetically, a minus and a plus can cancel each other. The problem is that plants
are not mathematicians and early season excesses do not compensate for late season
shortages. This is particularly true when full crop ET is used to determine what water
is potentially beneficially used. It is valuable to understand that no irrigation event or
series of events are 100% efficient in real life, but water applied under a deficit irriga-
tion condition is normally used beneficially.
It is imperative to obtain sufficient data on farm irrigation performance to deter-
mine a farm’s irrigation efficiency, a water district’s water use efficiency, or to make
an expression of the overall efficiency of agricultural water use. Without these data
and a well-founded understanding of what they represent, reported values are often
questionable, subjective and, in some cases, incorrect and misleading.
21.3.5 Field Characteristics
There are a number of measurements and field assessments that are important to
every evaluation of farm irrigation, including:
crop growth stage soil water status
root zone depth crop evapotranspiration
soil water-holding capacity field dimensions
These data are seemingly minor and easy to obtain in the total process of evaluating an
irrigation system, but unless a concerted effort is made by the evaluator to determine
each one, errors will commonly occur in the evaluation. Be assured, these field charac-
teristics are essential to the value and ultimate interpretation of irrigation performance.
It is also paramount that these values are included in the final evaluation report. It not
only allows the owner/operator to assess his/her future irrigation practices, but it gives
the next evaluator a frame of reference for any future evaluations.
21.4 FIELD EVALUATION METHODS

This chapter is not intended to provide the detailed evaluation procedures that each
irrigation method requires. However, it is important to discuss the design and opera-
tional concepts and attributes unique and critical to the general types of irrigation
methods. To facilitate this, the many irrigation systems being used today have been
grouped into three basic irrigation categories: surface, sprinkler, and microirrigation.
21.4.1 Surface Irrigation
In 2002, surface irrigation was reported to be the method of choice on land irrigated
in the U.S. (Irrigation Journal, 1997). Worldwide, it is estimated by the Food and Ag-
riculture Organization of the United Nations (FAO) to be the method used on 73% of
irrigated lands (Pallas, 1993).
Despite its widespread use, surface irrigation has the most controversy surrounding
its evaluation. To begin with, it has been labeled as “inefficient” by those who do not
completely understand the method. Secondly, the procedures for evaluating this
method are distinctly different from procedures used for sprinkler and microirrigation
methods. Thirdly, this irrigation method has been relegated to a position of limited
economic value by the general irrigation industry because it involves little or no hard-
ware (aluminum, brass, or plastic) and its design is most often done by a practitioner
(land leveler).
Soil characteristics have a major influence on surface irrigation performance. Soils
often dictate a field’s physical configuration (slope and length) as well as its opera-
tional requirements (irrigation frequency and duration). These components are all part
of standard evaluation procedures. What typically is not included in the evaluation
process is the uniformity of a field’s slope and the homogeneity of its soils. These fac-
tors can be partially defined through pre-evaluation activities and by discussions with
the owner/operator.
The evaluation of surface irrigation systems is time dependent. Measuring advance
and recession rates of water flows at enough locations to be representative of the entire
field can be time consuming. However, it is critical where surface system performance
is being compared to sprinkler and microirrigation methods that the data samples also
be comparable.
21.4.2 Sprinkler Irrigation
Sprinkler irrigation (i.e., high-volume sprinklers having discharge rates of at least
4 L/min.) is the most studied, best understood, and most fully documented irrigation
system of the three major methods. Flow, pressure, and discharge are system charac-
teristics that are easily measured and quickly recorded. Computer interpretation of
catch-can test data has simplified many aspects of data collection and interpretation. A
primary concern in sprinkler irrigation is uniformity of application since wind can
distort sprinkler patterns. However, this problem is generally intermittent and almost
impossible to design or operate around without turning off the system under severe
wind conditions.
There are considerable amounts of test data on sprinkler operational characteristics,
which are available from manufacturers and independent testing agencies. These pub-
lished performance data will facilitate pre-evaluation planning and assist in evaluation
interpretation.
21.4.3 Microirrigation
This method of irrigation (including low-volume sprinklers of less than 4 L/min
discharge, as well as trickle irrigation, low-volume bubblers, etc.) is relatively new
and has been evolving at a fast pace. Recent developments in its use on row crops and
in buried configurations have created a need for new evaluation techniques. Most of
the current procedures are minor modifications of sprinkler evaluation techniques.
Microsprinklers are being used extensively in permanent plantings and the designs do
not call for overlap of wetted patterns or even a totally wetted surface. Most designs
use one or more emission devices per tree.
A potentially beneficial operational feature of a microirrigation system is to irrigate
daily, or even more frequently. High-frequency irrigation has resulted in improved
yields and improved water use efficiency in many crops, particularly those with high
water requirements. This mode of operation requires extensive automation and as a
result many growers have been slow to utilize this potential benefit. Microirrigation
systems have an added requirement for accurate and current evaluation data because
their management may involve replacing the crop’s daily water use. Good manage-
ment is predicated on having a thorough understanding of the system and accurate
quantification of the crop’s daily water use and feedback control of the system.
The microirrigation system’s critical design feature is its wetted soil volume. Early
design specifications suggested a minimum wetted soil volume of not less than 33% of
the total root zone volume (Keller and Karmeli, 1975). Practice and experience have
shown that 50% wetted volume is a better design and, with good management (daily
operation), 60% to 80% of the wetted soil volume is realistic. Percent of wetted vol-
ume, as used here, is the percent of the total root zone volume (for row crops and per-
manent crops) that is wetted by irrigation.
21.5 ECONOMICS
To influence changes in any commercial enterprise, there needs to be some finan-
cial reward or economic return and this is true for conservation of water by production
agriculture. The reality of this becomes obvious when it is understood that, in many
cases, water conservation has no tangible benefit to the owner/operator who is con-
serving the water. In most cases, it is believed that some inefficiency is actually the
owner/operator’s insurance policy for those periods of drought and restricted water
supplies. There is a thread of commonality in this belief with the appropriation doc-
trine of water law—“use it or lose it.”
The way to address this issue is by establishing the economics of improved water
management. In evaluating the performance of an irrigation system, the operational
costs and system hardware costs are, in reality, the only two areas of opportunity to
develop system costs. Real and tangible benefits from system evaluations are associ-
ated with the quantity and quality of crop production. Improved irrigation performance
should translate into enhanced quantity and/or quality of the crop’s yield. To accu-
rately quantify these benefits is generally not within the scope of evaluating a farm
irrigation system. However, they should be discussed by the evaluator with the
owner/operator, when appropriate.
21.5.1 Operational Costs
The operational costs of an irrigation system are best determined in the pre-
evaluation period. Typically, these operational costs are water, energy, labor (includes
management), agronomic practices (fertilizers and cultivations), and pest and disease
control.
The cost of water should reflect the highest rate, particularly when tiered pricing is
used. When using groundwater, it should include the cost of a well and its mainte-
nance and development cost. Energy costs should include pump and motor efficiency
as well as the plant’s operation and maintenance costs. When it is beyond the scope of
the evaluation to determine these data, it is reasonable to assign values and record
these as assigned or estimated values. They should not become an issue as long as the
owner/operator recognizes that they are assigned.
Quantifying the cost of labor is even a greater challenge. Highly skilled labor costs
more than less-skilled labor and the change in labor cost may reflect changes in labor.
What is critical for the evaluator to recognize is that most farming operations are
highly motivated to conserve labor and operators will gladly trade water conservation,
if they have it, for labor savings.
21.5.2 System Costs
System costs principally translate into the hardware costs, i.e., re-nozzling a sprin-
kler system, or cutting a furrow length in half by buying/leasing/renting additional
gated pipe. When suggesting alternative systems or system changes, the full and rea-
sonable cost should be identified. The critical element in developing these costs is to
be sure the evaluation is practical and realistic. If one element of the cost estimate
looks unrealistic or a component of the recommended change such as labor is over-
looked, the grower will quickly disregard the suggested change.
21.6 ANALYSIS AND INTERPRETATION
Equally important to data collection is the analysis and interpretation and this goes
back to the “Who for Whom” criteria mentioned earlier. Where the owner/operator has
hired a consultant or formally requested an independent service (NRCS or Cooperative
Extension) to perform the evaluation, this question has less significance. However,
when a third party is paying for this service, or an agency is out soliciting fields to
evaluate, this “Who for Whom” question becomes more meaningful.
The initial step is to determine if this will be an evaluation of a single-event per-
formance or seasonal performance. It is more valuable if the evaluation is for the sea-
sonal performance of the system and the system is evaluated at least two times in the
same season. A seasonal water budget can then be developed. This is particularly im-
portant if the system is a large-volume, low-frequency system (surface or sprinkler). If
it is a low-volume, high-frequency (micro) system, a single evaluation may suffice.
These latter systems are not as dependent on soils, and, by knowing the following, a
reasonable seasonal performance can be calculated:
seasonal crop water use distribution uniformity
seasonal effective precipitation total hours of operation (season)
application rate
To estimate seasonal performance from a single evaluation of a surface or sprinkler
system requires some operational data not typically maintained by the owner/operator
or routinely available. These data are the soil water status, before and after each irriga-
tion throughout the season, including the initial soil water content and any pre-plant
irrigation. Without understanding the impacts of irrigation on the root zone reservoir,
the seasonal efficiency of any irrigation system is difficult to assess. With high-
volume, low-frequency systems, the error can be great, particularly where the soil’
intake characteristics change during the season. It is common for early irrigations to be
excessive and late season irrigations to be deficient. If the crop is stressed at any time,
this will invalidate the evaluation. The accumulated effects are easily misinterpreted as
efficient, or, if only one event is evaluated, it would not be representative.
Conversely, it should be understood that carefully managed crop water stress can
be important to the grower’s agronomic objective. For example, deficit irrigation is
very valuable to viticulture and grape quality, and also plays a valuable role in achiev-
ing top cotton production. With some small grains, managed crop water stress during
rapid stem elongation can shorten the straw length and reduce lodging without loss of
yield or quality.
Water districts can undertake an effort to evaluate district-wide irrigation events
during an irrigation season, and, through random sampling of irrigation events, de-
velop a valid representation. But care needs to be exercised in how cooperators are
selected. There have been several instances where participants were selected on a first-
come, first-served basis from a district-wide mailing. A cursory assessment of these
participants might show that these were the progressive owners/operators in the dis-
trict. The district has to recognize these potential biases and make an effort to evaluate
systems of some of the owners/operators who take a less active and consistent interest
in their operations.
21.7 MOBILE LABS
The Mobile Lab concept developed in California has some valuable features. These
labs or system evaluation teams were started in 1985 through a cooperative effort by
the California Department of Water Resources and the USDA NRCS. Technical sup-
port was provided by researchers at California State University at San Luis Obispo
(Cal Poly). The program has been instrumental in developing and standardizing field
test procedures for field evaluations of on-farm irrigation systems. The Mobile Lab
approach has effectively demonstrated to production agriculture and the water re-
sources industry the value of accurate and detailed data on irrigation system perform-
ance. Before evaluating on-farm water use, Westlands Water District made a case that
their district-wide farm irrigation efficiency was over 80%. After collecting data on
335 fields over two irrigation seasons, their efficiency was calculated to be 77% and
these were arguably the District’s more progressive owners/operators (Westlands Wa-
ter District, 1989b).
There have been several criticisms of the Mobile Lab program. Since the Mobile
Lab provides its services to growers for free, this tends to devalue the concept. The
Mobile Lab also employs summer students and recent college graduates to perform the
field work. This creates a situation where often a person with little practical field irri-
gation experience is evaluating a system being operated by someone with considerable
experience. The evaluator is also there by courtesy of the owner, and if the evaluator
wants to come back in the future to retest this system or another system, the evaluator
better not be too critical. The evaluations developed in these situations seldom fully
address the management problems.
21.8 OUTLINE FOR EVALUATIONS

21.8.1 Pre-Evaluation Interview with the Grower
The following information should be collected during an interview with the grower
prior to the field evaluation:
Operational information: How are irrigations scheduled? What is this event’s
objective? What is the fertilizer program? What is the cropping pattern, its his-
tory and future plans? What is the labor/work force situation? Are there any re-
cent data on soil salinity or irrigation water quality?
Field physical information, including a map showing water source and field di-
mensions, the row/furrow spacing (for furrow irrigation), the slope of the field,
and information about the system features (pumps, filters, etc.).
Crop information, including the date planted/age/emergence, date to be har-
vested or terminated, the depth of the root zone (actual or managed), and annual
estimated crop ET.
Economic information, including the cost of water, energy, and labor (including
management).
21.8.2 Collection of Field Data
Each type of system should have its own data sheet. These are available from most
manuals or how-to books. These forms should include the date, start and finish times,
name of evaluator, and weather conditions. The field map should be attached to the
data sheet and noted to reflect where readings and evaluations were taken (Merriam
and Keller, 1978; Burt and Lehmkuhl, 1991).
21.8.3 Report
The report should contain at least a procedures section, a results section, and a rec-
ommendations section. The results section should include a summary report along
with a copy of the actual field data. As a minimum, the report should record the fol-
lowing field characteristics and how they were determined: crop, root zone depth, soil
water-holding capacity, soil water status (starting and finish), and evapotranspiration
of crop to date. Also include the following measures of performance: application rate,
distribution uniformity, application efficiency, and irrigation efficiency.
21.9 CONCLUSIONS
A skillfully executed evaluation of on-farm irrigation system performance requires
more than simply measuring the hydraulic performance of the system. It requires an
evaluation of the entire system including the agronomic and social systems affecting
the operation. The evaluation should be done by either a qualified and experienced
professional or an individual operating under the supervision of an experienced pro-
fessional. There is no irrigation system for which a thorough and complete evaluation
will not identify one or more opportunities to improve the system or its performance.
However, these improvements are of little or no consequence to the owner/operator if
they only address the hydraulics and system performance and do not include the effect
major enhancements of irrigation performance will have on crop performance and real
economic returns. The effectiveness with which these opportunities are presented to
the operator/owner will determine how valuable the evaluation is to them.
REFERENCES
Burt, C. M., and K. Katen. 1988. Westside Resource Conservation District 1986/87
Water Conservation and Drainage Reduction Program. San Luis Obispo, Calif.:
Cal Poly.
Burt, C. M., and M. Lehmkuhl. 1991. Irrigation System Evaluation Manual. San Luis
Obispo, Calif.: Cal Poly.
FWR (Department of Water Resources). July 1996. Special 16th Anniversary Edition:
Water Conservation: Yesterday, Today, and Tomorrow. Sacramento, Calif.: Water
Conservation News.
Irrigation Journal. 1997. 1996 Annual irrigation survey reveals growth throughout
country. Irrigation J. Jan./Feb.: 27-42.
Keller, J., and D. Karmeli. 1975. Trickle Irrigation Design. Glendora, Calif.
Merriam, J. L., and J. Keller. 1978. Farm Irrigation System Evaluation: A guide for
Management. 3rd ed. Logan, Utah: Utah State Univ.
Merriam, J. L., M. N. Shearer, and C. M. Burt. 1983. Evaluating irrigation systems
and practices. In Design and Operation of Farm Irrigation Systems, 721-760. M. E.
Jensen, ed. St. Joseph, Mich.: ASAE.
Pallas, P. 1993. Water and sustainable agricultural development: The role of planning
and design of irrigation systems. In Proc. Trans. 15th Congr. Int’l Comm. on Irrig.
and Drain., lJ: 53-71. The Hague, Netherlands: ICID.
Small, L. E., and M. Svendsen. 1992. A Framework for Assessing Irrigation
Performance. Washington, D.C.: International Food Policy Research Institute.
Westlands Water District. 1989a. Water Conservation and Drainage Reduction
Programs 1987-1988. Fresno, Calif.: Westlands Water District.
Westlands Water District. 1989b. Summary Water Conservation and Drainage
Reduction Programs 1987-1988. Fresno, Calif.: Westlands Water District.
APPENDIX A
GLOSSARY
E. Gordon Kruse (USDA-ARS,
Ronald L. Elliott (Oklahoma State
University, Stillwater, Oklahoma)
Advance phase: A period, during surface irrigation, in which the water moves out
over the field from the point of inflow. Normally, advance ends when the stream
front encounters a field boundary.
Advance time: Time required for a given stream of irrigation water to move from the
upper end of a field to the lower end.
Aerosolization: Conversion of water to mist or fine droplets; from sprinklers, spray-
ers, or wind blowing over free water surfaces; that are suspended in air. Any con-
taminants in the water source may be transported with aerosol drift.
Affinity laws: Three equations that relate how the pump discharge rate, pressure, and
input power vary with change in pump speed, which is useful to estimate change in
pump performance with small changes in speed.
Agricultural weather networks: Arrays of weather stations, each of which may
measure temperatures, humidity, wind, solar radiation and other weather phenom-
ena. Each station records or transmits the measured values to a central site, perhaps
a state agency or water district. The data are commonly processed (summed, aver-
aged, checked for error) before they are made available to users for irrigation
scheduling and other purposes.
Agrochemicals: A broad range of products intended to enhance agricultural produc-
tion; includes fertilizers, pesticides, soil amendments, etc.
Air counterflow: Upward movement of air through soil during downward flow (per-
colation) of water.
Air drainage: (Soil) Renewal of soil air by diffusion.
Air entrapment: A soil condition that prevents complete saturation of soil pores by
water, because gases in the soil have no way to escape.
Air lift pump: A pump, used chiefly in wells, that introduces air under pressure to
water at the bottom of the well. This reduces the specific gravity of the air-water
mixture, which is caused by the pressure of surrounding, denser, water to rise to the
outlet (adapted from ICID MTD, 1996).
Air relief valve: (1) A device, commonly installed at the high points in a water pipe-
line, that allows air to escape the line as it is being filled, thus removing a restric-
tion to water flow. (2) A device installed on the high pressure side of the chemical
injection pump, and immediately upstream of the chemical injection line check
valve, to relieve trapped air and pressure that might otherwise affect the calibrated
injection rate or cause chemical discharge on personnel servicing the system.
Air-entry head: The negative pressure that must be exceeded for air to enter a satu-
rated soil.
Albedo: The ratio of electromagnetic radiation reflected from a soil and crop surface
to the amount incident upon it. In practice, the value is applied primarily to solar
radiation (from ASCE, 1990).
Alfalfa valve: An outlet valve attached to the top of a surface irrigation pipeline riser,
having an opening equal in diameter to the inside diameter of the riser pipe and an
adjustable lid or cover to control water flow. The circumference of the valve frame
is machined to provide a seat and seal for a portable hydrant, which may be at-
tached to direct water into a pipeline above ground.
Alkali soil: See Sodic soil.
Alternate set irrigation: A method of managing irrigation whereby, at every other
irrigation, alternate furrows are irrigated, or sprinklers are placed midway between
their locations during the previous irrigation.
Alternate side irrigation: The practice of furrow irrigating one side of a crop row
(for row crops or orchards) and then, at about half the irrigation time, irrigating the
other side. When used with deficit drip irrigation management, this practice is re-
ferred to as partial root zone drying.
Anisotropic soils: Soils whose physical properties vary with the direction of meas-
urement, especially soils with different permeabilities in different directions of
measurement.
Annual irrigation efficiency: The ratio of soil water replacement, plus water used for
cultural practices for all irrigations, plus the water used to satisfy the seasonal
leaching requirement; to the total water applied during all irrigations, including the
pre-plant irrigation.
Application efficiency: The ratio of the average depth of irrigation water infiltrated
and stored in the root zone to the average depth of irrigation water applied, ex-
pressed as a percentage.
Application rate: Rate that water is applied to a given area; usually expressed in units
of depth per time.
Aquiclude: Underground geologic formation that neither yields nor allows the pas-
sage of an appreciable quantity of water, although it may be saturated with water it-
self.
Aquifer: A geologic formation that holds and yields useable amounts of water. Aqui-
fers can be classified as confined (an aquifer whose upper, and perhaps lower,
boundary is defined by a layer of natural material that does not transmit water read-
ily) or unconfined (having no such confinement).
Aquitard: Underground geologic formation that is slightly permeable and yields in-
appreciable amounts of water as compared to an aquifer.
Arid climate: Climate characterized by low rainfall and high evaporation potential. A
region is usually considered as arid when precipitation averages less than 250 mm
(10 in.) per year.
Artesian aquifer: Aquifer that contains water under pressure as a result of hydrostatic
806 Appendix A Glossary
head. For artesian conditions to exist, an aquifer must be overlain by a confining
material or aquiclude and receive a supply of water. The free water surface stands
at a higher elevation than the top confining layer.
Atmosphere percentage: See One-third atmosphere percentage.
Atmospheric vacuum breaker: A valve that automatically opens, causing air to enter
a pipeline and stop backsiphonage, when a reversal or pressure gradient occurs that
could otherwise cause flow to siphon back toward the source. See also Pressure
vacuum breaker.
Available soil water: The portion of water in a soil that can be readily absorbed by
plant roots. It is the amount of water released between in situ field capacity and the
permanent wilting point.
Axial-flow pump: A rotary pump that develops head mostly by the propelling or lift-
ing action of the vanes on water; commonly referred to as a propeller pump.
Backflow prevention device: Safety device that prevents the flow of water from the
water distribution system back to the water source.
Backsiphonage: Flow of liquid in a direction opposite to the normal operating flow
condition. Caused by a reversal of normal pressure gradient. In chemigation sys-
tems, backsiphonage can cause contamination of the irrigation water supply.
Banding (of nutrients): The application of nutrients in narrow strips parallel to crop
rows. Proper location of the bands, relative to the crop, can minimize leaching by
irrigation and reduce the amount of fertilizer needed.
Basic intake rate: The rate at which water percolates into soil after infiltration has
decreased from an initial, higher rate to a nearly constant value.
Basin furrows: See Beds-in-basins.
Basin irrigation: Irrigation by flooding areas of level land surrounded by dikes. Used
interchangeably with level border irrigation, but usually refers to smaller areas.
Basin tillage: Specialized tillage to create a series of micro-catchments between crop
rows for retention of water under center pivots and lateral move irrigation systems.
Basin tillage is usually required under the outer spans of center pivot LEPA sys-
tems where application rates exceed intake rates. Also called reservoir tillage.
Beds (Bedding): (1) Preparation of furrow irrigated, row cropped field with wide,
flattened ridges between the furrows on which one or more crop rows are planted.
(2) The process of laying a pipe or other conduit in a trench with the bottom shaped
to the contour of the conduit; tamping earth around a conduit to form its bed.
Beds-in-basins: Ridges on a level basin on which crops are planted. During irrigation,
the water surface in the channels between the ridges remains below the ridge sur-
face. The channels are called dead-level furrows or basin furrows.
Bench flume: A water-conducting channel built on constructed terraces along hill-
sides or around mountain slopes when the ground is too rough, steep, or rocky to
permit an excavated canal.
Beneficial use of water: A use of water resulting in appreciable gain or benefit to the
user, consistent with state laws, which vary from state to state. Beneficial uses in-
clude satisfying the soil water deficit and leaching of salts from the root zone.
Berm: A strip or area of land, usually level, between the edge of a spoil bank and the
edge of a ditch or canal.
Best management practice (BMP): Any of several structural, nonstructural, and
managerial techniques recognized to be the most effective and practical means to

