Agriculture, Ecosystems and Environment 105 (2005) 401411

www.elsevier.com/locate/agee

Short communication

Reducing phosphorus in dairy diets improves farm nutrient

balances and decreases the risk of nonpoint pollution
of surface and ground waters
Gurpal S. Toora,*, J. Thomas Simsa, Zhengxia Doub
a

Department of Plant and Soil Sciences, University of Delaware, Newark, DE 19716, USA
Centre for Animal Health and Productivity, School of Veterinary Medicine, University of Pennsylvania,
382 West Street Road, Kennett Square, PA 19348, USA
Received 29 December 2003; received in revised form 8 June 2004; accepted 28 June 2004

Abstract
Reducing phosphorus (P) in dairy diets may result in different types of manure with different chemical composition.
Application of these manures to soils may affect the soil P solubility and lead to different environmental consequences. A
laboratory incubation study determined the impact of 40 dairy manures on P dynamics in two soil types, Mattapex silt loam
(Aquic Hapludult) and Kalmia sandy loam (Typic Hapludult). The manures were fecal samples of lactating cows, collected from
commercial dairy farms located in Northeastern and Mid-Atlantic United States, with a wide range of dietary P concentrations
(from 2.9 to 5.8 g P kg
1 feed dry matter, DM). Dried and ground fecal samples were mixed with surface horizon (015 cm) of
soils at 150 kg P ha
1 and the mixtures were incubated at 25 8C for 21 days. At the end of incubation, water soluble P (WS-P) and
Mehlich-3 P (M3-P) in the soilmanure mixtures were substantially higher than the control (soil alone) but were lower than the
soils receiving fertilizer KH2PO4 at 150 kg P ha
1. Similarly, the relative extractability of P in soils amended with low- and highP manures was always lower (<93%) than KH2PO4 suggesting that fertilizer P is more effective at increasing soil solution P in
the short-term. Concentrations of WS-P or M3-P in soilmanure mixtures did not differ regardless of the source of manure (i.e.
different farms and different diets). This suggests that when the same amount of P is added to soils through manure applications,
the solubility or bioavailability of P in soils will be the same. However, P concentrations in feces correlate significantly with that
in diets (r = 0.82**); and when the manures were grouped into high-P diets (averaging 5.1 g P kg
1) versus low-P diets (3.6 g P
kg
1), manure P was 40% greater in the high-P group (10.6 g kg
1 DM) than the low-P group (7.6 g kg
1 DM). Thus, lowering
excess P in diets would reduce P excretion in manures, P accumulation in soils, improve P balance on farms, require less area for
land disposal, and decrease potential for P loss to waters.
# 2004 Elsevier B.V. All rights reserved.
Keywords: Dairy diets; Manures; Soil incubation; Farm balances

* Corresponding author. Tel.: +1 302 831 4190; fax: +1 302 831 0605.
E-mail address: gurpal@udel.edu (G.S. Toor).
0167-8809/$ see front matter # 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.agee.2004.06.003

402

G.S. Toor et al. / Agriculture, Ecosystems and Environment 105 (2005) 401411

1. Introduction
Eutrophication has been linked with many aspects
of water quality degradation, including fish kills, loss
of biodiversity and recreational uses of waters, and the
onset of harmful algal blooms that can pose a threat to
human health (Burkholder et al., 1999). Phosphorus
(P) losses from agricultural production systems are
known to contribute to accelerated eutrophication of
natural waters (Sims et al., 1998; Toor et al., 2003).
This is especially true in areas with intensive animal
farming, where repeated manure applications have led
to excessive accumulation of P in soils. Substantial
evidence exists to show that higher P concentrations in
soils can result in increased P losses to natural waters
(Sharpley and Tunney, 2000; Sims et al., 1998). Given
this, there is an urgent need to develop best management practices (BMPs) that prevent P accumulation in
soils to values that are of concern for water quality.
One of the most important BMPs for animal
agriculture is the need to strike a balance between P
inputs and outputs in farms and/or watersheds. Many
modern animal operations have surpluses of P on the
farm because P inputs (feed, fertilizer) exceed P
outputs in animal and crop products. Research has
clearly shown that the principle P input on most US
and European dairy farms is animal feedstuffs. For
example, annual P surpluses for Brittany (92 kg P
ha
1), The Netherlands (39 kg P ha
1) and Belgium
(38 kg P ha
1) are due to import of animal feeds on
farms (Sibbesen and Runge-Metzger, 1995). Farm P
surpluses often cause regular over-application of P to
soils, increasing the potential for P loss to surface and
shallow ground waters. In the UK, Withers et al. (2001)
reported that an average P surplus of ca. 1000 kg ha
1
has accumulated over the last 65 years in grasslands
and arable areas because of continued applications of P
inputs to soils that contain adequate amounts of P for
optimum crop yields. Sharpley and Smith (1995)
reported 35 years of manure applications (37270 kg P
ha
1 year
1) to soils in Oklahoma and Texas increased
soil test P (Olsen P) from 15 to 187 mg kg
1.
Fortunately, research has begun to develop practical
means to reduce P accumulations in soils by reducing P
excretion in manures. For dairy farms, feeding P closer
to the animals nutritional requirement has been shown
to reduce manure P concentrations and P surpluses on
farms, and have no adverse effects on animal health or

