EPA 600/R 09/044 I June 2009 I www.epa.gov/ada
United States
Environmental Protection
Agency
Metrics for Nitrate Contamination
of Ground Water at
CAFO Land Application Sites -
Arkansas Dairy Study
Office of Research and Development
National Risk Management Research Laboratory, Ada, Oklahoma 74820
-------
-------
Metrics for Nitrate Contamination
of Ground Water at
CAFO Land Application Sites -
Arkansas Dairy Study
Philip A. Moore
University of Arkansas
Fayetteville, Arkansas
David Brauer
Conservation and Production Research Laboratory
Bushland, Texas
Office of Research and Development
National Risk Management Research Laboratory, Ada, Oklahoma 74820
-------
Notice
This work was supported through Interagency Agreement DW-12-921711-4 between EPA's
Ground Water & Ecosystems Restoration Division, National Risk Management Research
Laboratory (Elise Striz, Stephen Hutchins, Project Officers) and USDA-ARS's Conservation
and Production Research Laboratory (David Brauer, USDA-ARS Contact). Although this
work was funded substantially by the U.S. Environmental Protection Agency, it has not been
subjected to Agency review and therefore does not necessarily reflect the views of the Agency,
and no official endorsement should be inferred.
Contact Information:
Philip A. Moore, Jr.
USDA-ARS, Plant Sciences 115, University of Arkansas, Fayetteville, AR 72701.
Phone: 479-575-5724.
Fax: 479-575-7465.
E-mail: philipm!^uark.edu.
Mention of a trade name, proprietary product, or specific equipment does not constitute a
guarantee or warranty by the USDA and does not imply its approval to the exclusion of other
products that may be suitable.
-------
Foreword
The U.S. Environmental Protection Agency is charged by Congress with protecting the Nation's land, air, and water
resources. Under a mandate of national environmental laws, the Agency strives to formulate and implement actions
leading to a compatible balance between human activities and the ability of natural systems to support and nurture
life. To meet this mandate, EPA's research program is providing data and technical support for solving environmental
problems today and building a science knowledge base necessary to manage our ecological resources wisely,
understand how pollutants affect our health, and prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory is the Agency's center for investigation of technological
and management approaches for preventing and reducing risks from pollution that threatens human health and the
environment. The focus of the Laboratory's research program is on methods and their cost-effectiveness for prevention
and control of pollution to air, land, water, and subsurface resources; protection of water quality in public water
systems; remediation of contaminated sites, sediments and ground water; prevention and control of indoor air pollution;
and restoration of ecosystems. NRMRL collaborates with both public and private sector partners to foster technologies
that reduce the cost of compliance and to anticipate emerging problems. NRMRL's research provides solutions
to environmental problems by: developing and promoting technologies that protect and improve the environment;
advancing scientific and engineering information to support regulatory and policy decisions; and providing the technical
support and information transfer to ensure implementation of environmental regulations and strategies at the national,
state, and community levels.
This work was supported by EPA's Office of Research and Development through the Regional Applied Research Effort
(RARE) Program. This program is designed to 1) provide the Regions with near-term research on high priority, Region-
specific technical needs, 2) improve collaboration between Regions and ORD laboratories, 3) build the foundation
for future scientific interaction, and 4) develop useful tools for state, local and tribal governments to address near-
term environmental issues. EPA Region 6 and ORD's Ground Water & Ecosystems Restoration Division (GWERD)
recognized the need to evaluate whether properly-designed Comprehensive Nutrient Management Plans (CNMPs)
developed for land application of waste from Concentrated Animal Feeding Operations (CAFOs) are truly protective of
ground water quality. Funding ($13 OK total) was awarded to EPA Region 6 (Nancy Dorsey, EPA Region 6 Contact) and
administered through GWERD (Elise Striz, Stephen Hutchins, Project Officers), and was used by USDA's Agricultural
Research Service (David Brauer, USDA-ARS Contact) to conduct two separate site investigations at CAFO facilities
where CNMPs were being followed. The objective was to conduct comprehensive sampling of soil, soil water, and
crops for nutrients throughout the growing season to determine which simple soil/crop metrics are the best indicators
of the potential for nutrients to escape the root zone and become a threat to ground water. This report describes the site
investigation conducted by Dr. Philip A. Moore, Jr., and Dr. David Brauer for a dairy farm in Arkansas. The other site
investigation was conducted by Dr. Jerry L. Hatfield for a swine operation in Iowa and is described in the companion
report.
Robert W. Puls, Acting Director
Ground Water and Ecosystems Restoration Division
National Risk Management Research Laboratory
-------
-------
Contents
Notice ii
Foreword iii
Contents v
Diagrams and Figures vi
Abbreviations vii
Acknowledgements viii
Executive Summary ix
1.0 Introduction 1
2.0 Materials and Methods 3
Site Selection 3
Soil Samples 3
Lysimeter Samples 3
Effluent Application and Collection of Applied Effluent Samples 4
Forage Samples 5
Statistical Analyses 5
3.0 Results and Discussion 6
Precipitation Data 6
Lysimeter Data 6
Soil Data 9
Forage Data 11
Metrics for Nitrate Leaching 11
4.0 Conclusions and Recommendations 13
5.0 References 14
-------
Diagrams and Figures
Diagram 1. Schematic showing location of plots within Fields 1 and 2 3
Diagram 2.Stainless steel lysimeter installation 4
Figure 1. Daily precipitation between June 1, 2004 (Day -56) and September 1, 2005 (Day 401) 6
Figure 2. Lysimeter nitrate concentrations as a function of time 6
Figure 3. Concentrations of total N (TN) in lysimeter ground water samples as a function of time. ... 7
Figure 4. Ammonium concentrations in lysimeter ground water samples as a function of time 8
Figure 5. Relationship between concentrations of nitrate, organic N, and ammonium N and total N
concentrations in lysimeter ground water samples 8
Figure 6. Concentrations of soluble reactive P in lysimeter ground water samples as a function of
time 9
Figure 7. Concentrations of total P in lysimeter ground water samples as a function of time 9
Figure 8. Soil nitrate concentrations in samples taken from 0-5 cm depth as a function of time 10
Figure 9. Concentrations of soil inorganic N for samples taken from 0-5 cm depth as a function of
time 10
Figure 10. Depth distribution of soil nitrate (panel A) and inorganic N (panel B) 10
Figure 11. Depth distribution of soluble reactive P (panel A) and Mehlich III P (panel B)
concentrations in soil samples 11
Figure 12. Forage nitrate concentrations as a function of time 11
Figure 13. Relationship between lysimeter nitrate and forage nitrate concentrations 12
-------
Abbreviations
CAFOs Concentrated Animal Feeding Operations
HUA High Use Area
LSD Least Significant Difference
MCL Maximum Contaminant Level
N Nitrogen
NH4-N Ammonium Nitrogen
NMP Nutrient Management Plan
NO3-N Nitrate Nitrogen
P Phosphorus
-------
Acknowledgements
The authors wish to thank Scott Becton, Suzanne Horlick and Jerry Martin for their contributions to this work. The
authors also appreciate Dr. Charles MacKown (USDA/ARS, El Reno, OK) for conducting forage nitrate assays.
-------
Executive Summary
Nitrate is the most common chemical contaminant found in ground water. Recent research by U. S. EPA has shown
that land application of manure can cause nitrate contamination of ground water above the maximum contaminant
levels (MCLs) of 10 mg NO3-N/L at significant depths. This finding and similar ones across the nation are raising
concerns about the potential for manure to degrade ground water quality near concentrated animal feeding operations
(CAFOs). The objectives of this research were to determine if nutrient management plans (NMPs) for CAFOs are
inherently protective of ground water and which metrics can be used as red flags to identify when the land application
practices pose a risk to ground water.
