Control of
WATER POLLUTION
from cropland
mam^^mmmmmmam
Volumel— An overview
^^ Office of Research and Development
^ALfr-. Environmental Protection Agency
-------
REVIEW NOTICE
This report has been reviewed by the Office of Research and Development and the
Agricultural Research Service and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
DOCUMENT AVAI LABI LITY
While supply lasts, single copies may be requested from:
(1) U.S. Environmental Protection Agency
Agriculture and Non-Point Source
Management Division (RD-682)
Washington, D.C. 20460
To order please cite
REPORT NO. EPA-600/2-75-026b
(2) U.S. Department of Agriculture
ARS Information — Room 343A
Federal Center Building - No. 1
Hyattsville, Maryland 20782
To order please cite
REPORT NO. ARS-H-5-2
The public may also purchase this document from the National Technical Information
Service, 5285 Port Royal Road, Springfield, Virginia 22151 and from the Superintendent
of Documents, U.S. Government Printing Office, Washington, D.C. 20402
For sale by the Superintendent of Documents, U.S. Government Printing Office. Washington. D.C. 20402
-------
Control of
WATER POLLUTION
from cropland
Volume I —An overview
Authored by a Committee of Scientists of Agricultural Research Service, USDA.
B. A. Stewart, Bushland, Texas Coordinator
D. A. Woolhiser, Fort Collins, Colorado Hydrology
W. H. Wischmeier, Lafayette, Indiana Erosion
J. H. Caro, Beltsville, Maryland Pesticides
M. H. Frere, Chickasha, Oklahoma Nutrients
Appendix C prepared by K. F. Alt, Economic Research Service, USDA.
Prepared under an Interagency Agreement with the Office of Research and Development,
EPA. Project Officers were L. A. Mulkey, ORD, EPA, and C. W. Carlson, ARS, USDA.
JUNE 1976
Agricultural Research Service / f* \ Office of Research and Development
U.S. Department of Agriculture \53ZZ/ Environmental Protection Agency
-------
CONTENTS
Chapter
I. INTRODUCTION
B. A. Stewart and D. A. Woolhiser
2. HYDROLOGIC ASPECTS OF NONPOINT POLLUTION
D. A. Woolhiser [[[ 7
Fundamentals of Hydrology ............................................ 7
Components of the Hydrologic Cycle ...................................... 10
Agricultural Chemical and Sediment Transport Models ........................... 17
Agricultural Practices to Control Direct Runoff ............................... 18
Research Needs [[[ 23
Literature Cited [[[ 24
3. CROPLAND EROSION AND SEDIMENTATION
W. a Wischmeier [[[ 31
Sediment Sources and Quantities ......................................... 31
Cropland Erosion [[[ 33
Erosion Factors [[[ 36
Erosion Control Methods .............................................. 41
Sediment Delivery Ratios .............................................. 47
Tolerance Limits [[[ 48
Research Needs [[[ 50
Literature Cited [[[ 53
4. NUTRIENT ASPECTS OF POLLUTION FROM CROPLAND
M. a Frere [[[ 59
The Problems [[[ 59
Sources of Nutrients ................................................. 61
Transport from Cropland .............................................. 68
Effect of Control Practices ............................................. 73
Research Needs ......... .......................................... 81
Literature Cited [[[ 82
5. PESTICIDES IN AGRICULTURAL RUNOFF
J. a Caro [[[ 91
Extent and Trends in Use of Agricultural Pesticides ............................. 92
-------
6. INTERDISCIPLINARY RESEARCH NEEDS
B. A. Stewart 121
Appendix
A. SIMULATION OF DAILY POTENTIAL DIRECT RUNOFF 123
B. SIMULATION OF POTENTIAL PERCOLATION AND NITRATE LEACHING 149
C. ECONOMIC ANALYSIS METHODOLOGY 177
IV
-------
Control of
WATER POLLUTION
from cropland
Volume Il--An overview
CHAPTER 1
INTRODUCTION
B. A. Stewart and D. A. Woolhiser
Agricultural technology is one of the real strengths of
the United States. Although the population has in-
creased steadily, food and fiber production has met the
domestic needs and has also provided substantial
amounts for export, which is so important to the U.S.
trade balance. Fertilizers and pesticides have played a
major role in this accomplishment because the acreage of
cropland has changed little in the last 45 years—agricul-
tural chemicals and other technological inputs have been
substituted for land.
The marvels of agricultural technology have not gone
unchallenged. Much of the blame for polluted streams
and lakes is often placed on agricultural activities. Some
groups and individuals have even called for a total ban
on the use of agricultural chemicals. At the other
extreme, there are those who claim that the use of
chemicals has not had any adverse effect on the
environment and that there should be no restrictions on
or control of their use.
The ultimate decision as to whether agriculture is
contributing to pollution of particular water bodies to
such an extent that active control measures are required
rests with State or local authorities. To assist these
officials in reaching this decision and in choosing
appropriate controls, the Federal Water Pollution Con-
trol Act Amendments of 1972, Public Law No. 92-500,
specify that the Administrator of the Environmental
Protection Agency shall, in cooperation with other
agencies, provide guidelines for identifying and evaluat-
ing the nature and extent of nonpoint sources of
pollutants. This two-volume document on control of
potential water pollutants from cropland was written by
scientists of the U.S. Department of Agriculture in
response to this provision of the Act and at the request
of the Environmental Protection Agency. Volume 1 is a
User's Manual for guideline development. Here in Vol-
ume II we will review some of the basic principles on
which control of specific pollutants is founded, provide
supplementary information, and present some of the
documentation used in Volume I.
Management decisions relating to the control of
pollution from cropland involve a careful weighing of
potential costs and benefits. Some of the factors
affecting these decisions can be visualized by considering
the schematic drawing of an agricultural system in
Figure 1. The system itself is arbitrary and could consist
of a field, a state, or a river basin. Inputs to and outputs
from the system can be identified and inputs can be
classified as controlled or uncontrolled. Precipitation
and solar radiation are uncontrolled and contribute to
the stochastic nature of the outputs. The farmer has the
ultimate control over the controllable inputs, subject to
physical and legal constraints. The outputs can be
changed by varying the inputs or the system itself within
certain constraints imposed by physical laws.
Most people would agree that the system should be so
modified and that the inputs to the system should be
controlled at a level that maximizes the net benefits to
1
-------
society attributable to the system. Obviously the modifi-
cations and controls chosen depend strongly on the
concept of social welfare and must include many costs
and benefits not normally accounted for by a land
manager.
The costs and benefits of the agricultural use of land
depend on the weather and other elements that may be
considered as stochastic processes; therefore, the costs
and benefits associated with the agricultural use of the
land may themselves be considered as stochastic
processes. Consider the four sample functions shown in
Figure 2. Xi(t) represents the amount of daily precipi-
tation; Ci (t) represents the amount of a chemical applied
to a field and is a stochastic process because the time of
application depends on precipitation, stage of crop
growth and other factors associated with the particular
chemical; Yt(t) symbolizes daily surface runoff which
may transport the chemical to a stream or lake; Y2(t)
represents the amount of chemical transported to
surface water; and B(t) represents the benefit process
(costs are negative benefits). Social costs include those
incurred when surface runoff occurs shortly after a
chemical is applied and those due to sediment. Other
costs include those normally borne by the farmer. The
benefit from the sale of the crop will vary annually as a
result of yield variability and the demand of society for
the particular crop, expressed as the price. Conceptually
the management decision problem is not difficult, but
practically it is formidable. First there is the question of
uncertainty—we do not know the long-term effects of
low, intermittent concentrations of many chemicals on
living organisms, including man. Therefore, we cannot
estimate the cost attributable to the specific transport of
a given chemical. Since this, among many other uncer-
tainties, prevents the selection of control practices and
institutional mechanisms that maximize net social bene-
fits, we may wish to state the objective in physical
terms. As an example, one could select those control
practices which maintained the average chemical concen-
tration below some threshold value for a given percent
of the time. If social costs associated with the presence
of this chemical in a stream exceeded the benefits
attributable to it, this procedure would at least lead to
an improved situation. However, as will be shown in
subsequent chapters, our technology in predicting the
effects of changes in inputs and in the system itself on
Uncontrolled Inputs
Precipitation
Solar Radiation
System
(Field, Region)
Controlled Inputs
Seed
Fertilizer
Energy
Pesticides
Management
Capital
Labor
Outputs
Crop Production
Water
Sediment
Nutrients
Pesticides
Figure 1.-Agricultural production system.
-------
X
Precipitation
o
Chemical
Application
Surface
Runoff
-»-t
Chemical Transport
to Stream
Benefits and Costs
GO
1
e
I I
b d
Figure 2.-Sample functions of hydrologic processes and social costs and returns for agricultural system.
a: planting cost to farmer; b: cost of applying chemical; c: social cost when chemical is transported to
stream; d: harvest cost; e: return from sale of atop.
-------
concentrations of potential pollutants In surface waters
Is not developed well enough to moke this approach
feasible. As u last resort, we can use direct runoff,
percolation und erosion as surrogate variables with the
assumption that u change in any of these variables will
a Hoc I water quality. It must bo recognized that reduc-
tion of direct runoff or deep percolation may adversely
ulfcct wutei quality in Home Instances and, therefore,
may create water quality und quantity problems for
downstream water users who depend on runoff from
agricultural lands as u water supply.
Before dealing with specific potential pollutants, it Is
important u> know something about the land resources
ol the U.S. because this has u significant bearing on the
use ol agricultural chemicals, Only the land in the
contiguous 48 slates will hi- discussed, since there Is so
little cropland In Alaska and Hawaii.
The contiguous 48 states contain 1,899,322,000 acres
of land. The nonfederal rural land comprises
1,431 ,()30,000 acres, or 75 percent of the total. The use
>l this rural land is nearly equally divided between
Cropland, pasture and range, and forest land (Table 1). A
land capability classification system1 has been developed
hy Hie U.S. Department of Agriculture and u summary
of the amounts of various classes of soils and their use Is
given in Figure 3. Class i soils are nearly level, have a low
erosion hazard, and arc suited to a wide range of plants.
They are deep, have high permeability and water-holding
capacity, are well drained, and are fairly well supplied
with plant nutrients or are highly responsive to fertili-
sers, Soils in Class 11 have some limitation* that reduce
the choice of plants or require moderate conservation
practices. They often require special soil-conserving
cropping systems, soil conservation practices, water-
control devices, or tillage methods when used for
cultivated crops. Class U soils usually have gentle slopes
and arc moderately susceptible to wind and water
erosion.
('lass III soils are usually found on moderately steep
slopes and are more susceptible to water and wind
erosion than soils in Clash II. They can be used for
cultivated crops but require highly effective conservation
practices that may be difficult to apply und maintain if
erosion IK controlled. Class IV soils ure also suited for
cropland, but they require careful management and ure
often well suited for only two or throe common crops.
They ure usually found on steop slopes und are highly
susceptible to wind and water erosion.
Table I. Predominant land uxc for nonfmloral rural lond In Hie
contiguous 48 unto* (USDA Sutlitlcal llullulln No. 461)
Acre*
CROPLAND:
Row crop* 160,041,000
Cloie-ntruwn cropland liillow , , . . , 132,620,000
I'orunucropi 77,629,000
Cunwrvullon 11*0 .,,,,, , , 39,026,000
Temporary Idle II ,235,000
Orchard*, vlncyurdi, and built I'nilti 5,060,000
Open linul formerly cropped 11,592,000
437,203,000
101,061,000
379,929,000
I'ASTURI AND RANCJI-
Piiiturolund
Ranuoliind .......
I'ORI'ST LAND
Commercial 396,078,000
Noimiiniiu'rriul ^62.860.000
458,938,000
OTIII'K I AND:
In I'limi* , , ,
Nut In liirini ,
27,779,000
27,020,000
1 National Inventory of Soil and Water Conwrvutlon Need*,
1967. I'.S. Department of Agriculture Stutlitlcul Hullolln No,
461, January 1971.
Soils In Classes V, VI, VII, and VIII are limited in
their use and ure generally considered unsuitable for
cultivation. Class V soils have little or no erosion ha/.ard
but have other limitations that are impractical lo
remove, lixamples arc bottom lands subject to frequent
overflow, stony soils, and ponded soils where drainage is
unfeasible, Class VI soils are usually limited lo pasture,
range, forest, or wildlife habitat. However, some Class VI
soils can be used for common crops with careful
management. Some of the soils are also adapted to
special crops such us sodded orchards, blueberries, und
similar crops. Class VII soils ure not suited for cropland,
and Class VIII soils are not only unsulted for cropland,
but have limitations so severe that they are restricted
primarily to recreation, wildlife habitat, water supply,
and esthetic uses.
The erosion ha/.ard of cropland increases sharply
from Class I through Class IV soils. Therefore, the larger
the cropland acreage on Class III and IV soils, the greater
the hu/.urd of erosion. Also, since sediment Is u principal
transport mechanism for agricultural chemicals, the
potential for their loss is much greater on these soils, For
example, It is estimated that from one-third to one-half
of America's agricultural production depends on fertil-
izer use. Therefore, If fertilizer use were eliminated,
cropland acreage would have to be greatly expanded.
Figure 3 shows that any large increase in cropland would
have to come from Class III and IV soils, These soils ore
-------
300
250
200
~ 150
100
50
m
&W;
!•:•'•
m
CROPLAND
PASTURE
& RANGE
FOREST
OTHER
assa-a
\y v
LAND CLASS
v\
V\\
V\\\
Figure 3.-UM of varloui clauoi of land In the 4ftcontlguoui itatoi (baud on data from USDA Statlitlcal Bulletin No. 461).
leu desirable, not only because they are more erocltble,
but because they are lower In fertility and yield
considerably less than Class I and II soils, particularly
when fertilizers are not used.
Unless sediment is controlled at a given level, the loss
of agricultural chemicals from equal treatments will
usually Increase as the soil class number increases. This
suggests that one approach to control water pollution
from cropland is to concentrate crops to the fullest
extent possible on Class I and II soils. These soils are
naturally more productive, more responsive to fertilizers
because of higher water-holding capacities, and easier to
control with respect to sediment losses. In'all likelihood,
therefore, a high level of food and fiber production with
the least Impact on the environment would result from
using fertilizers and pesticides on the better lands whore
their effectiveness Is high and their loss is small. The use
of chemicals on the more erosive soils presents a
substantially greater throat to the environment. How-
ever, it is possible to use them safely on these soils if a
higher level of management Is practiced to control
sediment and associated chemical losses, The treatments
necessary to reduce losses are given In Volume I.
How much agricultural chemicals are affecting the
environment is certainly not clear. However, it appears
that sediment, nutrient, and pesticide losses urn be
controlled at an acceptable level by the selection of
proper management systems. The challenge, therefore, is
to develop appropriate assessment techniques and insti-
tutional mechanisms so that controls are used only when
needed. Also, recommending control practices for a large
area is extremely difficult because the practices are often
site-specific. The concepts presented in Volume I and
the material presented in the following chapters must,
therefore, be considered only us general aids to the
decision-making process. Control recommendations for
specific sites must be developed by specialists within the
area.
5
-------
CHAPTER 2
HYDROLOGIC ASPECTS OF NONPOINT POLLUTION
D. A. Woolhiser
Water, running over the land surface or percolating
through the soil mantle to eventually appear as ground-
water runoff, is a potential carrier of pesticides, nutri-
ents and sediment to streams and lakes. Any discussion
of nonpoint water pollution from agricultural sources
necessarily involves hydrology because water is the
primary transport medium.
In this chapter we will consider some hydrologic
fundamentals, including basic physical principles and a
brief discussion of the stochastic nature of hydrologic
processes. An understanding of the stochastic nature of
hydrologic processes is important because it affects the
interpretation of experimental data. Components of the
hydrologic cycle will be described to illustrate the
physical basis for modifying surface runoff by agro-
nomic and engineering practices. These components have
been aggregated into fairly general mathematical models
with the objective of describing agricultural chemical
transport. Finally, documentation is provided for most
of the 18 direct runoff control practices presented in
Section 4.2 of Volume I.
Only those aspects of the hydrologic cycle that are
important in nonpoint pollution wfll be emphasized in
this report. Readers interested in a more comprehensive
discussion of hydrology are referred to several texts (23,
32, 64, 115}. Although results of experimental investi-
ptions of the effects of land use and treatment on
runoff from agricultural lands in the United States were
reported as early as 1927 (84), only recent experimental
work wfll be considered here because dramatic changes
in agricultural practices have introduced time trends in
the amount of direct runoff from cropland (114).
To understand how nonpoint pollutants move from
fields to surface waters, we must first consider the
physical form and placement of agricultural chemicals,
including nutrients and manures. Then we must consider
the various paths they must follow and the conditions
(such as temperature, oxygen status, biological activity)
they may encounter from field to stream or lake. Form
and placement of the potential pollutants are considered
in subsequent chapters; in this chapter, we wfll concen-
trate on the pathways.
FUNDAMENTALS OF HYDROLOGY
Basic Physical Principles of Hydrology
Two basic physical principles governing the amount
and distribution of water on the earth are those of mass
conservation and energy conservation. These principles,
along with several empirical relationships, form the basis
for most mathematical descriptions of hydrologic phe-
nomena.
The principle of mass conservation is frequently
illustrated by the hydrologic cycle or by the water
budget for an arbitrary volume of sofl. Morton's (52)
qualitative representation of the hydrologic cycle, Figure
1, is useful for introducing some hydrologic terms and
expressing the concept that the mass of water on earth is
assumed to be constant.
If we consider the sector labeled "surface disposition
of precipitation-all forms" in Figure 1, applied to an
arbitrary volume of sofl with surface area, A, and depth,
d, as shown in Figure 2, we can write the conservation
equation for some arbitrary period of time, At:
(D
where:
P
W
precipitation received on the area, A
water imported as a result of man's activities
Qs = net surface runoff (surface runoff leaving A
less surface runoff entering A)
Op = net lateral outflow (may include ground
water flow or unsaturated flow)
-------
Figuie l.-The hydrologic cycle-a qualitative representation [Horton (52)].
AD = increase in surface storage (depression stor-
age and detention storage)
AS = increase in soil water storage
U = net vertical outflow through soil or rock
E = evaporation including evaporation from
plants (transpiration).
All dimensions are in appropriate depth units. The
total water yield for this area, both surface and
subsurface, is the difference between the total input of
precipitation and imported water, and evaporation,
assuming changes in storage are insignificant. Each of
these components will be discussed in more detail in the
next section.
The amount of evaporation is controlled by the
amount of energy available at the layer of soil and air in
which plants grow. A conservation of energy equation
may be written at this interface, expressing the relation:
Net rate of incoming energy per unit area = net rate
of outgoing energy per unit area
where:
Rs
P =
RL -
G =
H =
L =
E =
flux density of total short-wave radiation at
the ground surface
albedo of the ground surface (fraction of
incoming short-wave radiation that is re-
flected)
net flux density of long-wave radiation
heat flux density into the ground
sensible heat transfer into the atmosphere
latent heat of vaporization of water
evaporation rate
Rs(l-p)=RL
(2)
Changes in heat storage in the vegetation and the heat
used in photosynthesis have been ignored in Equation
(2). They would be about 1% of RS. The terms in
Equation (2) are in units of heat energy per unit area per
unit time. The magnitude of the terms in Equation (2)
may vary substantially. If the soil surface is wet or
covered by actively transpiring vegetation, most of the
available solar energy may be used to evaporate water. If
the soil surface is dry, most of the incoming energy may
be used to heat the air.
-------
Equations (1) and (2) are linked by the evaporation
term, E. The magnitude of E in Equation (1) is
effectively limited by the amount of heat energy
delivered to the surface A.
Stochastic Nature of Hydrologic Processes
A set of daily precipitation amounts on a particular
field, arranged chronologically, is an example of a time
series. Other examples include the daily direct runoff
from a field, the daily amount of water percolating
below the depth d, or any of the terms in Equations (1)
or (2) for an arbitrary period of time. An essential
feature of these time series or processes is that they are
unpredictable in a deterministic sense. That is, we
cannot predict with certainty how much rain will fall
tomorrow. These series can be viewed as sample func-
tions of stochastic processes. A stochastic process may
be informally defined as a process developing in time in
a manner controlled by probabilistic laws (81). Many
chance mechanisms are important in agriculture. Precip-
itation is perhaps the most important, but plowing,
planting and harvesting dates, and fertilizer and pesticide
application dates are certainly not deterministic.
To analyze a time series, one must first assume a
mathematical model for the stochastic process which is
completely specified except for parameter values that
can be estimated on the basis of an observed sample.
When the parameter values have been estimated, one can
obtain certain probability expressions that may be
valuable in decision making. For example, what is the
AS
•Q
B
Figure 2.-Control volume for water balance.
-------
probability that the concentration of some substance in
runoff from a field will exceed a certain level for 24
hours or more? In an ideal situation, our stochastic
models would be constructed in accordance with Equa-
tions (1) and (2) so that one might see how a change in
land management could affect the probability statement.
We cannot construct such ideal models. However, the
stochastic nature of hydrologic phenomena must be
appreciated because, if a substance is applied on the
land, we cannot guarantee that it will never be trans-
ported to a stream. The probability of such transport
happening in a particular year may become infinitesi-
mally small, however.
Another important concept is that of stationarity. A
stationary process is one where the chance mechanism
does not change with time. If we consider the process of
surface runoff from a field in continuous corn, we can
see that it is not stationary. Not only are there periodic
changes within a year caused by seasonal phenomena but
also there are long-term trends introduced by changes in
agricultural technology such as new tillage implements,
new crop varieties and increased fertilizer use. Therefore,
one cannot use long time series to estimate parameters
because the parameters are changing with time.
COMPONENTS OF THE HYDROLOGIC CYCLE
In this section we will describe individual components
of the hydrologic cycle and review some of the
mathematical models that have been proposed or used to
describe these elements. The discussions will not be
comprehensive but will consider those aspects deemed
most significant for chemical transport or for reducing
surface runoff.
Interception
When rain begins, drops strike plant leaves and stems
and are retained on these surfaces by the forces of
adhesion and cohesion until a sufficiently thick film of
water accumulates that gravitation overcomes these
forces. If rain continues, the storage on an individual leaf
will become nearly constant, with as much water falling
from the leaf as falls upon it. Water will also be lost from
the film on vegetation by evaporation. There is some
disagreement as to whether this evaporation is a net loss
insofar as the water balance of a volume of soil is
concerned (119). If transpiration is limited by the
energy available, evaporation from the water stored on
leaves is essentially equal to the amount of water that
would be lost by transpiration unless the albedo of wet
vegetation is less than that for dry vegetation. If
transpiration were limited by soil water content, how-
ever, the evaporation from a water film would be greater
and part of it could be considered as a net loss. Water
evaporated from mulch, dead leaves, stems or trunks
could be considered a net loss if energy were not
limiting. Rain intercepted by the canopy may subse-
quently reach the ground by dripping from the leaves or
flowing down the stem. If stemflow is significant, it can
produce substantial differences in soil-water content
over rather small distances (65).
Although several have attempted to develop a mathe-
matical description of interception based on physical
reasoning (51, 63), the models have been rather crude.
Many mathematical watershed models do not include an
explicit component for interception (27, 50). Crawford
and Linsley (26) combine interception and depression
storage into a single lumped storage with depletion by
evaporation and transfer to a lower zone storage.
Boughton's model (10) and the Tennessee Valley Au-
thority model (110) assume that precipitation will
accumulate in interception storage until a threshold or
capacity value is reached. The TVA model uses capa-
cities for forested watersheds of 0.05 inch in winter and
0.25 inch in summer. Saxton et al. (95) used a storage
amount of 0.10 inch for agricultural crops and showed
that evaporation from this source can be several inches
per year in a semi-humid climate.
Zinke (119) concluded:
"A survey of the data in the literature indicates
interception storage amounts for rain of from 0.25 mm
to 9.14 mm (0.01 to 0.36 in.) and a similar range for
snow, 0.25 mm to 7.62 mm (0.01 to 0.30 in.).
The storages indicate that one would not be greatly in
error to estimate about 1.3 mm (0.05 in.) storage
capacities for rain for most grasses, shrubs and trees; and
3.8 mm (0.15 in.) for snow for trees."
Jones (58) concluded that "a consistent difference in
storage capacity for trees, crops, and grass of various
heights was not evident."
From this brief review, interception does not appear
to have an important influence on runoff or deep
percolation from fields. However, an increase in inter-
ception is partly responsible for the reduction in runoff
caused by conversion from clean-tilled crops to pasture
or meadow.
10
-------
Depression Storage
After interception storage has been filled and the
infiltration capacity of the soil is exceeded so that all or
part of the soil surface is saturated, water will accumu-
late in surface depressions. Water stored in depressions
either evaporates or infiltrates into the soil—none of it
runs off the surface.
Depression storage can be increased by agronomic or
engineering practices and, therefore, can be important in
reducing direct runoff from fields. For example, under
ideal circumstances, as much as 2.5 inches may be stored
in contour furrows constructed with a range furrowing
machine commonly used in the west (77). Level bench
terraces with a capacity of over 2 inches have been
installed in the deep loess soils of western Iowa (96).
Doty and Wiersma (28) found that the maximum
potential depression storage capacity for conventional
contouring and for bedding and listing practices ranged
from approximately 1 inch for contouring to as much as
3 inches for listing and bedding. The potential surface
water storage decreases as land slope increases and is
approximately half as great for a 7 percent slope as for a
1 percent slope.
Agronomic and engineering practices to increase
depression storage have a transient effect. With annual
cropping systems, storage capacity usually is maximum
in the planting to first cultivation period, which is
frequently the most important for reducing losses of
agricultural chemicals by surface runoff. The storage
capacity then decreases and reaches a minimum during
the harvest to plowing period (28). Contour furrows in
range and pasture have maximum storage immediately
after installation. Erosion and trampling by livestock
gradually reduce this storage capacity. For example,
contour furrows in eastern Montana had only half their
original storage capacity after 6-10 years, and the
average effective life (storage > .05 inch) was about 25
years (77).
Mathematical descriptions of depression storage
usually represent it as a volumetric threshold that must
be exceeded before surface runoff occurs (10, 50). The
Stanford model (26) lumps depression storage with
interception but does not assume a fixed threshold
value. This approach can partially account for the spatial
variability of surface detention over the watershed. The
parameters in these models are usually found by trial
and error or by optimization techniques. Very little
information is available that could serve as a guide in
choosing values for depression storage based on physical
measurements in the field. The work of Doty and
Wiersma (28) is one exception for fairly simple geo-
metric shapes. Boughton (JO) suggested that the "ran-
dom roughness"of soil surface microtopography de-
scribed by Burwell and others (15, 16) might be an
adaptable measure of the depression storage.
Although manipulation of depression storage is an
obvious method of affecting surface runoff, the amounts
of change can be deduced only indirectly by analysis of
rainfall and runoff. The curve numbers for contoured
and contoured and terraced areas for the Soil Conserva-
tion Service method of estimating direct runoff shown in
Appendix A reflect some empirical data mixed with
judgment. Mathematical models that include depression
storage explicitly could be used to predict changes.
However, transciency of depression storage and its
dependence on precipitation, runoff, and erosion make
prediction difficult.
Infiltration
As snow melts or rain falls on the soil surface or drips
from the vegetation, the phenomenon of infiltration
governs the amount of water that will enter the soil and
thereby greatly affects the amount of surface runoff.
Some of the physical, chemical and biological charac-
teristics of soil that affect infiltration can be manipu-
lated by man through agronomic and engineering prac-
tices. Therefore, changing the infiltration characteristics
of soils can profoundly affect the amount of surface
runoff as well as the amount of water stored in the soil
for plant use.
Characteristics of both the porous medium and the
fluid affect infiltration. The porosity, pore-size distribu-
tion and tortuosity of soil pores all substantially affect
infiltration rates. Sands have higher infiltration rates
than silts or clays, which have a higher porosity but
much smaller pores. Soil compaction by the trampling of
livestock reduces infiltration capacity and increases
surface runoff (85). From this evidence it can be
inferred that compaction by machinery would also
decrease infiltration rates and that practices that reduce
machine traffic on a field should reduce surface runoff.
Raindrop impact on bare soil breaks up soil aggre-
gates into their component particles or much smaller
aggregates. These particles or small aggregates can be
carried into larger pores by water and form a thin
surface layer that has low hydraulic conductivity. This
surface layer may then control the infiltration rate (33,
47). Vegetation or mulches protect the soil surface from
raindrop impact and can prevent crust formation.
Dense vegetation with massive root systems and
farming systems that leave substantial amounts of plant
residues near the surface maintain high soil organic
matter content and promote aggregate stability, thus
maintaining high infiltration rates. Vegetation also has a
11
-------
higher evapotranspiration rate between rains than evapo-
ration from bare soil, thus the soil water content is
reduced at the beginning of the next rain which increases
the rate and amount of infiltration.
Tillage can increase the volume of large pores near the
soil surface and thereby increase infiltration rates. The
effect is transient, however.
Frozen soil usually has a lower infiltration rate than
unfrozen soil. If the soil is frozen while wet, a dense,
nearly impermeable mass may result. However, if frozen
while dry, some soils will show little change in infiltra-
tion rate (76). The effect of increased viscosity of the
water is apparently compensated for by the structural
change caused by freezing. Frost usually penetrates
deeper if the soil is bare than if it is snow covered.
Therefore, practices that prevent snow from blowing
away tend to lessen frost penetration but the additional
snow deposited may increase runoff.
Modern infiltration theory based on the theory of
unsaturated flow or two-phase flow in porous media has
provided a basis for understanding infiltration behavior.
This theory has been presented in several recent texts or
reviews dealing with theoretical aspects of infiltration (8,
38, 46, 75, 83).
Although infiltration theory is useful in explaining
observed infiltration phenomena, it has just begun to be
used in quantitatively estimating the effects of agro-
nomic or engineering practices on infiltration and
surface runoff. The partial differential equations describ-
ing infiltration must be solved by numerical methods—a
time consuming and costly task if one wishes to find
long-term average effects or distribution functions of
surface runoff. Also, this approach, with its strong
physical basis, requires costly and difficult measure-
ments of soil conductivity and diffusivity (109).
Because of these difficulties, several infiltration equa-
tions, either entirely empirical or based on simplifi-
cations of the more general formulations, have been
used. Equations presented by Norton (55) and Holtan
(48) are examples of the former. Gre.en and Ampt (39),
Philip (82, 83), Smith (104), Mein and Larsen (69), and
Brustkern and Morel-Seytoux (14) used either simplifi-
cations of the basic equations or algebraic approxi-
mations of numerical solutions of the basic equations.
The first three of these apply only to infiltration from a
ponded surface rather than to rainfall conditions.
Although solution of these equations is simpler and less
costly than solution of the more rigorous partial
differential equations, the question of parameter esti-
mation remains. Usually they are estimated for different
soil and cover conditions from infiltrometer experiments
on small plots or data from small watersheds. Musgrave
and Holtan (76) reviewed much of the data available
before 1964. Holtan and his associates attempted to
develop techniques for estimating parameters in the
Holtan equation by using information available in soil
surveys (34) or by estimating parameters for various
land-use or cover factors (49).
Of the hydrologic models considered, none includes
an infiltration component based on the numerical
solution of unsaturated flow or two-phase flow in
porous media. The USDAHL model (56) utilizes
Hoi tan's equation. The Bough ton model, the TVA
model and the Stanford model utilize empirical lumped
storage infiltration components, although the Stanford
model attempts to account for spatial variability by
assuming an invariant statistical distribution of infiltra-
tion capacity. The USGS model (27) utilizes an adapta-
tion of the Green and Ampt equation.
Soil Water and Ground water
Water stored in the soil and rock is frequently
separated into two components: the saturated or
groundwater zone and the unsaturated zone between the
groundwater and the surface. Water moves within the
unsaturated zone in response to gravitational and capil-
lary potential gradients. It may move generally down-
ward during rainfall or snowmelt and generally upward
after a long, dry period, or it may move upward near the
surface and downward in the lower part of the profile
simultaneously. In general, water movement in the
unsaturated zone will be predominantly vertical.
In some soils, a rather permeable topsoil is underlain
by a slowly permeable clay layer. If infiltration is rapid
enough, the surface soil may become saturated, resulting
in flow which is predominantly in a lateral downslope
direction and is known as interflow. This water may
reappear on the surface some distance downslope or at
the foot of the slope. Hydrologists generally agree that a
flow mechanism such as interflow exists; however, there
is some argument about its importance. Dunne (31)
concluded from his measurements of subsurface storm
flow in Vermont that interflow (subsurface storm flow
in his terminology) did not contribute significantly to
flood hydrographs. This does not mean, however, that
interflow is unimportant to water quality in some
regions. Minshall and Jamison (71) presented data
suggesting that interflow can exist on Midwest claypan
soils.
An interflow runoff component is included in the
Stanford model, the TVA model and the USDAH1/70
model. However, the volume of interflow runoff has not
been compared with field measurements because of the
difficulty in making such measurements. Therefore, it is
12
-------
difficult to ascertain if computed volumes are realistic or
are merely a result of curve-fitting procedures.
As Amerman (3) has pointed out, the separation
between the saturated zone and the unsaturated zone is
unnecessary from the physical point of view and is
possibly misleading. Figure 3(a) shows a hypothetical
transverse cross section through a valley during a
relatively dry period and Figure 3(b) shows a similar
section during a wet period. Under steady-state condi-
tions, the streamlines would represent the path lines of
water molecules or dissolved materials. However, hydrol-
ogic systems are usually unsteady so the streamlines are
continually shifting. The medium shown in this sketch is
isotropic so the streamlines are perpendicular to the
equipotential lines. In an anisotropic porous medium
that contained a relatively impervious layer, for ex-
ample, this would not be true.
Figure 3 illustrates some important points about
transport of dissolved chemicals. Suppose that in Figure
3(b) the soil surface from point A to point B was within
a single field. If we assume a steady state, the path line
from A to the stream is much shorter than that from B
to the stream. Therefore, a soluble chemical that leached
below the root zone on a particular day would take
BEDROCK
Figure 3(a).-Cross section of hypothetical hydrological system during a relatively dry period.
Figure 3(b).-Section of hypothetical hydrologic system during a wet period. [From Amerman, (3)]
13
-------
much longer to reach the stream from point B than from
point A. Hydrodynamic dispersion will also affect the
arrival time at a stream but has a relatively small effect
compared with that of macroscopic flow (78). How
much delay time might be involved between the arrival
times of chemical constituents at the stream? This time,
of course, depends on the path length and velocity.
Some numerical models can answer this question if the
system geometry and hydraulic characteristics of the
medium are known (12, 86, 100). As a crude approxi-
mation we might use the results of Carlston (20), who
found that the mean residence time of groundwater
recharge in a Wisconsin drainage basin was about 45
days. If we assume that the time of travel for a particle
following the streamline originating between A and B is
equal to the mean residence time and that the path
length of A is half the mean and the path length of B is
1.5 times the mean, the arrival of a slug of chemical
distributed uniformly over the field extending from A to
B might appear at the stream over an interval of 45 days.
Of course, the physical, chemical and biological
processes that affect the particular constituent during its
travels through the porous medium must also be
considered. For example, if we are concerned with
nitrate transport, some zones along the flow path may
be anaerobic and contain carbon. Under these circum-
stances, bacteria may convert the nitrate to harmless N
gas. Such conditions might well exist in the seep area
shown in Figure 3(b).
Legrand (62) discussed the patterns of contaminated
zones of water in the ground. Although he considered
contamination sources of small areal extent (point
sources), his concepts can be readily applied to nonpoint
sources. He noted that when contaminants move
through the unsaturated zone and reach the water table,
"enclaves" of contaminated water extend from the
source in the direction of, groundwater movement, as
shown in Figure 4. If the contaminant is not adsorbed or
chemically or biologically transformed, the enclave will
terminate at a stream (Field A, Fig. 4) and may cause
pollution. Because of additional water entering the
stream, the contaminant may be diluted to a harmless
level at a point C downstream.
The boundary of the enclave shown in Figure 4
assumes a constant inflow of the chemical uniformly
distributed over the field. As a rule, inputs will be
intermittent; therefore, the pattern may consist of a
series of smaller enclaves moving toward the stream
completely surrounded by uncontaminated water. The
dashed line emanating from the lower boundary of field
A terminates before reaching the stream, illustrating the
situation in which some of the chemical may pass
through a zone where chemical or biological reactions
may reduce its concentration to harmless levels before it
reaches a stream. The same situation holds for the
contaminant moving from field B. Enclaves will change
in areal extent and in shape as the water table changes its
configuration naturally or by pumping of wells.
Robbins and Kriz (92) presented a comprehensive
review of groundwater pollution caused by point and
nonpoint agricultural sources. Their concern was pri-
marily with measurements of water quality within the
enclaves of groundwater contamination, not with the
effects on water quality in streams and lakes. For an
excellent review of mathematical models describing
movement of chemicals in soils, see Boast (9).
The subsurface transport of agricultural chemicals
from a field to water bodies is obviously very compli-
cated. Although we have a qualitative understanding of
such transport, much uncertainty is involved in predict-
ing when and how much of a chemical may reach a
stream or lake, or how much the amount can be reduced
by control practices.
We can, however, identify certain goals of subsurface
water management on agricultural land. Maintaining
adequate water in the root zone and encouraging a
vigorous crop are advantageous from both the crop
production and water quality standpoints. Deep percola-
tion will occur in most humid and sub-humid climates.
Variations in soil characteristics will lead to substantial
differences in annual percolation, as shown in Figure 12,
Vol. I, and in Appendix B of this volume. In those areas
with substantial deep percolation, soluble agricultural
chemicals must be applied with more care.
Evapotranspiration
The sum of evaporation from the soil surface and
transpiration from plants is called "evapotranspiration"
and represents the transport of water from the earth to
the atmosphere. It is important in agriculture because it
is required for crop growth. It is important in the loss of
potential pollutants from cropland because it affects the
volume of direct runoff and the amount of soil water
that percolates to the saturated zone. Evapotranspiration
is obviously a major component in the hydrologic
cycle—it transports about 70 percent of the water that
falls on the conterminous United States back to the
atmosphere. This percentage can vary from 100 in arid
regions to about 50 or less in some mountainous areas of
the U. S.
Three physical requirements must be met for evapora-
tion from a surface to continue: 1) There must be a
supply of heat to convert liquid water to vapor, 2) the
14
-------
Direction of Ground
Water Concentration
Pumped
Well
C(Downstream Limit of Harmful Concentration)
Figure 4.-Plan view of water-table aquifer showing enclaves of groundwater with high concentrations of a soluble material added to
fields A and B.
vapor pressure of the air must be less than that of the
evaporating surface, and 3) water must be continually
available. Evapotranspiration, through the latent heat
term, is a component of the energy balance, Equation
(2), as well as the hydrologic water balance, Equation
(1).
When water is not limiting, the evapotranspiration
rate is limited by the radiant energy and advected energy
available. Therefore, a lower limit exists for total
runoff-the difference between precipitation and poten-
tial evapotranspiration. Potential evapotranspiration is
defined as the hypothetical rate of water loss from a
large, homogeneous area of continuous green crop,
under the given meteorological conditions, when there is
no resistance to water supply at the evaporating surface
(112).
At most locations in the United States, soil water is
limiting some time during the year, so the actual
evapotranspiration will be less than the potential even if
the ground is fully covered by a crop. With annual row
crops, the ground will not be covered by a transpiring
crop canopy for a substantial period of time, so
evapotranspiration will be less than from grasses and the
total runoff will be greater. This is one reason why
conversion from row crops to meadow or pasture usually
reduces runoff.
The physics of evaporation and evapotranspiration is
discussed in several texts (79, 93, 108). Here, we will
briefly outline the approaches that have been used to
estimate actual evapotranspiration from cropland. In
general, the models used consist of a continuity relation-
ship, a means of computing potential evapotranspiration,
15
-------
E0, which serves as the upper limit of the actual
evapotranspiration rate, a method of computing actual
evapotranspiration, E, as a function of soil water content
when it is below some critical level, and a means to
modify E if the ground is not fully covered by a crop
canopy.
Potential evapotranspiration can be computed by the
energy budget method, the aerodynamic method, or a
combination of the two (113). It can also be estimated
empirically from evaporation pan data. The equation
frequently used is:
where K is a "pan coefficient" and Ep is the evaporation
from a standard pan. Saxton et al. (94) found that daily
evapotranspiration computed by adjusted pan evapora-
tion was highly correlated (R2 = 0.87) with Eo calcu-
lated by the combination method.
It now seems to be accepted that actual evapotrans-
piration can be less than potential at soil water contents
above the wilting point. Baier (4) presented a compre-
hensive review of this subject. The procedure used in the
simulations of potential percolation in Volume I and
documented in Appendix B of this volume uses a
relationship between E/EO and available water that is
similar to those presented in the literature. When
evapotranspiration estimates are needed for different
stages of crop development, the evaporation rate can be
corrected by using a crop coefficient, Kp, that varies
according to the stage of growth of the crop (57), or the
ratio E/EO may be related to the leaf area index, LAI
(41, 90). The methods used for the simulations pre-
sented in Volume I are described in more detail in
Appendix B. For the extensive simulation study of
percolation and nitrate leaching in Volume 1, a phys-
ically more realistic but more complex model such as
presented by Richardson and Ritchie (89) or Saxton et
al. (95) would have been difficult to use with existing
time constraints.
Surface Runoff
As the transport medium for dissolved chemicals and
for sediments with their adsorbed chemicals, surface
runoff is an important link between fields and streams or
lakes.
Surface runoff begins when the rainfall (or snowmelt)
rate exceeds the infiltration rate of the soil and
depression storage is filled. Surface runoff is classified
somewhat arbitrarily as either overland flow or channel
flow. Overland flow is sometimes considered to be thin
sheet flow over a relatively smooth surface. However, a
more general and realistic definition would be the flow
that is outside of the well-defined channel system. The
mean velocity of overland flow is directly related to the
slope (laminar flow) or the square root of the slope
(turbulent flow) and is inversely related to the hydraulic
resistance of the surface. The hydraulic resistance varies
widely, depending on the surface characteristics, from a
Mannings resistance coefficient of 0.02 for bare soil to
0.4 for a dense turf (118). Such differences in hydraulic
resistance would result in water being about six times
deeper on the turf than on the bare soil for the same
discharge. The velocity, of course, would be only
one-sixth of that on the bare soil. The greater depth on
the dense sod would allow much more time for
infiltration after the rainfall stopped, resulting in less
runoff even if the infiltration characteristics of the soils
were the same. The decreased shear stress on the soil
with a sod cover would also result in a much lower
erosion potential.
Bailey, Swank and Nicholson (5) described the modes
of pesticide transport into and within the moving liquid
boundary during rainfall. The same processes would also
apply to nutrient transport by surface runoff. This
transport process consists of four mechanisms, as shown
in Figure 5:1) diffusion and turbulent transport of the
dissolved chemical from the soil water into the overland
flow film, 2) desorption of the chemical from soil
particles into or toward the moving film, 3) dissolution
of stationary paniculate matter trapped at the bound-
ary, and 4) scouring of paniculate matter and its
subsequent dissolution.
From a consideration of these processes, one could
infer that practices that reduce runoff velocities and
prevent scour of particulate matter might reduce chem-
ical transport even if the total volume of runoff were not
reduced. However, the most effective practices would be
those that increased infiltration rates so that more
chemicals could be carried into the soil by bulk-flow
transport. An increase in depression storage would have
a similar effect.
16
-------
Moving Liquid Boundary
(Film Thickness Increasing
Down Slope )
Soil Surface
( D) Moving
Dissolution
PP = Pesticide Particulate
in Motion
(C) Stationary
Dissolution
ps - Pesticide in
Solution in Motion
Water
Movement
(B) Pesticide (A) Liquid-Liquid
Desorption into or Diffusion Interchange
towards Moving (Mass Transfer of
Liquid Boundary Pesticide)
D Pesticide Particle
O Soil Particle
OPesticide Adsorbed on Soil Particle
/// Soil Solution Containing Pesticide
Figure 5.-Modes of pesticide transport into and within the moving liquid boundary during a rainfall event. [From Bailey, Swank and
Nicholson (5)]
AGRICULTURAL CHEMICAL AND SEDIMENT TRANSPORT MODELS
Models of agricultural chemical and sediment trans-
port (which may be interpreted to include predictions
made by them) represent our descriptions of how water,
sediment, and chemicals move on fields or watersheds
under existing or proposed conditions. The models
should not violate the basic physical principles of
hydrology and should incorporate principles of chem-
istry and biochemistry needed to describe chemical
behavior in a biological system. They will also include a
number of empirical relationships.
Comprehensive models of the transport of water
(hydrologic models) have been used for about 10 years.
Several of them have been discussed in previous sections
of this paper. Although special purpose water quality
models were developed as early as 1925 (JOT), general
transport models were not developed until 1967 (54,
55).
Development of agricultural chemical transport
models started around 1970. There have been several
reviews of the "state of the art" and the philosophy of
modeling chemical transport (1, 24, 37, 59, 61, 116,
117).
Bailey et al. (5) have developed a conceptual model
of pesticide runoff from agricultural lands, and several
quite general models are in the development and testing
stage (13, 25, 37).
The model developers usually started with a hydrol-
ogic model that was developed for some other purpose
and added components for chemical transport. The
structure of the hydrologic model used thus served as a
constraint on the transport model, imposing all of its
constraints and shortcomings. When these models have
been tested more thoroughly, many of the shortcomings
may be shown to be in the structure of the hydrologic
model.
These detailed models may be quite useful in an
intensive study of a particular field or watershed, but
they are far too complex for the extensive scope of this
report. Field data available for calibrating or testing
these models is also limited. Because of the present lack
of knowledge in modeling movement of agricultural
chemicals, we used potential direct runoff and potential
percolation as surrogate variables in Volume I. We
implicitly assumed that if surface runoff were reduced,
the transport of chemicals would also be reduced. The
models used to estimate potential direct runoff and
potential percolation are described in Appendices A and
B, respectively.
17
-------
AGRICULTURAL PRACTICES TO CONTROL DIRECT RUNOFF
Eighteen practices for controlling direct runoff, desig-
nated as Rl though R18, are presented in Volume I
(Table 14 and Section 4.2). Practices that reduce erosion
will usually reduce direct runoff, although to a lesser
extent. Therefore, the first 16 runoff control measures
have been assigned the same reference numbers as the
identical erosion control measures; only the alphabetical
prefixes differ. These agronomic or engineering practices
constitute means whereby direct runoff may be reduced
as compared to direct runoff from an index crop-
summer row crop (corn) with straight rows. These
practices are discussed in Volume I without supporting
documentation. Table 1 contains citations of articles
supporting statements made in Volume I and of articles
containing closely related information that may be
helpful in evaluating individual practices.
Most of the research cited in Table 1 was completed
within the last 15 years. Earlier work is not cited
because of possible nonstationarity caused by changes in
agricultural practices. The percentage reductions in
runoff are shown without any indication of statistical
significance. However, the decreases reported are consist-
ent with the physical basis of hydrology discussed in
previous sections. Ranges in response for individual
practices are usually attributable to soil and climatic
differences and to sampling variability.
Additional documentation of land use and treatment
is given in the reports of the Cooperative Water Yield
Procedures Study Project (101, 102) and in several re-
cent reviews (11, 61, 74).
Table 1. Bibliography on practices to control direct runoff (Volume I, Section 4.2)
No.
Runoff Control Practice
Page No.
in
Vol. I.
Description
Citations
Significant Subjects
Rl
71
No-till Plant in Residues
of Previous Crop
R2
71
Conservation Tillage
Harrold, Triplett and
Youker (42)
Harrold and Edwards (45)
Harrold, Triplett and
Youker (43)
Harrold, Triplett and
Youker (44)
Smith and Whitaker (103)
Allis (2)
Free and Bay (36)
Manner ing and Burwell
(66)
Comparison of runoff and soil loss from
no-till and conventional tillage corn at
Coshocton, Ohio.
Single-storm runoff from a no-till field of
corn on a 21% slope was less than that from
straight-row corn field on a 6.6% slope but
slightly greater than that from contoured
corn on the 6.6% slope.
Comparison of 3 years of runoff and soil
loss data from no-till and conventional
tillage corn at Coshocton, Ohio.
Five-year average May-Sept, runoff was 0.44
inch for conventional and 0.04 inch for
no-till corn at Coshocton, Ohio.
In a 3-year period at McCredie, Mo., runoff
from corn with conventional tillage
averaged 5 in; runoff from no-till fields was
6.7 in.
Over a 9-year period direct runoff from a
subtitled field in a corn-oats-wheat
rotation was 19% less than from straight-row
fields in the same rotation.
Runoff from a field of corn with mulch
tillage was greater than runoff from
conventional tillage.
Review runoff and erosion data from
various mulch tillage practices.
18
-------
Table 1. (continued)
Runoff Control Practice
No.
Page No.
in
Vol. I.
R3
72
R4
72
R5
73
R6
73
Description
Sod-Based Rotations
Meadowless Rotations
Winter Cover Crop
Improved Soil Fertility
Citations
Significant Subjects
Moldcnhauer et al (72)
Onstad (79)
Jamison, Smith and
Thornton (56)
Epstein and Grant (35)
Barnett (7)
Burwell and Holt (]7)
Carter, Doty and Carroll
(22)
Mannering, Meyer and
Johnson (67)
Moldenhauer, Wischmeier
and Parker (73)
Saxton and Whitaker (98)
Soil Conservation Service
(105)
Jamison, Smith and
Thornton (56)
Richardson (88)
Mannering and Burwell
(66)
Smith and Whitaker (103)
Jamison, Smith and
Thornton (56)
Runoff from a till-plant field was slightly
less than from a conventionally-tilled field
for rainfall applied with a rainfall simulator
in early June.
Till-plant tillage up and down the slopes
reduced runoff 42% over a 6-year period as
compared to conventional tillage.
Review of experiments at McCredie, Mo..
Comparison of runoff and erosion from
continuous potatoes and potatoes-sod-oats
rotation at Presque Isle, Me.
Rotation studies at Watkinsville, Ga.
Compares runoff from corn-oats-hay
rotation with runoff from continuous corn
in west-central Minnesota.
Runoff from Bermudagrass-corn rotation
compared with continuous corn at Holly
Springs, Miss.
Evaluated effect of sod-based rotation on
soil loss and infiltration using a rainfall
simulator.
Runoff measured from corn-oats-meadow
rotation and from continuous corn.
Runoff measured from corn, small grain,
meadow rotation and from continuous
row crops.
Runoff curve numbers established for
rotation meadow and row crops in
rotation.
Review of crop rotation experiments at
McCredie, Mo.
Measurements of runoff from cotton, corn,
oats rotation and oats, clover, cotton and
grain sorghum rotation at Riescl, Tex.
No runoff reduction for corn interseeded
with legumes at LaCrosse, Wis.
A small grain cover crop planted after
corn was removed for silage reduced runoff
substantially (5.5 in. vs. 11.5 in.)
Average annual runoff from plots at
McCredie Mo. ranged from 6.8 to 11.7
inches during 1941-50. Lowest runoff was
for well fertilized pasture; highest runoff
was for unfertilized corn-oats rotation.
19
-------
Table I. (continued)
No.
Runoff Control Practice
Page No.
in
Vol.1.
Description
Citation*
Significant Subject»
R7
RX
73
73
R9
73
Timing of I idd Operation*
Plow-Plant Systems
Contouring
Carter, Dendy and Doty
! Mofdenhauer. Wuchmeier
and Parker (73j
Saxton and Whitaker (98)
Wuchmeier (114)
None
Free and Bay (36.)
Mannering and BurweO
(66)
Wi*chmeier(H4)
Harrold and Edward* (45)
Alfc* (2;
Carter, Doty and
J Carroll (22)
Onstad (79)
Wi*chmeier(iJ4>
SiilUfetal (9l>
Average runoff wa* 9.01 inches from
improved fertilized pasture* and 17,9 from
unfertilized pasture* for a 6-year period.
fcxperiment wa* at Holly Spring*, Mm.
For a 10-year period at Oarinda, Iowa.
runoff from continuous corn receiving
nitrogen fertilizer wa* 2S'/' let* than runoff
from unfertilized corn.
Average annual runoff from row crop* with
fall fertilization wa* 1.10 inche* at
compared with 2.16 inche* from row crop*
receiving only ttarter fertilizer.
The ratio of runoff from corn land to
runoff from adjacent fallow decrease* with
increatci in corn yield. Much of the
increase in corn yield » cauted fy higher
fertilizer u*e.
Average growing *ea*on runoff for plow-
plant com wa* lew than runoff from con-
ventional corn at MarceNu*, NY,
Runoff from nnrulated rainfall on pkm -
plant corn wa* tew than that from con-
ventional com.
Report* three *tudiet that
-------
Table I. (continued)
No.
Runoff Control Practice
Page No.
in
Vol. I.
Description
Citation*
Significant Subject*
RIO
RH
R12
73
73
R13
RI4
R15
RI6
74
74
Graded Row*
Contour Strip Cropping
Terraces
Grassed Outlets
Ridge Planting
74
76
Contour Listing
Change in Land Use
| OnstadandObonrap)
Pfest(I06)
Soil Con«ervation Service
: Moldenhauer et al (72)
I
| None
i Baird and Richardson (6)
Richardion (f§)
Spomer,, Hetnentann and
PiesHUM)
Saxton and Spomer (96)
Saxton, Spomer and
Kramer (97)
Soil Conservation Service
None
Maimeringand Burwell
(66)
HoUtntuuer ft aJ
Hitter ft al (91)
Manneringand BurweU
(66)
Jamison, Smith and
Thornton (56)
Runoff from contoured and conservation
tillage field* in corn.
Compare* runoff from contoured corn with
level terraced com and with meadow.
Runoff curve number* established for con-
toured row crop*.
Graded row* on dope* of 3 A to 9% did not
reduce runoff.
Effects inferred from runoff reduction by
meadow.
Terracing alone on heavy day toil* of
Texas Blacklandt had little effect on runoff
volume.
Effect* of conservation practice* including
terracing on runoff.
Level terrace* drastically reduced turface
runoff in we*tem Iowa but groundwater flow
increased.
Level terrace* reduced discharge of water,
sediment, nitrogen and phosphorus.
Fourteen percent of water yield from level
terraced watershed wa* surface runoff.
Sixty-four percent of water yield from
contour watershed wa* surface runoff.
Effects of level terracing on runoff and
erosion.
Runoff curve numbers established for
graded terrace*.
No data available on effect* of grassed
outlets on surface runoff.
Ridge planting on contour reduced direct
runoff.
Ridge planting on graded rows did not
reduce direct runoff from a simulated rain.
Ridge planting reduced pesticide runoff.
Cite Iowa study where annual direct
runoff from contour-listed com wa* 55% less
than that from straight-row planting up and
down the slope.
Direct runoff from pasture and meadow wa*
lower than that from com in central
Missouri.
21
-------
Table 1. (continued)
Runoff Control Practice
No.
Page No.
in
Vol. I.
Description
Citations
Significant Subjects
R17
76
Other Practices
R18
76
Construction of Ponds
Dragoun (29)
McGuinness and Harrold
(68)
Rice and Dragoun (87)
Saxton and Whitaker (98)
Spomer, Heinemann and
Piest (106)
Thomas, Carter,
and Carreker (111)
Wischmeier (114)
Hanson et al (40)
Soil Conservation Service
(105)
Dragoun and Kuhlman
(30)
Mickelson (70)
Neff (77)
Schwab and Fouss (99)
Langbein, Hains and
Culler (60)
At Hastings, Nebr. average annual direct
runoff was 0.20 inch from watershed in
grass and 5 .24 inches from cultivated fields
in row crops.
Water yield decreased when watershed was
reforested.
Reseeding cropland with perennial prairie
grasses reduced runoff by 94% in a 2-year
period.
Comparison of direct runoff from pasture
and meadow.
Comparison of direct runoff from perennial
grass, contoured corn and level-terraced corn.
Bermudagrass meadow reduced runoff.
For nearly 5000 plot-years of data analyzed,
runoff from row crops averaged 12% of
total rainfall, while that from meadow
averaged 7%.
Effects of grazing intensity on direct
runoff from rangeland.
Runoff curve numbers established for
pasture and meadow.
Contour furrowing reduced runoff from
pastures.
Storing runoff in leveled areas for crop
production.
Storage capacity of contour furrows in
rangeland.
Surface runoff and tile flow from fields
with com and grass cover.
Hydrology of ponds and stock-water
reservoirs.
22
-------
RESEARCH NEEDS
Research on the effect of land management practices
on hydrology has usually involved three steps: 1)
intensive experimental measurements on plots and
watersheds, 2) analysis of the data using some type of a
mathematical model, and 3) generalization of results for
more extensive application.
Measurements made in the first step are frequently
governed by the model that is to be used in the second
step. For example, in most of the experimental work
examined, only rainfall and runoff were measured. This
was adequate when the only question was "Will treat-
ment A reduce surface runoff?" and when the study
could be maintained for enough time to obtain statis-
tically significant results. Unfortunately, we can no
longer afford this luxury of time. Policy decisions must
often be made quickly and by the time we have
statistical significance, the practice may be obsolete.
The alternative is to develop more detailed models to
use as a framework in analyzing the data and to obtain
more intensive measurements in a shorter time. The
experimental data can be used to estimate model
parameters and techniques must be developed for
predicting parameters from readily obtainable physical
measurements. Simulation can then be used to evaluate
the stochastic properties of the system and to examine
long-term effects.
Stochastic models of point and areal precipitation
must be developed and the parameters regionalized by
mapping or other techniques. As plant growth models
and other biological processes are included in hydrologic
models, the stochastic inputs must be expanded to
include temperature and radiation. Obviously, the joint
probability structure of precipitation, radiation and
temperature must be maintained.
Prediction of runoff from complex areas is still
difficult and needs a great deal of work from the
standpoint of water quality. If concentrations of the
chemical are important we must estimate the joint
probability structure of discharge of chemicals to a
stream and the quantity of water in the stream.
A second generation of agricultural chemical trans-
port models should be developed after the first gene-
ration models have been tested and their strengths and
weaknesses identified. Material models, systems which
retain many of the important characteristics of real
watersheds but are easier to manipulate and control,
may play an important part in model testing and in
understanding the significance of parameters. These
models, which would be less than an acre in size but
much larger and more complex than a soil column or a
lysimeter, would allow deliberate departures from homo-
geneity. The sensitivity of model parameters to such
variations could then be established under controlled
conditions.
The third aspect of past research, generalization of
results for more extensive application, needs much more
emphasis. The SCS curve number procedure for estimat-
ing direct runoff and the Universal Soil Loss Equation
are examples. These techniques were developed before
the advent of, or during the infancy of high-speed
computers and the models used were accordingly simple.
This constraint has been relaxed considerably so it
appears that significant improvements could be made.
For example, the direct runoff estimation procedure
could be improved by incorporating a simple soil
moisture accounting instead of the antecedent rainfall
index. The functional form of the equation could be
changed to more closely approximate results predicted
by modern infiltration theory. Results from complex
hydrologic models should be used along with experi-
mental results to develop a new procedure for estimating
direct runoff.
It should be emphasized that no single model will
meet all needs. We need a set of models, involving
increasing abstraction, and an objective procedure for
selecting the appropriate one for the job at hand.
23
-------
LITERATURE CITED
1. Aleti, A., Chiu, S. Y., and McElroy, A. D. 1974.
Methods for identifying and evaluating the nature
and extent of nonpoint sources of pollutants from
agriculture. Proc. 1974 Cornell Agr. Waste Manag.
Conf., 10-23.
2. Allis, J. A. 1952. The story of two watersheds. Jour.
Soil and Water Conserv. 7(5): 243.
12. Bredehoeft, J. D. and Finder, G. F. 1973. Mass
transport in flowing groundwater. Water Resour.
Res. 9(1): 194-210.
13. Bruce, R. R., Harper, L. A., Leonard, R. A., Sriyder,
W. M., and Thomas, A. W. (Unpublished). A model
for runoff of pesticides from small upland water-
sheds.
3. Amerman, C. R. 1973. Hydrology and soil science
in field soil water regime. Soil Sci. Soc. Amer. Spec.
Pub. No. 5, Madison, Wis.: 167-180.
4. Baier, W. 1967. Relationship between soil moisture,
actual and potential evapotranspiration. In Soil
Moisture, Proc. Hydrology Symp. No. 6, Nat. Res.
Counc. Canada: 151-191.
14. Brustkern, R. L., and Morel-Seytoux, H. J. 1970.
Analytical treatment of two-phase infiltration. Jour.
Hydr. Div. Proc. ASCE 96(HY12):2535-2548.
15. Burwell, R. £., Allmaras, R. R., and Amemiya, M.
1963. A field measurement of total porosity and
surface micro-relief of soils. Soil Sci. Soc. Amer.
Proc. 20:697.
5. Bailey, G. W., Swank, R. R., and Nicholson, H. P.
1974. Predicting pesticide runoff from agricultural
land: A conceptual model. Jour. Environ. Qual.
3(2):95-102.
6. Baird, R. W., and Richardson, C. W. 1969. Effects
of conservation treatments on water yield. In W. L.
Moore and C. W. Morgan, eds. Effects of watershed
changes on streamflow: 69-78. Univ. Texas Press,
Austin, Texas.
7. Barnett, A. P. 1965. Using perennial grasses and
legumes to control runoff and erosion. Jour. Soil
and Water Conserv. 20(5):212-215.
8. Bear, J. 1972. Dynamics of fluids in porous media.
American Elsevier Pub. Co., New York, 764 p.
9. Boast, C. W. 1973. Modeling the movement of
chemicals in soils by water. Soil Sci. 115:224-230.
10. Bough ton, W. C. 1968. A mathematical catchment
model for estimating runoff. Jour. Hydrol. (N.Z.)