performance (Karn, 2001). The National Research

Council (National Research Council, 2001) recommends 3.23.8 g P kg
1 in the diets of lactating dairy
cows. However, on many US dairy farms, P is often
formulated to contain 4.5 to 5.0 g P kg
1 (2025%
excess P), with levels sometimes as high as 8.0 g P
kg
1. The reasons for overfeeding P are mainly related
to concerns that the amount of available P in the
feedstuff may be inadequate for satisfactory animal
health and reproduction (Dou et al., 2003; Wu et al.,
2000). A recent survey in the Mid-Atlantic USA by
Dou et al. (2003) concluded that P fed to lactating cows
averaged 34% above NRC recommendations. The
excess P in diets is excreted in manures that must be
applied to cropland because of the lack of alternative
off-farm uses. This results in the accumulation of P in
soils to values that can increase the potential for P
losses. Ebeling et al. (2002) studied the influence of
two dairy diets (3.1 and 4.9 g P kg
1) on P losses in
runoff from a manure-amended silt loam soil in
Wisconsin. Dissolved reactive P (DRP) concentrations
in a June runoff event from the high-P diet manure were
four times higher (1.18 mg l
1) than the low-P diet
manure (0.30 mg l
1); similar trends were seen in
subsequent runoff events. In addition to increasing the
potential for P loss to water, excessive additions of P to
dairy diets annually costs $1015 per cow or over $100
million in the US alone (Council for Agricultural
Science and Technology, 2002). Thus, reducing the P
content in dairy diets will have both environmental and
economic benefits and, if done properly, should have
no deleterious effects on animal performance or crop
production.
Many approaches can be taken to reduce P in dairy
diets, such as reducing the amount of mineral P
supplemented or using forages, grains, and byproducts with lower P concentrations. Changing the
sources of P fed to dairy cows may also alter the forms
of P in manures and thus affect the potential for P loss
from manured soils by runoff, erosion, or leaching. In
this study, we evaluated the effects of manures
produced on a wide range of Mid-Atlantic and
Northeastern dairy farms, where low- or high-P diets
were fed, on the forms and solubility of soil P. Our
hypothesis was that reducing dietary P to reduce P
excretion would not affect the solubility or plant
availability of P in manure-amended soils, when
manures are applied at the same P rate.

G.S. Toor et al. / Agriculture, Ecosystems and Environment 105 (2005) 401411

403

thus would be expected to have a greater capacity to

sorb soluble P added in manures or fertilizers.

2. Materials and methods

2.1. Soil collection

2.2. Soil preparation and analyses

Two soils with different physical and chemical
properties were used in this study; Mattapex silt loam
(USDA: fine loamy, mixed, mesic Aquic Hapludult;
FAO: Acrisols) and Kalmia sandy loam (USDA: fine
loamy, siliceous, thermic Typic Hapludult; FAO:
Acrisols) (Table 1). The soils are typical of two major
physiographic regions in the Mid-Atlantic where dairy
operations are located. The Mattapex soil is a finetextured Piedmont soil while the Kalmia soil is
representative of the sandy, low-organic matter soils
of the Coastal Plain. Both soils were moderately acidic
and had rather low-organic matter (OM) contents and
cation exchange capacities (CEC). Soil test P (Mehlich
3) values for the Mattapex and Kalmia soils were in the
medium (25 to 50 mg P kg
1) and optimum (50
to 100 mg P kg
1) categories of P sufficiency for crops
(Sims et al., 2002). The Mattapex soil had much higher
contents of clay, Al and Fe than the Kalmia soil and
Table 1
Selected properties of two soils used in the study
Property
pH
Organic matter (g kg
1)
Sand (g kg
1)
Silt (g kg
1)
Clay (g kg
1)
CEC (pH 7) (meq 100 g
1)
Water soluble (mg kg
1)
P
Ca
Mehlich-3 (mg kg
1)
P
Ca
Total (mg kg
1)
P
Ca
Mg
Fe
Al
M3-PSRa
DPSb
a
b

Mattapex silt
loam soil

M3
PSR
Kalmia sandy
loam soil

5.3
18
180
610
210
8.4

6.3
15
700
190
110
4.2

1.8
33.7

3.8
10.5

34
529

80
321

415
789
1721
15890
13980

284
470
388
4406
8523

0.05
20

Soil samples were collected from topsoil horizons

(015 cm), air-dried, and ground to pass through a 2mm sieve prior to chemical analysis. Field capacity
was measured by the method of Tan (1996). Soil pH,
OM, CEC, and sand, silt and clay were determined by
standard methods of the University of Delaware Soil
Testing Program (Sims and Heckendorn, 1991).
Mehlich-3 P (M3-P), Ca (M3-Ca), Fe (M3-Fe) and
Al (M3-Al) were extracted with 0.2 M CH3COOH +
0.25 M NH4NO3 + 0.015 M NH4F + 0.13 M HNO3 +
0.001 M EDTA using a soil to solution ratio of 1:10
(Mehlich, 1984) and analyzed by inductively coupled
plasma atomic emission spectroscopy (ICP-AES). The
Mehlich-3 P saturation ratio (M3-PSR), which is a
quantitative measure of the amount of P sorbed by a
soil relative to its total P sorption capacity, was
calculated by the following equation, with values for
M3-P, M3-Al and M3-Fe (Sims et al., 2002):

0.11
36

Mehlich 3-P saturation ratio, P/[Fe + Al].

DPS: degree of P saturation, Pox/0.5[Alox + Feox].

