United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
Ada OK 74820
Research and Development
EPA-600/S2-83-01 7 June 1983
Project Summary
Rangeland Watershed Water
Budget and Grazing Cattle Waste
Nutrient Cycling
Jeff Powell, Frank R. Crow, and Donald G. Wagner
This research project was designed
to determine baseline data concerning
the source, movement concentration
and factors affecting nonpoint pollu-
tants in runoff from a representative
tallgrass prairie watershed grazed by
cattle in North Central Oklahoma A
60-hectare (ha), tallgrass prairie water-
shed moderately grazed yearlong by
beef cows and calves was instrumented
to determine precipitation and runoff
amounts and concentrations of sedi-
ment, nitrogen, phosphorus, potas-
sium, BOD, COD, and TOC. Soil types,
range sites and topography were sur-
veyed, mapped and described. Soil
water content, live and standing dead
vegetation biomass, dung pat density,
ground cover and herbage utilization
were determined at 25 locations on
the watershed periodically from April
1976 through October 1978. Concen-
trations of N, P, K, Ca, and structural
carbohydrates were determined in live
and standing dead vegetation and dung
collected periodically from different
locations on the watershed. Stocking
density and grazing pressure were cal-
culated. Independent site factors were
used in regression equations to predict
plant species abundance, live and
standing dead vegetation biomass,
utilization and dung pat density and
biomass. Changes in dung chemical
composition overtime were determined
for dung deposited during different
seasons and in place for different peri-
ods of time.
The amount of nonpoint source pol-
lution contributed to receiving waters
by runoff from the watershed was
comparable to that from tallgrass prairie
watersheds in other parts of the United
States and was minimal when com-
pared to other nonpoint sources of
pollution such as cropland and fer-
tilized pastures. Significant runoff oc-
curred in every season, but spring was
the season with the greatest runoff
and pollution potential because pre-
cipitation and soil water content were
greatest and ground cover was lowest
at this time. Sediment was the most
significant pollutant and was contri-
buted primarily from the sides of drain-
ageway channels. Direct overland move-
ment of dung into stream channels
was minimal because standing vegeta-
tion and ground litter on the lower
slope positions acted as a filter.
Plant transpiration rapidly depleted
soil water content between mid-May
and mid-July, especially in the upper
35-cm of the soil profile. This draw-
down reduced the probability of runoff
from all precipitation events except
intense thunderstorms during the sum-
mer and fall. Tallgrasses, standing
vegetation and ground cover were
much greater on loamy prairie range
sites on lower slope positions, where-
as dung density and biomass were
greater on the upland, more xeric shal-
low prairie sites. Areas selected by
cattle for bedgrounds and resting areas
had a much greater influence on dung
distribution than did areas selected for
grazing. Nutrient concentrations in
dung decreased to a steady state level
in less than six months regardless of
the season of deposition.
This Project Summary was developed
by EPA's Robert S. Kerr Environmental
Research Laboratory, Ada, OK. to an-
-------
nounce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
As the demands on rangelands for meat,
water and other products and uses be-
come more intense, the need to under-
stand the fundamental relationships in
rangeland ecosystems becomes more crit-
ical. Grazing by herbivores and water
production have historically coexisted on
rangelands. They will continue to coexist
as long as we choose to utilize meat from
grazing animals and water from runoff and
as long as rangeland managers have and
use the knowledge necessary to make
grazing and water production compatible.
Relatively little research has been or is
being conducted on potential water pollu-
tion from rangeland watersheds grazed by
livestock. The need for research on water
budgets and nutrient inventories and cyc-
ling on rangeland is critical and immediate.
Rangelands and forest-ranges could pro-
vide either a large area for animal waste
disposal or be a potentially significant
source of water pollution. The realization
of a beneficial and safe land use instead of
a detrimental consequence of natural
herbivore grazing depends largely on the
capacity of different rangelands to receive
and hold animal wastes without exceeding
the acceptable level of pollutants in water
runoff.
Because of the greater demands on
rangeland to produce red meat while
maintaining production of water of accept-
able quality (P.L 92-500) and the limited
amount of information concerning poten-
tial water pollution from rangeland water-
sheds grazed by cattle, this study was
designed to determine on a North Central
Oklahoma rangeland watershed grazed by
beef cattle 1) the source, transfer and
transformation of potential pollutants, 2)
the hydrologic and meteorologic param-
eters necessary to establish the water
budget and movement of selected nutri-
ents, 3) effects of environmental condi-
tions on the rate of degradation of grazing
cattle dung, and 4) effects of cattle waste
concentration, chemical composition and
distribution on levels of nutrients in soils.