7(2):75-100.
11. Boughton, W. C. 1970. Effects of land management
on quantity and quality of available water. Report
No. 120, Water Res. Lab., Univ. New South Wales,
Manly Vale, N.S.W., Australia, 330 p.
16. Burwell, R. E., Allmaras, R. R., and Sloneker, L. L
1966. Structural alteration of soil surfaces by tillage
and rainfall. Jour. Soil and Water Conserv. 21:61.
17. Burwell, R. E., and Holt, R. F. 1967. Soil and water
losses from Barnes soil in west-central Minnesota.
Minnesota Science, Univ. Minnesota Agr. Expt. Sta.
23(4):10-12.
18. Burwell, R. E., Schuman, G. E., Piest, R. F.,
Spomer, R. G., and McCalla, T. M. 1974. Quality of
water discharged from two agricultural watersheds
in southwestern Iowa. Water Resour. Res.
10(2):359-365.
19. Byers, H. R. 1959. General meteorology. Third ed.
McGraw-Hill, New York, 540 p.
20. Carlston, C. W. 1964. Tritium-hydrologic research:
Some results of the U. S. Geological Survey Re-
search Program. Science 143(3608):804-806.
21. Carter, C. E., Dendy, F. E., and Doty, D. W. 1966.
Runoff and soil loss from pastured Loess soils in
north Mississippi. Proc. Mississippi Water Resour.
Conf., Jackson, Miss., April: 104-118.
22. Carter, C. E., Doty, C. W., and Carroll, B. R. 1968.
Runoff and erosion characteristics of the brown
loam soils. Agr. Engin. 49(5):296.
24
-------
23. Chow, V. T. ed. 1964. Handbook of applied
hydrology. McGraw-Hill, New York.
24. Cooper, C. F. 1973. Hydrologic modeling-a vehicle
for quantifying man's impact on the environment.
Trans. ASAE 16(3):578, 579, 581.
25. Crawford, N. H., and Donigian, A. S., Jr. 1973.
Pesticide transport and runoff model for agricultural
lands. EPA-660/2-74-013, 211 p. U. S. Govt. Print-
ing Off., Washington, D.C.
26. Crawford, N. H., and Linsley, R. K. 1963. A
conceptual model of the hydrologic cycle. I.A.S.H.
Publication No. 63. Symp. Surface Waters: 573-587.
27. Dawdy, D. R., Lichty, R. W., and Bergman, J. M.
1972. A rainfall-runoff simulation model for estima-
tion of flood peaks for small drainage basins. USGS
Prof. Paper 506-B, 28 p.
28. Doty, C. W., and Wiersma, J. L. 1969. Geometric
shaping and contouring of land as related to
potential for surface-water storage. Trans. ASAE
12(3):322-325,328.
29. Dragoun, F. J. 1969. Effects of cultivation and grass
on surface runoff. Water Resour. Res.
5(5):1078-1083.
35. Epstein, E., and Grant W. J. 1966. Rock and crop
management effects on runoff and erosion in a
potato-producing area. Trans. ASAE 9(6):832-833.
36. Free, G. R., and Bay, C. E. 1969. Tillage and slope
effects on runoff and erosion. Trans. ASAE
12(2):209-219.
37. Frere, M. H.,Onstad, C. A., and Holtan, H. N. 1975.
ACTMO, an agricultural chemical transport model.
ARS H-3, U.S. Dept. Agr., Washington, D.C., 56 p.
38. Gardner, W. R. 1967. Development of modem
infiltration theory and application in hydrology.
Trans. ASAE 10:379-381, 390.
39. Green, W. H., and Ampt, G. A. 1911. Studies on soil
physics: I. Flow of air and water through soils. Jour.
Agr. Sci.4:l-24.
40. Hanson, C. L., Kuhlman, A. R., Erickson, C. J., and
Lewis, J. K. 1970. Grazing effects on runoff and
vegetation on western South Dakota rangeland.
Jour. Range Manag. 23(6) :418-420.
41. Hanson, C. L. 1973. Model for predicting evapo-
transpiration from native rangeland in the northern
Great Plains. Ph.D. Thesis, Utah State Univ., Logan,
Utah,116 p.
30. Dragoun, F. J. and Kuhlman, A. R. 1968. Effect of
pasture management practices on runoff. Jour. Soil
and Water Conserv. 23(2):55-57.
31. Dunne, T. 1970. Runoff production in a humid
area. ARS 41-160, U.S. Dept. Agr., Washington,
D.C, 107 p.
32. Eagleson, P. A. 1970. Dynamic hydrology. McGraw-
Hill, New York, 462 p.
33. Edwards, W. M., and Larson, W. E. 1969. Infiltra-
tion of water into soils as influenced by surface seal
development. Trans. ASAE 12(4):463,465,470.
34. England, C. B. 1970. Land capability: A hydrologic
response unit in agricultural watersheds. ARS
41-172, U.S. Dept. Agr., Washington, D.C., 12 p.
42. Harrold, L. L., Triplett, G. B., Jr., and Youker, R.
E. 1967a. Less soil and water loss from no-tillage
corn. Ohio Agr. Res. Develop. Ctr., Ohio Rep.
52(2):22-23.
43. Harrold, L. L., Triplett, G. B., Jr., and Youker, R.
E. 1967b. Watershed test of no-tillage corn. Jour.
Soil and Water Conserv. 22(3):98-100.
44. Harrold, L. L., Triplett, G. B., Jr., and Youker, R.
E. 1970. No-tillage-characteristics of the system.
Agr.Engin.51(3):128-131.
45. Harrold, L. L. and Edwards, W. M. 1972. A severe
rainstorm test of no-till corn. Jour. Soil and Water
Conserv. 27(l):30-35.
46. Hillel, D. 1971. Soil and water. Physical principles
and processes. Academic Press, New York, 288 p.
25
-------
47. Hillel, D., and Gardner, W. R. 1970. Transient
infiltration into crust-topped profiles. Soil Sci.
109:69-76.
48. Holtan, H. N. 1961. A concept for infiltration
estimates in watershed engineering. ARS 41-51, U.S.
Dept. Agr., Washington, D.C., 25 p.
58. Jones, J. R. 1969. An objective method for estimat-
ing runoff from small rural catchments. M.E. Thesis,
Univ. of N.S.W., Sch. Civil Eng, 138 p. (cited by
Boughton, W. C. 1970.) Effects of land management
on quantity and quality of available water. Rep. No.
120, Water Res. Lab., Univ. New South Wales,
Manly Vale, N.S.W., Australia, 330 p.
49. Holtan, H. N., and Creitz, N. R. 1967. Influence of
soils, vegetation and geomorphology on elements of
the flood hydrograph. I.A.S.H. Publ. No. 85,
2(3): 756-767.
50. Holtan, H. N. and Lopez, N. C. 1971. USDAHL-70
model of watershed hydrology, ARS Tech. Bui. No.
1435, U.S. Dept. Agr., Washington, D.C., 84 p.
51. Horton, R. E. 1919. Rainfall interception. Mon.
Weather Rev. 47: 603-623.
52. Horton, R. E. 1931. The field, scope and status of
the science of hydrology. Trans. Amer. Geophys.
Union 12:189-202.
53. Horton, R. E. 1940. An approach toward physical
interpretation of infiltration capacity. Soil Sci. Soc.
Amer. Proc. 5:399-417.
54. Huff, D. D. and Kruger, P. 1967a. The chemical and
physical parameters in a hydrologic transport model
for radioactive aerosols. Proc. Internatl. Hydrology
Symp., Ft. Collins, Col., 1:128-135.
55. Huff, D. D., and Kruger, P. 1967b. A numerical
model for the hydrologic transport of radioactive
aerosol from precipitation to water supplies. Isotope
Techniques in the Hydrologic Cycle, Am. Geophys.
Union, Geophys. Monogr. Ser. No. 11: 85-96.
56. Jamison, V. C., Smith, D. D., and Thornton, J. F.
1968. Soil and water research on a claypan soil.
ARS Tech. Bui. No. 1379, U.S. Dept. Agr., Washing-
ton, D.C., 111 p.
57. Jensen, M. E. 1966. Empirical methods of estimat-
ing or predicting evapotranspiration using radiation.
Evapotranspiration and its role in water resources
management. Conf. Proc. Am. Soc. Agr. Engr. St.
Joseph, Mich., December 5-6:49-53,64.
59. Konrad, J. C., and Cain, J. M. 1973. Hydrologic and
watershed modeling for regulating water quality.
Trans. ASAE 16(3):580-581.
60. Langbein, W. B., Hains, C. H., and Culler, R. C.
1951. Hydrology of stock-water reservoirs in
Arizona. U. S. Geol. Surv. Circ. 110,18 p.
61. Larson, C. 1973. Hydrologic effects of modifying
small watersheds- Is prediction by hydrologic
modeling possible? Trans. ASAE l6(3):560-564,
568.
62. Legrand, H. D. 1965. Patterns of contaminated
zones of water in the ground. Water Resour. Res.
1(1): 83-95.
63. Leonard R. E. 1967. Mathematical theory of inter-
ception. In Sopper, W. E., and Lull, H. R., eds.,
Forest hydrology. Pergamon Press, New York:
131-136.
64. Linsley, R. K., Jr., Kohler, M. A., and Paulhus, J. L
H. 1958. Hydrology for engineers. McGraw-Hill,
New York, 340 p.
65. Lull, H. W. 1964. Ecological and silvicultural as-
pects. In V. T. Chow ed., Handbook of applied
hydrology: 6-1,6-30. McGraw-Hill, New York.
66. Mannering, J. V., and Burwell, R. E. 1968. Tillage
methods to reduce runoff and erosion in the corn
belt. ARS Agr. Inf. Bui. No. 330, U.S. Dept. Agr.,
Washington, D.C., 14 p.
67. Mannering, J. V. Meyer, L. D., and Johnson, C. B.
1968. Effect of cropping intensity on erosion and
infiltration. Agron. Jour. 60: 206-209.
68. McGuinness, J. L., and Harrold, L. L. 1971.
Reforestation influences on small watershed stream-
flow. Water Resour. Res. 7(4): 845-852.
26
-------
69. Mein, R. G., and Larson, C. L 1973. Modeling
infiltration during a steady rain. Water Resour. Res.
9(2):384-394.
70. Mickelson, R. H. 1966. Level pan system for
spreading and storing watershed runoff. Soil Sci.
Soc. Amer. Proc. 30(3): 388-392.
71. Minshall, N. E., and Jamison, V. C. 1965. Interflow
in claypan soils. Water Resour. Res. 1(3): 381-390.
72. Moldenhauer, W. C., Lovely, W. G., Swanson, N. P.,
and Currence. H. D. 1971. Effect of row grades and
tillage systems on soil and water losses. Jour. Soil
and Water Conserv. 26(5): 193-195.
81. Parzen, E. 1962. Stochastic processes. Holden-Day,
Inc., San Francisco, 324 p.
82. Philip, J. R. 1964. An infiltration equation with
physical significance. Soil Sci. 77: 153-157.
83. Philip, J. R. 1969. Theory of infiltration. Advan.
Hydrosci. 5. Academic Press, New York: 215-305.
84. Ramser, C. E. 1927. Runoff from small agricultural
areas. Jour. Agr. Res. 34(9): 797-823.
85. Rauzi, F., and Hanson, C. L. 1966. Water intake and
runoff as affected by intensity of grazing. Jour.
Range Manag. 19: 351-356.
73. Moldenhauer, W. C., Wischmeier, W. H., and Parker,
D. T. 1967. The influence of crop management on
runoff, erosion, and soil properties of a Marshall
silty clay loam. Soil Sci. Soc. Amer. Proc. 31(4):
541-546.
74. Moore, W. L., and Morgan, C. W. eds. 1969. Effects
of watershed changes on streamflow. Univ. Texas
Press, Austin, 289 p.
75. Morel-Seytoux, H. J. 1973. Two-phase flows in
porous media. Advan. Hydrosci. 9, Academic Press,
New York: 119-202.
76. Musgrave, G. W., and Holtan, H. N. 1964. Infiltra-
tion, Chap. 12 in V. T. Chow, ed. Handbook of
applied hydrology: 12-1, 12-30. McGraw-Hill, New
York.
77. Neff, E. L. 1973. Water storage capacity of contour
furrows in Montana. Jour. Range Manag. 26(4):
298-301.
78. Nelson, R. W., and Eliason, J. R. 1966. Prediction
of water movement through soils—a first step in
waste transport analysis. Proc. 21st Industr. Waste
Conf., Purdue Univ. Engr. Ext. Ser. 121, Part 2:
744-758.
79. Onstad, C. A. 1972. Soil and water losses as affected
by tillage practices. Trans. ASAE 15(2): 287-289.
80. Onstad, C. A., and Olson, T. C. 1970. Water budget
accounting on two corn cropped watersheds. Jour.
Soil and Water Conserv. 25(4): 150-152.
86. Reddell, D. L., and Sunada, D. K. 1970. Numerical
simulation of dispersion in ground water aquifers.
Hydrol. Paper 41, Colorado State Univ., 79 p.
87. Rice, W. L., and Dragoun, F. J. 1965. Effects on
runoff volume from perennial prairie grass seeded
on cultivated land. Jour. Soil and Water Conserv. 20
(2): 63-64.
88. Richardson, C. W. 1972. Changes in water yield of
small watersheds by agricultural practices. Trans.
ASAE 15(3): 591-592.
89. Richardson, C. W., and Ritchie, J. T. 1973. Soil
water balance for small watersheds. Trans. ASAE
16(1): 72-77.
90. Ritchie, J. T., and Burnett, E. 1971. Dryland
evaporative flux in a subhumid climate: II. Plant
influences. Agron. Jour. 63: 56-62.
91. Ritter, W. F. Johnson, H. P., Lovely, W. G., and
Molnau, M. 1974. Atrazine, propachlor and dia-
zinon residues in small agricultural watersheds.
Runoff losses, persistence and movement. Environ.
Sci. Technol. 8(1): 38-42.
92. Robbins, J. W. D., and Kriz, G. J. 1969. Relation of
agriculture to groundwater pollution, a review.
Trans. ASAE 12(3): 397-403.
93. Rose, C. W. 1966. Agricultural physics. Pergamon
Press, New York, 230 p.
27
-------
94. Saxton, K. E., Johnson, H. P., and Shaw, R. H.
1974a. Watershed evapotranspiration estimated by
the combination method. Trans. ASAE 17(4):
668-672.
95. Saxton, K. E., Johnson, H. P., and Shaw, R. H.
1974b. Modeling evapotranspiration and soil mois-
ture. Trans. ASAE 17(4): 673-677.
96. Saxton, K. E., and Spomer, R. G. 1968. Effects of
conservation on the hydrology of loessial water-
sheds. Trans. ASAE 11(6): 848-849,853.
97. Saxton, K. E., Spomer, R. G., and Kramer, L. A.
1971. Hydrology and erosion of loessial watersheds.
ASCE Proc. Hydr. Div., 97 (HY11):1835-1851.
98. Saxton, K. E., and Whitaker, F. D. 1970. Hydrology
of a clay pan watershed. Univ. of Missouri-
Columbia, College Agr. Expt. Sta. Res. Bui. 974,47
P-
99. Schwab, G. O., and Fouss, J. L. 1967. Tile flow and
surface runoff from drainage systems with corn and
grass cover. Trans. ASAE 10(4): 492-496.
100. Schwartz, F. W., and Domenico, P. A. 1973.
Simulation of hydrochemical patterns in regional
groundwater flow. Water Resour. Res. 9(3):
707-720.
101. Sharp, A. L., Gibbs, A. E., and Owen, W. J. 1966.
Development of a procedure for estimating the
effects of land and watershed treatment on stream-
now. Tech. Bui. 1352, U.S. Dept. Agr., Washing-
ton, D.C. 57 p.
102. Sharp, A. L., Gibbs, A. E., and Owen, W. J. 1968.
A history of the cooperative water yield proce-
dures study project. U.S. Dept. Agr. and U. S.
Dept. Int. Res. Rep. No. 403, 1077 p.
103. Smith, G. E., and Whitaker, F. D. 1974. Losses of
fertilizers and pesticides from claypan soils. Report
on Project R-801-666 prepared for Office Res. and
Monitoring, U.S. Environ. Protec. Agcy. 75 p.
104. Smith, R. E. 1972. The infiltration envelope:
Results from a theoretical infiltrometer. Jour.
Hydrol. 17(1/2): 1-21.
105. Soil Conservation Service. 1972. National engineer-
ing handbook. Section 4. Hydrology. U. S. Govt.
Printing Off., Washington, D. C.
106. Spomer, R. G., Heinemann, H. G., and Piest, R. F.
1971. Consequences of historic rainfall on western
Iowa farmland. Water Resour. Res. 7(3): 524-535.
107. Streeter, H. W., and Phelps, E. B. 1925. A study of
the pollution and natural purification of the Ohio
River. U. S. Pub. Health Bui. No. 146.
108. Sutton, 0. G. 1953. Micrometeorology. McGraw-
Hill, New York, 333 p.
109. Swartzendruber, D., Skaggs, R. W., and Wiersma,
D. 1968. Characterization of the rate of water
infiltration into soil. Tech. Rep. 5, Purdue Univ.
Water Resour. Ctr., 120 p.
110. Tennessee Valley Authority. 1972. Upper Bear
Creek Experimental Project. A continuous daily-
streamflow model. Res. Paper No. 8. Knoxville,
Tenn., 99 p.
111. Thomas, A. W., Carter, R. L, and Carreker, J. R.
1968. Soil, water and nutrient losses from Tifton
loamy sand. Trans. ASAE 11(5): 677-679, 682.
112. Thornthwaite, C. E. 1948. An approach towards a
rational classification of climate. Geogr. Rev. 38:
55-94.
113. Thornthwaite, C. E., and Hare, F. K. 1965. The
loss of water to the air. In Agricultural meteor-
ology. Meteorological Monographs, Amer. Mete-
orol. Soc., Boston, Mass.6(28): 163-180.
114. Wischmeier, W. H. 1966. Relation of field-plot
runoff to management and physical factors. Soil
Sci. Soc. Amer. Proc. 30(2): 272-277.
115. Wisler, C. O., and Brater, E. F. 1959. Hydrology.
John Wiley and Sons, New York, 408 p.
116. Woolhiser, D. A. 1973. Hydrologic and watershed
modeling state of the art. Trans. ASAE 16(3):
553-559.
117. Woolhiser, D. A. 1975. The watershed approach to
understanding our environment. Jour. Environ.
Qual.4(l): 17-21.
28
-------
118. Woolhiser, D. A. 1975. Simulation of unsteady 119. Zinke, P. J. 1967. Forest interception studies in
overland flow. Chap. 12 in K. Mahmood and V. the United States. In W. E. Sopper and H. R. Lull,
Yevjevich, eds. Unsteady flow hi open channels, eds., Forest hydrology. Pergamon Press, New
Vol. II. Water Resour. Publications, Fort Collins, York: 137-161.
Col.: 485-508.
29
-------
CHAPTER 3
CROPLAND EROSION AND SEDIMENTATION
W. H. Wischmeier
Erosion is the wearing away of the land surface by
water, wind, ice, or other geological agents. Sediment is
defined as solid material, both mineral and organic, that
has been moved from its original source by these agents
and is being transported or has come to rest on the
earth's surface (66). Sediment impairs the quality of the
water resources in which it is entrained and often
degrades the location where it is deposited. It may carry
pesticides, toxic metals, and plant nutrients absorbed on
the soil particles (25, 69).
This chapter documents technical background and
methodology for estimating and controlling cropland
sediment production. It supplements the material given
in Volume I, Sections 3.3 and 4.1. Only sediment from
cropland erosion by water was considered pertinent to
the purposes of this manual, but literature on wind
erosion control is cited (12, 60, 61, 99, 100). Observed
quantities of sediment from geological erosion and from
nonagricultural sources are cited to help portray crop-
land sediment in its proper perspective, and land
classifications pertinent to large-area appraisals of crop-
land sediment potential are reviewed. Brief overviews of
(a) existing erosion research data, (b) the mechanics of
the soil-erosion process, and (c) progressive improve-
ments in prediction equations, provide pertinent back-
ground information for erosion-control technology. The
Universal Soil Loss Equation, soil loss tolerances, and
sediment delivery ratios are reviewed as potential tools
for pollution-control planning. The major emphasis is on
discussions of erosion factors and important features of
erosion-control practices.
SEDIMENT SOURCES AND QUANTITIES
Sediment concentrations in rivers of the United
States range from 200 to 50,000 ppm, with an occa-
sional concentration as high as 600,000 ppm (21). The
amount of sediment moved by flowing water has been
reported to average at least 4 billion tons a year, with
about one billion tons reaching major streams (19).
Estimates ascribe about 30% of this country's total
sediment to geological erosion and about half of it to
erqsion of agricultural lands (77).
Geological Erosion
The erosion that occurs under natural environmental
conditions of climate and vegetation, undisturbed by
man, is called geological, natural, or normal erosion (66).
Estimates of annual rates of geologic deposition in the
United States range from less than 0.30 to 0.74 ton per
acre (38, 65). Even at such relatively low rates, a large
drainage area will produce large quantities of sediment.
The Missouri River's name attests to the turbidity of its
waters before it was discovered by Europeans. The rate
of erosion under natural vegetation reaches a maximum
where the mean annual rainfall is between 10 and 15
inches. Under higher rainfall rates, improved vegetation
inhibits erosion; under rates of less than 10 inches
sediment-entraining runoff becomes more rare (29).
Natural erosion over long geologic periods can be quite
dramatic, as evidenced by the wearing away of moun-
tains and building up of flood plains.
The more rapid erosion that is primarily a result of
activities of man is called accelerated erosion (66).
Sediment produced by accelerated erosion comes from
many sources.
Nonagricultural Sources
Some major nonagricultural sources of sediment are:
erosion from construction activities, roadside erosion,
stream channel and streambank erosion, scouring of
flood-plain land by floodflow, mining and industrial
wastes dumped into streams or left in positions suscepti-
ble to erosion, and mass wasting from landslides.
In some watersheds, the sediment that originates
from these sources may far exceed that from cropland.
31
-------
A 1969 report by the Secretary of Agriculture and the
Office of Science and Technology (19) gave the follow-
ing statistics: During road construction in Scott Run
Watershed, Fairfax County, Virginia, sediment at the
rate of about 140 tons per acre was produced at the
source, and about half of this amount was measured at a
downstream gaging station. Erosion losses at rates of 42
to 289 tons per acre per year were measured on bare
roadside cuts near Carterville, Georgia, and comparable
rates were measured on 35 road cuts in the Baltimore
area. As much as 2,000 cubic yards of sediment per
square mile of access road has been measured in
mountainous country. Sediment from construction ac-
tivities in urbanizing areas near Lake Bareroft, Virginia,
was reported equivalent to 39 tons per acre annually.
Studies in southeastern Kentucky showed that sediment
yields from strip-mined coal land can be 1,000 times
that from forested land; there are about 2.3 million acres
of strip-mined lands in the United States. Erosion is a
serious problem on at least 300,000 miles of stream-
bank.
Cropland Sediment
Cropland does not produce the greatest amount of
sediment per unit of area, but because of the large area
involved, our 437 million acres of cropland as a whole
produce more sediment than any other source. Annual
soil loss from cropland ranges from about one ton to
more than 100 tons per acre, depending on the crop
system, management practices, rainfall, soil characteris-
tics, and topographic features. A 1967 Conservation
Needs Inventory by the USDA (75, 77) showed that
about half of our country's cropland averages between 3
and 8 tons of soil loss per acre per year, 30% averages
less than 3 tons, and 20% averages more than 8 tons.
Individual states have published the adjusted inventory
data, and the reports are available from state offices of
the Soil Conservation Service, USDA (75).
Large-Area Estimates of Cropland
Sediment Hazards
In 1940, Baver (7) listed the major erosion factors as
climate, topography, vegetation, soils, and the human
factor. The principal influence of climate is the type,
amount, and temporal distribution of the rainfall. The
human factor includes such items as crop sequence, soil
and crop management, and conservation practices. Each
of these factors often varies widely within a single
watershed or land resource area. All except climate often
vary appreciably even among different fields on a single
farm. Therefore, soil-loss estimation and control plan-
ning are most effective on a local basis, by procedures
given in Volume I.
On a large-area basis, the cropland contribution to
sediment in streamflow is influenced by: the amount of
sediment produced on the cropland (gross erosion), the
density of cropland in the drainage area, and the portion
of the eroded soil that actually reaches a continuous
stream system (sediment delivery ratio).
Gross Erosion
The erosion potential on a relatively homogeneous
drainage area can be estimated by using representative
soil, cover, and topographic features to evaluate the
factors in the Universal Soil Loss Equation. Published
maps and standard land classifications also provide
helpful information for appraisals of cropland sediment
hazard on a large-area basis.
The map given in Volume I as Figure 9 shows
relative potential contributions of cropland in the
conterminous United States by major land resource
areas.
Soil survey maps are the best sources of information
on soil characteristics and associated land features. These
maps generally include classifications of erosion and land
slope. The mapped erosion class is primarily an indica-
tion of the extent of prior erosion; quantitative erosion
rates are not mapped because of their local nature. The
slope class indicates whether the land is nearly level,
gently sloping, moderately sloping, strongly sloping,
steep, or very steep, but it does not provide information
on the slope shapes and lengths.
Land resource units are geographic areas of land,
usually several thousand acres in extent, that are
characterized by particular patterns of soil (including
slope and erosion), climate, water resources, land use,
and type of farming (73).
Major land resource areas consist of geographically
associated land resource units. The 156 major land
resource areas of the 48 conterminous states were
selected as the basis for mapping hydrologic and
erosion-potential data in Volume I. Major characteristics
of the 156 individual areas are given in Agriculture
Handbook No. 296 (73).
Capability classes (27) are interpretive soil groupings
made primarily for agricultural purposes. The classifica-
tion begins with the individual soil mapping unit. A
capability unit is a grouping of soils that are suited to
32
-------
the same kinds of cultivated crops and pasture plants
and have about the same responses to systems of
management. A capability subclass is a grouping of
capability units having similar kinds of limitations or
hazards. Four kinds of limitations or hazards are
recognized: erosion, wetness, root-zone limitation, and
climate. In the broadest category of capability classifica-
tion all the soils are grouped in eight classes.
The eight capability classes were briefly described in
Chapter 1. A more detailed description is given in
Appendix II of USDA Statistical Bulletin No. 461 (77).
Generally, erosion hazards increase as the capability-class
number increases (except Class V), but it is important to
recognize that for some areas the higher classifications
are due to wetness, root-zone limitations, or climatic
limitations rather than erosion. The class and subclass
designations, together, provide information about both
the degree and the kind of limitation.
Cropland Density
Acreage data for land in each subclass of each of the
eight capability classes, by states and several land-use
classifications can be obtained from the Conservation
Needs Inventory (75). These data were used in the
development of Figures 6 through 9, Volume I. Wind-
erosion limitations were included in the capability
subclass data used for Figure 7. The other tables and
charts in the erosion sections of Volume I are for water
erosion only.
Sediment Delivery Ratio
This is the factor that adjusts the gross sediment
estimate to compensate for deposition along the path
traveled by the runoff as it moves from a field slope to a
continuous stream system. The delivery ratio will be
discussed in more detail at the last of this chapter.
CROPLAND EROSION
Erosion Research Data
Measurements of runoff and soil loss from field plots
in the United States began about 1917, in Missouri (64).
Between 1929 and 1933 the U. S. Department of
Agriculture established ten Federal-State erosion re-
search stations, in regions where the problem had
become most critical. In the next 25 years, erosion plot
studies were established at 32 more locations. Precise
measurements of precipitation, runoff, soil loss, and
related field conditions at the 42 stations in 23 states
were continuous for periods of 5 to 30 years (85). In
1960, studies were underway on 18 soils. Fundamental
studies of erosion mechanics were conducted concur-
rently and have received increased emphasis since about
1960.
In 1954, the Agricultural Research Service established
a national runoff and soil-loss data center at Purdue
University. The basic data from more than 10,000
plot-years of erosion studies at 42 research stations were
assembled, standarized in units, and transferred to
punched cards for summarization and overall statistical
analyses (79). Data from continuing studies were added
annually for analysis with the previously assembled data.
The plot studies and fundamental investigations
identified the major erosion factors and provided a
wealth of information on erosion mechanics and control.
Inherent limitations of the plot data will be pointed out
in the discussion of soil loss equations.
Field-plot rainfall simulators are now used to expe-
dite filling voids in existing plot data and field testing of
new erosion control concepts and practices. This equip-
ment can simulate the drop sizes and terminal velocities
of natural rain at common intensities, apply simulated
rainfall on several 75-foot plots simultaneously, and
apply identical storms to plots on physically separated
soils and topographies (43).
The Erosion Process
Soil erosion is a process of detachment and transpor-
tation of soil materials by erosive agents (16). It is a
mechanical process that requires energy. Much of this
energy is supplied by falling raindrops. The dead weight
of the water falling in 30 minutes of a Midwest
thunderstorm may exceed 100 tons per acre. The
billions of drops which comprise this 100-ton volume of
water strike the soil, if unprotected, at an average
velocity of nearly 20 miles an hour. The impact energy
during the 30 minutes may exceed 1,000 foot-tons per
acre (93).
When raindrops strike bare soil at a high velocity,
they shatter soil granules and clods and detach particles
from the soil mass. Splash action and shallow overland
flow transport some of the detached particles directly
down the slope and others to implement marks and
other small channels, where the more concentrated
runoff provides transportation for them. This soil
movement is called sheet erosion (6), or interrill erosion
(40). This type of erosion occurs rather uniformly over
the slope and may go unnoticed until much of the
productive topsoil has been removed. In sheet erosion.
33
-------
nearly all of the soil-particle detachment is by raindrop
impact (32, 36, 101).
The erosive potential of flowing water depends on its
velocity, depth, turbulence, and type and amount of
material it transports (77). Water moving down the slope
follows the path of least resistance and concentrates in
tillage marks, eroded flow channels, and depressions in
the natural land surface, where it gains in depth and
velocity. Erosion in these flow concentrations is directly
related to the hydraulics of the concentrated flow (40).
The concentrated runoff may remove enough soil to
form small but well defined channels, or rills. Rills are
often the first readily apparent evidence of erosion, but
tillage usually obliterates them.
Rill erosion has been defined as an erosion process in
which numerous channels only several inches deep are
formed (66), and as the erosion occurring in flow
channels (40). In rill erosion, soil particles are detached
by the shearing action of water flowing over the soil
surface and by slumping of undercut side walls and small
headcuts. The detached particles are transported by a
combination of rolling, saltation, and suspension. Parti-
cles transported by suspension may travel long distances
before being deposited on the land surface. The capabil-
ity of runoff to detach soil material is proportional to
the sheer stress raised to a power of approximately two
(17). Consequently, rill erosion increases rapidly as
steeper or longer slopes increase runoff flow depth.
Under continued rainfall, sheet erosion continues be-
tween the rills. Field soil losses are usually a combina-
tion of sheet and rill erosion, and their relative contribu-
tions to total soil loss differ with soils and surface
conditions.
When water accumulates in narrow channels and, over
short periods, removes the soil from this narrow area to
depths of 1 to 2 feet, or more, the process is called gully
erosion (66). Cully erosion produces large amounts of
sediment but can usually be prevented on cropland.
A soil's inherent ability to resist erosion by rainfall
and runoff depends on its physical and chemical
properties. Erosion control is accomplished by reducing
the mechanical forces of the water acting on the soil
particles or by increasing the soil's resistivity to erosion,
or both.
Soil Loss Equations
soil properties, topographic features, and numerous
management details occurred at different levels and in
different combinations in the various studies.
Plot data predict specific-field soil losses only if the
influence of each of the major contributing parameters
can be isolated and evaluated relative to the level at
which the parameter was present in the study, so that
the various influences can be combined in different
proportions to simulate other situations. However, ef-
fects of rainfall characteristics and soil properties cannot
be isolated in a one-location study, where rainfall and
soil are either constant for the plot series or vary in
unison. Also, many relevant secondary variables cannot
be controlled in plot studies. Some of these vary
randomly over time. Some differ with seasons, and
others', such as rainfall distribution and storm character-
istics, show long-term trends at a given location but
fluctuate unpredictably for short time periods. The
uncontrolled variables interact with controlled variables,
and these interactions can substantially bias brief-period
research results. Assembling all the available erosion
research data at one location for overall statistical
analyses (79) counteracted many of these limitations. It
enabled combining basic data from various locations in
analysis designs capable of providing in formation on the
major factor effects individually and on some of the
most important interaction effects. It also helped mini-
mize bias of results by random variables.
Mathematical relationships were derived whose basic
and theoretical validity has been substantiated by
subsequent fundamental research. When these factor
relationships are combined in a general soil loss equa-
tion, planners can determine what the average annual
soil loss rate and the potential soil loss reductions from
various alternative crop and management systems are
likely to be at specific locations other than that of a plot
study.
The most accurate soil loss equation that is now
field-operational is the Universal Soil Loss Equation.
This equation has been used as an erosion-control
planning tool for more than a decade in the 37 states
east of the Rocky Mountains and is now used to a more
limited extent also in the Western States, Hawaii, and
several foreign countries. However, the following brief
overviews of four soil loss equations are pertinent to the
subsequent discussion of erosion factors.
The literature of the past 40 years includes many
reports of local erosion studies. These reports may
appear to a casual reader as inconsistent, and sometimes
incompatible, because of wide differences in the re-
ported results. However, most of these differences can
be accounted for by the fact that the rainfall pattern,
The Slope-Practices Equation
This initial soil loss equation was developed gradually
in the early I940's. Zingg (102) developed factors for
the effects of length and steepness of slope. Smith (62)
added crop and conservation practice factors and the
34
-------
concept of a limiting annual soil loss. Browning and
coworkers (10) proposed soil-erodibility and manage-
ment factors for Iowa, but their work was not published
until 1947. With the cooperation of program leaders in
the North Central Region of the Soil Conservation
Service, these initial developments were combined in the
Slope-Practice Equation for use throughout the Corn
Belt.
This equation used several dimensionless factors to
adjust an initial basic soil loss to specific field condi-
tions. Its basic soil loss was the average annual loss from
corn-oats-meadow rotations on research plots in the
North Central States. Factors for other crop systems
were estimated relative to this rotation. The equation
had no rainfall factor, and its soil factor was expressed
relative to 1.0 for Marshall silty clay loam. Zingg's slope
length and steepness exponents (0.6 and 1.4) were used
to adjust the soil-loss computations to field slope
dimensions.
The Musgrave Equation
In 1946, a national committee, with G. W. Musgrave
as chairman, was assembled in Ohio to reappraise the
factors in the Slope-Practice Equation and add a rainfall
factor. The modified model became known as the
Musgrave Equation (48). A graphical solution of the
equation was published in 1952 for the Northeastern
States (35).
The 1.75 power of the 2-year, 30-minute rainfall was
adopted as the rainfall factor, and Zingg's slope-length
and percent-slope exponents were lowered to 0.35 and
1.35, respectively. Annual cover factors were estimated
relative to a value of 1.0 for either continuous fallow or
continuous rowcrop. A quantitative soil factor was
derived by adjusting annual soil losses for effects of
rainfall, slope and cover. Subsequent research did not
confirm the adequacy of 2-year, 30-minute rainfall as an
index of local differences in rainfall erosivity. The
lowered slope-length factor was compatible with some
early sets of data but too low for others. Numerous plot
studies showed that continuous fallow and continuous
rowcrop are not interchangeable and that the cover
effect of continuous rowcrops is highly variable.
The Musgrave Equation has been widely used for
estimating gross erosion from large heterogeneous water-
sheds. Its highly generalized factor values are more easily
assigned to broad areas than are factors based on more
specific descriptions of the erosion-influencing parame-
ters. However, erosion hazards are highly localized. For
resource-conservation and pollution-control planning,
soil loss equations need to reflect local conditions as
accurately as possible.
The Universal Soil Loss Equation (USLE)
The Universal Soil Loss Equation (80, 94, 95),
developed in 1958, overcame many of the deficiencies of
its predecessors. Its form is similar to that of the
Musgrave Equation, but the concepts, relationships and
procedures underlying the definitions and evaluations of
the erosion factors are distinctly different (see section
on Erosion Factors). The major improvements (84)
included:
1. More complete separation of factor effects so that
results of a change in the level of one or several
factors can be more accurately predicted.
2. An erosion index that provides a good estimate of
the erosive potential of rainfall and its associated
runoff.
3. A quantitative soil-erodibility factor that is evalu-
ated directly from research data without refer-
ence to any common benchmark.
4. An equation and nomograph capable of comput-
ing the credibility factor for numerous soils from
soil-survey data.
5. A method of including effects of interactions
between cropping and management parameters.
6. A method of incorporating effects of local rainfall
pattern and specific crop cultural conditions in
the cover and management factor.
The Universal Soil Loss Equation computes average
annual soil loss as the product of two quantitative
factors (soil-erodibility and rainfall-erosivity) and four
qualitative factors (96). The equation is:
A=RKLSCP
where A is the average soil loss, in tons per acre, for the
time period used for factor R (usually average annual).
R is the rainfall and runoff erosivity index.
K, the soil erodibility factor, is the average soil loss in
tons per acre per unit of R, for a given soil on a
"unit plot" which is defined as 72.6 feet long,
with 9% slope, continuously fallowed, and tilled
parallel to the land slope.
L, the slope-length factor, is the ratio of soil loss
from a given length of slope to that from a
72.6-foot length with all other conditions identi-
cal.
S, the slope-steepness factor, is the ratio of soil loss
from a given percent-slope to that from a 9% slope
with all other conditions identical. (In practice,
factors L and S are usually combined in a single
topographic factor denoted by LS.)
C, the cover and management factor, is the ratio of
the soil loss with specified cover and agronomic
35
-------
practices to that from the fallow condition on
which factor K is evaluated.
P, the practice factor, is the ratio of soil loss with
supporting practices such as contouring or strip-
cropping to that with straight-row farming up and
down the slope.
The concepts and relationships underlying the evalua-
tions of these factors are reviewed in the discussion of
Erosion Factors.
Basic Erosion Models
Basic mathematical models are being developed that
combine fundamental principles, concepts and relation-
ships of erosion mechanics, hydrology, hydraulics, soil
science, and meteorology to simulate the erosion and
sedimentation processes. Substantial progress has been
made in developing static and dynamic models capable
of predicting spatial and temporal variations in erosion
and sedimentation (14, 17, 40, 52). To the extent that
these simulation models reflect direct and interacting
effects of more of the uncontrolled and secondary
variables, they will enhance analyses of erosion systems
and control practices. These models have not become
field operational because additional research is needed to
bridge certain information gaps. However, they have
already improved the understanding of erosion proc-
esses, helped explain some of the seeming inconsistencies
in the field-plot data, and improved the accuracy of
some of the factor evaluations for the USLE.
The initial basic models have added several important
new concepts. One is the treatment of soil detachment
by rainfall, detachment by runoff, and transport by
runoff, as individual subprocesses that bear substantially
different relationships to the erosion factors and that
occur in widely differing combinations (39, 44). Either
detachment capacity or transport capacity can limit
erosion at a given site. Another new concept is the
separation of rill erosion from interrill erosion (17, 40).
This distinction will help clarify unexplained differences
in the credibilities of soils and effectiveness of crop
canopies. Some soils allow very substantial sheet erosion
without rilling; others are much more susceptible to
rilling.
EROSION FACTORS
The climatic, soil, topographic, and management
parameters that largely determine erosion rates have
wide ranges of possible values, or levels, that can occur
in any of an extremely large number of possible
combinations. The six major erosion factors discussed in
this section estimate the effects of different levels of
these parameters on soil erosion by water. In a soil loss
equation, each factor must be represented by a number
that reflects the specific local conditions, and all the
numbers must be relative to the same, clearly defined,
benchmarks. The benchmark conditions for the Uni-
versal Soil Loss Equation are free of geographic bounds
and are defined as follows.
The benchmark management condition is continuous
fallow that receives primary and secondary tillage each
spring and is periodically tilled during the summer to
prevent vegetation and serious crusting. The tillage
operations are up and down the slope. This condition
was selected because: (a) continuous fallow is the only
condition under which soil effect could be evaluated
independently of cover, management, and residual ef-
fects, and (b) it is a more constant condition than would
exist with any type of cropping. The fact that this
condition rarely exists in practice is immaterial because
the soil loss computed by the equation as a whole does
reflect existing field conditions.
The slope length of 72.6 feet was selected as a
benchmark because most of the erosion research plots
since 1930 were of this length. It is sufficient for
measurement of runoff effect as well as raindrop-impact
effect. Slope steepness of 9% was the most representa-
tive for the existing plot data. Straight-row fanning up
and down the slope represents complete absence of
support practices. The "unit plot" on which the quanti-
tative soil factor is measured has these benchmark
conditions and factors L, S, C, and P have values of 1.0.
The values of factors R, K, L, and S are essentially
firm for a particular location and, together, determine
the location's characteristic erosion potential. The
farmer or planner has no control over rainfall pattern or
steepness of the slope. The effective slope length can be
reduced by use of terraces or diversions, but this
reduction can be classified as a practice effect. Manage-
ment systems that gradually improve soil structure and
increase its organic-matter content can affect its credi-
bility, but an appreciable change in the soil factor would
require many years. Factors C and P, on the other hand,
are highly responsive to executed management decisions.
Good management and erosion control practices reduce
sediment production primarily through their effects on
these two factors. The following discussions of the six
major erosion factors include the concepts and relation-
ships underlying their definition and evaluation for the
USLE.
36
-------
Rainfall and Runoff Erosivity (Factor R)
Most cropland erosion by water is directly associated
with rain events and is influenced both by the rain
intensities and by the amount and rate of runoff. The
function of factor R is to quantify these interrelated
erosive forces. The parameter used to evaluate R must be
predictable on a probability basis from meteorological
data. It must be definable for specific storms and for
specific time periods other than annual, and its seasonal
or annual evaluation must be influenced by all signifi-
cant rains rather than only by annual maxima.
The Rainfall-Erosion Index, El
The assembled plot data showed that when all factors
other than rainfall are constant, storm soil losses from a
cultivated field are directly proportional to an inter-
action term, which is the product of the rainfall energy
and the maximum 30-minute intensity. This product is
the El parameter (80, 93). The relation of soil loss to El
is linear; therefore, individual-storm values of El can be
summed to obtain seasonal or annual values of the
parameter. Frequency distributions of annual, seasonal,
or annual-maximum-storm El values follow the log-
normal type of curve that is typical of many hydrologic
data(S0).
Median raindrop size increases as rain intensity
increases, to about 3 in/hr, and terminal velocities of
free-falling waterdrops increase with increased drop size
(22, 33). Since the kinetic energy of a given mass in
motion is proportional to velocity squared, rainfall
energy is directly related to rain intensity. Analyzing
published dropsize and terminal-velocity data,
Wischmeier and Smith (93) derived the equation E = 916
+ 331 logioi, where E is the kinetic energy in foot-tons
per acre-inch of rain, and i is intensity in inches per
hour. The energy of a rainstorm can be computed from
recording-raingage data. The storm is divided into
successive increments of essentially uniform intensity,
and a rainfall energy-intensity table (93) derived from
the above formula is used to compute the energy of each
increment. Thus, the energy of a rainstorm is a function
of all its component intensities and rain amount.
In exploratory analyses of data from bare fallow
plots, rainfall energy was the best single predictor of
associated runoff, but was not a good predictor of soil
loss. For sheet erosion, soil detachment is primarily by
raindrop impact on the surface, but the capacity of the
associated runoff to detach and transport soil material is
directly related to its depth and velocity. These are
directly related to the maximum prolonged intensity of
the storm. Therefore, the erosive potential of a rain-
storm is a function of its kinetic energy, maximum
prolonged intensity, and their interaction, all three of
which are reflected in the El parameter.
The published rainfall energy-intensity table (84, 93)
applied the equation given above to intensities up to 10
in/hr. Two recent studies showed that median drop size
does not continue to increase when intensities exceed
about 3 in/hr (11, 26). Therefore, the energy given in
the table for a 3 in/hr intensity should be used for all
higher intensities as well. This change does not signifi-
cantly affect El computation in the United States
because prolonged intensities greater than 3 in/hr are too
rare to have much effect on average annual El values.
For computation of average annual El values, contin-
uous records of from 20 to 22 years are desirable in
order to avoid bias by cyclical variations in rainfall
pattern (49). Erosion index values were computed for
about 2,000 locations fairly uniformly distributed over
the 37 states east of the Rocky Mountains. By interpo-
lating between the computed point values, lines of equal
value (iso-erodents) were plotted on a map that included
county lines as references (82, 96). The mapped values
represent 22-year rainfall records (1937-1958). At sta-
tions where 40-year records were available, the 40-year
average annual rain amounts generally coincided very
closely with the corresponding averages for the 22 years
used in development of the iso-erodent map.
The computed annual El values are reasonably well
correlated with the 2-yr, 6-hr rainfall probabilities
published by the Weather Bureau (74). The relationship
is expressed by El = 27.38P2-17, where P = the 2-yr,
6-hr rainfall (87). The El values given in Figure lOa,
Volume I, for the 11 Western States were estimated by
this equation. Those for the other 37 states were taken
from the original iso-erodent map (#2).
Factor R in the USLE usually equals the pertinent El
value. For prediction of average annual soil loss, it is the
annual-El value available from Figure lOa; for short
specific time periods, it is the actual local El for that
period. However, there are two conditions for which the
computed El must be modified to evaluate factor R.
1. Where snowmelt runoff on moderate to steep
slopes is significant, the El value must be adjusted
upward to add the erosive effects of this runoff to
the R value. The Palouse Region of the Northwest
exemplifies this condition. Numerical evaluation
of the erosivity of runoff that is not an immediate
consequence of rainfall is an area of needed
research. Only tentative estimates of the adjust-
ment factor for the Palouse Region are presently
available.
37
-------
2. Experience has shown that on the Coastal Plains
of the Southeast, factor R is less than the El
values computed by the standard procedure. This
discrepancy may be due to the combination of
hurricane-associated storms and flat slopes. The
hurricane storms compute very high El values,
but the gentle slopes are soon largely covered by
very slowly moving runoff that shields the soil
surface from raindrop impact. In a study on a
similar soil and slope in the Maumee Basin of
Northeastern Indiana, using a rainfall simulator
and inflow at the upper end of the plot, drop
impact on the soil surface was needed to obtain
significant soil loss from a 35-foot plot (36). A
maximum of 350 for El values in the Southeast
was recently adopted as a temporary measure
until research can provide "effective El" values
for these conditions.
Runoff
Surface runoff is not a separate factor in the
Universal Soil Loss Equation or its predecessors because:
(a) no satisfactory prediction equation for cropland
runoff existed, and (b) the respective roles of rainfall
and runoff in the erosion process had not been separated
in erosion research. Runoff data alone do not predict
soil loss. The sediment content of an acre-inch of runoff
can range from a mere trace to many tons. For soil loss
prediction, the factors in the USLE would need to be
combined with the runoff factor, and the runoff would
first need to be predicted as a function of essentially the
same parameters. Therefore, it was advantageous to
relate the factors directly to soil loss in an equation for
widespread field use. The El parameter combines esti-
mates of runoff amount and rate with the potential of
the rainfall to detach soil material by drop impact and
splash action.
Researchers have recently made good progress in
separating rainfall-induced (interrill) erosion from run-
off-induced (rill) erosion (17, 40). With this separation, a
runoff factor added to the soil loss equation should have
substantial potential for improved accuracy. An equa-
tion that predicts the two types of erosion as separate
components of the total soil loss could largely solve the
aforementioned problems with factor R in the North-
west and Southeast. Also, some erosion-control practices
greatly reduce soil loss without appreciable effect on
runoff. Onstad and Foster (52) obtained good results
from adding a runoff factor to the USLE when using the
equation to route sediment through a watershed. How-
ever, more research is needed to make this approach
field operational.
Soil Erodibility (Factor K)
The susceptibility of a given land area to erosion is a
function of all the factors in the soil loss equation, but
some soils erode more readily than others even when
rainfall, topography, cover and management are identi-
cal. Soil credibility refers to a soil's inherent suscepti-
bility to erosion by rainfall and runoff. This is a function
of complex interactions of soil physical and chemical
properties. Numerous researchers have measured differ-
ences in the credibilities of a few soils, and some have
related credibility to specific soil properties (5, 10, 34,
46, 50, 51, 55, 76, 91). Water infiltration into soils was
reviewed by Pan and Bertrand (54).
The relation of soil loss to El is linear, and the
average increase in soil loss for each additional unit of El
differs for different soils (93). The average soil loss per
unit of El, measured under the previously defined "unit
plot" conditions, is the numerical soil-erodibility factor
of the USLE. For 23 benchmark soils for which K was
measured in long-term plot studies under natural rain, its
value ranged from 0.03 to 0.69 (50, 96).
Rainfall simulators were used in the Corn Belt, the
Southeastern States, and Hawaii to evaluate other soils
and obtain soil loss data for study of the relationships of
various soil properties to credibility (5, 89, 91). In a
Corn Belt study of about 60 soils selected to include a
broad range in soil properties, 24 primary and inter-
action terms were statistically significant in multiple
regression analysis of the data (91). This illustrates the
complexity of the problem, but for practical purposes
many of these terms can be neglected either because of
relatively small effect or because they are closely related
to particle-size distribution, organic-matter content, soil
structure or permeability.
Two recent findings were particularly helpful for
simplifying the prediction of inherent soil credibility:
(a) that from the viewpoint of erodibility, very fine sand
(0.05 - 0.10 mm) would be more properly classified as
silt than as sand (91), and (b) that percentages of sand,
silt and clay must be considered in relation to each
other, because of strong interaction between particle
sizes. The most informative particle-size parameter in the
Corn Belt study was M = % silt(100 - % clay), where the
very fine sand is included in the silt fraction. When this
parameter was included with organic-matter content, a
soil structure index, and the profile permeability class,
prediction of the erodibility factor was well within the
accuracy needed for field use (89). The equation is:
K = (2.1 x 10-*) (12 -On^M1 •'4 + 0.0325(8 - 2)
+ 0.025(P-3),
38
-------
where Om = percent organic matter, M = the particle-size
parameter presented above, S = structure index, and P =
permeability class (92). Permeability is a profile para-
meter; the other three pertain to the upper few inches of
soil.
The soil-erodibility nomograph presented in Volume
I, Figure 11, provides a quick graphic solution to this
equation. However, the relationship changes when the
silt fraction exceeds 70%. This change is reflected in the
nomograph by the bend in the percent-sand curves, but
is not reflected in the above equation. The structure-
index and permeability-class codings are defined on the
nomograph (89).
For a few special conditions, the nomograph solution
may be modified to improve K-value accuracy: (1)
Fragipans and claypans reduce permeability in wet
seasons, but do not greatly reduce it for thunderstorms
that occur when soil is relatively dry. Separate credi-
bilities can be computed for dry and wet seasons by
using different permeability ratings in the nomograph
formula. (2) The mulching effects of stone, gravel, or
shale on the surface are not accounted for in the
nomograph equation. If used on such soils, it would be
applied to mechanical-analysis data for the soil exclusive
of the large fragments, and the indicated K value would
then be reduced by treating the large fragments as partial
mulch cover. (3) The nomograph lacks sensitivity to
differences in erodibilities of desurfaced high-clay sub-
soils, because other chemical properties become impor-
tant under those conditions. Recent studies showed free
iron and aluminum oxides were important for high-clay
subsoils but not for most topsoils (58).
Standard texture classes are too broad to be accurate
indicators of erodibility. Therefore, the K values listed in
Table 2a of Volume I are only first approximations.
Nomograph solutions will show a broad range of
erodibilities within a texture class.
Topographic Features (Factors L and S)
Soil loss per unit area increases as slopes become
longer or steeper. The USLE denotes effects of slope
length by L and effects of steepness by S. Both are
dimensionless and expressed relative to the "unit plot"
dimensions defined for factor K. In practice, the two are
combined in a single topographic factor denoted by LS.
Slope length is the distance from the point of origin
of overland flow to the point where either the slope
decreases enough that deposition begins, or the runoff
water enters a well-defined channel (63). The effect of
slope length on runoff per unit area is generally not of
practical significance, although there have been instances
of statistically significant direct and inverse relationships
(83). Neither is soil detachment per unit area by
raindrop impact greater on long slopes. The effect of
slope length is, therefore, primarily due to greater
accumulation and more channelization of runoff on the
longer slopes. This increases the capability of the runoff
to detach and transport soil material.
Factor L in the USLE is dimensionless. For slopes
steeper than 4% it is generally computed by the formula
L = (A/72.6)0-5, where A = slope length in feet and 72.6
feet is the benchmark length. The exponent of 0.5 is the
average of values obtained in 10 independent studies in
which the observed values ranged from 0.3 to 0.9 (97).
Field observations indicate that the exponent is prob-
ably about 0.3 for slopes of less than 3%, and 0.4 for 4%
slopes. Increasing the exponent to 0,6 when slopes
exceed 10%, as suggested in Agriculture Handbook No.
282, is of questionable validity. The higher exponents
observed in the length-effect studies were associated
with plowed-out bluegrass sod or abnormally severe rain
events, on slopes that did not exceed 10%. Both L and S
are believed to be influenced by density of cover, soil
erodibility, and rainstorm characteristics, but existing
data are inadequate for mathematical evaluations of
these interaction effects.
There have been field indications that the slope-
length exponent becomes smaller for extremely long
slopes. This is logical because slopes approaching a
thousand feet in length would rarely have a constant
slope steepness along their entire length, and upslope
depositional areas would be likely.
Slope steepness affects both runoff and soil loss. In
the assembled plot data, runoff from small grain tended
to increase linearly with increases in slope. For row
crops the increase was curvilinear, increasing at an
increasing rate (83). Soil loss increases more rapidly than
runoff as slopes steepen.
The combined effects of length and steepness for
uniform slopes were shown in Table 3, Volume I. The
table was derived by the formula
LS =
72.6
430 sin2 6 + 30 sin 6
6.574
+ 0.431
where m = 0.5 if the slope is steeper than 4%, 0.4 for 4%
slopes, and 0.3 for slopes of 3% or less; and 6 = the angle
of slope.
The last quantity in this equation is an unpublished
conversion of an earlier formula (96) to an expression in
terms of the sine of the angle of slope. Within the range
of the research data, the two forms are equally accurate,
but an expression in terms of sin 0 is more logical and
39
-------
computes more realistic values when extrapolated to
steeper slopes.
The research data used to derive relationships of slope
length and steepness to soil loss were from plots not
longer than 270 feet and slopes not steeper than 18
percent. The extrapolated values shown in Table 3 of
Volume I for slopes that exceed these dimensions,
although speculative, are the best estimates presently
available. Soil loss estimates for slopes steeper than
about 30% are potentially subject to considerable error.
Research on steep slopes is a major need.
Shape of the slope is also important. \vhen a slope
steepens or flattens significantly toward the lower end,
or is composed of a scries of convex and concave
segments, its overall average gradient and length do not
correctly indicate the topographic effect on soil loss. An
irregular slope can be viewed as a series of segments such
that the gradient within each segment can, for practical
purposes, be considered uniform. The segments cannot
be evaluated as independent slopes when runoff flows
from one segment to the next. However, the amount of
soil detached on each segment can be computed by a
recently published formula (18) and summed for the
entire slope length. For each segment, the effective slope
length is the distance from the top of the overall slope to
the foot of the particular segment.
If the segments are selected so that they are also of
equal length, the slope-effect table for uniform slopes
can be used with appropriate adjustment factors for
position of the segment on the overall slope. For most
field slopes, three segments should be sufficient. The
procedure is as follows (87):
Ascertain the percent slope for each segment. Enter
the slope-effect chart or table with the total slope length
and read the LS value corresponding to the steepness of
each of the three segments. Multiply the chart LS value
for the upper segment by 0.58, the middle-segment value
by 1.06, and the lower-segment value by 1.37. The
average of the three products is a good estimate of the
effective LS value for that slope. The three products also
indicate the relative magnitudes of soil loss on the three
slope segments. (If two segments are sufficient, use the
multiples: 0.71 and 1.29. For four segments: 0.50,0.91,
1.18, and 1.40. For five segments: 0.45, 0.82, 1.06,
1.25, and \A2.)(87).
Cover and Management (Factor C)
The ability of a soil to resist the erosive forces of
rainfall and runoff is profoundly influenced by the
direct and residual effects of vegetation, crop sequence,
management, and agronomic erosion-control practices.
The effects of cropping and management must be
estimated in combination, because of many interrelated
variables. Nearly any crop can be grown continuously or
in any one of numerous rotations. The sequence within a
system can be varied. Crop productivity can be low, or it
can be high. Crop residues can be removed, left on the
surface, incorporated near the surface, or plowed under.
The amount of residues can vary from scattered pieces
to complete surface cover. The crop can be planted in a
pulverized and smoothed seedbed, in a rough and cloddy
seedbed, or with extremely little soil disturbance. It can
be intertilled after emergence, or the weeds can be
controlled with chemicals. The effectiveness of crop-
residue management will depend on the amount of
available residue. This, in turn, depends on the rainfall
distribution, the fertility level, and various management
decisions made by the farmer. Also, the residual effect
of meadow sod depends on the type and quality of
meadow, on how the succeeding seedbed was prepared,
and on the length of time elapsed since the sod was
turned under. The erosion-reducing effectiveness of a
crop system depends on how the levels of all these
variables, and others, are combined on the field.
Factor C in the Universal Soil Loss Equation is the
ratio of soil loss from land cropped under specified
conditions to the corresponding loss from clean-tilled,
continuous fallow (96) and therefore includes the effects
of all these variables. If the actual soil loss equals the
potential loss predicted by the product of factors R, K,
L, and S, factor C=l. This would be clean-tilled
continuous fallow or land where mechanical desurfacing
has removed all of the surface vegetation and most of
the root zone. Where there is any vegetative cover, where
the upper layer of soil contains significant amounts of
roots or plant residues, or where cultural practices
increase infiltration and reduce velocity, soil loss is less
than the product RKLS. Factor C brings this reduction
into the soil loss computation. On cropped land, C
ranges from about 0.60 downward to less than 0.01.
This great flexibility in the value of C is extremely
important to erosion-control planners. If C is reduced,
soil loss is reduced by the same percentage.
The canopy protection of crops varies widely for
different weeks or months in the crop year. The overall
erosion-reducing effectiveness of a crop depends on how
much of the erosive rain falls while the crop provides the
least protection. The correspondence of periods of
highly erosive rainfall with periods of good or poor
vegetative cover differs appreciably between geographic
regions. Therefore the C value for a particular crop
system will not be the same in all parts of the country. A
field-tested routine is available for computing site
C-values that reflect the net effect of the interrelated
crop and management variables in whatever combination
40
-------
they occur at the site and in relation to the local rainfall
pattern.
The entire rotation cycle is divided into a series of
cropstage periods so defined that cover and management
may be considered approximately constant within each
period. The five cropstage periods are defined as follows,
for each crop-year in the system (57, 96):
Period F - Rough fallow. Turn plowing to final
seedbeed preparation. (No-plow systems omit this
period.)
Period 1 - Seedling. Seedbed preparation to 1 month
after seeding.
Period 2 - Establishment. The second month after
spring or summer seeding. For fall-seeded grain,
this period extends to about May 1 in the
Northern States, April 15 in the Central States,
and April 1 in the Southern States.
Period 3 - Developing and maturing crop. End of
period 2 to harvest.
Period 4 - Residue or stubble. Crop harvest to
plowing or new seeding. (When meadow is seeded
with small grain, period 4 ends about 2 months
after grain harvest. The vegetation is then classified
as established meadow.)
Probable calendar dates for the events that begin the
successive periods are selected on the basis of local
climate and farm practice. The fraction of the annual El
that normally occurs in that locality during each
cropstage for each of the crops in the rotation is
determined from the applicable El-Distribution Curve in
Agriculture Handbook No. 282. These fractions are
multiplied by the corresponding soil loss ratios from
Table 2 in the same handbook. The sum of the products
obtained for the cropstage periods in any one year is the
C-value for that particular crop in that system. The
crop-system C-value is the sum of all the partial
products, divided by the number of years in the system.
This procedure has been used by the Soil Conservation
Service to develop local C-value tables that are available
from their state offices. The illustrative C values given in
Table 4 of Volume I were also derived by this procedure,
but the seeding and harvest dates and the El-distribution
data were generalized and are not precise for any
particular location.
The 33 regional El-distribution curves in Agriculture
Handbook No. 282 were derived from the 22-year
rainfall records used to develop the iso-erodent (R-value)
map (82.) Corresponding data for the 11 Western States
and Hawaii are presently available only as tentative
estimates.
The soil-loss-ratio table (96) was derived from analy-
sis of more than 10,000 plot-years of erosion data. The
data in this table are percentages of soil loss from the
indicated combinations of cover and management to
corresponding losses from continuous fallow. The table
has limitations that need to be recognized. The "mini-
mum tillage" classification applies only to plow-based
systems in which disking and smoothing are omitted. A
partial list of ratios for no-plow systems that retain some
or all of the residues on the surface was published in
1973 (86). The ratios for corn in cropstage 4, residues
left, are for stalk cover as left by the picker. Shredding
the stalks provides more complete cover and reduces the
soil-loss ratio. Some of the crop systems in the Western
States and Hawaii are not represented, but approximate
values for these systems are now available from the Soil
Conservation Service, Western Technical Center, Port-
land, Oregon. Approximate C-values for range, wood-
land, and idle land were published in 1974 (87).