M3-P
M3-Fe M3-Al

(1)

Another measure of P sorption capacity of a soil,

referred to as a degree of phosphorus saturation (DPS)
was computed using oxalate extractable P, Fe and Al
(Pox, Feox and Alox), that were measured by shaking 1 g
of soil with 40 ml of 0.2 M acid ammonium oxalate
(pH 3) for 2 h in the dark (McKeague and Day, 1966)
and determining P, Al and Fe in solution by ICP-AES,
as follows:
DPS

Pox
100
aAlox Feox

(2)

Values reported for a have a typical range from

0.34 to 0.61; based on previous studies, an a-value of
0.5 was used (Breeuwsma and Silva, 1992; Pautler and
Sims, 2000; Sims et al., 2002).
Water soluble (WS)- P and -Ca in soil samples were
determined by extraction (soil to deionized water ratio
of 1:10) for 1 h followed by centrifugation for 15 min
before filtration through a 0.45 mm filter paper (Olsen
and Sommers, 1982). Concentrations of P in solution
were determined by ICP-AES. Total P, Ca, Mg, Fe and
Al concentrations in the soils were determined by

404

G.S. Toor et al. / Agriculture, Ecosystems and Environment 105 (2005) 401411

microwave digestion using concentrated HNO3

followed by ICP-AES analysis (USEPA, 1986).
2.3. Manure preparation and analyses
Dairy manures (rectal grab feces, lactating cows
only) were collected from 40 commercial dairy farms
in the Northeast and Mid-Atlantic regions (Pennsylvania: 16, New York: 13, Maryland: 9, Delaware: 1
and Virginia: 1). Manure collection was part of a larger
study undertaken at the University of Pennsylvania,
where the manure samples were collected from individual cows to investigate the effect of dietary P
modification on P excretion in dairy cows (Dou et al.,
2003). For this study, we composited the manure
samples collected from individual cows on each farm.
Manures from 19 dairy farms having 2.93.9 g dietary
P kg
1 dry matter (DM) were grouped into the low-P
diet category, as proposed by the National Research
Council (National Research Council, 2001), whereas
21 farms having 4.6 to 5.8 g kg
1 dietary P were
grouped into the high P diet category. All diets consisted of forages (4450%), grains (3135%), and
concentrate mix (1426%) as major feed ingredients,
together with small amounts of salt, trace minerals and
vitamins.
Manures were dried at 55 8C and ground to pass a
1-mm sieve prior to analysis. Total C, N and S were
determined using CNS analyzer (LECO CNS-2000,
LECO Corporation, St. Joseph, MI). Manures were
analyzed for WS-P, and total P, Ca, Mg, Fe and Al, by
the methods described above for soils.
2.4. Incubation methodology
Composited manures from each dairy farm were
incorporated with 100 g of air-dried soil, in triplicate,
at 150 kg P ha
1 (67 mg P kg
1) in polyethylene cups.
The amount of P applied is typical when manures are
applied on an N basis (Sharpley et al., 2003). The
polyethylene containers were cut to provide two holes
at the top to allow free gaseous exchange and prevent
build up of anaerobic conditions during the incubation.
An inorganic-P treatment, used to simulate fertilizer P,
received P at 150 kg P ha
1 in a liquid form from
KH2PO4 and an unamended control received no
inorganic P or manure. These soils were incubated
at 70% of field capacity for 21 days at room

temperature (25 8C). We chose an incubation of 21

days because past research using laboratory incubation
studies to compare P release in soils amended with
organic-P sources has shown that >90% of the change
in extractable P usually occurs in this time period
(Griffin et al., 2003; Sharpley, 1982; Stout et al., 1998).
The moisture content was maintained by weekly
weighing the containers and adding deionized water as
needed. Subsamples were removed at days 1 and 21, air
dried and analyzed as mentioned in the soil preparation
and analyses section. The relative extractability of P
(RPE) in soils amended with manures in the two diet
groups (low-P versus high-P) compared to inorganic P
was calculated at days 1 and 21 by WS-P and M3-P
according to Leytem et al. (2003):
Extractability of P %
Soil Pmanure or inorganic P mg=kg

Soil Pcontrol mg=kg
100

Total P added 67 mg kg
1

RPE %

P extractabilitymanure
P extractabilityinorganic P

100

(3)
(4)

An RPE value of 100% indicates that manure P is

equally extractable as inorganic P, while lower or
higher RPE values indicate lower or greater extractability, respectively.
2.5. Statistical analyses
Descriptive statistics and one-way ANOVA were
carried out using Genstat 4.2, fifth Edition (Lawes
Agricultural Trust, Rothamsted, UK) to calculate
means and standard errors, and to test for significant
differences (P < 0.05) between means.

3. Results and discussion

3.1. Properties of dairy manures produced from
low- or high-phosphorus diets
The total-P concentration for diets from the 40
sampled dairy farms in the Mid-Atlantic region ranged
from 2.9 to 5.8 g kg
1; 18 farms had dietary-P values
in the NRC recommended range (3.23.9 g kg
1).
One farm had 2.9 g kg
1 dietary P, while the other

