&EPA
United States
Environmental Protection
Agency
Environmental Research
Laboratory
Athens GA 30605
EPA-600- 3-79-106
October 1979
Research and Development
Effectiveness of
Soil and Water
Conservation
Practices for
Pollution Control
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1 Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and materials. Problems are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Se.vice, Springfield, Virginia 22161
-------
EPA-600/3-79-106
October 1979
EFFECTIVENESS OF SOIL AND WATER
CONSERVATION PRACTICES FOR POLLUTION CONTROL
Edited
by
Douglas A. Haith and Raymond C. Loehr
College of Agriculture and Life Sciences
Cornell University
Ithaca, New York 14853
Grant No. R804925010
Project Offreer
Lee A. Mulkey
Technology Development and Applications Kranch
Environmental Research Laboratory
Athens, Georgia 30605
ENVIRONMENTAL RESEARCH LABOPATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ATHENS, GEORGIA 30605
-------
DISCLAIMER
This report has been reviewed by the Environmental Research Laboratory,
U.S. Environmental Protection Agency, Athens, Ga., and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does men-
tion of trade names of commercial products constitute endorsements or recom-
mendation for use.
11
-------
FOREWORD
As environmental controls become more costly to implement and the
penalties of judgment errors become more severe, environmental quality man-
agement requires more efficient analytical tools based on greater knowledge
of the environmental phenomena to be managed. As part of this Laboratory's
research on the occurrence, movement, transformation, impact, and control of
environmental contaminants, the Technology Development and Applications
Branch develops management or engineering tools to help pollution control
officials achieve water quality goals through watershed management.
Efforts to reach water quality goals include the identification and
application of best management practices (BMPs) to control agriculturally
related water pollutants. This report examines soil and water conservation
practices used for many years in the United States to determine their poten-
tial as BMPs. Although these conservation practices are directed to erosion
and water control rather than direct control of nonpoint source pollutants,
their use in specific situations could provide environmental managers with an
additional tool for controlling pollutants of potential concern.
David W. Duttweiler
Director
Environmental Research Laboratory
Athens, Georgia
iii
-------
ABSTRACT
The principal objectives of this project were to identify the potential
water quality effects of soil and water conservation practices (SWCPs) and to
describe the economic implications of their use. A secondary objective was
to develop and test methodologies that could be used to estimate the effects
of SWCPs on pollutant losses from croplands. The project was limited to non-
irrigated field crops in the Eastern half of the United States. Potential
water pollutants considered were sediment, nutrients, and pesticides. Pro-
ject resources and duration did not permit the execution of experimental
field studies. Rather, research objectives were accomplished using three
related procedures. First, data from previous and ongoing field studies were
collected and reviewed. Second, mathematical simulation and linear program-
ming models were used to estimate the effects of SWCPs on edge-of-field losses
of sediment, nutrients, and pesticides and to evaluate the impacts on farm
income of implementing SWCPs for pollution control. The models were applied
to hypothetical farms and fields in New York, Iowa, Texas, and Georgia. Third,
a panal of outside experts from universities, government agencies, and con-
sulting firms was convened three times during the study to review methods
and results of analyses.
The major environmental benefit of SWCPs was determined to be their use-
fulness in controlling edge-of-field losses of sediment and total phosphorus
from croplands. It was established, however, that farm plans for control of
erosion are not necessarily equivalent to plans for sediment control. SWCPs
may not be effective at reducing total losses (runoff plus percolation of
dissolved nitrogen (nitrate) although runoff losses will generally be reduced.
SWCPs implemented for erosion or sediment control will have additional bene-
fits in terms of reducing losses of moderately or strongly adsorbed pesti-
cides although in many cases alternative methods are more effective than
SWCPs for control of pesticide losses. Conservation tillage will often be a
cost-effective method of reducing losses of sediment, nutrients, and pesti-
sides because it is a practice that can be implemented in many cases with
little or no decrease in crop yields.
This report was submitted in fulfillment of Grant No. R804925010 by
Cornell University under the sponsorship of the U.S. Environmental Protection
Agency. This report covers the period October 25, 1976, to October 24, 1978.
Work was completed as of January 31, 1979.