Conclusions
The amount of nonpoint source pollu-
tion contributed to receiving waters by
runoff from this watershed, representative
of tallgrass prairie watersheds grazed by
cattle in North Central Oklahoma, is com-
parable to that from tallgrass prairie water-
sheds in other parts of the United States
and is minimal when compared to other
nonpoint sources of pollution, such as
cropland and fertilized pastures. Signifi-
cant runoff can occur in any season, but
spring has the greatest runoff and pollu-
tion potential when precipitation is great-
est and ground cover is at the lowest level.
Sediment is the most significant pollu-
tant and is contributed primarily by the
sides of drainageway channels. The po-
tential danger of pollution from chemical
leaching from watershed vegetation is
very low compared to the potential pollu-
tion of sediment and dung movement into
drainageways. Direct overland movement
of dung into stream channels is minimal
because standing vegetation and ground
litter on the deeper soils of the lower slope
positions adjacent to stream channels act
as retardants and filters. Utilization of
standing vegetation is also relatively light
in areas adjacent to stream channels.
Soil water drawdown and recharge cycle
were much more consistent all three years
of the study than were precipitation a-
mounts and distribution. Soil water con-
tent decreases from maximum to mini-
mum within about 60 days (i.e., mid-May
to mid-July) primarily due to transpiration
of actively growing vegetation. Soil water
content in the lower part of the soil profile
is much more consistent than that in the
upper part of the soil profile. Soil water
content is greater in ungrazed areas where
standing vegetation and plant ground litter
are greater than in grazed areas. Soil water
use rates are closely related to soil water
availability and the amount of growing
vegetation. As vegetation matures in late
summer, soil water recharge is greater
than evapotranspiration use.
Peak vegetation production varied each
year and occurred earliest (June) during
the driest year. Accumulative production
of the different plant species classes
amounted to 25 to 30% more production
than peak standing crop. Therefore, ac-
cumulative production of rangeland vege-
tation should be used as a measure of total
production rather than peak standing crop.
Productivity of grazed vegetation is about
as high as that of ungrazed vegetation.
Abundance of different plant species is
closely related to slope position, aspect,
soil types, range sites, A horizon and total
water content and grazing pressure. From
80 to 90% of the variation in live and
standing dead vegetation at different
watershed locations can be accounted for
with regression equations and indepen-
dent site factors. Large differences in
plant species composition and biomass,
ground cover, herbage utilization and
dung distribution can be expected on the
basis of range sites.
The average effective duration of dung
biomass is about two years since the aver-
age dung biomass on the watershed is
about twice that deposited annually by
grazing cattle. Chemical concentrations of
N, P, K, and Cain fresh dung decrease with
time and appear to reach a steady state in
about six to eight months. Dung decom-
position is apparently affected more by
fragmentation and decomposer activity
than by leaching or chemical transforma-
tion. Decomposition appears to be slower
on the more xeric, shallow prairie range
sites than on loamy prairie sites. Cattle site
preference for bedgrounds and resting
areas has a greater influence on dung
distribution and potential pollution sources
than does site preference for grazing. The
effect of urine on potential pollution is
insignificant because of rapid movement
into the soil or volatilization of N-com-
pounds into the air and the absence of
perennial streams on the watershed.
Study Area
The study area watershed is located at
latitude 33°N, longitude 97°W about 16
km northwest of Stillwater, Oklahoma.
The watershed has been grazed with cows
and calves for many years prior to the
study. Grazing continued during the study
as before. The area is generally not grazed
during the last two weeks of April and
during the 75 days between August 1 and
October 15. The average grazing use for
the watershed during the study period
was 70 to 80 animal-unit-days/ha.
The watershed is part of a prairie-wood-
land complex on undulating terrain. The
elevation varies from 290 m at the main
runoff-measuring weir to 318 m at the
upper end. The weighted average slope
for the entire watershed is 5.7%. Topog-
raphy and drainage patterns are shown in
Figure 1. In addition to the contours, the
map also shows the location of all per-
manent reference points for soil water and
vegetation sampling, fenced exclosures,
locations of rain gages, runoff measuring
stations, and access roads.
The watershed is composed of two
principal drainageways. Drainage density
is 8.9 km/km2. The south drainageway is
715 m in length, has a channel gradient of
2.0% and drains 51% of the watershed.