G.S. Toor et al. / Agriculture, Ecosystems and Environment 105 (2005) 401411

farms had dietary P from 4.6 to 5.8 g kg
1. This

variation in P content in diets resulted in manures with
different amounts of P (5.514.0 g kg
1). A positive
correlation (r = 0.82**) observed between P in diets
and P excreted in manures (Fig. 1) supports past
research showing that P fed in excess of animals
requirement is excreted in manure. Wu et al. (2001)
reported that 3.1, 3.9 and 4.7 g kg
1 dietary P for highproducing cows resulted in P excretion of 43, 66 and
88 g fecal P per day. They concluded that all of the P
fed in excess of 3.1 g kg
1 was excreted in the feces.
Similarly, Dou et al. (2002) reported that increasing
dietary P from 3.4 to 6.7 g kg
1 resulted in increasing
P excretion from 5.2 to 12.7 g kg
1. In our study,
manures produced from high P diets had 40% higher P
(mean of 21 manures: 10.6 g kg
1) as compared with
manures produced from low-P diets (mean of 19
manures: 7.6 g kg
1) (Table 2).
Concentrations of WS-P in dried and ground
manures were <1212 mg kg
1 (<17% of P), with
mean concentrations of 478 mg kg
1 for the low-P
diet group and 487 mg kg
1 for the high-P diet
group. There was no indication, as supported by
analyses of the manures themselves (Table 2), that
the diets used on the low- and high-P farms were
resulting in significant differences in the amount of
Ca, Mg, Fe and Al available to react with P in the
manures. In general, it seems likely that the diets
used on the low-P farms resulted in essentially the

405

same type of manures, except for significantly lower

total-P contents, as those from high-P farms. This, in
turn, may result in similar concentrations of P, and
may not affect the solubility of manure P on addition
to soils.
3.2. Effect of phosphorus sources on soil
phosphorus concentrations during 21-day soil
incubation
Modifying dairy diets to reduce total excreted P in
manures had no significant effect on P in manured
soils, when the manures were applied at the same
total-P rate of 150 kg P ha
1 (Table 3). While manure
application did increase all forms of P measured (WSP, M3-P) relative to the unamended soils, no
significant differences were observed in any form of
soil P measured between manures in the low-P (2.9
3.9 g kg
1) and high-P (4.65.8 g kg
1) diet categories. Similar trends in soil P were observed
immediately after incorporation (1 day) and after 21
days of reaction of manures and fertilizer P with the
two soils. In general, only small changes, primarily
increases, were observed in any form of soil P between
the 1 and 21 days sampling periods.
Mean values for WS-P, soil test P (M3-P) and soil P
saturation (M3-PSR) at day 21 (averaged over both
soils) were 8.7 mg kg
1, 90 mg kg
1 and 0.116 for the
19 manures produced using low-P diets, and

Fig. 1. Relationship between total P in diets and manures.

406

G.S. Toor et al. / Agriculture, Ecosystems and Environment 105 (2005) 401411

Table 2
Concentrations of total and water soluble elements in manures
Percentage Total (g kg
1 dry matter)
of P in diet C
N
S
P
Ca

Diet category
a

Mg

Fe

Low P

Mean
Minimum
Maximum
Median
S.D.

0.36
0.29
0.39
0.37
0.04

454
381
493
462
28

31.2
23
52
31
5

2.0
1.6
2.5
1.9
0.4

High Pb

Mean
Minimum
Maximum
Median
S.D.

0.51
0.46
0.58
0.50
0.05

456
427
486
457
16

30.5
26
44
30
4

1.8 10.6 28.6 10.1 1.5

1.3 7.8 18.5 7.4 0.7
2.6 14.0 37.1 14.6 3.1
1.7 10.6 28.8 10.0 1.4
0.5 1.6 4.7 1.8 0.60

a
b
c
d
e

7.6 27.8 9.4 1.4

5.5 4.8 4.6 0.8
9.8 42.8 12.1 2.4
7.4 27.9 9.7 1.3
1.1 8.0 1.7 0.4

Al

Water soluble Percentage of

WS-P addede

1
P (mg kg ) total as WS-P (%) (kg ha
1)

0.8 478c
0.3 362
1.8 1212
0.8 488
0.4 204

6.3
4.7
17.0
6.2
3.1

9.5
7.1
25.5
9.3
4.6

1.0 487d
0.3 319
2.3 770
0.9 431
0.5 128

4.6
2.6
9.1
4.3
1.4

6.9
3.8
13.6
6.4
2.3

Means of 19 dairy manures.

Means of 21 dairy manures.
By excluding two samples that had unusually higher water soluble P (818 and 1212 mg kg
1).
By excluding two samples that had unusually lower water soluble P (319 and 330 mg kg
1).
Amount of WS-P added to each soil with 150 kg manure P ha
1.

8.2 mg kg
1, 89 mg kg
1, and 0.118 for the 20

manures produced using high-P diets. Dairy manures
also increased all forms of soil P less than fertilizer P,
where these same parameters averaged 12.7 mg kg
1,
118 mg kg
1, and 0.149 at day 21 (Table 3). The

higher WS-P concentrations in fertilizer amended

soils are due to the forms of P present in these sources:
mineral P fertilizer (KH2PO4) used in the present
study contained all of the P in the water soluble
fraction which is expected to raise the WS-P of soil to

Table 3
Mean concentrations of water soluble P, Mehlich 3-P and Mehlich 3-P saturation ratio after 1 and 21 days incubation with different phosphorus
sources for two soils
Incubation period (days)
Mattapex silt loam
1

21

Kalmia sandy loam

1

21

a
b
c

Treatments

Water soluble P (mg kg
1)

Mehlich 3-P (mg kg
1)

Mehlich 3-PSR

Control
Fertilizer control
Low-P dietb
High-P dietc

1.1
7.8
5.0
5.7

(0.09)a
(0.50)
(0.13)
(0.21)

32
70
63
68

(1.4)
(3.9)
(0.8)
(1.0)