IV
-------
CONTENTS
Foreword iii
Abstract iv
1. Introduction 1
2. Conclusions 7
3. Recommendations 12
4. Definitions and Qualitative Evaluation of Soil
and Water .Conservation Practices 14
5. The Effects of Soil and Water Conservation
Practices on Sediment 39
6. The Effects of Soil and Water Conservation
Practices on Edge-of-Field Nutrient Losses 72
7. Simulation of the Action of Soil and Water
Conservation practices in Controlling Pesticides 106
8. The Cost-Effectiveness of Soil and Water Conservation
Practices for Improvement of Water Quality 147
9. The Effectiveness of SWCPs in Comparison with Other
Methods for Reducing Pesticide Pollution 206
References 287
Appendices
A. Cornell Nutrient Simulation (CNS) Model 325
B. Cornell Pesticide Model .(CPM) 340
C. Soil Loss Prediction Models 360
D. Control of Nitrogen Losses by Management
of Fertilizer Applications 369
E. Calculation of Budgets, Soil Erosion, and Sediment
Delivery for the Linear Programming Model 375
F. The Effects of Soil and Water Conservation Practices
on Runoff and Pollutant Loss from Small Agricultural
Watersheds: A Simulation Approach 385
v
-------
SECTION 1
INTRODUCTION
Historically, water pollution abatement policies have focused on control
of municipal and industrial sources. Interest in other potential sources has
increased as the nation has expanded its water pollution concerns. The em-
phasis of national water pollution control policies now is on the amount of
wastes that can be kept out of surface waters, rather than on the amount
that can be assimilated by the waters. Areawide waste treatment management
plans that are to be prepared as a result of Section 208 of the Federal Water
Pollution Control Act Amendments of 1972 (PL 92-50CT) are to include identifi-
cation of "agriculturally and silviculturally related nonpoint sources of
pollution" and are to "set forth procedures and methods to control to the
extent feasible such sources". Nonpoint sources will be difficult to control
if control is defined as the ability to establish and enforce effluent
standards. However, if the control is approached by using appropriate
management practices, many of the sources are controllable. In the United
States, the emphasis for control of nonpoint source contributions is being
placed upon "best management practices" rather than on the collection, treat-
ment, and effluent standards approach used for control of point sources.
Best management practices (BMPs) are defined as:
"A practice or combination of practices that is determined by a de-
signated areawide planning agency after problem assessment, examin-
ation of alternative practices and appropriate public participation
to be the most effective, practicable (including technological,
economic, and institutional considerations) means of preventing or
reducing the amount of pollution generated by nonpoint sources to a
level compatible with water quality goals."
The Clean Water Act of 1977 (PL 95-217) established a cost-sharing pro-
gram whereby the Secretary of Agriculture, with the concurrence of the
Administrator of the Environmental Protection Agency, may subsidize the in-
stallation of BMPs on farmlands for the purpose of water pollution control.
This activity is known as the "Rural Clean Water Program" and is under the
direction of the U.S. Soil Conservation Service. Successful implementation
of the Rural Clean Water Program will depend on identification of suitable
BMPs for control of agriculturally related water pollutants. This is not
a simple task, for such pollutants are primarily associated with diffuse
sources which are difficult to monitor and control.
-------
Soil and water conservation practices (SWCPs) are agricultural practices
that have the potential of being best management practices. However, the soil
and water conservation practices utilized in the United States for over 150
years are designed for erosion and water control rather than for direct
management of potential pollutants such as nutrients, oxygen demanding mater-
ials, or chemicals used for increased crop production or pest management.
Because SWCPs are utilized to retain the principal carriers of agri-
cultural nonpoint sources (runoff water and eroded soil) on a field, there can
be the tendency to consider all SWCPs as agricultural BMPs. This may not
always be appropriate for reasons which include the following
1. SWCPs are effective at reducing soil erosion on cropland. However,
many pollutants, including nitrogen, phosphorus and pesticides,
also are lost from cropland in dissolved form in runoff and per-
colation. The effects of SWCPs on the transport of pollutants
in solution is much less certain than their effects on erosion.
2. There are additional practices which are capable of reducing
pollutant losses from cropland such as fertilizer management,
integrated pest management and management of animal manure appli-
cations.