A study was conducted for one year (August 2004 to July 2005) on a typical dairy farm in northwest Arkansas. The
dairy had 250 cows and utilized 27 ha (67 acres) for manure application. Effluent from the holding pond was sprayed
onto four 4.05 ha (10-acre) fields each year. According to the farmer's NMP, effluent applications were to occur
during the growing season after the soil had dried considerably, which did not occur until August in 2004. Four small
(10 x 10 m) plots were established in each of two of these spray fields. Stainless steel lysimeters were installed to
a depth of 1 m and sampled weekly. Three soil cores were taken periodically from seven depths (0-5, 5-10, 10-20,
20-40, 40-60, 60-80 and 80-100 cm). Soil samples were collected 20 times throughout the year. These samples were
analyzed for soluble components as well as exchangeable ammonium and Mehlich III extractable P. Plant samples
were also analyzed for nitrate.
Nitrate levels in lysimeter samples were high, with peaks in excess of 100 mg NO3-N/L. The amount of N applied
via effluent averaged 280 kg N/ha (250 Ibs N/acre), which was not believed to be sufficient to cause such high levels.
Lysimeter P concentrations were also very high. Beginning in November, it was observed that the farmer utilized
the 4.05 ha fields as a loafing area or high use area (HUA) for his cows. We estimate that at the observed stocking
rate (31 cows/ha) as much as 840 kg N/ha (750 Ibs N/acre) was being added to these fields via direct waste deposits
from the cows. When coupled with the effluent application, the total N loading to these fields was approximately
1100 kg N/ha (1000 Ibs N/acre) in one year.
When state officials were contacted to determine if direct deposits were taken into account when determining
maximum N application rates for farms spreading liquid manures in Arkansas, we discovered that these deposits by
the cows were not accounted for. Currently, the Arkansas Phosphorus Index is being revised. During this revision
we will attempt to change regulations so that growers cannot apply effluent to a HUA unless the direct deposits are
properly accounted for. The best predictor of high nitrate in the lysimeter samples was leaf tissue nitrate concentration
with R-square of 0.82. Since nitrate toxicity in cows is a problem that will negatively affect production, dairy farmers
could easily be convinced to monitor this parameter. High nitrate levels in forage would allow them to know which
fields were receiving too much N, and allow them to alter applications accordingly.
-------
-------
1.0
Introduction
Animal agriculture is important to the economy of both
Arkansas and the nation. Modern farms often have large
numbers of animals and a relatively limited land base to
apply the manure. This can lead to the problem of over
application of nutrients, particularly nitrogen (N) and
phosphorus (P) to agricultural lands.
Nitrate (NO3") is soluble in water, hence, it can be
easily leached from soils into ground water. As
a result, nitrate is the most ubiquitous chemical
contaminant in the world's ground-water supplies
(Spalding and Exner, 1993). High concentrations
of nitrate in drinking water can cause problems if
consumed by infants possibly resulting in a serious
illness called methemoglobinemia or "blue-baby
disease." Blue-baby disease is due to conversion of
nitrate to nitrite by the immature gastrointestinal tract,
which results in low oxygen levels in the blood.
As a result of this potential human health threat, the
U.S. EPA established a maximum contaminant level
(MCL) of 10 mg nitrate-nitrogen (NO3-N)/L for nitrate
in drinking water (U.S. EPA, 1995). Unfortunately,
Nolan et al. (1998) reported that U.S. EPA MCL was
exceeded 24 percent of the time in the United States
when data from 1400 wells sampled between 1993 and
1995 were analyzed. Approximately 20 percent of the
population of France (10 million people) depends on
ground water with nitrate levels above the European
Community's upper threshold of 11.3 mg NO3-N/L
(Spalding and Exner, 1993). Nitrate contamination of
ground water near intensive vegetable production has
been reported in Japan (Babiker et al., 2004).
Several surveys of ground-water nitrate levels have been
completed in the past in areas considered prone to high
nitrate, including areas with high poultry production, like
Arkansas (Arkansas CES, 1990; Steele and McCalister,
1991) and Delaware (Ritter and Chirnside, 1984). The
number of wells with nitrate levels greater than 10 mg
NO3-N/L was relatively low in Arkansas (Steele and
McCalister, 1991). This was not the case in Delaware,
particularly in the number one broiler producing county
(Sussex County) where 37 percent of the wells had
nitrate levels above the MCL (Ritter and Chirnside,
1984). Ritter and Chirnside (1984) concluded that
nitrate leaching from poultry manure was likely the
major source of the high ground-water nitrate levels in
this county. Only 3.2 percent of the 1232 wells sampled
in a ten county area of Arkansas had nitrate levels
above the MCL and most of these were in the poultry
producing areas (Arkansas CES, 1990). However, Steele
and McCalister (1991) reported the average nitrate
concentration was only 3 mg NO3-N/L in areas receiving
heavy applications of poultry litter in western Arkansas.
Adams et al. (1994) studied nitrate leaching in a Captina
silt loam soil in northwest Arkansas for 1 year after
being fertilized with various rates (0, 5, 10 and 20 Mg/ha
or approximately 2.5 to 10 tons/acre) of poultry litter and
laying hen manure, which corresponded to N application
rates of 0, 220, 440 and 880 kg N/ha. Lysimeter data
taken from 1.2 m depth showed that only the 10 and
20 Mg/ha rates resulted in soil solution nitrate values in
excess of 10 mg NO3-N/L with maximum concentrations
of 24 and 37 mg NO3-N/L, respectively. Adams et al.
(1994) concluded that if manure application rates are
made at the recommended rates in Arkansas, less than
11 Mg/ha or 5 tons/acre (Daniels et al., 2008), then
excessive nitrate leaching should not occur. These
results confirmed those of Marriott and Bartlett (1975),
who showed that manure application rate played an
important role in nitrate leaching. They indicated that
manure could be applied at twice the crop's N needs
with minimal threat of nitrate leaching to the ground
water, provided the manure is applied to a deep well-
drained soil and the crop is harvested.
Although rate of N application is an important
determinant of the leaching potential of nitrates, other
factors also influence the concentration of nitrates in soil
water below the plants' rooting depth. Both the amount
of and the timing of precipitation events and irrigation
water applications relative to time of N applications
affect the amount of nitrate found deep in the soil
profile (Cardenas et al., 2005; van Es et al., 2006).
There may be a value in the use of estimates of crop
evapotranspiration to scheduling irrigation amounts and
frequency to limit the amount of nitrate leaching through
the soil profile (Gardenas et al., 2005). Leaching
potential of nitrate through coarse texture soils is greater
than for finer texture soils (van Es et al., 2006). Source
of the N, including the type of manure, also influences
the rate at which nitrate moves through the soil profile
(Giullard and Kopp, 2004; Wu and Powell, 2007).
Spalding and Exner (1993) stated that high temperatures,
abundant rainfall and relatively high organic contents
in Coastal Plain soils of the southeastern United States
promote denitrification below the root zone and naturally
remediate nitrate leaching into ground water. In North
Carolina, Gilliam (1991) found that high levels of
-------
nitrates (15-20 mg NO3-N/L) occurred in soil solutions
in Coastal Plain soils cropped to corn. However, these
high concentrations were not measured below 4 m.
Gilliam (1991) attributed these low nitrate levels at
greater depths to denitrification (soluble organic carbon
compounds provide an energy source for microbial
reduction of nitrate).
Several methods for monitoring nitrate leaching have
been used in the past. Nitrate analyses of soil cores
provided estimates of nitrate leaching as reliable as two
different types of soil lysimeters (Zotarelli et al., 2007).
Zhu et al. (2002) and Toth et al. (2006) demonstrated
that passive capillary lysimeters also provided reliable
estimates of nitrate leaching, while being easier to install
and maintain than other types of lysimeters. The problem
with these methods is that they are labor intensive during
installation and/or sample collection, thus limiting their
use by farmers.
Recent work by U.S. EPA personnel in Oklahoma have
demonstrated that land application of swine manure can
cause nitrate contamination of soil water above the MCL
at depths greater than 10m (Elise Striz, unpublished
data). These findings, along with similar findings around
the country, are raising concerns for soil and ground-
water degradation on or adjacent to concentrated animal
feeding operations (CAFOs). Currently, land application
of manures from CAFOs must follow a nutrient
management plan (NMP) in most states. In Arkansas,
NMPs for liquid manures have been regulated for over
a decade by Regulation 5, which states that manure
applications will be based on the N needs of the crop.