Practices that depend on small rates of residue and/or
tillage-induced surface roughness for erosion-control
effectiveness will be ineffective if slopes are excessively
long. The precise length limits for various mulch rates,
slope steepnesses, kinds of soil, and rainfall patterns have
not been determined. Investigations by the Agricultural
Research Service are underway to improve or verify the
approximate limits given in Table 13, Volume I.
Supporting Practices (Factor P)
This factor is similar to C except that P accounts for
additional effects of practices that are superimposed on
the cultural practices, such as contouring, terracing,
diversion, and contour stripcropping. Approximate
values of P, related only to slope steepness, were listed in
Table 5, Volume I. These values are based on rather
limited field data, but factor P has a narrower range of
possible values than the other five factors. Influences of
type of vegetation, residue management, rainstorm
characteristics, and soil properties on the value of P have
not been evaluated to the point of predictability.
EROSION CONTROL METHODS
Specific types of erosion-control practices were dis-
cussed in Section 4.1 of Volume I. These discussions
included general information on the advantages, limita-
tions, and variability of each type of practice. The
indicated percentages of reduction in soil loss were based
on C values estimated from all the available data rather
41
-------
than on results of any specific local experiment. This
section discusses principles and relationships that deter-
mine the effectiveness of erosion-control practices.
If surface runoff can be eliminated, movement of
sediment from a field will be insignificant. Breakdown of
soil aggregates by raindrop impact, rearrangement of soil
particles, and surface sealing can occur before runoff
begins, but very little sediment will leave the field unless
surface runoff is available to transport it. Land treat-
ments that result in a deep, fertile topsoil, a high level of
organic matter, good tilth, and good vegetative cover
increase infiltration and reduce runoff. These conditions
may completely eliminate surface runoff from moderate
rainstorms on some areas. Generally, however, where the
rainfall is adequate for crop production some of it falls
at intensities greater than the soil can infiltrate even
when well managed, and runoff occurs.
Land treatments that increase infiltration and the
capacity of the soil to store water will reduce small-
watershed flooding that results from short, intense rains
during the growing season. However, when the soil
becomes saturated to a considerable depth, as is often
the case in major flood periods, cultural practices have
much less effect on runoff. Erosion-control practices
must also reduce the shear stress and transport capacity
of the runoff. This means reducing runoff amounts,
velocities, and depths and dissipating the flow energies
on plant residues rather than on the soil surface.
Erosion-control practices rely primarily on five means
of reducing erosion: 1) vegetation, 2) plant residues, 3)
improved tillage methods, 4) residual effects of crops in
rotation, particularly systems that include grass and
legume meadow, and 5) mechanical supporting practices.
A sixth potential approach would be use of chemical soil
stabilizers, but they have not yet become economically
feasible for field use.
Vegetation
Vegetation (a) intercepts rainfall and thereby reduces
runoff and soil-particle detachment by drop impact, (b)
increases the soil's water-storage capacity through tran-
spiration, (c) retards erosion by decreasing runoff
velocity, (d) physically restrains soil movement, (e)
improves aggregation and porosity of the soil, and (0
in eases biological activity in the soil (59).
Crop Canopies
Leaves and branches that are not in contact with the
soil reduce runoff from small rains but have relatively
little influence on the amount and velocity of runoff
from prolonged rains. In 542 plot-years of convention-
ally planted corn, the average runoff per thousand
foot-tons of computed rainfall energy was only about 15
percent less in cropstage 3 than before canopy had
developed (90). But soil loss per El unit from a field of
clean-tilled 90-bushel corn is about 60 percent less in
cropstage 3 than in cropstage 1 (96, Soil Loss Ratio
Table), primarily as a result of raindrop interception by
the canopy. Water drops from canopy may regain
appreciable velocity, but usually not the terminal veloci-
ties of free-falling raindrops. Therefore, canopy reduces
rainfall erosivity by reducing its impact energy at the soil
surface. The amount of reduction depends on its height
and density. Canopy effect can be viewed as a reduction
in the "effective" El of the rainstorms and as such can
be directly computed for specific situations.
Figure 1 shows the ratios of effective El's computed
for several drop fall heights to the El of unintercepted
rainfall (88). Percent cover was defined as the percentage
of the total ground area that could not be hit by
vertically falling raindrops because of the canopy. Soil
loss reductions due to canopy over a bare soil should be
approximately proportional to the reductions in effec-
tive El. Figure 1 assumes a median dropsize of 2.5 mm
for both the rain and the droplets formed on the
canopy. Where rainfall is characteristically of low inten-
sity and small drops, canopy effect would be less.
All crops develop some canopy, but this may require
several months, and most or all of it may be lost with
the crop harvest. Good soil-fertility management and
narrow row spacing hasten the development of a
protective canopy. Crop sequences can be selected that
substantially reduce the length of time between succes-
sive plant covers, and early seeded winter cover crops
can provide interim cover.
Vegetation At The Soil Surface
Stands of grass or small grain are much more effective
than a raised canopy. Much of the rain that such
vegetation intercepts moves down the blades and stems
to a point so near the ground that the droplets regain no
appreciable energy. The dense vegetation at the soil
surface also reduces runoff amount and velocity and
physically restrains soil movement. For about 5,000
plot-years of data, runoff from small grain averaged 9
percent of total precipitation, in contrast to 12 percent
for row crops (83). Meadow averaged 7 percent. The
soil-loss-ratio table shows that soil loss from established
small grain averages less than half of that under a canopy
of conventionally planted corn. Soil loss from a good
quality grass and legume meadow is generally negligible.
42
-------
1.00
,80
Lul
U.
U-
Ul
o
2:
<£
O
a:
o
o
6
,60
.40
,20
0
*Average fall height of
drops from canopy (feet)
±L
T
ill
0
20 40 60 80
PERCENT GROUND COVER BY CANOPY
Figure 1.-Effect of crop canopy on effective EL
100
Plant Residues
With one-crop systems, crop residues can supply very
effective cover during the approximately 8 months from
harvest until the next crop develops full cover. Stalky
residues such as corn and grain sorghum provide more
effective cover when shredded than when left partially
standing. Complete residue cover at the soil surface
virtually eliminates raindrop impact on the soil, greatly
reduces the detachment capability and transport capac-
ity of runoff, and usually increases infiltration. Runoff
through or over a complete cover of residue mulch is
very low in sediment content.
For rowcrop production, high residue rates are most
fully utilized in the no-till systems. The seeds are planted
in narrow slots opened through the residue by a fluted
coulter or other device, without tillage. No-till planting
for corn has been very highly effective in chemically
killed meadow or small grain, in grain stubble, in
chopped rowcrop residues, and in winter-cover crops.
However, the practice is not adaptable to all fields (see
Section 4.1, Volume I).
Reductions in soil loss by various rates of straw
mulch tested under simulated rain have varied with soil
type and surface conditions (1, 4, 9, 28, 37, 42, 45, 70).
Figure 2 shows the average relation of soil loss with
various rates of mulch to corresponding losses with no
mulch, as observed on 35-foot cropland slopes subjected
to 5 inches of simulated rain in two 1-hour storms (86).
How much the slope length or steepness could be
43
-------
increased before unanchored mulch would be undercut
or transported by the flowing water has, however, not
been fully determined. Mulch applied on plowed and
disked surfaces of silt loam soils with slopes of 3 to 5
percent substantially reduced runoff, but mulch on a
15'' slope of untilled loam from which oat stubble had
been removed with a scraper had no significant effect on
amount of runoff. Under both conditions, however, one
ton of straw per acre reduced the velocity of the runoff
by about 60 percent (45).
Partial incorporation of the residues by shallow
tillage, such as disking, reduces the percentage of surface
cover and loosens some of the soil for easy detachment.
The residues are then less effective than equal quantities
left undisturbed on the surface. In a test under 5 inches
of simulated rain, no-till planting without prior disking
reduced soil loss 83 percent in contrast to a 73 percent
reduction with similar planting after disking (86). The
amount that soil-loss is increased by shallow tillage will
vary in relation to initial amount of residue and how
much is covered by the tillage.
Moldboard plowing inverts the upper 6 to 8 inches of
soil and usually covers virtually all of the residue. The
surface is then quite susceptible to erosion. However,
even with annual turnplowing, leaving the residues on
the field is far better than removing them. Regular
incorporation of crop residues by plowing gradually
increases the amount of organic materials in the soil and
improves water intake and soil structure. For 82
plot-years of continuous corn with residues removed
each fall, runoff during the seedling and establishment
months averaged 83% of corresponding losses from
fallow; for 50 plot-years in which the residues were
plowed down each year, runoff in those months aver-
1.0
as
o:
o
2
a6
0.4
0.2
20
40
60
80
100
% OF SURFACE COVERED BY MULCH
Figure 2.-Effect of plant-residue mulch on soil loss.
44
-------
aged 51% of that from fallow (90). Annual soil loss was
about 20% less where residues were incorporated than
where they were removed.
Improved Tillage Methods
That influence of a tillage practice on the disposition
of crop residues is extremely important for erosion and
sediment control is evident from the preceding discus-
sion. The surface micro topography and condition of the
soil after tillage also strongly influence the amount of
soil erosion. Roughness of the soil surface and porosity
of the tilled layer are important parameters in describing
the structure of a tilled soil (31). Rough surfaces detain
considerable quantities of water in microdepressions
until they can enter the soil, and the porous soil layer
offers a channel system to funnel water throughout the
tilled layer. Random roughness reduces runoff velocity,
and the water that is temporarily ponded in the
depressional areas shields portions of the soil surface
from particle detachment by raindrop impact. Also,
some of the sediment detached by raindrop impact on
the exposed surfaces is deposited on the ponded
surfaces. Porosity and roughness are influenced by the
type of tillage and by the water content of the soil at the
time of tillage. Pulverizing the soil increases erosion by
increasing the soil's detachability and increasing the
amount and rates of runoff.
When soil dries after a rain that has broken down its
surface structure and washed fines into soil voids, crusts
develop that are strong enough to reduce seedling
emergence (20, 47). Surface seals and crusts reduce
water intake by the soil and substantially increase
erosion (41, 78), Rough soil surfaces tend to concentrate
the dispersed material in the microdepressions and leave
the peaks more porous, but mulches are more effective
for preventing surface seal.
Soil compaction by heavy equipment can hamper
root and plant development and thereby increase soil
erosion. Conservation tillage practices generally require
less use of heavy equipment on the field. Aspects of soil
compaction were recently summarized by the American
Society of Agricultural Engineers (2).
Larson (30) points out that the secondary soil
aggregates around the seed and seedling roots must be
small enough to prevent undue drying of the soil, must
provide sufficient soil-seed or soil-root contact for
moisture transfer, must provide adequate aeration, and
must not be so finely divided as to encourage surface
crusting or mechanical impedance when dry. However,
the area between the rows, which he designates as the
water management zone, may be rough and cloddy or
may be left untilled under a residue mulch.
Conventional tillage includes primary and secondary
tillage operations normally performed in preparing a
seedbed for a given crop grown in a given geographic
area. Where the term is used in this manual as a basis for
comparisons, it includes moldboard plowing and several
disking and smoothing operations.
Minimum tillage is the minimum soil manipulation
necessary for crop production under existing soil and
climatic conditions. The term is often loosely applied to
any system with fewer operations than a conventional
system, but it is most accurate when applied to
plow-plant and wheeltrack-plant systems, in which the
field is plowed but secondary tillage is omitted. These
systems are most effective for erosion control when the
rowcrop follows one or more years of meadow, and
before the clods disintegrate.
Conservation tillage includes tillage systems that
create as good an environment as possible for the
growing crop and that optimize conservation of the soil
and water resources, consistent with sound economic
practices. Conservation tillage includes maximum or
optimum retention of residues on the soil surface and
use of herbicides to control weeds (98)
No-till is a system whereby a crop is planted directly
into a seedbed untilled since harvest of the previous
crop.
Residual Effects of Previous Crops
The benefits of crop rotations for minimizing periods
of little or no vegetative cover were pointed out earlier,
but crops and management practices also have residual
effects that influence soil credibility under succeeding
crops.
Sod-based Rotations
The greatest residual effects are derived from grass
and legume meadows. In data assembled from conven-
tional seeding and tillage practices, soil losses from corn
following meadow ranged from 14 to 68 percent of
corresponding losses from corn on adjacent plots in
meadowless systems. Grass and legume mixtures were
more effective than legumes alone. The erosion-control
effectiveness of rotation meadows turnplowed before
corn planting was, in general, directly proportional to
the quality of the meadow, as measured by hay yields.
Erosion reduction was greatest during the fallow and
corn-seeding periods and decreased gradually for about 2
years. The effects of well-managed long-term meadows
were still apparent in the third year. When second-year
hay yield exceeded the first, 2-year meadows were more
45
-------
effective than one-year meadows, but when allowed to
deteriorate in the second year they were less effective
(81).
Meadowless Systems
Crops that are not sod forming also have beneficial
residual effects on soil credibility, but they are much
less pronounced than those of grass and legume mead-
ows. Corn generally leaves the soil less credible than
soybeans but more erodible than good quality small
grain. All crop systems have beneficial residual effects
relative to continuous fallow; brief periods of fallow in a
rotation are not as erodible as continuous fallow.
Removal of the crop residues year after year gradually
reduces soil organic matter and adversely affects soil
tilth. One-time incorporation of residues by moldboard
plowing had little effect on infiltration or erosion, but
repeated incorporation year after year had very substan-
tial effects (90).
Mechanical Support Practices
These are mechanical erosion-control practices used
when slopes are too long or too steep for agronomic
practices alone to control erosion.
Contouring
Furrows made by plowing, planting, and cultivating
form natural channels in which runoff accumulates. If
the tillage is up and down slope, the shear stress of the
runoff increases as the slope of the furrows increases,
and erosion may be serious.
In contouring, tillage operations are carried out as
nearly as practical on the contour. The general rule is to
lay out guidelines which assure that all tillage is within a
gradient limit of 1 to 2 percent (59). On gently sloping
land, contouring will reduce the velocity of overland
flow by channeling it around the slope. Contoured ridge
or lister planting substantially increases the storage
capacity of the furrows and permits storage of large
volumes of water. When contouring is used alone on
steep slopes or under high rainfall intensities and soil
credibility, the hazard of gullying is increased because
row breakovers may release the stored water. Breakovers
cause cumulative damage as the volume of water
increases with each succeeding row. If the contour lines
are not carefully laid out and rows are allowed to cross
natural depressions at gradients much greater than 2
percent, adverse results of breakovers may completely
offset the beneficial effects of contouring. The effective-
ness of contouring is also impaired by decreased infiltra-
tion capacity due to surface sealing, and by reduction in
depression storage after tillage operations cease and the
soil settles (59).
Graded Rows
Graded rows are land-formed to a precise gradient.
This improves surface drainage and decreases the likeli-
hood of row breakovers.
Contour Stripcropping
Alternating contoured strips of sod with strips of row
crops is more effective than contouring alone. The sod
strips serve as filters when rows break, and much of the
soil washed from a cultivated strip is filtered out of the
runoff as it spreads within the first several feet of the
sod strip (64). In the Mormon Coulee near LaCrosse,
Wisconsin, some fields are reported to have been
cropped in strips for more than 70 years. Where the
strips were on the contour, or nearly so, erosion control
was good. Where the strips were sufficiently off-contour
to give row slopes of 5 percent or more, soil losses from
flow of runoff down the rows were high (8).
Systems with alternate contoured strips in meadow
reduce soil loss to about half of that from the same
rotation with contouring alone. Three-year rotations of
sod, small grain, and row crop were slightly less
effective. Alternate strips of fall-seeded grain and row
crop have effected some reduction in soil loss, but
alternate strips of spring-seeded grain and corn on
moderate to steep slopes have not proved more effective
than contouring alone (64).
Buffer Stripcropping is a practice in which strips of
grass are laid out between contour strips of crops in the
regular rotations. The grass strips may be irregular in
width and may be placed on critical slope areas in the
field (59).
Terracing
Terracing with contour farming is more effective foi
erosion control than Stripcropping, because it divides the
slope into segments with lengths equal to the terrace
spacing. With Stripcropping or contouring, the entire
field slope length is the effective length. With Stripcrop-
ping, the saved soil is largely that deposited in the sod
strips; with terracing the deposition is in the terrace
channels and may be as much as 80 percent (96) of the
soil moved to the channel. Erosion control between
46
-------
terraces depends on the crop system and other manage-
ment practices; with stripcropping, an effective sod-
based rotation is built into the system.
If a control level is desired that will maintain soil
movement between the terraces within the soil-loss
tolerance limit, the P factor for terracing should equal
the P factor for contouring. However, if the soil loss
equation is used to compute gross erosion for watershed-
sediment estimation, a terracing P factor equal to 20
percent of the contour factor is warranted (96). Since
terracing shortens the slope length, it also reduces the
bet ween-terrace soil loss by decreasing the topographic
factor. Dividing a slope that is steeper than 4% into n
equal segments divides the value of factor L by \fn. In
the table of P values given in Volume I, this reduction
was included in the Pf factor for convenience.
The two major types of terraces are the bench terrace
and the broadbase terrace (59). Broadbase terraces are
broad-surface channels or embankments constructed
across the slope of rolling land. They may be either
channel type or ridge type. The primary purpose of the
graded or channel-type terrace is to remove excess water
in such a way as to minimize erosion. The primary
purpose of the level or ridge-type terrace is moisture
conservation; erosion control is a secondary objective.
The channel is level and is sometimes closed at both ends
to assure maximum water retention. The Zingg conserva-
tion bench terrace is designed for use in semiarid regions
for moisture conservation. It consists of an earthen
embankment and a very broad flat channel that re-
sembles a land bench.
The steep-backslope terrace is constructed with a
backslope of 50% or steeper, which is kept in grass (8).
It may be either a graded or a level terrace. Parallel
grass-backslope terraces with subsurface drains are now
gaining popularity. They release the excessive water
slowly and are also better adapted to use of large farm
implements than graded or level terraces.
SEDIMENT DELIVERY RATIOS
The sediment delivery ratio is the parameter that
bridges the gap between upslope erosion data and
drainage-area sediment yield. The sum of the soil-loss
estimates for the individual tracts constituting a drainage
area approximates the quantity of soil moved from its '
original general position. To compute drainage-area
sediment yield, this estimate must be adjusted down-
ward to compensate for deposition in terrace channels,
in sod waterways, in field boundaries, at the toe of field
slopes, in depressional areas, and along the path traveled
by the runoff as it moves from the field to a continuous
stream system or lake (96). Sediment additions from
sources along this path must also be taken into account.
Further changes in sediment content of runoff water will
occur during the stream transport phase. The Universal
Soil Loss Equation computes gross sheet and rill erosion,
but it does not compute deposition. Nor does it
compute sediment from gully, streambank, and channel
erosion. The sediment delivery ratio provides a method
of accounting for the sediment losses and gains that
occur below the areas where the USLE is applied.
The delivery ratio is usually estimated from natural
drainage-area parameters and therefore does not account
for deposition in terrace channels or in constructed
settling basins or traps. The amount of sediment
deposited in these man-made devices near the sediment
source is subtracted from the computed field erosion to
obtain the gross-erosion estimate to which the delivery
ratio is applied. Two methods of defining the delivery
ratio will be discussed.
Delivery Ratios for Dealing with Downstream
Sediment Problems
For this purpose, the delivery ratio is defined as the
ratio of sediment delivered at a given point in the stream
system to the gross erosion from all sources in the
watershed above that point. Guides for estimating this
ratio were given in Volume I, section 3.3c. The source of
most of the information presented there was the
Sedimentation Section of the National Engineering
Handbook developed by the Soil Conservation Service
(72). The approximate delivery ratios that were listed
relative to watershed size were obtained from the
relationship curve derived from published and unpub-
lished data assembled by L. C. Gottschalk, G. M. Brune,
J. W. Roehl, R. Woodburn, S. B. Maner, L. H. Barnes,
and L. M. Glymph and presented in the Engineering
Handbook. This curve relates the delivery ratio to the
negative 0.2 power of drainage-area size. There have
been indications that the 0.1 power would be more
accurate for large drainage areas (3).
Analyzing data from 14 Texas Blackland Prairie
drainage areas that ranged from 0.42 to 97.4 square
miles, Renfro (57) computed delivery ratios ranging
from 0.62 for a drainage area of 0.5 square mile to 0.28
for an area of 100 square miles. These are significantly
larger than would have been estimated from the SCS
general relationship curve, and emphasize the need to
consider the other factors listed in Volume I as well as
watershed size. Several other relevant publications are
47
-------
listed in the literature citations (3, 13, 23, 24, 53, 56,
67).
Delivery ratios derived on this basis are more appro-
priate for dealing with downstream sedimentation prob-
lems than for estimating the amount and composition of
cropland sediment that reaches a continuous stream
system. However, they arc presently more directly
available than those discussed below and can be helpful
also for the latter purpose.
Delivery Ratios for Purposes of this Manual
For evaluation of cropland contributions to sediment
in stream systems, the delivery ratio should be defined as
the ratio of sediment delivered at the place where the
runoff enters a continuous stream system or lake to the
gross erosion in the drainage area above that point. It
will then not be biased by sediment-content changes that
occur during the stream transport phase. Where this ratio
is known or can be closely approximated from drainage-
area parameters, multiplying it by the computed gross
erosion will estimate the amount of sediment delivered
to the stream system.
No general equation for sediment delivery ratios as a
function of drainage-area parameters is presently avail-
able. A generally applicable upslope-deposition equation
is a major research need. However, guides for approxi-
mating the average delivery ratio for a particular
drainage area arc available. The ratio can approach a
value of 1.0 for a particular field if the runoff drains
directly into a lake or stream system, with no obstruc-
tions and no flattening of the land slope. On the other
hand, a wide expanse of forest duff or dense vegetation
below the eroding area may filter out essentially all of
the sediment except some of the colloidal material.
These are the extremes.
Anything that reduces runoff velocity (reduction in
slope steepness, physical obstructions such as ridges or
living or dormant vegetation, ponded water, etc.) re-
duces its capacity to transport sediment. When the
sediment load exceeds the transport capacity of the
runoff, deposition occurs. The observed sediment reduc-
tions by terracing or contour stripcropping are examples
of the potential magnitude of upslope deposition. More
than 80% of the soil eroded between terraces may be
deposited in the terrace channels because of the large
reduction in runoff velocity due to the terraces. Most of
the soil eroded from cultivated strips has been observed
to be deposited in the sod strips when contoured strips
of sod were alternated with equal-width strips in
cultivated crops.
Relative to the sediment-source area, the delivery
ratio will generally be directly related to amount of
runoff and inversely related to soil particle size. Relative
. to the land between the source area and the stream
system, the ratio will be directly related to slope
steepness and amount of channel-type erosion, and
inversely related to: distance of the source area from the
stream system or lake; density of vegetation at ground
level; and number of flow obstructions such as field
boundaries, culverts, etc.
The delivery ratio for a given drainage area will not be
constant for all runoff events, because the depth and
velocity of runoff will differ with storm size and
antecedent surface conditions. These differences will not
only affect transport efficiency; a major runoff event
may also pick up some of the sediment deposited
en route to the stream or lake in prior events. The
average delivery ratio for a drainage area can be
estimated more closely and should suffice for estimates
of long-term average sediment yields.
TOLERANCE LIMITS
This section discusses merits and limitations of several
alternative methods of defining soil loss or sediment
limits, as background information for those who may be
involved in developing state sediment control standards.
Soil loss limits used to illustrate points are not intended
as specific recommendations.
Optimum soil-loss limits for preservation of cropland
productivity may differ substantially from optimum
sediment standards for control of runoff pollution from
nonpoint sources. The underlying considerations are
quite different, and specific differences must be recog-
nized. Standards will be most beneficial when they
achieve both objectives with the least possible adverse
effect on production of food and fiber.
Tolerances for Preservation of Cropland
Productivity
Tolerance limits on average annual soil loss have been
used in this country for a quarter century to guide soil
conservation planning. Limits ranging from 2 to 5 tons
per acre are applied to individual field slopes. Experience
has shown these limits to be feasible and generally
adequate for preservation of high productivity levels.
The 2 to 5 ton tolerances represent the collective
judgment of soil scientists in the Soil Conservation
Service, Agricultural Research Service, and State agricul-
tural experiment stations in the 1950's. Factors consid-
ered in defining these limits were published by the Soil
48
-------
Conservation Service in reports of five regional soil loss
prediction workshops held from 1960 to 1962.
One of the major considerations was longtime mainte-
nance of adequate soil depth for good plant growth. The
rate of natural soil renewal for mature soils has been
hypothesized to balance the rate of erosion under
natural conditions, without influences of man (65).
Since erosion in excess of renewal rates reduces soil
depth, shallow soils were assigned lower tolerance limits
than those for deep soils with subsoil characteristics
favorable for plant growth. Prior erosion was a factor
because of its effect on the soil profile. Other considera-
tions included the prevention of field gullying, sedimen-
tation problems, seeding losses, soil organic matter
reduction, and plant nutrient losses. Research directed
to precise definition of soil loss tolerances (65, 68) has
been extremely limited.
Tolerances for Sediment Control
Sediment-control standards that coincide with the
toleranfes established for purposes of soil conservation
have the distinct advantage that a farmer is in compli-
ance if he follows a conservation plan approved by the
SWCD. These plans include a safety factor in that they
are generally designed to protect the most erodible
portion of the field. Since field slope gradients are
seldom uniform, the average soil loss for the entire field
is usually less than that on the slope the plan is designed
to protect.
Uniformly applied sediment-control standards based
on average annual soil losses are perhaps the most
feasible starting point, because of their simplicity and
because knowledge of precisely how much upslope soil
movement can be tolerated is inadequate. But if the
initial standards fail to attain the desired level of water
quality control, the next step should be a range in
standards to suit the requirements of various local
conditions rather than successive lowerings of uniform
limits. Uniformly lowering soil loss limits to attain
higher water-quality goals would unnecessarily remove
substantial acreages from grain production.
Before quantifying gross-erosion limits for cropland,
specific objectives of the limits should be defined.
Uniform soil loss tolerances reduce the total quantity of
sediment produced. This is important for reduction of
direct damage by deposition on upslope areas, on flood
plains, and in lakes or drainage ditches. But for control
of water pollution from nonpoint sources, other aspects
of the problem may be more important than the amount
of soil eroded from a particular field slope. These
include: upslope deposition, composition of the sedi-
ment, the protection needs, and fluctuations in sediment
loads.
Upslope Deposition
For control of water pollution from nonpoint sources,
soil material eroded from a field slope but deposited in
terrace channels, field boundaries, or elsewhere along the
path followed by the runoff en route to the stream
system is irrelevant. The fractions of sediments eroded
from upslope areas that are delivered to a continuous
stream system or lake range from less than 10% to nearly
100%. Uniform limits on erosion rates will allow a wide
range in quantities of delivered sediments. Estimating
sediment delivery ratios was discussed in the preceding
section. Low sediment delivery ratios are of little
relevance to preservation of the eroding cropland, but
they are highly important for water quality control.
Basing sediment standards on gross erosion minus the
estimated upslope deposition would achieve more uni-
form control of sediment quantity and allow greater
cropping flexibility. This would be a great improvement,
but sediment quantity is not the only important
criterion.
Composition Of The Sediment
Sediment traps or settling basins trap primarily the
coarse material. Clay, fine silt, and light soil aggregates
remain in suspension much longer than the coarse
material and are the greatest concern as a source of
turbidity and carrier of chemical compounds. There is
some particle-size selectivity in erosion, but generally the
composition of washoff material as it leaves the field is
closely related to that of the soil from which it is
derived. There is substantial size selectivity in the
transport and deposition phases, but the composition of
the sediment as it leaves the field will determine the
proportion of fine material available for transport in
suspension. Thus, for pollution control, variability in
soil-loss limits should be related to soil texture.
Protection Needs
Sediment standards could also be selective in relation
to the needs of the particular body of water being
protected. For example, controls need to be more
intensive for land draining into recreational waters and
urban water supplies than for land draining into major
river channels.
49
-------
Fluctuations In Sediment Loads
Short-time high sediment yields are much more
relevant for pollution control than for preservation of
the land resource. The average annual soil loss from a
particular crop system on a given field is the mean of
yearly losses that may differ tenfold, or even a hundred-
fold, due to differences in the cover and management
effects of the crops in the system, fluctuations in rainfall
erosivity, and intermittent crop failures.
The soil loss equation shows that under conditions
where a 6-year rotation of corn-com-corn-wheat-
meadow-meadow would average 5 tons of soil loss per
acre per year with conventional planting and tillage, the
first-year corn would average about 4 tons, second-year
com 9 tons, third-year corn 14 tons, wheat 2.7 tons, and
meadow 0.2 ton. On the average, at least half of the soil
loss from the corn would occur during the first month
after preparation of the clean-tilled seedbed. Appendix
tables in Agricultural Handbook No. 282 (96) show that
about one year in ten the rainfall-erosivity factor is
likely to exceed its local average value by 50 percent,
and one year in twenty by 75 percent. If a 20-year
rainfall occurred in the third corn year, the predicted
soil loss for that year on this field would be 1.75 times
14 tons, or nearly 25 tons, even though the longtime
crop-system average would not exceed the 5-ton limit.
Soil loss variability due to fluctuations in rainfall or
occasional seeding failures cannot be prevented, and
yearly or seasonal rainfall differences can be predicted
only on a probability basis. Because these differences
interact with other erosion factors, specific-storm soil
losses can presently not be accurately predicted. How-
ever, for each crop in the system, the effects of
fluctuations in rainfall tend to average-out over long
time periods, and the differences in cover and manage-
ment effects of the crops in a particular system are
reasonably well known. Therefore, the average annual
soil loss for each year in a crop sequence can be
predicted by use of the Universal Soil Loss Equation
with about the same accuracy as crop-system averages.
This is done by deriving factor C on a yearly basis by the
method illustrated in Agriculture Handbook No. 282.
Limits prescribed on a crop-year basis would reduce
the frequency of very high single-year or single-event soil
losses. In the preceding example, a 5-ton limit on the
design loss in any year of the cropping system would
require the use of good residue management for the
second-year corn and no-till planting in shredded-corn-
stalk mulch for the third corn year. If the soil and
climate were not compatible with no-till planting in
residue cover, the third-year corn would need to be
omitted from the cropping system.
However, crop-year soil loss limits would need to be
higher or more flexible than the present rotation-average
tolerances. If not, they would prevent production of
corn, soybeans, or other rowcrops on numerous fields
where these crops can be grown in rotation with
meadow and small grain, and they could also eliminate
the acceptability of periodic clean plowing for weed and
pest control on fields that are usually no-till planted.
The reason for this is that crop-year limits would allow
no credit for much more drastic reductions in soil loss
during other years in the crop system.
Modified Sediment Standards •
A possible alternative would be to continue the
present crop-system-average tolerances and superimpose
limits on the maximum design loss for any one year in
the system. The first limit would allow credit for
meadows and other low-erosion crops in a system
designed to preserve the productive capacity of the land.
The second limit could be sufficiently higher to avoid
unnecessary restrictions on land use and yet guard
against frequent occurrence of very high single-year
sediment yields.
The second limit would take into account such
factors as the intermittent more erodible conditions that
cannot be avoided, upslope deposition, soil texture, and
specific control needs for the location. Upslope deposi-
tion could be increased by requiring use of sediment
traps or filter strips of grass or small grain across the
lower end of a field in the years when it is plowed. The
same requirement could apply to the second and third
corn years in sod-based rotations.
RESEARCH NEEDS
The past 40 years have brought great progress in
erosion control, but serious erosion and sediment dam-
ages are still far too frequent. Population pressures,
increased export demands for agricultural products, and
more substitution of large machines for manpower
changed the erosion problems and intensified the hazard
on many millions of acres of productive cropland. Larger
fields generally mean longer continuous slopes. Exten-
sive monoculture sacrifices the potential residual effects
of sod-based systems. Large equipment greatly increases
production per man-hour but is not compatible with
following the field contours on much of the cropland.
50
-------
Furthermore, soil conservation and sediment control are
two individual goals and do not have the same require-
ments. Personnel and resources available for erosion
research in recent years have been insufficient to keep
pace with the changing needs. Research is particularly
needed in the following general areas. This research will
involve many preliminary and secondary investigations
that are not listed.
Sediment Delivery to Stream Systems
Average annual erosion rates on cropland can be
predicted with reasonable accuracy, but the percentage
of this eroded soil that reaches a continuous stream
system cannot. Sediment delivery ratios as usually
defined by geologists for dealing with downstream
sediment problems are influenced too much by stream
transport efficiency and sediment accretions from non-
agricultural sources to provide the information needed
for control of water pollution from nonpoint cropland
sources. If the sediment delivery ratio is used for this
purpose, it should be defined as the ratio of sediment
delivered at the place where the runoff water enters a
continuous stream system to the gross erosion from the
drainage area above that point.
The sediment delivery ratio, by either definition,
represents an attempt to reflect deposition and sediment
accretions in a single factor. The net effect of the two
processes would be difficult to relate to drainage-area
parameters in a regression equation because large
amounts of deposition and large sediment accretions can
occur in the same drainage area and balance each other.
Prediction equations for deposition and for sediment
accretions from runoff-induced erosion below the field
areas need to be separately derived, each as a function of
the drainage-area parameters pertinent to that process.
Such equations will facilitate estimating the effects of
cropland erosion control not only on the amount of
sediment delivered to the stream system but also on the
composition of sediment yields farther downstream.
Development of a better understanding of the basic
sedimentation and erosion processes involved between
the time when runoff leaves a field area and when it
reaches a continuous stream system is one of the greatest
erosion and sediment research needs.
Mathematical Modeling
Recent progress in development of mathematical
models for simulating the erosion and sedimentation
processes on field-size areas and on watersheds has
demonstrated the potential of such models to predict
temporal and spatial distribution of erosion and sedi-
mentation and to predict specific events more accu-
rately. However, some of the basic relationships assumed
for these initial models need research testing, and the
parameters need to be defined for a wide range of field
conditions. Relationships describing erosion and deposi-
tion in channels and gullies also need to be derived.
These models are generally complex and difficult to
use in the field. A relatively simple model that computes
individual-storm soil losses more accurately than the
Universal Soil Loss Equation is needed. Such a model
can use the basic format of the USLE, but it will need
separate erosivity factors for rill erosion and interrill
erosion, and their relationships to the other factors in
the equation will need to be determined. Use of volume
and peak rate of runoff to predict rill erosion shows
promise, but it requires derivation of a cropland-runoff
prediction equation.
Improvement of the basic models, and research to
determine the needed parameter relationships, should be
emphasized. Such models can provide more dependable
interpretation and extrapolation of field-plot data, and
the predictions of spatial and temporal variations in
erosion and deposition are needed for both conservation
and pollution-control planning.
Residue Management
Residue management is one of the major tools for
erosion control. In the densely populated countries, few
residues are usually available for erosion-control use
because they are needed for other purposes. We may
soon have similar problems in this country if crop
residues become economically profitable sources of
energy, concentrated feeds, or building materials. Re-
search must determine the optimum treatment and
placement of very limited residues and the optimum
amount and type of associated tillage required to
minimize erosion in the absence of what we now
consider adequate cover.
Neither has the optimum amount and placement of
residues where they are abundantly available been
determined. Optimum placement of a portion of the
residue may permit incorporation of the remainder into
the topsoil. This may reduce soil-temperature and
wetness problems without decreasing the erosion con-
trol.
Critical Slope-Length Limits for Practice
Effectiveness
Critical slope-length limits for effectiveness of partial
mulch covers and favorable microtopographies provided
by conservation tillage practices need to be determined.
51
-------
If limits can be defined in terms of depth and velocity of
runoff, they can then be related to soil, topography, and
rainfall characteristics for guidance in field application.
Clear definition of critical slope-length and drainage-area
limits is needed for improved terrace spacing design and
to prevent unexpected failures of some agronomic
practices. The investigations should include evaluation of
anchored versus loose mulches and of different types of
residues.
Slope-length limits for effective contouring need to
be more accurately defined in relation to permeability,
soil stability, crop cover, and other factors. Successive
row breakovers can result in rill erosion that more than
offsets the reduction in sheet erosion effected by the
contouring.
Erosion Index for Special Conditions
The El parameter is a good indicator of the erosive
potential of the rainfall and runoff in most of this
country, but there are a few conditions for which
further investigation of this factor is urgently needed.
The erosivity of surface runoff that is not directly
associated with drop impact needs to be evaluated, such
as runoff from thaw and snowmelt. This item is
particularly important in the Palouse Region of the
Northwest. The effects of soil-surface shielding by
ponded or very slowly moving runoff also need to be
identified and evaluated. These effects may account for
the difficulties experienced with the El parameter on the
Coastal Plains of the Southeast.
Topographic Factor
The topographic factor needs further research, both
with reference to factor interactions and with reference
to long or steep slopes. The effects of slope length and
steepness on soil erosion are known to be more variable
than indicated by existing formulas. There is evidence
that they are significantly influenced by mutual inter-
action and by interactions with cover, soil texture, and
rainstorm characteristics (or runoff rate). These inter-
action effects need to be quantified so that variations in
the topographic factor can be predicted. This is impor-
tant both for soil conservation planning and for pollu-
tion control guides.
Topographic effect also needs to be determined for
steep roadbank and construction slopes and for long
watershed slopes. Existing slope-length and steepness
formulas were derived from data on slopes not steeper
than 18% and, with only one exception, not longer than
270 feet. Extrapolation of the formulas to slopes that
far exceed these dimensions is quite speculative.
Soil Erodibility
The soil-erodibility nomograph needs to be aug-
mented for greater accuracy on high-clay subsoils and on
sandy loams. Effects of soil chemistry on credibility
need further investigation and quantification. Suscepti-
bilities of soils to sheet erosion and to rill erosion should
be evaluated separately, and influences of montmoril-
linitic clays need further study. Stripmine areas and
spoilbanks need specific research attention.
Particle size sorting in erosion and sedimentation is
important for pollution control and has not been.
adequately investigated.
Runoff Equation
A cropland runoff equation designed for general field
use would be a valuable asset in pollution control
planning. It could also provide an additional factor for
the Universal Soil Loss Equation that would improve its
accuracy, particularly for moderate storms.
Sediment Traps
Vegetated filter strips, settling basins, and sediment
traps can be used to cause sediments to deposit near the
point of origin, but more needs to be learned regarding
their design for optimum trapping efficiency and their
particle size selectivity. Also, extreme runoff events may
pick up substantial amounts of sediment from the traps.
The probabilities of such occurrences and methods of
minimizing them need to be investigated.
52
-------
LITERATURE CITED
I. Adams, J. E. 1966. Influence of mulches on runoff,
erosion, and soil moisture depletion. Soil Sci. Soc.
Amer. Proc. 30:110-114.
2. American Society of Agricultural Engineers. 1971.
Compaction of agricultural soils. St. Joseph, Mich.,
471 p.
3. American Society of Civil Engineers, Task Com-
• mittee on Preparation of Sedimentation Manual.
1970. Chapter IV. Sediment sources and sediment
yields. Jour. Hydraulics Div. 96:1283-1327.
4. Barnett, A. P., Diseker, E. G., and Richardson, E. C.
1967. Evaluation of mulching methods for erosion
control on newly prepared and seeded highway
backslopes. Agron. Jour. 59:83-85.
5. Barnett, A. P., and Rogers, J. S. 1966. Soil physical
properties related to runoff and erosion from
artificial rainfall. Trans. Amer. Soc. Agr. Engin.
9:123-125.
6. Baur, A. J. (dim). 1952. Soil and water conserva-
tion glossary. Jour. Soil and Water Conserv.
7:41-52,93-104, 144-156.
7. Baver, L. D. 1940. Soil physics. John Wiley & Sons,
New York, 370 p.
8. Beasley, R. P. 1972. Erosion and sediment pollution
control, Iowa State Univ. Press, Ames, 320 p.
9. Borst, H. L., and Woodburn, R. 1942. The effect of
mulches and surface conditions on the water rela-
tions and erosion of Muskingham soils. Tech. Bui.
825, U.S. Dept. Agr., Washington, D.C., 16 p.
10. Browning, G. M., Parish, C. L., and Glass, J. A.
1947. A method for determining the use and
limitation of rotation and conservation practices in
control of soil-erosion practices in Iowa. Soil Sci.
Soc. Amer. Proc. 23:246-249.
11. Carter, C. E., Greer, J. D., Braud, H. J., and Floyd,
J. M. 1974. Raindrop characteristics in South
Central United States. Trans. Amer. Soc. Agr.
Engin. 17:1033-1037.
12. Chepil, W. S., and Woodruff, N. P. 1963. The
physics of wind erosion and its control. Advan. in
Agron. XV:211-302. Academic Press, New York.
13. Corinth, R. L. 1970. Effects of watershed charac-
teristics on reservoir sediment deposition. Misc. Pub.
No. 970, U.S. Dept. Agr., Washington, D.C.
14. David, W. P., and Beer, C. E. 1975. Simulation of
soil erosion. Trans. Amer. Soc. Agr. Engin.
18:126-133.
15. Dideriksen, R. I., and Grunewald. A. R. 1974. What
is happening to our soil resources: the current
inventory of rural land. Proc. 29th Ann. Mtg., Soil
Conserv. Soc. Amer.: 15-17.
16. Ellison, W. D. 1947. Soil erosion studies part 1.
Agr. Engin. 28:145-146.
17. Foster, G. R., and Meyer, L. D. 1975. Mathematical
simulation of upland erosion using fundamental
erosion mechanics. In Present and prospective tech-
nology for predicting sediment yields and sources:
190-207. ARS-S40, U.S. Dept. Agr., Washington,
D.C.
18. Foster, G. R., and Wischmeier, W. H. 1974. Evaluat-
ing irregular slopes for soil loss prediction. Trans.
Amer. Soc. Agr. Engin. 17:305-309.
19. Freeman, 0. L., and Bennett, I. L. 1969. Control of
agriculture-related pollution. A report to the Presi-
dent, by U. S. Dept. Agr. and Off. Sci. and
Techno!., 102 p.
20. Gamer, T. H., and Bowen, H. D. 1966. Plant
mechanics in seedling emergence. Trans. Amer. Soc.
Agr. Engin. 9:650-653.
21. Glymph, L. M., and Carlson, C. W. 1968. Cleaning
up our rivers and lakes. Agr. Engin. 49:590,607.
22. Gunn, R., and Kinzer, G. D. 1949. The terminal
velocity of fall for water droplets. Jour. Met.
6:243-248.
23. Happ, S. C. 1971. Discussion of sediment sources
and sediment yield. Amer. Soc. Civ. Engin., Jour.
Hydraulics Div. 96:597-599.
53
-------
24. Happ, S. C. 1972. Valley sedimentation as a factor
in environmental impact on rivers: 27.1-27.25. H.
W. Shen Pub., Ft. Collins, Col.
25. Holt, R. F., Johnson, H. P., and McDowell, L. L.
1973. Surface water quality. Proc. Natl. Conserv.
Tillage Conf., Soil Conserv. Soc. Amer.: 141-156.
26. Hudson, N. W. 1971. Raindrop size. In Soil conser-
vation:50-56. Cornell Univ. Press, Ithaca, New
York.
27. Klingebiel, A. A., and Montgomery, P. H. 1961.
Land capability classification. Agr. Handbook No.
210. U. S. Govt. Printing Off., Washington, D. C.,
21 p.
28. Kramer, L. A., and Meyer, L. D. 1969. Small
amounts of surface mulch reduce soil erosion and
runoff velocity. Trans. Amer. Soc. Agr. Engin.
12:638-641,645.
29. Langbein, W. B., and Schumm, S. A. 1958. Yield of
sediment in relation to mean annual precipitation.
Trans. Amer. Geophys. Union 39:1076-1084.
30. Larson, W. E. 1962. Advances in minimum tillage
for corn. 17th Ann. Hybrid Corn Industry Res.
Conf., Amer. Seed Trade Assoc. 17:44-51.
31. Larson, W. E., and GUI, W. R. 1973. Soil physical
parameters for designing new tillage systems. Proc.
Natl. Conserv. Tillage Conf., Soil Conserv. Soc.
Amer.: 13-21.
32. Lattanzi, A. R., Meyer, L. D., and Baumgardner, M.
F. 1974. Influences of mulch rate and slope
steepness on interrill erosion. Soil Sci. Soc. Amer.
Proc. 38:946-950.
33. Laws, J. O. 1941. Measurements of fall velocity of
water drops and rain drops. Trans. Amer. Geophys.
Union 22:709-721.
34. Lelland, J. H., Rogers, H. T., and Elson, J. 1941.
Effect of slope, character of soil, rainfall, and
cropping treatments on erosion losses from Dun-
more silt loam. Va. Polytech. Inst. Tech. Bui. No.
72,31 p.
35. Lloyd, C. H., and Eley, G. W. 1952. Graphical
solution of probable soil loss formula for North-
eastern Region. Jour. Soil and Water Conserv.
7:189-191.
36. Mannering, J. V., Johnson, C. B., and Wischmeier,
W. H. 1974. Erodibilities of selected soils in the
Maumee Basin. 1974 Agron. Abs., Amer. Soc.
Agron.:167.
37. Mannering, J. V. and Meyer, L. D. 1963. Effects of
various rates of surface mulch on infiltration and
erosion. Soil Sci. Soc. Amer. Proc. 27:84-86.
38. Menard, H. W. 1961. Some rates of regional erosion.
Jour. Geol. 69:154-161.
39. Meyer, L. D. 1971. Soil erosion by water on upland
areas. In River mechanics II, Chapter 27. H. W. Shen
Pub., Ft. Collins, Col., 25 p.
40. Meyer, L. D., Foster, G. R., and Romkens, M. J. M.
1975. Source of soil eroded by water from upland
slopes. In Present and prospective technology for
predicting sediment yields and sources: 177-189.
ARS-S40, U.S. Dept. Agr., Washington, D.C.
41. Meyer, L. D., and Mannering, J. V. 1961. Minimum
tillage for com: its effect on infiltration and
erosion. Agr. Engin. 42:72-75.
42. Meyer, L. D., and Mannering, J. V. 1963. Crop
residue mulches for controlling erosion on sloping
land under intensive cropping. Trans. Amer. Soc.
Agr. Engin. 6:322,333, 337.
43. Meyer, L. D., and McCune, D. L. 1958. Rainfall
simulator for runoff plots. Agr. Engin. 39:644-648.
44. Meyer, L. D., and Wischmeier, W. H. 1969. Mathe-
matical simulation of the process of soil erosion by
water. Trans. Amer. Soc. Agr. Engin. 12:754-758,
762.
45. Meyer, L. D., Wischmeier, W. H., and Foster, G. B.
1970. Mulch rates required for erosion control on
steep slopes. Soil Sci. Soc. Amer. Proc. 34:928-931.
54
-------
46. Middleton, H. E., Slater, C. S., and Byers, H. G.
1932-1934. Physical and chemical characteristics of
the soils from the erosion experiment stations.
Tech. Bui. No. 316 and No. 430, U. S. Dept. Agr.,
Washington, D.C.
47. Morton, C. T., and Buchele,W. F. 1960. Emergence
energy of plant seedlings. Agr. Engin. 41:428431,
453-455.
48. Musgrave, G. W. 1947. The quantitative evaluation
of factors in water erosion, a first approximation.
Jour. Soil and Water Conserv. 2:133-138.
49. Newman, J. E. 1970. Climate in the 1970's. Crops
andSoils22(4):9-12.
50. Olson, T. C., and Wischmeier, W. H. 1963. Soil
erodibility evaluations for soils on the runoff and
erosion stations. Soil Sci. Soc. Amer. Proc.
27:590-592.
51. O'Neal, A. M. 1952. A key for evaluating soil
permeability by means of certain field clues. Soil
Sci. Soc. Amer. Proc. 16:312-315.
52. Onstad, C. A., and Foster, G. R. 1975. Erosion
modeling on a watershed. Trans. Amer. Soc. Agr.
Engin. 18:288-292.
53. Pacific Southwest Inter-Agency Committee. 1974.
Report of water management subcommittee, ero-
sion and sediment yield methods, 45 p.
54. Parr, J. F., and Bertrand, A. R. 1960. Water
infiltration into soils. In Advan. in Agron. XII:
311-363. Academic Press, Inc., New York.
55. Peele, T. C., Latham, E. E., and Beale, 0. W. 1945.
Relation of the physical properties of different soil
types to erodibility. South Carolina Agr. Expt. Sta.
Bui. No. 357.
56. Piest, R. F. and Bowie, A. J. 1974. Gully and
streambank erosion. Proc. 29th Ann. Mtg., Soil
Conserv. Soc. Amer.: 189-196.
57. Renfro, G. W. 1972. Use of erosion equations and
sediment delivery ratios for predicting sediment
yield. Proc. Sediment Yield Workshop, Oxford,
Miss. In press, U. S. Govt. Printing Off., Washington,
D.C.
58. Roth, C. B., Nelson, D. W., and Romkens, M. J. M.
1974. Prediction of subsoil erodibility using chem-
ical, mineralogical and physical parameters. Rep.
EPA-660/2-74-043, 111 p. U.S. Govt. Printing Off.,
Washington, D. C.
59. Schwab, G. O., Frevert, R. K., Edminster, T. W.,
and Barnes, K. K. 1966. Soil and water conservation
engineering, Second Ed. John Wiley & Sons, Inc.,
New York.
60. Skidmore, E. L., Fisher, P. S., and Woodruff, N. P.
1970. Wind erosion equation: computer solution
and application. Soil Sci. Soc. Amer. Proc.
34:931-935.
61. Skidmore, E. L., and Woodruff, N. P. 1968. Wind
erosion forces in the U. S. and their use in
predicting soil loss. U.S. Dept. Agr., Agr. Handbook
No. 346. U. S. Govt. Printing Off., Washington, D.
C.,42p.
62. Smith, D. D. 1941. Interpretation of soil conserva-
tion data for field use. Agr. Engin. 22:173-175.
63. Smith, D. D., and Wischmeier, W. H. 1957. Factors
affecting sheet and rill erosion. Trans. Amer. Geo-
phys. Union 38:889-896.
64. Smith, D. D., and Wischmeier, W. H. 1962. Rainfall
erosion. In Advan. in Agron. XIV: 109-148. Aca-
demic Press, Inc., New York.
65. Smith, R. M.,and Stamey.W. L. 1965. Determining
the range of tolerable erosion. Soil Sci.
100:414424.
66. Soil Conservation Society of America. 1970. Re-
source conservation glossary. 52 p.
67. Stall, J. B., and Bartelli, L. J. 1959. Correlation of
reservoir sedimentation and watershed factors. 111.
State Water Survey Rep. No. 37.
68. Stamey, W. L., and Smith, R. M. 1964. A conserva-
tion definition of erosion tolerance. Soil Sci.
97:183-186.
69. Stanford, G., England, C. B., and Taylor, A. W.
1970. Fertilizer use and water quality. ARS 41-168,
U.S. Dept. Agr., Washington, D. C., 19 p.
55
-------
70. Swanson, N. P., Dedrick, A. R., Weakly, H. E.,and
Haisc. H. R. 1965. Evaluation of mulches for
water-erosion control. Trans. Amer. Soc. Agr.
Engin. 8:438-440.
71. U. S. Department of Agriculture. 1971. Basic
statistics-national inventory of soil and water con-
servation needs. Statis. Bui. No. 461. Washington,
D.C.
72. U. S. Department of Agriculture. Soil Conservation
Service. 1971. Sediment sources, yields and delivery
ratios. In National engineering handbook. Sect. 3.
Washington, D.C.
73. U. S. Department of Agriculture. Soil Conservation
Service. 1972. Land resource regions and major land
resource areas. Agr. Handbook No. 296. U. S. Govt.
Printing Off., Washington. D. C.
74. U. S. Weather Bureau. 1958. Rainfall intensity-
frequency regime. Tech. Paper No. 29, 5 parts.
75. Van Doren. C. A., and Bartelli, L. J.- 1956. A
method of forecasting soil losses. Agr. Engin.
37:335-341.
76. Voxnesensky, A. S., and Artsruui, A. B. 1940. A
laboratory method for determining the anti-erosion
stability of soils. Soils and Pert., Commonwealth
Bur. Soils 10:289.
77. Wadleigh, C. H. 1968. Wastes in relation to agricul-
ture and forestry. Misc. Pub. No. 1065, U. S. Dept.
Agr., Washington, D. C., 112 p.
78. Whitaker, F. D., Heinemann, H. G., and Wischmeier,
W. H. 1973. Chemical weed controls affect runoff,
erosion, and corn yields. Jour. Soil and Water
Conserv.73:174-176.
79. Wischmeier, W. H. 1955. Punched cards record
runoff and soil-loss data. Agr. Engin. 36:664-666.
80. Wischmeier, W. H. 1959. A rainfall erosion index for
a universal soil-loss equation. Soil Sci. Soc. Amer.
Proc. 23:246-249.
81. Wischmeier, W. H. I960. Cropping-management
factor evaluations for a universal soil-loss equation.
Soil Sci. Soc. Amer. Proc. 24:322-326.
82. Wischmeier, W. H. 1962. Rainfall erosion potential -
geographic and locational differences of distribu-
tion. Agr. Engin. 43:212-215.
83. Wischmeier. W. H. 1966. Relation of field-plot
runoff to management and physical factors. Soil Sci.
Soc. Amer. Proc. 30:272-277.
84. Wischmeier, W. H. 1972. Upslope erosion analysis.
Chap. 15 I'M Environmental impact on rivers. Water
Resour. Publications, Fort Collins, Col.
85. Wischmeier, W. H. 1972. Erosion and sediment
research in the United States. Proc. Second Allerton
Conf. on Environmental Quality and Agriculture.
Spec. Pub. No. 25, College Agr., Univ. lll.:21-23.
86. Wischmeier, W. H. 1973. Conservation tillage to
control water erosion. Proc. Nad. Conserv. Tillage
Conf., Soil Conserv. Soc. Amer.: 133-141.
87. Wischmeier, W. H. 1974. New developments in
estimating water erosion. Proc. 29th Ann. Mtg., Soil
Conserv. Soc. Amer.: 179-186.
88. Wischmeier, W. H. 1975. Estimating the soil loss
equation's cover and management factor for undis-
turbed areas. In Present and prospective technology
for predicting sediment yields and sources: 118-124.
ARS-S40, U.S. Dept. Agr., Washington, D.C.
89. Wischmeier, W. H., Johnson, C. B., and Cross, B. V.
1971. A soil erodibility nomograph for farmland
and construction sites. Jour. Soil and Water Con-
serv. 26:189-193.
90. Wischmeier, W. H., and Mannering, J. V. 1975.
Effect of organic matter content of the soil on
infiltration. Jour. Soil and Water Conserv.
20:150-152.
91. Wischmeier, W. H., and Mannering, J. V. 1969.
Relation of soil properties to its erodibility. Soil Sci.
Soc. Amer. Proc. 33:131-137.
92. Wischmeier, W. H., and Meyer, L. D. 1973. Soil
erodibility on construction areas. Highway Res.
Board, Nat. Acad. Sci. Spec. Rep. 135:20-29.
93. Wischmeier, W. H., and Smith, D. D. 1958. Rainfall
energy and its relationship to soil loss. Trans. Amer.
Geophys. Union 39:285-291.
56
-------
94. Wischmeier, W. H., and Smith, D. D. 1961. A
universal soil-loss estimating equation to guide
conservation farm planning. Trans. 7th Cong. Inter-
natl. Soil Sci. Soc. 1:418425.
95. Wischmeier, W. H., and Smith, D. D. 1962. Soil-loss
estimation as a tool in soil and water management
planning. Pub. No. 59, Internati. Assoc. Sci. Hy-
drol.: 148-159.
96. Wischmeier, W. H., and Smith, D. D. 1965. Predict-
ing rainfall-erosion losses from cropland east of the
Rocky Mountains-Guide for selection of practices
for soil and water conservation. Agr. Handbook No.
282, U. S. Govt. Printing Off., Washington, D. C.,
47 p.
97. Wischmeier, W. H., Smith, D. D., and Uhland, R. E.
1958. Evaluation of factors in the soil loss equation.
Agr. Engin. 39:458462,474.
98. Witmuss, H. D., Triplett, G. B. Jr., and Greb, B. W.
1973. Concepts of conservation tillage systems using
surface mulches. Proc. Natl. Conserv. Tillage Conf.,
Soil Conserv. Soc. Amer.:5-12.
99. Woodruff, N. P. 1966. Wind erosion mechanics and
control. Proc. First Pan Amer. Soil Conserv. Cong.
Brazil:253-262.
100. Woodruff, N. P., and Siddoway, F. H. 1965. A
wind erosion equation. Soil Sci. Soc. Amer. Proc.
29:602-608.
101. Young, R. A., and Wiersma, J. L. 1973. The role of
rainfall impact in soil detachment and transport.
Water Resour. Res. 9:1629-1636.
102. Zingg, A. W. 1960. Degree and length of land slope
as it affects soil loss in runoff. Agr. Engin.
21:59-64.
57
-------
CHAPTER 4
NUTRIENT ASPECTS OF POLLUTION FROM CROPLAND
M. H. Frere
Nutrients are naturally occurring chemicals essential
for plant growth. Sixteen elements are essential for the
growth and reproduction of most plants. Most soils are
lacking in adequate amounts of nitrogen, phosphorus,
and potassium. Hence, fertilizers containing these nutri-
ents are essential to maintain the current level of
agricultural production. The other nutrient elements
may be added as impurities in the fertilizer or to treat
specific nutritional problems. Present evidence indicates
that nitrogen and phosphorus are the principal nutrient
pollutants and, therefore, only these nutrients are
considered in this chapter.
A major source of nutrients reaching water bodies in
this country is sewage, both from municipal treatment
plants and nonsewered residences. These are point
sources of pollution and extensive efforts are underway
to limit their contributions. Runoff from rural land is
another major source. Unlike point sources, runoff
integrates the contribution of nutrients and water from a
wide variety of landscapes that are continuously chang-
ing with time. It must be recognized that nutrients leak
from the system even when fertilizer is not applied and
while we cannot eliminate nutrient losses, it is desirable
to minimize them.
While a number of review papers have been written
about nutrient losses (47, 76, 89, 98, 145, 167) the
emphasis of this chapter is on the practices that can
control nutrient losses and the background necessary to
use these practices effectively. This chapter reviews the
literature and summarizes the data for nitrogen and
phosphorus leaving cropland by runoff, erosion, and
leaching. The material covered is limited to precipita-
tion-induced transport of the nutrients from cropland
and improved pastures. Beyond the scope of this
overview are the very important problems of irrigation
return flow, wind erosion, and losses from waste disposal
areas.
It must be emphasized at the beginning that the
dynamic system under consideration is very complex.
The wide variety of climates and landscapes provides
such a wide range of results that there is no typical case.
Complications are introduced by difficulties in chemical
analysis for nutrients in water samples. Numerous
procedures have been followed for chemical analysis.
Changes in nutrient form can occur between the time
the sample is taken and when it is analyzed. Some of the
nutrient reported as soluble could have been associated
with colloidal material not removed. The practical
significance of these complications is unknown, but they
are noted at this time to warn the reader of the
limitations of the data associated with nutrient pollu-
tion.
THE PROBLEMS
Two problems are associated with nutrients in the
aquatic environment: the water may be toxic to humans,
animals, or fish when the concentration of certain
nutrient forms exceeds a critical level; and eutrophica-
tion may be accelerated.
Toxicity
Case (28), Lee (92), and Winton (181) reviewed the
problems associated with nitrates and nitrites in drinking
water. The nitrite form of nitrogen, which is the most
toxic, interacts with components in the blood to
interfere with oxygen transport. Methemoglobinemia,
the technical name given to this illness, is often called
"the blue baby syndrome" because infants are very
susceptible. Most of the problems with drinking water
have been associated with farm wells with faulty well
casings and located close to manure concentrations such
as barnyards.
Nitrate is 5 to 10 times less toxic than nitrite and
healthy mature animals with single stomachs are able to
59
-------
excrete nitrate in the urine. Cattle, young animals, and
children convert some of the nitrate to nitrite in their
stomachs and can develop methemoglobinemia. Since
food also contains nitrite and nitrate, the response to
nitrate in drinking water could be quite variable. The U.
S. Public Health Service Drinking Water Standards of
1962 set the limit for nitrate at 10 mg N per liter (45
ppm nitrate). Armitage (5) reported the Recommended
Drinking Water Standards of the World Health Organi-
zation to be: 0-50 ppm nitrate = recommended, 50-100
ppm nitrate = acceptable. There is some concern,
however, that even nontoxic nitrate levels (chronic
conditions) may lower resistance to environmental
stresses and interfere with normal metabolism.
Dissolved ammonia is another form of nitrogen that
can occur at levels toxic to fish. Microorganisms can
generate free ammonia from organic matter in lake
bottoms during summer stagnation periods (164). Trout
are sensitive to 1-2 ppm ammonia (35) while goldfish
appear to be less sensitive (48).
Eutrophication
Eutrophication is the enrichment of waters by nutri-
ents and the ensuing luxuriant growth of plants. Much
has been written about this subject in the last few years
(100, 111, 153, 269). Rapid growth of algae is the
greatest and most widespread eutrophication problem in
most states (2). Algae can create obnoxious conditions
in ponded waters, increase water treatment costs by
clogging screens and requiring more chemicals, and cause
serious taste and odor problems (7 7). When a large mass
of algae dies and begins to decay, the oxygen dissolved
in the water decreases and certain toxins are produced,
both of which kill fish (49). The complexities of the
ecosystem are illustrated by the observation that the
nutritional status of a species of algae can vary from lake
to lake or even from different areas and depths of the
same lake on the same day (39). Streams, however, do
not age in the same sense as lakes, but their biological
productivity can be increased by added nutrients (69).