0.043
0.092
0.083
0.083

(0.0027)
(0.0044)
(0.0011)
(0.0025)

Control
Fertilizer control
Low-P diet
High-P diet

2.5
8.3
6.6
6.6

(0.06)
(0.24)
(0.12)
(0.21)

33
78
59
60

(0.6)
(2.7)
(0.7)
(0.8)

0.042
0.099
0.076
0.080

(0.0009)
(0.0052)
(0.0008)
(0.0013)

Control
Fertilizer control
Low-P diet
High-P diet

2.8
16.1
8.6
9.4

(0.09)
(0.81)
(0.39)
(0.31)

82
133
122
126

(2.7)
(9.0)
(0.8)
(1.2)

0.109
0.183
0.163
0.159

(0.0004)
(0.0044)
(0.0017)
(0.0029)

Control
Fertilizer control
Low-P diet
High-P diet

4.1
17.1
10.7
9.8

(0.21)
(0.38)
(0.27)
(0.25)

92
158
120
117

(0.9)
(7.1)
(1.4)
(1.1)

0.119
0.198
0.155
0.155

(0.0005)
(0.0046)
(0.0019)
(0.0023)

Data in parentheses are standard errors of the mean.

Means of 19 dairy manure amended soils (mean P in diet: 3.6 g kg
1).
Means of 21 dairy manure amended soils (mean P in diet: 5.1 g kg
1).

G.S. Toor et al. / Agriculture, Ecosystems and Environment 105 (2005) 401411

a greater extent immediately following its application

compared with manures that had <17% of P as WS-P.
Total WS-P addition was 9.5 kg ha
1 (4.2 mg kg
1)
for low-P diet group manures and 6.9 kg ha
1
(3.1 mg kg
1) for high-P diet group manures (Table
2). The presence of lower amounts of WS-P in these
manures resulted in a very similar increase in WS-P
concentrations in these soils. Studies have shown that
fertilizer P is readily available to plants while manures
slowly release P depending upon the inorganic P
content, and C:P ratio (Bahl and Toor, 2002; Toor and
Bahl, 1999). Ebeling et al. (2003) reported that after
16 weeks of incubation available soil P as measured by
Mehlich-3, BrayKurtz P1, and Olsen-P was highest
in a mineral P fertilizer amended soil than in one
amended with dairy manures.
The M3-PSR ratio was higher in the fertilizer
treatment than control or manure amended soils.
However, M3-PSR was similar for the low- and high-P
manure-amended soils. Sims et al. (2002) observed
M3-PSR values ranging from <0.01 to 0.59 for 465
Mid-Atlantic moderately acidic soils (median pH =
5.7). They reported ranges of M3-P and M3-PSR as
follows: (i) below optimum: 50 mg kg
1 and <0.06,
(ii) optimum: 51100 mg kg
1 and 0.060.11, (iii)
above optimum: >100 mg kg
1 and >0.11, and (iv)
above environmental: >150 mg kg
1 and >0.15.
According to this criteria, a single application of
fertilizer or manure increased M3-P and M3-PSR from
below optimum to optimum in Mattapex soil. While in
the Kalmia soil, these values were already in optimum
category in control treatment and were increased into
the above optimum (manure-amended soil) and
environmental (fertilizer-amended soil) categories.
These changes in M3-P (3050 mg kg
1) and M3PSR (0.040.09) from a single application of fertilizer
or manure are significant in the context of increasing
build-up of soil P that may increase potential for
increased P losses.
Soil type also influenced observed increases in
all forms of soil P due to manure or fertilizer
application. Increases in WS-P, M3-P, and M3-PSR
at day 21 were consistently greater for the Kalmia
soil than the Mattapex soil. For example, mean
increases in WS-P, M3-P, and M3-PSR (averaged
over P source, relative to the unamended soil) were
8.4 mg kg
1, 40 mg kg
1, and 0.077 for the Kalmia
soil, which had a higher P saturation value and,

407

because of its sandier texture, less capacity to retain

additional P, as compared to 4.7 mg kg
1,
32 mg kg
1, and 0.043 for the fine-textured Mattapex soil. The practical implications of lower
increases in WS-P and M3-P in the Mattapex soil
than the Kalmia soil are that the Mattapex soil, due to
higher P fixing capacity, can withstand a higher P
application rate before having higher P in soil
solution and runoff. This reiterates for different P
management practices on soils that have diverse
chemical characteristics.
3.3. Relative soil phosphorus extractability from
fertilizer and manure
The RPE in soils amended with low- and high-P
manures was evaluated in comparison with fertilizeramended soil. Manure P was always less extractable
than fertilizer P for both soils after 1 and 21 days of
incubation. The average RPE determined with WS-P
ranged from 43 to 51% and from 59 to 69% in the
Kalmia and Mattapex soils, respectively, and changed
only slightly with time. On the other hand, RPE
determined using stronger extractant (M3-P) ranged
from 80 to 93% after 1 day of incubation for both soils,
and decreased to 3961% with time (after day 21),
reflecting the tendency for M3-P to continue to
increase in the fertilized soils (e.g., Mattapex:
8 mg kg
1, Kalmia: 25 mg kg
1) while decreasing
in the manured soils (e.g., Mattapex: 48 mg kg
1,
Kalmia: 29 mg kg
1). The higher values of RPE
(WS-P, M3-P) for Mattapex soil than the Kalmia soil
after both incubation periods are attributed to the
higher P sorption in the Mattapex soil that resulted in
lower increase in WS-P for the fertilizer-amended soil
over control (5.86.7 mg kg
1) as compared with a
higher increase in the Kalmia soil (1313.3 mg kg
1).
Manures from high-P diets tended to have higher RPE
values (for WS-P and M3-P) initially than those from
low-P diets; however, little difference was observed
between manure types by day 21. In a related study,
Leytem et al. (2003) compared RPE from nine organic
and one inorganic (KH2PO4) P sources in an Evesboro
loamy sand soil that was incubated for 8 weeks. The
RPE by WS-P and M3-P was higher for KH2PO4,
followed by liquid and deep pit dairy and swine
manures, and solid manures and biosolids treated with
metal salts.

