United States
Environmental Protection
Agency
Robert S. Kerr Environmental Researe
Laboratory
Ada OK 74820
Research and Development
EPA-600/S2-82-079 Sept. 1982
Project Summary
Nitrates in Groundwater
Resulting from Manure
Applications to Irrigated
Cropland
Wynn R. Walker and Bruce E. Kroeker
Yearling beef manure was applied in
three rates and three frequencies to
57 experimental plots located within a
9.3 ha field near Grand Junction,
Colorado. The basic management
practices investigated were irrigation
and manure loading frequency, irriga-
tion efficiency, and manure loading
rates. Climatological conditions, soil
moisture, evapotranspiration, ground-
water outflows, and the furrow irriga-
tion system performance were meas-
ured in the field plots. The soil was
sampled and tested for nitrate-nitro-
gen (NO3-N), total Kjeldahl-nitrogen
(TKN), ammonium-nitrogen (NH4-N),
total organic carbon (TOC) and the
common salinity constituents. Vacuum
extractors below the soil root zone in
10 plots were used to measure and
sample the water percolating down-
ward. Crop N-uptake was evaluated
by periodic analysis of plant samples.
A soil-plant-water-N simulation model
reported in the literature was used to
evaluate the field data and predict the
conditions in the field plots.
Model simulations were run for a
wide range of typical Grand Junction
conditions to evaluate the effect of
irrigation and manure utilization
practices on NOs-N contamination of
groundwater. The results indicate the
time required to reach a steady-state
condition and the quantitative effects
of irrigation efficiencies, irrigation fre-
quency, manure loading rates, manure
loading frequency, and initial soil or-
ganic matter content.
This Project Summary was devel-
oped by EPA's Robert S. Kerr Environ-
mental Research Laboratory, Ada,
OK, to announce key findings of the
research project that is fully docu-
mented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
In the United States, annual excre-
mental animal manures are estimated
to exceed human wastes by an order of
magnitude. For the feeding operations
that are increasingly using confinement
systems, the large manure volume
creates a sizeable disposal problem.
Land application of animal manure has
been a fertilization practice for thou-
sands of years and today remains the
cheapest and most common manure
disposal method. Most land surfaces
used for applying manures are agri-
cultural croplands, many of which
require irrigation to supplement natural
precipitation.
Animal manure contains a variety of
salt and organic materials that easily
dissolve or suspend in water and there-
by degrade water quality. The spreading
of manure over agricultural lands gives
rise to two broad water quality detri-
ments. First, precipitation and irrigation
overland flows erode surface materials
including the organic and mineralogical
components of the manure and trans-
port them to nearby receiving waters
where the nutrients can lead to eutro-
phication and the salts to toxic impacts
on native biota. The second problem
centers on the infiltration of water
through the soil toward groundwater
basins. Soluble N, C, .and salt constit-
uents are transported downward by the
water and are partially transformed by
-------
the soil and its microbial population. In
both aspects of this environmental
problem, the control of the water move-
ment through improved irrigation and
tillage practices can substantially re-
duce the detriments.
The least understood problem is the
management of irrigation and land dis-
posal systems in conjunction with an
objective of groundwater quality con-
trol. The literature does not describe
many studies in which manure applica-
tion, irrigation, and groundwater sys-
tems are investigated collectively. This
study evaluated the effects of various ir-
rigation practices and manure loading
rates on the movement of pollutants
toward groundwater basins. Thus, the
study was designed to provide a range of
conditions broad enough to verify a sim-
ulation model of the physiochemical-
biological processes involved. The
model can be used to evaluate alterna-
tive management practices as long as
the range of conditions for which it had
proven reliable is not exceeded.
The scope of this study was necessar-
ily limited by time, space andf unding. Of
the variety of animal manures being
applied to irrigated lands, only yearling
beef cattle manures were studied, the
characteristics of which are summa-
rized in Table 1. To lessen groundwater
impacts, the surface runoff component
was minimized by incorporating the
manure into the upper 20 cm of the soil
and controlling the field tailwater runoff
rates. The field was planted to corn and
was watered with conventional furrow
irrigation practices.
Conclusions
A computer simulation model report-
ed in the technical literature was tested
with three years of field data from an ir-
rigated site near Grand Junction, Col-
orado, where variable applications of
yearling beef manure had been made.
With adjustments to the rate coefficients
for the mineralization of the manure,
the model predicted field measure-
ments within 10% to 25% for more than
80% of the total data base. Since individ-
ual components of the model have been
verified elsewhere under other soil,
crop and climatic conditions, it is con-
cluded that it represents a usable and ef-
fective simulation tool.
The mathematical and conceptual
sophistication of the model is within the
expertise of practicing engineers and
soil scientists. Input data are easily
collected as part of routine monitoring
and evaluation studies.