3. Water pollution from agricultural nonpoint sources results from
a chain of circumstances. Pollutants are lost from a cropped
field, transported through the landscape to surface or ground-
water, and may or may not result in an objectionable problem
(Figure 1-1). The relationship between any practice used on
a field and surface or groundwater quality is uncertain.
This research project was designed to evaluate the effectiveness of
SWCPs as potential BMPs and to clarify, where possible, the above uncertain-
ties. Conclusions in this report regarding the effectiveness of SWCPs for
nonpoint source pollution control are not reflections on their value for soil
conservation. Prevention of cropland erosion has long been a national ob-
jective and the soil conservation benefits of SWCPs were not evaluated in
this project.
DESCRIPTION OF THE PROJECT
Objective
The main objective of the project was to identify the potential water
quality effects of SWCPs and to describe the economic implications of their
use. The project evaluated the hypothesis that soil and water conservation
can be an economical way of significantly reducing agricultural nonpoint
source pollution. A secondary objective was to develop and test methodologies
which could be used to estimate the effects of SWCPs on pollutant losses from
arbitrary sites (cropped fields).
-------
Ul
edge of
field losses
losses to
surface water
agricultural field
and practices
intervening
land and
practices
overland flow
and constituents
impact of
agricultural land
use and management
practices
intermediate transport,
input, deposition and
transformations
pollution transport,
dispersion, and
water quality effects
FIGURE 1-1. TRANSPORT, TRANSFORMATION AND IMPACT OF POTENTIAL AGRICULTURAL NONPOINT SOURCES
-------
The project had a two-year duration and was limited to non-irrigated
field crops' in the Eastern half of the United States. Project duration and
resources did not permit the execution of experimental field studies. The
potential water pollutants which were considered in the study were sediment,
nutrients and pesticides. Effects of animal manures were not included.
Project analyses were generally limited to edge-of-field (Figure 1-1)
pollutant losses in surface runoff, subsurface water (interflow) and percola-
tion. Where possible, implications for water quality management were explored.
The project did not evaluate SWCPs in terms of any ultimate water quality
benefits that might occur.
Approach
Research objectives were accomplished using three related procedures.
1. Data from previous and ongoing field studies were collected and
reviewed to abstract information relevant to the effectiveness
of SWCPs.
2. Mathematical simulation and linear programming models were used to
estimate the effects of SWCPs on edge-of-field losses of sediment,
nutrients and pesticides and to evaluate the economic impact on
farms of implementing SWCPs for control of these pollutants. The
models were used to study SWCPs on hypothetical farms and fields
in New York, Iowa, Georgia and Texas.
3. A panel of outside experts (Table 1-1) was convened three times dur-
ing the two-year study to review project direction, methods of anal-
yses, regional differences in SWCP use and effectiveness, interpreta-
tion of information and results, and to review the conclusions.
The general methodology used to reach the conclusions could be of interest
to individuals having responsibility for agriculturally related nonpoint
source pollution control. It is recognized that each SWCP can control
some of the pollutants of potential concern. However, the key to their use
as a BMP is to understand the conditions under which they are effective and,
for specific situations, to be able to determine their potential effectiveness.
While the project results will be useful to interested individuals, many
individuals and agencies will wish to reach their own conclusions of the
effectiveness of SWCPs as BMPs in a site-specific situation. The approach
used in this study to determine SWCP effectiveness may permit others to
reach conclusions in their own situations.
PROJECT PARTICIPANTS
A multidisciplinary team of Cornell University faculty, staff and
graduate students were involved in the study. The disciplines included were
agricultural economics, agricultural engineering, agronomy, and environmental
engineering. The individuals having major project responsibility have been
-------
TABLE 1-1. EXPERT PANEL PARTICIPANTS
George W. Bailey
Environmental Research Laboratory-Athens
U.S. Environmental Protection Agency
Donald Boelter
U.S. Dept. of Agriculture
Forest Service
Blair T. Bower
Resources for the Future
Lee A. Christensen
U.S. Dept. of Agriculture
Economics, Statistics §
Cooperatives Service
Neil Cook
U.S. Dept. of Agriculture
Economics, Statistics §
Cooperatives Service
Norman H. Crawford
Hydrocomp, Inc.