For example, phosphate from farm land was a very
beneficial and important factor in the high production
of brook trout in Canadian streams (137).
Aquatic plants require a number of nutrients for
growth, but nitrogen and phosphorus appear to be the
ones accounting for most of the excessive growth.
Sawyer (128) concluded that eutrophication becomes a
problem when the concentration of inorganic nitrogen
exceeds about 0.3 ppm and inorganic phosphorus
exceeds about 0.015 ppm. These concentrations of the
inorganic forms of nutrients are maintained by microbial
conversion of organic forms so the total input of
nitrogen and phosphorus per unit area of the lake
(loading rate) is important (57). Current international
guidelines for eutrophication control are 1.8 to 4.5 Ibs.
of P and 45 to 90 Ibs. of N per surface acre of lake per
year (] 69).
The various roles of nitrogen in eutrophication have
been recently reviewed (24, 53). Aquatic organisms
assimilate nitrate and ammonium. Ammonia and amino
acids are excreted by live organisms and released by
decaying organisms. Fungi and bacteria can convert the
organic nitrogen in dead plant material and sediment to
ammonia and nitrate. Whenever the environment be-
comes anaerobic, in the presence of decomposable
organic matter, nitrates are denitrified to gaseous nitro-
gen compounds. Bouldin (20) estimates that the daily
loss of the nitrate in the bottom sediments can be 7 to
15 percent by microbiological denitrification and 2 to
28 percent of the ammonia by volatilization.
Some additional inputs of nutrients are often over-
looked, such as aquatic birds, leaves, dust, and pollen.
Another source is the fixation of atmospheric nitrogen
into organic nitrogen by a number of organisms such as
the blue-green algae. This process is considered to be
adaptive in that it occurs when other sources of nitrogen
are depleted.
Kramer et al. (86) and Lee (93) reviewed the role of
various phosphorus compounds in eutrophication. Sol-
uble orthophosphate is usually regarded as completely
available for algal growth. Soluble organic phosphates
and polyphosphates are probably not too available, but
are readily converted to orthophosphate. Finally, phos-
phate in participate organic matter and adsorbed to
mineral sediments is usually only slowly released. The
adsorption capacity of the sediment for phosphate
ranges from low for quartz sand to very high for certain
silicate clays.
Sediment low in phosphate will usually remove
phosphate from solution as it settles out (63, 64, 86, 93,
179). When the sediments have high P contents and the
environment around the clay is electrochemically re-
duced, then some of the phosphate can be released to
the solution (86). This released phosphate may form the
mineral apatite, which is relatively insoluble (179), or if
there is a mixing process, such as caused by wind, the
phosphate is redistributed through the lake (63,178).
Because a lake's ecological system is so complex,
Shannon and Brezonik (132) devised an index of seven
parameters to quantitatively characterize the trophic
60
-------
state of a lake. For 55 lakes in Florida, the relation
between this index and the loading rates of nitrogen and
phosphorus showed that an increased phosphate loading
was statistically the more important. An additive form
of the loadings accounted for 60 percent of the
variability in the index.
SOURCES OF NUTRIENTS
Fertilizers are well known as a source of nutrients on
cropland, but they are not the only source and some-
times are not the most important source. Fertile
cropland soil is a pool of nutrients with different degrees
of availability to the crop and to the transport processes
of runoff, erosion, and leaching. Precipitation and
animal wastes are other sources. In addition, legume
crops, with the assistance of microorganisms, can biolog-
ically convert atmospheric nitrogen into organic nitro-
gen.
The relative importance of each of these sources
depends on a number of factors, such as the geographical
location with its relation to climate and soils and the
crop management practices previously and presently
used. Table 1 shows the estimated 1969 national
nitrogen inputs (119) and estimates for several water-
sheds in Wisconsin (13). The national input for phos-
phorus in fertilizer is for 1973 (79), the manure input is
from Table 3, and the inputs from plant residues and
precipitation were calculated from the N inputs and N/P
ratios of 7.5 (Table 17, Vol.1) and 100 (Fig. 3),
respectively.
On the national level, fertilizer is a major input and
manure is a relatively small input. The relative contribu-
tion of these sources can be reversed for a specific
geographic location, as shown by the data for Wisconsin
watersheds.
Nutrient Cycles
Nitrogen
Volumes have been written about nitrogen behavior.
Two comprehensive reviews are: "Soil Nitrogen" (10)
and "Soil Organic Matter and its Role in Crop Produc-
tion" (3). Figure 1, adapted from Stevenson (149),
illustrates the numerous compartments and pathways of
nitrogen.
Most of the reactions in the soil portion of the cycle
are microbial and thus the rates are sensitive to
temperature and moisture. Warm (90° F) and moist
(water in 80 percent of the voids) are optimum
conditions for cycling within the soil. The conversion of
organic nitrogen to nitrate (ammonification and nitrifi-
cation) is often called mineralization. A study of 39 soils
from across the United States showed that the rate of
mineralization was proportional to the pool of mineral-
izable nitrogen. The size of this pool was not highly
correlated with the total organic matter or total nitrogen
(148). Thus, some forms of organic matter are readily
converted to mineral forms whereas other organic forms
are not. Part of the stable forms may exist as metal-
organic and organic-clay complexes.
Soils contain 0.075 to 0.3 percent total nitrogen or
1,500 to 6,000 Ibs. per acre in the top 6 inches (5). Soils
Table 1. Sources of nitrogen and phosphorus on a national and a watershed scale
Source
Fertilizer
Fixation
Manure
Kant residues
Precipitation
Total
National
Nitrogen
Million tons
6.8
3.0
1.0
2.5
1.5
14.8
Percent
45.9
20.3
6.8
16.9
10.1
Phosphorus
Million tons
2.2
0
0.4
0.3
0.01
2.9
Percent
76
0
14
10
0
Wisconsin watersheds
Nitrogen
Ibs/acre
10
12
42
45
8
117
Percent
8.5
10.3
35.9
38.5
6.8
Phosphorus
Ibs/acre Percent
8 32
0 0
12 48
5 20
0 0
25
61
-------
o\
K>
AIR
ADSORPTION
HARVEST
PRODUCTS
ANIMALS
PLANTS I
VOLATILIZATION
RESIDUES
EROSION
IMMOBILIZATION
SOIL ORGANIC
MATTER
AMMONIFICATION
CLAY AND
ORGANIC COMPLEXES
IMMOBILIZATION
LEACHING
NITRIFICATION
Figure l.-The nitrogen cycle in agriculture.
-------
cultivated for 100 years can still release 30 to 60 pounds
of nitrogen per acre per year and have drainage waters
with 5 to 10 ppm nitrate nitrogen. Fertile soils in the
Corn Belt are estimated to release 120 Ibs. of N per acre
(7). In the semiarid west, some fields are fallowed (kept
free of vegetation) every other year to accumulate
moisture. The increased moisture also promotes mineral-
ization of 20 to 50 Ibs. of N/acre (37).
When ammonia (a gaseous form of nitrogen) reacts
with the water it forms positively charged ammonium
ions. These ions are held by the negative charge of the
clays as exchangeable cations. Some of the ammonium
ions can be trapped or fixed between clay platelets.
Ammonia absorption by the soil has not been
considered in the past as a major path of nitrogen input.
However, in areas of high ammonia concentrations, such
as downwind of industries or feedlots, the soil, lakes,
and plants have absorbed from 20 to 70 pounds of
N/acre/yr (55, 67, 68,138).
Immobilization, the reverse of mineralization, is the
part of the N cycle that converts nitrate and ammonium
into organic forms. It occurs under aerobic or anaerobic
conditions and basically involves the uptake of the
mineral forms by microorganisms in the synthesis of cell
tissue. Whenever organic residues low in nitrogen are
being decomposed, mineral nitrogen must be used
because the carbon-to-nitrogen ratio of microbial tissue
is on the order of 5-10:1. Dead microbial tissue then
becomes part of the organic matter pool that can be
mineralized. A major problem in quantifying the immo-
bilization process is that it is impossible, without
nitrogen tracers, to measure the small amount of the
product in the large organic-matter pool.
Denitrification is not well understood but appears to
be a very important part of the nitrogen cycle affecting
environmental quality. Denitrification is the use of
nitrate by anaerobic microbes for oxygen and results in
the production of nitrogen and nitrogen oxide gases. The
necessary anaerobic conditions are most prevalent when
the water content of the soil is high, which is the same
condition needed for leaching. Another requirement is a
supply of carbon for an energy source. Lack of useable
carbon may be the factor that prevents all the nitrate in
the leachate from being denitrified. Carbon is usually
not very mobile and, therefore, once the nitrate passes
below the root zone, the opportunity for denitrification
is limited (114). Complete waterlogging is not essential
for denitrification. Since the soil contains a wide range
of pore sizes, an unsaturated soil can have areas where
the water contents and microbial activity are sufficient
to produce an anaerobic environment and denitrifica-
tion. Quantification of this process has been limited.
Most of the estimates have been based on nitrogen
budgets where all unaccounted-for nitrogen is assigned
to denitrification. The nitrogen gases produced are very
difficult to measure under field conditions.
Average losses have been estimated at 10 to 30
percent of the total yearly mineral nitrogen input (26).
When excessive rates of nitrogen arc applied, as much as
50 percent can be lost (99).
The inputs of fertilizer, biological fixation, animal
wastes, and precipitation; and the losses by leaching,
runoff and erosion, and plant uptake (avoiding excessive
fertilizer use) will be covered in subsequent sections.
Phosphorus
The phosphorus cycle shown in Figure 2 is a lot less
complicated than the nitrogen cycle, although it has
some of the same paths. Before considering the similar
paths, we will examine those reactions in the soil which
are unique for phosphorus. Black (7.5), Olscn and
Flowerday (775) and Ryden et al. (725) have prepared
comprehensive reviews of the subject.
The phosphate concentration in the soil solution is
low, usually 0.01 to 0.1 ppm P, although the total P in
the soil ranges from 100 to 1,300 ppm (73). The mineral
forms of phosphorus—calcium, iron, and aluminum
phosphates-have very low solubilities and the phosphate
is highly adsorbed to clay minerals. The organic forms of
phosphate have not been studied as extensively as have
the inorganic forms. However, the organic part of the
total phosphorus can range from 3 to 75 percent.
Because of the low solution concentrations and high
degree of adsorption, phosphate tends not to leach.
After 286 Ibs. of P had been applied in 11 years, the
plant "available" level had increased only 18 Ibs. and
very little had moved below 12 inches (32). After 82
years of fertilization, the total P had been doubled, but
no added P was found below 54 inches (76). The
availability of phosphate in the soil decreases exponen-
tially with time. However, soils vary greatly in their
conversion of added P to insoluble forms (90). Recent
work (52) indicates that chemical reactions immobili/.e
more than 50 percent of added soluble phosphate in a
few hours and an additional 10 percent in a month or so.
The amount of biological immobilization into organic
phosphates that occurs simultaneously with the chemical
reactions depends upon the amount of biological activ-
ity.
Precipitation
The amounts of nitrogen and phosphorus added in
precipitation are generally low and, for cropland, they
63
-------
HARVEST
PRODUCTS
ANIMALS
X FERTILIZERS
RESIDUES
EROSION
ADSORPTION
IMMOBILIZATION
SOIL ORGANIC
MATTER
/
MINING
DESORPTION
MINERALIZATION
Figure 2.-The phosphorus cycle in agriculture.
-------
are negligible in comparison to other inputs. On forests,
unimproved pasture and rangeland, and lakes, the input
of nutrients added by precipitation can be significant.
The concentrations of nitrogen and phosphorus in rain
and snow vary not only across the country, but within
short distances and during a storm.
Uttormark et al. (166) list 40 factors that can
influence the concentration of nutrients in precipitation.
Feedlots, industrial urban centers, power plants, disposal
sites, etc., are all relatively local sources that can increase
the general regional level of nutrients. They prepared a
map of the United States delineating areas of different
nitrogen contribution in rainfall. The Lake States had
the highest contribution, about 2 to 3 Ibs. of N/acre/yr.,
whereas the Western States had less than 1 Ib. of
N/acre/yr. Dry fallout was not included.
Seasonal maps of ammonium and nitrate in rain
across the United States (73) show that the highest
concentrations are in the spring and summer. It appeared
that the rainfall concentrations might be related to the
soil because the Southeast acidic soils are the lowest.
Ammonium concentrations may change as much as
10-fold during the year, but nitrate changes much less.
When rainfall samples are taken at different times
during a storm, the nitrogen concentrations are often
slightly higher during the first part (24, 50). This could
be due to evaporation from the drops into dry air or the
washout of dust.
The phosphate input in most places may be associ-
ated with dust, either as dry fallout between storms or
washed out of the atmosphere with rain. Dust or dry
fallout is a very important component associated with
the precipitation input of nutrients. It has been esti-
mated (166) that in arid regions 70 percent of the
nitrogen in uncovered rain gages is from dust. Storms
blowing in from oceans are quite low in phosphorus.
Phosphorus can also be associated with ash and smoke,
as indicated in concentrations of 0.24 ppm phosphate in
rain at Cincinnati compared to threefold less at a rural
location near Coshocton, Ohio (174). A yearly input of
0.2 to 0.6 Ib. P/acre/yr has been estimated (24, 110).
One problem associated with evaluating the importance
of dust is the absence of a measure of the loss by dust
into the atmosphere. There is no evidence that snow has
any different concentrations of nutrients than rain
would at the same time of year.
Fertilizers
Fertilizer use to improve crop yields dates from
antiquity. Sanskrit writings of 3,000 years ago note the
value of dung for fertilizer (103). Guano and Chilean
nitrate were available in Europe as a fertilizer over 100
years ago. Yields from field tests started at England's
Rothamsted Experiment Station in 1840 have been
maintained near maximum by the use of manure and
fertilizer while those from unfertilized plots have de-
creased to an uneconomic level.
Organic forms of nitrogen were the cheapest sources
of nitrogen until about 1900 (70). Before the 1950's,
sodium nitrate and ammonium sulfate were the principal
nitrogen sources. Then ammonium nitrate became the
leading source, only to be surpassed by anhydrous
ammonia and urea in the 1960's. Phosphate fertilizers
also changed in the 1950's from normal superphosphate
to concentrated superphosphates. These are trends to-
wards the use of more concentrated forms, thus reducing
shipping charges per pound of nutrient. The source of
the nutrient makes no difference to the plant because of
the extensive cycling in the soil. Some sources, such as
ammonium sulfate, can increase the soil acidity if used
for extended periods.
Table 2 shows how fertilizer use on four major crops
has changed over a 10-year period. The acreage of corn,
wheat, and cotton harvested has remained relatively
constant except in 1974 when acreage controls were
removed. The acreage of soybeans has steadily increased.
Except for cotton, the percentage of acres fertilized has
consistently increased over the years. Note also that the
yield per acre has tended to increase. Fertilizer rate has
increased except for nitrogen in 1974 when short
supplies and increased costs because of the energy crisis
caused many farmers to reduce their application rate.
The relative plateau of fertilizer use on cotton deserves
some comment. Cotton is relatively sensitive to nitrogen
and excess nitrogen can reduce yields. Cotton has been
fertilized intensively for a long time and the optimum
rates have evidently been found. Pests, such as insects,
weeds, and disease, are probably limiting yield more
than fertility.
Speculation on future fertilizer use is fraught with
uncertainties. The problem is basically economics. How
much fertilizer should the farmer apply to maximize his
profits? The yield response to fertilizer is less for each
subsequent increment of fertilizer and as the yield
increases some other costs also increase. In the 1960's,
farmers received $2.50 for each dollar invested in
fertilizer and so they substituted an investment in
fertilizer for additional land or labor (62). Real estate
costs and wages increased over 250 percent since 1950
while plant nutrient costs decreased (38). While the
energy crisis will tend to increase fertilizer costs faster
than other costs, it will probably be sometime before
they reach the level of other farm costs. Thus, farmers
will probably continue to fertilize, but will carefully
appraise the size of the applications and eliminate
65
-------
Table 2. The change in fertilizer use on four crops in the past 10 years
Crop
Corn ....
Wheat . .
Soybeans .
Cotton ....
Year
1964
1966
1968
1970
1972
1974
1964
1966
1968
1970
1972
1974
1964
1966
1968
1970
1972
1974
1964
1966
1968
1970
1972
1974
Acres
harvested1
Millions
55.4
56.9
55.9
57.2
57.4
63.7
49.8
49.9
55.3
44.1
47.3
64.1
30.8
36.5
41.1
42.1
45.7
52.5
14.1
9.6
10.2
11.2
13.0
13.1
Area fertilized2
N
82
91
92
94
96
94
47
49
56
61
62
66
6
17
21
21
22
22
75
75
73
72
77
79
P
Percent
75
85
88
90
90
87
36
38
43
44
44
46
10
24
27
27
29
28
56
58
55
48
55
58
Fertilizer rate2
N
Ibs./ac.
45
83
102
112
115
103
28
32
37
39
46
46
13
14
12
14
14
15
69
77
71
75
75
78
P
18
24
28
31
29
27
12
14
14
13
16
17
11
15
17
16
18
18
21
23
23
24
24
23
Yield1
bu.lac.
63
73
80
72
97
26
26
28
31
33
23
25
27
27
28
Ibs.lac.
517
480
516
438
507
1 United States Department of Agriculture, "Agricultural Statistics", Years 1964 to 1974.
2 1964-1970 data from "Cropping Practices" SRS-17, Statistical Reporting Service, USDA.
1972-1974 data from "Fertilizer Situation", FS-5, Economic Research Service, USDA.
excessive use. A food crisis that increases the price of
farm products relative to fertilizer costs would stimulate
fertilizer use. Fertilizer use in the United States will
increase 5 percent per year because of rising populations,
improved diets, and increased exports, according to a
recent estimate (78).
Animal Wastes
Before commercial fertilizers came into common use,
animal manure supplied most of the nutrients added to
the soil. In the 1960's it was more expensive to load,
haul manure several miles, and spread it than it was to
purchase and apply commercial fertilizer. In addition to
the economics, the convenience of commercial fertilizer
caused animal wastes to become a disposal problem
rather than a nutrient source.
The following discussion is based on information
from several references (29, 30, 96, 170,180).
Manure is the excrement of animals that contains the
undigested food and the urine. The nutrient content of.
manure is different for different animals, type of feed,
and amount of water consumed. In addition, the amount
of bedding used to absorb the urine or presence of
superphosphate to react with the ammonia, the method
of storage, and the duration of storage influence the
nutrient content of manure being applied to the soil.
Fresh manure contains 50 to 90 percent moisture, 0.2
to 6 percent total nitrogen, and 0.06 to 2.5 percent total
phosphorus, on a dry weight basis. The old rule of
thumb was that the typical ton of moist cow manure
contained about 10 Ibs. of total nitrogen and 1 Ib. of
total phosphorus when it was applied to the field.
Present methods of confinement, feeding, and manure
handling vary enough that this rule of thumb is no
longer adequate.
Yeck et al (183) recently estimated that 5.8 million
tons of nitrogen are excreted annually by livestock in
the United States. The percentage that can be collected
varies from zero for cattle on the range to nearly 100 for
caged poultry. The work of a number of investigators
indicates that about half the nitrogen collected is lost
during storage, handling, and spreading before it can be
incorporated into the soil. Thus, of the 2.4 million tons
66
-------
of nitrogen that they estimate can be collected, only 1.2
million tons are available as a substitute for fertilizer and
a large part of this is already being applied.
The pollution potential isn't a result of animals per
se, but of the manure they produce. Since there are
maps showing the geographical distribution of animals
(Figs. 20-24, Vol. I), we needed a method for conversion
to manure and its nitrogen and phosphorus contents.
From the collectible nitrogen previously described for
each class of livestock, a 50 percent loss before
incorporation was assumed. This can then be geographi-
cally distributed with the animals. To calculate the
amount of manure associated with nitrogen one must
assume some percentage of N in the manure being
spread. The amount of phosphorus can likewise be
calculated from the percent phosphorus. .These percent-
ages of N and P in the manure were estimated for the
different kinds of animals from data of various re-
searchers. These calculations are summarized in Table 3.
Not all of the nitrogen in manure is immediately
available. Most of the nitrogen left in manure when it is
incorporated into the soil is in organic compounds. Like
soil organic matter, microbes must mineralize the or-
ganic nitrogen into ammonium and nitrate, which are
taken up by the plant. About 40 or 50 percent of the
organic nitrogen is mineralized during the first cropping
season (772). Additional amounts are released in subse-
quent years so that yearly applications can build up the
nitrogen-supplying capacity of the soil. Of course this
"slow release" of nitrogen also means that some of it can
be released when plants are not rapidly growing (fall and
spring) and thus be available for leaching.
Biological Fixation of Nitrogen
One of the most recent reviews of the biological
fixation of nitrogen is by Allison (3). Another good
source of information is three chapters in "Soil Nitro-
gen" (10). Nonsymbiotic (free-living) organisms prob-
ably-fix about 10 Ibs. of N per acre annually. This
amount is of little practical importance in the nitrogen
balance of a cultivated field, but since it is comparable
to the precipitation contribution, it is important in
nonfertilized grass sods.
Nitrogen-fixing bacteria in a symbiotic relation with
roots of certain plants, principally legumes, can fix
sufficient nitrogen to support a grass-legume pasture.
With low soil nitrogen, the amount of nitrogen fixed in
effective legume nodules correlates closely with the dry
weight of the legume tissue produced. Generally, the
fixed nitrogen is produced only as the plant needs it and,
therefore, plants with poor growth will not fix much
nitrogen. Similarly, high soil nitrogen levels, such as
from fertilization, will reduce the amount of nitrogen
fixed because it is not needed for plant growth. There is
some indication (12) that fresh organic matter from
previously fertilized crops stimulates N fixation. Biggar
and Corey (13) report 200 Ibs of N per acre per year as
being fixed in legume systems. Allison (3) cites several
references where 150 to 300 Ibs. are common and as
much as 600 Ibs. of N/acre/year was observed. He
concludes that a grass-legume mixture may fix more N
per acre than legumes alone because the grass will
continually remove any nitrogen mineralized from the
soil or plant material.
Table 3. Estimates of nitrogen and phosphorus in manure that is available for application to cropland
Animals
Beef cattle
Dairy cattle
Swine
Laying hens
Broilers ....
Totals
Vol.1.
Figure
No.
21
22
23
24
25
Collectible N1
Thousand tons
650
670
600
250
240
2410
Available N2
Thousand tons
330
330
300
125
PO
1 205
N content'
in manure
Percent
2.5
2.0
2 8
4.5
38
Available
manure
Million tons
13
17
M
2.8
3.T
47.0
P content3
in manure
Percent
0.8
0.6
1.0
1.7
1 3
Available Ps
Tltoitsand tons
100
100
110
50
40
400
2 Estimated amounts of nitrogen that can be collected (183).
3 Available after application assuming 50% losses during handling, storage, and spreading.
4 Estimated contents based on data from various authors.
5 Calculated from available N and N content.
Calculated from available manure and P content.
67
-------
TRANSPORT FROM CROPLAND
Water pollution by nutrients from cropland involves
one of three transport processes: leaching, runoff, and
erosion. In nature the results of these processes are not
easily distinguishable. Water may infiltrate into the soil
and thus cause leaching, but a few feet or yards
downslope the water with its dissolved nutrients may
come to the surface and join the overland flow or
runoff. Similarly, the distinction between runoff and
erosion may be quite difficult when nutrients in the
runoff are adsorbed on or released from the eroded
sediment that the runoff water is carrying. The larger the
drainage area of the stream sampled, the more mixed are
the three transport processes. Some cases where these
processes operated independently will be examined so
that the system can be better understood.
The usual range of values is presented in Figures 3
and 4. Occasionally, more extreme values will be
observed because the system is highly variable. Concen-
trations are not provided for sediment transport because
the concentrations of sediment vary so much. Soils
contain 0.075 to 0.3 percent nitrogen and 0.01 to 0.13
percent phosphorus and deposition of coarse material
during transport can increase the concentrations in the
transported sediment by 2- to 6-fold.
Leaching
Leaching is the process whereby soluble chemicals are
dissolved and removed from the soil in water that is
percolating through the soil. Nitrate is the principal
nutrient form found in drainage waters because it is
seldom adsorbed to the soil minerals. Some red subsoils
in the southeast are an exception (156). Organic forms
of nitrogen and phosphorus and the orthophosphate ion
are seldom found to any extent in drainage water
because they are held by the soil.
Three principal ways of evaluating leaching are by
lysimeter studies, monitoring tile drains, and taking core
samples. Each method has some limitations and it is
important to recognize these limitations when interpret-
ing the data.
Lysimeters
Drainage lysimeters are columns of soil in the field
isolated by impermeable cylinders with facilities for
collecting the water that drains out of them. Cylinder
walls at or above the surface can prevent or reduce
runoff and, therefore, increase the amount of leaching.
Also, if suction is not applied, the bottom of the column
must be saturated before drainage will occur. This
saturated zone can provide an opportunity for denitrifi-
cation and dilution. Some lysimeters do not contain
undisturbed profiles while others are very shallow. Both
conditions limit their usefulness for field interpretation.
Lysimeters have provided some valuable information
about leaching. Kolenbrander (55) found that as the clay
content of the soil increased from 10 to 50 percent, the
nitrogen loss decreased from 40 Ibs. N/acre/year at 18
ppm to 4 Ibs. N/acre/year at 2 ppm. Wild (177) found
that nitrogen mineralized in the soil didn't leach quickly
in fine soil with cracks, but Kissel et al. (80) found that
applied fertilizer, simulated by chloride, did move
quickly through the cracks of a clay soil. Kolenbrander
(&5) reported phosphorus losses of 0.2 Ib. P/acre/year at
0.08 ppm for both cropped and grass lysimeters, with
and without fertilizer.
Tile Drains
Tile drains have recently provided most of the
measurements of nutrient leaching. While tile drains
sample a much larger area and thus provide a more
integrated value, they may not accurately reflect the
nutrient content and water volume leaving the field. One
problem is to define the boundaries of the drainage field
so that losses can be calculated on an area basis. Another
is that the tiles short-circuit the drain paths with aerated
conditions reducing the time and opportunity for
denitrification (757). As an illustration of this condition,
Thomas and Barfleld (157) monitored an area where
one-third of the water flow was from tile lines with
nitrate levels of 15 ppm N while the rest of the seepage
had only 3 ppm N. At lower flows, the seepage
accounted for nearly 90 percent of the flow with nearly
zero nitrate while tile lines had 10 percent of the flow
with concentrations of 9 ppm N.
Many studies have been reported concerning the
concentration and the average annual loss per unit area
of nutrients in drainage waters (83, 96, 125, 166). The
first feature to be recognized is the extreme variability in
the data, both the loss per unit area and the concentra-
tion. Thus, an average value has little utility. One of the
major factors causing the variability in the loss data is
that the water flow, and thus the load of nutrients
transported, varies greatly between dry years and wet
years. The type of crop grown is another major factor.
Nutrient concentrations are consistently lower in the
drain water from grasslands and woodlands than in that
from cropland. Grasslands and woodlands have a longer
68
-------
CONCENTRATION-ppm
O.QI QJ LQ 10.0 IQO.O
PRECIPITATION Ezzzzzzzzzzza N
CROPLAND RUNOFF Pi *--• ^tu (11 tutui n«\H
¥// ///////////// ////////////J
N NON-CROPLAND RUNOFF
P' i DRAINAGE
Figure 3.-Range of nitrogen and phosphorus concentrations in different waters.
-------
SPATIAL RATE - Ibs./acre/year
0.01
O.I
1.0
10.0
100.0
PRECIPITATION
N
r/ / / / / /
/ / / / / / / // / / // / // / //// // // / / // / / / / / / 1
N CROPLAND RUNOFF
NON-CROPLAND RUNOFF
PC
DRAINAGE
i /////////////
N
NON-CROPLAND
SEDIMENT
CROPLAND SEDIMENT [//////////////////////////////////TTT*
P
PC
i// / / //////// (/ / / // / / ////i i
figure 4.-Range of spatial rates of nitrogen and phosphorus in waters and sediments.
-------
growing season in which to remove nutrients and reduce
water flow. Also, even though they are located on soils
of lower fertility, these lands are seldom fertilized
because the economic return is insufficient. Generally
then, the fertile land that is cropped will have the
highest concentrations in the drain water and the most
drain water. Fertilizer applied to these lands, tends to
increase the concentration of nitrogen in the drain
water, but seldom that of phosphorus.
Soil Cores
Cores of soil can be taken from plots and fields, then
separated into segments and analyzed. This provides a
means to follow the movement of the nutrients down
through the soil profile. In most soils, the distance the
band of applied nitrate moves down through the soil is
proportional to the amount of water in excess of that
required to raise the water content to field capacity (84,
94). Water in excess of crop needs and soil properties
such as texture and structure are important factors in
determining the rate of leaching. In Wisconsin, Olsen et
al. (114) estimated the annual percolation below the
root zone to be 6 inches. A silt loam soil has a water
capacity of 30 to 50 percent, thus nitrate could move 12
to 18 inches per year under these conditions and it
would take at least 20 years to reach a water table at 30
feet.
In the semiarid Great Plains, leaching is not a hazard
except when additional water is added by irrigation or
fallow land prevents normal transpiration by plants.
North Dakota soils wetted to 6 feet only occasionally in
40 years (121), but in Colorado, nitrate accumulated
below the root zone in a wheat-fallow system (150). The
average nitrate content in a 20-foot profile was 260 Ibs
N/acre for cultivated dryland compared to 90 Ibs N/acre
in native grass. Since very little fertilizer is applied to
dryland crops in eastern Colorado, the difference be-
tween these averages is probably a result of nitrate
production and leaching under cultivation and fallowing.
Runoff
Runoff occurs when the rate of precipitation exceeds
the rate of infiltration. A heavy mulch apparently acts as
a sponge to hold water as do the numerous small surface
depressions. Surface-applied fertilizers can dissolve into
this water held in the depressions and by the mulch.
When this water becomes runoff it carries a load of
nutrients. This is why the highest concentrations in
runoff occur when there is runoff soon after surface
-------
because there is more plant cover more of the time.
Overgrazed lands have higher nutrient losses because of
increased runoff and erosion. Concentrations of nutri-
ents will also be high occasionally if animals have direct
access to a stream, if animals are fed near streams, or if
fertilizers and manures are surface applied. Phosphate
concentrations in runoff may be higher from grasslands
than from adjacent cropland after harvest because
freezing and drying cause a release of nutrients from the
vegetation (163, 176). Timber harvesting also causes a
sudden release of nutrients to runoff waters from the
decay of the trimmed foliage (18). The reduced water
consumption after harvest also increases the annual
nutrient loss. Uttormark et al. (166) and Brezonik (24)
provide a summary of losses from land in various uses.
Ryden et al. (125) provide a recent review of phos-
phorus losses.
Sediment
Sediment is the major transport vehicle for phos-
phorus and organic nitrogen. Raindrop splash and
overland flow of water detach soil particles containing
adsorbed phosphorus and associated organic matter. The
flowing water transports the particles off the field.
Transport capacity depends primarily on the volume and
velocity of water flow. Whenever the velocity is reduced,
such as by a flatter slope, the transport capacity is
reduced and any sediment in excess of the reduced
capacity settles out. Since the larger and heavier particles
settle out first, the remaining sediment contains a larger
percentage of the finer particles. The finer particles have
a higher capacity per unit of sediment to adsorb
phosphorus and, also, organic matter is lighter and tends
to be associated with the fine particles. Thus, the
transported sediment is richer in phosphorus and nitro-
gen than the original soil (59).
Bedell et al. (11) found that 98 percent of the
sediment samples from 2- to 4-acre watersheds contained
as much or more organic matter, phosphorus, and
nitrogen than the soil. The degree of enrichment has
often been described by an enrichment ratio, the ratio of
the nutrient concentration in the sediment to its
concentration in the. soil of the watershed. Recently,
enrichment ratios of 2 to 6 have been found for
phosphorus (130,155). Barrows and Kilmer (9) reported
an average enrichment ratio of 2.7 for nitrogen and 3.4
for "available" phosphorus. Available means that frac-
tion by chemical extraction that is expected to be
available to plants.
Massey et al. (104, 105) found that the enrichment
ratio was inversely related to the concentration of
sediment in the runoff and the total amount of sediment
lost. Doty and Carter (34) found that when the sediment
concentration was highest at peak flow, the chemical
and physical composition of the sediment was similar to
that of the soil. At lower flows, the sediment concentra-
tion was lower, the percentage of clay in the sediment
increased, and the enrichment ratio increased.
The loss of total N and P in sediment from cropland
ranges from 1 to 50 or 100 Ibs. per acre per year. Two
factors can contribute to the high loss from cropland:
the soil is fertile and contains a lot of nutrients per
pound of soil, and tillage operations often leave the soil
very susceptible in the most erosive part of the year.
Noncropland areas often have a lower nutrient content
than the cropland and also lower erosion rates because
of more vegetative cover. An exception is the large
amounts of sediment produced by gullies that are often
prevalent in noncropland.
Nutrient Losses from Large Watersheds
As indicated at the beginning of this section, the
nutrient composition of a stream reflects a mixture of
the three basic transport mechanisms. The composition
of streams changes with amount of flow, season of the
year, and distance down the stream as new material is
added from tributaries, seepage, and outfalls.
The Environmental Protection Agency (165) is under-
taking a national eutrophication study, but at present
only a preliminary analysis is available. The data indicate
less variability in total nitrogen loss than in total
phosphorus loss, and increases in the concentrations of
nutrients in the water are correlated with increased
density of animals in the watershed. A recent summary
of the spatial loss for streams (166) reports a range of 1
to 13 Ibs. N/acre/year and 0.03 to 2 Ibs. P/acre/year,
with averages of 5 and 0.4. Both studies indicate that
forest land yields less nutrients than agricultural land
and that the Midwestern states have higher losses than
other states.
Most perennial streams have a number of outfalls for
municipal and industrial wastes, which complicates any
budgeting of sources. For example, agricultural land is
estimated to have contributed all of the inorganic N, 49
percent of the total N, and 13 percent of the total P to
the Potomac River in 1966 (71). The figures for the
agricultural contribution to the Hudson River were 37
percent of the total N and 27 percent of the total P.
Streams are dynamic systems with living organisms
assimilating the nutrients and particulate matter ad-
sorbing or releasing the nutrients. Some of the nutrients
removed from solution may go undetected in the debris
moving along the bottom of the stream or floating on
72
-------
the surface. Phosphate-deficient sediments will remove
phosphate from the solution {154). Thus, the concentra-
tion of phosphate in a stream changes as soil is added
from noncropland and stream banks (88, 173) or gullies
(130). Keup (75) provides a good discussion of the
behavior of phosphorus in flowing streams. He cites
studies on two rivers where the overall loss from solution
was logarithmically related to the distance of streamflow
below a point source of phosphorus. The coefficient was
0.01 to 0.02 per mile; that is, 14 to 30 miles for a 25
perceni reduction.
EFFECT OF CONTROL PRACTICES
There is every reason to believe that nutrient loss
from cropland can be controlled at an acceptable level if
proper management practices are used. As pointed out
by Klingebiel (82), soil surveys are an important basis
for planning the optimum use of each field. By intensive
use of fields with high crop production potential and
low hazards from runoff, erosion, and leaching, the use
of marginal land with higher hazards can often be
reduced.
The effectiveness of management practices for the
control of nutrient losses has not been quantitatively
evaluated to the same degree as have the impacts on
runoff and erosion. Only in recent years have adequate
data on nutrients been collected.
The possibility of creating another problem by
solving one problem should be the concern of all who
make recommendations. The nitrate contamination of
ground water in Runnels County, Texas (87) can be used
as an example of this possibility. This land, which had
been dryland farmed since 1900, had nitrate formed and
leached below the root zone but not down to the water
table. Extensive terracing after the drought in the early
I950's increased water retention and leached the nitrate
on down to the water table. Thus, a nitrate leaching
problem was created by the terracing done to solve a
problem of limited moisture. While hindsight is much
clearer than foresight, we should learn from previous
experiences and examine our recommendations for
secondary effects.
Erosion and Runoff Control Practices
Sediment is a major pollutant in itself. That it also
carries nutrients and pesticides means that the first goal
in controlling pollution from cropland should be to
-:>ntrol erosion. For example, more than 97 percent of
'lie N and P lost from some watersheds was associated
with sediment lost primarily in the 2 months after
1'lanttng (27).
Soil conservation practices have been stressed for over
40 years and sufficient data have been collected to
permit fair predictions of average annual soil loss. Old
punciples are continually being used to create new
methods of control (140). Data on the effectiveness of
these practices for controlling nutrient losses are ade-
quate for only qualitative predictions. Generally, a
reduction in sediment loss provides less of a reduction in
nutrient loss.
Residue Management
The greater the amount of residue left on a field, the
greater the reduction in erosion. Zwerman et al. (184)
reported that leaving the crop residues instead of
removing them decreased runoff 50 percent and did not
change the nutrient content of the runoff water. Thus,
losses of nutrients in runoff were reduced 50 percent.
Losses of nutrients with sediment were also reduced.
Romkens et al. (124) used simulated rain and observed
that several conservation tillage systems reduced sedi-
ment loss but increased the loss of soluble nitrogen in
the runoff. Ketcheson and Onderdonk (74) found that
covering broadcast fertilizer with a chopped cornstalk
mulch reduced soil phosphorus losses 65 percent and
fertilizer losses 97 percent.
No-till or zero tillage (7) is one of the most effective
practices for reducing erosion. The effect of these
practices on the amount of runoff is variable; sometimes
runoff is increased and sometimes it is decreased. Smith
et al. (136) reported that nitrogen in runoff was not
greatly affected by the no-till practice, whereas phos-
phorus increased 5- to 8-fold, probably from leaching of
residues. Since runoff is sometimes decreased, nitrate
leaching can be increased (158). Schwab et al. (131)
found little difference in the nitrogen and phosphorus
content of tile drainage from conventional and no-tillage
plots.
Cropping
Sod reduces runoff and permits very little erosion.
Therefore, on an average annual basis, the rotations with
sod should show a reduced nutrient loss. Results from
corn-wheat-clover plots (102, 134) indicate that the
rotation reduces the total N and P loss 3- to 6-fold
compared to continuous corn or wheat. Schuman et al
73
-------
(729, 130) reported that a sod pasture lost about 10-fold
less nitrogen than continuous' corn. The phosphorus
losses from the pasture were lower by a factor of two,
even though the concentration in the runoff and on the
sediment was higher. A higher concentration of phos-
phorus in the runoff, particularly snowmelt, from
pasture or hay lands has been reported by several
workers (27, 161, 176).
When little or no residues are left on a field, as when
corn is harvested for silage, then planting a cover crop
such as a small grain will protect the soil during the
winter. Smith et al. (136) recorded a 50 percent
reduction in runoff and a 40-fold reduction in losses of
sediment, total nitrogen, and total phosphorus when
corn was planted into a ryegrass cover crop. They found
little effect on the soluble nutrients lost in runoff, which
were already low.
Supporting Practices
Several erosion control practices such as contouring
and terraces are physical rather than agronomic. They
can be used by themselves with regular cultivation or in
conjunction with reduced tillage systems to achieve even
greater reductions. Bedell et al. (11) reported that
contouring a corn-wheat-meadow rotation reduced sedi-
ment, nitrogen, and phosphorus losses from 3- to 5-fold
on all crops of the rotation. Schuman et al. (729, 130)
found terraces reduced water, sediment, and total
nitrogen losses a little over 10-fold and phosphorus
losses a little less than 10-fold compared with contour
tillage. The enrichment ratio doubled from 2 to 4 and
the phosphorus concentration in solution doubled.
Farm ponds are constructed for a variety of reasons
such as stock watering, recreation, etc. They are also an
effective trap for sediment and nutrients (77 7). In some
soils, the ponds are difficult to seal and a local seepage
problem can be created.
Nutrient Management Practices
Erosion control practices will probably solve most of
the phosphate pollution problems and many of the
nitrogen pollution problems. These practices will have
less effect on controlling nutrients dissolved in runoff
than in sediment. They have no effect and may even
aggravate a nitrate leaching problem. In these cases, it is
necessary to use alternative or additional practices to
achieve the desired degree of control. These practices
involve changing the use of nutrients. Table 4 contains a
list of these practices and some of the references used in
the following discussion.
Eliminating Excessive Fertilization
For preventing nitrate leaching, Olson (776) sug-
gested that only enough nitrogen be applied to satisfy
the crop needs, that the soil's capacity for producing
nitrate be accounted for, that the nitrate already present
in the root zone be taken into account, and that
adequate levels of other nutrients be supplied so that
there is maximum efficiency. These suggestions can be
reduced to a single concept of eliminating excessive use
of fertilizer.
Nitrate builds up in the soil when excessive levels of
nitrogen fertilizers are used (750), But .when only
adequate amounts are used at the proper ti^ne, little of
the nitrogen is left after harvest (95, 99, 714,135).
The greatest difficulty in preventing excessive fertili-
zation is in predicting what levels of fertilizer should be
applied so that the resulting level in the soil is adequate.
The first requirement is to predict the potential yield of
the crop and thus the nutrient requirements. Then, the
soil's ability to meet these requirements must be
evaluated. Finally, the efficiency of the applied fertilizer
in meeting the remaining nutrient requirements must be
considered.
This difficulty in accurately predicting fertilizer needs
and low nitrogen costs has led some growers to
overfertilize so that lack of nutrients would not limit
yields. Recommendations based simply on "mainte-
nance" or "balance" approaches to replace nutrients
removed by the crop should be discouraged (720). They
fail to account for either the nutrient supplied by the
soil or the losses of applied fertilizers.
The yield of any crop and its response to applied
fertilizer depends upon many different soil, plant,
climatic, and cultural factors (759). For example,
experiment station reports from Maryland and Michigan
show the yield of corn can vary 2-fold across a single
state. Both climate and soil properties can be involved.
As the precipitation during the growing season decreases,
the water stored in the soil when the plant starts to grow
becomes the major yield determinant.
Stanford (141, 143) has published extensively on
estimating nitrogen fertilizer requirements. He argues
persuasively that there is an internal nitrogen require-
ment of the crop for the expected yield. To adequately
estimate this requirement requires considerable field
work. A first approximation can be made by considering
the expected yields and nutrient contents (Table 17,
Vol. I). Good farmers in fertile farming areas will
probably produce higher yields but it is anticipated that
the nutrient content of the crops will be proportionately
74
-------
Table 4. Bibliography fur nutrient management practices (Volume I, Section 4.3).
Nutrient management practice
No.
Page No.
in
Vol.1.
Description '
Citations
Significant subjects
General
Nl
76
Kliminating Fxcessive Fertili-
zation
Herronffo/(60)
Linville and Smith (95)
Petersen and Sander (120)
Stanford (143)
Stewart (150)
Thomas and Hanway (159)
Thomas and Peaslee (160)
Viet.s(168)
Nil rale already in soil
Little left with adequate amount*
Discourages maintenance and balanced
approaches
tstimating N fcrtili/er requirement
Build-up with excessive use
Response to fertilizer
Phosphorus recommendations
Implication of banning all fertiliser
Leaching Com rol
N2
N3
N4
N:7
78
79
79
80
83
83
Timing Fertilizer Application
Using Crop Rotations
Using Animal Wastes for
Fertilizer
Plowing-under Green Legume
Crops
Using Winter Cover Crops
Controlling Fertilizer Release
or Transformation
Aldrich M)
Bouldin, Reid, and
Lathwell (22)
Lathwell. Bouldin. and
Reid<2J)
Bezel icek. Mulford. and
Magee(F2)
Olsen (113)
Stewart, Viets, and
Hutchinson (152)
Ashraf (6)
Uttormark, Chapin, and
Green (166)
Zwerman et al (185)
Lyon. Buck man. and
Brady (97)
Frink (46)
Tliomas (156)
Anonymous (4)
Boswel) and Anderson (]_9)
Broadbent (25)
Hauck and Koshino (58)
Conditions for fall fertilization
Argument for summer sided ressim:
Period of maximum use
Soybeans don't need fertiliser N
Profile N proportional to amount applied
Alfalfa removes deep N
Cost of storage
N and P lost in snowmelt
Manure increases infiltration
References for amount of fixation
Reduced leaching by cover crops
Recommended planting time
Cost estimate
Field experiment with inhibitors
Poor future prospects
Advantages of slow release
75
-------
Table 4 (continued)
Nutrient management practice
No.
Page No.
in
Vol. I.
Description
Citations
Significant subjects
Control of Nutrients in Runoff
N8
N9
N10
83
83
83
Incorporating Surface Appli-
cations
Controlling Surface Appli-
cations
Using Legumes in Haylands and
Pastures
Timmons, Burwell, and
Holt (162)
Wagner and Jones ( 171 )
Allison (3)
Plow-down reduces fertilizer loss
Less-frequent P and K applications needed on
fertile soils
Grass uses N from legumes
Control of Nutrient Loss by Erosion
Nil
83
Timing Fertilizer Plow-down
None
higher. More accurate data for the area under con-
sideration are usually available from the State experi-
ment station.
Given a yield estimate and a crop requirement, the
next step is to estimate the amount of nitrogen the soil
will supply without fertilizer. There are several factors to
be considered. One is the capacity of the soil to produce
nitrate by mineralizing organic nitrogen in the soil. An
incubation method would appear to give a reliable
estimate of the potential (144, 148) which is adjusted
for temperature and moisture effects (146, 147). How-
ever, the time required for the incubation prohibits soil
testing laboratories from using it (33). A hot water or
steam extraction of the soil sample may provide an
adequate estimate of the potential mineralizable nitro-
gen (139,142).
Also to be considered is the amount of nitrate already
in the soil (60, 61). In the more humid regions, any
nitrate remaining in the soil after harvest will be leached
out of the root zone before the crop can use it the
following season. But in the more arid areas such as the
Great Rains, this leaching doesn't occur regularly. A
final factor that needs to be considered for the soil
supply is the amount of nitrogen that is available from
residues and/or cover crops.
The final step is to estimate the fraction of the
fertilizer that the crop will use. Many field experiments
show that the plant takes up less than 70 percent and
often less than 50 percent of the applied nitrogen
fertilizer (91, 96, 116). Some of the fertilizer is
immobilized into organic matter, some is denitrified, and
some can be leached out of the root zone. These changes
can be reduced to some extent by applying the fertilizer
when the plant is growing.
The behavior of nitrogen is quite complex and several
estimates are required to predict the amount of fertilizer
needed. A simpler, but often less accurate approach, is
presently used in most cases. This approach relies on the
results of previous experiments in the area where
different rates of nitrogen were applied to the crop.
Figure 5 is a summary of such an experiment (66). In
this particular case, the soil and residues supplied enough
nitrogen for a yield of 65 bu/acre. Applying nitrogen
fertilizer at the rate of 120 Ibs/acre produced 141
bu/acre, which was close to the maximum yield observed
(147 bu/acre). Such information would be the basis for
recommending that 120 pounds of nitrogen be applied
to corn under similar conditions.
The phosphorus cycle is less complicated than the
nitrogen cycle and the phosphorus fertilizer recom-
mendations are also much easier to make (760). While
less than 20 percent of the applied phosphate is usually
taken up because of the reactions with the soil, there are
essentially no losses by leaching or volatilization. In
addition, soil tests have been extensively correlated with
yield responses so that the fertilizer requirement is more
readily predicted from soil tests.
Reducing fertilizer application to a less-than-adequate
level doesn't always decrease pollution and may in fact
increase it. Inadequate fertilization decreases growth and
can increase runoff, erosion, and leaching. Smith (134)
reports a 9-fold increase in nitrate loss from inade-
quately fertilized corn. Viets (167) discusses the implica-
tions of banning the use of all fertilizer. The effect
76
-------
would range from very little with soybeans in Iowa to
over a 90 percent reduction in per acre yield of
grapefruit in Florida. Acreage would need to be in-
creased 20 to 30 percent for the major crops of corn,
wheat, and cotton in the first year while there was still
some residual fertility. In addition, the added acreage
would be of lower fertility and more erodible, thereby
creating additional problems.
Mayer and Hargrove (706)used an economic model to
examine the impact of restricting fertilization to certain
levels nationally and only in Iowa. Reduced use through-
out the country would eliminate foreign exports, in-
crease cropland acreage, and increase prices of farm
products. If only a single state such as Iowa restricted
fertilizer, the impact on the farmer would be very great,
since his yield per acre would be reduced but the price
for his product would not go up because of supplies
from adjacent states.
Timing Fertilizer Application
The time of the fertilizer application can be an
important tool for increasing the efficiency of fertilizer
use and reducing fertilizer loss. Fertilizer nitrogen use is
maximized when fertilizer is applied near the time of
maximum vegetative growth (21, 91). Most crops grow
the fastest several weeks after the plant emerges, as
illustrated by corn in Figure 33 of Volume I (56). The
application of nitrogen fertilizers several weeks after the
plant has started to grow is commonly called summer
sidedressing. Bouldin et al. (22) provide a number of
arguments for the summer sidedressing of corn based on
experiments in New York. Since the fertilizer is used
more efficiently, less fertilizer is needed and the lower
fertilizer cost offsets the added cost of application. They
argue that if the field is too wet for sidedressing, then
previously applied fertilizer will probably be lost by
175
150
125
100
2
o
I 75
o
o
50
25
40
80
120
160
200
240
280
32O
360
APPLIED NITROGEN, Ibs./acre
Figure 5.-An example of the yield response of corn to applied nitrogen (66).
77
-------
leaching or denitrification. Conversely, if it is too dry to
move the nitrate down to the roots, then the growth will
be retarded by lack of water anyway.
If leaching is not a problem, then applying fertilizers
preplan! in the spring or even in the fall (except on
sandy soils) may be acceptable. For fall fertilization it is
usually recommended that an ammonium type fertilizer
be applied after the soil cools below 50° F (7). This
recommendation is based on the facts that while nitrate
is mobile, ammonium is relatively immobile and is
converted to nitrate very slowly below 50° F (40, 126,
182).
Nelson and Uhland (112) were among the first to
show regional variation in leaching. Using Thorn-
thwaite's calculations with a constant 4 inches of
water-holding capacity and implicitly accounting for the
temperature effects on nitrification, they divided the
area east of the Rocky Mountains into four regions of
different leaching potential (Fig. 20, Appendix B).
We have attempted to provide a more detailed
mapping by combining a nitrification model with a
percolation model. These models are described in detail
in Appendix B. The results of the simulations are also
presented in Appendix B as maps of percentage loss for
various soil groups. These maps can be used to estimate
average leaching losses for an area if the hydrologic
characteristics of the soils are similar to one of the
groups modeled. Losses of fall-applied and spring-applied
ammonia (Figs. 34 and 35, Vol. I) by nitrate leaching
were prepared from these maps by selecting the appro-
priate loss for the predominant soil group in each Land
Resource Area.
These results still represent coarse approximations in
spite of the numerous factors that were incorporated
into the model. Only a few combinations of soil
characteristics were modeled. Another limitation was the
relation used for the ammonium-nitrate conversion.
Other factors such as moisture and pH could modify the
actual conversion rate for a particular soil. Also, immobi-
lization and denitrification could remove some of the
nitrate produced, and the leaching process may not be
exactly plug flow. The model assumes little transpiration
and no uptake during the winter, which would tend to
overestimate leaching in southern states where there
could be plant growth. Thus, these maps can serve only
as first approximations and more detailed information
on locally important variations must be obtained from
the Soil Conservation Service, State experiment stations,
and extension staffs for each area.
Split applications, applying part of the nitrogen in the
fall or spring and the rest as a summer sidedressing,
combines some of the features of each time of applica-
tion. The first application provides enough fertilizer for
a poor year. The last application is not large enough to
cause toxicity problems that sometimes occur and it
provides an opportunity to adjust for favorable weather,
increased plant population, and optimum planting dates.
Using Crop Rotations
Crop rotations can be used to reduce the average
amount of nitrogen fertilizer required. High nitrogen-
requiring crops such as corn, cotton, and sorghum can be
rotated with crops requiring less nitrogen, such as small
grains, or legumes which require only small amounts of
starter fertilizer, such as soybeans or alfalfa. Olsen (113)
found the amount of nitrogen in the profile was
proportional to the amount of nitrogen applied during
the rotation. Thus, the average nitrogen content in
drainage from a watershed with diversified crops should
be lower than if the watershed were completely in crops
like corn.
Alfalfa is particularly useful because its deep root
system can remove some nitrate from deeper depths
than most crops can (752). Soybeans are high cash value
legumes that don't require nitrogen fertilization but
appear to respond to high levels of soil nitrogen from
previous crops (12, 175). The major limitations in crop
rotation are the loss of cash income and/or the cost of
additional equipment.
Using Animal Wastes for Fertilizer
Animal wastes, or manure, have been used as a source
of plant nutrients for thousands of years. A previous
section discussed many of the properties of manure.
Zwerman et al. (184, 185) report the increased infiltra-
tion and reduced runoff from long periods of manure
use. This section will be concerned with the problems
associated with using manure as a substitute for fertil-
izer.
The most serious problem from a water quality
standpoint is the loss of nutrients in runoff. Animal
manure produced during the winter must be either
stored or applied when the crops are not growing and
chances of loss are greater. Since equipment can't enter
fields that are wet with fall or spring rains, farmers with
little or no storage capacity are forced to spread the
manure on frozen or snow-covered fields. The resulting
runoff from rains or snowmelt can carry 10 to 20
percent of the nitrogen and phosphorus in the manure
(109, 166). The losses from a manure application
containing 100 pounds of N per acre are 10 to 20
pounds of N per acre and 3 to 10 pounds of P per acre
(107).
78
-------
Flowing-down manure soon after application is the
most appropriate method of controlling losses from
broadcast applications. This method also prevenls nitro-
gen loss as volatile ammonia. However, meadows and
haylands can't be plowed, nor can frozen croplands. For
these cases, the State of Maine guidelines (101) recom-
mend that only upland fields with less than 3 percent
slope be used for manure spreading when frozen or snow
covered. They also recommend against spreading on any
fields with slopes greater than 25 percent or within 100
feet of wells, springs, ponds, or lakes, or when there is a
high possibility of runoff.
While most manure is in a relatively solid form, some
stored wastes are in a slurry form. Slurries, principally
from dairy and swine operations, can be injected directly
into the soil, thus almost eliminating runoff losses. A
large part of the nutrients are in solution and can move
into the soil even if surface applied. Storage of manure is
a large added expense with little economic return. The
investment needed for manure storage on a dairy farm
has been estimated as three to five times the cost of
daily spreading (6).
A second problem of substituting manure for fertil-
izer is determining how much nitrogen is being applied.
Not only does the nutrient content of manure change
with different animals, but also with their feed and
environmental conditions. If bedding is used, this pro-
vides added bulk with no added nutrients, but it absorbs
liquids and prevents nutrient loss. The largest losses are
probably by volatilization of nitrogen during storage,
spreading, and before incorporation. As much as 40
percent of the nutrient value can be lost by delaying
incorporation for 4 days (127). Ammonia gas is con-
tinually being lost, while denitrification of nitrate and
nitrite occurs in anaerobic storage.
Other problems include the fact that not all the
nitrogen is available during the first growing season.
Organic nitrogen compounds are similar to the soil
organic matter in that some are more easily mineralized
by microbes than others. This is not too serious since
repeated yearly applications will build the organic
nitrogen up so that the total mineralized is equivalent to
the amount added. The nutrients may not be in the
proper ratio fora particular crop on that field. However,
this can easily be corrected by adding nutrients to the
'icld or to the manure in storage.
?lowing-Under Green Legume Crops
This practice was frequently used to supply nitrogen
before the development of commercial fertilizers, but
little research has been done on it since commercial
fertilizers became available. Thus, the principal sources
of information are older soil science books (97, 108).
The practice is based on the symbiotic relation between
some types of microorganisms and legume plums in
which nitrogen from the atmosphere is converted into
plant protein. From 40 to 60 pounds of N can be
supplied per ton of dry forage. Part of the nitrogen in
the plant comes from mineralized soil nitrogen, but this
is probably balanced by not considering the nitrogen in
the plant roots. Obviously, the greatest limitation of this
practice is loss of any return from this crop. If the forage
is harvested, then the net gain in soil nitrogen is small.
Using Winter Cover Crops
Although winter cover crops arc recommended for
control of soil erosion during the fall and winter (see
Erosion Control Practices), they can also reduce nitrate
leaching through plant uptake of nitrate and reduce
percolation by drying the soil out. An oat crop reduced
nitrate leaching 4-fold on one soil and eliminated it on
another. Vetch, however, reduced leaching only slightly
and because it is a legume, added nitrogen (14). Cover
crops of oats, timothy, and rye reduced leaching 40 to
60 percent (46). Thomas (156) recommends that the
cover crop be planted by October for the most effective
control of leaching.
The major expense of this practice is planting the
crop. Some economic return can be obtained by using
the crop for winter grazing. Also, in some areas it is
possible to double crop; that is, grow a winter grain such
as wheat and then plant a short season crop such as
soybeans. A serious limitation of cover crops is that they
can remove so much of the soil water that the main
summer crop suffers, particularly in a dry year.
Controlling Fertilizer Release or
Transformation
Many researchers have explored the possibility of
controlling fertilizer release or availability. Recently the
interest has been very great and over 50 papers were
presented at the 1974 Annual Meeting of the American
Society of Agronomy dealing with this subject. Two
basic approaches are being used: a slow release fertilizer
and a nitrification inhibitor.
Slow release fertilizers offer three advantages: i)
reduction in nutrient loss by leaching and runoff, ii)
reduction in immobilization before plant uptake, and iii)
reduction in losses by denitrification and volatilization
(58, 77, 119, 122). Three processes are used to slow the
release of the fertilizer from the granule: i) controlling
dissolution by a physical barrier, ii) using compounds of
-------
limited water solubility, and iii) using a barrier that
decomposes. Of the 13 different slow release fertilizers
developed, sulfur-coated urea seems to be the most
promising. Most of the commercial production is being
used on turf. The present cost is 25 to 40 percent more
than uncoated urea (4). The greatest problem is to
control the release so that the fertilizer is available when
the plant needs it. If the release doesn't occur in a short
period of time for row crops, then the remaining N is
susceptible for leaching after harvest. The future looks
promising, but more research is needed before large scale
recommendations can be made.
Nitrification inhibitors are chemicals that prevent
microbes from converting ammonium to nitrate. Five
chemicals olfer possibilities. The most widely tested are:
2-chloi o-6-( t richloromethyl) pyridine, sold as
N SERVE by Dow Chemical; 2-amino-4-chIoro-6-
rnethy I pyridine, sold as AM by Mitsui-Toatsu Industries;
and sodium a/ide. Experiments with soil in plastic bags
in the field from November to April show that most of
the conversion was prevented by N-SERVE (19). One of
the greatest difficulties has been to keep the inhibitor
near the ammonium; usually percolating water separates
them. Broadbent (25) doesn't consider the prospects of
developing a practical method to be very good.
Incorporating Surface Applications
Immediate incorporation of surface-applied fertilizers
and manure can prevent significant losses of nutrients. A
number of studies have shown that the losses are greatest
when the runoff occurs soon after application (see the
section on runoff losses). Timmons et al. (162) report
that deep incorporation of the fertilizer by plowing
clown and subsequent disking reduced the nutrient losses
to levels similar to those in runoff from unfertilized
plots. Broadcasting on a plowed surface is adequate if no
additional tillage is performed because the infiltration
is very high. Disking instead of plowing broadcasted
fertilizer was not effective. Up to 30 tons/acre of
manure have been incorporated into the soil with little
increase in the nitrate and ammonium contents of the
tail water from irrigation (151).
Controlling Surface Applications
The time of application and the type of fertilizer can
be controlled to some extent. Fertilizer should not be
applied during periods of expected runoff. Fall-seeded
grains are often top-dressed with fertilizer in the spring.
If leaching is not a problem, then fertilization at planting
would reduce runoff losses. Pasture and haylands usually
require surface application of fertilizers and thus runoff
losses could be a problem. Wagner and Jones (777)
report that if a high level of fertility is maintained, then
timing of phosphorus and potassium fertili/er applica-
tions is not critical. In fact, on slightly deficient soils
enough P and K can be plowed under when the forage is
planted to last 4 or 5 years. Since nitrogen is so mobile,
the greatest efficiency is obtained by applying it shortly
before or during the growing season. Cool season forages
such as the brome and blue grasses make their growth in
the spring and fall, whereas warm season grasses such as
the Bermudagrasses make their best growth during the
summer.
Spraying a fluid fertilizer on the surface might have
lower losses in runoff than broadcasting granular ferti-
lizers even though there is no published research on the
subject. Fluid fertilizers are liquids or suspensions of
micro crystals and therefore should come in quicker
contact with the soil than granules that must dissolve.
Ammonia cannot be used because it volatilizes too
easily. Urea is also subject to some losses because it is
converted to ammonium.
Using Legumes in Haylands and Pastures
Legumes can be planted with grass in pastures and
haylands to supply much of the nitrogen requirement
and thus reduce the need to fertilize with nitrogen. The
legumes can be very effective in this situation because
the grass receives nitrogen in leakage from the legumes as
well as the mineralized soil nitrogen (3). As pointed out
by Kilmer (76), leaching losses will be higher from
legumes than from grasses. However, if leaching is not a
problem and runoff losses are, then legumes can be used
effectively. A major problem is that competition for
other nutrients, water, and sunlight causes the grass to
crowd out the legume in a few years.
Timing Fertilizer Plow-Down
When nutrients are being lost by sediment transport,
erosion control practices are the obvious answer in most
cases. An additional procedure that can be recom-
mended is plowing during the least erosive period and
leaving the field in the least erosive condition. For
example, if erosion is less in the fall than in the spring,
phosphorus and potassium fertilizer might be plowed
under in the fall using stubble mulching techniques or
followed by a cover crop so that an erosive period in the
spring can be avoided.