The north branch is 493 m in length, has a
slightly flatter gradient, 1.5%, and drains
42% of the watershed. The remaining 7%
of the watershed drains directly into the
-------
Watershed Boundary
pond Watershed Boundary
—•»— Contours
Intermittent Waterway
-=- = Access Road
—«—»- Fence
• Recording Rain Gage
T Runoff Gaging Station
A Transit Station
a So/7 Water Station
Environmental Research Watershed
Agricultural Experiment Station
Oklahoma State University
NE 1/4 Sec. 31 and NW 1/4 Sec. 32
T20N. /?/£, Noble County, Okla.
Figure 1. Topographic map of experimental research watershed in North Centra/ Oklahoma.
-------
main drainageway, which has a 1.0%
slope.
Total drainage area of the watershed is
57.7 ha Land use includes 53.8 ha
rangeland, 3.6 ha cultivated land planted
each year to winter wheat, and a 0.3 ha
stock water pond located at the upper end
of the north drainageway. The pond has a
drainage area of 6.9 ha, approximately
13% of the total watershed area. In
general soils of very-fine or fine-loamy,
mixed thermic Vertic Haplustalfs occupy
70% of the watershed. The proportion of
soil orders is 78% Alfisols, 16% Mollisols
and 6% Inceptisols. Loamy soils devel-
oped under tallgrass prairie vegetation are
most common. Soil fertility in the A
horizon is high. A 62 horizon is generally
present Soil water-holding capacity is
good except on the limited area of coarse-
textured soils.
Shrubs and trees are common along
drainageways and on shallow, coarse-tex-
tured soils overlying fractured sandstone.
Range sites include Loamy Prairie, Claypan
Prairie, Shallow Prairie, Shallow Savannah
and Sandy Savannah.
Results
Precipitation and runoff data per 10-day
period from April 1976 through December
1978 are shown in Figure 2. The limited
number of runoff events is readily ap-
parent None occurred during the first 13
months of the study. The first runoff event
produced the greatest amount of runoff. In
that 10-day period about 25% of the
160+ mm precipitation moved off the
watershed. Although runoff events were
infrequent, runoff occurred at least once
during all seasons.
Table 1 is a summary of the number and
magnitude of rainfall events and related
runoff events for each calendar year of the
study. There was a total of 274 rainfall
events, of which 42 resulted in runoff. In
1976 only one of 20 rainfall events re-
sulted in runoff. At that, the runoff was
almost insignificant In 1977 and 1978
there was an average of one runoff event
for every nine rainfall events. However,
the total observed runoff was five times
greater in 1977 than in 1978. During
1979 when rainfall was higher than nor-
mal there was a runoff event for each 3.3
rainfall events.
Table 2 shows sediment loss rates for
the 12 runoff events in terms of water-
shed area and amount of runoff. As
expected, the larger runoff events pro-
duced the greatest sediment losses per
hectare. However, when expressed in
terms of kilograms per hectare per milli-
160
140
720
•o 700
•§ so
S
I 60
js- Runoff
r__n
JL-ll
F M A M J J A
7977
F M A M J J A S 0 N D
7975 - '
Figure 2. Precipitation and runoff for 10-day periods during study period.
Table 1. Summary of Rainfall and Runoff Events (April 1976 - December 1979)
Environmental Research Watershed, Oklahoma State University
Precipitation
Year
1976*
1977
1978
1979
No. of
Events
40
77
84
73
Largest
Event
(mm)
39.5
101.6
51.3
75.0
Observed
Total
(mm)
420.4
722.2
676.6
886.0
Percent of
Long-Term
Average
59
87
82
107
No. of
Events
2
9
9
22
Runoff
Largest
Event
(mm)
0.97
44.27
5.82
23.60
Observed
Total
(mm)
1.96
62.18
12.44
101.73
* Period includes April - December only.