James M. Davidson
Dept. of Soil Science
University of Florida
Clinton Johnson
U.S. Dept. of Agriculture
Soil Conservation Service
Victor J. Kilmer
formerly with Soils and
Fertilizer Branch
Tennessee Valley Authority
Joseph A. Krivak
Office of Water § Hazardous Materials
U.S. Environmental Protection Agency
Patrick J. Lawler
Lawler, Matusky § Skelly Engineers
William H. Luckmann
Dept. of Entomology
University of Illinois
L.L. McDowell
U.S. Dept. of Agriculture
Science § Education Administration
Robert B. McKusick
U.S. Dept. of Agriculture
Economics, Statistics §
Cooperatives Service
James W. Meek
Office of Water Planning § Standards
U.S. Environmental Protection Agency
Robert L. Metcalf
Dept. of Entomology
University of Illinois
L. Donald Meyer
U.S. Dept. of Agriculture
Science $ Education Administration
Lee A. Mulkey
Environmental Research Laboratory-
Athens
U.S. Environmental Protection Agency
Graham T. Munkittrick
U.S. Dept. of Agriculture
Soil Conservation Service
H. Page Nicholson
formerly with Environmental Research
Laboratory-Athens
U.S. Environmental Protection Agency
A.R. Robinson
U.S. Dept. of Agriculture
Science § Education Administration
William J. Sallee
U.S. Dept. of Agriculture
Agricultural Stabilization and
Conservation Service
Wesley D. Seitz
Institute for Environmental Studies
University of Illinois
-------
(cont'd.) TABLE 1-1. EXPERT PANEL PARTICIPANTS
Thomas N. Shiflet
U.S. Dept. of Agriculture
Soil Conservation Service
Donald F. Smith
formerly with Water Planning Division
Nonpoint Sources Branch
U.S. Environmental Protection Agency
Thomas E. Waddell
Environmental Research Laboratory-Athens
U.S. Environmental Protection Agency
Walter H. Wischmeier
formerly with U.S. Dept. of Agriculture
Science § Education Administration
Purdue University
David A. Woolhiser
U.S. Dept. of Agriculture
Science £ Education Administration
Colorado State University
listed as the authors and editors of the report. Many other individuals
played key roles in the project and several are acknowledged in the speci-
fic chapters. We particularly would like to recognize the input of Nelson
L. Bills and Fred N. Swader to the development of the project. Lee A.
Mulkey, the project officer for the study, made many contributions of inform-
ation, advice and encouragement which greatly facilitated work progress.
The project also had the benefit of assistance by Hydrocomp, Inc., Palo
Alto, California. Hydrocomp utilized its capabilities to run the Agricultur-
al Runoff Management (ARM) model to evaluate certain SWCPs. Appendix F pre-
sents the results of the Hyrocomp study. We wish to acknowledge and express
appreciation for the participation of Hydrocomp, Inc.
We also greatly appreciate the interest, perception, positive criticism,
and helpful input of the project's expert panel. The individuals who parti-
cipated in the panel are identified in Table 1-1. Their advice was given
freely throughout the project and was a significant contribution to the pro-
ject. Arrangements for the panel meetings were handled by Colleen S. Martin
in her usual competent fashion.
Last, but certainly not least, we wish to thank Doreen E. Kirchgraber
for her many hours spent organizing meetings, typing and proofing reports,
and carrying out countless other tasks which contributed to the successful
completion.of the project.
-------
SECTION 2
CONCLUSIONS
The effects of soil and water conservation practices (SWCPs) on losses
of agricultural chemicals from croplands in runoff and percolation are
largely determined by the site-specific effects SWCPs have on sediment
losses and water movement. Since most SWCPs significantly reduce sediment
losses they also reduce losses of chemicals associated with sediment,
such as organic nitrogen and phosphorus, inorganic particulate phosphorus
and organochlorines. Most SWCPs reduce surface runoff water, but to a lesser
extent than reductions in sediment. Reductions in runoff are usually associated
with increases in subsurface drainage, most particularly percolation to
groundwater aquifers.