reduce surface and groundwater contamination while still allowing the productive
use of resources.
Bioaccumulator: A crop capable of biofiltration.
Biofiltration: The culture of specific plants that have the ability to accumulate large
quantities of undesirable constituents (e.g. selenium, nitrate, boron, molybdenum)
from soil water. Potentially toxic constituents may therefore be removed by har-
vesting and disposing of the harvested matter. This management practice is most at-
tractive where drainage disposal is an environmental concern, the bioaccumulator
has economic value, or other treatment processes are unavailable or too expensive.
Biological oxygen demand (BOD): The amount of dissolved oxygen required to meet
the metabolic needs of microorganisms that degrade carbonaceous organic material
in a water sample.
Black alkali soil: A soil with a high pH and chemistry dominated by sodium carbon-
ate. Dissolved organic matter may be deposited on the soil surface as soil water
evaporates, causing a dark color. (Preferred term is saline-sodic soil.)
Black water: Water containing liquid and solid human body waste.
Booster pump: A pump, in a pipeline, used to increase the fluid pressure. In center
pivot sprinklers, may be located near the end of the lateral to provide adequate
pressure for a high volume sprinkler head.
Border dike: An earth ridge or small levee built to guide or to hold irrigation or re-
charge water in a field.
Border ditch: A small excavated channel for irrigation, typically running downslope
at the border of an irrigated strip or plot. Temporary dams in the ditch cause water
to be spread from one or both sides onto the land to be irrigated.
Border irrigation: Irrigation by flooding strips of land, rectangular in shape and cross
leveled, bordered by dikes. Water is applied at a rate sufficient to move it down the
strip in a uniform sheet. Border strips having no downfield slope are referred to as
level border systems. Border systems constructed on terraced lands are commonly
referred to as benched borders.
Bubbler irrigation: The application of water to flood the soil surface using a low-
pressure, small stream or fountain. The discharge rates for point-source bubbler
emitters are greater than for drip or subsurface emitters but generally less than
225 L/h (1 gpm). A small basin is often required to contain or control the water.
Bulk density: (Soil) The mass of dry soil per unit bulk volume. The bulk volume is
determined before the soil is dried to a constant mass at 105°C (220°F).
Cablegation: A surface irrigation technique in which a gradually decreasing inflow is
provided to each furrow of a set, by using a cable to pull a plug through a gated
pipe.
Capillary barrier effect: A restriction to the downward movement of soil water,
caused by a fine-textured soil layer overlying a coarser-textured layer.
Capillary fringe: A zone in the soil just above the water table that remains saturated
or nearly saturated. The vertical extent depends upon the size distribution of pores.
Capillary pressure head (Capillary rise): The height that water will rise by surface
tension above a free water surface in the soil.
Capillary tension (Capillary suction): The negative pressure that exists when the
adhesion of liquid molecules to soil particles is stronger than the cohesive forces
between the molecules. This condition causes the water to rise above a saturated
zone, creating a capillary fringe. The height to which water will rise by capillary
action is limited by surface tension and the soil pore size distribution.
Capillary zone: See Capillary fringe.
Capped riser: An irrigation pipeline riser extending above ground, with a watertight
cap over its top and outlet gates on its sides slightly above the ground surface. To
accommodate more outlet gates, a cap with a diameter larger than that of the riser
pipe is sometimes installed on the top of the riser, then known as a capped pot.
Catchment area: The area from which a lake, waterway, or reservoir receives surface
flow that originated as precipitation; a watershed (from ICID MTD, 1996).
Centrifugal pump: A pump consisting of rotating vanes (an impeller) enclosed in a
housing and used to impart energy to a fluid through centrifugal force.
Centrifugal separator: A device that removes solid particles from water, as for mi-
croirrigation use, by causing the water to spin in a chamber. The denser solids are
forced to the chamber’s boundary where they can be flushed away.
Check irrigation: Modification of a border strip with small earth ridges or checks,
constructed at intervals, to retain water as the water flows down the strip.
Check: (1) Noun. A structure to control water depth in a canal, ditch, or irrigated field.
(2) Verb. To control water depth by using some device to impede the water’s flow.
Chemical oxygen demand (COD): A quantitative measure of the amount of oxygen
required for the chemical oxidation of organic carbonaceous material; an indicator
of water quality.
Chemigation: Application of chemicals to crops through an irrigation system by mix-
ing them with the irrigation water.
Christiansen’s uniformity coefficient: A measure of the uniformity of irrigation wa-
ter application. Defined as the average depth of irrigation water infiltrated minus
the average absolute deviation from this depth, all divided by the average depth in-
filtrated.
Claypan: A dense, compact layer in the subsoil having a much higher clay content
than the overlying material, separated by a sharply defined boundary. Claypans are
usually hard when dry, and plastic and sticky when wet. They typically impede the
movement of water and air, and the growth of plant roots.
Colloids: Negatively charged soil particles smaller than 1 µm in diameter.
Column separation: A condition in pipelines that occurs at isolated points when the
water pressure drops below the vapor pressure, resulting in the water vaporizing,
separating the sections of fluid.
Combination air valve: A valve to be used at the high points of a water pipeline for
release of air and prevention of vacuum, having operating features of both air and
vacuum valves and air release valves.
Cone of depression or influence: The water table or piezometric surface, roughly
conical in shape, produced by the extraction of groundwater from a well.
Conjunctive use: (1) Of saline irrigation water: The beneficial use of saline water for
irrigation by blending or seasonal cycling with a source of water of low salinity. (2)
Of water supply: The combining of water from ground- and surface-water sources,
especially for irrigation of a farm or a project.
Consumptive use: The total amount of water taken up by vegetation for transpiration
or building of plant tissue, plus the unavoidable evaporation of soil moisture, snow,
and intercepted precipitation associated with vegetal growth.
Continuous-flow irrigation: A system of irrigation water delivery where each irriga-
tor receives the allotted flow of water continuously.
Contour furrows: Furrows plowed approximately on the field surface contour to re-
duce soil loss and increase infiltration.
Control head: All of the pumps, valving, filters, injectors, controllers, monitoring
equipment and other facilities required to deliver water at sufficient pressures and
appropriate quantity and quality to an irrigation (especially microirrigation) system.
Head as used in this term does not imply water pressure.
Control station: (Microirrigation) Facilities upstream of microirrigation distribution
and application piping for purposes that may include water measurement, filtration,
treatment, flow and pressure control, timing of application, and/or backflow pre-
vention. Sometimes called a control head.
Controlled deficit irrigation: An irrigation strategy in which less water is applied
late in the growing season than is being consumptively used.
Controlled drainage: Regulation of the water table by means of pumps, control
dams, or check drains, or a combination of these, for maintaining the water table at
a depth favorable to crop growth.
Conveyance efficiency: The ratio of the volume of water delivered to the field bound-
ary to the volume of water diverted from the source.
Conveyance loss: Loss of water from a channel or pipe during transport, including
losses due to seepage, leakage, evaporation, and transpiration by plants growing in
or near the channel.
Corner span: See Corner system.
Corner system: A type of center pivot sprinkler system designed to increase the irri-
gated area on a square field. Consists of a corner span attached to the end of a typi-
cal center pivot lateral that revolves about the end of the lateral in the field corners,
but is tucked behind the main lateral otherwise.
Corrugated plastic drain tubing (pipe): Tubing, circular in cross section, usually
made from polyethylene, with continuous, deep corrugations in the walls, such that,
although lightweight, it has high resistance to crushing. Available in a wide range
of diameters. Can be perforated in the factory, to allow water entry.
Coupler: (Sprinkler) A device, either self-sealing or mechanically sealed, that con-
nects two lengths of pipe or pipe to hose.
Critical suction head: See Net positive suction head.
Crop area: (Irrigation) The field surface area allocated to each plant. The crop area is
the plant spacing multiplied by the row spacing.
Crop coefficient: A factor for modifying estimates of reference crop evapotranspira-
tion (ET) to derive an estimate of ET for some other (non-reference) crop.
Crop irrigation requirement: The quantity of water, exclusive of effective precipita-
tion, that is needed for crop production, while avoiding water stress on the crop.
Cross: (Pressurized irrigation systems) A pipe fitting with 4 outlets or connections,
each 90° apart.
Cut: The portion of the land surface from which earth or rock has been removed or
will be removed by excavation; the depth below the original ground surface to the
excavated surface.
Cut-and-fill: Process of earthmoving by excavating part of an area and depositing the
excavated material on other areas or using it for embankments.
Cutback irrigation: The reduction of a furrow or border inflow stream after water has
advanced partially or completely through the field, in order to reduce runoff.
Darcian flow: Flow of a viscous fluid in isotropic porous media, proportional to and
in the direction of decreasing hydraulic gradient (adapted from ICID MTD, 1996).
Dead-level furrows: See Beds-in-basins.
Deep percolation: Water that moves downward through the soil profile below the
root zone and cannot be used by plants.
Deficit irrigation: Management of irrigation applications to apply less water than
would satisfy the soil water deficiency in the entire root zone, often during specific
crop growth stages. Sometimes called limited, stress irrigation, regulated deficit ir-
rigation, or partial root zone drying.
Delivery box: (Irrigation) A structure used to divert water from a canal to a farm, of-
ten including a measuring device. Also called a turnout.
Demand irrigation system: An irrigation water delivery policy whereby each irriga-
tor supplied by a district canal or pipeline may request irrigation water in the
amount needed and at the time desired.
Denitrification: The complete reduction of nitrates to atmospheric nitrogen and ox-
ides of nitrogen (from ICID MTD, 1996).
Depletion phase: The period, in surface irrigation, between the soaking and recession
phases, during which the water surface drops.
Depression storage: See Surface storage.
Design application efficiency: The computed water application efficiency when wa-
ter is applied at design rate and design time. See Water application efficiency.
Design emission uniformity: A measure of the uniformity with which irrigation water
is discharged from individual emitters by a microirrigation system. A calculated
measure of how uniformly an overall system design can distribute water from each
emission device in the field. Often confused with emission uniformity, which is the
result of actual field measurements of discharge from emission devices.
Desilting area: See Filter strip.
Desorption curve: A graphical description of empirically derived relationships be-
tween water content, suction, and hydraulic conductivity of a soil. Also referred to
as a water retention curve or water release curve.
Dimension ratio: The ratio of the average pipe diameter to the minimum wall thick-
ness of plastic pipe. The pipe diameter may be outside or inside diameter, depend-
ing on the type of plastic.
Disk filter: A small, high-capacity filter used in microirrigation systems that utilizes
grooved, annular, plastic, or metal epoxy-coated disks or wafers stacked tightly in-
side a cylinder. The filters are generally manifolded together and have a capacity of
about 1 L/s per filter. Often known as a wafer filter.
Distributary: (1) A small conduit or channel taking water from a canal for delivery to
farms. (2) Any system of secondary conduits. (3) A river channel flowing away
from the main stream and not rejoining it, in contrast to a tributary.
Distribution system: A system of ditches, or conduits and their controls, that conveys
water from the supply canal to headgates of farms served by the canal.
Distribution uniformity of low quarter: The ratio of the average of the lowest one-
fourth of measurements of irrigation water infiltrated to the average depth of irriga-
tion water infiltrated, expressed as a percentage.
Distribution uniformity: A measure of the uniformity of irrigation water distribution
over a field.
Diversion box: A structure built into a canal for dividing the water into predetermined
portions and diverting it to other canals.
Diversion dam: A barrier built in or across a stream for the purpose of diverting part
or all of the water from the stream into a canal.
Doctrine of prior appropriation: A legal doctrine of water rights that asserts that all
claims to water are based on beneficial use.
Double suction pump: A horizontal centrifugal pump having two suction inlets, one
on either side of the impeller.
Drainable porosity: See Specific yield.
Drainage coefficient: The design value at which shallow groundwater is to be re-
moved from an area to be drained, often expressed in units of depth per day.
Drainage: The process of removing surface or subsurface water from a soil or area.
Drain-out period: The time of water table recession during a drainage cycle.
Drift: The movement of water droplets in air by wind, especially the redistribution of
water drops from sprinklers or field sprayers by wind. Often adversely affects irri-
gation water distribution or causes irrigation to wet unintended areas. Can be haz-
ardous if irrigation water droplets containing pesticides reach areas outside the field
being irrigated.
Drip irrigation: A method of microirrigation wherein water is applied to the soil sur-
face as drops or small streams through emitters. Discharge rates are generally less
than 8 L/h (2 gal/h) for single-outlet emitters and 12 L/h (3 gal/h) per meter for
line-source emitters.
Drip tapes: Collapsible, thin-walled, low-pressure polyethylene tubes with built-in
emitters or orifices.
Drop size distribution: The gradation of diameters of water droplets, either as emit-
ted from an operating sprinkler, or as received at the soil surface at a given distance
from the sprinkler.
Drop structure: A hydraulic structure for safely transferring water in a channel to a
lower level channel without causing erosion.
Drop tube: A small-diameter, flexible tube, located at intervals along a center pivot or
traveler lateral to transmit water to lower-elevation water emission devices. Typi-
cally connected to the supply pipe by a small, U-shaped conduit called a gooseneck.
Dryland farming: Crop production without irrigation. Referred to as rainfed agricul-
ture when the annual precipitation exceeds consumptive use.
Duty of water: See Crop irrigation requirement.
Dynamic equilibrium: (Water table) A situation where the annual range of water
table fluctuations is relatively consistent from year to year.
Ebb-flow systems: An irrigation system used with potted plants in which an impervi-
ous basin is periodically flooded and then drained. The drainage water is used for
irrigation of the next basin.
Economic efficiency: The maximum economic value derived from the use of water,
commonly expressed per unit volume of water consumed rather than per unit di-
verted or applied, so as to properly account for return flows.
Eddy covariance: A statistical measure used in the analysis of high frequency wind
and scalar atmospheric data series to obtain values of fluxes of water vapor, carbon
dioxide, and sensible heat, representing relatively large areas.
Effective full cover: A stage in crop phenology that begins at the end of the crop de-
velopment period. Can be variously defined as (a) the time when some leaves of
plants in adjacent rows begin to intermingle so that soil shading becomes nearly
complete, (b) when plants reach nearly full size, (c) when the average fraction of
the ground surface covered by vegetation is about 0.7 to 0.8, (d) the initiation of
flowering, or (e) when the leaf area index reaches a value of about 3. The crop can
continue to grow in both height and leaf area after the attainment of effective full
cover.
Effective precipitation: That portion of total precipitation that becomes available for
plant growth.
Effluent irrigation: Land application of wastewater for irrigation and beneficial use
of the nutrients it contains.
Electrical conductivity: A measure of the ability of water to conduct electricity; used
to estimate the amount of soluble salts in irrigation or drainage water, or the solu-
tion extract of a soil.
Electrical resistance block: A small block consisting of electrodes set in an absorbent
material, such as gypsum, used to estimate soil water content.
Elevation head: Energy possessed by a fluid as a result of its position above some
datum.
Emission point: The location where water is discharged from an emitter.
Emission uniformity: An index of the uniformity of emitter discharge rates through-
out a microirrigation system. Takes account of both variations in emitters and
variations in the pressure under which they operate.
Emitter: A device through which water is discharged from an irrigation system onto a
field surface. May be any of the following types.
Flushing emitter: An emitter designed to have a flushing flow of water to clear the
discharge opening, either continuously or every time the system is turned on.
Labyrinth emitter (turbulent or turbulent-path emitter): A microirrigation
emitter having long, intricate passageways that create turbulent flow at normal
operating pressures.
Line-source emitter: A microirrigation device in which water is discharged from
closely spaced perforations, emitters, or a porous wall along tubing; an alterna-
tive to discrete emitters.
Long-path emitter: An emitter that employs a long capillary-sized tube or channel
to dissipate pressure.
Multi-outlet emitter: An emitter that supplies water to two or more points through
small-diameter auxiliary tubing.
Orifice emitter: An emitter that employs one or more orifices to dissipate pressure.
Pressure-compensating emitter: An orifice emitter designed to make use of a