80
-------
Mathematical Models
The soil-plant system is very complex and dynamic
and therefore the impact of various management prac-
tices can vary considerably. One of the most efficient
ways of testing alternatives on complex and dynamic
systems is to describe the system with a mathematical
model. If (he model adequately represents the behavior
of the system for known responses and is based on
sound fundamental relations, then its responses for
various alternatives can be quickly tested with modern
computer equipment.
A number of programs have been started in recent
years to develop models for sediment and chemical
transport from watersheds. Most of these efforts are still
in the development stage and have yet to be adequately
tested against field results.
AC'TMO, an agricultural chemical transport model
(44, 45} linked together a hydrology model (6,5), an
erosion model (//
-------
LITERATURE CITED
1. Aldrich, S. R. 1970. The influence of cropping
patterns, soil management and fertilizer on nitrates.
Proc. Twelfth Sanitary Engin. Conf., Univ. Illinois,
Urbana:152-169.
2. Alexander, M., Moyle, J. B., Patrick, R., Weinberger,
L., and O'Bryan, D. 1965. Aquatic blooms, Appen-
dix Y9. In Restoring the quality of our environ-
ment. Environ. Poll. Panel, President's Science
Advisory Committee, The White House.
3. Allison, F. E. 1973. Soil organic matter and its role
in crop production. Elsevier Scientific Pub. Co.,
Amsterdam, The Netherlands.
4. Anonymous. 1974. Enter: Sulfur-coated urea, a
slow-release fertilizer. Chem. Engin., 32 and 34.
5. Armitage, E. R. 1974. The runoff of fertilizers from
agricultural land and their effects on the natural
environment. In D. E. G. Irvine and B. Knights, eds.,
Pollution and the use of chemicals in agriculture.
Butterworths Pub., London.
6. Ashraf, M. 1974. An analysis of the impact of
manure disposal regulations on dairy farms. Amer.
Jour. Agr. Econ. 56(2): 331-336.
7. Baeumer, K., and Bakermans, W. A. P. 1973.
Zero-tillage. Advan. in Agron. 25: 77-123.
8. Barnett, A. P., Carreker, J. R., Abruna, F., Jackson,
W. A., Dooley, A. E., and Holladay, J. H. 1972. Soil
and nutrient losses in runoff with selected cropping
treatments on tropical soils. Agron. Jour. 64:
391-395.
9. Barrows, H. L., and Kilmer, V. J. 1963. Plant
nutrient losses from soils by water erosion. Advan.
in Agron. 15:303-316.
10. Bartholomew, W. V., and Clark, F. E. (Eds). 1965.
Soil nitrogen. Amer. Soc. Agron., Madison, Wis.
11. Bedell, G. D., Kohnke, H., and Hickok, R. B. 1946.
The effects of two farming systems on erosion from
cropland. Soil Sci. Soc. Amer. Proc. 11: 522-526.
12. Bezdicek, D. F., Mulford, R. F., and Magee, B. H.
1974. Influence of organic nitrogen on soil nitrogen
nodulation, nitrogen fixation, and yield of soy-
beans. Soil Sci. Soc. Amer. Proc. 38(2): 268-273.
13. Biggar, J. W., and Corey, R. B. 1969. Agricultural
drainage and eutrophication. In Eutrophication:
causes, consequences, correctives. Natl. Acad. Sci.,
Washington, D. C.: 404-445.
14. Black, C. A. 1966. Crop yields in relation to water
supply and soil fertility./« W. Pierre, D. Kirkham, J.
Pesek, and R. Shaw, eds. Plant environment and
efficient water use. Amer. Soc. Agron., Madison,
Wis.: 177-206.
15. Black, C. A. 1970. Behavior of soil and fertilizer
phosphorus in relation to water pollution. In T. L.
Willrich and G. E. Smith, eds. Agricultural practices
and water quality. Iowa State Univ. Press, Ames,
la.: 72-93.
16. Blanchar, R. W., and Kao, C. W. 1971. Effects of
recent and past phosphate fertilization on the
amount of phosphorus percolating through soil
profiles into subsurface waters. Missouri Water
Resour. Res. Ctr. Proj. No. A-031-Mo.
17. Borchardt, J. A. 1970. Secondary effects of nitro-
gen in water. In Nitrate and water supply: source
and control. Proc. Twelfth Sanitary Engin. Conf.,
Univ. Illinois, Urbana, 111.: 66-76.
18. Bormann, F. H., Likens, G. E., Siccama, T. G.,
Pierce, R. S., and Eaten, J. S. 1974. The export of
nutrients and recovery of stable conditions follow-
ing deforestation at Hubbard Brook. Ecol. Monogr.
44(3): 255-277.
19. Boswell, F. C., and Anderson, 0. E. 1974. Nitrifica-
tion inhibitor studies of soil in field-buried poly-
ethylene bags. Soil Sci. Soc. Amer. Proc. 38(5):
851-852.
20. Bouldin, D. R., Johnson, R. L., Burda, C., and Kao,
C. W. 1974. Losses of inorganic nitrogen from
aquatic systems. Jour. Environ. Qual. 3(2):
107-114.
82
-------
21. Bouldin, D. R., and Lathwell, D. J. 1968. Timing of
application is key to crop nitrogen use. Food and
Life Sci. 1:9-12.
22. Bouldin, D. R., Reid, W. S., and Lathwell, D. J.
1971. Fertilizer practices which minimize nutrient
loss. In Agricultural wastes: principles and guide-
lines for practical solutions: 21-35. Cornell Univ.,
Ithaca, N. Y.
23. Bradford, R. R. 1974. Nitrogen and phosphorus
losses from agronomy plots in north Alabama.
EPA-660/2-74-033. U. S. Govt. Printing Off., Wash-
ington, D. C.
24. Brezonik, P. L. 1972. Nitrogen: Sources and trans-
formations in natural waters. In H. E. Allen and J.
R. Kramer, eds. Nutrients in natural waters: 1-50.
John Wiley and Sons, New York.
25. Broadbent, F. E. 1971. Nitrogen in soil and water.
In Proc. symp. on nitrogen in soil and water:56-77.
Dept. Soil Sci., Univ. Guelph, Ontario.
26. Broadbent, F. E., and Clark, F. E. 1965. Denitrifica-
tion. In W. V. Bartholomew and F. E. Clark, eds.
Soil nitrogen. Amer. Soc. Agron., Madison, Wis.
27. Burwell, R. E., Timmons, D. R., and Holt, R. F.
1975. Nutrient transport in surface runoff as influ-
enced by soil cover and seasonal periods. Soil Sci.
Soc. Amer. Proc. (In press).
28. Case, A. A. 1970. The health effects of nitrates in
water. In Nitrate and water supply: source and
control. Proc. Twelfth Sanitary Engin. Conf., Univ.
Illinois, Urbana, El: 40-46.
29. Committee on Agriculture and the Environment.
1973. Productive agriculture and a quality environ-
ment. Agr. Board, Div. Biol. and Agr., Natl. Res.
Council.
30. Cornell University. 1969. Animal waste manage-
ment. Conf. on Agr. Waste Management, Syracuse,
N.Y.
31. Crawford, N. H., and Donigian, A. S.Jr. 1973.
Pesticide transport and runoff model for agricultural
lands. EPA-660/2-74-013, 211 p. U. S. Govt. Print-
ing Off., Washington, D. C.
32. Cywin, A. and Ward, D. 1971. Agricultural pollu-
tion of the Great Lakes basin. Combined report by
Canada and the United States. EPA-13020-7/71.
33. Dahnke, W. C., and Vasey, E. H. 1973. Testing soils
for nitrogen. In L. M. Walsh and J. D. Beaten, eds.
Soil testing and plant analysis: 97-114. Soil Sci. Soc.
Amer., Inc., Madison, Wis.
34. Doty, C. W., and Carter, C. E. 1965. Rates and
particle-size distributions of soil erosion from unit
sources areas. Trans. ASAE 8(3): 309-311.
35. Downing, K. M., and Merkens, J. C. 1955. The
influence of dissolved oxygen concentration on the
toxicity of unionized ammonia to rainbow trout.
Ann. Appl. Biol. 43: 243-246.
36. Dunigan, E. P., Mondart, C. L., Jr., Phelan, R. A.,
and Shamsuddin, Z. H. 1974. Surface runoff losses
of fertilizers. Louisiana Agr., Summer 17(4): 4-7.
37. Eckert, R. E., Jr., Klomp, G. J., Young, J. A., and
Evans, R. A. 1970. Nitrate-nitrogen status of fal-
lowed rangeland soils. Jour. Range Manag. 23(6):
445^47.
38. Engibous, J. C. 1972. Economics of fertilizer use.In
The fertilizer handbook: 143-153. The Fertilizer
Institute, Washington, D. C.
39. Fitzgerald, G. P. 1972. Bioassay analysis of nutrient
availability. In H. E. Allen and J. R. Kramer, eds.
Nutrients in natural waters: 147-169. John Wiley
and Sons, New York.
40. Frederick, L. R. 1956. The formation of nitrate
from ammonium nitrogen in soils: I. Effect of
temperature. Soil Sci. Soc. Amer. Proc. 20:
496-500.
41. Frere, M. H. 1971. Requisite sampling frequency for
measuring nutrient and pesticide movement with
runoff waters. Jour. Agr. Food Chem. 19(5):
837-839.
42. Frere, M. H. 1975. Integrating chemical factors with
water and sediment transport from a watershed.
Jour. Environ. Qual. 4(1): 12-17.
83
-------
43. Frere, M. H., Jensen, M. E., and Carter, J. N. 1970.
Modeling water and nitrogen behavior in the soil-
plant system. Proc. 1970 Summer Computer Simu-
lation Conf.: 746-750.
44. Frere, M. H., Onstad, C. A., and Holtan, H. N. 1974.
Modeling the movement of agricultural chemicals.
Proc. 1974 Summer Computer Simulation Conf.:
271-274.
45. Frere, M. H., Onstad, C. A., and Holtan, H. N. 1975.
ACTMO, an agricultural chemical transport model.
ARS-H-3, Agr. Res. Serv., U. S. Dept. Agr., Washing-
ton, D. C.
46. Frink, C. R. 1970. The nitrogen cycle of a dairy
farm. In Relationship of agriculture to soil and
water pollution: 127-133. Cornell Univ., Ithaca,
N.Y.
47. Frink, C. R. 1971. Plant nutrients and water
quality. Agr. Sci. Rev. 9(2): 11-25.
48. Fromm, P. O. 1970. Toxic action of water soluble
pollutants on fresh-water fish. U. S. Water Qual.
Off., Environ. Protec. Agcy. Grant No. 18050 DST.
49. Fry, F. E. J. 1969. Some possible physiological
stresses induced by eutrophication. In Eutrophi-
cation: causes, consequences, correctives: 531-536.
Natl. Acad. Sci., Washington, D. C.
50. Gambell, A. W., and Fisher, D. W. 1964. Occurrence
of sulfate and nitrate in rainfall. Jour. Geophys.
Res. 69(20): 4203-4210.
51. Gburek, W. J., and Heald, W. R. 1970. Effects of
direct runoff from agricultural land on the water
quality of small streams. In Relationship of agricul-
ture to soil and water pollution: 61-68. Cornell
Univ., Ithaca, N. Y.
52. Ghoshal, S. 1974. Fate of fertilizer phosphorus
under aerobic decomposition. Plant and Soil 40(3):
685-688.
53. Goering, J. J. 1972. The role of nitrogen in
eutropnic processes. In Water pollution micro-
biology: 43-68. Wiley-Interscience, New York.
54. Hagin, J., and Amberger, A. 1974. Contribution of
fertilizers and manures to the N- and P-load of
waters. Final report submitted to the Deutsche
Forschungs Gemeinschaft.
55. Hanawalt, R. B. 1969. Environmental factors influ-
encing the sorption of atmospheric ammonia by
soils. Proc. Soil Sci. Soc. Amer. 33(2): 231-234.
56. Hanway, J. J. 1966. How a corn plant grows. Spec.
Rep. No. 48. Iowa State Univ., Ames, la.
57. Harmeson, R. H., Larson, T. E., Henly, L. M.,
Sinclair, R. A., and Neill, J. C. 1973. Quality of
surface water in Illinois, 1966-71. Bui. 56, Dept.
Registration and Education, 111. State Water Surv.
58. Hauck, R. D. and Koshino, M. 1971. Slow-release
and amended fertilizers. In R. A. Olson, T. J. Army,
J. J. Hanway, and V. J. Kilmer, eds. Fertilizer
technology and use: 455494. Soil Sci. Soc. Amer.,
Inc., Madison, Wis.
59. Heinemann, H. G., Holt, R. F., and Rausch, D. L.
1973. Sediment and nutrient research on selected
corn belt reservoirs. In W. C. Ackermann, G. F.
White, and E. B. Worthington, eds. Man-made lakes:
their problems and environmental effects:381-386.
Amer. Geophys. Union, Geophys. Monogr. 17.
60. Herron, G. M., Dreier, A. F., Flowerday, A. D.,
Colville, W. L., and Olson, R. A. 1971. Residual
mineral N accumulation in soil and its utilization by
irrigated com. Agron. Jour. 63: 322-327.
61. Herron, G. M.,.Terman, G. L., Dreier, A. F., and
Olson, R. A. 1968. Residual nitrate nitrogen in
fertilized deep loess-derived soils. Agron. Jour. 60:
477-482.
62. Hildreth, R. J., and Williams, M. S. 1968. Fertilizer
use economics. In L. B. Nelson, M. H. McVickar, W.
C. White, R. D. Munson, L. F. Seatz, and S. L.
Tisdale, eds. Changing patterns in fertilizer use.'433.
Soil Sci. Soc. Amer., Inc., Madison, Wis.
63. Holt, R. F. 1969. Runoff and sediment as nutrient
sources. In Water pollution by nutrients-sources,
effects and control. Bui. 13: 35-38. Minnesota
Water Resour. Res. Ctr., Univ. Minn., Minneapolis.
84
-------
64. Holt, R. F., Dowdy, R. H., and Timmons, D. R.
1970. Chemistry of sediment in water. In T. L.
Willrich and G. E. Smith, eds. Agricultural practices
and water quality: 21-34. Iowa State Univ. Press,
Ames, la.
65. Holtan, H. N., and Lopez, N. C. 1971. USDAHL-70
model of watershed hydrology. ARS Tech. Bui. No.
1435, U. S. Dept. Agr., Washington, D. C.
66. Hunter, A. S., and Yungen, J. A. 1955. The
influence of variations in fertility levels upon the
yield and protein content of field corn in eastern
Oregon. Soil Sci. Soc. Amer. Proc. 19: 214-218.
67. Hutchinson, G. L., Millington, R. J., and Peters, D.
B. 1972. Atmospheric ammonia: absorption by
plant leaves. Science 175: 771-772.
68. Hutchinson, G. L., and Viets, F. G., Jr. 1969.
Nitrogen enrichment of surface water by absorption
of ammonia volatilized from cattle feed lots. Sci-
ence 166: 514-515.
69. Hynes, H. B. N. 1969. The enrichment of streams.
In Eutrophication: causes, consequences, correc-
tives: 188-196. Natl. Acad. Sci., Washington, D. C.
70. Ibach, D. B., and Mahan, J. N. 1968. Fertilizer use
patterns and potentials. In L. B. Nelson, M. H.
McVickar, R. D. Munson, L. F. Seatz, S. L. Tisdale,
and W. C. White, eds. Changing patterns in fertilizer
use: 1-31. Soil Sci. Soc. Amer., Inc., Madison, Wis.
71. Jaworski, N. A., and Hetling, L. J. 1970. Relative
contributions of nutrients to the Potomac River
basin from various sources. In Relationship of
agriculture to soil and water pollution. Cornell
Univ., Ithaca, N.Y.
72. Johnson, J. D., and Straub, C. P. 1971. Develop-
ment of a mathematical model to predict the role of
surfacp runoff and ground water flow in over-fertili-
zation of surface waters. Minnesota Water Resour.
Res. Ctr., Univ. Minn., Minneapolis.
73. Junge, C. E. 1958. The distribution of ammonia and
nitrate in rain water over the United States. Trans.
Amer. Geophys. Union 39(2): 241-248.
74. Ketcheson, J. W., and Onderdonk, J. J. 1973. Effect
of corn stover on phosphorus in runoff from
nontilled soil. Agron. Jour. 65(1): 69-71.
75. Keup, L. E. 1968. Phosphorus in flowing waters.
Water Res. 2(5): 373-386.
76. Kilmer, V. J. 1974. Nutrient losses from grasslands
through leaching and runoff. In D. A. Mays, ed.
Forage fertilization: 341-362. Amer. Soc. Agron.,
Madison, Wis.
77. Kilmer, V. J., and Barber, J. C. 1971. Fertilizer use
and misuse. In Proc. symp. on nitrogen in soil and
water. Dept. Soil Sci., Univ. Guelph, Ontario.
78. Kilmer, V. J., Davis, C. H., Douglas, J., Harre, E.,
Jones, U. S., McVickar, M. H., Nelson, L. B., Reed,
J. F., and White, W. C. 1974. The U. S. fertilizer
situation and outlook. A report by a task force of
the Council for Agr. Sci. and Technol., Ames, la.
79. Kilmer, V. J., Gillian, J. W., Joyce, R. T., and Leitz,
J. F. 1972. Loss of fertilizer nutrients from soils to
drainage waters. I. Studies on grassed watersheds in
western North Carolina. Water Resour. Res. Inst.
North Carolina, Rep. No. 55.
80. Kissel, D. E., Ritchie, J. T., and Burnett, E. 1973.
Chloride movement in undisturbed swelling clay
soil. Soil Sci. Soc. Amer. Proc. 37(1): 21-24.
81. Kling, G. F. 1974. A computer model of diffuse
sources of sediment and phosphorus moving into a
lake. PhD Thesis, Cornell Univ., New York.
82. Klingebiel, A. A. 1972. Soil and water management
to control plant nutrients in natural waters. In
Effects of intensive fertilizer use on the human
environment: 153-178. Soils Bui. No. 16, FAO.
Pub. by Swedish Internatl. Devlpmt. Authority.
83. Kolenbrander, G. J. 1969. Nitrate content and
nitrogen loss in drain water. Neth. Jour. Agric. Sci.
17: 246-255.
84. Kolenbrander, G. J. 1970. Calculation of parameters
for the evaluation of the leaching of salts under field
conditions, illustrated by nitrate. Plant and Soil 32:
439-453.
85
-------
85. Kolenbrander, G. J. 1972. Eutrophication from
agriculture with special reference to fertilizers and
animal waste. In Effects of intensive fertilizer use on
the human environment: 305-327. Soils Bui. No.
16, FAO. Pub. by Swedish Internatl. Devlpmt.
Authority.
86. Kramer, J. R., Herbes, S. E., and Allen, H. E. 1972.
Phosphorus: analysis of water, biomass, and sedi-
ment. In H. E. Allen and J. R. Kramer, eds.
Nutrients in natural waters: 51-100. John Wiley and
Sons, New York.
94. Levin, I. 1964. Movement of added nitrates through
soil columns and undisturbed soil profiles. Eighth
Internatl. Congr. Soil Sci., Bucharest, Romania:
1011-1022.
95. Linville, K. W., and Smith, G. E. 1971. Nitrate
content of soil cores from corn plots after repeated
nitrogen fertilization. Soil Sci. 112(4): 249-255.
96. Loehr, R. C. 1974. Agricultural waste management.
Academic Press, Inc., New York.
87. Kreitler, G. W., and Jones, D. C. 1974. Natural soil
nitrate: the cause of the nitrate contamination of
ground water in Runnels County, Texas. Proc.
Second Natl. Ground Water Quality Symp., Denver,
Col.: 179-188.
88. Kunishi, H. M., Taylor, A. W., Heald, W. R.,
Gburek, W. J., and Weaver, R. N. 1972. Phosphate
movement from an agricultural watershed during
two rainfall periods. Jour. Agr. Food Chem. 20:
900-905.
89. Kurtz, L. T. 1970. The fate of applied nutrients in
soils. Jour. Agr. Food Chem. 18: 773-780.
97. Lyon, T. L., Buckman, H. 0., and Brady, N. C.
1952. The nature and properties of soil. The
MacMillan Co., New York.
98. McCarthy, P. L., Hem, J. J., Jenkins, D., Lee, G. F.,
Morgan, J. J., Robertson, R. S., Schmidt, R. W.,
Symons, J. M., and Trexler, M. V. 1967. Sources of
nitrogen and phosphorus in water supplies. Jour.
Amer. Water Works Assoc. 59: 344-366.
99. MacGregor, J. M., Blake, G. R., and Evans, S. D.
1974. Mineral nitrogen movement into subsoils
following continued annual fertilization for corn.
Soil Sci. Soc. Amer. Proc. 38(1): 110-112.
90. Larsen, S., Gunary, D.,and Sutton, C. D. 1965. The
rate of immobilization of applied phosphate in
relation to soil properties. Jour. Soil Sci. 16:
141-148.
91. Lathwell, D. J., Bouldin, D. R., and Reid, W. S.
1970. Effects of nitrogen fertilizer application in
agriculture. In Relationship of agriculture to soil
and water pollution: 192-206. Cornell Univ., Ithaca,
N.Y.
92. Lee, D. H. K. 1970. Nitrates, nitrites, and methemo-
giobinemia. Environ. Res. 3(5/6): 484-511.
100. Mackenthun, K. M. 1965. Nitrogen and phos-
phorus in water. U. S. Dept. Health, Education and
Welfare, Washington, D. C.
101. Maine Experiment Station. 1972. Maine guidelines
for manure and manure sludge disposal on land.
Misc. Rep. No. 142, Agr. Expt. Sta., Orono, Me.
102. Martin, W. P., Fenster, W. E., and Hanson, L. D.
1970. Fertilizer management for pollution control.
In T. L. Willrich and G. E. Smith, eds. Agricultural
practices and water quality. Iowa State Univ. Press,
Ames, la.
93. Lee, G. F. 1973. Role of phosphorus in eutrophica-
tion and diffuse source control. In S. H. Jenkins and
K. J. Ives, eds. Phosphorus in fresh water and the
marine environment: 111-128. Pergamon Press,
Oxford, England.
103. Mason, H. C. 1974. The need for chemical control
of pests and the use of fertilizer. In D. E. G. Irvine
and B. Knights, eds. Pollution and the use of
chemicals in agriculture. Butterworths Pub.,
London.
86
-------
104. Massey, H. F and Jackson, M. L. 1952. Selective
erosion of soil fertility constituents. Soil Sci. Soc.
Amer. Proc. 16: 353-356.
105. Massey, H. F., Jackson. M. L., and Hayes, 0. E.
1953. Fertility erosion on two Wisconsin soils.
Agron. Jour. 45: 543-547.
106. Mayer. L. V., and Hargrove, S. H. 1971. Food
costs, farm incomes, and crop yields with restric-
tions on fertilizer use. Dept. Econ., Center for Agr.
and Econ. Development, Iowa State Univ., CAED
Report No. 38, Ames, la.
107. Midgley, A. R. and Dunklee, D. E. 1945. Fertility
runoff losses from manure spread during the
winter. Vermont Agr. Expt. Sta. B.ul. No. 523.
108. Millar, C. E., and Turk. L. M. 1951. Fundamentals
of soil science. John Wiley and Sons. New York.
115. Olsen, S. R.,and Flowerday. A. D. 1971. Fcnili/cr
phosphorus interactions in alkaline soils. In R. A.
Olson, T. J. Army, J. J. Hanway. and V. J. Kilmer.
eds. Fertilizer technology and use: 153-185. Soil
Sci. Soc. Amer., Inc., Madison, Wis.
116. Olson, R. A. 1972. Maximixing the utilization
efficiency of fertili/er N by soil and crop manage-
ment. In Effects of intensive fertili/.er use on the
human environment: 34-52. Soils Bui. No 16,
FAO. Pub. by Swedish Intcrnatl. Devlpmt. Au-
thority.
117. Olson, R. A., Seim, E. C.. Muir, J.,and Moshcr, P.
N. 1972. Influence of fertili/.cr practices on water
and the quality of the environment. Univ.
Nebraska, Proj. No. B-004-NEB. Rep. No.
PB211081, Natl. Tech. Inform. Scrv.. Springfield.
Va.
109. Minshall, N. E. 1970. Stream enrichment from
farm operations. Jour. Sanitary Eng. Div., Proc.
ASCE96(SA2): 513-524.
110. Murphy, T. J., and Cesarotti, D. J. 1974. Phos-
phorus inputs from the atmosphere. W75-01199.
OWRT-A-065-ILL(1). Rep. No. PB237-341/3WP.
Natl. Tech. Inform. Serv., Springfield, Va.
111. National Academy of Sciences. 1969. Eutrophi-
cation: causes, consequences, correctives. Natl.
Acad. Sci., Washington, D. C.
112. Nelson, L. B., and Uhland, R. E. 1955. Factors
that influence loss of fal applied fertilizers and
their probable importance in different sections of
the United States. Soil Sci. Soc. Amer. Proc. 19:
492496.
113. Olsen, R. J. 1970. Effect of various factors on
movement of nitrate nitrogen in soil profiles and
on transformations of soil nitrogen. PhD Thesis,
Univ. Wisconsin, Madison.
114. Olsen, R. J., Hensler, R. F., Attoe, O. J., Witzel, S.
A., and Peterson, L. A. 1970. Fertilizer nitrogen
and crop rotation in relation to movement of
nitrate nitrogen through soil profiles. Soil Sci. Soc.
Amer. Proc. 34(3): 448452.
118. Onstad, C. A., and Foster, G. R. 1975. Erosion
modeling on a watershed. Trans. ASAE 18:
288-292.
119. Parr, J. F. 1972. Chemical and biochemical con-
siderations for maximizing the efficiency of ferti-
lizer nitrogen. /// Effects of intensive fertilizer use
on the human environment: 53-86. Soils Bui. No.
16, FAO. Pub. by Swedish Internal!. Devlpml.
Authority.
120. Pctersen, G. A., and Sander, D. H. 1974. The
continuing need for soil testing. Soil Sci. Soc.
Amer. Proc. 38(5): 859-860.
121. Power, J. F. 1970. Leaching of nitrate-nitrogen
unHer dryland agriculture in the Northern Great
Plains. In Relation of agriculture to soil and water
pollution: 111-122. Cornell Univ., Ithaca. N. Y.
122. Prasad, R., Rajale. G. B., and Lakhdine, B. A.
1971. Nitrification retarders and slow-release nitro-
gen fertilizers. Advan. in Agron. 23: 337-383.
123. Rogers, H. T. 1942. Losses of surface-applied
phosphate and limestone through runoff from
pasture land. Soil Sci. Soc. Amer. Proc. 7: 69-76.
87
-------
124. Romkens, M. J. M., Nelson, D. W., and Mannering,
J. V. 1973. Nitrogen and phosphorus composition
of surface runoff as affected by tillage method.
Jour. Environ. Qual. 2(2): 292-295.
125. Ryden, J. C., Syers, J. K., and Harris, R. F. 1973.
Phosphorus in runoff and streams. Advan. in
Agron. 25: 1-45.
126. Sabey, B. R., Bartholomew, W. V., Shaw, R., and
Pesek, J. 1956. Influence of temperature on
nitrification in soils. Soil Sci. Soc. Amer. Proc. 20:
357-360.
127. Salter, R. M., and Schollenberger, C. J. 1939. Farm
manure. Bui. No. 605, Ohio Agr. Expt. Sta., 69 pp.
128. Sawyer, C. N. 1947. Fertilization of lakes by
agricultural and urban drainage. New England
Waterworks Assoc. 61: 109-127.
129. Schuman, G. E., Burwell, R. E., Piest, R. F., and
Spomer, R. G. 1973. Nitrogen losses in surface
runoff from agricultural watersheds on Missouri
Valley loess. Jour. Environ. Qual. 2(2): 299-302.
130. Schuman, G. E., Spomer, R. G., and Piest, R. F.
1973. Phosphorus losses from four agricultural
watersheds on Missouri Valley loess. Soil Sci. Soc.
Amer. Proc. 37(3): 424427.
131. Schwab, G. O.. McLean, E. O., Waldren, A. C.,
White, R. K., and Michener, D. W. 1973. Quality
of drainage water from a heavy textured soil.
Trans. ASAE 16: 1104-1107.
132. Shannon, E. E., and Brezonik, P. L. 1972. Rela-
tionships between lake trophic state and nitrogen
and phosphorus loading rates. Environ. Sci. Tech.
6(8): 719-725.
133. Sievers, D. M., Lentz, G. L., and Beasley, R. P.
1970. Movement of agricultural fertilizers and
organic insecticides in surface runoff. Trans. ASAE
13(3): 323-325.
134. Smith, G. E. 1967. Are fertilizers creating pollu-
tion problems? In Soil and America's future:
108-114. Soil Conservation Soc. Amer., Ankeny,
la.
135. Smith, G. E. 1969. Control of waterborne pollut-
ants. Res. Com., Great Plains Agr. Council Publ.
No. 34, Vol. 1: 147-155. Kansas State Univ.,
Manhattan.
136. Smith, G. E., Whitaker, F. 0., and Heinemann, H.
G. 1974. Losses of fertilizers and pesticides from
clay pan soils. EPA-660/2-74-068, U. S. Govt.
Printing Off., Washington, D. C.
137. Smith, M. W. 1959. Phosphorus enrichment of
drainage waters from farmlands. Jour. Fisheries
Res. Board Canada, 16(6): 887-895.
138. Smith, P. F. 1974. Soil absorption of atmospheric
nitrogen in Florida. Commun. Soil Sci. and Plant
Anal. 5(4): 341-353.
139. Smith, S. J., and Stanford, G. 1971. Evaluation of
a chemical index of soil nitrogen availability. Soil
Sci. Ill: 228-232.
140. Soil Conservation Society of America. 1973. Con-
servation tillage. A handbook for farmers. Soil
Conservation Soc. Amer., Inc., Ankeny, la.
141. Stanford, G. 1966. Nitrogen requirements of crops
for maximum yield. In M. H. McVickar, W. P.
Martin, I. E. Miles, and W. H. Tucker, eds.
Agricultural anhydrous ammonia; technology and
use. Soil Sci. Soc. Amer., Madison, Wis.
142. Stanford, G. 1968. Extractable organic nitrogen
and nitrogen mineralization in soils. Soil Sci. 106:
345-351.
143. Stanford, G. 1973. Rationale for optimum nitro-
gen fertilization in corn production. Jour. Environ.
Qual. 2: 159-166.
144. Stanford, G., Carter, J. N., and Smith, S. J. 1974.
Estimates of potentially mineralizable soil nitrogen
based on short-term incubations. Soil Sci. Soc.
Amer. Proc. 38:99-102.
145. Stanford, G., England, C. B., and Taylor, A. W.
1970. Fertilizer use and water quality. ARS
41-168, Agr. Res. Serv., U. S. Dept. Agr., Washing-
ton, D. C.
88
-------
146. Stanford, G., and Epstein, E. 1974. Nitrogen
minerali/ation-water relations in soils. Soil Sci.
Soc. Amer. Proc. 38(1): 103-107.
147. Stanford, G., Frere, M. H., and Schwaninger, D. H.
1973. Temperature coefficient of soil nitrogen
mineralization. Soil Sci. 115: 321-323.
148. Stanford, G., and Smith, S. J. 1972. Nitrogen
mineralization potentials of soils. Soil Sci. Soc.
Amer. Proc. 36(3): 465472.
149. Stevenson, F. J. 1965. Origin and distribution of
nitrogen in soil. /// W. V. Bartholomew and F. E.
Clark, eds. Soil nitrogen. Amer. Soc. Agron.,
Madison, Wis.
150. Stewart, B. A. 1970. A look at agricultural
practices in relation to nitrate accumulations. In O.
P. Engelstad, ed. Nutrient mobility in soils: accum-
ulation and losses: 47-60. Soil Sci. Soc. Amer.,
Inc., Madison, Wis.
151. Stewart, B. A., Mathers, A. C., and Thomas, J. D.
1974. Manure effects on rate of advance, intake,
and quality of runoff from irrigated Pullman clay
loam. Agron. Abstr., 170.
152. Stewart, B. A., Viets, F. G., Jr., and Hutchinson,
G. L. 1968. Agriculture's effect on nitrate pollu-
tion of groundwater. Jour. Soil and Water Conserv.
23(1): 13-15.
153. Stewart, K. M., and Rohlich, G. A. 1967. Eutro-
phication—a review. Calif. State Water Qual. Con-
trol Board, Sacramento, No. 34, 188 pp.
154. Taylor, A. W., and Kunishi, H. M. 1971. Phosphate
equilibria on stream sediment and soil in a water-
shed draining an agricultural region. Jour. Agr.
FoodChem. 19:827-831.
155. Thomas, A. W., Carreker, J. R., and Carter, R. L.
1969. Water, soil, and nutrient losses on Tifton
loamy sand. Univ. Georgia, College of Agr. Expt.
Sta. Res. Bui. 64.
156. Thomas, G. W. 1970. Soil and climatic factors
which affect nutrient mobility. In Orvis P. Engel-
stad, ed. Nutrient mobility in soils: accumulation
and losses: 1-20. Soil Sci. Soc. Amer., Inc.,
Madison, Wis.
157. Thomas, G. W., and Barfield, B. J. 1974. The
unreliability of tile effluent for monitoring subsur-
face nitrate-nitrogen losses from soils. Jour. Envi-
ron. Qual. 3(2): 183-185.
158. Thomas, G. W., Blevins, R. L., Phillips, R. E.,and
McMahon, M. A. 1973. Effect of a killed sod
mulch on nitrate movement and corn yield. Agron.
Jour. 65(5): 736-739.
159. Thomas, G. W., and Hanway, J. 1968. Determining
fertilizer needs. In L. B. Nelson, M. H. McVickar,
R. D. Munson, L. F. Seatz, S. L. Tisdale, and W. C.
White, eds. Changing patterns in fertilizer use:
119-140. Soil Sci. Soc. Amer., Inc., Madison, Wis.
160. Thomas, G. W., and Peaslee, D. F. 1973. Testing
soils for phosphorus. In L. M. Walsh and J. D.
Beaton, eds. Soil testing and plant analysis:
115-132. Soil Sci. Soc. Amer., Inc., Madison, Wis.
161. Timmons, D. R., Burwell, R. E., and Holt. R. F
1968. Loss of crop nutrients through runoff. Minn.
Sci. 24(4): 16-18.
162. Timmons, D. R., Burwell, R. E., and Holt, R. F.
1973. Nitrogen and phosphorus losses in surface
runoff from agricultural land as influenced by
placement of broadcast fertilizer. Water Resour.
Res. 9(3): 658-667.
163. Timmons, D. R., Holt, R. F., and Latterell, J. J.
1970. Leaching of crop residues as a source of
nutrients in surface runoff water. Water Resour.
Res. 6(5): 1367-1375.
164. U. S. Environmental Protection Agency. 1973.
Measures for the restoration and enhancement of
quality of freshwater lakes. EPA-430/9-73-005. U.
S. Govt. Printing Off., Washington, D. C.
165. U. S. Environmental Protection Agency. 1974.
Relationships between drainage area characteristics
and nonpoint source nutrients in streams. Working
Paper No. 25, Pacific Northwest Environ. Res.
Lab., Corvallis, Ore.
166. Uttormark, P. D., Chapin, J. D., and Green, K. M.
1974. Estimating nutrient loadings of lakes from
non-point sources. EPA-660/3-74-020. U. S. Govt.
Printing Off., Washington, D. C.
89
-------
167. Viets, F. G., Jr. 1971. Fertilizer use in relation to
surface and ground water pollution. In R. A.
Olson, T. J. Army, J. J. Hanway, and V. J. Kilmer,
eds. Fertilizer technology and use: 517-532. Soil
Sci. Soc. Amer., Inc., Madison, Wis.
168. Viets, F. G., Jr. 1971. Water quality in relation to
farm use of fertilizer. BioScience 21: 460-467.
169. Vollenweider, R. A. 1971. Scientific fundamentals
of the eutrophication of lakes and flowing waters,
with particular reference to nitrogen and phos-
phorus as factors in eutrophication. Organ, for
Econ. Coop, and Devlpmt., Paris.
170. Wadleigh, C. H. 1968. Wastes in relation to
agriculture and forestry. Misc. Pub. No. 1065. U.S.
Dept. Agr., Washington, D. C.
171. Wagner, R. E., and Jones, M. B. 1968. Fertilization
of high yielding forage crops. In L. B. Nelson, M.
H. McVickar, R. D. Munson, L. F. Seatz, S. L.
Tisdale, and W. C. White, eds. Changing patterns in
fertilizer use: 297. Soil Sci. Soc. Amer., Inc.,
Madison, Wis.
172. Walsh, L. M. 1973. Soil and applied nitrogen. Fact
Sheet No. A2519, Univ. Wisconsin - Extension.
173. Wang, Wun-Cheng. 1974. Adsorption of phosphate
by river particulate matter. Water Res. Bui. 10(4):
662-671.
174. Weibel, S R., Weidner, R. B., Cohen, J. M., and
Christiansen, A. G. 1966. Pesticides and other
contaminants in rainfall and runoff. Jour. Amer.
Water Works Assoc. 58(8): 1075-1084.
175. Welch, L. F. 1974. Nitrogen on soybeans. Crops
and Soils 26(9): 9-10.
176. White, E. M. 1973. Water-leachable nutrients from
frozen or dried prairie vegetation. Jour. Environ.
Qual. 2(1): 104-107.
177. Wild, A. 1972. Nitrate leaching under bare fallow
at a site in northern Nigeria. Jour. Soil Sci. 23(3):
315-324.
178. Wildung, R. E., Schmidt, R. L., and Gahler, A. R.
1974. The phosphorus status of eutrophic lake
sediments as related to changes in limnological
conditions total inorganic and organic phos-
phorus. Jour. Environ. Qual. 3(2): 133-138.
179. Williams, J. D. H., and Mayer, T. 1972. Effects of
sediment diagenesis and regeneration of phos-
phorus with special reference to Lakes Erie and
Ontario. In H. E. Allen and J. R. Kramer, eds.
Nutrients in natural waters: 281-315. John Wiley
and Sons, New York.
180. Willrich, T. L. and Smith, E. G. 1970. Agricultural
practices and water quality. Iowa State Univ. Press,
Ames.
181. Winton, E. F. 1970. The health effects of nitrates
in water. In Nitrate and water supply: source and
control. Proc. Twelfth Sanitary Engin. Conf.,
Urbana, 111.
182. Yaalon, D. H. 1965. Downward movement and
distribution of anions in soil profiles with limited
wetting. In Experimental pedology: 157-164.
William Cloves and Sons, Ltd., London.
183. Yeck, R. G., Smith, L. W., and Calvert.C. C. 1975.
Recovery of nutrients from animal wastes an
overview of existing options and potentials for use
in feed. Proc. Internatl. Symp. on Livestock
Wastes, Univ. Illinois, Urbana.
184. Zwerman, P. J., Bouldin, D. R., Grewelding, T. E.,
KJausner, S. D., Lathwell, D. J., and Wilson, D. O.
1971. Management of nutrients on agricultural
land for improved water quality. Project No.
13020 DPB, Cornell University, New York. Rep.
No. PB209-858, Natl. Tech. Inform. Serv., Spring-
field, Va.
185. Zwerman, P. J., Klausner, S. D., Bouldin, D. R.,
and Ellis, D. 1972. Surface runoff nutrient losses
from various land disposal systems for dairy
manure. In Waste management research: 495-502.
Cornell Univ., Ithaca, N. Y.
90
-------
CHAPTER 5
PESTICIDES IN AGRICULTURAL RUNOFF
J. H. Caro
The use of "chemical pesticides has resulted in
enormous benefits to mankind, chiefly in the areas of
public health and agricultural production. In public
health, insect control programs have saved millions of
lives by combatting such diseases as malaria, yellow
fever, and typhus. In India, for example, the use of DDT
has been credited with increasing the average life
expectancy from 32 to 47 years of age. In Sri Lanka,
annual malaria cases dropped from 2 million in 1950 to
17 in 1963; when use of DDT was then discontinued,
the number of cases immediately rose and again reached
1 million in 1968 (45). Agricultural benefits are many:
pesticide chemicals have promoted higher crop yields
and improved quality of produce; aided the mechani-
zation of agricultural production, with substantial reduc-
tion in labor requirements; and have helped to improve
the utilization and management of land. Use of agricul-
tural pesticides also has resulted in important economic
benefits to both the farmer and the consumer of food
and fiber. Insecticides are widely estimated to return $5
to the farmer for every $1 expended (127), which often
tips the scales to economic profit from a crop rather
than economic loss. Agricultural products would prob-
ably cost the consumer two or three times more than at
present if the use of chemicals were eliminated (150).
Despite all the far-reaching benefits, the use of
pesticides has brought about a conflict of interest
because of the possibility of harmful impact on environ-
mental quality. Conservationists have often indicted
pesticide residues as being responsible for a variety of
injurious effects, including fish kills, reproductive fail-
ures in birds, and acute illnesses in man and animals.
Agricultural applications have just as often been charged
with being primary sources from which the chemicals
dissipate into the environment. Although acute adverse
occurrences have indeed taken place, the sources of the
damaging pesticides have been a matter of some dispute.
With respect to chronic effects, the true significance of
low residue levels of most pesticides in the general
environment resulting from long-term use of the chem-
icals is still not well understood. In any event, it is
widely expected that the use of chemical pesticides will
remain an integral part of agricultural technology for
many years and will in fact increase at least through the
next decade. Consequently, information on the path-
ways by which pesticides leave the site of application
and distribute throughout the environment will continue
to be actively sought so that appropriate controls can be
instituted.
One such pathway is the movement of pesticides
away from treated fields in runoff water and on
sediment carried along in the water. In Volume I, we
presented guidelines for identifying areas of potential
pollution problems arising from this movement and also
described appropriate pesticide management practices
that would alleviate the problems. In this chapter, we
will provide documentation to support the recom-
mended practices. We will also indicate the size of the
potential problem by showing the extent of agricultural
use of pesticides, and we will examine the state of
knowledge concerning pesticide transport in runoff.
Related areas to be covered include (1) information on
pesticide persistence in soil, which affects the relation-
ship between amounts of residues moved in runoff and
the time elapsed since application of the pesticide to the
field; (2) characteristic levels of pesticides found in the
aquatic ecosystem; and (3) the impact of pesticides on
aquatic organisms, which will permit some assessment by
the reader of potential hazards of the reported levels. In
addition, we will summarize information on methods for
removing pesticide residues from the aquatic environ-
ment, and we will show the areas within the broad
subject of pesticides in runoff that clearly require
additional research.
91
-------
EXTENT AND TRENDS IN USE OF AGRICULTURAL PESTICIDES
Because of the demands of our increasing population
for space for cities, roadways, and recreational areas, and
because economic trends have made smaller and less
efficient farm units unprofitable, the amount of crop-
land that supports each of us has, until very recent
times, declined steadily since the 1920's (Figure 1). The
same pressures have favored ever more intensive farming
of the acreage still cultivated. An important component
of this modem, high-efficiency agriculture is the use of
chemicals to combat the pests and blights that attack
our crops and agricultural products. The insects, weeds,
and plant diseases that cause significant agricultural
damage are many and varied. In the United States alone,
for example, there are an estimated 10,000 species of
damaging insects and mites, about 600 of which are
serious pests that require control every year (114).
The chemicals employed to control these pests have
been highly effective. The general impact of pesticide
use on crop yields is indicated in Table 1, which shows
an apparent relationship between rates of pesticide
application and crop yields in major geographic areas.
Although yields may be increased by any of a number of
improved agricultural practices, the importance of pesti-
cide use is undeniable.
The number of specific chemicals is impressive. The
current domestic market for pesticides includes more
than 1800 biologically active compounds sold in over
32,000 different formulations. In 1971, 833 million
pounds of the compounds were used in the United
States, of which almost 60% were accounted for by use
on farms (Table 2). Of all pesticides used on farms,
herbicides comprised 52%, insecticides 39%, and chemi-
cals for control of plant diseases 9%.
The extent and distribution of pesticide use on major
crops in 1971 is shown in Table 3. Three crops—corn,
soybeans, and cotton—accounted for almost 80% of all
herbicide use on farms; two crops—cotton and corn-
accounted for nearly 70% of insecticide use; and fruit,
nut, and vegetable crops accounted for 85% of fungicide
use. The most extensively treated crop was peanuts,
whereas only small percentages of alfalfa acreage were
treated. The pests of most concern in specific crops are
shown clearly in the table. Weeds, for example, are a
severe problem in soybeans, but insect damage is limited.
Conversely, tobacco generally requires protection from
insects but not from weeds.
The geographic distribution of cropland treated with
pesticides is shown in Table 4. Although the table shows
400-
200
250
50
1910
1920
1930
1950
I960 1970
1940
Year
Figure l.-Land in crops and population in the United States. From Barrens (17).
92
-------
Table 1. Rates of pesticide application and yields of major
crops in countries and geographic areas'
Table 3. Acres of crops grown and percentage treated with
pesticides, United States, 1971
Country or Area
United States
Latin America
Oceania ...
India
Africa
Pesticide
application
rates
Ounces/A
154 0
26 7
21.3
3.1
2.8
2.1
1.8
Yields of
major
crops
Ib/A
4 890
3 050
2 320
1 760
1 400
730
1,080
From T.nnisetal (61).
Table 2. Use of pesticides and percentage used by farmers,
United States, 19711
Type of pesticide
Herbicides
Insecticides
Fungicides
Total
Million pounds
used2
359
319
155
833
Percentage used
by farmers
63
53
27
59
1 From Andrilenas (5).
2 Active ingredients.
data for 1969, the relative distribution has probably not
changed significantly in succeeding years. By far the
largest acreage treated is in the two North Central
regions, which basically comprise the Corn Belt and
much of the Wheat Belt. These 12 states contain 63% of
the cropland receiving herbicides and 47% of that
receiving insecticides. The treated area is lowest in the
Northeast (New England and Middle Atlantic regions).
In the United States, the use of pesticides has
continued to increase. Total annual sales for domestic
use exceeded one billion pounds in 1973 (6), as
contrasted with 833 million pounds in 1971 (Table 2),
and much of the increase is undoubtedly attributable to
expansion in agricultural demand. There are two main
reasons for this. Within the past year or two, farm prices
have risen, so that the crop is worth more to the farmer
and he will tend to apply pesticides at a lower level of
infestation to protect it. Second, and also as a result of
changing farm economics, land is once more being
converted to cropland after a long period in the reverse
direction (Figure 1). The U.S. Department of Agricul-
ture has projected an annual increase in cropland within
the near future of approximately 4%. This corresponds
to over 3.5 million additional acres of corn, of which
about 57% will require herbicides and 33% insecticides,
Crop
M..|. Percentage of total acres
acres treated with Pesticides for
' control of
Alfalfa
Corn
Cotton
Fruit crops
Rice
Small grains1 . . . .
Sorghum
Soybeans
Sugarbeets . . . .
Tobacco . .
Vegetable crops . . .
27.5
74.0
12.4
4 0
1 5
1.8
91.7
20.8
43.5
1.4
0.8
4.8
Weeds
1
79
82
28
92
95
37
46
68
75
7
43
Insects
8
35
61
81
87
35
5
39
8
30
77
62
Diseases
<0.5
1
<0.5
53
85
<1
<0.5
2
13
7
27
1 Adapted from Andrilenas (5).
2 Includes wheat, oats, barley, rye, and mixed grains.
Table 4. Cropland acreage treated with pesticides, by
geographic region, 19691
Region
Cropland acreage treated for
control of
Weeds
Insects'*
New England ,
Middle Atlantic ,
East North Central
West North Central . .
South Atlantic ,
East South Central
West South Central . . . . ,
Mountain
Pacific ,
United States
1000 acres
. . . . 260
. . . . 1645
. . . . 20060
. . . 33375
. . . 4030
. . . . 4265
. . . . 9990
. . . . 5655
. . . . 5635
84915
1000 acres
245
675
7890
10885
4^10
2670
7545
1985
3780
39880
1 From U.S. Bureau of the Census (154).
States in the regions are:
NE: ME,NH,VT,MA,RI,CT
MA: NY.NJ.PA
ENC: OH,IN,IL,MI,WI
WNC: MN.IA.MO.ND.SD.NB.KS
SA: DE,MD,VA,WV,NC,SC,GA,FL
ESC: KY,TN,AL,MS
WSC: AR, LA, OK, TX
MT: MT, ID, WY, CO, NM, AZ, UT, NV
PA: WA, OR, CA, AK, HI
Not including land for hay crops.
and more than 8 million additional acres of wheat, of
which about 2.5 million acres will need herbicide
treatment.
Some economists estimate that the use of chemical
pesticides will increase up to 15% annually over the next
few years (45) because of the economic situation and
93
-------
potential fuel shortages. To save fuel, fanners may well
replace band treatment with broadcast applications, with
a resultant increase in pesticide applied per acre. Within
the growing market, the patterns of use of specific
chemicals will differ from the present because of shifts
in cropping patterns and pest populations and the
buildup of resistance in target species of pests. In the
North Central States, for example, there has been a
marked movement away from production of extensive
crops (wheat, oats, barley, rye, flax) and toward row
crops (corn, soybeans). Between 1961 and 1973, row
crops increased from 58 to 80 million acres, whereas
extensive crops fell from 49 million to 34 million (141).
Weed populations will change because the varieties that
are more easily controlled will be selectively removed.
Despite the difficulties in moving chemicals through the
registration process, new pesticides will continue to be
introduced into the marketplace to combat stubborn
weeds and to counteract pest resistance to older chem-
icals.
DISSIPATION OF PESTICIDES FROM TREATED LANDS
During application to soil or foliage, pesticides may
be lost in spray drift or by volatilization; after applica-
tion, they disappear from the site of application by
various pathways. The chemicals may undergo biological
or chemical degradation; on foliage or the soil surface,
they may degrade under the action of sunlight or they
may evaporate; they may be taken up into the plant and
removed in the harvested crop; they may be adsorbed
onto soil particles and moved off the treated area in
eroded material; or they may dissolve in rainwater (or
irrigation water) and move away in surface runoff or
down through the soil in the soil solution, perhaps later
to reappear in surface runoff or groundwater. The rates
of disappearance and the fractions moving by each
pathway depend primarily on the properties and formu-
lation of the pesticide; the type, microbial population,
moisture level, and type of management of the soil; the
extent and intensity of rainfall; and the soil and air
temperatures.
The movement in runoff water, eroded sediment, and
subsurface water is of direct concern here and will be
examined in detail. A number of excellent reviews are
available that deal with the other pathways: Kaufman
(98) on microbial degradation of pesticides, Crosby (54)
and Armstrong and Konrad (9) on nonbiological degra-
dation, Spencer (145) and Guenzi and Beard (76) on
pesticide volatilization, Caro (33) and Nash (120) on
uptake of insecticides by plants, and Foy et al (71) on
uptake of herbicides by plants.
Factors Influencing Pathway of Movement into
Water Courses
Adsorption and Solubility
The pathway that a pesticide takes in its movement
away from the site of application through the action of
water-that is, whether it moves in runoff water or
eroding sediment or leaches down into the soil-is
governed by sorption and solution equilibria that depend
primarily on the water solubility of the chemical, the
degree and strength of its adsorption on soil, and on the
interaction of both soil and pesticide with water (144).
Generally, compounds that are more water-soluble will
move primarily in runoff water and those more strongly
adsorbed will move mostly on sediment. An inverse
relationship exists between solubility and extent of
adsorption, but only within families of compounds.
Some pesticides, such as paraquat and diquat, are very
water-soluble but will move only on the sediment
because of strong, irreversible adsorption; others have
low water solubility but will nevertheless move in the
water except when applied to a rather adsorptive soil
(153).
Soil characteristics are clearly very important in
determining the degree of adsorption, as illustrated by
the fact that adsorptivity of a pesticide can vary by as
much as 1 S-fold over a range of soil types. The most
important soil property that influences the way a
pesticide partitions between soil and water, as deter-
mined by a number of investigators, is the organic
matter content, which usually gives a good direct
correlation with the degree of adsorption. Other prop-
erties that may be important are the acidity, cation
exchange capacity, moisture content, temperature, and
clay mineral content. With some pesticides such as the
triazine and triazole herbicides, adsorption depends
primarily on soil acidity: in acid soils, they associate
with free hydrogen ions to form cations that adsorb
strongly to the negatively charged soil; in neutral or
alkaline soils, they are in molecular form and are held
much more weakly by the soil (163). The acid herbi-
cides, such as 2,4-D and picloram, also adsorb more
strongly in acid soils. Bailey and White (13) showed
correlation coefficients for a number of soil properties
and adsorption as part of a comprehensive review of the
adsorption and desorption of pesticides in soil.
94
-------
Water in the soil competes with pesticides for
adsorption sites on soil particles, so that as the moisture
level in the soil decreases, the fraction of the chemical
adsorbed increases (82, p. 100). Rain falling on a dry soil
will therefore desorb a portion of the pesticide, which
would then move with the water in any ensuing runoff.
Generally, adsorption decreases as soil temperature
increases, and the response to changes in temperature
becomes less as the adsorption bonds weaken (82, p.
l>7).
Pesticides adsorb preferentially on smaller soil par-
ticles because of the high surface area per unit weight of
these particles. When runoff occurs, the small particles
are transported greater distances than coarser material.
Since rill and sheet erosion primarily involve surface soil,
such erosion will tend to favor movement of the more
strongly adsorbed pesticides (707, p. 431). Higher
pesticide concentrations in eroded material do not
necessarily mean that gross losses will be greater in the
sediment than in runoff water; the reverse sometimes is .
true because the amounts of water moved are so much
greater (134, 166).
teachability
Pesticides in the soil or on its surface may move down
through the soil profile dissolved in water. The principal
factors affecting the movement are the same as those
controlling overland movement-adsorption and solubi-
lity-because a pesticide is partitioned between soil and
water in leaching as well as in runoff. Other parameters
influencing leaching are water flow rate and amount, and
the formulation, concentration, and rate of degradation
of the pesticide (89). The correlation of solubility and
adsorption with pesticide movement through soil sug-
gests that solubility may be important in the initial
movement from the point of application, whereas
adsorption may be the determining process in later
movement. Therefore, adsorption will, in general, be a
better indicator of overall potential movement than will
solubility (21). As examples of this, prometryne (48
ppm water solubility) moves less in soil than simazine (5
ppm) because it is more strongly adsorbed, and monuron
(230 ppm) moves about the same as atrazine (33 ppm)
for the same reason (85).
The pesticide moves downward through the profile
either by mass flow of water from impacting rainfall or
by molecular diffusion in the soil solution. Diffusion,
which is influenced by bulk density and temperature in
addition to soil moisture, is slow in comparison with
mass flow and is important only over short distances
(-/), so that mass flow is the primary means of
movement under most conditions. Water does not,
however, continuously move downward, except in very
high rainfall areas. Evaporation at the surface causes
upward movement of subsurface water and its comple-
ment of dissolved pesticides, which then concentrate at
the surface (107, p. 42°). Pesticides in the water can also
move laterally when they encounter a /.one of water
saturation or when they reach the boundary between
two areas of different soil moisture, since water will
move laterally into the drier soil (59). If the laterally
moving water intercepts the sloping surface of the land,
the dissolved pesticide will join the overland tlow and
appear in surface runoff.
Despite occasional reports of low-level groundwatcr
contamination by pesticides, measurements have not in
general shown that groundwater pollution by leaching of
pesticides through soil is extensive or significant. Much
excess water must be applied even to a relatively mobile
chemical to move it deeply into the profile. Many
reported findings in experiments with picloram, a
relatively teachable herbicide, are in agreement: except
in sandy soils, picloram does not leach below the 2-foot
depth (107, p. 430). For a more strongly adsorbed
pesticide such as dieldrin, several hundred years would
be required for the chemical to be transported in
solution at a residual concentration of 20 ppb to a depth
of 1 foot in neutral soils (67). The groundwatcr
contamination that does occur may be caused by
pesticides being carried on soil particles washed down
into deep cracks in the soil in drought-breaking rainfalls
(125. 169).
Formulation
The formulation in which a pesticide is applied also
may affect the pathway of movement, especially if
runoff occurs shortly after application, before the
chemical has equilibrated with the soil. For example,
ester formulations of the herbicide 2,4-D applied to a
sandy loam soil in a set of experiments (16) were tar
more susceptible to washoff than an amine salt formula-
tion. The amine formed a true solution with water and
leached into the soil, whereas the relatively insoluble
esters were adsorbed and moved on the eroded sediment.
Factors Influencing Amounts of Pesticides
Moved into Water Courses
The quantity of a pesticide moving into a water
course from a treated area in any given runoff occur-
rence depends on a number of associated factors. Its
relationship to topography, intensity and duration of
rainfall, soil crodibility, and land management and
cropping practices are discussed in the chapters on
erosion and runoff. Obviously, the amount moved will
95
-------
increase with the amount of pesticide initially applied to
the area. It also depends on the following parameters.
Time After Application
Characteristically, pesticide losses are highest in the
first runoff occurring after application of the chemical,
and the magnitude of the loss generally decreases as time
between application and runoff increases. The effect of
elapsed time is particularly noticeable with short-lived
pesticides and with pesticides that are not incorporated
into the soil. Concentrations of the chemical in subse-
quent runoff events decrease at a rate that depends
largely on the persistence of the pesticide in the soil.
Field experiments with the carbamate insecticides car-
baryl and carbofuran showed that pesticide concentra-
tions in both runoff water and sediments in the third
runoff, which occurred within 1 or 2 months after
application, were less than 5% of the concentrations in
the first runoff (35, 36). By contrast, concentrations of
the persistent insecticide dieldrin in the third runoff, 3
or 4 months after application, were about 15% (water)
and over 30% (sediment) of those in the first runoff
(34). The pattern of relatively high concentrations in the
first runoff, decreasing with time to eventually negligible
concentrations in succeeding runoffs, has been noted in
experiments with many pesticides. Quantitative ex-
amples are illustrated below in the section on pesticide
levels in runoff.
Persistence in Soil
As mentioned, the persistence of a pesticide in soil
affects the change with time in amounts lost in runoff.
However, many factors influence the persistence of an
individual pesticide and, consequently, it can be quite
variable. Picloram, for example, has been reported in
specific instances to effectively disappear from the soil
in as little as 50 days (110) or as long as 6 years (30),
but its persistence under moderate conditions is gen-
erally about 1.5 years.
A pesticide applied to the soil is subject to a sequence
of overlapping loss processes—application losses, volati-
lization, sorption, leaching, and eventually chemical and
biological degradation (92). As a result, the loss rate
changes rapidly during the early period when the
chemical is distributing and equilibrating in the soil, then
becomes nearly constant over a relatively longer time.
This later rate is dictated by many conditions, including
among others the weather; cultural practices; and type,
temperature, moisture level, and acidity of the soil.
Pesticides that are subject to microbial degradation will
have reduced persistence when applied to an area that
had received an earlier application of the same chemical
because of growth in populations of active micro-
organisms after the first treatment (99). If, as is often
the case, more than one pesticide is applied, interactions
between the chemicals may also markedly alter the
persistence of the individual compounds in both soil and
aquatic environments (97).
Antecedent Soil Moisture
Some pesticides will have greater losses in runoff if
applied to wet soil than if applied to dry soil, partic-
ularly if runoff occurs soon after application. Experi-
ments showing this have been reported for 2,4-D (16)
and for fluometuron (75). The effect is probably related
to the competition of water with the pesticide for
adsorption sites on the soil particles.
Proximity to Water Course
Sloping cropland rarely abuts continuous streams.
Consequently, pesticide-containing runoff usually must
traverse some untreated land before reaching the water.
This intervening area can trap some of the pesticide,
resulting in lowered contamination of the stream. Large
decreases can be obtained. In one set of measurements,
flow over only 5 feet of untreated soil with slopes of 3%
or 8% reduced picloram losses in runoff from small field
plots more than 50% (151); in another case, dieldrin
applied at 10 to 20 times normal levels to strips of land
12 to 15 feet away from the edges of ponds, with
shallow to steep slopes intervening, did not appear in the
pond water or bottom mud, except for very low (0.3
ppm) contamination of the mud when the pesticide was
left on the surface of the soil (58).
Placement of the Pesticide
The most important effect of pesticide placement
with respect to environmental contamination is that
soil-incorporated pesticides will not be lost in runoff to
as great an extent as those applied and left on the
surface or sprayed on foliage. The subject is discussed in
more detail in section 4.4 of Volume I.
96
-------
PERSISTENCE AND FATE OF PESTICIDE RESIDUES IN THE
AQUATIC ENVIRONMENT
Distribution of Pesticides on Entering
Water Bodies
A pesticide carried in agricultural runoff entering a
receiving water body-stream, pond, or lake-will dis-
tribute within the aquatic system in a manner and at a
rate that depends primarily on whether the chemical is
initially dissolved in the water or adsorbed on particles
of eroded soil suspended in the water.
A dissolved pesticide will be diluted in the larger
volume of water and will be subject to processes that
dissipate it. In a flowing stream, it will simply be
transported away from the point of entry, later to
undergo degradation or removal from the water. In a
pond or lake, it may sorb or concentrate in algae and
aquatic vegetation or it may attach to suspended
sediment and other particulates in the water such as
bacterial floes, diatoms, and general organic or inorganic
fragmentary material. In either case, it is eventually
deposited on the bottom of the lake unless it is
chemically or biologically degraded before it reaches the
bottom or is taken up by living organisms. The sorption
processes appear to be generally quite rapid and effi-
cient, as shown in measurements of organochlorine
insecticide sorption on algae (91) and bacterial floes
(106). Highly soluble pesticides that are only weakly
adsorbed may be hydrolyzed or biologically degraded in
solution at a rate that depends on the types and numbers
of microorganisms in the water.