One of the main underlying assumptions of using a well
designed and executed NMP is that ground water will
be protected from excessive amounts of nitrate or other
nutrients.
In the southern United States, attention has focused
on the potential of animal manure from poultry and
integrated poultry and beef cattle operations contributing
to ground-water contamination because of the number
of these operations. In 1997, poultry and beef cattle
production comprised 50 percent of the agricultural
income in Arkansas and over 90 percent of the animal
manure production (VanDevender, 1997). Dairy farms
in Arkansas, on the other hand, accounted for 2 percent
of the farm income and 5 percent of the animal manure
production. The stocking rate of cattle (number of
head/ha or acre) tends to be greater with dairy than
beef cattle. Dairy cattle routinely require supplemental
feeding with both energy- and protein-rich feeds, thus
leading to an importation of nutrients to the farm. A
disproportionate amount of these nutrients are fed
to dairy cattle during that period of the year when
forages are dormant. The potential for nutrient run-
off or leaching is greater when cattle are being fed
on dormant grasslands (Owens and Shipitalo, 2006).
The amount of water soluble P in dairy manure is
relatively high (Shigaki et al., 2006). Several studies
have indicated that the amount of soluble P applied as
manure is directly associated with the potential for P
losses in run-off (DeLaune et al., 2004a; 2004b). Soupir
et al. (2006) indicated that nutrients are more readily
lost from direct deposition of dairy manure onto the
soil than losses from other application methods. These
previous findings indicate that there is the potential for
ground-water contamination on or adjacent to southern
dairy operations, especially when rainfall exceeds
evapotranspiration and soils are permeable to water.
The objectives of this research were to determine
if an NMP, when properly executed, consistently
protects ground water, and, if not, to determine what
soil/crop/manure metrics could be utilized to predict
potential ground water degradation.
-------
2.0
Materials and Methods
Site Selection
The site selected for this study was a dairy farm
located in Washington County in northwest Arkansas.
The initial meeting with the landowner was made on
July 20, 2004. The dairy had approximately 250 cows
on 27.1 ha (67 acres, 47 acres were grazed; 20 acres
were hayed). Lagoon effluent from the farm was
applied to four 4.05 ha (10-acre) fields of bermudagrass
(Cynodon dactylon (L.) Pers). Two of the four fields,
designated as Field 1 or 2, were randomly chosen for
the study. Within each of the two fields, four plots
(10 x 10 m) were established (Diagram 1). According to
the Washington County Soil Survey (USDA, 1969), the
soil type in both fields was exclusively a Jay silt loam
(fme-silty, mixed, active, thermic Oxyaquic Fragiudalf)
on 1-3 percent slope. Characteristics of soil recovered
from coring to 1 m depth were consistent with this
determination (data not shown). According to the
Washington County Soil Survey (USDA, 1969) the Jay
series consists of well-drained, slowly permeable soils
that have a subsoil fragipan between 0.4 and 1.0 m of
the soil surface. A full description of the Jay silt loam
is available (NRCS, 2000). No attempt was made to
determine the depth at which the fragipan occurred. The
landowner was able to utilize these fields according to
his management scheme, thus cattle were not excluded
from the plots. Rainfall data were obtained from a
USGS gauge station located near the fields (USGS,
2009).
Field 1
Field 2
1174'
Field Plot Schematic; One Lysimeter Per Plot
Diagram 1. Schematic showing location of plots within
Fields 1 and 2.
Soil Samples
Three soil cores (5 cm in diameter) were collected
from each plot on each sampling date. The cores were
divided into 7 discrete samples by depth (0-5, 5-10,
10-20, 20-40, 40-60, 60-80 and 80-100 cm). Soils
from each depth of the three cores were combined for
each plot. Thus, there were a total of 56 soil samples
(8 plots x 7 depths) taken at each sampling time. First
soil samples were taken on July 24, 2004. Soil samples
were collected 19 additional times over the next year,
with the bulk of the samples being collected just after
dairy effluent applications. Collection of soil samples
terminated prior to effluent application in 2005.
Soluble reactive phosphorus, nitrate-N, and
ammonium-N were extracted using a 1:10 (soil:water)
extraction for one hour (Self-Davis et al., 2000). A
sequential KC1 extraction was conducted on the soil
for exchangeable ammonium. Nitrate (+nitrite) was
determined using the Cd reduction method on filtered
(0.45 |jm) samples (Method 418-F; APHA, 1992).
Soluble reactive P samples were also filtered (0.45 |jm),
and acidified to pH 2.0 with HC1. Soluble reactive
P was determined using the ascorbic acid technique
(APHA Method 424-G; APHA, 1992). Ammonium was
determined with the salicylate-nitroprusside technique
on filtered (0.45 |jm), acidified samples (Method 351.2;
U.S. EPA, 1979). Soil samples were also analyzed
for Mehlich III P (Mehlich, 1984). Levels of soil
constituents were expressed as mg per kg of dry soil.
These concentrations were converted to pounds per
acre assuming 2,000,000 pounds of soil per acre-foot of
topsoil.
Lysimeter Samples
One stainless steel lysimeter (50 cm long and 5 cm
in diameter) was installed within each of the four
10 x 10 m2 plots in each field on August 2, 2004
(Diagram 2). Both the equipment and installation
were essentially as described by Adams et al. (1994).
Installation began by digging a hole of sufficient size to
accommodate an armor valve box (approximately 15 cm
in diameter and 20 cm in height). From the bottom
of this hole a second hole was dug at a 45-degree
angle to the soil surface, in which the stainless steel
lysimeter and the attached vacuum tubes were placed.
Excavated topsoil and subsoil were kept separately.
After placement of the lysimeter, the hole was filled
with the removed material starting with excavated
subsoil. Such an installation minimizes the disturbance
-------
vacuum grade tygon tubing
vv equipped with 3-way stopcocks
Dashed lines represent soil that was excavated x
which was later refilled using the same soil
stainless steel lysimeter with 0.45 urn holes
Diagram 2. Stainless steel lysimeter installation.
to the soil immediately above the sample collection
chamber. Leachate from the sample collection chamber
was collected by the application of a vacuum. Attempts
to gather ground-water samples from the lysimeters
were made on a weekly basis. Between September 9,
2004 and October 31, 2004, ground-water samples were
obtained from less than half of the lysimeters. During
November 2004, ground-water samples were obtained
from over hah0 of the lysimeters, but not all. From
December 1, 2004 through May 14, 2005, ground-
water samples were obtained from all lysimeters. After
May 14, 2005, the number of lysimeters with ground-
water samples declined until only two lysimeters in
Field 2 had samples. Attempts to gather ground-water
samples were terminated prior to effluent application in
2005. Chemicals present in the soil water at a depth of
1 m are likely to continue to percolate downward until
reaching ground water because a vast majority (over
95 percent) of roots and associated organic C are located
at a soil depth of less than 0.9 m in bermudagrass
dominated agro-ecosystems (Carreker et al. 1977;
Franzluebbers and Stuedemann, 2005). Lysimeter
samples were analyzed for pH, electrical conductivity
(EC), total organic C (TOC), total nitrogen (TN),
ammonium (NH4+), nitrate (NO3"), soluble reactive
phosphorus (SRP), and total P (TP). Ammonium was
determined with the salicylate-nitroprusside technique
(Method 351.2; U.S. EPA, 1979). Soluble reactive P
samples were filtered (0.45 |jm), acidified to pH 2.0
with HC1 and frozen. Soluble reactive P was determined
using the ascorbic acid technique (Method 424-G;
APHA, 1992). Total N and total organic C were
analyzed using a Skalar total N and C analyzer. Total P
and metals were determined using a Spectro inductively
coupled argon plasma emission spectrometer following
acid digestion.
Effluent Application and Collection of
Applied Effluent Samples
Effluent applications were made according to an
approved NMP. Application amounts were based
on N needs of the forage. Liquid manure from the
holding pond was applied via a travel gun on Field 2
on August 3, 2004 and on Field 1 on August 18, 2004.