Table 2. Summary of Sediment Loss by Runoff Environmental Research Watershed,
Oklahoma State University
Runoff
Date
20 May 77
23 May 77
27 May 77
1 July 77
14 Nov 78
18 March 79
22 March 79
10 April 7 '9
2 May 79
9 June 79
5 July 79
17 July 79
Rainfall
(mm)
101.55
38.61
22.35
45.72
47.75
20.32
55.37
26.92
99.31
75.95
52.07
82.55
Runoff
(mm)
44.27
13.56
1.85
1.72
0.94
0.56
16.43
1.85
32.03
8.76
2.36
23.47
Sediment
Loss
(kg/ha)
687.6
193.7
6.9
32.1
6.8
11.8
411.3
34.7
294.7
126.0
25.9
737.6
Sed/Runoff
Ratio
(kg/ha/mm)
15.53
14.28
3.75
18.68
7.30
21.21
25.03
18.70
10.55
14.39
10.98
31.43
TOTAL 668.47 147.80 2569.1
Weighted Average Sediment/Runoff Ratio = 17.38 kg/ha/mm
-------
meter (kg/ha/mm) of runoff, the sedi-
ment loss rates were quite similar for most
of the events. Very little of the sediment
loss from the environmental watershed
was caused by raindrop impact on bare
areas. Inspection of the watershed during
rainfall leads us to believe bank erosion
was largely responsible for the sediment
load that was collected at the lower end of
the watershed.
Total soil water in the soil profile during
the 32-month sampling period varied
from a low of about 90 mm in August of
each year to a high of 200 mm in late May
of 1977 and 1978. The soil water storage
capacity of the watershed soil was about
or slightly greater than 110 mm. The soil
water depletion and recharge patterns
were relatively consistent each of the three
years. Soil water depletion was very rapid
in late spring during the period of most
rapid plant growth. Most of the soil water
depletion was due to transpiration rather
han evaporation. Recharge occurred after
he low point in August regardless of the
amount of precipitation received. In gen-
eral maximum depletion occurred in about
3 three-month period. The average soil
i/vater contents at all depths in the soil
arofile were consistently greater where
grazing was excluded than where grazing
/vas allowed. Although soil characteristics
e.g., texture) for the paired soil water
sampling locations differed slightly, the
differences in soil water content appear to
ie due primarily to vegetation cover. The
average standing, aboveground vegetation
kg DM/ha) for ungrazed and grazed areas
/vere 3160 and 1 730, respectively.
The rate of soil water use or loss from
he soil due to evapotranspiration is shown
n Figure 3. Soil water use rates were
calculated as (precipitation - runoff -
difference in total soil water in succeeding
sampling dates) divided by the number of
days between sampling dates. The rate of
ioil water use was related to soil water
availability and vegetation growth. In
general the soil water use rates shown in
rigure 3 appear to be more realistic mea-
sures of evapotranspiration than pan evap-
aration or any other parameter that does
not consider temperature, available soil
water or water use by the actively growing
Dlants present at a particular time in the
/ear.
Ground litter values were most variable,
x>th within a single sampling date and
jetween sampling dates. Although runoff
hrough the weir was not frequent, intense
^instorms of short duration often moved
jcently deposited ground litter on steep
slopes and ridgetops down to lower slopes
•s
o
I
I
AMJJASONDJFMAMJJASONDJFMAMJJASOND
' 7376 " 7977 " 1978 '
Figure 3. Soil water use rate (mm/day) in relation to precipitation and insoak (mm).
where it was accumulated. Greater a-
mounts of standing vegetation on the
lower slopes acted as a barrier to overland
flow of runoff water and ground litter.
Ground litter was generally least in early
spring after winter decomposition.
Standing dead biomass was more con-
sistent than ground litter both within a
single sampling date and between sam-
pling dates. Standing dead biomass was
least in early spring and greatest in early
winter. Standing dead biomass was less
in 1977 and 1978 because of reduced
live plant biomass in 1976 and 1977,
respectively. Live plant biomass was less
in 1976 and 1977 than in 1978. Peak
production was 1420 kg/ha in mid-June
1976, 1440 kg/ha in early June 1977
and 1630 kg/ha in mid-August 1978.
Summer rains in 1977 caused a second
peak production in early September. Sum-
mer rains in 1978 maintained peak pro-
duction from late June through mid-
August In this study, the accumulative or
sum-of-the-peaks production was 24%,
14%, and 1 7% greater than the greatest
single peak production estimate in 1976,
1977, and 1978, respectively.
The watershed occupies portions of two
pastures. About 80% of the watershed is
in one pasture and 20% is in another
pasture. Stocking density, grazing pres-
sure and herbage utilization were calcu-
lated for each portion of the watershed in
separate pastures and then adjusted to
show average stocking density (animal
units/hectare [a.u./ha]), grazing pressure
(animal units/1000 kg of standing herb-
age) and utilization (kg herbage/hectar;
%) for the watershed. The stocking densi-
ty on the watershed for the study period,
ranged from a low of 0.14 a u./ha to a high
of 0.38 a.u./ha when the watershed was
grazed. The yearlong average stocking
rate was 0.20 a.u./ha in 1976, 0.34
a.u./ha in 1977 and 0.28 a.u./ha in
1978. In general grazing pressure was
less variable than stocking density. How-
ever, the potentially high pollution period
in late winter is shown more clearly.