flexible diaphragm so as to discharge water at a constant rate over a wide range
of lateral line pressures.
Vortex emitter: An emitter that employs a vortex effect to dissipate pressure.
End gun: A large sprinkler, similar to that used on a traveling sprinkler, mounted on
the end of a center pivot lateral to throw water a long distance, thereby increasing
the irrigated area for a given lateral length.
End suction pump: A horizontal centrifugal pump having one suction inlet, at one
side of the impeller.
Endotoxins: Complex polysaccharide molecules that occur in the outer cell mem-
branes of certain gram-negative bacteria and that may elicit an antigenic response.
Exposure may cause toxic hemorrhagic shock and severe diarrhea. May be present
in inadequately treated wastewater used as an irrigation water source.
Enterobacteria: Bacteria that may inhabit the intestines, including those that may
cause disease. May be present in inadequately treated wastewater used as an irriga-
tion water source (adapted from ICID MTD, 1996).
Enthalpy: In meteorology, enthalpy is sensible heat. In thermodynamics, a function of
a system, equivalent to the sum of the internal energy of the system plus the prod-
uct of its volume multiplied by the pressure exerted on it by its surroundings.
Entrance head: Head required to establish flow into a conduit or structure.
Entrance loss: Energy lost in eddies and friction at the inlet to a conduit or structure.
Equalizing ditch: A secondary ditch, usually parallel to a field ditch, used to supply
irrigation water to two or more furrows.
Evapotranspiration (ET): The combination of water transpired from vegetation and
evaporated from soil and plant surfaces.
Exchangeable sodium percentage: The percentage of the cation exchange capacity
of a soil occupied by sodium ions.
Feedback control: Control and managements systems that respond to real-time inputs
from sensors to adjust inputs or outputs.
Fertigation: The application of nutrient solutions to plants in irrigation water.
Field capacity: The amount of water remaining in a soil when the downward water
flow due to gravity becomes negligible. Also called the upper drainable limit.
Field ditch: A ditch constructed within a field, either for irrigation or drainage.
Fifteen atmosphere percentage: Water retained in an air-dried and screened sample
of soil that has been wetted and then brought to equilibrium on a permeable mem-
brane at a soil moisture tension of 155 m of water, expressed as a percentage of the
dry mass of soil; an estimate of the permanent wilting point or lower limit in some
soils.
Filter strip: A permanent vegetated strip between fields and receiving waters or run-
off conveyance structures to retard surface runoff and remove sediment, nutrients,
on other contaminates from surface runoff. Also called a desilting area.
Filter: (1) Sands, gravels, or fibrous materials placed around a well screen or perfo-
rated casing to increase permeability near the well and prevent aquifer formation
material from entering the well. (2) Any of various devices used in micro- and
sprinkler irrigation systems to remove debris from the water that might clog or oth-
erwise foul the emitters or sprinklers.
Final infiltration rate: See Basic intake rate.
Flashboard: A wooden plank, generally held horizontally in vertical slots on the crest
of a dam or check structure, to control the upstream water level. Sometimes called
crestboard.
Flexible membrane liner: A synthetic barrier to water movement, commonly fabri-
cated of polyvinyl chloride or high density polyethylene sheets, used to line water
storage facilities to prevent water loss by seepage or to line waste containment fa-
cilities to prevent flow of contaminants to groundwater.
Float valve: A valve, actuated by a float that automatically controls the flow of water.
Flood irrigation: A method of irrigation whereby water is applied to the soil surface
without such flow controls as furrows, borders, or corrugations. Also used region-
ally to refer to all forms of surface irrigation.
Flow meters: Any of several types of devices designed to measure rate of flow of
liquids or gases in closed conduits. Sometimes used to refer to open channel flow
measurement devices.
Flume: (1) An open conduit for conveying water across obstructions; (2) An entire
canal elevated above natural ground; an aqueduct; (3) A structure, calibrated for a
specified length and shape, for measuring open channel flow rate.
Fragipan: A soil having compacted horizons created by a close-packed arrangement
of silt and sand particles (usually low in clay) that are nearly impermeable to water.
See also Hardpan.
Freeboard: The vertical distance between the maximum water surface elevation an-
ticipated in design and the top of retaining banks, pipeline vents, or other struc-
tures, provided to prevent overtopping because of unforeseen conditions.
Friction head: The energy required to overcome friction caused by fluid movement
relative to the boundaries of a conduit or containing medium.
Friction slope: Friction head loss per unit length of a conduit.
Full irrigation: Management of a water application to fully replace the soil water
deficiency over an entire field.
Furrow dike: A small dike formed in a furrow to prevent runoff.
Furrow irrigation: A method of surface irrigation whereby the water is supplied to
small ditches or furrows to guide it across the field. May also be referred to as rill
or corrugation irrigation when the furrow size is relatively small.
Furrow: (1) A trench in the soil made by a tillage tool. (2) A small channel for con-
veying irrigation water.
Gate: A device used to control the flow of water to, from, or in a pipeline, or open
channel. It may be opened and closed by screw action, slide action, or hydraulic or
pneumatic actuators.
Gated pipe: Portable pipe with small gates installed along one side for distributing
irrigation water to corrugations or furrows.
Geographic information system, GIS: A computer system capable of capturing,
storing, analyzing, and displaying geographically referenced information.
Global positioning system , GPS: A satellite-based navigation system composed of a
network of satellites placed into specific orbits around the earth. Signals from the
satellites allow the latitude and longitude of receivers to be precisely determined;
useful in applications such as surveying, navigation, and precision agriculture.
Grade: (1) Noun. The degree of slope of a road, channel, or ground surface. (2) Verb.
To finish the surface of a canal bed, roadbed, top of embankment, or bottom of ex-
cavation.
Gradually varied flow: Steady nonuniform open-channel flow in which the changes
in depth and velocity from section to section are gradual enough such that accelera-
tive forces are negligible.
Gravel filter: A layer of granular solids, perhaps several centimeters thick, placed
around a drainage conduit or well screen to prevent the infiltration of fine materials
into the drain or well.
Gravitational potential: The elevation above or below a reference level. The pre-
ferred term is soil water potential.
Gravitational water: Soil water that moves into, through, or out of the soil under the
influence of gravity.
Gravity flow: Water flow that is not pumped but flows due to the force of gravity.
Occurs in irrigation, drainage, inlets, and outlets.
Gray water: Domestic wastewater other than that containing human excreta, such as
sink drainage, washing machine discharge, or bath water.
Green-Ampt equation: An equation that describes infiltration of water into soil as a
function of time.
Gross diversion requirement: The total quantity of water diverted from a stream,
lake, or reservoir, or removed from an aquifer in order to irrigate a crop.
Gross irrigation water requirement (IR): The total annual crop evapotranspiration,
less effective precipitation, then divided by efficiency and leaching factors.
Ground-fault circuit device: An electrical safety device that instantaneously cuts off
electrical current upon detection of current leakage.
Groundwater mining: Pumping of groundwater, for irrigation or other uses, at rates
significantly faster than the rate at which the groundwater is being recharged. Ulti-
mately results in aquifer depletion.
Groundwater: Water that occurs in the zone of saturation in an aquifer or soil.
Growing season: The period, often the frost-free period, during which the climate is
such that crops can be produced.
Gypsum block: An electrical resistance block, used to indicate soil water content, in
which the absorbent material is gypsum.
Hard hose: The conduit that supplies water, under pressure, to a traveling sprinkler.
Flexible enough to remain connected to both the water source and the sprinkler dur-
ing moves, but rigid enough to maintain a circular cross section when not pressur-
ized.
Hardpan: The general term for a dense, distinct soil layer that is largely impervious to
water, usually found below the uppermost topsoil layer, which may also restrict
plant root growth and farm tillage equipment. Hardpans may be formed by dis-
solved silica deposits in the soil that fuse and bind the soil particles or by rigid ma-
trices formed by deposits of iron oxides and calcium carbonates (caliche). Other
hardpans are formed by compaction caused by natural migration and packing of
soil particles (fragipans), by heavy traffic on wet soils, or by repeated tillage with
equipment such as moldboard plows or rototillers.
Head ditch: A ditch across the upper end of a field, used for distributing water in sur-
face irrigation.
Head gate: A water control structure at the entrance to a conduit or canal.
Head loss: The energy loss in fluid flow.
Head works: Diversion structures at the upper end of a conduit or canal.
Head: The energy in a liquid system expressed as the equivalent height of a column of
the liquid above a given datum.
Helminths: Parasitic nematodes, roundworms, hookworms, tapeworms, and whip-
worms. May be present in soils or in irrigation or drainage waters. May adversely
affect human health.
Horizontal centrifugal pump: A centrifugal pump having a horizontal shaft.
Humid climate: A climate characterized by high rainfall and low evaporation poten-
tial. A region is usually considered as humid when precipitation averages more than
500 mm (20 in.) per year.
Hydrant: An outlet, usually portable, used for connecting surface irrigation pipe to an
alfalfa valve outlet.
Hydraulic coefficient of variation: A measure of that discharge nonuniformity of
microirrigation emitters that is caused by pressure variations.
Hydraulic conductivity: The ability of a porous medium to transmit a specific fluid
under a unit hydraulic gradient; a function of both the characteristics of the medium
and the properties of the fluid being transmitted. Often determined by a laboratory
measurement, corrected to a standard temperature and expressed in units of
length/time. (Although the term hydraulic conductivity is sometimes used inter-
changeably with the term permeability, the user should be aware of differences.)
Hydraulic efficiency: (1) The efficiency with which a pump imparts energy to water
or a turbine extracts energy from water. (2) A measure of the loss of energy when
water flows through a hydraulic structure.
Hydraulic gradient: A change in the hydraulic head per unit distance.
Hydraulic jump: An abrupt turbulent rise in water level from a flow stage less than
critical depth to a flow stage greater than critical depth, within which the velocity
passes from supercritical to subcritical.
Hydraulic radius: The cross-sectional area of a fluid stream or conduit divided by its
wetted perimeter.
Hydraulic resistance: The friction along the wetted boundary of a channel or conduit
that causes a loss in head.
Hydraulic thrust: The tension on a pump shaft, consisting of the sum of the weight of
the impellers, the weight of the lineshaft, and the hydraulic thrust on the impellers.
Hydrograph: A graphical or tabular representation of the flow rate of a stream with
respect to time.
Hydrologic cycle: A term used to describe the movement of water in and on the earth
and in the atmosphere. Numerous processes, such as precipitation, evaporation,
condensation, and runoff, comprise the hydrologic cycle.
Hydrostatic pressure: Force per unit of area exerted by a liquid at rest.
Hygroscopic moisture (Hygroscopic water): Moisture absorbed by dry soil from a
saturated atmosphere.
Immobilization: (Soil) The biological conversion of inorganic compounds in the soil
to organic compounds such that the compounds are not usable by plants, but may
still be leached.
Impeller meter: A rotating mechanical device for measuring flow rate in a pipe or
open channel.
Impermeable layer: (Soil) A layer of soil that is resistant to penetration by water, air,
or roots.
Infiltrability: See Infiltration capacity.
Infiltration capacity: The maximum rate that water will infiltrate, as limited by soil
properties. Assumes that the water supply to the soil surface is not limiting.
Infiltration opportunity time: The time that water inundates the soil surface, with
opportunity to infiltrate.
Infiltration pit: A depressions in the soil near the control head of a microirrigation
system for disposal of water drained from the system.
Infiltration rate: The quantity of water that enters the soil surface in a specified time
interval. Often expressed in volume of water per unit of soil surface area per unit of
time.
Infiltration: The downward entry of water through the soil surface into the soil.
Infiltrometer: A device for measuring the infiltration rate of water into soil.
Instream flow requirements: The flow regime necessary to provide for the combined
needs of fish, wildlife, recreation, navigation, hydropower production, and down-
stream conveyance in a stream.
Intake: (1) Noun. Head-works of a conduit. (2) The place of diversion. (3) Verb. Wa-
ter infiltration into soil.
Interception: That portion of precipitation caught by vegetation and thus prevented
from reaching the soil surface.
Interceptor drains: A buried tile drainage system, gravel-filled trench, or open drain
installed at a slight angle to the contour to collect and remove groundwater as it
flows across an impermeable layer. Thus, excess water is prevented from locally
saturating the crop root zone or from causing embankments to become unstable.
Used when water originates outside the area being protected by the drains.
Intercropping: The practice of cultivating an additional crop in the spaces available
between the main crop plants or rows, in order to produce a greater yield per unit
area during a given time period.
Interlock: A safety device on a chemigation system that prevents the operation of the
water supply or chemical solution pump if the other pump stops, thus preventing
chemical spills.
Interseeding: The general practice of sowing a crop into another standing, growing,
crop late in the season, usually to enhance biomass production, erosion control, or
to increase soil organic matter.
Intrinsic permeability: The property of a porous material that describes the ease with
which gases or liquids flow through it. See Permeability.
Invert: The lowest element of the internal cross section of a channel or pipe.
Inverted siphon: A closed conduit with end sections above the middle section; used
for carrying canal flow below a depression or under a highway by gravity. Does not
involve siphonic action.
Irrecoverable water loss: Water loss that, because of evaporation, phreatophyte tran-
spiration, or groundwater recharge that is not economically recoverable, becomes
unavailable for reuse.
Irrigable area: The area in a project capable of being irrigated, principally as regards
to availability of water, suitable soils, and topography.
Irrigating stream: (1) Flow for irrigation of a particular tract of land. (2) Flow of
water distributed at a single irrigation. Sometimes called irrigating head.
Irrigation check: A small dike or dam used in the furrow alongside an irrigation bor-
der to make the water spread evenly across the border.
Irrigation district: A cooperative, self-governing, semi-public organization set up as
a subdivision of a state or local government to provide irrigation water to members.
Irrigation efficiency: The ratio of the average depth of irrigation water that is benefi-
cially used to the average depth of irrigation water applied, expressed as a percent-
age. See Beneficial use.
Irrigation hose: A closed conduit for supplying water to moving irrigation systems,
flexible when subjected to normal operating pressure; may be collapsible to a flat
cross section when purged of water.
Irrigation interval: The time from the start of one irrigation until the beginning of the
next irrigation for a given field. Sometimes called irrigation frequency.
Irrigation runoff percentage: The equivalent depth of irrigation water running off a
field expressed as a percentage of the depth of irrigation water applied.
Irrigation set: The area irrigated at one time within a field.
Jet pump: A centrifugal pump equipped with a jet or ejector assembly that allows
water to be lifted from greater depths than would be possible without the jet. Typi-
cally used for domestic water supply.
Kinematic wave: A method of mathematical analysis of unsteady open channel flow
in which the dynamic terms are omitted because they are small and assumed to be
negligible.
Kostiakov equation: A mathematical description of the rate of infiltration into soil as
it varies with time.
Lag time: (Surface irrigation) The interval between the time water is turned off at the
upper end of a field and the time that it recedes (disappears) from that point.
Land forming: A general term encompassing both land grading and land smoothing.
Land grading: The shaping of the surface of land to predetermined grades so each
row or surface slopes to a drain or is configured for efficient irrigation water appli-
cation. Also called land forming or land shaping. See Land leveling as a special
case.
Land leveler: A machine with a long wheel base, used for land smoothing or leveling
operations.
Land leveling: The process of shaping the land surface to a level surface; a special
case of land grading.
Land smoothing: The removal of nonuniformities from the soil surface.
Landscape coefficient: A factor for modifying estimates of reference crop evapotran-
spiration to derive an estimate of ET for a given, perhaps urban, landscape.
Laser leveling: Land leveling in which a stationary laser transmitter and a laser re-
ceiver on earthmoving machines are used for grade control.
Laser receiver: An electronic device normally mounted on earthmoving machines or
trenchers, which receives signals from a laser transmitter and indicates to the opera-
tor or sends signals to control points on the machine to adjust the machine to follow
the slope established by a laser transmitter.
Laser transmitter: A device that generates a collimated laser light beam. By rotating
the beam, a plane of laser light may be created.
Lateral: A branch from a larger irrigation canal. Commonly refers to the channel that
carries water from a main canal to a farm headgate.
Lath box: See Spile.
Leachate: Water that moves downward through some porous media; often contains
substances dissolved from the media.
Leaching fraction: The ratio of the depth of subsurface drainage water (deep percola-
tion) to the depth of infiltrated irrigation water. See Leaching requirement.
Leaching requirement: The fraction of water entering the soil that must pass through
the root zone in order to prevent soil salinity from exceeding a specified value.
Leaching requirement is used primarily under steady-state or long-time average
conditions (USDA, 1954).
Leaching: Removal of soluble material from soil or other permeable material by the
passage of water through it.
Length of run: The distance water must flow in furrows or borders over the surface of
a field from the head to the end of the field.
LEPA: See Low energy precision application.
Lewis and Milne equation: An equation that describes infiltration of water into soil
as a function of time.
Life cycle cost: The sum of the initial investment costs and the other fixed and operat-
ing costs, including interest on the money invested, over the life of a piece of
equipment such as a pumping plant or sprinkler system.
Lift: (1) The depth of the layer of soil deposited with each pass of earthmoving
equipment. (2) The vertical distance water is raised by a pump, measured from the
surface of the water being pumped to the elevation of the discharge.
Limited irrigation: See Deficit Irrigation.
Line gate: A hub-end, screw-type valve that is installed in a pipeline.
Line source: (1) A single line of relatively closely spaced sprinklers that applies water
in triangular-shaped distribution pattern perpendicular to the sprinkler line, used for
crop water response research. (2) A source of water emitted uniformly along a line,
such as a drip tape.
Longitudinal smoothing: A land smoothing operation where all soil movement is
done parallel to the crop row direction for the purpose of obtaining a grade.
Low energy precision application (LEPA): An irrigation system, either center pivot
or lateral move, equipped with extended-length drop tubes and application devices
designed to apply small, frequent irrigations at or near ground level to individual
furrows. The LEPA concept involves soil surface management to increase water
storage in the furrows until infiltration is complete.
Lysimeter: An isolated block of soil, usually undisturbed and in situ, for measuring
the quantity, quality, or rate of water movement through or from the soil. The term
lysimeter is often used erroneously for suction cup samplers or other passive or ac-
tive soil water capture devices.
Macropores: Noticeable cracks or tubes (such as worm holes or cavities left by de-
cayed roots) in the soil profile. See Preferential flow.
Management allowed depletion: The desired soil water deficit at the time of irriga-
tion; can vary with crop growth stage.
Manifold: (1) A pipeline having multiple outlets to supply water to laterals. (2) A
piped arrangement for connection of multiple lines or devices to an irrigation sys-
tem. (3) A pipeline connecting multiple subsurface drainage lines in order to collect
effluent for disposal.
Manufacturer’s coefficient of variation: A measure of the variability of discharge
from a random sample of a given make, model, and size of microirrigation emitters,
as they are produced by the manufacturer and before any field operation or aging
has taken place; equal to the ratio of the standard deviation of the discharge of the
emitters to the mean discharge of the emitters.
Matric potential: A measure of the attraction that the solid soil matrix has for water.
Defined as the amount of work that must be done to transfer a unit quantity of pure
water, identical in composition to the soil water, from a pool the elevation and ex-
ternal gas pressure of the point under consideration, to the soil water.
Media filters: Pressurized tanks containing a uniform, loose bed of sharp-edged sand
or crushed rock to entrap organic and inorganic contaminants from water; often
used for microirrigation.
Meter gate: A calibrated, adjustable irrigation valve used for flow measurement and
control.
Microirrigation emitter: A small dispensing device designed to dissipate pressure
and discharge a small, uniform flow or trickle of water at a constant rate, which
does not vary significantly because of small variations of pressure. Also called a
dripper or trickler. See Emitter.
Microirrigation: The frequent application of small quantities of water as drops, tiny
streams, or miniature spray through emitters or applicators placed along a water de-
livery line. Microirrigation encompasses a number of methods or concepts such as
bubbler, drip, trickle, mist, and spray.
Microsprinkler (Microspinner, Microsprayer): An emission device, often with a
simple orifice; includes small, low-pressure minisprinklers, foggers, spitters, spin-
ners, jets, and sprayers that are installed in the field on tubing. A microsprinkler
typically applies water to a larger area than a drip emitter, but does not uniformly
cover the entire cropped area.
Miner’s inch: Discharge from an orifice, one square inch in area, under a definite
head, fixed by statute or practice but differing from state to state. In Colorado, a
miner’s inch equals 0.00074 m3/s. In Arizona, northern California, Montana, Ne-
vada, and Oregon, a miner’s inch equals 0.00071 m3/s. In southern California,
Idaho, Kansas, New Mexico, North Dakota, South Dakota, Nebraska, and Utah, a
miner’s inch equals 0.00057 m3/s. (No longer in common use.)
Mist irrigation: A method of microirrigation in which water is applied in very small
droplets, usually overhead, to modify air temperature or humidity.
Mixed-flow pump: A pump in which the pressure is developed partly by centrifugal
force and partly by the lifting action of the impellers on the water; combines the
characteristics of axial-flow and centrifugal pumps.
Moisture equivalent: An arbitrary soil water content used as an estimate for the field
capacity of soils. It is the weight of water remaining in a soil sample (expressed as
percentage of the dry weight of soil) after the soil has been saturated and subjected
for 30 min to a centrifugal force 1,000 times gravity. (No longer in common use.)
Multiple-state (Multi-stage) pump: A pump having more than one impeller, all
mounted on a common shaft.
Nappe: A sheet or curtain of water flowing from a structure, such as a weir or dam.
Net irrigation water requirement: The depth of water needed to fulfill the ET re-
quirement in excess of any effective precipitation for a disease-free crop growing in
a large field under non-restricting soil and soil water conditions and under adequate
fertility. Also considers contributions of shallow groundwater and change in stored
soil water during the period of interest.
Net positive suction head (NPSH): The head that causes liquid to flow through the
suction piping and enter the eye of the pump impeller. The required NPSH , which
should be specified by the pump manufacturer, is a function of the pump design
and varies with the capacity and speed of the pump. Available NPSH is a function
of the system in which the pump operates and represents the energy level in the wa-
ter over vapor pressure at the pump inlet. The available NPSH must equal or ex-
ceed the required NPSH or cavitation will occur.
Neutron scattering: The measurement of soil water content by means of an instru-
ment that introduces fast neutrons into the soil region to be measured. Fast neutrons
are “thermalized” by impact with hydrogen ions; thus quantity of thermalized neu-
trons can be related to water content.
Nonbeneficial use: The consumption of water by non-economic vegetation and
evaporation from bare soil and water surfaces. See Beneficial use.
Non-point source pollution: Pollution originating from diffuse areas (land surface or
atmosphere) having no well-defined source.
Non-process depletion: The use or removal of water from a water basin such that it is
permanently unavailable for further use and such use or removal does not produce
an intended good. Includes evaporation from soil and water surfaces, depletion by
weeds and other non-economic vegetation.
Nonsaline-alkali soil: Soil containing sufficient exchangeable sodium to interfere
with the growth of most crops. Preferred term is sodic soil.
Nonuniform flow: Flow in which the average cross-sectional velocity is not the same
at successive channel cross sections. If the velocity at a given cross section is con-
stant with time, it is referred to a steady nonuniform flow. If the velocity changes
with time at each cross section, it is known as unsteady nonuniform flow.
Normal depth: Depth of flow in an open channel during uniform flow.
Nozzle: The discharge opening or orifice of a sprinkler head; used to control the vol-
ume of discharge, distribution pattern, and droplet size.
One-third atmosphere percentage: Water retained in an air-dried and screened sam-
ple of soil that has been wetted and then brought to equilibrium on a permeable
membrane at a soil moisture tension of 3.45 m of water, expressed as a percentage
of the dry mass of soil. An estimate of field capacity in some soils. Also referred to
as upper drainable limit.
Operational efficiency: The ratio of the attained efficiency to the design efficiency.
Operational waste: Water that is lost or otherwise discarded from an irrigation sys-
tem after having been diverted into it as part of normal operations.
Orchard valve: An outlet valve installed inside a pipeline riser with an adjustable
cover or lid for flow control, similar to an alfalfa valve but with lower flow capac-
ity.
Orifice: An opening with a closed perimeter through which water flows. Certain
shapes of orifices are calibrated for use in measuring flow rates.
Osmotic potential: The amount of work that must be done, per unit quantity of water,
in order to transport an infinitesimal quantity of water from a pool of pure water, at
atmospheric pressure, to a pool of water identical in composition to the soil water
(adapted from ICID MTD, 1996; SSSA, 2001).
Outlet gate: A valve, usually a slide valve that controls the flow of water from an
outlet.
Outlet: An appurtenance to deliver water from a pipe system to the land. An outlet
may consist of a valve, a sprinkler head, a riser pipe, and/or an outlet gate.
Overflow stand: A standpipe in which water rises and, at a given elevation, overflows
into a pipe or a containment vessel.
Overhead irrigation: Any sprinkler irrigation method where the water is applied
above the crop.
Overland flow: The flow of storm water over the soil surface before becoming chan-
nelized (from ICID MTD, 1996).
Packing gland: A seal that surrounds the drive shaft of a centrifugal pump, located
where the sbaft exits the pump’s volute case; allows shaft rotation with minimal
friction and minimal water leakage.
PAM: See Polyacrylamide.
Parshall flume: A calibrated device used to measure the flow of water in an open
channel, based on the principle of critical flow. Formerly called the improved Ven-
turi flume.
Particle-size analysis: The determination of the various amounts of the different par-
ticle size fractions in a soil sample, usually by sedimentation, sieving, or microme-
try.
Peak use rate: The maximum rate of consumptive use of water by plants.
Penman-Monteith procedure: A method of estimating the rate of evapotranspiration
based on a net balance of energy fluxes and vapor transport.
Perched water table: An aquifer condition that occurs when there is an impermeable
layer (aquiclude) above the main aquifer but below the land surface. Water perco-
lating downward is trapped above the aquiclude, forming the perched water table.
May result in the need for localized drainage.
Percolation pond: A man-made pond established to recharge an aquifer by allowing
water to percolate through layers of soil and gravel. The water is filtered and
cleaned of sediments and pathogens as it slowly moves toward the aquifer.
Percolation: The downward movement of water through the soil profile or through
other porous media.
Perforated casing: A section of well casing with openings for water entry. See Screen
(1).
Peristaltic pump: A pump that pressurizes liquid through action that resembles an
animal’s heart beat.