Length of effluent application was 1 hr 45 min for
Field 1 and 2 hr for Field 2. The pressure at which
the liquid was sprayed was 207 kPa (30 pounds per
square inch) on both fields. Four shallow plastic pans
(16.5 x 30.5 cm) were placed in each of the four plots
during effluent application. The effluent collection area
represented approximately 2 percent of the plot area.
The volume of effluent captured in each was recorded.
The captured effluent was analyzed for total N and
nitrate concentrations. The captured effluent had an
average total N content of 651 and 514 mg N/L for
Fields 1 and 2, respectively. Nitrate concentrations were
0.30 and 0.38 mg NO3-N/L, respectively, for effluent
collected on Fields 1 and 2. These results indicate
that most of the N in the effluent was not in the nitrate
form. The average application rate applied to the plots
in Field 1 was 602,100 L/ha (or 64,100 gallons/acre).
The amount of effluent captured in the collection pans
in the plots in Field 2 was considerably less, averaging
330,000 L/ha (or 35,100 gallons/acre). Using the total N
content of the collected effluent, N application rates of
391 and 168 kg N/ha (or 349 and 151 Ibs N/acre) were
calculated for Fields 1 and 2, respectively. The average
total N application rate for both fields was 280 kg N/ha
(or 250 Ibs N/acre).
What was captured in the pans was probably
representative of what the plots received, but may not
have been representative of what was applied to the
entire field. The traveling gun was spraying effluent
on Field 2 for a slightly longer period of time than on
Field 1, 2 hours versus 1.75 hours (see above). Hence,
Field 2 should have had slightly more manure applied
than Field 1. The reason for discrepancy between
what was applied to the field and what fell on the plots
is probably the non-uniform application which was
made by the gun. The location of the plots, direction
and speed of wind, etc., relative to the position of the
traveling gun in the field were not the same. Very few,
if any, growers spreading liquid manure in Arkansas
have center pivot irrigation systems. They have
traveling guns, like the one this grower utilized. Hence,
uneven applications within fields are probably typical of
on-farm situations. No weather station was installed at
the experimental site. Daily precipitation amounts were
obtained from the USGS stream gauge 07048600, which
was located within 10 km of the research site (USGS,
2009).
-------
Forage Samples
Leaf tissue was collected from vegetation growing in
each of the 8 plots every 30-40 days. Above ground
biomass was clipped within 2 cm of the soil surface
from 10 randomly selected areas within each plot, each
approximately 15 cm in diameter. A sub-sample from
the leaf material collected from each plot was dried at
60°C for 48 hours, ground to pass a 40-mesh screen and
nitrate-N was determined via nitrate reductase assay
(MacKown and Weik, 2004).
Statistical Analyses
Data were analyzed by analysis of variance using SAS
(SAS Institute, 1985). The experimental unit was a plot
within one of the two fields. The experimental design
included sampling dates, plots within field and fields as
the main factors. In the statistical analyses, missing data
from a sampling date were treated as missing values.
Least significant differences (LSD) at P = 0.05 are
included in figures and text for mean comparison tests.
Days mentioned in text and accompanying figures refer
to days after the study was initiated, i.e. July 27, 2004.
-------
3.0
Results and Discussion
Precipitation Data
Daily precipitation from June 1, 2004 to September 1,
2005 is presented in Figure 1. Sufficient rainfall
occurred in June and July 2004 (Day -56 to Day 0) to
prevent the application of effluent to the bermudagrass
fields until early August, as to comply with the soil
moisture parameters of the farm's NMP. A substantial
rainfall event occurred immediately after the effluent
applications. Approximately 8-10 cm of water were
applied to the fields by effluent application and rainfall
between August 1 (Day 5) and September 9, 2004
(Day 44), the date that the first soil water samples were
obtained from the lysimeters. Little rainfall fell during
the fall of 2004. Several rainfall events occurred around
December 1, 2004 (Day 127) and February 1, 2005
(Day 189). Frequent rainfalls of over 1 cm occurred
between April 1 and July 1, 2005 (Days 248-339).
5 -
4 •
3 •
2 -
1 •
^~~1 1 T~
Jl
-56 0 56
i
.
JUlii
1 1
JiLii
1
id
1,
112 168 224 280 336 392
Time (days)
Figure 1. Daily precipitation between June 1,2004 (Day -56)
and September 1,2005 (Day 401). Lime zero is
July 27, 2004. Daily precipitation was obtained
from USGS gauge station 07048600 located on
the White River near Fayetteville AR. Lhis gauge
station is located within 10 km of the experimental
site.
Lysimeter Data
The first lysimeter ground-water samples were
collected on September 9, 2004, approximately three
to four weeks after effluent applications were made.
Nitrate levels in the lysimeters were exceedingly high
(Figure 2), exceeding 200 mg NO3-N/L at Day 44 of
the study (September 9, 2004). Nitrate levels were
above 100 mg NO3-N/L from Day 0 to Day 150 (mid-
December 2004). Nitrate levels steadily decreased
until around Day 200 (February 2005), then increased
again. The average nitrate concentration in Field 1 was
significantly (P < 0.05) higher than Field 2 (149 versus
90 mg NO3-N/L). This may have been due to the higher
N application rate in the plots in Field 1 versus Field 2.
Ground-water samples were collected from at least one
lysimeter in Field 2 for a longer period of time compared
to Field 1, thus accounting for the data beyond Day 320
for Field 2. Trends in total N levels in lysimeter samples
were similar to those of nitrate (Figure 3).
Field 1
Field 2
0 50 100 150 200 250 300 350
Time (days)
Figure 2. Lysimeter nitrate concentrations as a function of
time. Time zero is July 27, 2004, date of the first
soil sampling. Days 100, 200 and 300 correspond
to November 4, 2004, February 12,2005 and
May 23, 2005, respectively. Effluent applications
occurred on August 3 and 18, 2004 for Fields 2 and
1, respectively, or Day 6 and 21, respectively. Data
are means across four plots (n = 4) within each
field at a sampling date (LSD = 22.0 mg N/L).
The discovery of higher than expected nitrate values
at the beginning of the study was hard to explain at
first, since N applications from effluent appeared
to be insufficient to cause such elevated levels. As
stated earlier, Adams et al. (1994) only found a high
of 37 mg NO3-N/L in a soil fertilized with manure
containing 880 kg N/ha; an N application rate far in
excess of what was applied to these fields. Moore et
al. (2000) measured much lower nitrate levels in soil
solutions on two typical poultry/beef farms in northwest
Arkansas during a two year study, with the highest
observed nitrate concentrations being 2.8 mg NO3-N/L,
roughly 100 fold lower than the peak values found in
this study.
-------
100
0 50 100 150 200 250 300 350
Time (days)
Figure 3. Concentrations of total N (TN) in lysimeter
ground-water samples as a function of time.
Time zero is July 27, 2004, date of the first soil
sampling. Days 100, 200 and 300 correspond
to November 4, 2004, February 12, 2005 and
May 23, 2005, respectively. Effluent applications
occurred on August 3 and 18, 2004 for Fields 2
and 1, respectively. Data are means across four
plots within each field (n = 4) at a sampling date
(LSD = 29.4 mg/L).
Beginning in early November 2004, it was observed
that the dairy farmer was using the two fields where
effluent was applied for a loafing area for the cattle.
Within a month of using this area for this purpose, the
soil surface was covered in a layer of manure that was
1-2 cm deep (direct deposits from the dairy cows). The
assumption was made that using the effluent application
fields for loafing areas was a direct violation of the
nutrient management plan. However, personnel from
USDA/NRCS and the University of Arkansas indicated
that no such restrictions were in place and they routinely
did not account for direct deposits of N and P when
evaluating or developing NMP for effluent applications.
