You are on page 1of 1

Systems Modeling to Improve the Hydro-Ecological

Performance of Diked Wetlands H31I-1234

Omar Alminagorta ,
1 David E. , Karin M. Rosenberg 2 Kettenring3

AGU Fall Meeting, San Francisco, December 2012


Abstract Model Formulation Model Application More Results
A systems model was developed to recommend water allocations and vegetation control The model identifies potential wetland management actions to maximize wildlife habitat
We apply the model at the Bear River Migratory Bird Refuge (The Refuge), Utah. The
actions among diked wetland units to improve wetland habitat for target bird species. Model while simultaneously satisfying hydrological, ecological, and management constraints. We
Refuge is the largest wetland complex on the Great Salt Lake and an important stopping and
recommendations are subject to constraints such as water availability, spatial connectivity of formulated the model in six phases (Fig. 4).
resting area for migratory birds on the North American Pacific and Central Flyway (Fig. 9).
wetland units, hydraulic infrastructure capacities, vegetation growth, responses to Phase Model Components The Refuge covers 118.4 km2 and includes 25 wetland units maintained through a series of
management activities, plus financial and time resources available to manage water and canals, gates, and dikes.
1.Wetland management purposes Maximize wildlife habitat
invasive vegetation. We apply the model at the Bear River Migratory Bird Refuge, Utah. Refuge managers participated throughout this study from identifying the problem,
Habitat suitability index (H)
Results show that there are opportunities to increase by almost two-fold the hydro-ecological 2.Performance metrics collecting data, through interpreting results.
Weighted usable area for wetlands (WU)
performance of wetland habitat by water allocation among wetland units and control of Storage (S)
invasive vegetation such as Phragmites australis (common reed). Water depth (WD) Flood area (A)
3.Decision variables
Flow rate (Q)

Background Invasive vegetation coverage (IV) Vegetation removal (RV)

4.Relationship between decision


Habitat suitability related to water depth (HW)
Water shortages, wetland drainage, agricultural variables and performance
Habitat suitability related to invasive vegetation (HV)
and sub/urban land use, and invasive vegetation metrics Figure 11. Spatial and temporal distribution of combined habitat suitability index for
have degraded wetland ecosystems (Fig. 1). Water availability (in) optimized condition in 2008. Dark shading denotes areas less suitable for priority species.
Thus, there is a need to assess management Budget to control invasive vegetation (b)
options to improve wetland habitat. 5.Constraints
Gate operation (ag) 700
Refuge water Wet(1997)
Management of wetlands includes the Species requirements (sw) right
Q2006 Q2011

Weight usable area wetland (km2)


manipulation of water levels to create a diversity Identify values for water depth and invasive vegetation cover that 650

of wetland habitats and improve habitat for water 6.Formulate Optimization Model Q1996
Q2005
maximize the weighed usable area while satisfying all constraints
birds (Smith, Euliss et al. 2008). Fig. 9. Location of the Bear River Migratory Bird Refuge and its conveyance network 600 Q2010
Q2009
Fig. 4. Model phases Excellent Black Necked Stilt
Wetland management also involves control of 1.0 American Avocet
Q2007

Results
BaseC(2008)

Habitat Suitability Index


Fig. 1. Scarcity of water in wetlands Wetland habitat performance
invasive vegetation such as Phragmites (Fig. 2). 0.9 Tundra Swan 550 Q2004
Excessive spread of invasive vegetation can It is quantified using 2 performance metrics: 0.8
1. Habitat suitability index: 0 to 1, from poor 0.7
reduce plant species diversity and limit nesting 0.6
to excellent habitat quality (Fig. 5). 0.5 We ran the model for a base case (2008 hydrologic conditions) and seven scenarios. 500
habitat and food availability for birds ( Zedler
2. The weighted usable area for wetlands 0.4 Comparison between the base case scenario of optimized management and past management Dry(1992)
and Kercher 2004). 0.3 450
represents the available surface area that activities show that there are opportunities to increase by almost two-fold the suitable
Water allocation and vegetation management 0.2
provides suitable hydrological and 0.1 wetland habitat area.
actions require resources that in many cases are
ecological conditions for priority bird Poor 0.0 Results of seven scenarios suggest that the performance of wetland habitat is more affected 400
0 50000 100000 150000 200000 250000 300000 350000
limited. In these instances, wetland managers 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00
species. by variables related with vegetation responses, limited gate operation, and water availability Water availability (Hm3/year)
need tools to help them decide where to allocate Water Depth (m)
Objective Function rather than financial budget (Table 1).
water and control vegetation to improve the Fig. 5. Habitat suitability for 3 priority bird species Figure 12. Relationship between water availability and Weight Usable Area for Wetlands
The model recommends water allocations and management of invasive vegetation to
ecological performance of a wetland ecosystem. Fig. 2. Invasive vegetation in wetlands Maximizes the sum of the weighted usable area (WU) considering two main terms: (i) the indicator. Data points represent model scenarios including dry, base case, wet
improve hydro-ecological performance (Fig. 10).
combination of individual habitat suitability indices (HW, HV), and weighting parameter (sw) conditions, and annual discharges from 2004 to 2011 (Q2004 to Q2011).
Research Objective and (ii) the flooded area (A), which is a function of the variable storage (S). Table 1. Model results for changes in water availability, budget, vegetation response, The model shows that water availability below the Refuges existing water right critically drops
,, , ,
and gate management scenarios. wetland performance. Thus, Refuge managers should be concerned about new upstream water
. = , , ,
,
We developed and applied a system model that recommends water allocations among wetland Inputs Result
abstractions that reduce available water.
Subject to : Water Budget WU
units and vegetation management actions to improve the ecological performance of a wetland
Conclusions
Scenario
Hydrological Constraints Availability ($1000/year) (km2/year)
ecosystem. Mass balance equation at each network node i (year)
Base case 2008 180 571
The optimization model identifies water depths and reduction of invasive vegetation
Systems Model
, + , , , ,, Simulation 2008 180 295
1 Dry condition 1992 180 468 coverage with the objective to maximize the area with suitable hydrological and ecological
, , = , 1, , , 2 Wet condition 1997 180 690
3 Increase budget by 50% 2008 270 583 conditions for priority bird species while simultaneously satisfying constraints related with
The model was developed for wetland units (w) where managers can control water Ecological Constraints Fig. 6. Gate operation and invasive vegetation control 4 Decrease budget by 50% 2008 90 563 water availability, spatial connectivity, hydraulic capacities, vegetation responses, and
allocation and invasive vegetation in each unit (Fig. 3). Invasive vegetation reduction 5 Increase vegetation response 15% per year 2008 180 470 financial resources.
Managers control water levels through canals, gates, weirs, and other hydrologic , = 1, , + , , Inputs Modeling Tools Outputs 6 Increase vegetation response 50% per year 2008 180 317
Refuge managers should continue controlling invasive vegetation, protecting the Refuges
Available surface 7 Limiting gate management operation 2008 180 403
infrastructure. Water allocations are affected by water availability, conveyance network, Management Constraints Hydrological Hydro Platform water right and allocating water according to model recommendations to reach desired
canal capacities, evaporation rates, and operation of gates. Water availability Manages inputs area that provides
Limited budget to control invasive vegetation
Network conveyance and displays suitable conditions for
wetland management goals.
Managers control invasive vegetation cover using herbicide and burning. The effectiveness , ,
1 1A 1B 2A 2B
Evaporation loss network priority bird species 3 3 3 3 3