Permanent wilting point: The soil water content below which plants cannot readily
obtain water and will permanently wilt. Sometimes called permanent wilting per-
centage, fifteen atmosphere percentage, or lower limit. It is estimated as the soil
moisture content at about –1.5 MPa matric potential.
Permeability: (1) Qualitatively, the ease with which gases, liquids, or plant roots
penetrate or pass through a layer of soil or porous media.( 2) Quantitatively, the
specific soil property describing the rate at which gases and liquids can flow
through soil or porous media.
Phenology: A description of the stages of a plant’s development as affected by its
environment.
Phreatophyte: A plant that draws its water supply from groundwater (from ICID
MTD, 1996).
Phytoplankton: Small plants that float passively near the surface of water bodies
(adapted from ICID MTD, 1996).
Phytotoxin: A chemical substance that, when applied to growing plants, by man or by
pathogens, will adversely affect their health.
Piezometric head: The pressure of a fluid at a given elevation in a conduit or porous
media, represented by the height to which the fluid would rise in a tube open to the
fluid at its lower end and to the atmosphere at its upper end.
Pitcher irrigation: An irrigation method that uses loosely covered, unglazed clay pots
to distribute water to sparsely planted crops by diffusion and capillarity through the
pot walls. Typically, the pots are filled with water manually. Potentially highly ef-
ficient. Also called pot irrigation.
Pitot gage: A device for measuring velocity of fluid flow at a point. Typically a tube,
several mm in diameter and several cm in length, aligned with the flow, and having
an opening at its upstream, streamlined tip that senses total head, and another open-
ing to measure pressure head. Velocity can be computed from the head differential.
Plastic or sheet paper mulches: Sheets of plastic or organic materials, generally nar-
rower than crop row spacing, that are placed on the soil on rows of high-value
crops to restrict evaporation, prevent weed growth, and/or regulate soil tempera-
ture. Holes in the mulch allow desired plants to emerge.
Playa: A dry, flat-floored basin, often lined with fine sediments, which holds water
and becomes a lake temporarily after rains; generally restricted to arid regions
Point-rows: Crop rows at the edges of a field that, because of field shape, are shorter
than the typical row, thus more difficult to water with furrow irrigation.
Polyacrylamide (PAM): A chemical that tends to cause aggregation of soil particles,
thus inhibiting water-induced erosion and enhancing infiltration. Can be applied in
irrigation water or directly to the soil.
Ponding depth: The depth of water that will be infiltrated in an event before water
begins to pond on the soil surface, assuming the water application is by sprinkling
or rainfall.
Pore-size distribution index: The volumes of various sizes of pores in a soil, ex-
pressed as percentages of the bulk volume (from SSSA, 2001).
Porosity: (1) In soil, the volume percentage of the total bulk not occupied by solid
particles. (2) In an aquifer, the sum of the specific yield and the specific retention.
Porous pipes: Tubing having permeable walls, allowing water to seep more or less
uniformly along the length of the tube for microirrigation.
Porous trickle tubing: Microirrigation tubing with a uniformly porous wall. The
pores are small and ooze water under pressure.
Portable pipe: Irrigation pipe, which is or can be moved between irrigation sets, such
as sprinkler or gated pipe.
Positive displacement pump: A pump that moves a fixed quantity of fluid with each
stroke or rotation, by means of a piston or gear; thus having a discharge rate nearly
independent of discharge pressure.
Potential evapotranspiration: The rate at which water, if available, would be re-
moved from wet soil and plant surfaces by evaporation and transpiration (from
ASCE, 1990).
Power unit: A motor or engine used to drive a machine such as a pump.
Precipitation excess: That amount of precipitation that exceeds retention, such as
interception by vegetation, infiltration, and storage in depressions on the soil sur-
face. Generally considered equivalent to the surface runoff of an area.
Precipitation intensity: The rate of precipitation, generally expressed in units of
depth per time.
Precision agriculture (PA) or Precision farming: Any of a variety of farming prac-
tices that emphasize the need to spatially apply production inputs in the proper
amounts and at the proper times to maximize their effectiveness. May involve in-
tentional spatial variation in seeding rates and varying amounts of irrigation water,
fertilizer, or pesticides on different areas of a single field, to give the most eco-
nomical production from those areas.
Preferential flow: Flow into and through porous media or soil by way of cracks, root
holes, and other paths of low resistance rather than uniformly through the whole
media.
Preferential flow: Flow into and through porous media or soil by way of cracks, root
holes, and other paths of low resistance rather than uniformly through the whole
media.
Preplant irrigation: Irrigation applied prior to seeding. Sometimes called preirriga-
tion.
Pressure head: The pressure energy in the liquid system, indicated as the equivalent
height of a water column above a given datum.
Pressure potential: (Soils) See Matric potential.
Pressure rating: The estimated maximum internal pressure that can be continuously
exerted in a pipe or container with a high degree of certainty that failure will not
occur.
Pressure regulator: (1) A pressure-reducing device designed to maintain a constant
pressure at an irrigation sprinkler regardless of elevation changes of the sprinkler.
(2) A device used to maintain a constant, desired pressure on a pipeline.
Pressure vacuum breaker: A device that automatically opens, causing air to enter a
pipeline and stop backsiphonage, when a reversal or pressure gradient occurs that
could otherwise cause flow to siphon back toward the source. Differs from an at-
mospheric vacuum breaker by having a spring to help the valve open against inter-
nal line pressure.