The grower at this farm was using the two 4.05 ha
(10-acre) spray fields as the dairy loafing area in the
winter for all of his cows. Hence, 250 cows were
being kept on roughly 8.1 ha (20 acres) all day,
except when the cows were moved to and from the
milking parlor. Thus, the stocking rate was about
31 cows/ha (or 12.5 cows/acre). Atypical dairy
cow weighing 500 kg (1,100 Ibs) excretes roughly
0.23 kg N/day (0.5 Ibs N/day) in manure and urine
(USDA, 1992). Therefore, approximately 7 kg N/ha
(6 Ibs N/acre) were being added each day by the cattle
via direct deposits. These fields were used as a loafing
area for 3 to 4 months during the year of the study.
Responses from the farmer indicate that use of these
fields as a loafing area for 3 to 4 months annually had
been a common practice for several years. If the fields
were used as a loafing area for 90 to 120 days, then the
total direct deposit of N from cattle would be 620 to
820 kg N/ha (560 to 750 Ibs N/acre). These estimates
indicate that direct deposit of manures added three times
the average amount of N that was added via effluent
applications (i.e. 280 kg N/ha or 250 Ibs N/acre).
Thus, the total annual N loading on these fields was
at least 800 kg N/ha and could easily have exceeded
1,000 kg N/ha.
As mentioned earlier, nitrate concentrations in ground
water have been shown to be directly influenced by the
amount of manure applied (Adams et al., 1994; Marriott
andBartlett, 1975). Results of Adams etal. (1994)
indicated that N application rates from poultry litter
additions had to exceed 440 kg N/ha for ground-water
samples taken at 120 cm to exceed 10 mg NO3-N/L.
Results of Adams et al. (1994) predict that lysimeter
NO3-N concentrations in this study should have been
less than 10 mg NO3-N/L if N was being added only
from the effluent applications. Lysimeter NO3-N
concentrations in this study typically exceeded that
observed by Adams et al. (1994) suggesting that the N
application rate was in excess of the previous study's
highest rates, 880 kg N/ha. Therefore, total N input of
at least 1,000 kg N/ha seems plausible, based on the
observed nitrate-nitrogen concentrations in lysimeter
samples.
If most of the N additions to these soils were occurring
from direct manure deposits during the winter, why did
lysimeter nitrate values decline during this time period
(Figure 2)? One possible explanation is that manure
deposited on the soil surface during the winter did not
start to break down until warmer weather in spring. The
dominant N forms in dairy manure were not nitrate.
Nitrate is the form of soil N most susceptible to leaching.
Thus, significant mineralization of the N in the manure
would have to occur to produce nitrate for leaching
to be detected. This mineralization could account for
the increases in lysimeter nitrate and total N levels
beginning between Days 225 and 250 (mid-March
2005) in Field 1 and between Days 275 and 300 (mid-
June 2005) in Field 2. Neither of these increases in
lysimeter concentrations was due to additional effluent
applications. Precipitation amounts during the winter
months of 2005 were less than historical means
(Figure 1), thus downward movement of nitrate and
other N compounds into the soil profile would have been
minimal. As spring progressed, warmer temperatures
coincided with increases in precipitation. Increased
precipitation and temperatures would have promoted
both N transformation and downward movement of
soluble water N compounds, thus explaining increases in
lysimeter N concentrations after Day 250.
Although the total N concentrations in lysimeter samples
followed the same patterns as nitrate, the values were
-------
about 25 percent higher (Figure 3). Total N is higher
because it includes other forms of N, such as ammonium
and organic N forms. Mean total N concentrations
for Field 1 were higher than for Field 2 (189 versus
108 mg N/L), as was the case for nitrate, and consistent
with the higher N application rate on Field 1. Differences
in average lysimeter concentrations between the
two fields suggest that the addition of N via effluent
application was sufficient to alter ground-water levels of
nitrate and total N.
Lysimeter ammonium values were less than those
for nitrate (Figure 4). Twice during the experiment,
ammonium values from the lysimeters in Field 1
exceeded 4 mg N/L. One event occurred around
Day 130 (early December 2004) and the other was
observed around Day 230 (mid-March 2005). It is
unclear what caused these spikes; they did not coincide
with either effluent applications or high amounts of
rainfall. The mean ammonium concentrations in soil
solutions were significantly higher in Field 1 than
Field 2 (4.4 versus 0.6 mg NH4-N/L). These higher
concentrations in the plots in Field 1 may have been due
to the higher application rate of effluent, and thus, N.
50 100 150 200 250 300 350 400
Ammonium concentrations in lysimeter
ground- water samples as a function of time.
Time zero is July 27, 2004, date of the first soil
sampling. Days 100, 200 and 300 correspond
to November 4, 2004, February 12, 2005 and
May 23, 2005, respectively. Effluent applications
occurred on August 3 and 18, 2004 for Fields 2
and 1, respectively. Data are means across four
plots within each field (n = 4) at sampling date
(LSD= 1.97mg/L).
The amount of nitrate in the lysimeters was highly
correlated with total N, as was organic N (Figure 5).
However, ammonium was poorly correlated to total N
concentrations. The slopes from these three relationships
indicate that about 77 percent of the total N was nitrate,
22 percent was organic N and 1 percent was ammonium.
0.200
organic N nitrate N ammonium N
0 100 200 300 400 50C
Lysimeter Total Nitrogen Concentration (mg N/L)
Figure 5. Relationship between concentrations of
nitrate, organic N, and ammonium N to total N
concentrations in lysimeter ground-water samples.
Regression equations and regression coefficients
(R) are presented in figure.
It is somewhat unusual for shallow lysimeters installed
in soils in northwest Arkansas to have ground-water
samples for a majority of the year as observed in this
study. Normally lysimeter samples can only be taken a
few weeks of the year following heavy rainfall events.
The soils in these two fields were a Jay silt loam, which
is known to have a fragipan in the subsoil. The fragipan
restricts the downward movement of water, thus leading
to a perched water table in the subsoil for at least a
portion of the year. Therefore, it is quite possible that
the lysimeters were sampling water from this perched
water table and that is the reason why lysimeter samples
could be collected for a large part of the year.
The chemical composition of the lysimeter samples,
which contained high concentrations of nitrate and low
levels of ammonium, indicate that the soil solution
remained oxidized (under anaerobic conditions ammonia
will not be oxidized to nitrate). This is somewhat
surprising, since the deeper soil samples were often dark
grey to blue in color, which normally indicates anaerobic
conditions. However, the average iron and manganese
concentrations in soil solution were only 0.21 and
0.14 mg/L, respectively (data not shown), which also
indicate oxidized conditions. Had anaerobic conditions
been present, then denitrification should have occurred,
since plenty of organic C was present. Concentrations of
total organic C were typically near 50 mg C/L or above
(data not shown).
Soluble reactive P levels in the lysimeters were
typically between 0.5 and 2.0 mg P/L, but increased to
around 7 mg P/L in Field 1 halfway through the study
(Figure 6). Soluble reactive P levels were significantly
higher in Field 1 than Field 2 (1.60 versus 1.02 mg P/L).
As with total N and nitrate concentrations, the
-------
differences in soluble reactive P between the two fields
indicate that the P additions via the effluent application
were sufficiently large to affect ground-water sample
concentrations. Typically, soil water P values taken from
lysimeters at 1 m depth are much lower in northwest
Arkansas, ranging between 0.01 to 0.05 mg P/L (Philip
Moore, unpublished data). Total P concentrations of
water samples taken from the lysimeters were also high
(Figure 7). Total P values were only slightly higher than
those for soluble reactive P (Figures 6 and 7), indicating
that most of the P in the lysimeter samples was soluble
reactive P. These data indicate that this soil was
saturated with P.
s!
o
C/J
Field 1
Field 2
0 50 100 150 200 250 300 350
Time (days)
Figure 6. Concentrations of soluble reactive P in lysimeter
ground-water samples as a function of time.
Time zero is July 27, 2004, date of the first soil
sampling. Days 100, 200 and 300 correspond
to November 4, 2004, February 12, 2005 and
May 23, 2005, respectively. Effluent applications
occurred on August 3 and 18, 2004 for Fields 2
and 1, respectively. Data are means across four
plots within each field (n = 4) at a sampling date
(LSD=1.97mg/L).