The fate of pesticides entering water bodies adsorbed
on sediment has been discussed in detail by Pionke and
Chesters (126). The pesticide will distribute first with
the carrying sediment, then will equilibrate with the
remainder of the aquatic system. Sediments entering
water bodies will segregate on a particle-size basis: in a
stream, the fractionation will depend on stream velocity;
in a lake, the particles will settle on the bed in decreasing
order of particle size. The finer particles containing the
highest concentrations of pesticides will be transported
farthest and will be localized in a stream; in a lake, these
particles will settle last and remain at the water-sediment
interface. In large, thermally stratified lakes, density
currents may control the movement and mixing of
incoming sediments and settling may be very slow.
Conditions in the water body may affect the adsorp-
tion and desorption of the pesticide on the sediment. If
the pH of the lake or stream is higher than that of the
inflow, desorption of acidic compounds (such as 2,4-D,
2,4,5-T, or picloram) or weakly basic compounds (such
as the triazine or urea herbicides) will be favored, and
the reverse will be true if the pH of the body is lower
than that of the inflow. Salinity in the lake will favor
adsorption of acidic pesticides and desorption of basic
pesticides, but the effect is generally minor. A lower
temperature in the water body will increase pesticide
adsorptivity, but this too is a minor effect under field
conditions (126). If there is any oil pollution in the
water, adsorbed oil will significantly concentrate the
pesticides on the sediment (87).
Post-Distribution Processes
Pesticides never reach true equilibrium in water
bodies because the systems are dynamic, with many
processes continually operating to remove the chemicals
from the system at rates that change as conditions
change. Pesticides sorbed on bottom muds may be
churned up and carried along with sediment during
periods of turbulent flow or they may remain where
originally deposited. Since sorption on participate
matter is generally reversible, the bottom muds provide a
continuous supply of desorbed pesticides to the over-
lying water. In the water, the pesticides may be
chemically or biologically degraded, they may reach the
surface and volatilize, or they may be decomposed near
the surface by the action of sunlight. (Sunlight energy is,
however, probably too weak to induce much photode-
gradation in natural waters.) At the surface of the
bottom muds, pesticide degradation is extensive. Be-
cause organic matter accumulates there, it is an area of
high microbial activity. The microbial populations may
consume so much oxygen that the environment becomes
anaerobic, a condition that favors the degradation of
many pesticides. For example, most chlorinated hydro-
carbon insecticides, although normally highly persistent,
will degrade at an appreciable rate under anaerobic
conditions when the temperature is 20° C or higher (91).
Pesticide Persistence in Aquatic Environments
Measurements have been made of the persistence in
aquatic systems of a large number of specific pesticides,
as summarized by Pionke and Chesters (126), Paris and
Lewis (123), and Echelberger and Lichtenberg (60),
97
-------
among others. Information of a more general nature is
presented here.
Organophosphorus Insecticides
As a class, organophosphorus insecticides are among
the less hazardous pesticides in the aquatic system
because they hydrolyze rapidly. Within the pH range
6.0-8.5, which covers most systems, most organophos-
phorus compounds hydrolyze within 8 to 12 days, with
hydrolysis occurring both in water and on sediments
(118, 126). Parathion is an exception, being somewhat
resistant to chemical hydrolysis. It is, however, readily
susceptible to microbial degradation in both aerobic and
anaerobic environments. In the absence of an active
microbial population, parathion remains in the aquatic
environment for several months; in the presence of an
active population, it is degraded in a matter of weeks
(75).
Organochlorine Insecticides
As noted earlier, these compounds will degrade very
slowly, if at all, in aerobic aquatic systems, but will
decompose more rapidly in anaerobic environments
through the action of microorganisms. The degradation,
which generally involves a simple dechlorination of the
molecule, may require up to several months for comple-
tion in natural systems. There is, however, some varia-
tion among the individual members of the class, persist-
ence increasing in the following order: lindane,
heptachlor, endrin, DDT, DDD, aldrin, heptachlor
epoxide, dieldrin (90, 91).
Other Pesticides
The carbamate insecticides have been shown to
degrade in slightly alkaline river water in less than 4
weeks (60), but since the stability of these compounds is
pH-dependent, appreciably longer persistence would be
expected in a more acid environment.
The decomposition of many herbicides is primarily
biological. In aerobic lake water, for example, 2,4-D
persisted for up to 120 days, whereas in lake muds, it
was substantially decomposed in 24 hours once the
microbial populations had adapted to the chemical (2).
The herbicide dicamba also dissipates from water most
rapidly under nonsterile conditions. The rate of disap-
pearance depends greatly on the temperature, especially
in the presence of sediments containing active microbial
populations (139).
CHARACTERISTIC LEVELS OF PESTICIDES IN THE AQUATIC ECOSYSTEM
When pesticide-containing runoff occurs from agricul-
tural land, the chemicals are quickly diluted in the water
bodies receiving the runoff and are also partitioned
among the various components of the environment-
water, bottom sediments, and living organisms—so that
each component eventually bears a concentration of the
pesticide. The magnitudes involved have been measured
by many investigators under a variety of conditions and
are summarized here, including levels in the runoff itself
and in drainage streams, farm ponds, lakes, and oceans,
so that some appreciation may be gained of the impact
of agricultural activities on water quality.
The importance of maintaining a constant surveil-
lance of the aquatic system in the United States has been
recognized by the Federal government. Comprehensive
programs for continuous monitoring of pesticides in fish,
estuarine shellfish, water, and bottom sediments, among
other components of the general environment, are
conducted by various agencies to establish baseline levels
and to signal significant trends. The overall program is
coordinated by an interagency committee (122). With
respect to the quality of our waters, standards have been
set for acceptable limits of certain pesticides in drinking
water (Table 5). An obvious goal of pesticide control
programs is to assure that the natural waters of the
country are sufficiently pesticide-free to drink without
purification.
Pesticide Levels in Runoff
Concentrations of pesticides leaving treated fields in
runoff water and entrained sediments during the crop
season following application are almost always measur-
able, so that there is little doubt that agriculture does
contribute to the pesticide residues found in the general
aquatic ecosystem. Both concentrations and gross
amounts lost depend on numerous factors, including
among others the intensity of rainfall, the time after
pesticide application that runoff occurs, and the mode
of application. Concentrations of a given pesticide may
therefore differ substantially in runoff occurrences at
separate locations under different sets of conditions.
Nevertheless, it is useful to examine the results of
specific measurements reported in the literature, not
only because an understanding may be gained of the
orders of magnitude involved, but also because some
98
-------
Table 5. Recommended limits for pesticides in drinking water1
Pesticide
Recommended
Limit
Organochlorine Insecticides
Chlordane
DDT
Dieldrin
Endrin
Heptachloi
Heptachloi epoxide
Lindane
Methoxychlor
Toxaphene
ppb
1
3
.... 50
. . . 1
.... 0 5(0 2)2
. . . . 01
. . . . 01
"•' .
.... 5(4)
. . . 1000(1 OO)2
. . 5
Phenoxy Herbicides
2, 4-D
Silvex
2,4,5-T
20(1 OO)2
30(10)
2
1 From Environmental Protection Agency (62).
Numbers in parentheses indicate limits proposed by the
Environmental Protection Agency to take effect in December
1976.
informative general conclusions may be drawn. Table 6
lists a number of such measurements made on water-
sheds under normal agricultural conditions.
The figures in the table show that pesticide concen-
trations in runoff are generally, but not always, highest
in the first runoff occurrence following application of
the chemical. Presumably, small flows can occasionally
produce higher concentrations than more intensive flows
preceding them. Concentrations are always lower in
runoffs occurring later in the season, irrespective of
which pesticide is applied, and the reduction is generally
greater for water-borne pesticides than for those carried
on sediment. The table also shows that the organochlo-
rine insecticides, as is well known, adsorb to a great
extent onto sediments, leaving only very low concentra-
tions in water. Control of erosion will therefore reduce
the movement of these chemicals substantially. Atrazine,
on the other hand, moves with both water and sediment.
In general, the concentrations of pesticides in runoff are
considerably above drinking water standards (Table 5)
and must be diluted substantially in drainage streams to
avoid acute harmful effects on aquatic organisms in the
streams. Finally, the table shows, particularly in the
-ases of carbaryl and picloram, how widely runoff
concentration of a pesticide may vary under different
conditions.
Though amounts of a pesticide lost in runoff may
vary among specific treatments, it is almost always true
that the gross loss over the course of the year following
application will represent only a small percentage of the
amount of chemical that had been applied to the
cropland. Results of watershed and field-plot experi-
ments reported in the literature (Table 7) show the wide
applicability of the relationship. Except for one experi-
ment with atrazine in which a heavy rain occurred one
week after application, total amounts lost in runoff
water and sediment were always 5 percent or less of the
application. This appears to hold true irrespective of the
soil type or degree of incorporation of the pesticide into
the soil, and applies even on relatively steep slopes.
Simulated rainfall experiments, in which the amount and
intensity of water falling onto a small sloping plot can be
controlled, show that rainfalls of the size and intensity
that might be expected to occur every year produce only
small percentage losses of pesticides in runoff, but that
heavier "10-year" rains may cause larger losses (11,
166).
Pesticide Levels in Drainage Streams
Almost all the measurements that have been made of
pesticide concentrations in flowing streams draining
treated areas indicate that, as might be expected,
concentrations are substantially lower than in direct
runoff. With many herbicides, residues were always
below detectable limits in waters a few hundred yards
below sprayed areas (66, 78). Even where detected,
levels of such herbicides as 2,4-D, 2,4,5-T, dicamba, and
picloram were well below median tolerance limits for
trout. In one instance, the herbicide fenuron was applied
at a high rate of 23 Ib/A along stream channels. Only
2.4% of the application was lost in the stream water over
a 27-month period, with a maximum concentration of
430 ppb following a heavy rain (56). Results for
insecticides were much the same as those for herbicides.
Phosphorus insecticides were just at detection levels in
drainage streams in California (14). Chlorinated insecti-
cides were generally at low but measurable levels. In
streams draining sugarcane fields, waters contained a
maximum of 820 ppt (parts per trillion) of endrin over a
4-year period, generally decreasing to 30 to 40 ppt 3
months after treatment. Streambed material averaged
100 ppb, with levels decreasing as the season progressed.
Dieldrin, BHC, and DDT were also found in the waters
in the parts-per-trillion range (104). In drainage streams
from a commercial orchard that had received substantial
applications of Organochlorine insecticides, no residues
were found in the water, but detectable levels of DDT
compounds, dieldrin, and endrin appeared in the silt,
organic debris, and bottom organisms (117).
99
-------
8
Table 6. Characteristic concentrations of pesticides in runoff: maxima and rates of decrease
Pesticide
Chlorinated
Insecticides
DDT
DDT
Dieldrin
Dieldrin
Dieldrin
Endosulfan . .
Endrin
Endrin
Endrin
Methoxychlor ....
Other
Insecticides
Carbaryl
Carbaryl
Phorate
Herbicides
Atra^ine .......
Fluometuron
Picloram
Picloram
Picloram
Picloram
2,4,5-T
2,4,5-T
Application
rate
Ib/A
15
065
1.5
5 0
5 0
0.31
0 3
0.3
025
200
45
15
0.67
8.0
1 0
4.0
30
2.0
1.0
025
3.0
1 2
Runoff occurrence giving Maximum concentration
maximum concentration
Number after
application
1st
1st
1st
2nd
1st
1st
1st
2nd
1st
1st
1st
1st
3rd
1st
3rd
1st
3rd
3rd
Days after
application
1
13
13
4
2
6
1
18
17
24
24
86
30
6
10
86
22
Runoff occurrence giving
reduced concentration
In water In sediment Num*er **" Dl»? ^ er
application application
ppb ppb
70.0 30,000
83.01
70.0 30,000
20.0
14,200
19.01
2.73
5.02
49.01
8.8
248 12,200
1220
19.0
4600 6,200
700 950
870
19.0
14.4
89.7
17.0
287
380
5th
3rd
10th
3rd
3rd
3rd
4th
3rd
2nd
2nd
4th
3rd
8th
2nd
4th
5th
46
18
82
14
27
23
33
29
35
35
197
98
30
20
197
38
Reduced concentration
In water In sediment
ppb ppb
1.0 10,000
7.01
1.0 30,000
6.7
, 5,000
2.01
0.53
2.88
8.01
1.0
8.4 80.0
980 3,300
180 550
9.0
LO
6^0
50
Citation
Haan(77)
Epstein and Grant (63)
Haan (77)
Caro era/ (34)
Caro et al (34)
Epstein and Grant (63)
Willis and Hamilton (169)
Willis and Hamilton (169)
Epstein and Grant (63)
Edwards and Glass (59)
Caro, Freeman, and Turner
(36)
Fahey (68)
Fahey (68)
Hall, Pawlus, and Higgins (81)
Hall, Pawlus, and Higgins (81 )
Wiese (167)
Bovey et al (26)
Baur, Bovey, and Merkle U9)
Baur, Bovey, and Merkle (19)
Baur, Bovey, and Merkle (19)
Bovey if a/ (26)
Edwards and Glass (59)
1 Water-sediment mixture.
-------
Table 7. Percentages of applied pesticides lost in runoff1 in field experiments
Pesticide
Atrazine
Carbofuran . . .
DDT
DDT
Dieldrin
Dieldrin ... . . . .
Methyl parathion . . .
Methyl parathion . . .
Propachlor
Toxaphene . . . .
Trifluialin
Trifluralin
Incorpo-
rated depth
In.
... 0
0
0
2
... 3
2
0
0
.... 3
3
.... 0
0
.... 0
0
.... 0
0
0
.... 0
.... 6
.... 6
Soil texture
Silty clay loam
Silty clay loam
Silt loam
Silt loam
Silt loam
Silt loam
Loamy sand
Gravelly loam
Silt loam
Silt loam
Gravelly loam
Gravelly loam
Silty clay loam
Various
Loamy sand
Sandy loam
Silt loam
Loamy sand
Loamy sand
Sandy loam
Slope
%
14
14
10-15
10
9
10
2-4
8
14
10
8
8
02
0 1-4
4
2
10-15
2-4
4
2
Pesticide
in runoff
% ofappln.
4.8-5.0
2.6
2.5-15.9
0 1
0.9
1.9
1.0-2.8
0.7
2.3
0.02
0.25-0.35
0.01-1.0
0 1
<3.0
0.01-0.02
0.13-0.25
3.1
0.4-0.6
03-05
0.5-0.8
Citation
Hall (80)
Hall, Pawlus, and Higgins (81)
Ritterera/034)
Caro, Freeman, and Turner (36)
Caro et al (35)
Caro et al (35)
Bradley, Sheets, and Jackson (27)
Epstein and Grant (63)
Caro et al (34)
Caro et al (34)
Epstein and Grant (63)
Epstein and Grant (63)
Willis and Hamilton (169)
Wiese(167)
Sheets, Bradley, and Jackson (142)
Sheets, Bradley, and Jackson (142)
Ritterera/(134)
Bradley Sheets, and Jackson (27)
Sheets Bradley and Jackson (142)
Sheets Bradley, and Jackson (142)
1
Both water and sediment.
Pesticide Levels in Farm Ponds
Concentrations of pesticides in the waters of farm
ponds adjacent to treated areas are clearly sensitive to
the amount of pesticide applied in the area and the
length of time between application and the first heavy
rain. In a pond near cotton plots, for example, DDT and
toxaphene concentrations in the water were always
significant after application and were especially high
when intense rain closely followed the application.
Concentrations ranged from 0.4 to 13.4 ppb for DDT
and from 2.9 to 65.2 ppb for toxaphene (142).
Similarly, in measurements of the herbicide picloram in
ponds at several locations in Texas, concentrations in the
water ranged from 55 to 184 ppb if the first rainfall
occurred within 2 weeks after application, but were only
2 to 29 ppb if the first rainfall was delayed for 6 weeks.
Concentrations in all cases dropped to 1 ppb or less
within 6 months (78).
The few measurements that have been made indicate
that farm ponds are not always contaminated despite
proximity to treated areas. In Virginia in 1966, when use
of organochlorine insecticides was high, heptachlor was
1 >und in the water of only 10 of 35 ponds examined, at
levels up to 5 ppb. In the bottom muds, the conversion
product heptachlor epoxide was found in 14 of the
ponds at levels of 1 to 60 ppb. As is evident, the
majority of the ponds contained no detectable residues
(162). In another set of measurements, the waters of a
farm pond near a 5-lb/A treatment with carbaryl
contained no detectable residues throughout the crop
season, and a pond near a 0.67 Ib/A treatment with
phorate showed a maximum of only 4 ppb, becoming
undetectable later in the season (68). Of course, ponds
or any water body can receive relatively heavy doses of
pesticides by drift or inadvertent direct spray from aerial
applications.
Pesticide Levels in Rivers
Virtually no published information is available on
pesticide residues in river waters of the United States
showing concentrations occurring later than about 1970.
Nevertheless, the pattern is clear: contamination of the
streams peaked about 1966 (Figure 2) and then de-
creased steadily, probably right up to the present, with
decline in domestic use of the most significant contami-
nants, the organochlorine insecticides. Concentrations
have always been low and are often at trace levels, which
are less than about 1 to 5 ppt for most of the chemicals.
The national situation is perhaps best summarized by
examining the results of a few broad-scale monitoring
studies. Lichtenberg et al (108) combined five annual
synoptic surveys of U.S. rivers for 1964 through 1968.
Except for dieldrin in 1964, no pesticide appeared in
more than 40% of the samples, and the frequency of
101
-------
Dieldrin
0
25
0
25
0
r no
| data
Heptachlor
i
Lindane 8 BHC
WTTTft
no
r
l data
no
data
Chlordane
1964 1965 1966 1967 1968
Figure 2.-Percent positive occurrences of ten organochlorine
insecticides in river waters of the United States, 1964-68.
From Lichtenberg era/ (108).
positive occurrences declined sharply after 1966 (Figure
2). Maximum concentrations found during the 5-year
period for the 10 compounds shown in the figure ranged
from 0.84 ppb for ODD to 0.048 ppb for heptachlor. An
intensive examination of Mississippi and Missouri River
drinking water samples during the same period (137)
showed frequencies of occurrence similar to those of the
broader surveys: over 40% of the samples were positive
for dieldrin, over 30% for endrin and total DDT, and
20% for chlordane. Little or no aldrin, heptachlor,
toxaphene, or methoxychlor was found. For the later
period from 1968 through 1971, measurements of
organochlorine insecticides and three herbicides (2,4-D,
2,4,5-T, silvex) in waters taken from 20 stations in
Western U.S. rivers (138) showed the characteristic
decline in positive insecticide occurrences. In 1967-68,
there were 165 insecticide occurrences and 70 herbicide
occurrences; corresponding figures in 1970-71 were 53
and 54. Maximum concentrations found were 0.46 ppb
DDT and 0.99 ppb 2,4-D. In sum, contamination of river
waters by pesticides is at low levels, sporadic, and
decreasing. Even at worst, residues were well below
acceptable limits for drinking water (Table 5).
Pesticide residues in riverbed sediments are also
relatively low. In the Mississippi River, high concentra-
tions were found only in sediments located just below
pesticide manufacturing and formulating plants (18).
Large amounts of organochlorine insecticides applied to
crops in the Mississippi River Delta did not contaminate
streambed sediments widely (112). Dieldrin and endrin
occurred in only 18 to 20% of the sediments taken from
a Louisiana estuary in 1968-69; maximum levels were 4
or 5 ppb (136). In Rhode Island streams, measurements
made about 1970 showed that DDT and its metabolites
occurred in almost all sediments at concentrations
generally below 500 ppb. Chlordane and dieldrin were
also found, but not as often and usually at less than 50
ppb (133). The reported results suggest that riverbed
surfaces of streams draining extensive row-crop farmland
could be contaminated by locally used pesticides at low
part-per-million levels.
Pesticide Levels in Lakes
Pesticide concentrations are generally lower in large
lakes than in rivers. Water samples from Lake Erie in
1971-72 mostly showed no measurable residues. In the
few positive samples obtained, low levels of diazinon,
dieldrin, atrazine, and simazine were found. It was
concluded that the contribution by agricultural sources
to pesticide pollution of the lake was negligible and
insignificant (156). In 1967-68 in Lake Poinsett, the
largest natural lake in South Dakota, DDT and its
metabolites predominated, averaging 80 ppt in the
water, with lesser amounts of other organochlorine
insecticides appearing in most of the samples (83). The
waters in Lake Michigan contained only 2 or 3 ppt of
organochlorine insecticides, chiefly the DDT family, in
1969-70. Upper sediment layers contained median con-
centrations of 18.5 ppb total DDT and 2.0 ppb dieldrin,
with traces of heptachlor epoxide and lindane (105).
Being large, diverse impoundments, lakes provide
excellent environments for measurement of the process
of biological magnification of pesticides through food
chains. Two studies clearly illustrate the effect. In one,
DDT in Lake Michigan appeared at levels of about 2 ppt
in the water, 14 ppb in the bottom muds, 410 ppb in
sand fleas, 3 to 6 ppm in fish, and as much as 99 ppm in
herring gulls, which are near the top of the chain (114).
102
-------
In the second study, in which organochlorine insecti-
cides in Lake Poinsett were measured, the effect was
considerably smaller. Residues found were in the ratio:
water 1, bottom sediments 18, crayfish 18, zooplankton
37, algae 37, and fish 790 (83). Whatever the magnitude,
biological magnification must be considered when evalu-
ating the significance of very low concentrations of
persistent pesticides in waters and bottom sediments.
Pesticide Levels in Oceans
Pesticides in the marine environment originate from
many sources and it is clearly not possible to define the
contribution from agricultural runoff. Taken together,
however, all sources have contaminated the ocean to
only extremely low levels, and only in coastal areas.
DDT, the most ubiquitous pesticide, is undetectable in
Atlantic deep-sea water or sediments, but does appear in
water, sediments, shellfish, and finned fish along the
coast. However, concentrations in shellfish have declined
to insignificant levels since the curtailment of DDT use
(88) and, although not reported, concentrations in the
other components of the ocean environment have
probably also declined. Where they occur, concentra-
tions are highest in surface slicks; in 1968, coastal slicks
contained up to 13 ppb of the organochlorine insecti-
cides, whereas the underlying waters were at low ppt
levels (140). Whole seawater off the Pacific Coast in
1970 also contained DDT, but only at a maximum of
less than 6 ppt (50).
The pattern of decreasing DDT concentration with
distance from shore and with depth was confirmed in a
more recent investigation of Pacific Ocean waters (168).
DDT concentrations in surface film samples taken in
1971 and 1972 were 11-15 ppt in coastal waters, 0.4 ppt
in the offshore California current, and less than 0.02 ppt
in the North Central Pacific. In subsurface waters, by
contrast, DDT levels were less than 0.01 ppt in the
North Central Pacific and only 0.1 ppt in the offshore
current. The measurements also showed that concentra-
tions of polychlorinated biphenyls were ubiquitous and
were as much as two orders of magnitude higher than
those of DDT.
IMPACT OF PESTICIDES ON THE AQUATIC ENVIRONMENT
System planners can best evaluate the need for
instituting controls for pesticides in agricultural runoff if
they understand the effects that the chemicals may
produce in the downstream aquatic environment. The
information presented here will emphasize some major
effects but does not cover all that is known on the
subject.
A variety of pesticide-induced effects on aquatic
birds, fish, aquatic plants, invertebrates and microorgan-
isms has been documented. Acute responses have been
measured quantitatively and subtle effects produced by
long-term exposure to low, sublethal pesticide concen-
trations have been recognized, even though not
measured directly. Perhaps the most prominent effects
have been the catastrophic fish kills during the 1960's in
which pesticides were implicated, though never posi-
tively established as the cause. One such took place in
the Mississippi River, presumably caused by endrin; a
second occurred in the Rhine River and was attributed
to endosulfan. These and similar kills probably resulted
from high pesticide concentrations emanating from a
point source rather than from nonpoint agricultural
runoff. Runoff was, however, probably partly responsi-
ble for contamination of Lake Michigan salmon by DDT
in 1969 that, though not fatal to the fish, resulted in
widespread confiscation of the commercial catch, with
an economic loss estimated at $3 to $4 million (95).
Unless pesticide concentrations are very high, re-
sponses of organisms in a particular aquatic system are
extremely difficult to predict because of the great
variability and natural complexity of ecosystems and the
assortment of environmental insults that man imposes
on the systems. Ponds, lakes, and streams vary in their
content of water, dissolved salts, temperatures, acidity,
and nature and populations of plants and animals on the
bottom. Pesticide contamination in small bodies of
water, for example, has been judged to be more
transitory and less serious than in large bodies because of
greater bottom surface area per unit volume of water,
higher flushing rates, and greater biological activity (95).
Furthermore, different forms of the active pesticidal
ingredient may differ in toxicity, and additives in the
formulation, such as wetting agents or binders, may be
more toxic to aquatic organisms than the active ingredi-
ent itself (119).
In general, herbicides are less toxic to aquatic
organisms than insecticides (Table 8), though there are a
number of exceptions (Vol. I, Tables 8a and 9a).
Herbicides, especially those applied directly, also have
desirable as well as undesirable effects on water bodies.
They may be responsible for opening and maintenance
of navigable waterways, saving of irrigation water,
increase in aesthetic and monetary value of waterfront
property, control of mosquitoes and snails by removal of
103
-------
Table 8. Relative toxicity of selected pesticides to
aquatic organisms1
Pesticides
Organism
Plankton Shrimp Crab Oyster Fish
Herbicides
Phosphorus ....
Insecticides
DDT.
1
0.5
2
1 i
1000 800
400 200
1
1
300
1
2
700
From Butler (31). Based on the arbitrary assignment of
the value of 1 to herbicides.
aquatic weeds, restoration of recreational waters, im-
provement in fish management, and elimination of
flavors and odors from algal blooms. On the debit side,
many herbicides are acutely toxic to fish (Vol. I, Table
8a). Serious losses of fish and other aquatic fauna may
also occur when herbicides kill aquatic weeds, which
gravitate to the bottom and decompose, removing
necessary oxygen from the water. Moreover, phenols
resulting from the hydrolysis of phenoxy herbicides such
as 2,4-D may impart objectionable flavors and odors to
water. In comparison with insecticides, however, the
hazards of herbicides in the aquatic environment are
small. Most herbicides have little or no toxicity to
humans, wildlife, or livestock; they may reduce phy-
toplankton populations initially, but recovery generally
occurs within 2 or 3 weeks; they do not undergo
biological magnification in food chains, and shellfish are
tolerant of them, accumulating residues only tempo-
rarily after exposure (72).
Acute Toxicity
The toxicity of a pesticide to fish is affected by
numerous parameters, including the size, age, and species
of the fish; water temperature and acidity; and physical
differences at the aquatic site. Survival time after
exposure generally correlates directly with body weight,
probably because the smaller fish consume a propor-
tionately greater diet and have less fat for storage
detoxification. Higher water temperature increases toxi-
city of some pesticides and decreases it for others. DDT
and methoxychlor are examples of insecticides that are
less toxic at higher temperature; toxaphene, endrin,
malathion, and parathion are more toxic (32). The effect
can be very pronounced, as shown in Table 9. There are
differences in response within species as well as between
species. Diquat, for example, was toxic to female
mosquitofish under conditions in which the males were
not affected (170).
It is important to recognize that the high toxicity to
fish of some herbicides (trifluralin, for example) is
tempered in nature by processes that inactivate the
compound, such as strong adsorption and relative
immobility on soil surfaces, so that contamination of
natural water bodies is minimal if the chemical is used
according to label directions.
Another aspect of toxicity to fish is the development
of resistance to pesticides, which has been documented
for several species of fish. For example, fish in a
contaminated lake in Mississippi had higher tolerance for
endrin, DDT, and toxaphene than those in a relatively
clean lake (22). Toxicity will also result in the flourish-
ing of one group of organisms in an aquatic environment
while others are suppressed. Thus, elimination of algae-
eating species will produce increases in algae and
anaerobic bacteria populations (130).
Chronic Toxicity
Fish mortalities have been observed to occur in
nature by long-term, low-level exposure to pesticides.
Numerous pathological effects on the tissues and organs
of fish have also been noted, including lesions of liver
and gills and changes in the intestines, kidneys, brain,
and blood. However, some chemicals—notably meth-
oxychlor and carbamate insecticides such as carbofuran
and carbaryl—are rapidly hydrolyzed on ingestion and
therefore do not have chronic effects.
Fish Reproduction and Growth
Persistent pesticides such as DDT and dieldrin can
have strong adverse effects on reproduction in fish. In
one typical case, the hatching success of landlocked
Atlantic salmon from a lake contaminated with DDT
was 36% lower than in control fish (109). Mortality
usually occurs in salmon sac-fry during the period of
Table 9. Effect of water temperature and exposure time on
the toxicity of trifluralin to bluegills1
Water temperature
V
85
75
65
55
45
48-Hour
LC50
tig/liter
8.4
66
200
380
590
24-Hour
LC50
tig/liter
10
120
360
530
1300
1 From Cope (48).
104
-------
yolk-sac absorption. Some herbicides also may produce
reproduction problems in fish, causing atrophy of
spermatic tubules and production of abnormal sper-
matozoa (48). In addition to reproductive failures, loss
of appetite and restricted growth have been reported to
result from exposure of fish to pesticides.
Fish Behavior
Cases have been reported of changes in the condi-
tioned responses and locomotor patterns of fish as a
result of exposure to pesticides. One well-documented
effect is an increase in sensitivity to low water tempera-
tures, including active avoidance, in fish exposed to DDT
(32).
Effects on Aquatic Plants
Herbicides, by their very nature, are more toxic to
aquatic plants than insecticides, but adverse effects do
not always occur where they might be expected.
Phytoplankton are sometimes unaffected by pesticides,
sometimes multiply when predators are removed by the
chemicals, and sometimes are seriously inhibited in
growth, depending on the particular conditions at hand.
One apparent and obviously serious effect is the disrup-
tion of photosynthesis in phytoplankton: in a compre-
hensive series of tests, carbon fixation by estuarine
phytoplankton was reduced by 45 of 54 chemicals
tested, with reductions of over 90% for several of the
compounds (158). Some aquatic plants act as concen-
trating agents for the organochlorine insecticides, so that
when the plants die, concentrations of the pesticides, as
well as of plant nutrients, are released into the water.
Odor and Taste
Several of the organochlorine insecticides, including
toxaphene, endrin, and heptachlor, impart objectionable
odors to water at concentrations of only a few parts per
billion; a number of herbicides generate a strong odor;
and the solvents used in many formulations are highly
odorous in concentrations as low as 16 ppb (131). The
herbicide 2,4-D hydrolyzes in water to 4-chlorophenol
and 2,4-dichlorophenol, both of which impart dis-
pleasing flavors and odors at low ppb levels. However,
the phenols are only rarely detected in natural waters
(72). Decomposing aquatic plants that have been killed
by herbicides are a major source of foul odors in aquatic
environments.
REMOVAL OF PESTICIDES FROM THE AQUATIC ENVIRONMENT
Obviously, no removal of pesticide residues from
bodies of water would be needed if chemicals that are
rapidly degradable in the aquatic environment were the
only ones being used, but such is not the case. Pesticides
that are relatively persistent pose a threat to water
quality that has prompted investigations of methods for
their removal, chiefly in connection with the protection
of drinking water supplies. Residues can be removed
directly from water bodies by such measures as dredging
sediments and removing weeds, debris, and coarse fish.
However, these methods are generally not economically
feasible and the method that is generally followed for
nondrinking waters is to simply allow a period of time
for natural renovation to occur.
With public water supplies, available data indicate
that present-day conventional water-treatment processes,
such as lime-alum coagulation, sedimentation, sand
filtration, chlorination, and pH adjustment, will reduce
high pesticide levels substantially, but are inadequate for
removal of chronic contamination at low levels (39). The
degree of removal depends on the water solubility and
adsorptivity of the individual pesticides, with more
efficient removal for compounds of low solubility or
high adsorptivity. In one series of tests, for example,
conventional treatment eliminated less than 10% of the
lindane and only 20% of the parathion in the water, but
removed 55% of the dieldrin, 63% of the 2,4,5-T ester,
and 98% of the DDT (135).
Much research effort has been expended on adsorb-
ents to purify the water beyond the levels attainable by
conventional treatment. Activated carbon is clearly the
most effective adsorbent for pesticides, having a removal
efficiency about 4 orders of magnitude greater than that
of soil, 3 orders greater than that of algae, and 2 orders
greater than that of coal (100). The effectiveness of
removal depends on the contact time, the concentration
of activated carbon, the concentration of the pesticides,
and the presence of organic material in the water that
may compete with the pesticides for adsorption sites on
the carbon. If the water is clarified by other processes
before the activated carbon is introduced, the competi-
tive organic matter can be at least partially controlled
(121). Removal efficiency decreases at pesticide con-
centrations below 1 ppb. Although it is possible to
remove lower concentrations of organochlorine insecti-
cides, inordinately large amounts of carbon are required
(39). At the 1-ppb level, organochlorine insecticides in
drinking water account for only about 5% of the dietary
105
-------
intake of these pesticides and would not pose a threat to
human health, at4east with respect to acute effects (94).
Other techniques for pesticide removal from water
have been explored, with limited success. Chemical or
biological oxidation will degrade some pesticides, but
toxic products are formed in many cases (39); ozone will
attack even the stable organochlorine insecticides, but
only at large and impractical concentrations and with
unknown products (135); and strongly basic anion-
exchange resins will remove high concentrations of such
pesticides as 2,4-D salts (3). Use of reverse osmosis
membranes has shown promise for removal of a wide
variety of pesticides (40). Other effective processes may
yet be developed that will take advantage of specific
characteristics of individual pesticides, one possible
example being the introduction of microorganisms that
are specific for rapid inactivation of certain classes of
chemicals.
PRACTICES FOR REDUCING ENTRY OF PESTICIDES
INTO THE AQUATIC ENVIRONMENT
A total of 15 pesticide management practices, desig-
nated as P 1 through P 15, are presented in Volume I
(Table 18 and Section 4.4). TTiese practices reduce
pesticide losses in runoff from treated fields by manipu-
lation of the chemical itself and are meant to supple-
ment the basic control of runoff and erosion carrying
the pesticide, which is dealt with in other sections of
Volume I. The various aspects of the practices are
discussed in Volume I without supportive documen-
tation. Appropriate documentation for each of the
practices is, however, presented in Table 10.
Table 10 contains citations of articles supporting
direct statements made in Volume I and of articles
containing closely related information that may be of
benefit in evaluating the individual practices. Informa-
tion for both Volume I and this volume was obtained
not only from the published papers cited, but also from
discussions and meetings with agricultural and pesticide
specialists and from internal progress reports of the
CRIS (Current-Research-in-Science) information re-
trieval system of the U.S. Department of Agriculture and
the State agricultural experiment stations.
RESEARCH NEEDS
The length of this review constitutes first-hand evi-
dence that much is already known about pesticides in
the aquatic environment, yet numerous important ave-
nues for future research remain. Further efforts should,
of course, be directed to minimization of those agricul-
tural practices that contribute to erosion and runoff
from cropland and thereby produce excessive losses of
applied chemicals, but our concern here is with aspects
of the system that deal with the pesticides directly. Such
needs appear to fall into five general areas: (1) predic-
tion of pesticide behavior in the aquatic ecosystem; (2)
definition of significance of residues occurring in water
bodies; (3) investigation of means for lowering rates and
frequency of application of pesticides, so that the
potential for contamination of waters would be lessened;
(4) development of new pesticides having environ-
mentally favorable properties; and (5) research on
corrective measures to reduce or remove contamination
by applied pesticides. No order of priority among these
is intended in the brief discussions that follow.
Prediction of Pesticide Behavior
The development of mathematical models to predict
the behavior of pesticides after application is a relatively
new and important area of research. The ideal model is
one that is able to predict the consequences of any given
practice or set of conditions; applying it, one can
identify optimum modes of pesticide use with respect to
some particular attribute. However, a large amount of
work is required to construct useful models of the
typically complex agricultural systems. Moreover, data
collection could be a limiting step; a complete model
may demand such an elaborate input of hydrologic,
chemical, biologic, and management data that collection
of real-world numbers, including analyses, might take so
long that events would outrun predictions.
Despite these difficulties, development and refine-
ment of exploratory models is being actively pursued.
With respect to pesticide movement in runoff, a model
has been developed with the objective of minimizing
water pollution (12, 53). It takes into account condi-
tions both during and between runoff events, and has
given satisfactory predictions for pesticides moved en-
tirely on sediment, but not yet for those moved in both
water and sediment. Models in related areas have also
been proposed. The movement of agricultural chemicals
through the soil profile has been described mathemati-
cally, but existing models do not take into account the
ongoing natural soil-forming processes, so that their
106
-------
Table 10. Bibliography on pesticide management practices (Volume I, Section 4.4)
Pesticide management practice
No.
Page No.
in
Vol. I.
Description
Citations
Significant subjects
PI
85
Using Alternative Pesticides
P2
85
Optimizing Pesticide
Placement With Respect
to Loss
P3
85
Using Crop Rotation
Craig et al (52)
Bailey (10)
Spencer (144)
Erbach and Lovely (64)
CaroeM/(35)
Ritterefa/(134)
Apple (1)
Constien ef a/(4J)
Moomaw and Robison (116)
Reid and Peacock (132)
Erbach, Lovely, and
Bockhop(65)
Bode and Gebhardt (25)
Wax (159)
Epstein and Grant (6j)
Wax (159)
Stockdale, DeWitt, and
Ryan (147)
Daniels (55)
Fleming (6J), p. 327
Blakely, Coyle, and Steele
(23), p. 305
Wade (155)
Typical examples of alternative pesticides
effective against same pests in same crops
Discussion of pesticide properties pertinent
to movement in runoff
Discussion of pesticide properties pertinent
to movement in runoff
Desirability of rotating equally effective
pesticides in succeeding years on same crop
Comparative pesticide loss in runoff: in-furrow
vs. broadcast applications
Comparative pesticide loss in runoff: ridge
planting vs. surface-contour planting
Comparative toxicity to crop seed of insecti-
cides when placed in seed furrow
Necessity for placing insecticides in seed furrow
in no-till management
Satisfactory performance of herbicides placed
in narrow bands
Subsurface sweep applicators
High efficiency of precise spacing of herbicides
Advantage of disk over other implements in
incorporating pesticides to minimize loss
in runoff
Current trends toward broadcast application
Lower runoff loss of pesticides from rotation
Weed reduction by crop rotation
Improved insect control by crop rotation
Improved insect control by crop rotation
Typical examples of improved insect control
by crop rotation
Reduction in erosional losses by rotation
Intercropping of corn and peanuts to reduce
attack by corn borers
107
-------
Table 10. (continued)
No.
Pesticide Management Practice
Page No.
in
Vol. I.
Description
Citations
Significant subjects
P4
P5
P6
P7
86
86
86
87
Using Resistant Crop
Varieties
Optimizing Crop Planting
Time
Optimizing Pesticide
Formulation
Using Mechanical Control
Methods
Kuhlman, Cooley, and
Walt (103)
AUaway(l),p. 391-2
Sprague and Dahms (146)
Chant (38), p. 204
Hoffman (93)
Fleming (69), p. 328
Craig era/(52), p. 48
Wellhausen (164)
Waxefaf(160)
Bulkhead et al (28)
Wax (159)
t
Foy and Bingham (70)
Mullison (119)
Barnett effl/(16)
Miles and Woehst (115)
Clack (42)
Depew (57)
Burnside and Colville (29)
Wax (159)
Whitaker, Heinemann, and
Wischmeier (165)
Crafts and Robbins (51),
p. 140-154
Behrens (20)
Comparative acreage receiving insecticides:
continuous crop vs. rotations
Comparative weed control: continuous crop
vs. rotations
Review of crop resistance to insects and
reductions in use of insecticides
Wheat varieties resistant to Hessian fly
Examples of resistant crop varieties and insects
resisted
Time of crop planting: summary of effects
on insect infestations
Advantages of early plantings for combatting
European corn borer
Break in sorghum plantings to combat sorghum
midge
Advantages of late plantings of soybeans to
combat weeds
Summary of planting dates of field crops
Addition of surfactants to increase penetration
of herbicides; comparability of liquids and
granules
Surfactants or oils to enhance herbicide
penetration in plants
Toxicity of components of formulations other
than active ingredient
Comparative runoff potential: 2,4-D esters
vs. amine salt
Controlled release formulations
Use of foam formulations for weed control
Superiority of granular formulations over
liquids in seed-furrow applications
Superiority of tillage-herbicide combination
for weed control
Discussion of tillage and flame cultivation for
weed control
Ability of cultivation to reduce erosive soil
loss
Discussion of tillage methods in weed control
Disadvantages of cultivation and tillage
108
-------
Table 10. (continued)
Pesticide management practice
No.
Page No.
in
Vol. I.
Description
Citations
Significant subjects
P8
87
P9
87
P10
87
Pll
88
P12
88
PI 3
Eliminating Excessive
Treatment
Optimizing Time of Day for
Pesticide Application
Optimizing Date of Pesticide
Application
Using Integrated Control
Programs
Using Biological Control
Methods
Using Lower Pesticide
Application Rates
Turnipseedgfa/(152)
Chiang (41)
Shore (143)
Ware et al (157)
Cooperative Exten. Serv.,
Illinois (47), p. 238
Apple, Walgenbach, and
Knee (8)
Harrison and Press (86)
Anderson (4)
Texas Agr. Exten. Serv.
(149)
Erbach and Lovely (64)
Summers, Byrne, and
Pimentel (148)
Hanson (84)
Council Environ. Qual. (49)
Chant (38)
Casey, Lacewell, and
Sterling (37)
Giese, Peart, and Huber (73)
Knipling (101)
Quraishi (129)
Putnam and Duke (128)
Patti and Camer (124)
Tumipseed et al (152)
Casey, Lacewell, and Sterling
(37)
Effectiveness of lower than recommended rate
for insect control in soybeans
Development of recommendations for deter-
mining threshold pest damage
Mathematical computation of optimum dosages
Higher efficiency of early morning spraying
Timing of spray to avoid harm to honeybees
Comparative effectiveness: planting-time vs.
cultivation-time treatments for corn
rootworm
Timing of sprays against corn borer
Optimization of time of foliar application of
herbicides
Recommendations for long-interval preplan!
applications of herbicides on cotton and
peanuts
Critical period for applying herbicides on corn
and soybeans
Advantages of early insecticide application
for alfalfa weevil control
Aspects of integrated pest control programs
Aspects of integrated pest control programs
Aspects of integrated pest control programs
Example of reduction in insecticide use by
introduction of pest management strategy
Reliability of computer-based pest manage-
ment systems
Overview of biological control of insects
Aspects of biological control of insects
Example of biological control of weeds
Example of usefulness of Bacillus thuringiensis
Examples of effectiveness of rates less than
recommended levels
Example of lower dosage in an integrated
control program
109
-------
Table 10. (continued)
Pesticide Management Practice
No.
Page No.
in
Vol. I.
Description
Citations
Significant subjects
P14
89
Managing Aerial Applications
P15
89
Planting Between Rows in
Minimum Tillage
Johnstone (96)
Wax (159)
Marstonefg/(lll)
Cole, Barry, and Frear (43)
Glotfelty and Caro (74)
Coop. Exten. Serv. Ohio (46)
Way (161)
Insecticide application in ultra-low-volume
sprays
Advantages and disadvantages of aerial
application of herbicides
Pesticide in stream water from aerial spraying
DDT in environment after aerial application
Movement of airborne pesticides
Practice recommended for reduction of corn
rootworm populations
Disadvantages of practice
ability to predict field behavior is limited (24). A
relatively simple model has been developed (79) in
which the quantities of pesticide to be applied are
optimized with respect to profits to the grower. Refine-
ment of this model could well lead to reduced use of
pesticides and lessened environmental contamination.
Significance of Residues
The true significance of pesticide residues in the
environment is perhaps the least understood aspect of
the system, particularly with regard to chronic contami-
nation at very low concentrations. Changes in behavioral
patterns of aquatic organisms have been observed as a
result of chronic exposure, but little is truly known of
possible long-term, subtle effects (118). To aid in
assessment of the hazard, we need (1) in-depth studies of
declining species; (2) studies of the gain, loss, or change
in residues in both living and nonliving components of
the environment, to relate trends to observable effects;
and (3) lexicological measurements under conditions
that simulate the natural environment more closely than
the conditions used in such tests in the past.
Reducing Pesticide Use
Research directed to reduction in use of pesticides by
more efficient application of the chemical or by substi-
tution of nonchemical methods of pest control offers
the broadest opportunity for decreasing the potential for
environmental contamination by pesticides. One impor-
tant facet that will require considerable future effort is
the development of large-scale integrated control pro-
grams in which minimum amounts of chemical pesticides
are used. Much more information is needed for inte-
grated control than is generally required to use pesticides
alone. A successful program for insect control, for
example, requires knowledge of the dynamics of the pest
population, life history of the pest, natural enemies,
nutritional requirements, host plants, economic thres-
hold of the insect population, and behavior of chemicals
and organisms used in the program with respect to
effects on nontarget environmental components (93,
114). Each of these must be examined in detail and
interrelated and there are also researchable associated
matters, such as the selection of the most suitable of
alternative methods of control for incorporation into a
program, the use of adverse natural phenomena to signal
the appropriate time for attacking insects by integrated
control, and the bringing of experimentally proven
integrated control methods up to practical application
(38, 93).
Several approaches to more efficient application and
utilization of pesticides are being actively investigated,
but require additional effort. One such is the develop-
ment and testing of foams, gels, and polymer-encapsu-
lated slow-release formulations that can reduce drift,
minimize movement of the pesticide in the environment,
and perhaps decrease the number of applications needed
110
-------
because the chemical will be used more efficiently. A
second is the use of electrostatically charged sprays to
decrease drift and optimize deposition of the chemicals
onto plant surfaces. Optimization of spray droplet
particle sizes for efficient on-target deposition is also
being investigated. A third avenue under investigation is
the reduction of pesticide volatilization from plant and
soil surfaces, and a fourth is the precision placement of
pesticides in the soil by devices such as subsurface sweep
applicators. Another aspect worthy of further study is
the development of computer programs to predict
occurrences of pest infestations. Such a program has
already been successful with potato late blight, saving
growers an average of 4 sprays annually in comparison
with the normal practice of spraying at 10-day intervals
(102).
Development of New Pesticides
Present pesticides, although effective, are far from
perfect. Opportunity is still great to develop new
chemicals that are highly specific for the pest, safe to
man and wildlife, and have little effect on the quality of
the environment. The ideal compound would have low
solubility in fats to minimize the possibility of biomag-
nification; would be biodegradable, but only after its
intended function is completed; and would be nontoxic,
but convertible to a toxicant in the presence of the pest
(113).
Corrective Measures
Research should be conducted on means for shorten-
ing the persistence of relatively stable compounds at the
site of application or in the aquatic environment by
deliberate manipulation of their modes of dissipation.
Work of this type would be directed to such goals as
enhancement of volatility and photodecomposition or
adjustment of adsorptivity and teachability, perhaps by
use of adjuvants. Other possible corrective efforts could
involve the use of aquatic plants as traps to remove
residues from water by absorption, addition of an
inoculum of microorganisms to degrade the pesticides
after their biocidal activity is no longer needed, direct
chemical inactivation in soil by addition of a reactant,
use of additives to control the metabolic degradation of
herbicides within plant tissues, and use of new adsorp-
tive media or ion exchange resins to remove residues
from water (70).
Ill
-------
LITERATURE CITED
1. Allaway, W. H. 1957. Cropping systems and soil./w
Soil, the yearbook of agriculture, 1957: 386-395.
U.S. Dept. Agr., Washington, D.C.
2. Aly, O. M., and Faust, S. D. 1964. Studies on the
fate of 2,4-D and ester derivatives in natural surface
waters. Jour. Agr. Food Chem. 12(6): 541-546.
3. Aly, O. M., and Faust, S. D. 1965. Removal of
2,4-dichlorophenoxyacetic acid derivatives from
natural waters. Jour. Amer. Water Works Assoc.
57(2): 221-230.
4. Anderson, L. E. 1973. Factors affecting herbicide
performance. Agron. 8-1. Exten. Div., Univ.
Missouri, Columbia, 2p.
5. Andrilenas, P. A. 1974. Farmers' use of pesticides in
1971. Econ. Res. Serv., U.S. Dept. Agr., Agr. Econ.
Rep. No. 252, 56 p.
6. Anonymous. 1974. Pest. Chem. News, 3(17). Food
Chem. News, Inc., Washington, D.C.
7. Apple, J. W. 1971. Response of com to granular
insecticides applied to the row at planting. Jour.
Econ. Entomol. 64(5): 1208-1211.
8. Apple, J. W., Walgenbach, E. T., and Knee, W. J.
1969. Northern corn rootworm control by granular
insecticide application at planting and cultivation
time. Jour. Econ. Entomol. 62(5): 1033-1035.
9. Armstrong, D. E., and Konrad, J. G. 1974. Nonbio-
logjcal degradation of pesticides. In W. D.
Guenzi, ed. Pesticides in soil_and water: 123-131.
Soil Sci. Soc. Amer., Inc., Madison, Wis.
10. Bailey, G. W. 1966. Entry of biocides into water
courses. Univ. Calif. Water Resour. Center Rep. 10:
94-103.
11. Bailey, G. W., Barnett, A. P., Payne, W. R., Jr., and
Smith, C. N. 1974. Herbicide runoff from four
coastal plain soil types. EPA-660/2-74-017. U.S.
Govt. Printing Off., Washington, D.C., 98 p.
12. Bailey, G. W., Swank, R. R., Jr. and Nicholson, H.
P. 1974. Predicting pesticide runoff from agricul-
tural land: a conceptual model. Jour. Environ. Qual.
3(2): 95-102.
13. Bailey, G. W., and White, J. L. 1964. Review of
adsorption and desorption of organic pesticides by
soil colloids, with implications concerning pesticide
bioactivity. Jour. Agr. Food Chem. 12(4): 324-332.
14. Bailey, T. E., and Hannum, J. R. 1967. Distribution
of pesticides in California. Jour. San. Eng. Div.,
Proc. Amer. Soc. Civil Eng. 93(SA5): 27.
15. Baldwin, F. L., Santelmann, P. W., and Davidson, J.
M. 1975. Movement of fluometuron across and
through the soil. Jour. Environ. Qual. 4(2):
191-194.
16. Barnett, A. P., Hauser, E. W., White, A. W., and
Holladay, J. H. 1967. Loss of 2,4-D in washoff of
cultivated fallow land. Weeds 15(2): 133-137.
17. Barrens, K. C. 1971. Environmental benefits of
intensive crop production. Agr. Sci. Rev. 9(2):
33-39.
18. Barthel, W. F., Hawthorne, J. C., Ford, J. H.,
Bolton, G. C., McDowell, L. L., Grissinger, E. H.,
and Parsons, D. A. 1969. Pesticide residues in
sediments of the lower Mississippi River and its
tributaries. Pestic. Monit. Jour. 3(1): 8-66.
19. Baur, J. R., Bovey, R. W., and Merkle, M. G. 1972.
Concentration of picloram in runoff water. Weed
Sci. 20(4): 309-313.
20. Behrens, R. 1975. Corn and weeds. Weeds Today
6(1): 15-18.
21. Biggar, J. W. 1970. Pesticide movement in soil
water. In Pesticides in the soil: ecology, degrada-
tion, and movement. Internal!. Symp. on Pesticides
in the Soil: 107-119. Michigan State Univ., East
Lansing.
22. Bingham, C. R. 1969. Some effects of pesticides on
Mississippi waters. Proc. Miss. Water Resour. Conf.,
MSU Water Resour. Res. Inst., State College, Miss.
23. Blakely, B. D., Coyle, J. J., and Steele, J. G. 1957.
Erosion on cultivated land. In Soil, the yearbook of
agriculture, 1957: 290-307. U.S. Dept. Agr., Wash-
ington, D.C.
112
-------
24. Boast, C. W. 1973. Modeling the movement of
chemicals in soils by water. Soil Sci. 115(3):
224-230.
25. Bode, L. E., and Gebhardt, M. R. 1969. Equipment
for incorporation of herbicides. Weed Sci. 17(4):
551-555.
26. Bovey, R. W., Burnett, E., Richardson, C., Merkle,
M. G., Baur, J. R., and Knisel, W. G. 1974.
Occurrence of 2,4,5-T and picloram in surface
runoff water in the Blacklands of Texas. Jour.
Environ. Qual. 3(1): 61-64.
27. Bradley, J. R., Sheets, T. J., and Jackson, M. D.
1972. DDT and toxaphene movement in surface
water from cotton plots. Jour. Environ. Qual. 1(1):
102-105.
28. Burkhead, C. E., Max, R. C., Karnes, R. B., and
Reid, E. 1972. Usual planting and harvesting dates,
field and seed crops. Agr. Handbook No. 283, 84 p.,
Statis. Rep. Serv., U.S. Dept. Agr., Washington, D.C.
29. Burnside, 0. C., and Colville, W. 1964. Soybean and
weed yields as affected by irrigation, row spacing,
tillage, and amiben. Weeds 12(2): 109-112.
30. Burnside, 0. C., Wicks, G. A., and Fenster, C. R.
1971. Dissipation of dicamba, picloram, and
2,3,6-TBA across Nebraska. Weed Sci. 19(4):
323-325.
31. Butler, P. A. 1965. Effects of herbicides on estu-
arine fauna. Proc. So. Weed Conf. 18: 576-580.
32. Cairns, J., Jr., Heath, A. G., and Parker, B. C. 1975.
Temperature influence on chemical toxicity to
aquatic organisms. Jour. Water Poll. Contr. Fed.
47(2): 267-280.
33. Caro, J. H. 1969. Accumulation by plants of organo-
chlorine insecticides from the soil. Phytopathology
59(9): 1191-1197.
34. Caro, J. H., Edwards, W. M., Glass, B. L., and Frere,
M. H. 1972. Dieldrin in runoff from treated
watersheds. Proc. 1970 Symp. on Interdiscip. As-
pects of Watershed Mangt., Bozeman, Mont.:
141-160. Amer. Soc. Civil Engin., New York.
35. Caro, J. H., Freeman, H. P., Glotfelty, D. E.,
Turner, B. C., and Edwards, W. M. 1973. Dissipation
of soil-incorporated carbofuran in the field. Jour.
Agr. Food Chem. 21(6): 1010-1015.
36. Caro, J. H., Freeman, H. P., and Turner, B. C. 1974.
Persistence in soil and losses in runoff of soil-incor-
porated carbaryl in a small watershed. Jour. Agr.
Food Chem. 22(5): 860-863.
37. Casey, J. E., Lacewell, R. D., and Sterling, W. 1975.
An example of economically feasible opportunities
for reducing pesticide use in commercial agriculture.
Jour. Environ. Qual. 4(1): 60-64.
38. Chant, D. A. 1966. Integrated control systems. In
Scientific aspects of pest control: 193-218. Natl.
Acad. Sci.-Natl. Res. Council, Washington, D.C.
39. Chesters, G., and Konrad, J. G. 1971. Effects of
pesticide usage on water quality. BioScience 21(12):
565-569.
40. Chian, E. S. K., Bruce, W. N., and Fang, H. H. P.
1975. Removal of pesticides by reverse osmosis.
Environ. Sci. Technol. 9(1): 52-59.
41. Chiang, H. C. 1973. Ecological considerations in
developing recommendations for chemical control
of pests: European corn borer as a model. FAO
Plant Protec. Bui. 21(2): 30-39.
42. Clack, J. E. 1972. Application of herbicides in foam
to reduce drift. Proc. South. Weed Sci. Soc. 25:
495^97.
43. Cole, H., Barry, D., and Frear, D. E. H. 1967. DDT
levels in fish, streams, stream sediments, and soil
before and after DDT aerial spray application for
fall cankerworm in northern Pennsylvania. Bui.
Environ. Contam. Toxicol. 2(3): 127-155.
44. Constien, E. J., Anderson, L., Murphy, W. J.,
Woodruff, C. M., Palm, E., Thomas, G., and
Workman, H. 1974. Conservation tillage-planting
systems. Exten. Div., Univ. Missouri, Columbia, 4 p.
45. Cooperative Extension Service, The Ohio State
University. 1973. Pesticides and the environment.
Bui. 536,20 p.
113
-------
46. Cooperative Extension Service, The Ohio State
University. 1974. Insect control in no-tillage corn.
Bui. L-208,4 p.
47. Cooperative Extension Service, Univ. Illinois. 1974.
Guide for the custom application of pesticides in
Illinois. In Summary of 26th Illinois custom spray
operators training school: 234-252. Univ. Illinois,
Urbana-Champaign.
48. Cope, O. B. 1965. Some responses of fresh water
fish to herbicides. Proc. South. Weed Conf. 18:
439-445.
49. Council on Environmental Quality. 1972. Integrated
pest management. Washington, D.C., 41 p.
50. Cox, J. L. 1971. DDT residues in seawater and
particulate matter in the California current system.
Fish. Bui., Natl. Mar. Fish. Serv. 69(2): 443-450.
51. Crafts, A. S., and Robbins, W. W. 1962. Weed
control. 3rd ed., McGraw-Hill, New York, 660 p.
52. Craig, W. S., Brown, K. E., Huggans, J. L., Thomas,
G. W., and Jones, F. G. 1974. Insect control
recommendations, Missouri, 1974. Coop. Exten.
Serv., Univ. Missouri, Columbia, 155 p.
53. Crawford, N. H., and Donigian, A. S., Jr. 1973.
Pesticide transport and runoff model for agricultural
lands. EPA-660/2-74-013, 211 p. U.S. Govt. Print-
ing Off., Washington, D.C.
54. Crosby, D. G. 1970. The nonbiological degradation
of pesticides in soils. In Pesticides in the soil:
ecology, degradation, and movement. Internatl.
Symp. on Pesticides in the Soil: 86-94. Michigan
State Univ., East Lansing.
55. Daniels, N. 1971. Detection of insecticidal residue
and control of soil insects. Jour. Econ. Entomol.
64(1): 175-177.
56. Davis, E. A., and Ingebo, P. A. 1970. Fenuron
contamination of stream water from a chapparal
watershed in Arizona. Res. Prog. Rep., West. Soc.
Weed Sci.: 22.
57. Depew, L. 1971. Evaluation of foliar and soil
treatments for greenbug control on sorghum. Jour.
Econ. Entomol. 64(1): 169-172.
58. Edwards, C. A., Thompson, A. R., Beynon, K. I.,
and Edwards, M. J. 1970. Movement of dieldrin
through soils. I. From arable soils into ponds. Pestic.
Sci. 1: 169-173.
59. Edwards, W. M., and Glass, B. L. 1971. Methoxy-
chlor and 2,4,5-T in lysimeter percolation and
runoff water. Bui. Environ. Contam. Toxicol. 6(1):
81-84.
60. Eichelberger, J. W., and Lichtenberg, J. J. 1971.
Persistence of pesticides in river water. Environ. Sci.
Technol. 5(6): 541-544.
61. Ennis, W. B., Jr., Jansen, L. L., Ellis, I. T., and
Newson, L. D. 1967. In The world food problem. A
report of the President's Science Advisory Com-
mittee, Vol. Ill: 130-175. The White House.
62. Environmental Protection Agency. 1973. Water
quality criteria, 1972. EPA-R3-73-033, 594 p. U.S.
Govt. Printing Off., Washington, D. C.
63. Epstein, E., and Grant, W. J. 1968. Chlorinated
insecticides in runoff water as affected by crop
rotation. Soil Sci. Soc. Amer. Proc. 32(3): 423426.
64. Erbach, D. C., and Lovely, W. G. 1974. Weed
control with conservation tillage production of corn
and soybeans. Jour. Soil and Water Conserv. 29(1):
44-45.
65. Erbach, D. C., Lovely, W. G., and Bockhop, C. W.
1975. Area of influence of herbicide granules.
Submitted to Weed Sci.
66. Evans, J. 0., and Duseja, D. R. 1973. Herbicide
contamination of surface runoff waters. EPA-R2-
73-266, 99 p. U.S. Govt. Printing Off., Washington,
D.C.
67. Eye, J. D. 1968. Aqueous transport of dieldrin
residues in soils. Jour. Water Poll. Contr. Fed. 40(8)
part2:R316-R332.
68. Fahey, J. E. 1970. Contamination of farm ponds
following treatment of watersheds for pest control.
In Effect of pesticide residues and other organo-
toxicants on the quality of surface and ground
water resources: 9-13a. Purdue Univ. Water Resour.
Res. Center, Tech. Rep. No. 10.
114
-------
69. Fleming, W. E. 1957. Soil management and insect
control. In Soil, the yearbook of agriculture, 1957:
326-333. U.S. Dept. Agr., Washington, D.C.
70. Foy, C. L., and Bingham, S. W. 1969. Some research
approaches toward minimizing herbicidal residues in
the environment. Residue Rev. 29: 105-135.
71. Foy, C. L., Coats, G. E., and Jones, D. W. 1971.
Translocation of growth regulators and herbicides;
vascular plants. In P. L. Altman and D. S. Dittmer,
eds. Respiration and circulation: 743-791. Biol.
Handbooks, Fed. Amer. Soc. Expt. Biol., Bethesda,
Md.
72. Frank, P. A. 1972. Herbicidal residues in aquatic
environments. In Fate of organic pesticides in the
aquatic environment: 135-148. Advan. in Chem.