Primary treatment: The first stage of wastewater treatment. Removes floating debris
and solids by screening and sedimentation (from ICID MTD, 1996).
Productivity of water: See Water use efficiency.
Progressive cavity pump: A type of positive displacement pump used to transfer en-
ergy to a fluid by the rotary motion of a specially shaped impeller within an appro-
priately shaped casing.
Propeller pump: A pump that develops pressure primarily by the lifting action of
vanes on the water. See Axial-flow pump.
Proportional mixers: A chemical injector in which the quantity of material injected
depends on the flow rate through the injector. Thus, the concentration of injected
chemical in the water remains constant.
Protozoa: (Water) Single-celled organisms. Some, such as Giardia lamblia and G.
muris, may cause diarrhea in humans when their cysts are ingested.
Psychrometric constant: The ratio of sensible heat to latent heat at a given pressure:
y = CP P/ 0.622 λ , where CP is the specific heat of air, P is the atmospheric pres-
sure, and λ is the latent heat of vaporization.
Pump column: The pipe through which water from turbine or submersible pumps in a
well is conveyed to the ground surface.
Pump efficiency: The ratio of the water power produced by the pump, to the power
delivered to the pump by the power unit.
Pump stand: A structure to convey water from a pump to the inlet of a pipeline.
Pump: See the specific type of pump in question.
Pumpback system: Equipment that enables reuse of irrigation water on a given field,
by collecting runoff from the field and pumping it to the upper end of the field to be
reapplied.
Pumping plant or station: A complete installation of one or more pumps, together
with all necessary appurtenances such as power units, sumps, screens, valves, mo-
tor controls, motor protection devices, fences, and shelters.
Rack: A screen of parallel bars placed in a channel to catch debris. Sometimes called
a bar screen.
Radial gate: A horizontally pivoted water control gate whose face is usually a circular
arc with center of curvature at the pivot. Sometimes called a Taintor gate.
Radiation (extraterrestrial, clear-sky, and others): The process by which electro-
magnetic energy is propagated through free space, as distinguished from conduc-
tion and convection.
Range nozzle: One of two orifices on an impact sprinkler head. Produces a collimated
jet that maximizes the wetted radius, as opposed to the spreader nozzle.
Recession phase: The period, during surface irrigation, after inflow has ceased and
the soil surface below the stream is exposed, usually from the upstream end, but
occasionally from the downstream or both ends.
Recharge well: A well used to introduce excess surface waters into a groundwater
body.
Reciprocating diaphragm pump: A pump that uses the back-and-forth motion of a
flexible disk to draw fluid in through one check valve and expel it at higher pres-
sure through another.
Reciprocating piston pump: A pump that uses the back-and-forth motion of a piston
in a cylinder to draw fluid in through one check valve and expel it at higher pres-
sure through another.
Reclaimed water: Water that, after use for domestic, agricultural, or industrial pur-
poses, is made suitable for further beneficial use by appropriate treatment.
Redistribution: The movement of water within the soil profile, due to potential gradi-
ents, from the point where it first infiltrates the soil.
Reduced pressure backflow device: Two independently acting, internally loaded,
check valves separated by a reduced pressure zone, installed between two shut-off
valves in a chemigation line to prevent backflow caused by backpressure and back-
siphonage.
Reference evapotranspiration: The rate at which water is evaporated and transpired
from a specific, reference crop when soil water is readily available and under speci-
fied conditions of growth. Estimates of reference ET can be modified by appropri-
ate coefficients for other crops and non-standard conditions.
Regulated deficit irrigation: An irrigation strategy in which specific plant water
stresses are deliberately imposed during specific growth stages (usually early in the
season) using daily irrigations but only replacing a fraction of the plant’s daily wa-
ter use.
Relief drainage system: A network of subsurface drains that is used when the source
of the water causing the drainage problem originates within the area protected by
the drains.
Relief well: A small well installed in an area with a high water table for the purpose of
lowering the water table; often part of a network of multiple wells, usually used
when buried drainage systems are not feasible.
Reservoir tillage: See Basin tillage.
Residual water content: Water that cannot be withdrawn from a soil by suction.
Resistance coefficient: An empirical term in equations for estimating head loss for
flow in pipes or open channels. Allows the estimate to be adjusted according to the
degree of resistance to flow. Depends primarily on the roughness of the conduit
boundary, but may also consider sinuosity and obstacles in the flow.
Retention storage: See Surface storage.
Return flow: That portion of the water diverted from a stream which finds its way
back to the stream channel, either as surface or subsurface flow.
Return-flow or reuse system: A system of pipelines or ditches to collect and convey
surface or subsurface runoff from an irrigated field for reuse.
Revolving screen: A trash screen or rack in the form of a cylinder or continuous belt
that is rotated by water passing through it or some other power source.
Richard’s equation: An equation that describes unsaturated flow in porous media.
Riser: (Sprinkler irrigation) The vertical pipe used to connect a sprinkler head to a
lateral pipe; elevates the head above the crop being irrigated.
Root zone: The layer of soil that plant roots readily penetrate and in which the pre-
dominant root activity occurs.
Rotary gear pump: A pump that uses two meshed, rotating gears to force liquid
through the chamber enclosing them, and expel it at high pressure.
Rotary lobe pump: A pump that uses two rotating lobes in a chamber to force liquid
out of the chamber at high pressure.
Roughness coefficient: Preferred term is resistance coefficient.
Row crop: A crop, annual or perennial, planted in rows far enough apart to allow cul-
tivation between rows during the growing season.
Runoff: That portion of precipitation, snow melt, or irrigation that flows over and
through the soil, eventually making its way to surface water supplies.
Safe well yield: Amount of groundwater that can be withdrawn from an aquifer with-
out degrading quality or reducing pumping level.
Salination, Salinization: The increase in the concentration of various salts to soil or
water. The term usually implies that the use of the soil or water, by man, is im-
paired.
Saline soil: Nonsodic soil containing soluble salts in such quantities that they interfere
with the growth of most crops. The electrical conductivity of the saturation extract
is greater than 4 milliSiemens/cm (0.01 mho/in.), and the exchangeable-sodium
percentage is less than 15.
Saline-sodic soil: Soil containing sufficient exchangeable sodium to interfere with the
growth of most crops and containing appreciable quantities of soluble salts. The
exchangeable-sodium percentage is greater than 15, and the electrical conductivity
of the saturation extract is greater than 4 milliSiemens/cm.
Salinity stress function: A coefficient used to reduce the crop coefficient in
evapotranspiration estimates under conditions of salinity stress.
Salinization: See Salination.
Salt sink: An area where salts are immobilized, i.e., removed more or less perma-
nently from transport with water movements in the hydrologic cycle. Sinks may
occur in the subsoil or in shallow water bodies.
Saltwater intrusion: The invasion of a freshwater body, especially an aquifer, by
saline water from an ocean or other body of saltwater, when there is a hydraulic
connection between the two bodies and pressure gradients are favorable.
Saturated flow: Flow of water through a porous material under saturated conditions.
Screen: (1) Wells: A manufactured well casing with precisely dimensioned and
shaped openings (compare with Perforated casing). (2) Canals: A device used to
trap debris in an open channel.
Seawater intrusion: See Saltwater intrusion.
Secondary canals: Sub-main canals or laterals.
Secondary treatment: Biochemical treatment of wastewater, using bacteria to con-
sume organic wastes. May involve use of trickling filters, activated sludge and/or
chlorine disinfection. Removes floating and settleable solids and most suspended
solids and BOD (adapted from ICID MTD, 1996).
Sedimentation: The process by which solid fragments of inorganic or organic mate-
rial, deriving from soil erosion or weathering of rock, are transported and deposited
by wind, water, or ice.
Seep collar: An annular plate of concrete or other impervious material placed on the
outside surface of an underground conduit to lengthen the flow path and thus im-
pede seepage.
Seepage: The movement of water into and through the soil from unlined canals,
ditches, or water storage facilities.
Semiarid climate: Climate characterized as neither entirely arid nor humid, but in-
termediate between the two conditions. A region is usually considered as semiarid
when precipitation averages between 250 mm (10 in.) and 500 mm (20 in.) per yr.
Also called subarid.
Sensible heat: (1) The internal energy of a gas plus the product of pressure and vol-
ume. (2) The heat absorbed or given off by a substance that is not in the process of
changing its physical state (in contrast to latent heat).
Shutoff head: The pressure head at the outlet of a pump at which the discharge drops
to zero. The maximum pressure a pump will develop at a given speed.
Silt: (1) A soil separate consisting of particles between 2 and 50 µm in diameter.
(2) Colloquial: Deposits of sediment, which may contain soil particles of all sizes.
Siltation: See Sedimentation.
Single-leg distribution: The water application pattern about a single sprinkler as
measured along one radius. Commonly approximates a triangular or an ellipse.
Siphon tube: A relatively short, lightweight, curved tube used to convey water over
ditch banks to irrigate furrows or borders.
Site-specific farming: See Precision agriculture.
Slide gate: A head control valve, which slides on rails, used to control drainage or
irrigation water.
Slip-formed ditch: An open channel, typically for irrigation water conveyance. Con-
structed by excavating a channel of uniform cross-section in the earth, then lining
by feeding fresh concrete into a slip form that is moved continuously along the
channel. The slip form extrudes the concrete along the channel boundary with uni-
form thickness and relatively smooth surface.
Soaker hose: A microirrigation device consisting of a tube with porous walls or uni-
formly spaced perforations that emits water, under low pressure, as a line source.
Soaking phase: The period, during surface irrigation, after advance is complete, dur-
ing which inflow continues until sufficient water has been introduced to satisfy the
requirement in the field and inflow is stopped.
Sodic soil: A nonsaline soil containing sufficient exchangeable sodium to adversely
affect crop production and soil structure. The exchangeable-sodium percentage is
greater than 15 and the electrical conductivity of the saturation extract is less than 4
milliSiemens/cm (0.01 mho/in.).
Sodium adsorption ratio (SAR): The proportion of soluble sodium ions in relation to
the soluble calcium and magnesium ions in the soil water extract. (Can be used to
predict the exchangeable sodium percentage.)
Sodium percentage: Sodium cations as a percentage of total cations in water or a soil
solution.
Soft hose: (Sprinklers) A conduit, which collapses when not under pressure, used to
supply water, under pressure, to a traveling sprinkler.
Soil aeration: The process by which air and other gases enter or are exchanged in the
soil.
Soil compaction: Consolidation, reduction in porosity, and collapse of the structure of
soil when subjected to surface loads.
Soil heat flux density: The rate or quantity of vertical movement of heat through a
unit area of soil in response to temperature gradient. A factor in energy balance es-
timates for evapotranspiration calculations.
Soil horizon: A layer of soil differing from adjacent genetically related layers in
physical, chemical, and biological properties or characteristics.
Soil profile: A vertical section of the soil from the surface through all its horizons into
the parent material.
Soil structure: The combination or arrangement of primary soil particles into secon-
dary particles, units, or peds that make up the soil mass. The principal types of soil
structure are platy, prismatic, columnar, blocky, and granular.
Soil texture: Classification of soil by the relative proportions of sand, silt, and clay
present in the soil.
Soil water capacity: The quantity of water that a soil can hold and make available to
most plants, usually defined as water held between –33 kPa and –1500 kPa matric
potential; the amount of water stored in the soil at field capacity. Also called avail-
able water capacity.
Soil water characteristic curve: A soil-specific relationship between the soil water
matric potential and soil water content.
Soil water deficit or depletion: The amount of water required to raise the soil water
content of the root zone to field capacity.
Soil water potential: The amount of work that must be done per unit quantity of pure
water in order to transport reversibly and isothermally an infinitesimal quantity of
water from a pool of pure water at a specified elevation at atmospheric pressure to
the soil water at the point under consideration.
Soil water pressure: The pressure (positive or negative), relative to the external gas
pressure on the soil water, to which a solution identical in composition to the soil
water must be subjected in order to be in equilibrium through a porous permeable
wall with the soil water.
Soil water: Water stored in the soil.
Soil: The unconsolidated minerals and material on the immediate surface of the earth
that serves as a natural medium for the growth of plants.
Specific yield: The ratio of the volume of water drained from a soil or aquifer to the
bulk volume drained when gravity drainage from the soil or aquifer is induced by a
well or drainage system. Specific yield will always be less than or equal to the po-
rosity of the soil or aquifer. Also called drainable porosity.
Spile: A conduit, made of lath, pipe, or hose, placed through an irrigation ditch bank
to transfer water to a field.
Spillbox: A canal-stabilizing structure.
Spray irrigation: The application of water by a small spray or mist to the soil surface,
where travel through the air becomes instrumental in the distribution of water.
Spreader nozzle: An orifice on an impact sprinkler head designed to apply water
close to the head, as opposed to the range nozzle.
Sprinkler head: A device for distributing irrigation water under pressure.
Sprinkler irrigation: A method of irrigation in which the water is sprayed or sprin-
kled through the air to the ground surface.
Sprinkler irrigation systems:
Boom: An elevated, cantilevered sprinkler(s) mounted on a central stand. The
sprinkler boom rotates about a central pivot.
Center pivot: An automated irrigation system consisting of a pipeline that rotates
about a pivot point and is supported by a number of self-propelled towers. The
water is supplied at the pivot point and flows outward through the pipeline, sup-
plying outlets spaced along the pipeline.
Corner pivot: An additional span or other equipment attached to the end of a cen-
ter pivot irrigation system that allows the overall radius to increase or decrease
in relation to the field boundaries. Also called corner catcher or corner system.
Lateral move (Linear move): An automated irrigation machine consisting of a
sprinkler line supported by a number of self-propelled towers. The entire unit
moves in a generally straight path and irrigates a basically rectangular area.
Permanent: Underground piping with risers and sprinklers.
Portable (Hand move): A sprinkler system that is moved by disconnecting pipe
sections, carrying them to a new location, and reconnecting them, requiring no
special tools.
Side move sprinkler: A sprinkler system with the supply pipe supported on car-
riages and towing small diameter trailing pipelines, each fitted with several
sprinkler heads.
Side roll sprinkler: The sprinkler system in which the lateral pipe serves as the
axle for a number of wheels. The system is moved across the field between sets,
using a small engine to rotate the pipeline.
Solid set: A system that covers the complete field with pipes and sprinklers in such
a manner that the entire field can be irrigated without moving any of the system.
May be portable or permanent.
Towed sprinkler (Towline): A system where lateral pipelines are mounted on
wheels, sleds, or skids, and are moved from set to set by towing in a direction
approximately parallel to the lateral.
Trail line system: A side move sprinkler designed so that the main lateral can pull
multiple small diameter trail lines, each carrying several sprinkler heads. This
configuration increases the area that can be watered with each system move.
Sometimes called movable solid set.
Traveler system: A single large sprinkler, commonly referred to as a gun,
mounted on a moving cart.
Sprinkler pattern: The areal distribution of water applied by a single sprinkler head
or an array of sprinkler heads.
Stabilized grade: The slope of an earthen channel at which neither erosion nor depo-
sition occurs.
Staff gauge: A graduated scale, generally vertical, from which the water surface ele-
vation may be read.
Stage: The elevation of a water surface above or below an established datum; gauge
height.
Stand gate: A valve in a structure that covers an inlet into, or an outlet from, a pipe-
line and controls water flow into or out of the pipeline.
Stand: A structure, often to protect or control flow in a pipeline, formed from vertical
sections of pipe or from cast-in-place concrete (box stand). It may serve as a pump
stand, gate stand, or float valve stand. It may also function as a vent or sand trap.
Standpipe: A vent constructed of vertical pipe on a pipeline, to relieve pressure
surges and water hammer.
Static head: The potential energy due to elevation differences. See Head.
Static lift: The vertical distance between source and discharge water levels in a pump
installation.
Steady flow: Open-channel flow in which the rate and cross-sectional area remain
constant with time at a given station.
Stem flow: (1) Precipitation intercepted by vegetation that reaches the ground by
flowing down the stems or trunks of vegetation. (2) Flow of sap in the xylem of
plants.
Stilling well: A pipe, chamber, or compartment, having closed sides and bottom ex-
cept for a comparatively small inlet connected to a main body of water for attenua-
tion of waves or surges while permitting the water level within the well to rise and
fall with the major fluctuations of the main body. Used with water measuring de-
vices to improve accuracy of measurement.
Storage efficiency: The ratio of the average depth of irrigation water infiltrated and
stored in the root zone to the soil water deficit, expressed as a percentage.
Stratified soils: Soils that are composed of layers, usually varying in permeability and
texture.
Stress irrigation: See Deficit irrigation.
Sub, Subbing: (Colloquial) The lateral movement of water through soil from the wet-
ted portion of an irrigated furrow to the center of the ridge. Important when irriga-
tion is needed to germinate seeds.
Subarid climate: See Semiarid climate.
Subhumid climate: The condition in an area that has, generally, enough precipitation
to support all crops, but rainfall variability makes irrigation necessary to prevent
crop water stress. Natural vegetation in such an area in normally comprised of tall
grasses. Annual rainfall varies from 500 mm in cool regions to as much as 1525
mm in warm areas (adapted from ICID MTD, 1996).
Subirrigation: Application of irrigation water below the ground surface by raising the
water table near or within the root zone.
Submerged flow: Flow through any critical depth-measuring structure where the
downstream water depth is high enough to interfere with establishment of critical
velocity at the control section. Submergence is usually expressed as the ratio of
downstream to upstream water level.
Submersible turbine pump: A type of vertical turbine pump powered by an in-line,
close-coupled electric motor that is immersed in the well or sump with the pump.
Subsurface irrigation or Subsurface drip irrigation: Application of water below
the soil surface through emitters from buried plastic tubing, with discharge rates
generally in the same range as surface drip irrigation. This method of water applica-
tion is different from and not to be confused with subirrigation where the root zone
is irrigated by water table control.
Surface irrigation: Any of a broad class of irrigation methods in which water is dis-
tributed over the soil surface by gravity flow.
Surface pipe outlet: An outlet for attaching a surface pipe to a riser without using a
portable hydrant.
Surface runoff: Precipitation, snow melt, or irrigation in excess of what can infiltrate
or be stored in small soil surface depressions.
Surface sealing: Reorienting and packing of dispersed soil particles in the immediate
surface layer of soil by water, with clogging of surface pores, resulting in reduced
infiltration.
Surface soil: The uppermost part of the soil, ordinarily moved in tillage, or its equiva-
lent in uncultivated soils, ranging in depth from 10 to 20 cm (4 to 8 in.). Sometimes
called the soil management zone.
Surface storage: Water held in small surface depressions when irrigation applications
or precipitation rates exceed intake rates. The depressions may be caused by tillage
or by natural variations in the soil topography. Also called depression storage or
retention storage.
Surfacing: The upward movement of water from buried microirrigation emitters to
the soil surface, often when application rates exceed hydraulic conductivity of the
surrounding soil. Can increase weed growth, move salts to the surface, increase soil
evaporation and increase fungal disease, but is sometimes appropriate to ensure
adequate wetting across the bed width for germination, fumigation, and/or proper
growth of shallow-rooted crops.
Surge anticipator valve: A valve that opens when the pressure drops below a speci-
fied set point; designed to be installed in a bypass line and provide protection to
pipelines and related equipment from surges that can occur when a pump is sud-
denly stopped. The valve regulators sense the precursor (pump power failure or a
low pressure wave) of a high-pressure wave and opens in anticipation of a returning
high-pressure wave or “water hammer.”
Surge chamber: A protective device for pressurized irrigation systems that serves to
remove sharp pressure spikes and create smooth, controlled pressure oscillations.
Surge irrigation: A surface irrigation technique wherein flow is applied to furrows
(or less commonly, borders) intermittently during a single irrigation set to increase
uniformity of infiltration.
Surge plug: A device used at the end of a sprinkler lateral to release high-pressure
surges that may occur if the lateral is filled too fast, thus preventing damage to the
lateral.
Sustainability: Environmentally, a characteristic of agricultural development that
meets the needs of the present without compromising the ability of future genera-
tions to meet their own needs. Economically, the ability to make enough profit to
continue operations over time, which is influenced by market, physical, regulatory,
financial, and political issues.
System head curve: A graphical description of the total head required to move water
through a pumping system as a function of water flow rate.
System: A combination of elements of equipment, technical information, and infra-
structure, the use of which, working in conjunction, is necessary to efficiently ac-
complish some objective.
Tailwater ratio: The volume of surface runoff from an irrigated field to the volume of
water delivered to the field.
Tailwater: (1) Water, in a stream or canal, immediately downstream from a structure.
(2) Excess irrigation water that reaches the lower end of a field.
Tensiometer: An instrument, consisting of a porous cup filled with water and con-
nected to a manometer or vacuum gauge, used for measuring the soil water matric
potential at tensions below about 80 centibars.
Tertiary treatment: Advanced cleaning of wastewater that goes beyond the secon-
dary or biological stage. Causes removal of nutrients such as phosphorous, nitrogen
and most suspended solids (from ICID MTD, 1996).
Throat: The constricted flow area in a hydraulic structure, such as a measuring flume.
Torpedoing: The dragging of a heavy, smooth cylinder along the length of an irriga-
tion furrow to smooth its surface, thus reducing resistance to flow.
Total dynamic head: The head required to pump water from its source to the point of
discharge; equal to the static lift, plus head losses in pipes and fittings, plus the in-
crease in velocity head.
Total evaporable water: The maximum depth of water that can be evaporated from
the surface soil layer when the layer has been initially completely wetted.
Total suction head: The head required to lift water from the water source to the cen-
terline of the pump plus velocity head, entrance losses, and friction losses in suc-
tion pipeline.
Trapezoidal flume: A calibrated, open-channel structure with sidewalls inclined to
the horizontal, used to measure the flow of water, using the principle of critical
flow.
Trash rack: A screen or grate at the intake of a channel, drain, or pump structure, for
the purpose of trapping debris.
Trickle irrigation: A method of microirrigation wherein water is applied to the soil
surface as drops or small streams through emitters. (Preferred term is drip irriga-
tion.)
Tubewell: A circular well, too small to be entered by a human, bored into the ground
to tap one or more aquifers (adapted from ICID MTD, 1996).
Turbine pump: A type of pump having one or more stages, each consisting of an
impeller on a vertical shaft, surrounded by stationary and usually symmetrical
guide vanes. Combines the energy-imparting characteristics of axial-flow and pro-
peller pumps.
Turbulent flow: Flow in which the fluid particles move in an irregular random man-
ner, in which the head loss is approximately proportional to the second power of
the velocity.
Turbulent or turbulent-path emitter: See Emitter, labyrinth.
Turnout: See Delivery box.
Unavailable soil water: That portion of water in a soil held so tightly by adhesion and
other soil forces that it cannot be absorbed by plants rapidly enough to sustain
growth; soil water content at the permanent wilting point.
Unconfined aquifer: An aquifer whose upper boundary consists of relatively porous
natural material that transmits water readily and does not confine water. The water
level in the aquifer is the water table.
Uniform flow: Flow in an open channel in which the velocity and depth are the same
at each cross section.
Uniformity coefficient: (1) In irrigation, a characterization of the areal distribution of
water in a field as the result of an irrigation. (2) In soils, the ratio of the D60 size
particles passing a screen to the D10 size of a granular material.
Unit stream: The amount of water, per unit of width, turned into each border strip or
basin during irrigation. Sometimes called a unit width stream.
Unsaturated zone: That part of the soil profile in which the voids are not completely
filled with water.
Unsaturated flow: Movement of water in soil in which the pores are not completely
filled with water.
Vadose zone: The portion of the Earth’s mantle between the land surface and the
phreatic (groundwater) zone; the biologically active volume of soil where most
plant root growth, nutrient uptake, and microbial activity occurs.
Valve: A device to control flow, as in pipelines. Valves used in pressurized systems
include:
Air relief valve: A device that releases air from a pipeline automatically without
permitting loss of water.
Air vacuum, air relief valve: A device that releases air from a pipeline automati-
cally without permitting loss of water or admits air automatically if the internal
pressure becomes less than atmospheric.
Check valve: A valve used in a pipeline to allow flow in only one direction.
Drain valve: (1) Automatic: A spring-loaded valve that will automatically open
and drain the line when the pressure drops to near zero. (2) Flushing type: A
valve on the end of a line for the purpose of flushing out dirt and debris. May be
incorporated into an end plug or end cap.
Foot valve: A check valve used on the bottom of a suction pipe to retain the water
in the pump when it is not in operation or to prevent backflow.
Pressure relief valve: A spring-loaded valve set to open at a pressure slightly
above the operating pressure, used to relieve excessive pressure and surges. See
Air relief valve.
Vacuum relief valve: A valve used to prevent a vacuum in pipelines and avoid
collapsing of thin-wall pipe.
Vapor pressure deficit: The difference between the existing vapor pressure and that
of a saturated atmosphere at the same temperature.
Variable frequency drive: A means of controlling pump speed, and thus discharge
rate or pressure, by allowing the frequency of the electrical supply to the pump mo-
tor to be varied.
Velocity head: the head (energy) due to the velocity of a moving fluid; equal to the
square of the mean velocity divided by twice the gravitational acceleration.
Vent: An appurtenance to a pipeline that permits the passage of air to or from the
pipeline.
Venturi flume: An open channel flow measuring device having a contracted throat,
such that flow depth can be related to discharge. (Preferred term for improved Ven-
turi flume is Parshall flume.)
Venturi injection system: A system that causes a liquid, typically containing fertiliz-
ers or pesticides in solution, to be drawn into the primary irrigation stream at a re-
gion of lowered pressure created when the irrigation stream flows through a con-
striction in the line.
Vertical turbine pump: A centrifugal pump having one or more impellers on a verti-
cal shaft. The impellers and bowls, typically suspended below the pumping water
surface in a sump or well, are powered by the shaft from a power unit at the point
of discharge.
Victaulic coupling: A device for joining pipes having plane ends, by compressing a
flexible gasket between metal parts.
Volatilization: Conversion of a liquid or solid to a gas, by evaporation or sublimation.
Sometimes a means by which concentrations of pollutants are transformed to a less
harmful state by dilution of the vapor in the atmosphere.
Volute case: The fixed housing within which the impeller of a centrifugal pump ro-
tates. Typically shaped to help convert input energy to discharge pressure. Suction
and discharge lines are attached to the volute case.
Wastewater: Water carrying dissolved or suspended solids from homes, farms, busi-
nesses, or industries (from ICID MTD, 1996).
Water amendment: (1) A fertilizer, herbicide, insecticide, or other material added to
water for the enhancement of crop production. (2) A chemical water treatment to
reduce emitter clogging.
Water application efficiency: The ratio of the average depth of water infiltrated and
stored in the root zone to the average depth of water applied by irrigation.
Water conservation: Protection and management of water resources for maximum
sustained benefits.
Water conveyance efficiency: The ratio of the volume of irrigation water delivered
by a distribution system to the volume of water introduced into the system.
Water depletion: The use or removal of water from an area, such as a watershed,
such that the water is permanently unavailable for further use.
Water harvesting: Any practice that enhances the runoff from a watershed or catch-
ment area for collection and beneficial use.
Water leveling: A method of land grading wherein fields are divided into segments,
flooded, and the highs are scraped down until all soil is beneath the water surface.
Typically used with rice production.
Water management: The planned development, distribution and use of water re-
sources, in accordance with predetermined objectives and considering both quantity
and quality of the water resources (adapted from ICID MTD, 1996).
Water pressure driven injector: A device to inject a chemical solution into the irri-
gation stream in a pipeline. Water drawn from the mainline above a flow constric-
tion powers a small turbine or piston device, which power the injector.
Water right: The legal right to use water supplies, derived from common law, court
decisions, or statutory enactments.
Water spreading: (1) Application of water to land to enhance infiltration and storage
as groundwater for subsequent withdrawal. (2) A specialized form of surface irriga-
tion accomplished by diverting flood runoff from natural channels or watercourses
and spreading the flow over relatively level areas.
Water storage efficiency: The ratio of the average depth of irrigation water infiltrated
and stored in the root zone to the soil water deficit.
Water stress coefficient: A coefficient used to reduce the crop coefficient in
evapotranspiration estimates under conditions of water stress.
Water table: The upper surface of a saturated zone below the soil surface where the
water is at atmospheric pressure.
Water use efficiency: (1) An index of the productivity of water. (2) Dry matter or
harvested portion of crop produced per unit of water consumed. (3) Ratio of water
beneficially used to the water delivered to the area being irrigated. (4) The increase
in production per unit increase in evapotranspiration (ET) due to irrigation, or
WUEi = (Yirr – Ydry)/(ETi –Etd). (5) The response of the total crop biomass (above-
ground dry matter) produced per unit mass of water taken up by the crop.
Water-holding capacity: The amount of soil water available to plants. See Available
soil water.
Waterlogging: A condition in which the water table is located at or near the soil sur-
face, such that crop growth is inhibited, or cropping is infeasible (adapted from
ICID MTD, 1996).
Water-stressed: When a country’s renewable water supplies drop below about 1700
cubic meters per capita, some analysts label the country as being water-stressed.
Weir: (1) A structure across a stream to control or divert the flow. (2) A device for
measuring the rate of flow of water. Weirs are commonly sharp-crested or broad-
crested with rectangular, trapezoidal, or triangular cross section. Weirs include:
Broad-crested weir: Weir for water measurement having a rounded or wide crest
in the direction of flow.
Cipolletti weir: A sharp-crested trapezoidal weir with sides that slope 1 (horizon-
tal) to 4 (vertical).
Contracted weir: A weir having sufficiently sharp upstream edges and slow ap-
proach velocity to cause the nappe to contract. Sometimes called a sharp-crested
weir.
Rectangular weir: An open channel flow measurement structure having a rectan-
gular opening. If the weir crest extends across the full width of the channel, it is
suppressed (side contractions are suppressed).
Suppressed weir: A flow measuring weir with sides flush with the walls of the
channel, thus eliminating (suppressing) contractions of the overflowing water.
Trapezoidal weir: A sharp-crested weir of trapezoidal shape. See Cipolletti weir.
Triangular weir: A sharp-sided, usually 90°, V-notch weir.
Weir head: The vertical distance from the crest of a weir to the water surface in the
forebay above the weir, not including the velocity head of approach.
Weir pond or box: The pond upstream from a weir, usually used to reduce the veloc-
ity of approach and allow for full contraction of flow for measurement purposes.
Well casing: The pipe installed within a borehole to prevent collapse of sidewall ma-
terial, to receive and protect the pump and pump column, and to allow water flow
from the aquifer to the pump intake.
Well development: The process of removing fine formation materials or materials
introduced during well construction from the well intake zone for the purpose of
stabilizing and increasing the permeability of the intake zone and any filter pack
materials.
Well efficiency: The ratio of theoretical drawdown to measured drawdown in a well
being pumped. Theoretical drawdown is estimated from adjacent observation well
data obtained during well tests.
Well intake zone: The portion of the well surrounding the well inlet which is modi-
fied by the well construction and development processes. This includes the space
between the well inlet and the undisturbed aquifer.
Well screen: That part of the well casing that has openings through which water en-
ters. See Screen and Perforated casing.
Well test: A determination of the relationship between well yield and drawdown with
time.
Well yield: The discharge rate that can be sustained from a well through some speci-
fied period of time. See Safe well yield.
Wetted perimeter: The length of the wetted contact between a flowing liquid and the
open channel or closed conduit conveying it, measured in a plane at right angles to
the direction of flow.
Wetted volume: The percentage of the total root zone volume (for row crops and per-
manent crops) that is wetted by irrigation.
Wilting point: See Permanent wilting point.
Wind drift: See Drift.
Wire-to-water efficiency: The ratio of electrical input power to output power of an
electric motor-driven pump; the product of the pump and the motor efficiencies; the
overall pumping plant efficiency.
WUCOLS (Water Use Classification of Landscape Species): A procedure devel-
oped in California for estimating evapotranspiration over a wide range of landscape
vegetation and environmental conditions (Costello and Jones, 1994).
Yield plateau: Maximum yield of a crop where all inputs are optimized; further in-
creases in inputs do not result in increased yields.
Zero-inertia equations: A simplification of the equations of fluid motion in which
acceleration terms are discarded, thus simplifying the equations’ solutions.
Zooplankton: Small animals that float passively near the surface of water bodies
REFERENCES
ASABE. 2007. S526.2 JAN2001: Soil and water terminology. St. Joseph, Mich.:
ASABE.
ASCE (ASCE Water Quality Tech. Committee, Kenneth K. Tanji, chair). 1990. Agri-
cultural Salinity Assessment and Management. 2nd ed. ASCE Manuals and Reports
on Engineering Practice No. 71. New York, N.Y.: American. Soc. Civil Engineers.
Costello, L. R., and K. S. Jones. 1994. WUCOLS Water Use Classification of Land-
scape Species: A Guide to the Water Needs of Landscape Plants. (Rev. ed.). Univ.
California Cooperative Extension.
ICID MTD. 1996. ICID Multilingual Technical Dictionary on Irrigation and Drain-
age. Rev. Ed. New Delhi, India: English-French. International Commission on Irri-
gation and Drainage (ICID).
SSSA (Soil Science Society of America). 2001. Glossary of Soil Science Terms.
Madison, Wisc.: SSSA.
USDA. 1954. Diagnosis and Improvement of Saline and Alkali Soils. Agricultural
Handbook No. 60, USDA.
APPENDIX B
ANNOTATED
BIBLIOGRAPHY OF
IRRIGATION STANDARDS
Allen R. Dedrick (USDA-ARS, Beltsville,
Maryland)
Abstract. This appendix annotates all ASABE and ISO standards pertaining to irri-
gation and drainage used in combination with irrigation. Additionally, irrigation and
drainage-related Engineering Practices in ASABE are included as reference. A few
ISO Technical Reports are referenced when applicable. ASABE standards on drainage
are included in the references since, in many instances, drainage is irrigation related.
Those ASABE standards approved as American National Standards are identified. The
importance of standards in the irrigation field is discussed and sources of irrigation
standards for irrigation are presented.
Keywords. ANSI, ASABE, ASAE, Drainage, Irrigation, ISO, Microirrigation,
Sprinkler irrigation, Standard, Standardization, Surface irrigation.
B.1 INTRODUCTION
Standards represent the state of the art in their respective areas, and can be impor-
tant tools for people who select, design, purchase, install, operate, and evaluate irriga-
tion equipment or systems (Solomon and Dedrick, 1995, 2001). Suppliers, designers,
advisors, and end users should be aware of available standards on various equipment
and procedural issues. In covering important topics of significance to everyone inter-
ested in irrigation, standards:
address materials and construction requirements for equipment items;
identify the key parameters used to quantify equipment performance;
set requirements for acceptable or qualifying performance levels;
establish methods for testing, data analysis, and interpretation; and
establish methods and requirements for performance reporting.
Such standardized methods ensure that product information is comparable. Without
comparable performance reports, product evaluation and selection would be hindered,
and designers would never be sure what values for hydraulic or other relevant charac-
teristics should be factored into their designs.
Standards are important for both suppliers and consumers. They are often used to
specify or pre-qualify equipment that will be considered for purchase. Standards
clauses on data interpretation provide the basis for a common understanding of per-
formance evaluation and criteria for categorization. Using standards to guide purchase
decisions provides some level of protection for end users and suppliers. End users are
assured that minimum performance and safety criteria are satisfied by products meet-
ing the standard
Standards for field evaluation and acceptance provide protection for both supplier
and consumer against either acceptance of substandard goods or improper rejection of
acceptable goods. Field evaluation standards further aid irrigation advisors and users
by helping to identify and quantify problems and suggesting possible causes.
The irrigation community should participate in and carefully observe the standards
development process. Well-written standards are very beneficial, but poorly written or
out-of-date standards can lead to serious problems. A standard should focus on the
results required, not on how the results may be achieved. Specifying “how” can lead to
exclusionary language limiting design or manufacturing innovation or prohibiting
whole classes of “nonstandard” options for achieving the necessary results. For exam-
ple, if a standard specified an in-line emitter/lateral configuration, it could exclude the
possibility of on-line emitters, laterals with integral emitters, or porous pipe products.
Care, also, must be taken in specifying testing methods to avoid methods that may
be inappropriate to a class of products, too expensive for common use, and/or re-
quirements that may be too lax or too harsh. For example, a standard may specify that
a pressure measurement be taken at least 10 diameters downstream of a flow-
disturbing element, since otherwise, reliable readings are not possible. The method of
measuring pressure (manometer or gage) should be left to the discretion of the test
laboratory. Any device with the necessary accuracy (specified in the test method stan-
dard) should be acceptable.
The best safeguard against inappropriate standards is the active participation of
knowledgeable experts in the review of documents as they progress through all stages
of development—including the periodic reviews done at least every five years.
B.2 SOURCES OF STANDARDS
B.2.1 American Society of Agricultural and Biological Engineering (ASABE)
The ASABE is an educational and scientific organization dedicated to the ad-
vancement of engineering applicable to agricultural, food, and biological systems.
Founded in 1907 as ASAE and headquartered in St. Joseph, Michigan, ASABE com-
prises 9000 members in more than 100 countries. ASABE is the key organization
within the U.S. for irrigation standards (ASAE, 2005). Standards development is a
major priority area for the ASABE to help meet the Society objectives of advancing
the theory and practice of engineering in agricultural, food and biological systems.
Stetson (1982) reviewed the history of standards within ASABE.
A 14-member Board of Trustees, who ensures ASABE fulfills its missions, leads
ASABE. Activities of the Society are carried out by its four councils: Membership,
Meetings, Publications, and Standards, with input from eight Divisions or Institutes,
which represent ASABE’s technical specialty areas. The Soil and Water Division, one
840 Appendix B Annotated Bibliography of Irrigation Standards
of the eight, develops most standards and engineering practices pertaining to irrigation
and drainage. The Soil and Water Division is comprised of six technical groups, in-
cluding the Irrigation Group (SW-24). The group is subdivided into one administrative
and six technical committees. The Soil and Water Standards Committee (SW-03) pro-
vides oversight to the development of all soil and water standards.
The development of standards is the responsibility of a technical group, e.g., SW-
24. It determines whether the proposal is within the division’s area of interest, a draft
should be developed in light of existing standards, or other Standards Development
Organizations (SDOs) are considering subjects related to the proposed activities. If a
standard is merited, the technical group assigns responsibility for developing the draft
standard, usually to a technical committee, and assures that the technical group com-
pletes the task in a timely manner, according to the standards development procedures
(ASAE, 2005).
B.2.2 American National Standards Institute (ANSI)
ANSI is the national organization that coordinates the voluntary development of
national standards and establishes criteria for consensus standards development. ANSI
is a private, nonprofit organization originally formed in 1918 as the American Engi-
neering Standards Committee (Dedrick, 1982). Through its coordinating role, and by
establishing guidelines for uniform standardization procedures, ANSI adds credibility
to our national standardization program. They assure that consensus has been achieved
and that standards have been prepared using due process. ANSI requires that all con-
cerned national interests have had an opportunity to review or comment on a docu-
ment, including the right to appeal actions at various levels of review.
ANSI does not itself develop standards; ASABE and similar organizations develop
standards for their particular areas of interest and expertise. ANSI accredits ASABE
and similar organizations that develop American National Standards. Accreditation
includes periodic audits to verify compliance to established procedures. Standards
developed by accredited organizations may become American National Standards
once ANSI has determined that development was in accordance with ANSI procedural
guidelines (Dedrick, 1982). ANSI helps to avoid the conflict, overlap, and duplication
that might result if organizations developed standards independently.
ANSI is the U.S. representative to the International Organization for Standardiza-
ton (ISO) and authorizes U.S. representation on the various technical committees and
subcommittees developing international standards. ANSI has authorized the Irrigation
Association1 (IA) to serve as the U.S. administrator for ISO Technical Committee 23,
Subcommittee 18 (ISO/TC23/SC18). ISO/TC23/SC18 is the primary group responsi-
ble for developing irrigation and drainage standards for ISO.
B.2.3 International Organization for Standardization (ISO)
ISO is an international agency for standardization, headquartered in Geneva, Swit-
zerland. Members of ISO are the national standards bodies (like ANSI for the U.S.)
from about 90 countries (Dedrick, 1986). The objectives of ISO are to advance
worldwide standardization; facilitate international exchange of goods and services;
1
The Irrigation Association (IA) is a trade organization whose domestic membership represents numerous
segments of the irrigation community in the United States (IA, 2005). Members include equipment manu-
facturers, distributors and dealers, contractors, consultants and advisors, and individual members, including
representatives of universities, and local, state, and federal government agencies.
and promote cooperation in intellectual, scientific, technological, and economic activ-