As with N, a lot of the P present in these fields probably
originated from direct deposits. A 500 kg dairy cow
excretes roughly 0.03 kg P/day. If the field were used as
loafing areas for four months with an average density of
31 cows/ha, then roughly 110 kg P/ha (100 Ibs P/acre)
would have been deposited. While significant amounts
of N may be lost via gaseous emissions, such as
ammonia volatilization or possibly denitrification, the
only loss mechanisms for P would be runoff or leaching.
Soil Data
One set of soil samples was taken prior to effluent
application on Field 2. Three sets of soil samples were
taken prior to effluent application on Field 1, and one set
was collected the day after effluent application (i.e.,
0 50 100 150 200 250 300 350 400
Time (days)
Figure 7. Concentrations of total P in lysimeter ground-
water samples as a function of time. Time zero is
July 27,2004, date of the first soil sampling. Days
100,200 and 300 correspond to November 4, 2004,
February 12,2005 and May 23, 2005, respectively.
Effluent applications occurred on August 3 and 18,
2004 for Fields 2 and 1, respectively. Data are
means across four plots within each field (n = 4) at a
sampling date (LSD = 2.43 mg/L).
August 19, 2004). Soil nitrate concentrations in the
surface layer (0-5 cm) were high throughout the study
(Figure 8). There appeared to be a definite response of
soil nitrate concentrations in the surface layer to the
effluent application in Field 1. Soil nitrate levels at
the soil surface (0-5 cm) in Field 1 varied between 60
and 100 mg NO3-N/kg soil over the first four sampling
dates, three prior to effluent application and one day
after. Soil nitrate levels then increased over two-fold,
exceeding 150 mg NO3-N/kg soil, during September and
October 2004 (Days 30-100). This peak in soil nitrate
levels appeared to occur in response to precipitation
that occurred during September and October. Warm
temperatures and adequate moisture are necessary
for soil microbes to convert the N in the effluent to
nitrate. Such a trend was less apparent in Field 2. Soil
nitrate levels in the soil surface (0-5 cm) in Field 2
averaged almost 200 mg NO3-N/kg at the beginning of
the study, and increased slightly in the samples on the
second sampling date, which was 2 days after effluent
application. Surface concentrations of soil nitrate
decreased to approximately 80 mg NO3-N/kg soil by
the fourth sampling date (Day 23). As with Field 1,
soil nitrate concentrations in the surface layer (0-5 cm)
increased between the fourth and fifth sampling dates
and remained at an elevated level to approximately
Day 100 (Figure 8). These results are consistent with
a higher rate of N addition via the effluent to Field 1
compared to Field 2. Soil inorganic N (nitrate plus
ammonium) followed similar trends, but peaked between
250 and 300 mg N/kg (Figure 9). Comparisons of
the concentration of soil nitrate relative to inorganic
-------
N indicate that nitrate was the dominant inorganic
N species in the soil, as was found in the lysimeter
samples.
300
± 100
Figure 8.
Soil nitrate concentrations in samples taken
from 0-5 cm depth as a function of time. Time
zero is July 27,2004, date of the first soil
sampling. Days 100, 200 and 300 correspond
to November 4, 2004, February 12,2005 and
May 23, 2005, respectively. Effluent applications
occurred on August 3 and 18, 2004 for Fields 2
and 1, respectively. Data are means across four
plots (n = 4) within each field at a sampling date
(LSD=15.5mg/L).
300
250
200
150
= 100
50
Figure 9.
100
200
Time (days)
300
400
Concentrations of soil inorganic N for samples
taken from 0-5 cm depth as a function of time.
Time zero is July 27, 2004, date of the first soil
sampling. Days 100, 200 and 300 correspond
to November 4, 2004, February 12,2005 and
May 23, 2005, respectively. Effluent applications
occurred on August 3 and 18, 2004 for Fields 2
and 1, respectively. Data are means across four
plots within each field (n = 4) at a sampling date
(LSD = 21.7mg/L).
Most of the soil nitrate and inorganic N was concentrated
near the soil surface (Figures lOAand 10B). These
data represent the means over the 20 samplings. The
average nitrate concentration at the soil surface was
exceedingly high in both fields - around 100 mg N/kg.
Nitrate levels decreased in the soil profile to a minimum
value of around 10 mg N/kg at the 20-40 cm depth, and
then increased again as depth increased. Such increases
in soil nitrate at depths greater than 40 cm are consistent
with the existence of a layer at or below 1 m that was
impeding the downward movement of soil water and
nitrates. Soil inorganic N (which includes ammonium)
followed the same trends (Figure 10B), but was higher,
175 mg N/kg at the surface.
Average soil nitrate (mg N/kg) Average Total Inorganic N (mg N/kg)
0 25 50 75 100 125 150 175 200 0 25 50 75 100 125 150 175 200
•=•40
Figure 10. Depth distribution of soil nitrate (panel A) and
inorganic N (panel B). Data are means across
four plots within each field and 20 sampling dates
(n = 80). LSD values for comparing means are
8.34 and 11.6 mg/L for soil nitrate and inorganic N,
respectively.
The water soluble soil P values were above 150 mg P/kg
(300 pounds/acre) in the surface layer and decreased
with depth to near zero at the 60-80 cm depth and
below (Figure 11 A). These values at the surface are
extremely high and represent a "worst-case scenario"
for the potential for P losses via run-off. Likewise,
Mehlich III values between 1,250 and 1,500 mg P/kg in
the surface layer (2,500 to 3,000 Ibs P/acre) are higher
than necessary for optimum crop growth (Figure 1 IB).
Unlike soil N concentrations, there was no increase in
soil P values at lower soil depths. The distribution of soil
P in the upper 40 cm was unusual for pastures. Typically,
there is a sharp decline in soil P values with depth
(Braueretal.,2005).
In the past, Arkansas has made recommendations for
NMPs where liquid manure is applied based on N.
Hence, if the soil P values exceeded levels necessary
for optimum crop growth, manure applications
would still have been allowed. However, the state is
-------
currently changing the policy for farms permitted under
Regulation 5 (farms with liquid manure) so that they will
have to use the Arkansas Phosphorus Index to determine
application rates. At these levels of soil P, manure
applications would not be allowed by the Arkansas P
Index.
Average soluble P (mg P/kg)
Average Mehllch III P (mg P/kg)
0 25 50 75 100 125 150 175 200 0 500 1000 1500 2000
Figure 11. Depth distribution of soluble reactive P (panel A)
and Mehlich III P (panel B) concentrations in
soil samples. Data are means across four plots
within each field and 20 sampling dates (n = 80).
LSD values for comparing means are 5.46 and
39.7 mg/L for soluble reactive P and Mehlich III P,
respectively.
Forage Data
As in the case of soil samples, one set of forage samples
was collected prior to effluent application on Field 2 and
three sets of samples were collected from Field 1 prior
to effluent application. Nitrate concentrations in forage
started out very high (near 3,200 mg NO3-N/kg forage
in Field 1) at the beginning of the study (Figure 12).
These high levels of nitrate may have been due to the
nitrogen inputs from direct deposit of manures during
the winter when the fields were used as a loafing area.
Bermudagrass forage is usually considered a low risk
for inducing nitrate toxicity in cattle because it does not
tend to accumulate enough nitrate to induce toxicity
(Strickland et al., 2008). In other studies, nitrate-
nitrogen concentrations in bermudagrass forage did
not exceed more than 400 mg NO3-N/kg (dry weight)
with N fertilizer application up to 100 kg N/ha (Hojjati
et al., 1972). N fertilizer application rates of 200 to
400 kg N/ha increased concentrations to 1,200 to
1,600 kg NO3-N/kg. Toxicity of cattle to nitrates is low
if the forage is less than 3,000 mg NO3-N/kg (Strickland
et al., 2008). Forage nitrate concentrations in samples
from Field 1 declined over the first four sampling dates,
the last of which were collected the day after effluent
application. Forage nitrate concentrations in samples
from Field 1 were higher than the initial levels over
the next four sampling dates. These increases in forage
nitrate concentrations corresponded to increases in
soil nitrate levels in the samples from the surface layer
(0-5 cm). An increase in forage nitrate concentrations
for samples from Field 2 during the early part of the
study was not readily discernible. In general, forage
nitrate concentrations decreased to a minimum of 26 mg
NO3-N/kg forage at Day 220 (February 26, 2005) and
increased to over 2,000 mg NO3-N/kg by Day 274
(April 21, 2005) in Field 1. The forage on this site
was bermudagrass, which is a warm season grass. In
northwest Arkansas, bermudagrass will become dormant
around November 1 and will resume growth by May 1.