408

G.S. Toor et al. / Agriculture, Ecosystems and Environment 105 (2005) 401411

3.4. Long-term implications of low- and highphosphorus-diet generated manures

Many states in the Northeastern and Mid-Atlantic
region are now requiring, or strongly suggesting, that
P-based manure management practices be implemented for agricultural operations using animal
manures as soil amendments (Sharpley et al., 2003).
Clearly, modifying dairy diets to reduce the amount of
manure P on the farm is an important component of Pbased manure management. There was about 40% less
P to manage each year on the 19 farms using low-P
diets than the 21 farms feeding high-P diets. This
equates to a difference in the amount of P annually
added to soils of 60 kg P ha
1, when applied on an
equivalent mass basis (i.e., 20 Mg dry manure ha
1; a
typical manure-application rate on many dairy farms),
and 39 kg P ha
1 when applied to provide the amount
of plant available N (PAN) recommended to attain a
realistic corn (Zea mays L.) yield for this region
(9.5 Mg ha
1).
It is important to note, however, that even the lesser
amount of P that would be applied with low-P manures
exceeds the amount removed in the harvested portion
of corn grown for grain or forage grass hay (20
25 kg P ha
1). This points to the need for alternative
uses of dairy manures beyond land application to
cropland, particularly for soils that are already in the
optimum, or higher, soil test P-range for crop
production. These alternatives may include solid
liquid separation by sedimentation or screening, and
then separated solids can be composted, reused as
bedding or potting material, or directly applied to offfarm soils as an amendment. Other alternative
treatments for solid materials (e.g., poultry litter
and sewage sludges) are drying, incineration and
pyrolysis.
Because of the current lack of alternatives to land
application of dairy manure, one interim approach to
P-based management now being used in some states
(e.g., Delaware and Maryland), for soils considered
high in P from an environmental perspective (M3-P
>150 mg kg
1), is to allow farmers to apply, in the
current year, the amount of manure P equal to that
which will be removed in the harvested portion of the
next three years crops. This has been advocated for
two reasons. First, it balances the amount of P added to
soils during a crop rotation with the P removed and

thus prevents further buildup of soil P. Second, it is not

feasible, given current equipment limitations, to apply
manure at rates equal to the P removed by only one
crop in the rotation; hence a single-manure application
every 3 years is permitted. While perhaps advisable
from a P management perspective, it is important to
recognize that this approach will result in greater
manure applications when farmers use low-P rather
than high-P diets. For example, manure application
based on a three-year P crop removal of 208 kg ha
1
by a typical corn (average yield of 9.5 Mg ha
1, year
1), wheat (Triticum spp.) and soybean (Glycine max L.
Merr.) (4.7 Mg ha
1, 2.4 Mg ha
1; year 2), and corn
(9.5 Mg ha
1; year 3) rotation would be 27.3 Mg ha
1
for low-P and 19.6 Mg ha
1 for high-P manures. The
amount of N added with low-P manures (853 kg ha
1)
would exceed that from the high-P manures
(598 kg ha
1). In both cases, applying dairy manure
to meet 3-year crop P-removal rates would also add
considerably more N than is needed by most annual
crops, increasing the potential for N losses to ground
and surface waters. This suggests that the most
balanced approach would be to reduce P in diets (and
manures) and apply the amount of N needed to achieve
economically optimum yields.
Another important consideration for dairy farmers
is the amount of manure that can be applied before a
soil reaches the value where P-based management will
be required. For instance, in Delaware and Maryland,
soils with a M3-P value greater than 150 mg P kg
1
are required to implement some form of P-based
management for manures, such as only applying P at
crop removal rates. Related to this, and of importance
to our study, is whether or not modifying dairy diets by
changing the nature of the feed ingredients would
affect how soil P increases when manure is applied.
Specifically, would soil P increase more or less
rapidly, per kilogram of manure P, when manures from
low- or high-P diets were applied. To answer this
question, we quantified the increase in soil WS-P and
M3-P (mg P kg
1) per mass of dairy manure P added
(kg manure P ha
1) for both soils. For example, for the
Mattapex soil, increase in soil M3-P with application
of 1 kg of manure P ha
1 was 0.172 mg kg
1 for lowP manure and 0.181 mg kg
1 for high-P manure.
Similarly, for the Kalmia soil, the soil M3-P was
0.193 mg kg
1 for low-P manure and 0.176 mg kg
1
for high-P manure. For both soils, the increase in soil

G.S. Toor et al. / Agriculture, Ecosystems and Environment 105 (2005) 401411

M3-P with fertilizer KH2PO4 was twice as high

(Mattapex: 0.300 mg kg
1 and Kalmia: 0.444 mg
kg
1) as with manures. The relatively similar increases in soil M3-P with low- and high-P manure
applications show that diet will not affect the increase
in P, that these increases will be less than those for
fertilizer P, and that the rate of increase will vary with

409

soil type. This supports our hypothesis that the nature

of diets used to reduce overfeeding P will not alter the
solubility or bioavailability of P in soils. Thus, the risk
of P loss to water should decrease if dairy farms stop
overfeeding P.
We used the above values to determine the mass of
dairy manure that can be applied to these soils before

Fig. 2. Implications of adding low and high P in diet-produced manure on increase in Mehlich 3-P for two soils.