ity. ISO comprises over 2000 technical committees, subcommittees, and working
groups. Dedrick (1982, 1986) gives additional background on ISO.
ISO/TC23/SC18 is the subcommittee for irrigation and drainage standards.
Penkava (1986) reviewed the early history of SC18. Countries active in SC18 over the
years include Israel, France, the U.S., Canada, Spain, Hungary, and Mexico. Other
countries that have participated in the past are Belgium, the People’s Republic of
China, Denmark, Germany, Italy, and the United Kingdom. The Secretariat and Chair
of SC18 are held by Israel. A number of the ISO standards have been based on
ASAE/ASABE Standards.
B.3 IRRIGATION STANDARDS
This section identifies ASABE standards, national standards (ANSI), and interna-
tional standards (ISO) pertaining to irrigation and drainage. The ASABE standardiza-
tion program includes engineering practices. The categories are defined as follows
(ASAE, 2005):
Standard. A definite terminology, specification, performance criteria, or procedure
providing interchangeability; enhancing quality, safety, economy or compatibility;
and/or providing proper and adequate model or example. Standards may include (1)
definitions, terminology, graphic symbols, and abbreviations; (2) performance criteria
for materials, products, or systems; (3) testing procedures; and/or (4) specifications or
ratings regarding size, mass, volume, etc.
Engineering Practice. A practice, procedure, or guide accepted as appropriate,
proper, and desirable for general use in design, installation, or utilization of systems or
system components, and based upon current knowledge and the state of the art.
The ASABE standards and engineering practices are available by contacting
ASABE (www.asabe.org). International standards are available through ANSI (www.
ansi.org) or ISO (www.iso.org).
All standards, engineering practices, or technical reports pertaining to irrigation or
drainage used in combination with irrigation are identified in the following tables. The
information is a status report as of February, 2006. The reader is directed to the vari-
ous organizations identified above for more detailed information, either about all in-
cluded standards, the currently available standards published on irrigation, or the stan-
dardization process used within each organization.
Each standard, engineering practice, or technical report is identified by the develop-
ing organization using a number; date; title; and brief description or scope. ASABE2
uses a month/year designation for date (e.g., DEC01), while ISO uses year, only. An
ANSI designation of an ASABE standard indicates the standard has been approved by
ANSI as an American National Standard (e.g. ANSI/ASAE S261.7 DEC01). The “r”
in the Sxxx.r designation for ASAE/ASABE standards indicates the revision number.
The tables below list general irrigation standards applicable to several irrigation
system types (Table B.1), surface irrigation standards (Table B.2), sprinkler irrigation
standards (Table B.3), microirrigation standards (Table B.4), and drainage standards
used in combination with irrigation systems (Table B.5).
2
All standards developed as ASAE standards will continue to be designated as ASAE, even if revised. All
new standards (post July 2005) will be designated as ASABE.
Table B.1. General irrigation standards applicable to several irrigation system types.
Number and Title Description or Scope
ANSI/ASAE S261.7 This Standard is intended as a guide to engineers in the design and
DEC01 installation of low or intermediate pressure non-reinforced concrete
Design and installation of irrigation pipelines and for the preparation of detailed specifica-
non-reinforced concrete tions for a particular installation. It is restricted to pipelines with
irrigation pipe systems vents or stands open to the atmosphere or closed pipelines operat-
ing at less than 6 m of head.
ASAE EP267.7 FEB03 The following principles and practices are recommended for pre-
Prevention of mosquito vention of mosquito production sources (i.e., breeding habitat)
problems associated with associated with irrigation and drainage systems in humid, semiarid,
irrigation and drainage and arid areas. Although these practices apply to all mosquito spe-
systems cies, the primary targets are those that transmit diseases to humans
and animals. Information herein is directed towards engineering
practices. Chemical and biological control measures are also part
of a complete control program, and more complete, detailed infor-
mation on their use should be obtained from appropriate sources.
ASAE EP329.1 MAR02 This Engineering Practice concerns 3.7 kW and larger single-phase
Single-phase rural distri- motors, and three-phase motors with phase converters, that are
bution service for motors started infrequently (generally with less than 6 starts in a 24-hour
and phase converters period and not more than 1 start between 6 p.m. and midnight) and
that serve loads with non-pulsating power requirements. Some
examples are motors for fans on crop driers, feed grinders, silo
unloaders, irrigation pumps, and auger feeders.
ANSI/ASAE S362.2 This Standard provides detailed information for the application of
JAN83 electrical apparatus to electrically driven or controlled irrigation
Wiring and equipment for machines, covering all electrical equipment, apparatus, compo-
electrically driven or con- nents, and wiring necessary from the point of connection of electric
trolled irrigation machines power to the machine.
ANSI/ASAE S376.2 This Standard applies to underground, thermoplastic pipelines used
FEB04 in the conveyance of irrigation water to the point of distribution
Design, installation and and may or may not apply to potable water systems. Plastic pipe
performance of under- materials addressed in the standard include polyvinyl chloride
ground, thermoplastic irri- (PVC), chlorinated poly(vinyl chloride) (CPVC), and polyethylene
gation pipelines (PE).
ANSI/ASAE S397.2 This Standard provides a common document for use by all those
FEB03 involved in electrical irrigation systems. It applies to three-phase,
Electrical service and 240 V, or 480 V service for irrigation pump motors, irrigation ma-
equipment for irrigation chines, and auxiliary equipment.
ASAE EP 400.2T3 JAN94 This Tentative Engineering Practice is directed to wells constructed
Designing and constructing to obtain ground water for irrigation purposes; however, many of
irrigation wells the details presented herein also are suitable for domestic, munici-
pal, and industrial wells. It is intended as a guide for preparing
specifications for irrigation well construction.
(continued)
3
Classification of “tentative” may be assigned to any new standard, if sufficient justification for approval
exists pending clarification or introduction of minor changes. The tentative classification shall be used when
provisions of the standard are based on new technology that may be beyond or lead the current state of the
art, or when “hardware” required for conformance with the standard is not readily available.
Table B.1 continued.