During the winter months, nitrate, as well as other water
soluble constituents, is leached from the forage, thus
explaining the decrease in nitrate concentrations during
the winter.
E
•=- 2000
Field 1
•
Field 2
o
0 SO 100 150 200 250 300 350 400 450
Time (days)
Figure 12. Forage nitrate concentrations as a function of
time. Time zero is July 27, 2004, date of the first
soil sampling. Days 100, 200 and 300 correspond
to November 4, 2004, February 12, 2005 and
May 23, 2005, respectively. Effluent applications
occurred on August 3 and 18,2004 for Fields 2
and 1, respectively. Data are means across four
plots within a field (n = 4). Equation and regression
coefficient (R) appear in figure.
Metrics for Nitrate Leaching
Prior to conducting this study we hypothesized that soil
nitrate would be the best metric for predicting nitrate
leaching. However, the relationship between soil nitrate
concentrations at 80-100 cm and lysimeter nitrate levels
was poor (y = -0.79x + 172, r = 0.08 (P > 0.10), data not
shown). The best relationship between soil nitrate and
lysimeter nitrate was with the samples taken from the
0-5 cm depth (y = 0.56x + 92, r = 0.42 (P < 0.05), data
not shown). Although this relationship is statistically
significant, its predictive power is poor.
-------
The best indicator of high concentrations of nitrate
in soil solution was the forage nitrate concentrations
(Figure 13). Many growers often have their hay analyzed
for nitrate to make sure there is little or no potential
for nitrate toxicity in their cattle. Forage nitrate
concentrations followed the same seasonal patterns as
did lysimeter nitrate concentrations, with the highest
levels observed at the beginning and ending of study;
i.e., during summer (Figure 13).
3300
.§ 250
8 150
£
Field 1
•
Field 2
o
y = 0.096 + 64
R = 0.83
0 500 1000 1500 2000 250
Forage nitrate concentration (mg N/kg)
Figure 13. Relationship between lysimeter nitrate and forage
nitrate concentrations. Data are means across
four plots within a field (n = 4) at eight sampling
dates. Regression equation and coefficient (R) are
presented in figure.
-------
4.0
Conclusions and Recommendations
The lysimeter nitrate concentrations and soil P
concentrations observed in this study were orders of
magnitude higher than what is normally observed
in poultry/beef farms located in northwest Arkansas.
The reason for these higher nutrient levels did not appear
to be the effluent applications; but rather, they appeared
to be due to direct manure deposits from dairy cattle
as a result of using the fields as the loafing area for the
cows during winter months. Beginning in November
the surface of these fields became entirely covered with
cow manure. Although the levels of nitrate found in
lysimeters were often in excess of 100 mg NO3-N/L, it
is not known if ground-water contamination actually
occurred, since deeper wells were not installed to
investigate this.
When representatives from USDA/NRCS and University
of Arkansas Cooperative Extension Service were
interviewed about the mechanics of the NMP for liquid
manures using Regulation 5, we discovered that the
amount of nutrients deposited directly from the animals
was not taken into consideration. Under normal grazing
conditions, this practice would not cause a problem,
since manure rates from grazing animals is normally
less than what is applied via effluent. However, in dairy
loafing areas, when dozens of dairy cows consuming a
complete ration are stocked at a high number of animals
per land area, then excessive nutrient applications can
result.
In hindsight, accumulation of N and P in the soils at
this farm should have been expected. The 250 dairy
cows on this farm would be expected to generate about
21,000 kg N and 2,800 kg P in their manure annually,
based on USDA estimates referenced earlier in this
report. These amounts of excreted N and P calculate to
be annual loading rates in excess of 700 kg N/ha and
100 kg P/ha, assuming that only the 27 ha on the farm
were available for deposition.
The use of pastures or hay fields as HUA may be a
difficult situation to correct for many dairy farmers.
The cows must be kept within certain proximity of
the dairy when days are short. In this case, these four
fields receiving the effluent were the four closest fields
to the milking parlor. The solution would be to spread
the cows out over a greater area (make more fields
into loafing areas), so that the manure concentrations
would be reduced. In addition, the effluent from the
holding pond should be applied to other fields that
are not used for loafing. This is also problematic,
since it requires piping the liquid manure greater
distances on the farm, which increases the grower's
cost. However, this is one fairly simple solution that
is easily implemented. Dr. Karl VanDevender is an
Agricultural Engineer with the University of Arkansas
Cooperative Extension Service and is in charge of
Extension activities involving liquid manure applications
in Arkansas. His recommendation for this problem is
as follows: "Growers should decide whether they want
to keep a field in pasture or use it as a HUA or loafing
area. If it is kept in pasture, then the cow density and
use duration should be kept down to the point where
forage is maintained. These pastures would be subject
to Phosphorus Index if additional manure (effluent) is
applied, just as they would for a beef cattle operation. If
the grower decides to make it a HUA, then he/she should
design and manage to 1) minimize the area, 2) divert
runoff water while it is clean, 3) scrape/clean surface as
needed to avoid low spots, ponding, excessive manure
accumulations, and 4) treat filter runoff water from HUA
with filter strips and grass waterways." This advice
appears to be sound and well grounded. Efforts will also
be made to restrict effluent applications on dairy loafing
areas in the revised Arkansas P Index. It also appears that
annual loading rates of N and P need to be considered at
the farm level when NMPs are being developed. If such
calculations had been performed in the preparation of
this farm's NMP, it should have been obvious that the
31 ha was insufficient land area for application of the
amount of waste created to avoid the buildup of nutrients
in the soil.
The best metric to be indicative of nitrate leaching was
the forage nitrate concentration. Since growers do not
want to feed cattle forages containing excessive nitrate
levels, they would be willing to pay for forage nitrate
analyses. Hence, we recommend that forage nitrate be
routinely measured on spray fields receiving effluent
to determine if excessive nitrate leaching is occurring.
This would also serve a dual purpose of protecting
cattle from poor quality vegetation. Forage sampling
for nitrate analyses just prior to harvesting would
provide indicators of utility of the hay for dairy cows
and if future effluent application protocol needed to be
adjusted.
-------
5.0
References
Adams, P. L., T. C. Daniel, D. R. Edwards, D. J. Nichols,
D. H. Pote, and H. D. Scott. Poultry litter and
manure contributions to nitrate leaching through
the vadose zone. Soil Science Society of America
Journal 58: 1206-1211 (1994).
American Public Health Association. Standard methods
for the examination of water andwastewater; 18th
ed.; APHA: Washington, DC, 1992.
Arkansas Cooperative Extension Service. Rural well
water testing program in Arkansas; University of
Arkansas: Little Rock, AR, 1990.
Babiker, I. S., M. A. A. Mohamed, H. Terao, K. Kato,
and K. Ohta. Assessment of groundwater
contamination by nitrate leaching from intensive
vegetable cultivation using geographical
information system. Environment International
29: 1009-1017 (2004).
Brauer, D., G. E. Aikens, S. J. Livingston, L. D. Norton,
T. R. Way, and J. H. Edwards. Amendment effects
on soil test phosphorus. Journal of Environmental
Quality 34: 1682-1686 (2005).
Carreker, J. R., S. R. Wilkinson, A. P. Barnett, and
J. E. Box. Soil and water management systems
for sloping land; Agricultural Research Service/
USD A, Publication #ARS-S-160. U.S. Government
Printing Office: Washington, DC, 1977.