Specific conclusions based on the comparative analysis of previous
research and the results of mathematical simulation and linear programming
models are summarized in the remainder of this section.
CONTROL OF SEDIMENT LOSSES
1. Cropland soil erosion control practices are generally used to
hold soil in place. Sediment control practices are designed to pre-
vent eroded soil from entering streams and lakes. Erosion control
practices reduce sediment loss from fields. Cost effective sediment
control requires consideration of practices which do not control
erosion (e.g., sediment basins or stream buffer strips) and concen-
trating erosion control practices on fields with highest sediment
delivery.
2. SWCPs are more effective in controlling sediment yield from fields
than in reducing runoff.
3. SWCPs reduce field sediment yield by reducing soil detachment and
its transport in runoff. Control of detachment by rainfall requires
protection of the soil surface by residues or crop canopies. Soil
detachment and transport by overland flow are controlled by reduction
of the quantity and/or velocity of flow.
4. Adsorbed chemicals are associated with clay and organic fractions
of sediment. These constituents are most effectively controlled
by SWCPs which provide protection from raindrop splash, such as
-------
sod-based rotations, no-tillage, and conservation tillage. SWCPs
that rely on reducing sediment transport such as graded terraces
or contouring are effective in controlling losses of organic
matter and clay if these sediment fractions are in soil aggregates.
5. The potential for erosion is predictably variable from month to
month dependent on local characteristic weather conditions and
crop growth. Since the effectiveness of SWCPs is seasonally
predictable, in some cases SWCPs can be matched with highest
potential sediment loss periods so as to be most effective.
6. SWCPs that control erosion by reducing raindrop impact are less in-
fluenced by storm magnitude than practices which rely on control
of soil detachment by flow or reduction of transport capacity. As
the magnitude of runoff events increases, the effectiveness of some
transport and flow detachment control type practices such as strip
cropping or graded terraces decreases. Contouring can become in-
effective under conditions of very high runoff.
CONTROL OF NUTRIENT LOSSES
7. Practices such as contouring, terraces, sod-based rotations, conser-
vation tillage and no-tillage significantly reduce edge-of-field
losses of solid-phase nitrogen and phosphorus because they reduce
erosion.
8. Sod-based rotations significantly reduce losses of dissolved phos-
phorus and nitrogen in surface runoff.
9. Practices such as contouring, terraces, conservation tillage and
no-tillage which reduce surface runoff reduce loss of dissolved
nitrogen in surface runoff, although reductions are less than those
of solid-phase nitrogen. These practices may increase dissolved
nitrogen losses in subsurface drainage (percolation and/or interflow).
10. Practices such as no-tillage and conservation tillage which protect
the soil surface with crop residues during the non-growing season
have an uncertain effect on losses of dissolved phosphorus in
surface runoff.
11. SWCPs such as contouring and terraces which are not based on residue
management, decrease losses of dissolved phosphorus in surface runoff,
although reductions are less than those of phosphorus associated
with sediment.
12. Management of nitrogen fertilizer applications to meet crop needs
can reduce the losses of dissolved nitrogen in both runoff and
percolation.
13. Losses of all forms of nitrogen and phosphorus in runoff vary
significantly from year to year. More importantly, the effects of
SWCPs also show significant yearly variations.
-------
CONTROL OF PESTICIDE LOSSES
14. Wider implementation of techniques which reduce the amount of pesti-
cide used for crop protection offers an effective way of reducing
the quantities of pesticide leaving crop fields. Cost-effective
insect pest control programs are currently being developed which
reduce insecticide use. Most of these programs integrate several
pest control techniques and are generally described by the term
"integrated pest management" (IPM). One of the most widely im-
plemented IPM techniques consists of the use of population monitoring
("scouting") to determine when pesticide applications are necessary.
More complex integrated pest management programs involving a com-
bination of cultural and biological control methods have been imple-
mented for cotton and are being developed for other crops.
15. Improper disposal of pesticide containers and dilute pesticide mixtures
in applicaton tanks appear to .be important sources of pesticide
pollution. Legislative controls on pesticide disposal as well as
technological advances in application equipment can effectively
reduce these sources of pesticide pollution.