Table 1. Summary of Manure Characteristics
Constituent
Average Composition
(% Dry Weight)
Dry Matter
Volatile Solids
Biochemical Oxygen Demands
TKN
/W/4-/V
Total Inorganic-N
Total Carbon
TOC
Total Phosphorus
Othophosphate
Chemical Oxygen Demand
Calcium
Sodium
Potassium
Chloride
66.68
63.74
2.57
2.24
0.30
0.30
28.20
26.70
0.19
0.03
3.51
2.96
0.67
2.13
1.65
Results of the model testing indicate
that the most important irrigation of the
season in moving NOa-N toward a
groundwater system is the first water
application. In most surface irrigated
conditions, about 50 to 75 percent of
the annual deep percolation comes
from the preplant or emergence irriga-
tion. Subsequent irrigations occur
when the soil surface exhibits a lower
infiltration rate and the crop root zone is
active to a greater depth. For the Grand
Junction soil, improving the first
irrigation from a typical 24-cm applica-
tion to a 12-cm application reduced the
annual N flux below the 200-cm depth
by more than 50 percent.
Manure applications on irrigated
lands increase N losses from the system
by both denitrification and leaching in
direct proportion to the loading rates.
Increasing the net weight application
rate from about 12 metric tons/ha
(MT/ha) to 404 MT/ha increased de-
nitrification losses by 900 kg-N/ha/yr,
and leaching losses to the groundwater
basin by 360 kg-N/ha/yr.
Probably the most interesting conclu-
sion regarding the loading management
is the effect of application frequency on
the N losses due to denitrification and
leaching. Applying manure each year,
every other year, and every third year
over an extended period results in very
small changes (about 10 percent) in the
amount of N leached from the 200 cm
profile. The losses due to denitrification
on the other hand are reduced signifi-
cantly by less frequent loadings. Con-
sequently, as a disposal practice,
manure applications should be made
more frequently in order to maximize
the treatment efficiency of the soil-crop
system.
The salts in the manure applied to
irrigated lands create conditions detri-
mental to efforts aimed at controlling N
contamination of groundwater resources.
Heavy loadings and efficient irrigation
practices will result in very large ac-
cumulations of salts in the soil profile
between 40 cm and 120 cm and there-
fore will seriously affect crop yields. The
same heavy loadings coupled with tradi-
tional irrigation practices still result in
salination of the soil profile below 60 cm
but do not tend to accumulate salts
beyond a level dictated by the annual
leaching fraction.
Recommendations
The effects of animal manure addi-
tions to the upper soil profile on the
decomposition of soil and organic
materials under field conditions need
further attention in the model. The
modification in the mathematical forms
of the model should probably be made
in the mineralization rate coefficients.
The rate coefficients for other N and C
transformations worked well for local
conditions and have also been verified
elsewhere.
Controlling the applied depth of
irrigation water during the first irrigation
should be the most emphasized irrigation
practice for managing the contribution
of N to underlying groundwater basins.
More frequently, irrigation should also
be considered for controlling N leaching.
Annual manure loadings should be
encouraged over less frequent applica-
tions, where disposal rather than
fertilization is the primary objective of
the program. Careful evaluation of the
salts contained in the manures should
be made if crop production is to be
maintained, particularly if high irrigation
efficiencies can be achieved consistently.
-------
Modeling Irrigation and
Manure Loading Practices
Several modifications were made to
the original computer code in oYder to
study the consequences of various
irrigation and manure disposal practices
on the N system. Following the revisions,
a large array of initial conditions,
irrigation practices, and manure loading
strategies were evaluated. In order to
illustrate the results, the effects of
irrigation practices as represented by
irrigation efficiency and the effects of
manure loading practices as represented
by loading rates will be summarized.
Effects of Irrigation Efficiency
Using the normal site conditions,
leaching and denitrification losses
resulting from the range of annual
manure loadings were simulated over
the period of years required to reach
"equilibrium" at a 2-m depth in the soil.
These tests were made for irrigation
application efficiency typical of furrow
irrigation systems as well as an "im-
proved" condition in which the applied
depth during the first irrigation was re-
duced by one-half. The results are illus-
trated in Figure 1 and include results for
the case in which the natural organic
matter content in the soil profile was de-
fined at 20 percent of the normal condi-
tion.
For the normal condition, leaching
rates shown in Figure 1 increase
approximately linearly from about 400
kg-N/ha/yr to over 700 kg-N/ha/yr. By
improving only the first irrigation, these
rates are decreased by about 50
percent.
There are indications in the literature
that maintaining soil moisture at
various levels could have significant
effects on denitrification. The effect,
therefore, would reduce the leaching
losses of NO3-N and NH4-N. In the field
investigations, irrigations were initiated
when the available soil moisture had
been depleted by 60, 30, and 15
percent, respectively. To test the effects
of average soil moisture levels, the
normal site condition receiving spring
manure loadings over extended periods
were simulated at these three values.
The results indicated that changing the
frequency reduced NOa-N leaching
substantially, but the effect is primarily
one of irrigation efficiency rather than
frequency. However, the frequency of
applying irrigation water was important
if viewed from its actual effect on irriga-
tion efficiency. The wetter soil condi-
800
700
600
oj 500
co
CD
-------
than the denitrification losses. It is
important to note that when the total
nitrogen leaching was evaluated, the
differences between the three treatment
levels were less than 10 percent.
s:
s*.
O
8/.5 oo-o
o
1.0
------- |