410

G.S. Toor et al. / Agriculture, Ecosystems and Environment 105 (2005) 401411

they exceed the current regulatory limit now

established in Delaware and Maryland (150 mg M3P kg
1). It should be noted that these observations
were derived from an incubation study, which require
field scale verification with a diverse group of soils and
at variable rates of manure applications, however, in
the absence of that field data, our results can be used to
provide tentative information about manure P loading
on soils. For the Mattapex silt loam soil, an additional
12.8 Mg dry manure ac
1 can be applied from
manures in the low-P diet category than those in
the high P category before M3-P exceeds 150 mg kg
1
(Fig. 2a). For the sandier, high-P Kalmia soil, less
manure (3.9 Mg dry manure ac
1) can be applied
before the current M3-P limit is reached (Fig. 2b).
Thus, for a typical dairy farm in the Northeastern and
Mid-Atlantic region (200 ac), using a low-P diet
would allow for the annual application of an additional
780 (Kalmia) to 2560 (Mattapex) Mg dry manure per
farm before soils such as these reach the point where
P-based management would be required. The actual
amount of additional dairy manure that could be land
applied once a low-P diet is adopted will obviously
vary between farms as a function of soil type, soil test
P, and other parameters. However, the fact remains
that modifying dairy diets to reduce P excretion will
not alter the rate of increase in soil P at a given P
application rate due to manure additions.

manure management that will help to reduce P

excretion in manures, and associated P surplus in soils,
while reducing costs of animal production and
preserving non-renewable P mineral sources. However, other factors such as land area, and crops grown
in a given farm are the predominant factors influencing
the farm P balance, and should be considered while
assessing the impacts associated with reduction in
dietary P. Related to this, alternative off-farm uses of
dairy manures should be evaluated that are a key to
resolving the long-term P management problems on
dairy farms.

4. Conclusions

References

Increase in soil WS-P and soil test P (M3-P) did not

differ significantly due to application of manures
produced from low- and high-P diets. This suggests
that the nature of diets used to reduce overfeeding P
will not alter the solubility or bioavailability of P in
soils. Thus, the risk of P loss to water should decrease
if dairy farms stop overfeeding P. As many of the soils
in the intensive dairy production regions are already
saturated with P and due to the increased incidences of
eutrophication in the Chesapeake Bay Watershed,
regulatory authorities in the Northeastern and MidAtlantic region are strongly suggesting that P-based
manure management practices be implemented for
agricultural operations using animal manures as soil
amendments. Clearly, reducing P content in diets on
the farm is an important component of P-based

Bahl, G.S., Toor, G.S., 2002. Influence of poultry manure on

phosphorus availability and the standard phosphate requirement
of crop estimated from quantityintensity relationships in different soils. Bio. Res. Tech. 85, 317322.
Breeuwsma, A., Silva, S., 1992. Phosphorus fertilization and environmental effects in the Netherlands and the Po region (Italy).
Rep. 57. Agriculture Research Department, Wageningen, The
Netherlands.
Burkholder, J.M., Mallin, M.A., Glasgow, H.B., 1999. Fish kills,
bottom-water hypoxia, and the toxic Pfiesteria complex in the
Neuse River and Estuary. Mar. Ecol. Prog. Ser. 179, 301310.
Council for Agricultural Science and Technology, 2002. Animal diet
modification to decrease the potential for nitrogen and phosphorus pollution. In: Klopfenstein, T., Angel, C.R., Cromwell,
G.L., Erickson, G.E., Fox, D.G., Parsons, C., Satter, L.D.,
Sutton, A.L., (Eds.), Issue paper no. 21.
Dou, Z., Ferguson, J.D., Fiorini, J., Toth, J.D., Alexander, S.M.,
Chase, L.E., Ryan, C.M., Knowlton, K.F., Kohn, R.A., Peterson,
A.B., Sims, J.T., Wu, Z., 2003. Phosphorus feeding levels and

Acknowledgements
This work was funded by USDA Initiative for
Future Agriculture and Food Systems Grant no. 200152103-11334. We are thankful to our collaborators in
the animal science departments of the University of
Pennsylvania, University of Maryland, Pennsylvania
State University, and Cornell for providing manure
samples. Thanks are expressed to Sarah Alexander and
Jude Fiorini of University of Pennsylvania, and James
Hyde of University of Delaware for coordinating
manure sample collection, providing farm dietary
information, and help with the incubation study,
respectively.