16. The seasonal losses of pesticides due to drift and volatilization are
often greater than the amount of pesticide carried in runoff. Except
for practices like sod-based rotations which reduce the amount of
pesticide applied, SWCPs cannot usually reduce drift and volatiliza-
tion. Drift losses can be reduced by changes in formulation or
application procedures.
17. The effectiveness of SWCPs in controlling transport of five major
pesticides was compared with the effectiveness of alternative
practices. For the three insecticides studied (carbofuran, toxa-
phene, and methyl parathion) integrated pest management programs
combining scouting with cultural pest control methods seem to be a
more effective means of reducing the losses of insecticide in water
and sediment than SWCPs. SWCPs are effective at reducing the
quantity of paraquat in runoff. However, currently available infor-
mation indicates that paraquat in runoff is irreversibly adsorbed
to clay particles and does not appear to pose an environmental
hazard since it is not available to biological organisms. Of the
five pesticides considered, SWCP appeared to -be among the most
effective pollution control methods only for atrazine.
18. Most pesticides currently in use can be classified as moderately
adsorbed. Total mass flux of such pesticides in runoff occurs pri-
marily in the dissolved form. Thus SWCPs must reduce runoff water
in order to control the losses of these pesticides.
19. Pesticides such as paraquat; toxaphene, DDT, and dieldrin which are
strongly adsorbed to the soil are primarily associated with clay
and organic matter. Runoff losses of these pesticides will be most
effectively controlled by SWCPs which control erosion.
-------
20. The modelling results indicate that generally the effect the SWCP
on downward movement of pesticides below the root zone is minimal.
However, field and experimental data to support this conclusion
are lacking. In particular this conclusion would not hold for a
persistent, nonadsorbed pesticide.
21. The length of time between a pesticide application and the first
major runoff event following application has a very significant
effect on the amount of pesticide carried in the runoff, especially
for pesticides which are not persistent. Thus, improved timing
of applications could significantly reduce loss of these pesticides.
COST-EFFECTIVENESS
22. Although both the cost and effectiveness of all SWCPs may vary con-
siderably for different situations, conservation tillage methods
appear to have widespread potential for reducing soil erosion and
sediment delivery with little or no negative effect and sometimes a
positive effect on net farm income. Thus in many cases, these
practices should form the basis for farm plans designed to reduce
soil erosion and/or to reduce sediment delivery.
23. The most economical set of practices for a farm may depend on
whether the goal is to reduce sediment delivery to waterways or to
reduce levels of soil erosion. In particular, farm plans designed
primarily to control sediment delivery rather than soil erosion can
often specify location of crops and practices relative to waterways
and utilize sediment traps, thereby resulting in a reduction in
sediment delivery which is proportionally greater than the reduction
in soil erosion. Thus the cost of achieving a reduction in sediment
delivery can be significantly less than the cost of an equivalent
percentage reduction in soil erosion. The benefits of reducing both
soil erosion and sediment delivery should be considered in the
design of farm plans for improving water quality.
24. Although systems of practices designed to reduce soil erosion
generally achieve a reduction in sediment delivery roughly proportional
to the reduction in erosion, a few practices are notable exceptions.
On fields which have a relatively low sediment delivery ratio (SDR),
practices such as diversion ditches and sod waterways may, in some
cases, increase the SDR, resulting in increased levels of sediment
delivery. On fields with high SDRs, practices such as tile-outlet
terrace systems may decrease the SDR due to sediment deposition within
the terrace channel. Such practices will be relatively more efficient
in reducing sediment delivery than in reducing soil erosion.
25. For any given farm, the marginal cost of reducing sediment delivery
increases as required reductions become greater. To achieve relatively
small reductions, practices such as conservation tillage, contouring,
and strip-cropping (in areas where close-seeded crops are normally
grown) can often be implemented with little change in farm income.
To achieve greater reductions in sediment delivery, practices which
10
-------
are generally more costly, such as widely spaced terraces, must be
used. High levels of reduction require implementation of practices
generally associated with sharply increased farm costs. Typical
practices (usually in combinations) would include closely spaced
terrace systems, substitution of sod crops for row crops, and removal
of land from crop production. Implementation of such practices
generally results in a markedly greater farm cost and a considerable
reduction in the level of cost-effectiveness.