of invasive vegetation control is influenced by the natural growth of invasive vegetation


Authors
2 2 2 2 2
Limited gate management operation Storage, area, and GAMS Recommended : 1 1 1 1 1
and the available financial budget to reduce invasive vegetation. 1 water depth Optimization o Water allocations 0
1 2 3 4 5 6 7 8 9101112
0
1 2 3 4 5 6 7 8 9101112
0
1 2 3 4 5 6 7 8 9101112
0
1 2 3 4 5 6 7 8 9101112
0
1 2 3 4 5 6 7 8 9101112

Controlling water levels and vegetation allows managers to create habitat that supports a tan relationships for software to wetland units 2C 2D 3A 3B 3C
() = + 1 +
3 3 3 3 3
1 Ph.D. Candidate, Department of Civil and Environmental Engineering, Utah State
diversity of wetland bird species (s) and plant community types that mimic a well- 2 2 wetland unit MATLAB o Reduction of 2 2 2 2 2

Water Depth (m)


1 1 1 1 1
Channel capacity Output invasive vegetation 0 0 0 0 0 University, Logan, UT, o.alminagorta@aggiemail.usu.edu
functioning freshwater ecosystem with multiple birding, hunting, and other wetland (+)
= () ; ()
= () ;
1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112
2 Assistant Professor, Department of Civil and Environmental Engineering and Utah Water
(+) ()
Ecological visualization o Allocation of 3D 3E 3F 3G 3H
services. , = + , , ; , ,
Initial vegetation financial budget to
3 3 3 3 3

Gc 1
2
1
2
1
2
1
2
1
2
1
Research Laboratory, Utah State University, Logan, UT, david.rosenberg@usu.edu
k coverage reduce invasive 0 0 0 0 0 3 Assistant Professor, Ecology Center and Department of Watershed Sciences, Utah State
Species habitat
1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112
0.8 vegetation 3I 3J 3K 4A 4B
requirements Simulated water 3 3 3 3 3 University, Logan, UT, karin.kettenring@usu.edu
2 2 2 2 2
0.6 Species weights allocations 1 1 1 1 1

References
0 0 0 0 0
Management Quantify how 1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112 1 2 3 4 5 6 7 8 9101112
0.4 4C 5A 5B 5C 5D
Unit cost of removing changes in water 3 3 3 3 3
2 2 2 2 2
0.2 invasive vegetation availability, 1 1 1 1 1
Smith, L., N. Euliss, et al. (2008). "Application of a geomorphic and temporal perspective to
G+
G- Total budget to vegetation response, 0
1 2 3 4 5 6 7 8 9101112
0
1 2 3 4 5 6 7 8 9101112
0
1 2 3 4 5 6 7 8 9101112
0
1 2 3 4 5 6 7 8 9101112
0
1 2 3 4 5 6 7 8 9101112
Gd 0 manage vegetation financial budgets, gate Month wetland management in North America." Wetlands 28(3): 563-577.
0 xo X
Simulated Optimized
Fig. 7. Gate operation function Gate operation operation affect Zedler, J. B. and S. Kercher (2004). "Causes and consequences of invasive plants in
available wetland performance Fig. 10. Comparison of model recommended (optimized, red line) and historical wetlands: opportunities, opportunists, and outcomes." Critical Reviews in Plant Sciences
Simulation Capabilities (simulated, blue bars) water allocations during 2008 in each wetland unit.
Fig. 3. Major components of the systems model for diked wetlands , = , , , Fig. 8. Model inputs and outputs 23(5): 431-452.