G.S. Toor et al. / Agriculture, Ecosystems and Environment 105 (2005) 401411
critical control points on dairy farms. J. Dairy Sci. 86, 3787
3795.
Dou, Z.X., Knowlton, K.F., Kohn, R.A., Wu, Z.G., Satter, L.D.,
Zhang, G.Y., Toth, J.D., Ferguson, J.D., 2002. Phosphorus
characteristics of dairy feces affected by diets. J. Environ. Qual.
31, 20582065.
Ebeling, A.M., Bundy, L.G., Powell, J.M., Andraski, T.W., 2002.
Dairy diet phosphorus effects on phosphorus losses in runoff
from land-applied manure. Soil Sci. Soc. Am. J. 66, 284291.
Ebeling, A.M., Cooperband, L.R., Bundy, L.G., 2003. Phosphorus
source effects on soil test phosphorus and forms of phosphorus in
soil. Commun. Soil Sci. Plant Anal. 34, 18971917.
Griffin, T.S., Honeycutt, C.W., He, Z., 2003. Changes in soil
phosphorus from manure application. Soil Sci. Soc. Am. J.
67, 645653.
Karn, J.F., 2001. Phosphorus nutrition of grazing cattle: a review.
Anim. Feed Sci. Technol. 89, 133153.
Leytem, A.B., Sims, J.T., Coale, F.J., 2003. Determination of
phosphorus source coefficients for organic phosphorus sources:
laboratory studies. J. Environ. Qual. 33, 380388.
McKeague, J., Day, J.H., 1966. Dithionite and oxalate-extractable
Fe and Al as aids in differentiating various classes of soils. Can.
J. Soil Sci. 46, 1322.
Mehlich, A., 1984. Mehlich 3 soil test extractant: a modification of
Mehlich 2 extractant. Commun. Soil Sci. Plant Anal. 15, 1409
1416.
National Research Council, 2001. Nutrient Requirements of Dairy
Cattle, seventh ed. National Academy Press, Washington, DC.
Olsen, S.R., Sommers, L.E., 1982. second ed. In: Page, A.L.,
Miller, R.H., Keeney, D.R. (Eds.), Phosphorus. Methods of Soil
Analysis, Part 2. Agronomy No. 9. American Society of Agronomy, Madison, Wisconsin, USA, pp. 403430.
Pautler, M.C., Sims, J.T., 2000. Relationships between soil test
phosphorus, soluble phosphorus, and phosphorus saturation in
Delaware soils. Soil Sci. Soc. Am. J. 64, 765773.
Sharpley, A., Tunney, H., 2000. Phosphorus research strategies to
meet agricultural and environmental challenges of the 21st
century. J. Environ. Qual. 29, 176181.
Sharpley, A.N., 1982. Prediction of water-extractable phosphorus
content of soil following phosphorus addition. J. Environ. Qual.
11, 166170.
Sharpley, A.N., Smith, S.J., 1995. Nitrogen and phosphorus forms in
soils receiving manure. Soil Sci. 159, 253258.
Sharpley, A.N., Weld, J.L., Beegle, D.B., Kleinman, P.J.A., Gburek,
W.J., Moore, P.A., Mullins, G., 2003. Development of

411

phosphorus indices for nutrient management planning strategies

in the United States. J. Soil Water Conserv. 58, 137152.
Sibbesen, E., Runge-Metzger, A., 1995. Phosphorus balance in
European agriculture-status and policy options. In: Tiessen,
H. (Ed.), Phosphorus in Global Environment. John Wiley &
Sons, Chichester, UK, pp. 4358.
Sims, J. T., Heckendorn, S. E., 1991. Methods of Analysis of
the University of Delaware Soil Testing Laboratory, Cooperative
Bulletin No. 10. University of Delaware, Newark, DE
19716.
Sims, J.T., Maguire, R.O., Leytem, A.B., Gartley, K., Pautler, M.C.,
2002. Evaluation of Mehlich 3 as an agri-environmental soil
phosphorus test for the Mid-Atlantic United States. Soil Sci. Soc.
Am. J. 66, 20162032.
Sims, J.T., Simard, R.R., Joern, B.C., 1998. Phosphorus loss in
agricultural drainagehistorical perspective and current
research. J. Environ. Qual. 27, 277293.
Stout, W.L., Sharpley, A.N., Pionke, H.B., 1998. Reducing soil
phosphorus solubility with coal combustion products. J.
Environ. Qual. 27, 111118.
Tan, K.H., 1996. Measurement of field capacity water, Soil Sampling, preparation and analysis, Marcel Dekker Inc., New York,
pp. 6768.
Toor, G.S., Bahl, G.S., 1999. Kinetics of phosphate desorption from
different soils as influenced by application of poultry manure
and fertilizer phosphorus and its uptake by soybean. Bio. Res.
Tech. 69, 117121.
Toor, G.S., Condron, L.M., Di, H.J., Cameron, K.C., Cade-Menun,
B.J., 2003. Characterization of organic phosphorus in leachate
from a grassland soil. Soil Bio. Biochem. 35, 13171323.
USEPA, 1986. Method 3051. Acid Digestion of Sediments, Sludges
and Soils. Test Methods for Evaluating Solid Waste, vol. IA,
third ed. EPA/SW-846. National Technical Information Service,
Springfield, VA.
Withers, P.J.A., Edwards, A.C., Foy, R.H., 2001. Phosphorus cycling
in UK agriculture and implications for phosphorus loss from
soil. Soil Use Manage. 17, 139149.
Wu, Z., Satter, L.D., Blohowiak, A.J., Stauffacher, R.H., Wilson,
J.H., 2001. Milk production, estimated phosphorus excretion,
and bone characteristics of dairy cows fed different amounts
of phosphorus for two or three years. J. Dairy Sci. 84, 1738
1748.
Wu, Z., Satter, L.D., Sojo, R., 2000. Milk production, reproductive
performance, and fecal excretion of phosphorus by dairy cows
fed three amounts of phosphorus. J. Dairy Sci. 83, 10281041.