26. Farm plans designed to reduce soil erosion and/or sediment delivery
will not necessarily reduce losses of dissolved nutrients.
27. Reducing levels of sediment delivery to a waterway will reduce levels
of solid-phase nutrients reaching the waterway. The reductions,
however, may be proportionally less than the reductions in sediment
delivery in situations where the nutrient enrichment ratio increases
as levels of sediment delivery are decreased.
28. Results from the Cornell Pesticide Model and the budgeting studies
indicate that neither SWCPs nor incorporation of herbicides are
highly cost-effective for reducing losses of herbicides in runoff.
If implemented only for the purpose of reducing herbicide losses,
reductions will generally not exceed 1-2 grams in edge-of-field
losses per dollar of farm cost. However, if SWCPs are implemented
for the control of soil erosion and/or sediment delivery, reductions
in herbicide losses may be an important secondary benefit.
SUMMARY
The major environmental benefit of SWCPs is their usefulness in con-
trolling edge-of-field losses of sediment and total phosphorus from croplands.
SWCPs may not be effective at reducing total losses (runoff plus percolation)
of nonsorbed pollutants such as nitrate, although runoff losses may be reduced.
Total nitrate losses can be controlled by management of nitrogen fertilizer
applications. SWCPs implemented for sediment control will also have some
additional benefits in terms of reducing losses of pesticides, especially
those which are moderately or strongly adsorbed and are persistent. However,
in many cases available alternatives are more effective than SWCPs for con-
trol of pesticide losses.
11
-------
SECTION 3
RECOMMENDATIONS
The principal conclusions of the study (Section 2) reflect a recognition
that supporting evidence is largely limited to mathematical modelling re-
sults and fragmentary field data. The evaluation of economic and water
quality implications of soil and water conservation practices (SWCPs) is
a relatively new emphasis and more definitive conclusions must await further
study. It is recommended that further research be initiated in four major
areas.
1. Long-term field studies to measure the effects of SWCPs on either
pollutant losses or crop yields have been rare. Hence deter-
minations of cost-effectiveness are generally based on mathematical
models which have seen limited validation. Until systematic long-
term field studies are complete, it is likely that many uncer-
tainties will remain. However, future field studies must be
linked to modelling efforts, since the site-specific nature of
nonpoint source pollutants will always preclude experimental
study of all relevant variables.
Technical and economic evaluation of the water quality implica-
tions of individual SWCPs and SWCPs in combination should be
continued to more clearly identify reductions in pollutant
load, the level of management needed for continued effective-
ness, and applicability of use. The purpose of such field
studies would be two-fold: to provide an indentification of the
probable effects of SWCPs on pollutant losses and crop yields
and to provide systematic testing of mathematical models which
will be utilized as planning tools in most water quality man-
agement studies.
2. The linkages between edge-of-field pollutant losses and water
quality were only partially addressed in the project, generally
by the implicit assumption of a linear relationship. The
transport of pollutants from field to stream or aquifer and
subsequent movement and behavior of the pollutants in
the receiving water are important determinants of water quality.
Unfortunately, these phenomena are both imperfectly understood
and infrequently measured.
12
-------
Field and modeling studies to provide information on the funda-
mental relationships involved in the transport and transfor-
mation of pollutants from source to stream and as the pollutants
move in the stream will contribute needed understanding on the
impact of SWCPs on water quality.
3. In several sections of the report the use of SWCPs for water
pollution control was compared with their use for reducing
cropland erosion. The point of these comparisons was not to
show that erosion and pollution control represent opposing
resource management options, but rather to indicate that an
erosion control program can be an inefficient way to control
water pollution. While this is an important observation,
most individuals would agree that the issue is not management
of erosion or_ water quality. It is clearly essential to the
nation and the agricultural community that both problems be
solved.
Before this can be done more research is needed to develop
systems of practices which can simultaneously and efficiently
serve both soil conservation and water quality goals. Such
research should be particularly sensitive to the effects
of cost-sharing and technical design criteria on the potential
of SWCPs for achieving these two goals.
4. Certain components of the study noted that alternative approaches
can be used to help meet water quality objectives. These
alternatives include integrated pest management and fertilizer
and manure management.