During this period standing herbage bio-
mass is relatively low because of winter
grazing and pressure from snow and
freezing rain.
Utilization by cattle of the live vegeta-
tion, by species class, and of the standing
dead vegetation was used to determine
vegetation removed or trampled and vari-
ous factors related to forage value, diet
composition, selectivity and temporary
changes in herbage composition due to
grazing. Utilization values were deter-
mined by averaging caged and grazing
residue species class biomass values sep-
arately for all samples. All other utilization-
related factors were determined from
these averages.
The average, total live and standing
dead herbage utilized between April 12,
1976 and October 9, 1978 was 950
kg/ha or about 32% of the herbage
available. The exclosure study showed the
average total live vegetation in the ex-
closures was about 42% greater than that
outside the exclosures. Even with rela-
tively heavy winter grazing on the upper
-------
slopes, exclosures had only 50% more
standing litter than adjacent grazed areas.
The proportion of live vegetation utilized
(31.296) was about the same as the pro-
portion of standing dead utilized (32.5%).
Within the live vegetation about 30% of
the grass biomass and 34% of the forb
biomass was utilized. Utilization values
for tall and midgrasses were lower than
those for forb species class and lower than
those for shortgrasses, cool season grasses
(primarily Bromus japonicus) and Schiz-
achyrium scoparium (SCSC, little blue-
stem). The utilization figures are the result
of site selection as well as species selec-
tion. In addition, these values include
utilization data from two winter grazing
periods. Results from small, uniform
watersheds may be misleading if extrap-
olated to large, diverse rangeland water-
sheds.
Based on utilization (%) and available
herbage composition (%), grasses were
twice as important for forage as were
forbs. Of all species classes, SCSC had the
highest forage value index and was about
three times more important as forage than
were tallgrasses. Grasses were estimated
to contribute about 64% of the diet; forbs
contributed the remaining 36% during the
growing season.
Compared with chemical analyses of
tallgrass prairie vegetation in this area by
other scientists, the average nitrogen and
calcium contents in live vegetation was
much higher than usual, especially in
summer and fall. Potassium content was
about average. Phosphorus content was
about average in the spring, lower during
June and July and average or above aver-
age in August and September. The general-
ly higher than reported nutrient content in
our samples was because of a lower
percentage of tallgrasses with correspon-
ding higher percentage mid- and short-
grasses and forbs, lower rainfall and re-
duced production, and recycling of nutri-
ents through grazing animals.
Nitrogen, phosphorus and potassium
contents in live vegetation decreased from
a high in early spring to a low in summer at
a rate closely resembling the decrease in
soil water content. Changes in standing
dead composition were uniformly cyclic
with seasons and generally reflected the
effects of drought and plant maturity on
different species classes. Changes in
chemical composition of dung reflected
the effects of nutrient availability in forage
and supplemental feed.
Chemical yields from all plant materials
(biomass X chemical composition) were
50 kg N/ha, 25 kg Ca/ha, 1 6 kg K/ha and
3 kg P/ha on the watershed. Ground litter,
usually because of its greater biomass,
produced half or more of the total nutrient
yields for N, P, and Ca, but only 5 kg/ha of
the 16 kg K/ha Total yields were relatively
low for all nutrients as compared to those
from highly fertilized pastures and crop-
lands. Total digestible dry matter (DDM)
yield in standing vegetation ranged from a
low of about 200 to 300 kg DDM/ha in
late winter to a high of 1220 kg DDM/ha
in mid-summer of 1978. Live DDM yields
varied relatively more than those in stand-
ing dead vegetation.
The average effective duration of dung
biomass on the watershed is estimated to
be about two years. Based on the weight
of dung collected from 30-m diameter
areas around each of 25 locations on April
1, 1977 dung biomass on the watershed
averaged 900 kg/ha and ranged from 60
kg/ha to 2900 kg/ha. The average daily
dung deposition rate was about 5.0 kg/
day/a.u., and the annual dung deposition
was 460 kg/ha/yr (0.25 a.u./ha X 5.0
kg/day/a.u. X 365 days/yr). Since the
average dung biomass on the watershed
was 900 kg/ha and the annual deposition
rate was 460 kg/ha, the average weight
loss of dung biomass is about 50% per
year.