Ser. Ill, Amer. Chem. Soc., Washington, D.C.
73. Giese, R. L., Peart, R. M., and Huber, R. T. 1975.
Pest management. Science 187(4181): 1045-1052.
74. Glotfelty, D. E., and Caro, J. H. 1974. Introduction,
transport, and fate of persistent pesticides in the
atmosphere. In V. R. Deitz, ed. Removal of trace
contaminants from the air: 42-62. ACS Symp. Ser.
17, Amer. Chem. Soc., Washington, D.C.
75. Graetz, D. A., Chesters, G., Daniel, T. C., Newland,
L. W., and Lee, G. B. 1970. Parathion degradation
in lake sediments. Jour. Water Poll. Contr. Fed. 42:
R76-R94.
76. Guenzi, W. D., and Beard, W. E. 1974. Volatili-
zation of pesticides. In W. D. Guenzi, ed. Pesticides
in soil and water: 107-122. Soil Sci. Soc. Amer.,
Inc., Madison, Wis.
77. Haan, C. T. 1971. Movement of pesticides by runoff
and erosion. Trans. ASAE 14(3): 445-447,449.
78. Haas, R. H., Scifres, C. J., Merkle, M. G., Hahn, R.
R., and Hoffman, G. O. 1971. Occurrence and
persistence of picloram in grassland water sources.
Weed Res. 11(1): 54-62.
79. Hall, D. C., and Norgaard, R. B. 1973. On the
timing and application of pesticides. Amer. Jour.
Agr. Econ. 55(2): 198-201.
80. Hall, J. K. 1974. Erosional losses of s-triazine
herbicides. Jour. Environ. Qual. 3(2): 174-180.
81. Hall, J. K., Pawlus, M., and Higgins, E. R. 1972.
Losses of atrazine in runoff water and soil sediment.
Jour. Environ. Qual. 1(2): 172-176.
82. Hamaker, J. W., and Thompson, J. M. 1972.
Adsorption. In C. A. I. Goring and J. W. Hamaker,
eds. Organic chemicals in the soil environment, Vol.
I: 49-143. Marcel Dekker, Inc., New York.
83. Harmon, M. R., Greichus, Y. A., Applegate, R. L.,
and Fox, A. C. 1970. Ecological distribution of
pesticides in Lake Poinsett, South Dakota. Trans.
Amer. Fish. Soc. 99(3): 496-500.
84. Hanson, A. A. 1972. A directed ecosystem approach
to pest control and environmental quality. Jour.
Environ. Qual. 1(1): 45-54.
85. Harris, C. I. 1966. Adsorption, movement, and
phytotoxicity of monuron and s-triazine herbicides
in soU. Weeds 14(1): 6-10.
86. Harrison, F., and Press, J. 1971. Timing of insecti-
cide applications for European corn borer control in
sweet corn. Jour. Econ. Entomol. 64(6):
1496-1499.
87. Hartung, R., and Klingler, G. W. 1970. Concen-
tration of DDT by sedimented polluting oils. Envi-
ron. Sci. Technol. 4(5): 407-410.
88. Harvey, G. R. 1974. DDT and PCB in the Atlantic.
Oceanus 18(1): 19-23.
89. Helling, C. S. 1970. Movement of s-triazine herbi-
cides in soils. Residue Rev. 32: 175-210.
90. Hill, D. W. 1967. Degradation of selected chlori-
nated hydrocarbon pesticides in anaerobic and
aerobic water environments. Diss. Abstr. B27:
2399-2400.
91. Hill, D. W., and McCarty, P. L. 1967. Anaerobic
degradation of selected chlorinated hydrocarbon
pesticides. Jour. Water Poll. Contr. Fed. 39(8):
1259-1277.
115
-------
92. Hiltbold, A. E. 1974. Persistence of pesticides in
soil. In W. D. Guenzi, ed. Pesticides in soil and
water: 203-222. Soil Sci. Soc. Amer., Inc., Madison,
Wis.
93. Hoffman, C. H. 1970. Alternatives to conventional
insecticides—control of insect pests. Agr. Chem.
25(9): 14-19A.
94. Huang, J. C. 1972. Pesticides in water-effects on
human health. Jour. Environ. Health 34(5):
501-508.
95. Johnson, H. E., and Ball, R. C. 1972. Organic
pesticide pollution in an aquatic environment. In
Fate of organic pesticides in the aquatic environ-
ment: 1-10. Advan. in Chem. Ser. Ill, Amer.
Chem. Soc., Washington, D.C.
96. Johnstone, D. R. 1973. Insecticide concentration
for ultra-low volume crop spray application. Pestic.
Sci. 4(1): 77-82.
97. Kaufman, D. D. 1972. Degradation of pesticide
combinations. In A. S. Tahori, ed. Fate of pesticides
in environment. Pesticide Chemistry Vol. VI:
175-204. Gordon and Breach, Inc., New York.
98. Kaufman, D. D. 1974. Degradation of pesticides by
soil microorganisms. In W. D. Guenzi, ed. Pesticides
in soil and water: 133-202. Soil Sci. Soc. Amer.,
Inc., Madison, Wis.
99. Kearney, P. C., Nash, R. G., and Isensee, A. R.
1969. Persistence of pesticide residues in soils. In M.
W. Miller and G. G. Berg, eds. Chemical fallout.
Current research on persistent pesticides: 54-67.
Charles C. Thomas, Pub., Springfield, m.
100. King, P. H., Yeh, H. H., Warren, P. S., and Randall,
C. W. 1969. Distribution of pesticides in surface
waters. Jour. Amer. Water Works Assoc. 61(9):
483486.
101. Knipling, E. 1972. Use of organisms to control
insect pests. Jour. Environ. Qual. 1(1): 34-40.
102. Krause, R. A., and Massie, L. B. 1973. Application
and implementation of computerized forecasts of
potato late blight. Phytopathology 63(2): 203.
103. Kuhlman, D. E., Cooley, T. A., and Walt, J. 1974.
Adult com rootworm populations in Illinois,
August 1973. In Summary of 26th Illinois custom
spray operators training school: 159-164. Coop.
Exten. Serv., Univ. Illinois, Urbana-Champaign.
104. Lauer, G. J., Nicholson, H. P., Cox, W. S., and
Teasley, J. I. 1966. Pesticide contamination of
surface waters by sugarcane fanning in Louisiana.
Trans. Amer. Fish. Soc. 95(3): 310-316.
105. Leland, H. V., Bruce, W. N., and Shimp, N. F.
1973. Chlorinated hydrocarbon insecticides in
sediments of southern Lake Michigan. Environ.
Sci. Technol. 7(9): 833-838.
106. Leshniowsky, W. O., Dugan, P. R., Pfister, R. M.,
Frea, J. I., and Randies, C. I. 1970. Aldrin:
removal from lake water by flocculent bacteria.
Science 169: 993-995.
107. Letey, J., and Oddson, J. K. 1972. Mass transfer.
In C.A.I. Goring and J. W. Hamaker, eds. Organic
chemicals in the soil environment, Vol. I: 399-440.
Marcel Dekker, Inc., New York.
108. Lichtenberg, J. J., Eichelberger, J. W., Dressman,
R. C., and Longbottom, J. E. 1970. Pesticides in
surface waters of the United States—a 5-year
summary, 1964-1968. Pestic. Monit. Jour. 4(2):
71-86.
109. Locke, D. O., and Havey, K. 1972. Effects of DDT
upon salmon from Schoodic Lake, Maine. Trans.
Amer. Fish. Soc. 101: 638.
.110. Lutz, J. F., Byers, G. E., and Sheets, T. J. 1973.
The persistence and movement of picloram and
2,4,5-T in. soils. Jour. Environ. Qual. 2(4):
485-488.
111. Marston, R. B., Schults, D. W., Thiroyama, T., and
Snyder, L. V. 1968. Amitrole concentrations in
creek waters downstream from an aerially sprayed
watershed sub-basin. Pestic. Monit. Jour. 2(3):
123-128.
112. McDowell, L. L., Grissinger, H. H., Bolton, G. C.,
and Parsons, D. A. 1971. Chlorinated hydrocarbon
insecticide contamination of streambed sediments
in the Mississippi River Delta. Proc. Miss. Water
Resour. Conf., Water Res. Inst., MSU, State
College, Miss.
116
-------
113. McNew, G. L. 1972. Interrelationships between
agricultural chemicals and environmental quality in
perspective. Jour. Environ. Qual. 1(1): 18-22.
114. Metcalf, R. L. 1972. Agricultural chemicals in
relation to environmental quality: insecticides to-
day and tomorrow. Jour. Environ. Qual. 1(1):
10-14.
124. Patti, J. H., and Garner, G. R. 1974. Bacillus
thuringiensis investigations for control of Heliothis
spp. on cotton. Jour. Econ. Entomol. 67(3):
415-418.
125. Phillips, W. M., and Feltner, K. C. 1972. Persist-
ence and movement of picloram in two Kansas
soils. Weed Sci. 20(1): 110-116.
115. Miles, J. W., and Woehst, J. E. 1969. Formulations
for controlled release of Abate in water. In
Pesticidal formulations research: 183-191. Advan.
Chem. Ser. 86, Amer. Chem. Soc., Washington,
D.C.
116. Moomaw, R., and Robison, L. 1973. Broadcast or
banded atrazine plus propachlor with tillage vari-
ables in corn. Weed Sci. 21(2): 106-109.
117. Moubry, R. J., Helm, J. M., and Myrdal, G. R.
1968. Chlorinated pesticide residues in an aquatic
environment located adjacent to a commercial
orchard. Pestic. Monit. Jour. 1(4): 27-29.
118. Muirhead-Thomson, R. C. 1971. Pesticides and
fresh water fauna. Academic Press, New York, 245
P-
119. Mullison, W. R. 1970. Effects of herbicides on
water and its inhabitants. Weed Sci. 18(6):
738-750.
120. Nash, R. G. 1974. Plant uptake of insecticides,
fungicides, and fumigants from soils. In W. D.
Guenzi, ed. Pesticides in soil and water: 257-313.
Soil Sci. Soc. Amer., Inc., Madison, Wis.
121. Nicholson, H. P., and Thoman, J. R. 1965.
Pesticide persistence in public water, their detec-
tion and removal.//! L. O. Chichester, ed. Research
on pesticides: 181-189. Academic Press, New
York.
122. Panel on Pesticide Monitoring, Working Group on
Pesticides. 1971. National pesticide monitoring
program. Pestic. Monit. Jour. 5(1): 35-71.
123. Paris, D. P., and Lewis, D. L. 1973. Chemical and
microbial degradation of ten selected pesticides in
aquatic systems. Residue Rev. 45: 95-124.
126. Pionke, H. B., and Chesters, G. 1973. Pesticide-
sediment-water interactions. Jour. Environ. Qual.
2(1): 29-45.
127. President's Science Advisory Committee. 1965.
Restoring the quality of our environment. The
White House.
128. Putnam, A. R., and Duke, W. B. 1974. Biological
suppression of weeds: evidence for allelopathy in
accessions of cucumber. Science 185: 370-372.
129. Quraishi, M. S. 1971. Nontoxic approaches to
insect control, N. Dak. Farm Res. 28(4): 33-35.
130. Raghu, K., and MacRae, I. C. 1967. The effect of
the gamma-isomer of BHC upon the microflora of
submerged rice soil. Can. J. Microbiol. 13: 173 and
625.
131. Randall, C. W. 1967. Pollutional aspects of pesti-
cides in natural streams. Amer. Water Resour.
Assoc. Bui. 3(3): 47-53.
132. Reid, J. T., and Peacock, H. A. 1973. Subsurface
sweep for applying herbicides. Agron. Jour. 65(6):
996-998.
133. Rhode Island Water Resources Center. 1972.
Eighth annual report: 26-32.
134. Ritter, W. F., Johnson, H. P., Lovely, W. G., and
Molnau, M. 1974. Atrazine, propachlor, and
diazinon residues in small agricultural watersheds.
Runoff losses, persistence, and movement. Envi-
ron. Sci. Technol. 8(1): 38-42.
135. Robeck, G. G., Dostal, K. A., Cohen, J. M., and
Kreissl, J. F. 1965. Effectiveness of water treat-
ment processes in pesticide removal. Jour. Amer.
Water Works Assoc. 57(2): 181-199.
117
-------
136. Rowe, D. R., Canter, L. W., Snyder, P. J., and
Mason, J. W. 1971. Dieldrin and endrin concentra-
tions in a Louisiana estuary. Pestic. Monit. Jour.
4(4): 177-183.
137. Schafer, M. L., Peeler, J. T., Gardner, W. S., and
Campbell, J. E. 1969. Pesticides in drinking water.
Waters from the Mississippi and Missouri rivers.
Environ. Sci. Technol. 3(12): 1261-1269.
138. Schulze, J. A., Manigold, D. B., and Andrews, F. L.
1973. Pesticides in selected western streams—
1968-71. Pestic. Monit. Jour. 7(1): 73-84.
139. Scifres, C. J., Allen, T. J., Leinweber, C. L., and
Pearson, K. H. 1973. Dissipation and phytotoxi-
city of dicamba residues in water. Jour. Environ.
Qual. 2(2): 306-309.
140. Seba, D. B., and Corcoran, E. F. 1969. Surface
slicks as concentrators of pesticides in the marine
environment. Pestic. Monit. Jour. 3(3): 190-193.
141. Sharpies, J. A., and Walker, R. 1974. Shifts in
cropland use in the North Central region. Agr.
Econ. Res. 26(4): 106-110.
142. Sheets, T. J., Bradley, J. R., Jr., and Jackson, M.
D. 1972. Contamination of surface and ground
water with pesticides applied to cotton. Univ. No.
Car., Water Resour. Res. Inst., Rep. No. 60, 63 p.
143. Shore, J. 1974. Accurate residue prediction. Pest
Control 42(12): 18-21.
144. Spencer, W. F. 1970. Distribution of pesticides
between soil, water, and air. In Pesticides in the
soil: ecology, (degradation, and movement. Inter-
natl. Symp. on Pesticides in the Soil: 120-128.
Michigan State Univ., East Lansing.
145. Spencer W. F., Farmer, W. J., and Cliath, M. M.
1973. Pesticide volatilization. Residue Rev. 49:
147.
146. Sprague, G. F., and Dahms, R. G. 1972. Develop-
ment of crop resistance to insects. Jour. Environ.
Qual. 1(1): 28-34.
147. Stockdale, H. J., DeWitt, J., and Ryan, S. 1974.
Summary of Iowa insect pest control recommenda-
tions for 1974. Coop. Exten. Serv., Iowa St. Univ.,
Ames. IC-328(Rev.), 21p.
148. Summers, C. G., Byrne, H. D., and Pimentel, D.
1971. Spring timing applications for control of the
alfalfa weevil in New York. Jour. Ecoh. Entomol.
64(2): 478479.
149. Texas Agricultural Extension Service. 1974. Sug-
gestions for weed control with chemicals. Part
I-MP 1059, 32p. Texas A&M Univ., College Sta-
tion.
150. Texas A&M University, College of Agriculture.
1970. Special report: impact of drastic reduction
in the use of agricultural chemicals on food and
fiber production and cost to consumer. 62p.
151. Trichell, D. W., Morton, H. L., and Merkle, M. G.
1968. Loss of herbicides in runoff water. Weed Sci.
16(4): 447-449.
152. Turnipseed, S. G., Todd, J. W., Greene, G. L.,and
Bass, M. H. 1974. Minimum rates of insecticides on
soybeans: Mexican bean beetle, green cloverworm,
corn earworm and velvetbean caterpillar. Jour.
Econ. Entomol. 67(2): 287-291.
153. Upchurch, R. 1966. Behavior of herbicides in soil.
Residue Rev. 16: 47-85.
154. U.S. Bureau of the Census. 1973. Census of
agriculture, 1969. U.S. Govt. Printing Off., Wash-
ington, D.C.
155. Wade, N. 1975. International agricultural research.
Science 188 (4188): 585-589.
156. Waldron, A. C. 1974. Pesticide movement from
cropland into Lake Erie. EPA-660/2-74-032, 95p.
U.S. Govt. Printing Off., Washington, D.C.
157. Ware, G. W., Estesen, B. J., Cahill, W. P., and
Frost, K. R. 1972. Pesticide drift VI: target and
drift deposits vs. time of applications. Jour. Econ.
Entomol. 65(4): 1170-1171.
158. Ware, G. W., and Roan, C. C. 1970. Introduction
of pesticides with aquatic microorganisms and
plankton. Residue Rev. 33: 1545.
159. Wax, L. M. 1973. Weed control. In B. E. Caldwell,
ed. Soybeans: improvement, production, and uses:
417457. Amer. Soc. Agron., Madison, Wis.
118
-------
160. Wax, L. M., Stoller, E. W., Slife, F. W., and
Andersen, R. N. 1972. Yellow nutsedge control in
soybeans. Weed Sci. 20(2): 194-201.
161. Way, W. A. 1968. Soil tillage for corn: maxi-mini-
zero. Farm Facts No. 40, Exten. Serv., Univ.
Vermont. 2p.
162. Weatherholtz, W. M., Cornwell, G. W., Young, R.
W., and Webb, R. E. 1967. Distribution of hep-
tachlor residues in pond ecosystems in south-
western Virginia. Jour. Agr. Food Chem. 15(4):
667-670.
163. Weber, J. B., Weed, S. B., and Sheets, T. J. 1972.
Pesticides-how they move and react in the soil.
Crops and Soils 25(1): 14-17.
164. Wellhausen, H. W. 1972. Sorghum production for
grains or silage. Coop. Exten. Serv., Univ.
Arkansas, Leaflet 427 (Rev.), 8p.
165. Whitaker, F. D., Heinemann, H. G., and Wisch-
meier, W. H. 1973. Chemical weed controls affect
runoff, erosion, and corn yields. Jour. Soil and
Water Conserv. 28(4): 174-176.
166. White, A. W., Barnett, A. P., Wright, B. G., and
Holladay, J. H. 1967. Atrazine losses from fallow
land caused by runoff and erosion. Environ. Sci.
Technol. 1(9): 740-744.
167. Wiese, A. F. 1975. Movement and persistence of
fiuometuron in the South: II. Lateral movement in
runoff water. Proc. South. Weed Sci. Soc., in press.
168. Williams, P. M., and Robertson, K. J. 1975.
Chlorinated hydrocarbons in sea-surface films and
subsurface waters at nearshore stations and in the
North Central Pacific Gyre. Fish. Bui., Natl.
Ocean. Atmos. Admin. 73(2): 445-446.
169. Willis, G. H., and Hamilton, R. A. 1973. Agricul-
tural chemicals in surface runoff, ground water,
and soil. I. Endrin. Jour. Environ. Qual. 2(4):
463-466.
170. Yeo, R. R. 1967. Dissipation of diquat and
paraquat and effects on aquatic weeds and fish.
Weed Sci. 15(1): 4246.
119
-------
CHAPTER 6
INTERDISCIPLINARY RESEARCH NEEDS
B. A. Stewart
The preceding chapters have listed research needs
relative to particular disciplines. This chapter briefly
discusses their interrelationships as they relate to non-
point pollution from cropland. Nonpoint pollution can
best be controlled by controlling erosion. However,
knowledge is critically lacking concerning the degree of
erosion and sediment control necessary to control
nutrient and pesticide losses.
The Universal Soil Loss Equation is being used
extensively in predicting nonpoint pollution. It is an
excellent tool because it quantifies soil loss and the
effect of various practices on reducing this loss. The
Universal Soil Loss Equation, however, estimates the
average annual soil loss from a field area and does not
indicate the amount of sediment that is delivered to a
stream. The sediment delivery ratio is an attempt to
relate field losses to amounts reaching the stream. The
difficulty with this concept for predicting water pollu-
tion is that it takes into account both the deposition of
sediment as it moves toward the stream and the gains
from channel erosion. Pollutants, particularly pesticides
and nutrients from added fertilizers, are usually not
associated with sediment from channel erosion.
The enrichment ratio is a measure of the increase in
the concentration of a pollutant associated with the
sediment that actually reaches a stream compared to the
concentration in the watershed soil. The concentration
usually increases because more nutrients and pesticides
are adsorbed on fine-textured particles than on coarse
particles, and more coarse particles are deposited as the
sediment moves from the field area to the stream.
Consequently, the most pressing research need is to gain
a better understanding of how the Universal Soil Loss
Equation, the sediment delivery ratio, and the enrich-
ment ratio can be meshed. These measurements and
evaluations will not be made easily, cheaply, or quickly.
Data obtained from small plots or field area can provide
only crude applications because the size effect is so
significant. Since cropland is so diffuse, monitoring of
agricultural areas is impractical. The most likely ap-
proach, therefore, is to develop predictive models. A
major effort should be directed toward obtaining neces-
sary data and developing such models. Results from such
an effort cannot be expected to be precise but should
represent a statistical approach that establishes relation-
ships and relative levels.
The Universal Soil Loss Equation predicts average
annual soil losses. Additional accuracy is needed to
predict single events. These can be important in their
effect on water quality, especially if the time of loss is
associated with applications of agricultural chemicals.
Again, information is needed on large watershed areas. It
is fairly easy to instrument a field-sized area and measure
losses of sediment and associated pollutants. The diffi-
culty is in determining how much of these actually reach
the stream. The effects and proper design of filter strips,
settling basins, and sediment traps should also be
determined. Some evidence indicates that these can very
effectively reduce losses; however, there are also indica-
tions that this deposited material may be moved during
extreme runoff events.
No-till and conservation tillage systems are highly
effective for controlling soil loss. Additional research is
needed, however, because it is not known how widely
these practices can be used. Insect and disease hazards
are greater, and if these practices are used on vast
acreages, outbreaks could occur. Also, in some areas, it
may be necessary to occasionally plow to loosen the soil.
More data are also required concerning pesticide and
nutrient losses from no-till and conservation tillage
systems as compared to conventional systems. Sediment
losses are drastically reduced, but the meager data
available do not show proportionate losses in pollutants.
Another primary research need is to evaluate the
economic impact of control measures for controlling
sediment and chemical losses. Available technology is
adequate to control these pollutants in most instances.
Implementation of these measures could produce major
changes in cultivation practices, timing of chemical
applications, and location of production and general
121
-------
farming practices. These changes would have direct
effects on the profitability of agricultural production,
supplies of food and fiber, and rural land use. An urgent
need is to appraise alternative procedures for controlling
pollutants so that economic hardships can be minimized.
Since our landscapes are not uniform and all areas are
not equidistant from a water body, numerous combina-
tions of practices can achieve the same water-quality
goal. The most desirable approach is to give each farmer
the opportunity to select the appropriate combination
of practices. To do this, he needs to know how much
control each practice will provide. The costs and benefits
of implementing the various practices should also be
known. This should include regional and national im-
pacts. Thus, much research is needed in a variety of
disciplines and the first step in any of them is to
quantify the relationships.
Last, and perhaps most important, criteria must be
established as to what levels of sediment and chemicals
constitute pollution. The criteria might be absolute
limits (not to exceed X ppm) or conditional limits (not
to exceed Y ppm in Z years). Although the establish-
ment of these criteria will require research outside of
traditional agricultural areas—limnology, aquatic botany
and zoology, and water treatment—it is of great impor-
tance to the agricultural community. Without these
criteria it is impossible to determine which practice will
be adequate, much less decide what the economic
impacts will be.
122
-------
APPENDIX A
SIMULATION OF DAILY POTENTIAL DIRECT RUNOFF
INTRODUCTION
The amount and seasonal distribution of direct runoff
was estimated to assess potential transport of pesticides
and nutrients. The effects of some land management
practices on direct runoff were also estimated. Hydrol-
ogists have developed several rainfall-runoff models of
various degrees of complexity for making these esti-
mates. The more physically realistic models are quite
complicated and require a great deal of input informa-
tion and computer time. The national scope of this
report and the severe time constraints involved dictated
the use of a rather simple method of estimating runoff
from rainfall. Any input information required must also
be readily available. After considering several possi-
bilities, we decided to use the Soil Conservation Service
procedure for estimating direct runoff from storm
rainfall (4).
THE SOIL CONSERVATION SERVICE PROCEDURE FOR ESTIMATING DIRECT RUNOFF
FROM STORM RAINFALL
The Soil Conservation Service procedure for estimat-
ing direct runoff from storm rainfall (sometimes called
the SCS curve number method) was designed to use the
most generally available rainfall data: total daily rainfall.
For this reason rainfall intensity is largely ignored. The
basic relationship is the equation:
Q =
(P-Ia)2
Pj>L
(1)
where
Q
P
L
runoff in inches
rainfall in inches
initial abstraction in inches
u
S = potential maximum retention plus initial
abstraction.
The initial abstraction before runoff begins is con-
sidered to consist mainly of interception, infiltration and
surface storage. Utilizing limited data from small experi-
mental watersheds, the following empirical relationship
was developed:
!a = (0.2)S. (2)
Substituting this relationship into equation (1) gives
0= (P-C.2S)2
V P+0.8S
,P>(0.2)S
(3)
SCS
which is the rainfall-runoff relation used in the
method.
The parameter CN (runoff curve number of hydrol-
ogic soil-cover complex number) is defined in terms of
the parameter S as:
CN =
1000
S+10
(4)
Note that runoff equals rainfall when S = 0 and CN =
100.
The potential maximum retention, S , and therefore
the runoff curve number are related to sofl surface and
profile properties, the vegetative cover, management
practices, and the soil water content on the day of the
storm. Solutions of equation (3) are shown as a family
of curves in Fig. 1.
Soil water content on the day of the storm is
accounted for by an Antecedent Moisture Condition
(AMC) determined by the total rainfall in the 5-day
period preceding the storm.
Three AMC groups have been established with the
boundaries between groups dependent upon the time of
year as shown in Table 1.
The seasonal difference in the AMC groupings is an
attempt to account for the greater evapotranspiration
between storms during the growing season.
The different infiltration characteristics of soils are
accounted for by classifying soils into four groups based
123
-------
HYDROLOGY: SOLUTION OF RUNOFF EQUATION Q=(Clnil)2
P+0.8S
P= 0 to 12 inches
Q = 0 to 8 inches
Rainfall (P)
RUNOFF (Q)
With P* in: S2ln+F;
. L and F.P-I0-Q|;
• VTT i i • : • i : : • • I • • 11
.'Curves on this sheet are for
case Ia« 0.2 S, so that
;lmtiah
-abstraction In iTTTT
Infiltration-
curve1•H
4567
RAINFALL (P) IN INCHES
REFERENCE
Mockus, Victor; Estimating direct runoff amounts from storm rainfall:
Central Technical Unit, October 1955
u. a MFAKnmrr or AOUCUWJKE
SOIL OONSEBVATION SEHV1CB
ftmoBnn omnt - wotouoaf UANCB
STANDARD DUG. NO.
ES- 1001
SHEET 1 Of 2
OKIE 6-29-56
REVBED 10-1-64
Figure l.-Solutions of Eq. 3. [From SCS National Engineering Handbook (4)]
-------
Table 1. Seasonal rainfall limits for antecedent
moisture conditions1
AMC group
Total 5-day antecedent rainfall
Dormant season Growing season
I
II
III
inches
<0.5
0.5-1.1
>1.1
inches
. <1.4
1.4-2.1
>2.1
1 From SCS National Engineering Handbook (4).
upon the minimum rate of infiltration obtained for a
bare soil after prolonged wetting. The influences of both
the surface and the profile of a soil are included. The
hydrologic soil groups as defined by SCS soil scientists in
the National Engineering Handbook are:
A. (Low runoff potential). Soils having high infiltration
rates even when thoroughly wetted and consisting
chiefly of deep, well to excessively drained sands or
gravels. These soils have a high rate of water transmis-
sion.
B. Soils having moderate infiltration rates when
thoroughly wetted and consisting chiefly of moderately
deep to deep, moderately well to well drained soils with
moderately fine to moderately coarse textures. These
soils have a moderate rate of water transmission.
C. Soils having slow infiltration rates when thoroughly
wetted and consisting chiefly of soils with a layer that
impedes downward movement of water, or soils with
moderately fine to fine texture. These soils have a slow
rate of water transmission.
D. (High runoff potential). Soils having very slow
infiltration rates when thoroughly wetted and consisting
chiefly of clay soils with a high swelling potential, soils
with a permanent high water table, soils with a claypan
or clay layer at or near the surface, and shallow soils
over nearly impervious material. These soils have a very
slow rate of water transmission.
The SCS has classified over 9,000 soils in the United
States and Puerto Rico according to the above scheme.
A sample from the extensive table in the SCS National
Engineering Handbook is shown in Table 2. Rainfall-
runoff data from small watersheds or infiltrometer plots
were used to make the classifications where such data
were available, but most are based on the judgement of
'il scientists and correlators who used physical prop-
erties of the soils in making the assignments.
The interaction of hydrologic soil group (soil) and
land use and treatment (cover) is accounted for by
assigning a runoff curve number for average soil moisture
condition (AMC II) to important soil cover complexes
for the fallow period and the growing season. Rainfall-
runoff data for single soil cover complex watersheds and
plots were analyzed to provide a basis for making these
assignments. Average runoff curve numbers for several
soil-cover complexes are shown in Table 3. Average
runoff curve numbers (AMC II) are for the average soil
moisture conditions. AMC I has the lowest runoff
potential. AMC III has the highest runoff potential.
Under this condition the watershed is practically satu-
rated from antecedent rains. Appropriate curve numbers
for AMC I and III based upon the curve number for
AMC II are shown in Table 4.
Curve numbers for a "good hydrologic condition"
were used in the potential direct runoff simulations.
"Hydrologic condition" refers to the runoff potential of
a particular cropping practice. A row crop in good
hydrologic condition will have higher infiltration rates
and, consequently, less direct runoff than the same crop
in poor hydrologic condition. Good hydrologic condi-
tion seemed an appropriate description of corn under
modern management practices.
Seasonal variation not accounted for by the seasonal
dependency of the AMC classes is included by varying
the average moisture condition curve number according
to the stages of growth of a particular crop. For the
simulations reported here, with straight row corn as the
index crop, the average (AMC II) curve number was set
equal to that for fallow for the period from March 1
until the average emergence date for corn. Emergence
dates were assumed to be 2 weeks after the average
planting date reported by the USDA (5). During the
growing season, AMC 41 curve numbers for each day
were calculated by the following equation:
where
CNj = the curve number for the ith day for AMC
II.
F = fallow curve number.
Cj = crop coefficient for the ith day. Cj < 1.
Cave = average crop coefficient for the growing
season.
CNave= average growing season curve number for
AMC II.
The crop coefficients Cj are defined as the ratio of
the crop evapotranspiration to potential evapotranspira-
tion for a given day when soil water is not limiting. Crop
125
-------
Table 2.—Soil names and hydrologic classifications1 (Sample)
AA8ERG
AASTAD
ABAC
AdAjO
ABBOTT
AdBOTTSTUtfN
A8CAL
AifcGG
ABEL*
ABELL
AdtROctN
ABtS
ABUcNc
AalNbTON
ABIguA
ABO
ABOR
ABRA
ABRAHAM
AUSAnOKfct
A6SCOTA
Att&HER
ABSTcO
ACACIO
ACADEMY
ACAD1A
ACANA
ACASCO
A i. c IT UN AS
ACEL
ACKER
ACKMtN
ACMt
ACO
ACULITA
ACUHA
ACOVt
ACRtt
ACKELANc
Aw TON
A;UFF
ACWCIRTH
ACY
ADA
AOAJK
ADAMS
AOAMSUN
AOAMSTJMN
AdAMSVILLc
AOATUN
A3AVEN
ADOIELDU
ADD I SON
AODV
AUE
AOcL
AOELAIDfe
ADELANTU
AOEL INU
ADELPH1A
ADfcNA
AOCER
ADILIS
AOIRJNDACK
ADIV
ADJUNTAS
AOKINS
ADLtR
AOOLPH
ADRIAN
AENEAS
AETNA
AFTON
AGAi*
AGASSIZ
AGATe
A GAM AM
AiifcNCY
AutR
AGNcR
AuNEM
AGNUS
AGUA
A GU AD ILL A
AGUA OULtfc
AGUA FRIA
AGUALT
AGUEUA
AGU1L1TA
AGUIRRE
A GUST IN
AHATONE
C
B
0
C
D
C
0
B
B
B
0
0
^
0
C
B/C
D
C
B
C
B
0
u
C
c
0
b
0
B
0
B
B
C
B
B
C
C
c
c
B
B
B
C
a
0
A
B
C
0
D
C
D
C
A
A
D
B
a
c
c
D
A
B
C
B
C
D
A/0
B
B
0
B
0
0
B
C
u
b
d/C
a
B
A
C
B
B
8
B
U
B
0
NJTES A
AHL
AHLSTRON
AHMEtK
AHOLT
AHTANUM
AHMAHNbt
AIBtlNITQ
AIKEN
AIKMAN
AILEY
AINAKEA
AIRMONT
AIROTSA
AIRPORT
AITS
A JO
AKAKA
AKASKA
AKkLA
ALADDIN
ALAfc
ALAELOA
ALAGA
ALAKAI
ALA MA
ALAMANCE
ALAMO
ALAMOSA
ALAPAHA
ALAPAI
ALBAN
ALBANC
ALBANY
ALBATON
ALBEt
AL8EMARLE
ALBERWILLE
ALBIA
ALBION
ALBRIGHTS
ALCALDE
ALCtSTER
ALCOA
ALCONA
ALCOVA
ALOA
ALOAX
ALDEN
ALDER
ALDERDALE
ALDfcRMOOD
ALBINO
ALDMELL
ALEKNAGIK
ALCMEDA
ALEX
ALEXANDRIA
ALEXIS
ALFOKD
ALGANSEb
ALGERITA
ALGIERS
ALGOMA
ALHAMBRA
ALICE
ALICEL
ALICIA
ALIUA
ALIKCHI
ALINE
ALKO
ALLAGASH
ALLARD
ALLEGHENY
ALLEMANOS
ALLEN
ALLENUALE
ALLtNS PARK
ALLENSV1LLE
ALLENTINE
ALLENMOOD
ALLESSIO
ALLEY
ALLIANCE
ALLIGATOR
ALLIS
ALLISON
ALLOUEZ
ALLOMAV
ALMAC
ALMENA
ALMONT
BLANK HYDROLOGIC
TWO SOIL GROUPS SUCH AS
C
C
a
D
c
c
c
8/C
0
B
B
C
B
D
B
C
A
B
C
B
A
B
A
0
B
B
D
C
0
A
B
D
C
D
C
B
C
C
B
C
C
B
B
B
B
C
0
D
B
C
C
C
C
B
C
B
C
B
B
B
B
C/0
B/D
B
A
B
B
B
B
A
0
B
B
B
0
B
C
B
C
D
B
8
C
B
0
0
C
C
B
C
D
SOIL
ALMY
ALOHA
ALONSO
ALOVAA
ALPENA
ALPHA
ALPON
ALPOMA
ALPS
ALSfcA
ALSPAJGH
ALSTAD
ALSTON
ALTAMONT
ALTAVISTA
ALTO'ORF
ALTMAR
ALTO
ALTOGA
ALTON
ALTUS
ALTVAN
ALUM
ALUSA
ALVIN
ALVIRA
ALVISO
ALVOR
AHAOUR
AMAGON
AMALU
AMANA
AMARGOSA
AMARILLO
AMASA
AMBERSON
AMBOY
AMBRAM
ANEDEE
AMELIA
AMEN1A
AMERICUS
AMES
AMESHA
AMHERST
AMITY
AMMON
A MOLE
AMOR
AMOS
AMSDEN
AMSTERDAM
AMTOFT
AMY
ANACAPA
ANAHUAC
ANAMITE
ANAPRA
ANASAZ1
ANATONE
ANAVERDE
ANAWALT
ANCHO
ANCHORAGE
ANCHOR BAY
ANCHOR POINT
ANCLOTE
ANCO
ANDERLY
ANDERS
ANDERSON
ANDES
ANOORINIA
AND OVER
ANDREEN
ANDREESON
ANDRES
ANDRE* S
A NED
ANETH
ANGELICA
ANGELINA
ANGELO
ANGIE
ANGLE
ANGLEN
ANGOLA
ANGOSTURA
ANHALT
ANIAK
ANITA
ANKENY
GROUP INDICATES
B
C
B
C
B
C
B
B
C
B
C
B
B
D
C
D
B
C
C
B
B
B
8
D
B
C
D
C
0
D
D
B
D
B
B
C
C
A
B
B
A
C
B
C
C
B
C
B
C
B
B
0
D
B
0
D
8
B
D
8
D
B
A
D
D
D
C
C
C
B
C
C
0
B
C
8
C
0
A
D
B/D
C
C
A
B
C
B
0
0
0
A
ANLAUF
ANNABRLA
ANNANOALE
ANNISTON
ANOKA
ANONES
ANSARI
ANSEL
ANSELM3
ANSON
ANTELOPE SPRINGS
ANTERO
ANT FLAT
ANTHO
ANTHONY
ANT I GO
ANTILON
ANTIOCH
ANTLER
ANTOINE
ANTROBUS
ANTY
ANVIK
ANWAY
ANZA
ANZIANO
APACHE
APAKUIE
APISHAPA
APISON
APOPKA
APPIAN
APPLEGATE
APPLE TON
APPLING
APRON
APT
APTAKIS1C
ARABV
ARAOA
ARANSAS
ARAP I EN
ARAVE
ARAVETON
ARBELA
ARBONE
ARBOR
ARBUCKLE
ARCATA
ARCH
ARCHABAL
ARCHER
ARCH IN
ARCO
ARCOLA
ARD
AROEN
ARDENVOIR
ARCH LA
AREDALE
ARENA
ARENALES
AR.ENOTSVILLE
ARENOSA
ARENZVILLE
ARGONAUT
ARGUELLO
ARGVLE
ARIEL
ARIZO
ARKABUTLA
ARKPORT
ARLAND
A RLE
ARLING
ARLINGTON
AR10VAL
ARMAGH
ARMIJO
ARMINGTON
ARMO
ARMOUR
ARMSTER
ARMSTRONG
ARMUCHEE
ARNEGARO
ARNHART
ARNHE IM
ARNO
ARNOLD
ARNOT
ARNY
C
B
C
B
A
C
D
B
A
B
C
C
C
B
B
B
B
0
C
C
B
B
B
B
B
C
D
A
C
B
A
C
C
C
B
B
C
B
C
0
c
a
B
c
B
B
B
B
B
B
C
C
B
C
C
B
8
C
B
C
A
B
A
B
D
B
B
C
A
C
B
B
B
D
C
C
D
D
D
8
B
C
D
D
B
C
C
D
B
C/0
A
THE SOIL GROUP' HAS NOT BEEN
AROOSTOOK
AROSA
ARP
AURINGTON
AR(tIT3LA
ARROLIME
A!UON
ARROW
ARROMSMITH
ARROYO SECO
ARTA
ARTOIS
ARVADA
A«VANA
ARVESON
ARVILLA
ARZELL
ASA
ASBURV
AS;ALON
ASCHOFF
ASHBV
ASHCROFT
ASHDALE
ASHE
ASHKUN
ASHLAR
ASHLEY
ASH SPRINGS
ASH TON
ASXUE
ASNUELOT
ASHuaoo
ASKEW
ASO
AS3TIN
ASPEN
ASPERHONT
ASSINNIBOINE
ASSUMPTION
ASTATULA
ASTOR
ASTORIA
ATASCkDERO
ATASC3SA
ATCO
ATENCIO
ATEPIC
ATHELHOLD
ATHENA
ATHENS
ATHERLY
ATMERTON
ATHNAR
ATHOL
ATKINSON
ATLAS
ATLEt
ATHDRE
AT3KA
ATON
ATRVPA
ATSION
ATTERBERRV
ATTEM4N
ATTICA
ATTLEBORO
ATMATER
ATMELL
ATHOOD
AUBBEENAUBBEE
AUBERRV
AUBURN
AUBURNDALE
AUDI AN
AU GRES
AUGSBURG
AUGUSTA
AULO
AURA
AURORA
AUSTIN
AUSTKELL
AUXVASSE
AUZQUI
AVA
AVALANCHE
AVALON
AVERY
AVON
AVONBURG
AVONDALE
DETERMINED
C
C
B
0
C
0
B
B
B
C
C
D
C
D
B
C
B
B
B
B
C
B
B
B
C
B
A
C
B
B
C
C
C
C
C
B
8
8
B
A
A/D
B
C
D
B
B
D
B
B
B
B
B/D
C
B
B
D
C
B/D
C
B
C
C
B
A
B
B
C/0
B
B
B
C/0
0
B
C
B
C
D
B
C
C
D
D
B
C
B
B
B
C
D
£
B/C INDICATES THE OR AI NEO/UNDRAI N ED SITUATION
1 From SCS National Engineering Handbook
126
-------
Table 3.-Runoff curve numbers for hydrologic soil-cover complexes1
(Antecedent moisture condition II, and I = 0.2 S)
Cover
Land use Treatment or practice
Fallow Straight row
Row crops "
a
Contoured
"
" and terraced
Small grain Straight row
n
Contoured
n
" and terraced
n n n
Close-seeded legumes2 Straight row
or rotation meadow "
Contoured
n
" and terraced
n H n
Pasture or range
Contoured
"
"
Meadow
Woods
Farmsteads
Roads (dirt)3
(hard surface)
' From SCS National Engineering Handbook (4).
, Close-drilled or broadcast.
hydrologic sou group
Hydrologic condition
A
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Fair
Good
Poor
Fair
Good
Good
Poor
Fair
Good
....
77
72
67
70
65
66
62
65
63
63
61
61
59
66
58
64
55
63
51
68
49
39
47
25
6
30
45
36
25
59
72
74
B
86
81
78
79
75
74
71
76
75
74
73
72
70
77
72
75
69
73
67
79
69
61
67
59
35
58
66
60
55
74
82
84
C
91
88
85
84
82
80
78
84
83
82
81
79
78
85
81
83
78
80
76
86
79
74
81
75
70
71
77
73
70
82
87
90
D
94
91
89
88
86
82
81
88
87
85
84
82
81
89
85
85
83
83
80
89
84
80
88
83
79
78
83
79
77
86
89
92
Including right-of-way.
127
-------
Table 4.-Curve numbers (CN) and constants for the case Ia = 0.2S1
CN for
condi- CN.for
... _ conditions
"°n i HI
100
99
98
97
96
95
94
93
92
91
90
89
88
87
86
85
84
83
82
81
80
79
78
77
76
75
74
73
72
71
70
69
68
67
66
65
64
63
62
61
100
97
94
91
89
87
85
83
81
80
78
76
75
73
72
70
68
67
66
64
63
62
60
59
58
57
55
54
53
52
51
50
48
47
46
45
44
43
42
41
100
100
99
99
99
98
98
98
97
97
96
96
95
95
94
94
93
93
92
92
91
91
90
89
89
88
88
87
86
86
85
84
84
83
82
82
81
80
79
78
s a
values
(inches)
0
.101
.204
.309
.417
.526
.638
.753
.870
.989
1.11
1.24
1.36
1.49
1.63
1.76
1.90
2.05
2.20
2.34
2.50
2.66
2.82
2.99
3.16
3.33
3.51
3.70
3.89
4.08
4.28
4.49
4.70
4.92
5.15
5.38
5.62
5.87
6.13
6.39
Curve2
starts
where
P =
(inches)
0
.02
.04
.06
.08
.11
.13
.15
.17
.20
.22
.25
.27
.30
.33
.35
.38
.41
.44
.47
.50
.53
.56
.60
.63
.67
.70
.74
.78
.82
.86
.90
.94
.98
.03
.08
.12
.17
.23
.28
CNfor
condi-
tion
II
60
59
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
25
20
15
10
5
0
CNfor
conditions
1 III
40
39
38
37
36
35
34
33
32
31
31
30
29
28
27
26
25
25
24
23
22
21
21
20
19
18
18
17
16
16
15
12
9
6
4
2
0
78
77
76
75
75
74
73
72
71
70
70
69
68
67
66
65
64
63
62
61
60
59
58
57
56
55
54
53
52
51
50
43
37
30
22
i3
0
s 2
values
(inches)
6.67
6.95
7.24
7.54
7.86
8.18
8.52
8.87
9.23
9.61
10.0
10.4
10.8
11.3
11.7
12.2
12.7
13.2
13.8
14.4
15.0
15.6
16.3
17.0
17.8
18.6
19.4
20.3
21.2
22.2
23.3
30.0
40.0
56.7
90.0
190.0
infinity
Curve2
starts
where
P =
(inches)
1.33
1.39
1.45
1.51
.57
.64
.70
.77
.85
1.92
2.00
2.08
2.16
2.26
2.34
2.44
2.54
2.64
2.76
2.88
3.00
3.12
3.26
3.40
3.56
3.72
3.88
4.06
4.24
4.44
4.66
6.00
8.00
11.34
18.00
38.00
infinity
1 From SCS National Engineering Handbook (4).
2 For CN in Column 1.
128
-------
coefficient curves for corn were obtained by fitting a
Fourier Series to a curve presented by Kincaid and
Heermann (2).
Curve numbers for antecedent moisture conditions I
and HI were obtained from Table 4.
At harvesting date or when CNj = CNave, whichever
came first, the curve number was set equal to CNave and
remained a constant until the next March 1. The SCS
recommends that the after-harvest curve number be set
equal to the average growing season curve number if 1/3
of the soil surface is exposed. This simulation represents
a situation where residues are left on the field after
harvest.
Any precipitation that occurred when the mean air
temperature was less than O°C was assumed to be snow
and was accumulated as snow storage until the temper-
ature went above 0°C. Snowmelt was calculated by
using a degree-day factor:
S = KT (6)
where
S = snowmelt in inches
K = degree-day snowmelt factor (inches/day/0C)
T = mean daily temperature, °C
A degree-day snowmelt factor K = 0.18 in/day/°C
(0.10 in/day/0 F) was used in all calculations. This is
approximately the mid-range of the values quoted by
Linsley, Kohler and Paulhus, (0.06 -0.15in/day/°F)(5).
The snowmelt calculated in this manner was used to
estimate the antecedent moisture condition and the SCS
curve number procedure was used to estimate snowmelt
runoff. The SCS National Engineering Handbook does
not recommend the use of curve numbers in estimating
runoff from snowmelt, because there is no way to
account for frozen ground. SCS considers the entire
snowmelt as computed by equation (5) to be runoff,
which is good practice when one is concerned with
floods. Because this study was not concerned with
floods, it was deemed more appropriate to use the curve
number procedure to estimate snowmelt runoff despite
its limitations. Obviously, the snowmelt runoff as
calculated may have significant errors.
129
-------
SIMULATION PROCEDURE
Data
The daily precipitation data and temperature data
required for the simulations were obtained on magnetic
tape from the National Climatic Center, Environmental
Data Service, NOAA, U. S. Dept. of Commerce, at
Asheville, N. C. The data set obtained is termed Day
Deck 345. The normal period of record was from
January 1948, through December 1973. A year begin-
ning on March 1 was used in all simulations. The stations
used are listed in Table 5.
Simulations were performed only for stations east of
the Rocky Mountains for the following reasons:
1. This report is intended to cover only nonirrigated
cropland and much of the cropland in the West is
irrigated.
2. Rainfall gradients tend to be very steep in the
West because of orograpnic effects. Therefore, interpola-
tion between widely separated meteorologic stations
would be misleading.
Computer Program
The program SCSRO (Soil Conservation Service
Runoff) was written in FORTRAN IV. A generalized
flow chart is shown in Fig. 2.
Assigning Hydrologic Soil Groups to
Land Resource Areas (LRAs)
Land Resource Areas (LRAs) are shown in the map in
Fig. 2, Vol. I and are discussed in Section 3.1, Vol. I.
Although Land Resource Areas are defined as geographic
areas characterized by a particular pattern of soil type,
topography, climate, water resources, land use and type
of farming (1) they are large enough that each of these
factors varies significantly within the area. Therefore, it
is impossible to characterize an entire LRA by a single
soil series. In many cases, however, the major soil series
listed for a LRA have similar hydrologic characteristics
in that they fall into one hydrologic soil group. Where
there is a wide range of hydrologic characteristics within
a LRA the hydrologic soil group of the predominant
agricultural soil was used.
The simulation results shown in Vol. I and in this
Appendix should not be considered representative of the
entire LRA. However, they are representative of the
predominant agricultural soils of the LRA, subject, of
Table S.-Meteorological records used in simulations
Location
Wichita, KA
Columbia, MO
Dodge City, K A
Kansas City, MO
Springfield, MO
Chicago, IL
Cleveland, OH
Columbus, OH
Lansing, MI
Sault Ste. Marie, MI
Green Bay, WI
Fargo, ND
LaCrosse.WI
Des Moines, IA
Grand Island, NB
Huron, SD
Omaha, NB
Sioux Falls, SD
Bismark, ND
Williston, ND
Scottsbluff, NB
Rapid City, SD
Cairo, IL
Indianapolis, IN
Lexington, KY
Springfield, IL
Savannah, GA
Miami, FL
Houston, TX
Brownsville, TX
Raleigh/Durham, NC
New Castle/Wilmington,
DE
Charleston, SC
Columbia, SC
Jacksonville, FL
Memphis, TN
Mobile, AL
Lake Charles, LA
Dallas, TX
Little Rock, AR
Oklahoma City, OK
Buffalo, NY
Newark, NJ
Boston, MA
Portland, ME
Syracuse, NY
Wilkes-Barre/Scranton,
PA
El Paso, TX
Amarillo, TX
Cape Hatteras, NC
Tallahassee, FL
Pittsburgh, PA
Period of
record
48-67
48-67
48-67
43-67
43-67
43-67
48-67
48-67
49-53,60-69
47-66
50-69
48-67
48-67
46-70
50-70
43-67
48-67
43-67
48-68
35-62
48-67
49-68
Missing
years
none
"
"
'*
"
"
H
"
"
II
'"
"
"
It
"
"
"
"
62
48,49,50
none
a
30-67 48-53,56-62
48-67
48-67
43-67
51-71
48-68
48-68
48-68
48-68
48-68
48-68
48-68
48-68
48-68
48-69
48-58
48-68
48-68
48-67
48-69
48-68
48-68
48-68
48-68
49-53,56-71
48-68
48-68
57-69
48-68
48-68
none
ii
"
52
49
49
49
49
49
49
49
49
49
49,62
49
49
49
none
49,65
49
49
49
49
50
49
49
58
49
49
Total
years
20
20
20
25
25
25
20
20
15
20
20
20
20
25
21
25
20
25
20
25
20
20
25
20
20
25
20
20
20
20
20
20
20
20
20
20
20
10
20
20
20
20
20
20
20
20
20
20
20
12
20
20
130
-------
Read
Field
Character 1st ic s
CALL SUBROUTINE
CN
CALL SUBROUTINE
TPREAD
CO
oc
<
LLJ
LL.
O
LU
CO
LU
CC
53
LU
a
LU
CL.
Subroutine CN computes
an average curve number for
each day of the year.
Subroutine TPREAD reads in
.1 year of daily precipita-
tion and temperature data.
CALL SNOW
CALL AMCF
Subroutine SNOW determines
which daily values of precipi-
tation are snow, accumulates
it and converts it to snowmelt
on the appropriate days.
Subroutine AMCF computes the
• antecedent precipitation
index for each day.
CALL SRO
Subroutine SRO computes daily
surface runoff using the SCS
curve number procedure.
Calculate rainfall
and runoff statistics
for 14-day periods,
seasons , and year.
I
PRINT
Figure 2.-Program SCSRO flowchart
131
-------
course, to the limitations of the SCS runoff estimation
procedure. Predominant agricultural soil series in each
LRA were obtained from Austin (1) and hydrologic
classifications for the soil series were obtained from
Table 7.1 of the SCS National Engineering Handbook
(4). A list of the LRA's and the assigned hydrologic soil
groups is presented in Table 6. These assignments were
reviewed and modified by personnel of the Technical
Service Centers of SCS and their assistance is gratefully
acknowledged.
Table 6.—Hydrologic soil group and available water holding
capacities for predominant agricultural soils in land
resource areas
Table 6.-Hydrologic soil group and available water holding
capacities for predominant agricultural soils in land
resource areas-Continued
Lantl r>~m;—« \,,,Ar~\s>^.^ Available water
rpcr,nr™> Dominant hydrologic
resource soil croup capacity in
area sou B1™? 4.ft. root zone
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43-51
52
53
54
55
56
57
Mountains
B
Mountains
Soil Information Lacking
Forest
Mountains
B
B
C
C
B
Mountains
B
B
D
D
D
D
D
Mountains
D
Mountains
Soil Information Lacking
" " "
D
D
D
Desert
a
"
Irrigated Desert
B
Mountain
B
B
C
B
Mountain
"
Soil Information Lacking
n a n
B
Mountains
B
B
B
B
D
B
(inches)
8
8
8
8
8
8
8
8
6
6
6
6
6
6
6
6
6
8
8
8
6
8
8
8
8
*6
8
4
8
Land
resource
area
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
Dominant hydrologic
soil group
B
C
D
B
Mountains
D
B
A
B
B
B
B
C
B
B
B
B
B
D
C
C
A
C
D
C
D
B
D
D
D
Forest
Forest
B
A
Forest
Forest
Forest
B
A
B
B
D
B
B
B
B
C
B
B
Available water
capacity in
4-ft. root zone
(inches)
*6
8
4
8
4
8
2
8
8
8
8
6
8
8
8
8
8
*4
8
8
4
8
6
8
6
8
6
6
6
8
2
8
2
4
8
6
8
8
8
8
8
8
8
132
-------
Table 6--Hydrologic soil group and available water holding
capacities for predominant agricultural soils in land
resource areas-Continued
Land
resource
area
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
Dominant hydrologic
soil group
B
B
C
C
C
D
D
D
B
C
Mountains
D
Mountains
C
C
B
C
C
Mountains
C
Mountains
B
B
Mountains
D
D
B
C
D
B
A
B
C
C
C
D
Mountains
A
B
C
B
C
C
D
Swamp
D
C
A
B
Swamp
Available water
capacity in
4-ft. root zone
(inches)
8
8
*6
*6
8
*4
*4
4
8
*4
6
*4
8
8
8
8
8
8
8
6
6
8
8
6
8
4
4-
*4
8
8
6
4
8
8
8
8
8
6
6
6
4
8
* Available water-holding capacity reduced because root
zone is shallower than 4 feet.
133
-------
SIMULATION RESULTS
Program SCSRO output for each 14-day period for
the n years of record may include:
1. A listing of rainfall amounts ordered by magni-
tude.
2. A listing of simulated runoff ordered by magni-
tude.
3. The mean and standard deviation of rainfall and
runoff events.
The statistical summary for the n-year simulation
included:
1. A table showing the number of runoff events for
each 14-day period for each year.
2. The probability that there would be no runoff
events in any year for each 14-day period.
3. The mean annual simulated runoff.
4. The mean growing season simulated runoff.
Maps of the mean annual and seasonal simulated
runoff (potential direct runoff) are shown for each
hydrologic soil group in Figs. 3 through 10. Because of
the relatively small area of soils classified in hydrologic
group A, simulations with this group were not per-
formed for all rainfall stations. The growing season was
taken as the time interval between emergence and
harvest and varied with location. These maps can be used
to supplement the information presented in Figs. 3 and
4 in Vol. I.
Top few rainfall stations were used in this analysis to
depict climatic and orographic influences in the Appa-
lachian Mountains; therefore, care must be used in
interpreting the maps in these regions.
134
-------
Figure 3.-Mean annual potential direct runoff in inches. Straight-row corn in good hydrologic condition-Hydrologic Soil Group A.
U»
-------
OJ
0
Figure 4.-Mean annual potential direct runoff in inches. Straight-row corn in good hydrologic condition -Hydrologic Soil Group B.
-------
Figure 5.-Mean annual potential direct runoff in inches. Straight-row corn in good hydrologic condition-Hydrologic Soil Group C.
-------
OJ
oo
Figure 6.-Mean annual potential direct runoff in inches. Straight-row corn in good hydrologic condition -Hydrologic Soil Group D.
-------
94
Figure 7.—Mean growing season potential direct runoff in inches. Straight-tow corn in good hydrologic condition-Hydrologic Soil Group A.
UJ
-------
Figure 8.-Mean growing season potential direct runoff in inches. Straight-row corn in good hydrologic condition-Hydrologic Soil Group B.
-------
Figure 9.-Mean growing season potential direct runoff in inches. Straight-row corn in good hydrologic condition-Hydrologic Soil Group C.
-------
Figure lO.-Mean growing season potential direct runoff in inches. Straight-row corn in good hydrologjc condition-Hydrologic Soil Group D.
-------
CONSISTENCY CHECK AND DISCUSSION OF ERRORS
The accuracy of the simulated runoff amounts was
checked by comparing average annual potential direct
runoff with measured average annual runoff from several
small, single-crop watersheds in straight-row crops (pri-
marily corn and cotton) (7, 8). The watersheds used in
this comparison and a brief data summary are presented
in Table 7. The following linear regression equation was
obtained:
Qs = 1.365 +0.578 Qf
(7)
where
Qs = simulated average annual direct runoff.
Q0 = observed average annual runoff.
The coefficient of variation (r2) was 0.616. A scatter
diagram of computed versus observed average annual
runoff is shown in Fig. 11. The vertical lines emanating
from the plotted points indicate the range in simulated
runoff that would occur if the soils were assumed to
belong to adjacent hydrologic soil groups. This indicates
Table 7.-Surface runoff consistency check
Watershed
College Park, MDW-3
Americus, GA W-l
Lafayette, IN W-4
" W-5
" W-8
" W-10
" W-l 2
" W-l 3
" W-l 5
Clarinda, IA W-V
" W-W
" W-Y
Coshocton.OH W-l 15
" W-l 10
" W-l 18
" W-192
" W-l 06
Guthrie, OK W-2
Garland TX W-III
Spur,TX W-2
Riesel, TX Y-7
Hastings, NB 3-H
Oxford. MS WC-1
WC-3
C hickasha, OK C-l
Area
acres
6.06'
17.9
2.01
2.87
1.96
2.06
3.37
3.02
3.59
3.25
1.97
3.25
1.61
1.27
1.96
7.59
1.56
3.21
10.4
9.39
40.0
3.95
3.88
1.61
17.8
Crop
Soybeans
Sweet corn
Corn
Cotton
Corn
Soybeans
Corn
Soybeans
Corn
Soybeans
Corn
Soybeans
Corn
Soybeans
Corn
Soybeans
Corn
Soybeans
Corn
Corn
Corn
Corn
Corn
Corn
Corn
Corn
Cotton
Cotton
Corn
Cotton
Row crops
Corn
Corn
Corn
Corn
Hydrologic
soil group
B
B
B
84% B
16% C
B
66% B
34% C
89% B
11% C
80% B
20% C
B
90% B
10% D
90% B
10% D
72% B
28% D
C
C
C
C
C
B
D
50% B
50% C
D
B
C
C
81% C
Mean annual
Observed
2.47
1.27
2.26
4.14
4.89
5.88
3.98
3.93
4.33
1.41
3.37
1.06
2.85
2.41
1.97
3.13
3.23
7.67
11.67
2.70
7.00
4.85
14.99
13.77
1.24
runoff (in.)
Simulated
2.60
7.17
2.60
3.05
2.60
3.57
2.91
3.17
2.60
2.32
2.32
2.89
3.05
3.05
3.05
3.05
3.05
2.30
8.30
2.05
8.75
1.65
11.00
11.00
3.32
19% B
143
-------
15
V)
•o
0>
O
5
4
,_ v^
I'*
1
0
Qs= l.365-t-0.578 Q0
r2=0.6!6
s.e. = l.44"
I Std. Dev
of Mean
8. o
C CO
o
o ^
0»
u. O E
^ o O
1_
(O 1-)
a.
3
Regression
Line
0 I 2 3 4 5 6 7 8 9 10 II 12 13 14
Q0 , Observed Mean Annual Direct Surface Runoff, inches
Figure 11.-Simulated and observed mean annual direct surface runoff.
15
144
-------
the inherent limitation in the SCS method caused by
lumping all soils into four distinct groups. Horizontal
bars emanating from plotted points indicate estimated
standard errors of the mean of observed data. This
illustrates the problem of short, fragmented records. The
SCS method tended to underestimate runoff of more
than 3 inches.
Meaningful comparisons are difficult because the
observations were made over such a long time. Agricul-
tural practices have changed drastically so the data are
not stationary. For example, hybrid corn and higher
fertility levels have led to rapid canopy establishment
and more residues after harvest. The simulated condition
after harvest—approximately 67% cover by residues—
probably is not consistent with the practices on water-
sheds where the data were obtained. One would antici-
pate that the simulated runoff would be less than the
observed in this case. Data from some of the watersheds
listed in Table 7 were undoubtedly used in developing
the SCS curve number procedure so this is not an
independent test of its predictive capabilities.
Although the relationship between simulated and
observed direct runoff shown in Fig. 11 is not as good as
one would wish, it must be compared with the available
alternatives before one can judge its usefulness. One
alternate that has been suggested is to use the map of
surface-water runoff prepared by the U.S. Geological
Survey (6) as an indicator of potential loss by direct
runoff. To test this method, consider the following
regression relationship between the average annual run-
off from the USGS map for the locations in Table 7 and
the observed average annual direct runoff:
QG = 6.74 + 0.503 Q0 ; r2 = 0.138
(8)
where QQ is average annual surface-water runoff from
the USGS map.
The simulated results obviously are superior to those
obtained from the runoff map as indicators of potential
direct runoff.
One would anticipate that the sum of the simulated
average annual direct runoff and the average annual deep
percolation estimated by the procedures described in
Appendix B of this volume should be rather well
correlated with the average annual streamflow from the
USGS maps.