ASAE EP 409.1 FEB04 This Engineering Practice specifies necessary safety devices to be
Safety devices for chemi- used when injecting liquid chemicals into irrigation systems from a
gation dedicated irrigation water source. This Engineering Practice de-
scribes safety devices that will prevent three specific hazards: (1)
an unexpected shutdown of the irrigation pumping plant due to
mechanical or electrical failure while it is unattended, (2) an irriga-
tion pumping plant shutdown failure while the injection equipment
continues to operate, and (3) the chemical injection system stop-
ping while the irrigation pump continues to operate. It is not in-
tended that this Engineering Practice apply for irrigation systems
connected to a public or private water distribution system used as a
municipal, industrial, or residential water supply.
ASAE S491 FEB03 The purpose of this Standard is to establish a common set of sym-
Graphic symbols for pressur- bols that represent equipment components and can be used in the
ized irrigation system design design and planning of pressurized irrigation systems.
ASAE EP505 APR04 This Engineering Practice establishes minimum recommendations
Measurement and reporting for measurement, reporting, siting, operation, maintenance, and
practices for automatic data management procedures for automatic agricultural weather
agricultural weather Sta- stations. It applies to automatic weather stations installed individu-
tions ally, or as part of a network of stations, for the measurement and
reporting of specific weather variables in agricultural environments
and includes a recommended core set of measurements and general
siting considerations for agricultural weather stations.
ISO 7714:2000 This International Standard specifies general requirements and test
Agricultural irrigation methods for volumetric valves able to deliver automatically preset
equipment—Volumetric quantities of water. It is applicable to valves actuated by pipeline
valves—General require- pressure and flow alone, and which do not need any other external
ments and test methods source of energy.
ISO/TR 8059:1986 This Technical Report (TR) deals with automatic irrigation systems
Irrigation equipment— based on hydraulic devices using only the energy that can be ob-
Automatic irrigation sys- tained from the water in the irrigation system; gives main defini-
tems—Hydraulic control tions and a classification of these systems; and applies to automatic
control systems, in which the control of water application is
achieved by means of water quantity measurement.
ISO 9625:1993 This International Standard specifies the requirements and test
Mechanical joint fittings methods for mechanical fittings intended to join polyethylene (PE)
for use with polyethylene pressure pipes used in aboveground and underground irrigation
pressure pipes for irriga- systems conveying water at temperatures not exceeding 45°C. This
tion purposes International Standard does not apply to fittings used in drip irriga-
tion systems.
ISO 9635:1990 This International Standard specifies construction and performance
Irrigation equipment— requirements and test methods for hydraulically operated valves,
Hydraulically operated intended for operation in irrigation systems which may contain fertil-
irrigation valves izers and chemicals of the types and concentrations used in agriculture.
ISO 9644:1993 and This International Standard specifies a test method to determine the
Amd 1:1998 pressure loss in irrigation valves under steady-state conditions
Agricultural irrigation when water flows through them. The valve performance specifica-
equipment—Pressure tions presented are of such scope and accuracy as to assist irriga-
losses in irrigation valves tion system designers when comparing pressure losses through
— Test method various types of irrigation valves.
(continued)

ISO 9911:1993 This International Standard specifies the general requirements and
Agricultural irrigation test methods for manually operated small plastics valves intended
equipment—Manually for operation in agricultural irrigation systems.
operated small plastic
valves
ISO 9912-2:1992 This part of IS0 9912 specifies the general construction require-
Agricultural irrigation ments and test methods for automatic self-cleaning strainer-type
equipment—Filters—Part filters intended for operation in agricultural irrigation systems. It
2: Strainer-type filters does not cover filtration ability, efficiency, and capacity.
ISO 9912-3:1992 This part of IS0 9912 specifies the general construction require-
Agricultural irrigation ments and test methods for automatic self-cleaning strainer-type
equipment—Filters— filters intended for operation in agricultural irrigation systems. It
Part 3: Automatic self- does not cover filtration ability, efficiency, and capacity.
cleaning strainer-type
filters
ISO 9952:1993 This International Standard specifies the construction and perform-
Agricultural irrigation ance requirements, and test methods for check valves, intended for
equipment—Check valves operation in agricultural irrigation systems that may contain fertil-
izers and chemicals of types and concentrations used in agriculture.
ISO 10522:1993 This International Standard specifies construction and performance
Agricultural irrigation requirements and test methods for direct-acting pressure-regulating
equipment—Direct-acting valves intended for operation in irrigation systems that may contain
pressure-regulating valves fertilizers and chemicals of types and in concentrations commonly
used in agricultural irrigation.
ISO 11419:1997 This International Standard gives the structural and operational
Agricultural irrigation requirements and test methods for float-type air release valves
equipment—Float type air intended for operation in irrigation systems that may contain
release valves chemicals of the types and concentrations generally used in agri-
cultural irrigation.
ISO 11678:1996 This International Standard specifies minimum required properties
Agricultural irrigation and test methods for aluminum tubes intended for use in agricul-
equipment—Aluminum tural irrigation systems for the transport of water for irrigation
irrigation tubes purposes. It applies to hand-moved and towed tubes, and to tubes
intended for stationary or temporary installation.
ISO 11738:2000 This International Standard specifies requirements for the compo-
Agricultural irrigation nents and method of installation of pressurized irrigation system
equipment—Control heads control heads with a nominal size of up to and including 200 mm.
This International Standard is applicable only to the above-ground
components of irrigation control heads for sprinkler irrigation and
microirrigation (mini-sprinklers, drip irrigation, etc.). It is applica-
ble to the basic irrigation control head, on which other irrigation
control and command components (electrical, electronic, and hy-
draulic) may be assembled, but does not deal with these additional
components.
ISO 13457:2000 This International Standard specifies construction and operational
Agricultural irrigation requirements and test methods for water-driven chemical injector
equipment—Water-driven pumps. These water-driven injector pumps are used to inject
chemical injector pumps chemicals into irrigation systems. The chemicals include liquid
fertilizers and solutions of fertilizers and other soluble agricultural
chemicals such as acids, pesticides, and herbicides
(continued)

ISO 13460:1998 This International Standard specifies the required properties and
Agricultural irrigation test methods for plastics saddles for assembly on polyethylene
equipment—Plastics pressure pipes used in aboveground and underground irrigation
saddles for polyethylene systems.
pressure pipes
ISO 15081:2005 This International Standard establishes conventional basic graphi-
Agricultural irrigation cal symbols for use on drawings and diagrams concerning the in-
equipment—Graphical stallation of pressurized agricultural irrigation systems.
symbols for pressurized
irrigation systems
ISO/TR 15155:2005 This Technical Report is intended to provide guidance on the de-
Test facilities for agricul- sign, selection, installation, and use of the equipment required to
tural irrigation equipment establish basic test facilities for irrigation equipment evaluation. Its
purpose is to provide information sufficient to complement the
detailed procedures included in ISO 9261, ISO 15886, ISO 7714,
ISO 9635, ISO 9644, ISO 9911, ISO 9952 and ISO 10522 for the
testing of agricultural irrigation system components, specifically:
emitters, sprinklers and valves.
ISO 15873:2002 This International Standard specifies the construction of, and op-
Agricultural Irrigation erational requirements and test methods for, differential pressure
equipment—Differential Venturi-type liquid additive injectors—a component of systems
pressure Venturi-type used to inject chemicals, including liquid fertilizers, liquid solu-
liquid additive injectors tions of water-soluble fertilizers, acids, caustics, pesticides, herbi-
cides, and other liquid additives into irrigation systems. This Inter-
national Standard does not specify means of preventing backflow
of liquid additives to potable water supply systems, the assembly
of such means near to the Venturi injector being covered by water
protection regulations.
Table B.2. Surface irrigation standards.