Daniels, M. B., K. VanDevender, and T. Riley. Nutrient
management planning for livestock operations:
An overview. University of Arkansas/Division of
Agriculture/Cooperative Extension Fact sheet 9515,
2008. Available at: http://www.uaex.edu/Other
Areas/publications/PDF/FS A-9515 .pdf.
DeLaune, P. B., P. A. Moore, Jr., D. K. Carman,
A. N. Sharpley, B. E. Haggard, and T. C. Daniel.
Development of a phosphorus index for pastures
fertilized with poultry litter — Factors affecting
phosphorus runoff. Journal of Environmental
Quality?,?,: 2183-2191 (2004a).
DeLaune, P. B., P. A. Moore, Jr., D. K. Carman,
A. N. Sharpley, B. E. Haggard, and T. C. Daniel.
Evaluation of the phosphorus source component
in the phosphorus index for pastures. Journal of
Environmental Quality 33: 2192-2200 (2004b).
Franzluebbers, A. J., and J. A. Stuedemann.
Bermudagrass Management in the Southern
Piedmont USA: VII. Soil-Profile Organic Carbon
and Total Nitrogen. Soil Science Society of America
Journal 69: 1455-1462(2005).
Gardenas, A. I., J. W Hopmans, B. R. Hanson, and
J. Simunek. Two-dimensional modeling of nitrate
leaching for various fertigation scenarios under
micro-irrigation. Agricultural Water Management
74: 219-242 (2005).
Gilliam, J. W. Fertilizer nitrates not causing problems in
North Carolina ground water. Better Crops, Spring
Issue, 1991, pp. 6-8.
Guillard, K. and K. L. Kopp. Nitrogen fertilizer form
and associated nitrate leaching from cool-season
lawn turf. Journal of Environmental Quality
33: 1822-1827(2004).
Hojjati, S. M., T. H. Taylor, and W. C. Templeton.
Nitrate accumulation in rye, tall fescue and
bermudagrass as affected by nitrogen fertilization.
Agronomy Journal 64: 624-627 (1972).
MacKown, C. T. and J. C. Weik. Comparison of
laboratory and quick-test methods for forage nitrate.
Crop Science 44: 218-226 (2004).
Marriott, L. F. and H. D. Bartlett. 1975. Animal waste
contribution to nitrate nitrogen in soil. pp. 296-298.
In "Managing Livestock Wastes". Proc. Int. Symp.
Livestock Wastes 3rd Symposium, ASAE, St.
Joseph, MI.
Mehlich, A. Mehlich 3 soil test extractant:
A modification of Mehlich 2 extractant.
Communications in Soil Science and Plant Analysis
15:1409-1416(1984).
Moore, P. A., Jr., T. C. Daniel, and D. R. Edwards.
Reducing phosphorus runoff and inhibiting
ammonia loss from poultry litter with aluminum
sulfate. Journal of'Environmental Quality 29: 37-49
(2000).
Nolan, B. T, B. C. Ruddy, K. J. Hitt and D. R. Helsel.
A national look at nitrate contamination of
groundwater. Conditioning Purification 39: 76-79
(1998).
NRCS. 2000. Established Series: Jay Series. Available
at: http://www2.ftw.nrcs.usda.gOv/osd/dat/J/JAY.
html. Verified February 12, 2009.
Owens, L. B. and M. J. Shipitalo. Surface and
subsurface phosphorus losses from fertilized pasture
systems in Ohio. Journal of Environmental Quality
35: 1101-1109(2006).
-------
Ritter, W. F. and A. E. M. Chirnside. Impact of land
use on ground-water quality in southern Delaware.
Ground Water 22: 38-47 (1984).
SAS Institute. 1985. SAS user's guide: Statistics. Ver. 5
ed. SAS Inst, Gary, NC.
Self-Davis, M. L., P. A. Moore, Jr., and B. C. Joern,
2000. "Determination of water and/or dilute salt
extractable phosphorus in soil." In Methods
of Phosphorus Analysis for Soils, Sediments,
Residuals, and Waters. Southern Cooperative Series
Bulletin No. 396, 2000, pp. 24-26.
Shigaki, F., A. Sharpley, and L. I. Prochnow. Source-
related transport of phosphorus in surface runoff.
Journal of Environmental Quality 25: 2229-2235
(2006).
Soupir, M. L., S. Mostaghimi, and E. R. Yagow.
Nutrient transport from livestock manure applied to
pastureland using phosphorus-based management
strategies. Journal of Environmental Quality 35:
1269-1278 (2006).
Spalding, R. F. and M. E. Exner. Occurrence of nitrate in
ground water - a review. Journal of Environmental
Quality 22: 392-402 (1993).
Steele, K. F. and W. K. McCalister. Potential nitrate
pollution of ground water in limestone terrain
by poultry litter, Ozark region, USA. In Nitrate
contamination: Exposure, consequence, and control;
Bogardi, L, Kuzelka, R. D., Eds; Proceedings of the
NATO Advanced Research Workshop on Nitrate
Exposure; NATO ASI series, Series G: Ecological
Sciences, vol. 30. Springer-Verlag: Berlin,
Germany, 1991; pp 209-218.
Strickland, G., G. Selk, H. Zhang, and D. L. Step.
Nitrate toxicity in livestock; Oklahoma Cooperative
Extension Service Bulletin PSS-2903, 2008.
Available at: http://pss.okstate.edu/publications/
cornsorghum/PSS-2903.pdf.
Toth, J. D., D. Zhengxia, J. D. Ferguson, D. T. Galligan,
and C. F. Ramberg, Jr. Nitrogen- vs. phosphorus-
based dairy manure applications to field crops:
nitrate and phosphorus leaching and soil phosphorus
accumulation. Journal of Environmental Quality
35:2302-2312(2006).
USD A. 5*0/7 survey for Washington county, Arkansas.
USDA/SCS, Washington, DC, 1969.
USD A. 1992. Agricultural Waste Management Field
Handbook. Chapter 4 - Agricultural waste
characteristics. Available at: http://www.wsi.
nrcs.usda.gov/products/W2O/AWM/docs/handbk/
awmfh-chap4 .pdf.
USEPA. Methods for chemical analysis of water and
wastes, EPA/600/4-79-020. U.S. Environmental
Protection Agency, Washington, DC, 1979.
USEPA. Drinking water regulation and health
advisories. EPA, Office of Water, Washington, DC,
1995.
USGS. 2009. National Water Information System:
Web Interface; USGS 07048600 White River near
Fayetteville, AR. Available at: http://waterdata.usgs.
gov/nwis/nwisman/?site no=07048600&agency
cd=USGS. Verified February 13, 2009.
VanDevender, K. Manure Management: Concepts and
Environmental Concerns. University of Arkansas/
Division of Agriculture/Cooperative Extension Fact
Sheet 1037, 1997. Available at: http://www.uaex.
edu/OtherAreas/publications/HTML/FSA-1037.asp.
vanEs, H. M., J. M. Sogbedji and R. R. Schindelbeck.
Effect of manure application timing, crop, and soil
type on nitrate leaching. Journal of Environmental
Quality 35: 670-679 (2006).
Wu, Z. and J. M. Powell. Dairy manure type, application
rate, and frequency impact plants and soils. Soil
Science Society of America Journal 71: 1306-1313
(2007).
Zhu, Y, R. H. Fox, and J. D. Toth. Leachate collection
efficiency of zero-tension pan and passive capillary
fiberglass wick lysimeters. Soil Science Society of
America Journal 66: 37-43 (2002).
Zotarelli, L., J. M. Scholberg, M. D. Dukes and
R. Munoz-Carpena. 2007. Monitoring of nitrate
leaching in sandy soils: comparison of three
methods. Journal of Environmental Quality
36: 953-962 (2007).
-------
-------
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGES FEES PAID
EPA
PERMIT NO. G-35
Office of Research and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
Recycled/Recyclable
Printed with vegetable-based ink on
paper that contains a minimum of
50% post-consumer fiber content
processed chlorine free
------- |