The following regression equation was obtained be-
tween simulated direct runoff plus percolation and
runoff (streamflow) from the USGS map for 45 of the
52 meteorological stations used:
(Qs + Qp) = 0-409 + 0.979 QG ; r2 = 0
1.884
(9)
where Q_ is the simulated average annual deep percola-
tion. Seven stations in the karst area of Florida and in
the coastal area of the Southeastern United States were
omitted because anomalies on the USGS map indicate
that much of the groundwater runoff flows directly into
the ocean. These crude checks indicate that the simula-
tions provide reasonable estimates of annual direct
runoff and percolation.
Records of runoff from continuous straight-row corn
were not readily available for periods shorter than one
month so it was not possible to check the accuracy of
the time distribution of simulated potential direct runoff
within the year. However, 20 years of runoff data were
available for a small watershed in meadow at Coshocton,
Ohio. Direct runoff for 14-day periods simulated by the
SCS procedure was compared with observed data.
Simulated and observed mean runoff per event, mean
number of runoff events per period, and mean runoff
amount per period are shown in Fig. 12. The standard
deviation of the observed runoff per period is indicated
by a vertical line for each period.
The simulated runoff per period is within one
standard deviation of the observed runoff for 14 of the
26 periods. Assuming that the mean value is normally
distributed for each period and that the 26 periods are
independent trials, the null hypothesis cannot be re-
jected at the 10 percent level.
Sample distribution functions of runoff amount per
event for two periods are shown in Fig. 13. A Kolmo-
gorov-Smirnov test comparing the distribution functions
indicates that the null hypothesis cannot be rejected at
the 10 percent level for both periods.
Although it is impossible to make strong inferences
on the basis of the limited tests performed, the SCS
method appears adequate for arriving at a first estima-
tion of direct surface runoff.
145
-------
1.2 -
tn
c
0
H-
o 0.35
c
ir 0.30
0.25
0.20
0.15
0.10
0.05
" I 3 5 7 9 II 13 15 17 19 21 23 25 27 29
MAR MAR APR MAY JUN JUL AUG SEPT OCT NOV DEC JAN JAN
I 28 25 23 20 18 15 12 10 7 5 2 3O
Period
Figure 12.-Comparison of simulated and observed runoff records by 14-day periods. Coshocton, Ohio meadow.
Mean Runoff per Period
146
-------
c
o
o
c
=j
U-
c.
o
"5
.a
CO
O
CD
3
£
3
O
Coshocton Ohio, Meadow
o Observed
x Simulated
Period I, Mar 1-14
Period 11, July 18- Aug 2
I I I 1 I I
II I
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
x Runoff , inches
2.0
Figure 13.-Sample distribution functions foi simulated and observed records. Ooshocton, Ohio meadow.
-------
LITERATURE CITED
1. Austin, M. E. 1965. Land resource regions and major
land resource areas of the United States. SCS USDA
Agriculture Handbook No. 296,82 p.
2. Kincaid, D. C. and Heermann, D. F. 1974. Scheduling
irrigations using a programmable calculator. ARS
NC-12.5Sp.
3. Unsley, R. K., Jr., Kohler, M. A. and Paulhus, J. L.
H. 1958. Hydrology for Engineers. McGraw-Hill, N.
Y.
4. Soil Conservation Service, USDA. 1971. SCS Na-
tional Engineering Handbook, Section 4, Hydrology.
U.S. Govt. Printing Off., Washington, D.C.
5. USDA Statistical Reporting Service. 1972. Usual
planting and harvesting dates. Agr. Handbook No.
283.
6. U. S. Geological Survey. 1970. The National Atlas of
the United States of America. 417 p.
7. U. S. Dept. of Agriculture. Hydrologic Data for
Agricultural Watersheds in the United States. USDA
Misc. Publications No. 945 (1956-59); 994(60-61);
1070(1962); 1164(1963); 1194(1964); 1216(1965);
1226(1966); 1262(1967).
8. U.S. Dept. of Agriculture. 1957. Monthly precipita-
tion and runoff for small agricultural watersheds in the
United States. USDA ARS, Soil and Water Conservation
Research Branch, Washington, D.C. (June).
148
-------
APPENDIX B
SIMULATION OF POTENTIAL PERCOLATION AND NITRATE LEACHING
INTRODUCTION
Potential percolation was defined in Section 3.4,
Volume I, as the annual amount of water that would
percolate below the root zone in a field of straight-row
corn.
The relationship between potential percolation and
other hydrologic variables can be expressed by the
vertical water balance equation for a column of soil as
P =
(1)
where P is the precipitation, Qs is the direct runoff, E is
the evapotranspiration, Qp is the percolation from the
bottom of the root zone and AS is the change in soil
water storage in the root zone during the time under
consideration. In Eq. (7), it is assumed that imported
water, lateral porous media flow, and change in surface
detention storage are negligible.
Soluble agricultural chemicals that are not strongly
adsorbed by the soil may be carried below the root zone
by percolating water. After the percolating water has
reached the ground water table, it will move laterally
and eventually reappear in a stream, lake or possibly the
ocean. Some ground water may also flow into the root
zone in seepage areas and be transpired by vegetation.
Because precipitation can be considered as a stochas-
tic process, all of the other variables in Eq. (7) are
stochastic in nature. A simulation approach is necessary
to estimate each of the other terms because of the rather
complex relationships between them. To carry out the
simulation we need a soil water model, a direct runoff
model and an evapotranspiration model. The SCS
procedure as described in Appendix A was used as the
direct runoff model. The soil-water model and the
evapotranspiration model are described in this Appen-
dix.
THE SOIL-WATER MODEL
The soil-water model described in this section utilizes
the approximation that soil-water moves readily under
gravitational forces when its water content is above field
capacity. It is further assumed that water does not move
downward when the water content is below field
capacity and that when the water content reaches the
wilting point it is no longer available to plants.
The structure of the three-compartment soil-water
model used is shown in Fig. 1. The water content of
compartment i is designated as Sj(t) and the maximum
capacity of the i*h compartment is Kj, which corre-
sponds to field capacity. Compartments 1 and 2 repre-
sent the active root zone and compartment 3 represents
'he water storage below the current root zone and above
the maximum depth of rooting, d3- The depth of the
surface layer is dj and d2(t) is the time varying depth of
the root zone. The extraction of water from different
depth zones varies with stage of crop development, so
the capacities of compartments 2 and 3 vary with time
as the crop canopy expands but their sum is a constant.
The maximum capacities of the compartments can be
interpreted as the available water-holding capacity per
unit area for the depths dj, d2(t)-di and d3-d2(t).
Input to the system is Xj(t), which is the difference
between daily rainfall plus snowmelt and direct runoff.
System output is Yj(t), ¥2(1) and ¥3(1) where
Y! (t) = daily evaporation from the soil surface
Y2 (t) = daily transpiration
Y3 (t) = dafly seepage below the maximum root zone
149
-------
The evaporation and transpiration model will be
described in the next section. The rules for movement of
water between compartments are as follows:
y,2(t+l) = S,(t) + X1(t+l)-Y1(t+l)-Y4(t+l)-K1; otherwise
(2)
where yj2(t+l) is the flow from compartment 1 to com- transpiration and flow to compartment 2. Flow from
partment 2. compartment 1 to compartment 2 exists only if the
Equation (2) states that the water content of com- available waterholding capacity of compartment 1 is
partment 1 in period t+1 is equal to the water content in exceeded. No provision is made for upward flow in this
period t plus the infiltration on day t+1 less evaporation, model, although such flow is physically possible.
y12(t+l)-Ys(t+l)-y23(t+l)
t) + y12(t+l)-Y5(t+l)
-------
Transpiration Evaporation
Y2(t) Y,(t)
Infiltration
X,(t)
1 •
*
Y4(t)
Y5(t)
(
i
1 S,(t)
\
2 Q
2
y,2(t)
t)
K,
K,(t)
d2(t)
r
3
S3(t)
K,(t)
Y3(t)
Percolation
Figure 1.-Soil-water model schematic.
crude concept of "field capacity." The approximations
and assumptions included in the model probably would
not lead to serious errors in deep, well-drained soils.
Substantial errors may occur where water tables are
shallow or in soils with shallow, relatively impermeable
layers.
THE EVAPOTRANSPIRATION MODEL
The evapotranspiration (ET) model is based on the
i -quently used assumption that ET will take place at
the "potential" rate if the soil has adequate water and a
complete crop canopy or if the surface is wet. Actual
evapotranspiration rates will be less than potential as the
soil dries.
The model can be described by the following equa-
tions:
Evaporation:
Y1(t)=[l-C(t)]KpP(t) SiW
(9)
151
-------
O)
0)
o
c
0)
en
ai
LJ
d
o
c
o
O
i
o
Q
to
0)
o
I
Time
Figure 2.—Seasonal variation of root zone.
where C(t) is a time varying crop coefficient related to
the portion of the soil surface covered by vegetation, Kp
is a coefficient to convert pan evaporation to potential
ET, P(t) is the pan evaporation in inches per day and
Sj(t) and KI have been previously defined.
Total transpiration is given by:
Y2(t) = C(t)KpP(t)f . (10)
where f is the ratio of actual to potential evapotranspira-
tion and depends on the water content of compartments
1 and 2 as shown in Fig. 3.
The transpiration loss from compartment 1 is:
Y4(t)-Y2(t)
s,(t)
(ID
and the transpiration loss from compartment 2 is:
Y5(t) = Y2(t)-Y4(t) . (12)
An examination of Eqs. (9) through (12) reveals that
for bare fallow conditions, [C (t) = 0], only evaporation
occurs, i.e., Y2(t) = 0. For full canopy conditions C(t) =
l,Yj(t) = 0 and all loss is through transpiration. The
form of this evapotranspiration model is identical to that
used by Hanson (5).
NITRATE LEACHING MODEL
The purpose of this model is to gain a quantitative
insight into what percentage of nitrogen applied as
ammonium in the fall or spring would move below the
root zone before the roots had reached their full
extension in the following crop year. Any nitrogen
present in the profile at the time of fertilization is
ignored as are denitrification losses. It is also assumed
that there is no nutrient uptake by weeds or winter
152
-------
0
0.6
.0
S2(t)]/[K,(t)-hK2(t)]
Figure 3.—Relationship between actual and potential evapotranspiration and root zone water content.
cover crops. Water flow through the soil during the
dormant period is assumed to be piston flow. The upper
compartment with capacity Kt was retained in the soil
model. At the time ammonium was applied in the fall,
the water in compartments 2 and 3 was assumed to be
uniformly distributed in the depth d3 - d!. Any time the
capacity of compartment 1 is exceeded, the volume of
water yu(t) moves as a piston flow, displacing some of
the soil water ahead of it as shown in Fig. 4.
The depth increment of the i*h output from compart-
ment 1 to the lower zones is given by the expression
where E is an exclusion factor to account for the
fraction of the soil water not containing nitrate and 0pis
the volumetric field capacity (21,22), E as used in these
calculations is the ratio of the volume of soil water
excluding nitrate to the total soil volume. Each time
Xi2(t) is nonzero, a new increment is introduced,
pushing all those ahead of it down a distance AZ.
A number of studies (9, 12, 28) have shown that
anions, like nitrate, move with the wetting front through
dry soil. The depth to the peak concentration can often
be estimated by the ratio of infiltration to field capacity.
Laboratory leaching studies (21, 22) with water contents
near saturation show that these anions are excluded
from some of the soil water and thus move faster than
the total soil water does. Opinions differ as to the
importance of this factor for field conditions. Some
consider it important (23), others do not (1). We have
chosen to incorporate an exclusion factor because a large
part of the nitrate movement being modeled will occur
at moisture contents above field capacity and it will also
reflect some of the channelized flow in clay soils (8).
When ammonium fertilizer is applied, it is assumed to
be concentrated at depth, di. Ammonium is converted
153
-------
E(d3-d1)
0
e
Figure 4.—Piston flow soil water model.
to the nitrate according to the following temperature-
dependent relation after a lag of 5 days:
N(t+l)=N(t) + k(T)A(t)
(14)
where N(t) is the nitrogen in the nitrate form on day (t),
A(t) is the nitrogen in the ammonium form on day (t),
and k(T) is a temperature-dependent rate function given
by the following equations (4, 19):
= 0.0032T-.012 ; 10°C
-------
o
O
March I
Q)
O
Q
O
0)
Q
-
o
I
a
LL!
O
i /
K
o
0)
Wo
Hfcu
Figure 5.-Nitrate leaching model operation in time.
-------
zone extension, all Wj from NR-1 that are below the root
zone are summed and divided by the ammonium
nitrogen present at spring fertilizations to give the
percentage loss. Note that for the case portrayed in Fig.
5, no nitrate would be lost from spring fertilization
because the first increment of recharge after spring
fertilization was not below the root zone. In the piston
flow model for water movement and nitrate leaching it is
assumed that the nitrate is completely mixed within
each element AZ, but that there is no diffusion or
dispersion allowing nitrate exchange between elements.
SIMULATION PROCEDURE
Data
Daily precipitation and temperature data used in the
simulations were obtained on magnetic tape from the
National Climatic Center, Environmental Data Service,
NOAA, U.S. Dept. of Commerce, Asheville, N. C. The
data set obtained is termed Day Deck 345. The normal
period of record was from January 1948 through
December 1973. A year beginning on March 1 was used
in all simulations. The stations used are listed in Table 5,
Appendix A. Simulations were limited to stations east of
the Rockies because of the steep rainfall gradients in the
West and because most of the situations where leaching
may be a problem are in irrigated areas and thus are
excluded from this report.
Mean monthly pan evaporation data were obtained
for the stations used or for nearby stations from a U. S.
Weather Bureau publication (25). Fourier series were fit
to these monthly values and were converted to mean
daily values. A single harmonic explained more than 97%
of the variance for most of the stations. The mean daily
evaporation and the amplitude and phase angle of the
first harmonic were plotted on maps and isolines were
drawn. These parameters were then estimated by inter-
polation from the maps for stations where evaporation
pan data were not available. The pan coefficient, Kp,
was obtained from the map presented by Kohler,
Nordenson and Baker (10).
Estimation of Parameters
The index crop considered was straight-row corn.
Planting and harvesting dates for each locality were
obtained from maps prepared by the USDA Statistical
Reporting Service (24). Plowing and spring fertilization
were arbitrarily assumed to have been done 14 days
before planting. The fall fertilization date was the day
the 5-day moving average temperature went below 50° F
(10° C) or December 15, whichever occurred first. Corn
was assumed to reach full canopy 80 days after planting.
Root zone depths, available soil water capacities, field
capacities, and exclusion fractions most commonly used
in the simulations are shown in Table 1.
Table l.-Most commonly used soil-water model parameters
Parameter
Hydro logic soil group
B
D
d,
d,1
Ki ....
KI +K2 "*" 1^3 . .
0F . ...
E
4 in.
4 ft.
.33 in.
4.0 in.
.123
.04
4 in.
4 ft
.67 in.
8.0 in.
.237
.07
4 in.
4ft.
.67 in.
8.0 in.
.327
.10
4 in.
4 ft
.50 in.
6.0 in.
.345
.15
1 Root zone depths were reduced for shallow soils.
It was assumed that the available water-holding
capacity of a soil was the difference between the water
content at 0.3 bars and 15 bars tension. (Approximate
field capacity and wilting point). Typical textures of
soils in each hydrologjc soil group were then selected
and the total available water content was rounded to the
nearest inch. The storage capacities were assigned to land
resource areas on the basis of the characteristics of the
predominant agricultural soils. The assignments were
reviewed and corrected where necessary by soil scientists
of the SCS Technical Service Centers. The assignments
by land resource areas are shown in Table 6, Appendix
A. The values of the .exclusion fraction, E , were
estimated for the assumed water-holding capacities from
published data on 15 soils (21).
Computer Program
The subroutine. ETRANS (Evapotranspiration) was
written in FORTRAN IV. It is called from program
SCSRO described in Appendix A. A generalized flow
chart is shown in Fig. 6.
156
-------
Read
Field
Char ac t er i s t ±c s
Read evaporation
and crop coeffi.
series parameters
REPEAT FOR DESIRED NUMBER OF YEARS
Call CN
Call TPREAD
*
Call SNOW
i
Call AMCF
\
Call SRO
*
Call STEMP
For explanation of these sub
> routines, see Fig. 2,
Appendix A.
Call ETRANS
Subroutine STEMP computes a
5^day moving average temperature
for use in the nitrification
calculations.
Subroutine ETRANS computes evapo-
ration, transpiration, percolation
nitrification and nitrate move-
ment on a daily basis for 1 year.
Compute rainfall,
runoff, percolation
evapotranspiration
and N loss
statistics.
Write
Output
Summary
Figure 6.-Generalized flow chart Program SCSRO with percolation and nitrate leaching option.
157
-------
SIMULATION RESULTS
The program output for each 14-day period included: 5. Distribution functions of loss of fall-applied nitro-
1. An ordered listing of daily rainfall. gen, spring-applied nitrogen and leaching.
2. An ordered listing of simulated runoff. 6. Mean annual percolation.
3. An ordered listing of daily percolation. 7. Mean annual evapotranspiration.
4. The mean and standard deviation of each of the Maps of the mean annual percolation, fall-applied N
above. loss and spring-applied N loss for each of four available
The statistical summary for the n-year simulation soil water-holding capacities are shown in Figs. 7
included: through 18. Because of the limited area in which soils of
1. A table showing the number of runoff events for -Hydrologic Group A are predominant, simulations were
each 14-day period for each year. not performed for all stations. Therefore, isolines could
2. The probability that there would be no runoff in not be drawn in the east central portion of the United
any year for each 14-day period. States. The mean annual precipitation for the period of
3. The mean annual simulated runoff. record used in the simulation is shown in Fig. 19.
4. The mean growing season simulated runoff.
158
-------
Figure 7.-Mean annual percolation below a 4-foot root zone in inches. Hydrologic Soil Group A. Four inches available water-holding capacity. Straight-row com.
ct
\o
-------
12.0
14.0
O
Figure 8.-Mean annual percolation below a 4-foot root zone in inches. Hydrologic Soil Group B. Eight inches available water-holding capacity. Straight-row corn.
-------
Figure 9.—Mean annual percolation below a 4-foot root zone in inches. Hydrologic Soil Group Q Eight inches available water-holding capacity. Straight-row corn.
-------
to
Figure lO.-Mean annual percolation below a 4-foot root zone in inches. Hydro-logic Soil Group D. Six inches available water-holding capacity. Straight-row corn.
-------
50
0,0
Figure 11.-Mean percentage loss of fall-applied nitrogen. Straight-row com in good hydrologic condition. Hydrologic Soil Group A. Available water-holding capacity - 4 inches.
Ui
-------
Figure 12.-Mean percentage loss of fall-applied nitrogen. Straight-row corn in good hydrologic condition. Hydrologic Soil Group B. Available water-holding capacity - 8 inches.
-------
Figure 13.- Mean percentage loss of fall-applied nitrogen. Straight-row com in good hydrologic condition. Hydiologic Soil Group C Available water-holding capacity - 8 inches.
o\
-------
30
Figure 14.-Mean percentage loss of fall-applied nitrogen. Straight-row coin in good hydrologjc condition. Hydtologic Soil Group D. Available water-holding capacity - 6 inches.
-------
Figure 15.-Mean percentage loss of spring-applied nitrogen. Straight-tow com in good hydrologic condition. Hydrologic Soil Group A. Available water-holding capacity - 4 inches.
-------
Rgure 16.-Mean percentage loss of spring-applied nitrogen. Straight-row corn in good hydrologic condition. Hydrologic Soil Group B. Available water-holding capacity - 8 inches.
-------
v_
--)
^
I 0*0
I
0*0
0-0
x—-<
)
T
0«0
(-__.
!o«o
o-o !
i
0.0
^
0*0,
0«0-'—t—
^x
\
V
/ T """""T-J
/ ' i
\J—— / i±!
{ / T— i
/ / io«o
-/ I i L-
\ /
D.O°^___.V^(
0'° °Co c
°'° 'i °'° ?
-- — -,{, ,'
i 1
O«O 1 \
1 1
^/-^
£^~ /'
)io -,-- A
">V
0-0
/0«0
0«0.
I i °
6jfiuT*i.r26*
•^
'/
Figure 17.-Mean percentage loss of spring-applied nitrogen. Straight-row corn in good hydrologic condition. Hydrologjc Soil Group C Available water-holding capacity - 8 inches.
-------
•o yv-oj
°n>^
°'°0-0
/
0»0
Figure 18.-Mean percentage loss of spring-applied nitrogen. Straight-row corn in good hydrologic condition. Hydrologic Soil Group D. Available water-holding capacity - 6 inches.
-------
Figure 19.-Mean annual precipitation for period of record used in simulations.
-------
DISCUSSION
Simulated mean annual percolation and nitrate leach-
ing cannot be compared with observations because such
data are not generally available. However, the excellent
correlation between the sum of the simulated potential
direct runoff and percolation and the surface runoff
from USGS maps as presented in Appendix A suggests
that the simulated results are reasonable.
Isolated bits of data are available for additional
checks. Minshall (14) in a study of 25 years of runoff
data (1940-1964) on the Platte River in southwestern
Wisconsin found that the mean annual base flow was 5.7
inches. The mean annual potential percolation for
Hydrologic Group B is somewhat greater than 5 inches
(Fig. 8).
Hanway and Laflen (6) reported 3-year averages of
1.11 and 4.64 inches of subsurface drainage for tile
outlet terraces in Creston, Iowa and Charles City, Iowa,
respectively. From Fig. 8, the mean annual potential
percolation for these -sites is about 2 inches and 3 inches,
respectively. One would anticipate greater percolation
losses from tile outlet terraces than from straight row
corn but the period of record is too short to make valid
comparisons. Again it appears that the simulated per-
colation is reasonable.
Saxton, Spomer and Kramer (20) reported on meas-
urements of base flow for small watersheds with contour
corn near Treynor, Iowa. Six-year average base flow
from two watersheds was 2.52 and 2.47 inches. From
Fig. 8, simulated deep percolation in this area is about 2
inches. Rainfall during the 6-year period was above
average for 5 of the 6 years.
Simulated mean annual percolation is compared with
lysimeter data in Table 2. Only one set of data is for
corn (Coshocton, Ohio), and the simulated percolation is
very close to the observed. The other data sets agree
favorably with the simulated percolation when the crop
canopy differences are considered. The shallow lysime-
ters at Windsor, Conn, probably account for much of the
difference between observed and computed percolation.
We were unable to find any data showing the
percentage of fall-applied nitrogen lost during the winter
and spring.
Although the comparisons between simulated per-
colation and data cannot be considered as conclusive,
they do suggest that the simulations provide a reasonable
ordering of Land Resource Areas with respect to
percolation losses. The absolute amounts also appear to
be realistic.
The only way a technique such as this can be judged
is against readily available alternatives. The leaching
hazard map prepared by Nelson and Uhland (18) is
shown in Fig. 20. The material presented in this
Appendix and in Vol. I clearly presents a more detailed
picture of percolation and of the relative hazards of
nitrate leaching from fall fertilization.
Care should be used in interpreting the maps of
potential percolation and nitrate leaching where it is
known that the model assumptions are seriously in error.
For example, in the Southern United States the assump-
tion of no nutrient uptake or transpiration during the
winter would be inaccurate if winter cover crops are
planted. In this case, both the percolation and the
nitrate loss would be overestimated.
Table 2.-Comparison of simulated mean annual percolation with lysimetei data
Location Citation Soil Sf&Sp
Ithaca, N.Y. Bizzell (2) Pet oskey gritty A
sandy loam B
Average percolation
Crop Simulated
Observed (com)
inches
Vegetables 17.76
17.76
inches
13
11
Geneva, N.Y. Collison et al. (3) Ontario, Dunkirk
Knoxville, Tenn. Mooers et al. (15) Cumberland
Windsor, Conn.
Windsor, Conn.
Coshocton, Ohio
Morgan &
Jacobson(16)
Morgan, et al.
(11)
Harrold &
Dreibelbis (7)
Merrimac,
20" depth
Merrimac,
30" depth
Muskingum
B
A
A
C
Barley-clover 12.7 11
rotation
Fallow 22.5 15
Tobacco 13.63 20
Tobacco with 12.45 20
winter cover
Corn years in 7.43 7
CWMM rotation
172
-------
Figure 20.—Relation of degree of leaching to geographic area. Leaching ranges from nil in Area I to very high in Area IV.
From Nelson and Uhland (18).
173
-------
LITERATURE CITED
1. Appelt, H., Holtzclaw, K., and Pratt, P. F. 1975.
Effect of anion exclusion on movement of chloride
through soils. Soil Sci. Soc. Amer. Proc. 39:264-267.
2. Bizzell, J. A. 1943. Lysimeter experiments-VI: The
effects of cropping and fertilization on the losses of
nitrogen from the soil. Cornell Univ. Agr. Exp. Sta.
Memo. No. 252: 1-24.
3. Collison, R. C., Beattie, H. C., and Harlan, J. D.
1933. Lysimeter investigations: III. Mineral and
water relations and final nitrogen balance in legume
and nonlegume crop rotations for a period of 16
years. New York (Geneva) Agr. Exp. Sta. Tech. Bull.
212:1-81.
4. Frederick, L. R. 1956. Formation of nitrate from
ammonium nitrogen in soils: 2. Effect of population
of nitrifiers. Soil Sci. 83:481-485.
5. Hanson, C. L. 1973. Model for predicting evapotrans-
piration from native rangelands in the Northern Great
Plains. Ph.D. dissertation, Utah State University,
Logan.
6. Hanway, J. J. and Laflen, J. M. 1974. Plant nutrient
losses from tile-outlet terraces. Jour. Environ. Quality
3(4): 351-356.
7. Harrold, L. L. and Dreibelbis, F. R. 1958. Evaluation
of agricultural hydrology by monolith lysimeters
1944-55. U. S. Dept. Agr. Tech. Bui. 1179,166 p.
8. Kissel, D. E., Ritchie, J. T. and Burnett, E. 1973.
Chloride movement in undisturbed swelling clay soils.
Soil Sci. Soc. Amer. Proc. 37:21-24.
9. Kolenbrander, G. J. 1970. Calculation of parameters
for the evaluation of the leaching of salts under field
conditions, illustrated by nitrate. Plant and Soil
32:439-453.
10. Kohler, M. A., Nordenson, T. J., and Baker, D. R.
1959. Evaporation maps for the United States. U. S.
Weather Bureau Tech. Paper No. 37, 13 p.
11. Kohler, M. A., Nordenson, T. J., and Baker, D. R.
1963. Rainfall-runoff models. General Assembly of
Int. Assoc. of Scientific Hydrology Surface Waters,
Berkeley, Calif.: 479-491.
12. Levin, I. 1964. Movement of added nitrates through
soil columns and undisturbed soil profiles. Proc. 8th
Intern. Congress of Soil Science, Bucharest,
Romania.
13. Makkink, G. F. and van Heemst, H. D. J. 1966.
Water balance and water bookkeeping regions.
Verslagen en mededlingen van de Commissie voor
Hydrologisch Orderzock T. N. O. No. 12, 9-112.
Cited by F. E. Schulze, Chapter 1, Rainfall and
rainfall excess in recent trends in hydrograph
synthesis. Proc. of Tech. Meeting 21. Versl. Medel
Comm. Hydrol. Onderz. T.N.O. 12. The Hague. 103
P-
14. Minshall, N. E. 1967. Precipitation and base flow
variability. Proc. International Assoc. of Scientific
Hydrology, Bern: 137-145.
15. Mooers, C. A. Maclntire, W. H., and Young, J. B.
1927. The recovery of soil nitrogen under various
conditions as measured by lysimeters of different
depths. Tennessee Agr. Exp. Sta. Bui. 138: 1-30.
16. Morgan, M. F. and Jacobson, H. G. M. 1942. Soil
and crop interrelations of various nitrogenous fertil-
izers: Windsor lysimeter Series B. Connecticut Agr.
Exp. Sta. Bui. 458: 269-328.
17. Morgan, M. F., Jacobson, H. G. M., and Lecompte,
S. B. Jr. 1942. Drainage water losses from a sandy
sofl as affected by cropping and cover crops:
Windsor lysimeter Series C. Connecticut Agr. Exp.
Sta. Bui. 466: 729-759.
18. Nelson, L. B. and Uhland, R. E. 1955. Factors that
influence loss of fall applied fertilizers and their
probable importance in different sections of the
United States. Soil Sci. Soc. Amer. Proc.
19(4):492-496.
174
-------
19. Sabey, B. R., Bartholomew, W. V., Shaw, R., and
Pesek, J. 1956. Influence of temperature on nitrifi-
cation in soils. Soil Sci. Soc. Amer. Proc.
20:357-360.
20. Saxton, K. E., Spomer, R. G., and Kramer, L. A.
1971. Hydrology and erosion of loessial watersheds.
Proc. ASCE Jour. Hydr. Div. 97 (HY11): 1835-1852.
21. Smith, S. J. 1972. Relative rate of chloride move-
ment in leaching of surface soils. Soil Sci.
114(4):259-263.
22. Smith, S. J., and Davis, R. J. 1974. Relative
movement of bromide and nitrate through soils.
Jour. Environ. Qual. 3(2): 152-155.
23. Thomas, G. W. 1970. Soil and climatic factors
which affect nutrient mobility. In Engelstad, O. P.
ed., Nutrient mobility in soils: accumulation and
losses, Soil Sci. Soc. Amer., Inc., Madison, Wise.
24. USDA Statistical Reporting Service. 1972. Usual
planting and harvesting dates. Agr. Handbook No.
283.
25. U. S. Weather Bureau. Climatography of the United
States. Climatic summary of the United States-
Supplement for 1951 through 1960.
26. Van Wijk, W. F., Larson, W. E. and Burrows, W. C.
1959. Soil temperature and the early growth of corn
from mulched and unmulched soil. Soil Sci. Soc.
Amer. Proc. 23(6): 428-434.
27. Wiser, E. H. and van Schilfgaarde, J. 1964. Predic-
tion of irrigation water requirements using a water
balance. Proc. VI^1 International Congress of Agr.
Engr., Lausanne: 121-129.
28. Yaalon, D. H. 1965. Downward movement and
distribution of anions in soil profiles with limited
wetting. In Experimental Pedology. William Cloves
and Sons, Ltd. London: 157-164.
175
-------
APPENDIX C
ECONOMIC ANALYSIS METHODOLOGY
The following discussion details the application of the
method presented in Section 5, Volume I, to evaluate
the decision-maker's optimal choice in the example given
in Section 6.2, Volume I. This example is clearly
site-specific and cannot hope to show the full gamut of
variables which may potentially be of significance in other
situations, such as irrigation, hired labor, other crop
rotations and the like. The decision-maker will have to
adjust the budgeting system shown here to his particular
situation.
Several important assumptions were made. For cer-
tain computations, the size of the farm became a
parameter, and the assumption was made that the
example farm had 250 tillable acres. Other assumptions
are detailed in the following tables. In addition, it was
assumed that none of the macro effects described in
Section 5.2 of Volume I influence any of the decision
variables noted here. This assumption implied that any
machinery which may become obsolete due to a change
in cropping practices would be sold at a cost close to its
depreciated value and that, consequently, there was
no cost of disposing of obsolete equipment to be added
to the actual machinery costs.
The determination of the relevant production alterna-
tives resulted in five potential choices, namely, (1)
continuous corn no-till planted in 70 percent residue
cover, contoured, (2) a corn-corn-corn-wheat-meadow
rotation with moldboard plowing on the first year corn
and no-till planting on the second and third year corn,
contoured, (3) continuous corn with rotary strip tillage,
terraced, (4) continuous corn with chisel planting,
terraced, and (5) a corn-soybean rotation with no-till
planting, terraced. The costs and returns for a sixth
production method (i.e. continuous corn, residue left,
with moldboard plowing, straight row) are shown for
comparison purposes only. This particular production
method does not meet the soil erosion limitation and
can therefore not be considered as an available alterna-
tive.
Three of the five viable alternatives required terrac-
ing, which contributed an additional production cost,
summarized in Table 1. The example assumed a farm
Table 1. Broadbase terrace construction and maintenance costs
Item
Amount
Terrace spacing, feet 120
Slope length, feet 350
Number of terraces per slope ...... 2
Feet terrace/acre 249
Construction cost/foot terrace3, $ 0.60
Construction cost/acre, $ 149.40
Prorated construction costb, $ 13.74
Maintenance cost, foot3. $ 0.00023
Maintenance cost, acre, $ 0.06
Yearly terrace charge/acre, S 13.80
Total yearly terrace charge (250 acres), $. 3.450.00
a Source: Sidney James (ed), Midwest Farm Planning
Manual, 3rd Edition, 1SU Press, Ames, Iowa, 1973, p. 33.
b Assume 20 year life of terrace. Interest at 8 percent.
located on Monona silt loam with more than 3 percent
organic matter, a land slope of 6 percent, and an average
slope length of 350 feet. According to the technical
standards1 for terrace construction, the construction of
level broadbase terraces with a spacing of 120 feet would
be appropriate in this situation. Table 1 shows the
assumptions used in the computation of the cost of
terracing the entire farm.
Each of these production systems requires a specific
set of field operations and implements. Table 2 lists the
implements considered in this study and the computa-
tion of the fixed costs for each machine. The computa-
tion of the depreciation cost used the straight-line
method over the economic life of the implement. Not all
of the implements were used in any particular crop
production activity; Table 3 shows which implements
were used in each production alternative, the total hours
of machinery use, and the total implement cost. These
costs did not include the cost of the tractor (listed
1 U.S. Department of Agriculture, Soil Conservation Service,
Iowa, Technical Standards and Specifications for Conservation
Practices, Section 4A-Cropland, Work Unit Technical Guide,
Code No. 600 and 602, January 1973.
177
-------
00
Table 2. Machinery fixed costs
Machine
Size
Initial3
cost
Salvageb
value
Economic0
life
Yearly
depreciation
Taxes, insurance^
and housing
Interest6
Yearly
fixed cost
Dollars Percent Years Dollars Dollars Dollars Dollars
Stalk shredder 12'flail 2,350 13.7 12 169.00 70.50 101.83 341.33
Moldboard plow 5-16" 2,590 17.7 10 213.16 77.70 113.96 404.82
Chisel plow 15' 1,700 13.7 12 122.26 51.00 73.67 246.93
Disk, tandem 20' 4,385 17.7 10 360.89 131.55 192.94 685.38
Harrow 20' 340 17.7 12 23.32 10.20 14.73 48.25
Sprayer tractor mounted 680 17.7 10 55.96 20.40 29.92 106.28
Banter - conventional 4-38" 1,430 17.7 10 117.69 42.90 62.92 223.51
Rotary strip planter 4-38" 3,675 17.7 10 203.45 110.25 161.70 574.40
No-till plant (fluted coulters) ... 4-38" 4,375 17.7 10 360.06 131.25 192.50 683.81
Wheat drill (with grass seeding
attachments) 12' 2,740 9.7 14 176.73 82.20 117.43 376.36
Cultivator 4-38" 1,470 17.7 10 120.98 44.10 64.68 229.76
Duster 4-row 400 13.7 12 28.77 12.00 17.33 58.10
Combine, self-prop small 70-80 hp 16,100 18.9 10 1,305.71 483.00 708.40 2,497.11
00111 ho"1 2-38" 2,800 18.9 10 227.08 84.00 123.20 434.28
Hatform 13' 2,500 18.9 10 202.75 75.00 110.00 387.75
Hay mowers 7' 960 12.5 i2 70.00 28.80 41.60 140.40
Hay conditioners . . .-. 7' 1,300 12.5 12 94.79 39.00 56.33 190.12
Hay rake side delivery 980 12.5 12 71.46 29.40 42.47 143.33
Hay baler PTO 3,500 21.1 8 345.19 105.00 157.50 607.69
» Source: Background Information for Use with CROP-OPT System. FM 1628, ISU Cooperative Extension Service, Ames, Iowa, November 1974.
0 Source: George E. Ayres, Estimating Used Machinery Costs, A.E. 1078, ISU Cooperative Extension Service, Ames, Iowa, January 1974.
<• Source: Sidney James (ed), Midwest Farm Planning Manual, 3rd Edition, ISU Press, Ames, Iowa, 1973, Table IV-7, p. 129.
Taxes and insurance at 2 percent of initial cost; housing at 1 percent of initial cost. Source: George E. Ayres, Estimating New Machinery Costs, A.E. 1077, ISU Cooperative
Extension Service, Ames, January 1974.
e Assumed at 8 percent per annum.
-------
Table 3. Machinery costs
H(
Implement
)urs per
Acres of
useb
Times
over0
Total Repair cost per
hours 100 hours"
Corn, residue left, spring turn-plow, conventional
Stalk shredder
Moldboard plow
Sprayer
Disk
Harrow
Planter
Cultivator
Combine
Corn head
Total
Com, fall shred stalks, chisel
Stalk shredder
Chisel plow
Sprayer
Harrow
Planter
Cultivator
Combine
Corn head
Total
.18
.36
.21
.10
.10
.22
.21
.63
.63
250
250
250
250
250
250
250
250
250
1
1
1
1
1
1
2
1
1
45.0
90.0
52.5
25.0
25.0
55.0
105.0
157.5
157.5
Dollars
94.00
129.50
34.00
219.25
10.20
114.40
73.50
322.00
56.00
Total repair
cost
Dollars
42.30
116.55
17.85
54.81
2.55
62.92
77.18
507.15
88.20
Yearly fixed
cost
Dollars
341.33
404.82
106.28
685.38
48.25
223.51
229.76
2,497.11
434.28
Total
cost
Dollars
383.63
521.37
124.13
740.19
50.80
286.43
306.94
3,004.26
522.48
5,940.23
plant, 30-40% residue cover
.18
.17
.21
.10
.22
.21
.63
.63
Corn, residue left, strip-till row zones
Stalk shredder
Sprayer
Rotary strip-till planter . .
Cultivator
Total
Corn, fall shred, no-till plant.
Stalk shredder
Sprayer
No-till planter
Duster
Combine
Corn head
Total
.18
.21
.22
.21
.63
.63
50-70%
.18
.21
.22
.21
.63
.63
250
250
250
250
250
250
250
250
1
1
1
1
1
2
1
1
45.0
42.5
52.5
25.0
55.0
105.0
157.5
157.5
94.00
85.00
34.00
10.20
114.40
73.50
322.00
56.00
42.30
36.13
17.85
2.55
62.92
77.18
507.15
88.20
341.33
246.93
106.28
48.25
223.51
229.76
2,497.11
434.28
383.63
283.06
124.13
50.80
286.43
306.94
3,004.26
522.48
4,961.73
,40-50% residue cover
250
250
250
250
250
250
residue cover
250
250
250
250
250
250
Corn-corn-com-wheat-meadow. residue left, no-till
Stalk shredder
Moldboard plow
Sprayer
Disk
Harrow
No-till planter
Wheat drill
Duster
Combine wheat
Corn head
Platform
Hay mower
Hay rake
Hay baler
Total
.18
.36
.21
.10
.10
.22
.25
.21
.63
.30
.63
.30
.31
.31
.30
.63
150
50
150
100
50
150
50
150
150
50
150
50
50
50
50
50
1
1
1
2
1
1
1
1
1
1
1
1
olant 2nd
1
1
1
1
1
1
1
1
1
1
1
1
3
3
3
3
45.0
52.5
55.0
105.0
157.5
157.5
45.0
52.5
55.0
52.5
157.5
157.5
and 3rd corn
27.0
18.0
31.5
10.0
5.0
33.0
12.5
31.5
109.5
94.5
15.0
46.5
46.5
45.0
94.5
94.00
34.00
294.00
73.50
322.00
56.00
94.00
34.00
350.00
8.00
322.00
56.00
94.00
129.50
34.00
219.25
10.20
350.00
219.20
8.00
322.00
56.00
50.00
96.00
52.00
58.80
210.00
42.30
17.85
161.70
77.18
507.15
88.20
42.30
17.85
192.50
4.20
507.15
88.20
25.38
23.31
10.71
21.93
0.51
115.50
27.40
2.52
352.59
52.92
7.50
44.64
24.18
26.46
198.45
341.33
106.28
574.40
229.76
2,497.11
434.28
341.33
106.28
683.81
58.10
2,497.11
434.28
341.33
404.82
106.28
685.38
48.25
683.81
376.36
58.10
2,497.11
434.28
387.75
140.40
190.09
143.33
607.69
383.63
124.13
736.10
306.94
3,004.26
522.48
5,077.54
383.63
124.13
876.31
62.30
3,004.26
522.48
4,973.11
366.71
428.13
116.99
707.31
48.76
799.31
403.76
60.62
2,849.70
487.20
395.25
185.04
214.27
169.79
806.14
8,038.98
179
-------
Table 3. (continued)
u
Implement
Corn-soybeans, no-till plant,
Stalk shredder
Sprayer
No-till planter
Duster
Combine corn
Combine soybeans ....
Corn head
Platform
Total
r«r
fall shred
.18
.21
.22
.21
.63
.30
.63
.30
Acres of
useb
corn stalks
125
250
250
125
125
125
125
125
Times
over0
1
1
1
1
1
1
1
1
Total Repair cost per
hours 100 hound
22.5
52.5
55.0
26.25
116.25
78.75
37.5
Dollars
94.00
34.00
350.00
8.00
322.00
56.00
50.00
Total repair
cost
Dollars
21.15
17,85
192.50
2.10
374,32
44.10
18.75
Yearly fixed Total
cost cost
Dollars
341.33
106.28
683.81
58.10
2,497.11
434.28
387.75
Dollars
362.48
124.13
876.31
60.20
2,871.43
478.38
406.50
5,179.43
Source: Background information for use with CROP-OPT system, FM 1628, ISU Cooperative Extension Service, Ames, Iowa,
November 1974.
° Acres on which implement is used each year.
c Number of trips through field with implement.
d Computed as percentage of list price. Used 2% for combine, platform, corn head and duster; 4% for stalk shredder and hay condi-
tioner; 5% for moldboard plow, chisel plow, cultivator, sprayer, and disk; 6% for hay rake and hay baler; 8% for planters and wheat
drill; 10% for hay mower. Source: George Ayres, Estimating new machinery costs, AE 1077, ISU Cooperative Extension Service,
Ames, Iowa, January 1974.
180
-------
separately in Table 4) or fuel and lubrication (Table 5).
The implement hours per acre (from Table 3) were
aggregated for each production alternative. The total was
augmented by a 10 percent figure for traveling to field,
idling, etc., to result in the total tractor hours figure
listed in Table 4. The depreciation cost assumed a
straight-line depreciation over the economic life of the
tractor.
The fuel costs for the tractor and the combine were
computed as shown in Table 5. These fuel costs were
presented separately from the other machinery and
tractor variable costs for the purpose of emphasizing the
differences in fuel consumption among production
alternatives.
Table 6 shows the computations for the seed costs of
the five production alternatives. The assumption was
made that the no-till alternatives would be subject to a
higher seed mortality rate than the other alternatives,
due to the higher crop residue levels.
Agricultural chemicals were selected on the basis of
the recommended nutrient and pesticide practices. It
was assumed that the nitrogen was applied in NH3 form
and the phosphate and potassium in granular bulk form
(Table 7). The restrictions on optimal timing (N2) are
assumed to be met by fertilizer application just prior to
planting, which in the case of two alternatives (corn
chisel-plant and corn rotary-strip-till) also implies incor-
poration (N8 and 12).
The pesticide costs (Table 8) were estimated on the
basis of pesticide recommendations by the ISU Exten-
sion Service for control of the major pests.2' 3 The
pesticide costs for the several production alternatives
may vary since each rotation requires the use of a unique
mix of pesticides. For example, the herbicide cost for
the no-till alternatives was higher than for conventional
tillage because greater amounts of and more expensive
types of herbicides were assumed to be used with this
alternative. The insecticide cost for the rotation includ-
ing meadow was assumed greater than for continuous
corn alternatives due to the expected incidence of the
first-year corn insect complex. No insecticide cost was
assumed for soybeans, since the acreage of soybeans
ordinarily treated with insecticides was quite small.
Labor costs for the five production alternatives were
computed as shown in Table 9. The labor requirement
per acre was estimated as 130 percent of the tractor
hour requirement to account for overhead labor in
addition to the direct requirements. The labor cost per
hour was assumed equal to the present average wage rate
for Iowa.
Table 10 presents two additional cost components,
namely the corn drying costs and interest charges. It was
assumed that the costs (variable and fixed) of drying
corn amounted to 12 cents/bushel, which is the current
charge for custom drying in Iowa.4 It was assumed that
the out-of-pocket costs involved the use of borrowed
capital. The interest costs on machinery and the tractor
were included in their total costs and are not repeated
here.
The gross revenue for each of the production alterna-
tives was computed as shown in Table 11. The no-till
alternatives were assumed to have a slightly lower yield
than the more conventional tillage alternatives due to
increased production and harvesting complexities.
The final table, Table 12, summarizes all of the
preceding computations and shows the gross revenue
and net return figures for each of the six production
methods. A land cost was included based on an assumed
land value of $974.00 per acre5 and a cash rent of $7.40
per $100 value.6 Since this land charge applied equally
to all six production methods, any error in this land
charge will change only the absolute levels and not the
differences in net returns among the six alternatives.
It appears that the (unavailable) alternative of con-
tinuous corn with conventional moldboard tillage has a
significantly higher net revenue than any of the other
(available) production alternatives. There is only a small
variation in net return of the top three (available)
production alternatives with a major net return drop to
the corn-soybeans alternative. The corn-corn-corn-
wheat-meadow rotation has by far the lowest net return
among these six alternatives, indicating that the savings
in fertilizer cost generated by the nitrogen nutrient
credit from the legume meadow are not sufficient to
offset the increases in other costs.
2Harold J. Stockdale, Insect Pest Control Recommendations
for 1975, IC-328 (Rev.), ISU Cooperative Extension Service,
Ames, Iowa, January 1975.
3Vivan M. Jennings, Weed Control Guide for 1975. Pm 601
(Rev.), ISU Cooperative Extension Service, Ames, Iowa, January
1975.
^Estimated 1975 Iowa Custom Rates, ISU Cooperative
Extension Service, FM 1698, Ames, Iowa, January 1975.
s$974.00 is the November 1, 1974 average price for high
grade farmland in West Central Iowa, reported in William Murray
etal, Land Values Double in 5 years, FM 1681, ISU Cooperative
Extension Service, Ames, Iowa, January 1975.
6 A rent of $7.40 per $100 value is the average cash rental
rate for corn and soybean land reported in E. G. Stoneberg and
Ronald Winterboer, Cash Rental Rates from Iowa Farm Land,
FM 1626 (Rev.), ISU Cooperative Extension Service, Ames,
Iowa, August 1973.
181
-------
Table 4. Tractor costs
Item
Straight-row
C conv.
Contour
C no-t.
CCCWM no-t. C chisel
Terraced
C strip
CB no-t.
Tractor hours per acre3
Total tractor hours''
Tractor initial cost,0 dollars . . .
Economic life, years^
Salvage value, percent6 ....
Yearly depreciation, dollars . . .
Taxes, insurance and housing/
dollars
Average annual interest,^ dollars
Total fixed costs, dollars . . . .
Repair costs,11 dollars
Total tractor costs, dollars (excl.
fuel)
2.21
607.75
18,230.00
11
27.5
1,201.52
546.90
795.49
2,543.91
886.34
3,430.25
1.65
453.75
18,230.00
13
23.5
1,072.52
546.90
785.29
2,404.95
661.75
3,066.46
2.20
605.00
18,230.00
11
27.5
1,201.52
546.90
795.49
2,543.91
882.33
1.92
528.00
18,230.00
12
25.5
1,131.78
546.90
789.97
2,468.65
770.04
1.65
453.75
18,230.00
13
23.5
1,072.76
546.90
785.29
2,404.95
661.75
1.34
368.50
18,230.00
14
21.5
1,022.18
546.90
781.29
2,350.36
537.42
3,426.24 3,238.69
3,066.70 2,887.79
a Assume tractor is required for harvest hauling, in amount equivalent to time requirements for combine. Add 0.2 hours per acre
for application of fertilizer with rented implements.
b Increased by 10 percent for idling, travel to field, etc.
ciOOPTOhpdiesel.
d From Sidney James (ed.) Midwest Farm Planning Manual. 3rd Edition, ISU Press, Ames, Iowa, 1973, Table IV-7, p. 129.
e From George E. Ayres, Estimating Used Machinery Costs, A.E. 1078, ISU Cooperative Extension Service, Ames, Iowa, January
1974.
f Taxes and insurance at 2 percent and housing at 1 percent of initial cost. Source: George E. Ayres, Estimating New Machinery
Costs. AE 1077, ISU Cooperative Extension Service, Ames, Iowa, January 1974.
S Assume 8 percent interest.
h 0.8 percent of list price per 100 hours of use. Source: Ibid.
Table 5. Fuel costs
Item
Total tractor hours
Fuel cost per tractor hour,3 dollars .
Tractor fuel cost, dollars
Total combine hours
Fuel cost per combine hour, dollars
Combine fuel cost, dollars
Total fuel cost dollars
Straight-row
C conv.
607.75
2.071
1,258.65
157.50
1.106
174.20
1,432.85
Contour
C no-t.
453.75
2.071
939.72
157.50
1.106
174.20
1,113.92
CCCWM no-t.
605.00
2.071
1,252.96
109.5
1.106
121.11
1,374.07
C chisel
528.00
2.071
1,093.49
157.50
1.106
174.20
1,267.69
Terraced
C strip
453.75
2.071
939.72
157.50
1.106
174.20
1,113.92
CB no-t.
368.50
2.071
763.16
116.25
1.106
128.57
891.73
a Fuel consumption gallons per hour = 0.044xPTO hp. Lubrication costs at 15 percent of fuel cost. Source: Sidney James (ed.)
Midwest Farm Planning Manual. 3rd Edition, ISU Press, Ames, Iowa 1973, p. 125. Assume diesel fuel at $0.40/gal.
b Gasoline consumption = 2.35 gal./acre. Source: George E. Ayres, Fuel Required for Field Operations, AE 1079, ISU Cooperative
Extension Service, Ames, Iowa, March 1974. Lubrication costs at 15 percent of fuel costs. Assume gasoline at $0.40/gal.
182
-------
Table 6. Seed costs
Item
Straight-row
C conv.
Corn
Seeding rate (seeds/acre)
Assumed mortality, % .
Final stand
Seed amount8, bu. ...
Seed cost", dollars . . .
Wheat
Seed amount, bu.
Seed costc, dollars
Hay
Seed amount**, Ibs
Seed '-oste, dollars
Soybeans
Seed amount', bu. . . .
Seed costS, dollars . . .
Seed cost per acieh, dollars
Total seed cost, dollars . . .
23,000
10
20,700
0.274
6.85
6.85
1.712.50
Contour
Terraced
Cno-t. CCCWMno-t, C chisel
C strip
26,000
20
20,800
0.310
7.75
7.75
1,937.50
26.000
20
20,000
OJ10
7.75
1-5
11.25
15
24.45
11,79
2,947.50
244)00
13
0.286
7,15
24,000
13
20380
0.286
7.15
7.15
1,787.50
7.15
1,787.50
CB no-t.
26,000
20
20,800
0.310
7.75
1
9.50
8.62
2,155.00
a Based on 84,000 seeds per bushel.
b Assuming price of $25.00 per bushel (Iowa price, U.S. Department of AgikiJtme. AgrinOtvnlPrices. Apr. 15,1974.
c Price of S7.50 per bushel (U.S. Department of Agriculture, Agricultural Prim, Sept. 15.1974),
d Source: Sidney James (ed.), Midwest Farm Planning Manual. 3rd Edition, ISU Vuas, Aae*. Iowa, 1975, p. 18.
e Price of $163.00 per 100 Ibs. (U.S. Department of Agriculture, AgriadaniPrices, Sept. 15,1974).
' Source: Sidney James, op. cit., p. 20.
i Price of $9.50 per bushel (U.S. Department of Agriculture, Agricultural Prices. Sept. 15,1974).
" Average seed cost.
183
-------
Table?. Fertilizer costs
Item
Corn
N*
P20<
K20
Wheat
N
P20S
K20
Soybeans
P,0S
K20
Average amount0
N
p,0s
K20
Cost of fertilizer per acred, dollars. .
Total cost of fertilizer, dollars ....
Rental of application equipment6,
dollars
Total fertilizer cost, dollars
Straight-row
C conv.
Contour
C no-t. CCCWM no-t.
C chisel
Terraced
C strip
CB no-t.
170
30
20
170
30
20
30.90
7,725.00
187.50
7,912.50
170
30
20
170
30
20
30.90
7,725.00
187.50
7,912.50
113° 170
30 30
20 20
60
25
30
80 170
23 30
18 20
17.06 30.90
4,265.00 7,725.00
125.00 187.50
4,390.00 7,912.50
170
30
20
170
30
20
30.90
7,725.00
187.50
7,912.50
150b
30
20
30
30
75
30
25
18.38
4,595.00
125.00
4,720.00
a Fertilizer recommendations based on: Regis D. Voss, General Guide to Fertilizer Recommendations in Iowa, AG-65 (rev.), ISU
Cooperative Extension Service, Ames, Iowa, August 1973.
b Includes fertilizer credit from meadow or soybeans.
c Amount per year of rotation if other than continuous com.
d Assume N as NH3 and P20S as 46 percent P20S. Prices per pound are $0.136 for N, $0.206 for P20S, and $0.080 for K20.
Source: Iowa price in U.S. Department of Agriculture, Agricultural Prices, September 15,1974.
e Assume 50^/acre for NHs knife and 25^/acre for 4-ton bulk spreader.
TableS. Pesticide costs
Item
Straight-row
C conv.
Contour
Terraced
C no-t. CCCWM no-t.
C chisel
C strip
CBno-t.
Com
Herbicide, dollars
Insecticide, dollars
Acres
Total cost, dollars
11,00 16.00 16.00 11.00 11.00 18.00
7.00 7.00 9.00 7.00 7.00 7.00
250 250 150 250 250 125
4,500.00 5,750.00 3,750.00 4,500.00 4,500.00 3,125.00
Soybeans
Herbicide, dollars
Acres
Total cost, dollars
Total pesticide cost, dollars ....
11.00
125
1,375.00
4,500.00 5,750.00 3,750.00 4,500.00 4,500.00 4,500.00
184
-------
Table 9. Labor costs
Item
Straight-row
Contour
Terraced
C conv.
C no-t.
CCCWM no-t. C chisel
C strip
CB no-t.
Total direct labor, hours ....
Overhead (30%), hours
Total labor, hours
Cost per hour, dollars
Total labor cost, dollars
765.25
229.58
994.83
2.50
2,487.08
611.25
183.38
794.63
2.50
1,986.58
714.50
214.35
928.85
250
2,322.13
685.50
205.65
891.15
2.50
2,227.88
611 25
183.38
794 53
2 50
1 986 58
484 75
145 43
630 18
2 50
1 575 45
Table 10. Other costs
Item
Corn dry ing
Grain harvested
Cost per bushel
Total cost
Interest (8%) on operating capital
Fertilizer (8 mo.)
Seed (8 mo.)
Pesticide (6 mo.)
Fuel (3 mo.)
Labor (3 mo.)
Total interest
Total other costs
Straight-row
C conv.
27,500
0.12
3,300.00
558.98
91.28 '
180.00
28.66
36.08
895.00
4,195.00
Contour
C no-t.
26,250
0.12
3,150.00
558.98
103.27
230.00
22.28
23.56
938.09
4,088.09
CCCWM no-t.
jju . .
16,500
0.12
1,980.00
342.19
157.10
150.00
28.54
33.18
711.01
2,691.01
C chisel
27,500
0.12
3,300.00
558.98
95.27
180.00
25.35
31.36
890.96
4,190.96
Terraced
C strip
27,500
0.12
3,300.00
558.98
95.27
180.00
22.28
26.98
883.51
4,183.51
CB no-t.
13,125
0.12
1,575.00
340.32
114.86
180.00
18.75
17.71
671.64
2,246.64
18S
-------
Table 11. Revenue
Item
Straight-row
C conv.
Contour
C no-t.
CCCWMno-t. C chisel
Terraced
C strip
CB no-t.
Corn
Expected yield, bu./ac 110.0 105.0
Area cropped, acres 250 250
Total output, bu 27,500 26,250
Expected price, dollars/bu 2.75 2.75
Gross revenue, dollars 75,625.00 72,187.50
Wheat
Expected yield, bu./ac. .
Area cropped, acres
Total output, bu
Expected price, dollars/bu.. .
Gross revenue, dollars
Meadow
Expected yield, tons/ac
Area cropped, acres
Total output, tons
Expected price, dollars/ton. . .
Gross revenue, dollars
Soybeans
Expected yield, bu./ac
Area cropped, acres
Total output, bu
Expected price, dollars/bu
Gross revenue, dollars
Total gross revenue, dollars 75,625.00 72,187.50
105
150
15,750
2.75
43,312.50
45.0
50
2,250.0
4.00
9,000.0
4.0
50
200.0
45.00
9,000.00
110.0
250
27,500
2.75
75,625.00
110.0
250
27,500
2.75
75,625.00
105.0
125
13,125
2.75
36,093.75
61,312.50 75,625.00 75,625.00
40.0
125
5,000.00
6.00
30,000.00
66,093.75
Table 12. Summary
Item
Straight-row
C conv.
Contour
Terraced
C no-t.
CCCWM no-t. C chisel
C strip
CB no-t.
Gross revenue
Costs
Tractor (excl. fuel) . . .
Implements (excl. fuel).
Fuel
Seed
Fertilizer
Pesticides
Labor
Terracing
Other
Land charge (see text) .
Total cost
Net return
75,625.00
3,430.25
•5,940.23
1,432.85
1,712.50
7,912.50
4,500.00
2,487.08
0
4,195.00
18,020.00
49,630,41
25,994.59
72,187.50
3,066.46
4,973.11
1,113.92
1,937.50
7,912.50
5,750.00
1,986.58
0
4,088.09
18,020.00
48,848.16
23,339.34
dollars •
61,312.50 75,625.00
75,625.00 66,093.75
3,426.24
8,038.98
1,374.07
2,947.50
4,390.00
3,750.00
2,322.13
0
2,691.01
18,020.00
46,959.93
14,352.57
3,238.69
4,961.73
1,267.69
1,787.50
7,912.50
4,500.00
2,227.88
3,450.00
4,190.96
18.020.00
51,556.97
24,068.05
3,066.70
5,077.54
1,113.92
1,787.50
7,912.50
4,500.00
1,986.58
3,450.00
4,183.51
18,020.00
51,098.25
24,526.75
2,887.79
5,179.43
891.73
2,155.00
4,720.00
4,500.00
1,575.45
3,450.00
2,246.64
18,020.00
45,626.04
20,467.71
186
-------
As the discussion in Section 5, Volume I, points out,
there are a number of intangible variables not accounted
for in the net return figures. One of these intangibles,
that of scheduling, may be of only minor significance in
this example. The scheduling of alternatives with the
highest net returns does not differ sufficiently to
influence the decision.
A more important consideration is the variability of
yields. The variance of yield under no-till is higher than
for production alternatives utilizing more tillage. This
higher variance for no-till may be partly due to a lack of
familiarity with this method on the part of growers. In
the present example, the no-till production alternatives
were assumed to have a lower yield than the other
production alternatives to account for this potential
yield impact. A farmer who is a risk-averter or who is
utterly unfamiliar with no-till planting may be willing to
accept a lower net return with a higher degree of
certainty if that alternative excludes no-till planting.
One additional consideration related to the cost of
terracing. The present example assumes tacitly that the
full cost of terracing is borne by the farmer. Historically,
society has reimbursed terracing costs through various
government programs so that the farmer usually paid
half the cost or even less. Under any such cost-sharing
program, the relative differences in net revenue will
change. In the present example, a cost-sharing program
with a 50-50 split would give two of the terraced
alternatives a net revenue practically identical to the
(unavailable) conventional tillage continuous corn activ-
ity.
*US. GOVERNMENT PRINTING OFFICE:1977 722-669
187
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-75-026b
2.
4. TITLE AND SUBTITLE
Control of Water Pollution from Cropland: Volume II—
An Overview
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
June 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
E. A. Stewart, D. A. Woolhiser, W. H. Wischmeir,
J. H. Caro, and M. H. Frere
8. PERFORMING ORGANIZATION REPORT NO
ARS-M-5-2
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Agricultural Research Service
U.S. Department of Agriculture
Washington, D.C. 20250
10. PROGRAM ELEMENT NO.
1HB617
11. CONTRACT/GRANT NO.
IA6 D4-0485
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Research Laboratory - Athens
U.S. Environmental Protection Agency
Athens, Georgia 30601
13. TY.PE OF REPORT AND PERIOD COVERED
Final Jan. '74 - June '76
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
Prepared as a joint publication of Office of Research and Development, EPA, and
Agricultural Research Service, USDA.
16. ABSTRACT
Engineering and agronomic techniques to control sediment, nutrient, and
pesticide losses from cropland are identified, described, and evaluated. Method-
ology is developed to enable a user to identify the potential sources of pollutants,
select a list of appropriate demonstrated controls, and perform economic analyses
for final selection of controls. The basic principles on which control of specific
pollutants is founded are reviewed, supplementary information is provided, and
some of the documentation used in Volume I is presented. Volume I (Report Mo.
EPA-600/2-75-026a) is available from NTIS as report no. PB 249-517.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. cos AT I Field/Group
runoff
pesticides
nutrients
non-point source pollution
hydrology
sediment control
erosion
agriculture
cropland
13 B
1. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report)
21. NO. OF PAGES
Unlimited
20. SECURITY CLASS (Thispage)
22. PRICE
EPA Form 2220-1 (9-73)
------- |