ANSI/ASAE S289.2 This Standard provides standards and specifications for the instal-
FEB04 lation of concrete slip-form canal linings. The scope is limited to
Concrete slip-form canal establishing standard dimensions for trapezoidal shaped canal sec-
linings tions. These standard sections will enable manufacturers to stan-
dardize excavating and lining equipment, thus, reducing costs and
assuring quality control. This standard is restricted to irrigation
canals that have a bottom width less than 1.8 m, and a total depth
of lined section not greater than 2.1 m.
ASAE EP408.2 MAR95 This Engineering Practice is a guide to engineers, technicians and
Surface irrigation runoff farmers designing and operating systems to collect runoff from
reuse systems surface irrigated fields for subsequent use on the same or other
fields.
ASAE EP419.1 FEB03 The purpose of this Engineering Practice is to establish and stan-
Evaluation of irrigation dardize procedures for evaluating the performance of individual
furrows irrigation furrows. Such evaluations should be aimed at improving
the design, operation, and management of furrow irrigation sys-
tems.
ISO 16149:2006 This International Standard specifies the requirements for unplasti-
Agricultural irrigation cized polyvinyl chloride (PVC) piping, used to supply and to dis-
equipment—PVC above- tribute low-pressure irrigation water through gates. It is applicable
ground low-pressure pipe to PVC piping with diameters from 50 to 315 mm operating at low
for surface irrigation— pressures and exposed to sunlight.
Specifications and test
methods
Table B.3. Sprinkler irrigation standards.

ANSI/ASABE S395 This Standard is intended to improve the degree of personal safety
MAR79 for operators and others during the normal application, operation
Safety for self-propelled, and service of self-propelled, hose-drag agricultural irrigation
hose-drag agricultural irri- systems.
gation systems
ASABE S 398.1 JAN01 This procedure is used only to determine the radius of throw of
Procedure for sprinkler single sprinklers. No attempt is made here to define product use,
testing and performance design or application procedures; to determine uniformity of dis-
reporting tribution; or the uniformity of the radius of throw throughout the
area of coverage.
ANSI/ASABE S436.1 The purpose of this Standard is to define a method for characteriz-
DEC01 ing the uniformity of water distribution of sprinkler packages in-
Test procedure for deter- stalled on center pivots and lateral move irrigation machines. This
mining the uniformity of test produces data to be used in computing the coefficient of uni-
water distribution of center formity, which can assist in system design and/or selection, and
pivot and lateral move irri- can be used to quantify certain aspects of system performance in
gation machines equipped the field.
with spray or sprinkler noz-
zles
(continued)

ISO 7749-1:1995 This part of IS0 7749 specifies the design and operational re-
Agricultural irrigation quirements of rotating sprinklers and sprinkler nozzles for agricul-
equipment—Rotating sprin- tural irrigation equipment and their test methods. It applies to
klers—Part 1: Design and sprinklers intended for assembly in pipeline networks for irriga-
operational requirements tion and operation at the pressures recommended by the manufac-
turer.
ISO 8026:1995 and This international Standard specifies the general requirements and
Amd 1:2000 test methods for irrigation sprayers. It applies to sprayers intended
Agricultural irrigation for assembly in pipeline networks for irrigation and for operation
equipment—Sprayers— with irrigation water.
General requirements and
test methods
ISO 8224-1:2003 This part of ISO 8224 specifies the operational characteristics of,
Traveler irrigation ma- and laboratory and field test methods for, traveler irrigation ma-
chines—Part 1: Operational chines. It includes (1) user-oriented technical information for in-
characteristics and labora- clusion in the manufacturer’s accompanying product literature, (2)
tory and field test methods laboratory test procedures for evaluating the uniformity of water
application on an irrigated strip by a machine operating within a
specified range of conditions and for determining the maximum
traveling rates the drive mechanism is able to achieve in response
to specified operating conditions, and (3) field test procedures for
determining the uniformity of water application on a given irri-
gated strip under local conditions prevailing in the field at time of
testing. It is applicable only to traveler irrigation machines and not
to other types of irrigation machine such as center pivot and mov-
ing lateral irrigation machines.
ISO 8224-2:1991 This part of IS0 8224 specifies test methods for specific physical
Traveler irrigation ma- properties and accelerated durability tests for softwall irrigation
chines—Part 2: Softwall hose and couplings used with irrigation machines. It applies to
hose and couplings—Test such hose and couplings used with stationary hose-coiling, mobile
methods hose-laying or hose-coiling, and mobile hose-dragging irrigation
machines used in agriculture and forestry.
ISO 11545:2001 This International Standard specifies a method for determining the
Agricultural irrigation uniformity of water distribution in the field from center-pivot and
equipment—Center-pivot moving lateral irrigation machines equipped with sprayer and
and moving lateral irriga- sprinkler nozzles. This International Standard is not applicable to
tion machines with sprayer the evaluation of center-pivot irrigation machines equipped with
or sprinkler nozzles— various corner arm application devices.
Determination of uniform-
ity of water distribution
ISO 12374:1995 This International Standard provides detailed information for the
Agricultural irrigation— application of electrical apparatus to electrically driven or con-
Wiring and equipment for trolled agricultural irrigation machines, covering all electrical
electrically driven or con- equipment, apparatus, components, and wiring necessary from the
trolled irrigation machines point of connection of electric power to the machine.
(continued)

ISO 15886-1:2004 This part of ISO 15886 defines terms used in relation to sprinklers
Agricultural irrigation intended for agricultural irrigation and provides a means of classi-
equipment—Sprinklers— fying those sprinklers according to physical factors, characteristics
Part 1: Definition of terms of water spray, mechanism for operation and water distribution,
and classification approach to sealing, intended use, and additional functions incor-
porated into the sprinkler. Its scope is intentionally broad to cover
the widest possible range of sprinkler construction, performance
and intended-use alternatives.
ISO 15886-3:2004 This part of ISO 15886 specifies the conditions and methods used
Agricultural irrigation for testing and characterizing the water distribution patterns of
equipment—Sprinklers— sprinklers intended for agricultural irrigation. It deals both with
Part 3: Characterization of indoor and outdoor, radial and full grid tests and is organized to
distribution and test meth- deal first with conditions common to all the tests and then with
ods those unique to indoor and outdoor testing, respectively. The term
sprinkler is used here in a broad, generic sense, as defined in ISO
15886-1, with the intent of covering the wide variety of products
classified in ISO 15886-1. The specific performance measure-
ments addressed in this part of ISO 15886 include distribution
uniformity, wetted radius and water jet trajectory height: it is ap-
plicable to all irrigation sprinkler classifications for which uni-
formity of application, wetted radius and water jet trajectory
height evaluation are required for the design objectives, as defined
by the manufacturer.
Table B.4. Microirrigation standards.

ASABE EP405.1 FEB03 The purpose of this Engineering Practice is to establish minimum
Design and installation of recommendations for the design, installation and performance of
microirrigation systems microirrigation systems: including trickle, drip, subsurface, bub-
bler, and spray irrigation systems. Provisions of this Engineering
Practice are primarily those that affect the adequacy and uniform-
ity of water application, filtration requirements, water treatment,
and water amendments.
ASABE S435 FEB04 This Standard covers requirements and methods for testing of
Polyethylene pipe used for polyethylene (PE) pipe made in standard dimensions for drip irri-
microirrigation laterals gation. Included are criteria for classifying PE plastic materials and
PE pipe, a system of nomenclature for PE plastic pipe, and re-
quirements and methods of test for materials, workmanship, di-
mensions, sustained pressure, burst pressure, and environmental
stress crack resistance. Methods of marking are also given.
ANSI/ASAE S539 FEB03 This Standard defines a procedure to collect irrigation media filter
Media filters for test data and provides procedures to classify and characterize me-
irrigation—Testing and dia filter test data from manufacturers and independent testing
performance reporting laboratories. It establishes a consistent basis to validate and sup-
port manufacturer’s statements on the performance, reliability,
safety, and long-term effectiveness of individual irrigation media
filtration systems. Sufficient data are to be developed so irrigation
system designers and others can evaluate the suitability of a par-
ticular filter system for a specific application.
(continued)

ANSI/ASABE S553 This Standard specifies testing methods, performance require-
MAR01 ments, and data to be supplied by the manufacturer for collapsible
Collapsible emitting hose emitting hose products with discrete emission points along their
(drip tape)—Specifications lengths, commonly referred to as “drip tape,” and herein referred
and performance testing to as “collapsible emitting hose.” This Standard applies to col-
lapsible emitting hose intended for irrigation of which the emitters
form an integral or permanently attached part. It does not apply to
tubing that is porous along its entire length.
ISO 8779:2001 This International Standard specifies the test methods and required
Polyethylene (PE) pipes for properties for pipes made from polyethylene (PE), and intended to
irrigation laterals— be used for irrigation laterals. This International Standard is appli-
Specifications cable to pipes designated as PE 32, PE 40, and PE 63 in accor-
dance with ISO 12162:1995, having a nominal pressure of PN 2, 5,
PN 4, PN 6, and PN 10, a nominal outside diameter from 12 to 32
mm inclusive and intended to be used for the conveyance of water
under pressure at temperatures up to for irrigation purposes.
ISO 8796:2004 This International Standard specifies a method of determining the
Polyethylene PE 32 and susceptibility to environmental stress cracking of polyethylene
PE 40 pipes for irrigation pipes intended for use with insert-type fittings. It is applicable to
laterals—Susceptibility to pipes conforming to ISO 8779, manufactured from PE 32 and PE
environmental stress crack- 40 materials and intended for use with insert-type fittings.
ing induced by insert-type
fittings—Test method and
requirements
ISO 9261:2004 This International Standard gives mechanical and functional re-
Agricultural irrigation quirements for agricultural irrigation emitters and emitting pipes,
equipment—Emitters and and, where applicable, their fittings, and provides methods for
emitting-pipe—Specifica- testing conformity with such requirements. It is applicable to emit-
tion and test methods ters, emitting and dripping (trickling) pipes, hoses, including col-
lapsible hoses (“tapes”) and tubing of which the emitting units
form an integral part, to emitters and emitting units with or without
pressure regulation and with flow rates not exceeding 24 L/h per
outlet (except during flushing), and to fittings dedicated to the
connection of emitting pipes, hoses, and tubing. It is not applicable
to porous pipe (pipe that is porous along its entire length), nor does
it cover the performance of pipes as regards clogging.
ISO 9912-1:2004 This part of ISO 9912 defines terms used in relation to filters in-
Agricultural irrigation tended for agricultural microirrigation systems—in particular,
equipment—Filters for pressurized systems—and provides a means of classifying those
micro-irrigation—Part 1: filters according to filtration method, structure, operating principle,
Terms, definitions and and function. It does not deal with classification according to the
classification type of water intended to be filtered; nor does it apply to the classi-
fication of filters for potable or domestic water use.
Table B.5. Drainage standards used in combination with irrigation systems.

ANSI/ASABE EP302.4 This Engineering Practice is intended to improve the design, con-
FEB03 struction, and maintenance of surface drainage systems that are
Design and construction of adapted to modern farm mechanization.
surface drainage systems
on agricultural lands in
humid areas
ASABE EP369.1 DEC87 This Engineering Practice sets forth principles and practices useful
Design of agricultural to engineers in the planning and design of pumping plants for drain-
drainage pumping plants age of agricultural land. It is not intended as complete specifications
and does not include pumping plants for deep well drainage.
ASABE EP407.1 FEB03 The purpose of this Engineering Practice is to provide planning,
Agricultural drainage design, construction and maintenance information and criteria for
outlets—Open channels agricultural drainage outlets by means of open channels. This
Engineering Practice is compatible with ASABE Engineering
Practice EP302.3, Design and construction of surface drainage
systems on agricultural lands in humid areas, and ASABE Engi-
neering Practice EP260, Design and construction of subsurface
drains in humid areas.
ASABE EP463.1 FEB03 This Engineering Practice is intended as a guide to engineers in
Design, construction, and the design and construction of subsurface drains in arid and semi-
maintenance of subsurface arid regions where irrigation is often used to provide adequate
drains in arid and semiarid water for crops. This Engineering Practice is intended to comple-
areas ment ASABE Engineering Practice, ASABE EP260.4, Design and
construction of subsurface drains in humid areas.
ASABE EP479 MAR90 This Engineering Practice is intended as a guide for the design,
Design, installation and installation, and operation of a system of subsurface drainpipes,
operation of water table head control structures, and water conveyance facilities whose
management systems for purpose is the efficient management of the free water surface in
subirrigation/controlled the soil for subirrigation and drainage. This Engineering Practice
drainage in humid regions is limited to systems utilizing pipe drains to effect drainage and
supply subsurface water.
B.4 REFERENCES
ASAE. 2005. ASAE Standards 2005. 52nd ed. St. Joseph, Mich.: ASAE.
Dedrick, A. R. 1982. Progress through standards: Its dynamic future, national and
international. ASAE Paper No. 82-5544. St. Joseph, Mich.: ASAE.
Dedrick, A. R. 1986. International standards for irrigation and drainage equipment:
Why develop them. ASAE Paper No. 86-5525. St. Joseph, Mich.: ASAE.
IA (The Irrigation Association). 2005. The Irrigation Association 2005-2006
Membership Directory and Buyer’s Guide. Fairfax, Va.: IA.
Penkava, F. F. 1986. New developments in international standards. The Irrigation
Association’s Irrigation News X(4): 3-5.
Solomon, K. H., and A. R. Dedrick. 1995. Standards development for microirrigation.
In Proc. of the 5th Int’l Microirrigation Congress. Published as CATI publ.
#950601. Available at: www.wateright.org/950601.asp.
Solomon, K. H., and A. R. Dedrick. 2001. Standards benefit developing irrigation
markets. Agric. Mechanization in Asia, Africa and Latin America 32(2): 48-54.
Stetson, L. E. 1982. Setting the standards. Agric. Eng. 63(5): 12-13.

Design and Operation of Farm Irrigation Systems 2nd Edition PDF

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Design and Operation of Farm Irrigation Systems 2nd Edition PDF

Hochgeladen von

Copyright:

Verfügbare Formate

Design and

This publication may not be reproduced

Copy editing by Peg McCann

ASABE is an educational and scientific

For more information visit www.asabe.org.

The American Society of Agricultural and Biological Engineers

International Standard Book Number (ISBN) 1-892769-64-6

5.2 Irrigation Scheme Physical Models ..................................................................... 109

16.13 Summary ......................................................................................................... 627

20.3 Nutrients in Wastewater and Reclaimed Water ................................................. 764

Figure 1.1. Increase in irrigated area worldwide from 1800 to 2000

Table 1.1. Irrigated areas in the top 25 countries, by continents, and in

Table 1.3 . Comparison of irrigation methods in the U.S.

1.3.3 Trends in U.S. Irrigation

1.4 ISSUES FACING IRRIGATED AGRICULTURE

commitment to minimize such environmental issues continues to increase and irri-

It should also be mentioned that urbanization is creating large pollution loads on

To maximize the potential of existing and emerging technologies, irrigators must

USDA-NASS (U.S. Department of Agriculture National Agricultural Statistics Service).

2.1.2 Water, Plant Growth, Irrigation, and Crop Production

resources on adapting to climate change. A committee of the Board of Water Science

Irrigated Land, hectares, millions

Irrigated Land, hectares, millions

2.3 CROP PRODUCTION AND

2.3.2 Deficit Irrigation and WUE

Gleick, P. 2002. The World’s Water, 2002-2003: Biennial Report on Freshwater

Experience Value Monitor

Cause and Effect

Figure 3.1. A conceptual model of the decision process.

An additional feedback loop is the learning process. By monitoring the results of a

Resources Problems Goals & Values

3.2 PLANNING FOR IRRIGATION

Table 3.1. Factors for consideration in irrigation decision making.

needed, should be included in the economics and evaluation of irrigation options.

Table 3.2. Expected economic lives for irrigation equipment (years).

Table 3.4. Sprinkler irrigation system costs, U.S. irrigation practice.

Table 3.6. Microirrigation system costs, U.S. irrigation practice.

Table 3.7. Annual operating costs for microirrigation systems.

Table 4.1. Possible negative environmental impacts of irrigation.

Table 4.1 continued.

Table 4.2. Possible positive impacts of irrigation.

4.2 WATER STORAGE, DIVERSION, AND CONSUMPTION

Figure 4.1. Morrow Point Dam on

Another potential impact of diversions is increase in stream temperature below the

Over-application of irrigation water, non-uniform irrigation water distribution, non-

Pesticides include herbicides, insecticides, fungicides, soil fumigants, and nemati-

methods estimate leaching by applying the principle of conservation of mass to a con-

4.3.4.1 Irrigation system selection, design, and management. Irrigation system

Table 4.3. Technologies for remediation of groundwater. Technologies potentially

A variety of other technologies have been studied for groundwater remediation.

Figure 4.3. Sediment deposition in the

4.4.2 Runoff and Erosion Processes

Figure 4.4. Eroded irrigation furrows.

Figure 4.5. Collecting a furrow water

gravimetrically or with Imhoff (sedimentation) cones (Figure 4.5). Volumetric sam-

Corwin, D. L., B. L. Waggoner, and J. D. Rhoades. 1991. A functional model of solute

Hamilton, P. A., and D. R. Helsel. 1995. Effects of agriculture on ground-water

and R. M. Cruse, eds. Madison, Wis.: Soil Sci. Soc. America.

NASS (National Agricultural Statistics Service). 2004. 2002 Census of Agriculture.

Final Report (Rainbow Report). Sacramento, Calif.: Calif. Dept. of Water

computer graphics. Appl. Agric. Res. 2(4): 277-289.

cepts of efficiency and uniformity, as applied to irrigation, will be discussed in greater

Consumptive Use Nonconsumptive Use