The locations with the greatest dung
biomass were all along the upper slopes of
the watershed. Five of the six locations
with the greatest dung biomass were on
soils with a sandy loam surface horizon.
Since cattle usually defecate after rising
from their bedground, these areas may be
preferred bedgrounds and resting areas.
Cattle may prefer sandy soils for bed-
grounds in the winter since sandy soils are
drier and warmer than are fine-textured
soils with a higher soil water-holding ca-
pacity. Another factor affecting dung dis-
tribution may be a slower rate of decom-
position on sandy soils. Drier soils with
lower soil fertility, lower soil water content
and less standing vegetation and ground
litter would be expected to have lower
levels of metabolic activity of decomposer
microorganisms.
The simple linear correlation coefficient
between dung biomass per location and
dung pat density per location was +0.93.
The regression coefficient of +0.33 indi-
cates the average dung pat weighed about
0.33 kg (SE=0.03 kg).
Small differences in dung density and
biomass due to range site and plots of
accumulative dung density over time per
location indicated similarities in dung
deposition pre- and post-removal due to
range site, but differences in deposition
distribution patterns within range sites. '
Dung deposition was much greater on
shallow prairie range sites than on loamy
prairie range sites.
Regression equations developed to pre-
dict pre-removal dung density and bio-
mass and post-removal dung density ac-
counted for about 80 to 90% of the
variation. A horizon soil water and sodium
contents and moisture economy index
were inversely related to dung density and
biomass, whereas A horizon potassium
content was directly related to dung density
and biomass in both equations. In general
dung density and biomass were greater on
the warmer and more xeric sites.
Dung deposition measures were not
highly correlated with herbage utilization
measures. The magnitude of the correla-
tion coefficients between dung deposition
and utilization of standing dead were,
however, about twice the magnitude of
correlation coefficients between dung
deposition and utilization of live vegeta-
tion. Therefore, cattle apparently spent
more time grazing in the areas of greatest
dung deposition in the winter than during
the growing season. Standing dead utili-
zation was influenced primarily by abiotic
site factors, whereas live vegetation utili-
zation was influenced to a high degree by
live production and species composition
which were significant only during the
growing season.
The highest dung deposition values were
for those locations used as bedgrounds
and resting areas and where the dung
decomposition rates would be lowest.
These locations coincided with the loca-
tions with a high degree of standing dead
utilization, primarily on the relatively high,
wide and flat ridgetop in the west central
portion of the watershed. Both live and
dead utilization were relatively low on the
lower slope positions along drainageways;
however, live utilization was frequently
much higher than dead utilization in these
areas. Cattle apparently prefer to graze
green tallgrass material during the growing
season, but not dead tallgrass material in
any season.
Changes in N, P, K, and Ca concentra-
tions in dung in place on the watershed for
various periods of time were not consistent
with respect to nutrient or season of
deposition. Averaged over all deposition
seasons, there was only a slight decrease
in N content over time. Essentially, those
factors influencing the N content of dung
at the time of deposition strongly out-
weighed environmental factors influencing
loss of N after deposition. The phosphorus
content in dung also varied due to season
-------
of deposition and period of time subject to
degradation. In general P concentrations
increased during the first 30 to 90 days
after deposition and decreased after about
90days in place. The change in potassium
contents over time was much more con-
sistent than those of N and P contents. K
content decreased rapidly in the first 15 to
30 days in place, then stabilized over the
'emainder of the period. Changes in
:alcium content in dung over time were
iimilar to those of K content. However, the
elative loss of Ca over time was not as great
is that of K. Changes in concentrations of
itructural fiber components (i.e., ADF,
kDL and CEL) were determined over a
140-day degradation period for dung de-
losited in July 1976. ADF and ADL
icreased over time, whereas CEL de-
reased over time.
J. Powell, F. R. Crow, and D. G. Wagner are with Oklahoma Agricultural
Experiment Station, Oklahoma State University, Stillwater, OK 74078,
Lynn R. Shuyler is the EPA Project Officer (see below).
The complete report, entitled "Rangeland Watershed Water Budget and Grazing
Cattle Waste Nutrient Cycling," fOrder No. PB 83-180 844; Cost: $26.50,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Robert S. Kerr Environmental Research Laboratory
U.S. Environmental Protection Agency
Ada, OK 74820
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
PS 0000329
230 S DEARBORN STREET
CHICAGO IL 60604
------- |