Investigations to identify the effectiveness of these and other
nonstructural management approaches to reduce pollutant loads
will permit a better understanding of the proper application and
value of both these approaches and SWCPs to help meet water
quality objectives.
13
-------
SECTION 4
DEFINITIONS AND QUALITATIVE EVALUATION OF
SOIL AND WATER CONSERVATION PRACTICES
Tammo S. Steenhuis and Michael F. Walter
DESCRIPTION OF SOIL AND WATER CONSERVATION PRACTICES
A soil and water conservation practice (SWCP) is an agricultural practice
which reduces soil erosion and/or increases water retention. Selected prac-
tices discussed in this report are of interest because (1) they are in common
use and recommended by the Soil Conservation Service, and (2) data from field
studies are available for reference and comparison of results. The included
practices were selected from the SCS National Engineering Handbook, Section 2,
(1966) and from control of Water Pollution From Cropland, Volume 1 (Stewart
et_ al_. 1975). Definitions of other key terms used throughout this report are
In" Table 4-1.
SWCPs can be divided into two groups. The first, cultural practices, are
nonstructural in nature. This group includes no-tillage, conservation tillage
contour farming, graded rows, ridge planting, contour listing, contour strip
cropping, sod-based rotations, and cover crops. The second group consists of
structural practices which usually require off-farm help for construction. The
practices in this group include terraces, diversion terraces, grassed waterways,
filter strips, and artificial drainage.
Definition of Conventional Tillage
To compare the relative effectiveness of these practices, a base or
standard must be defined. For the purposes of this report, conventional
tillage is used as the standard and is defined as moldboard plowing, followed
by secondary tillage at least three times to smooth and pulverize the soil,
and planting and cultivating where appropriate. Operations are assumed to be
done up and down the land slope. In general, all operations, including
fertilizer application, are assumed to be done in the spring. In some locations,
fall plowing and/or fertilizer application may be conventional, and shifting
these operations to spring would be classified as a SWCP. Thus, the concept of
"conventional tillage" varies with crop grown and location, but in all cases
the surface is free from residues for a period of time.
Cultural Practices
No Tillage--
The term no-till refers to a method of planting crops that involves no
seedbed preparation other than opening the soil for the purpose of placing
14
-------
TABLE 4-1. SELECTED KEY DEFINITIONS OF TERMS USED IN THIS REPORT
Adsorption
Erosion
Sediment
Sediment Yield
The adhesion of a substance to the surface of
soil particles.
Detachment and movement of soil or rock fragments
by water, wind, ice or gravity (for purposes of
this study, erosion refers only to that caused by
water and gravity). Types of water erosion
include:
Gully Erosion: Advanced stage of erosion which
produces large channels which cannot be smoothed
over by normal tillage operations.
Natural Erosion: Wearing away of the earth's
surface by water under natural environmental
conditions of climate, vegetation, etc., undis-
turbed by man. Also called geologic erosion.
Rill Erosion: An erosion process- in which
numerous small but well defined channels only
several inches deep are formed. The rills can
easily be removed by normal tillage operations.
Sheet Erosion: The removal of a thin, relatively
uniform layer of soil particles. Also called
interri 11 erosion (Foster and Meyer, 1977).
Splash Erosion: The spattering of small soil
particles caused by the impact of raindrops on
wet soils. The loosened and spattered particles
may or may not be subsequently removed by surface
runoff.
Any solid material, either mineral or organic, that has
been eroded, transported and deposited on land or in
water.
Amount of sediment which moves off a unit area
(e.g., edge of field losses).
Sediment Delivery Sediment that is transported to a specific point,
generally a stream.
Sediment Delivery The ratio between sediment delivery and soil erosion.
Ratio (SDR)
Gross Soil Erosion Soil erosion as defined by the Universal Soil Loss
Equation,
15
-------
the seed at the intended depth, usually accomplished by opening a small slit
or by punching holes in the soil. There is generally no cultivation during
crop production, and chemicals are used for weed control. In some parts of
the country, no-till has a wider meaning. A band 15 to 20 cm is tilled with
a rototiller attachment. Unless otherwise specified in this report, the term
no-till is exclusively associated with tillage of a very narrow band of up to
5 cm during seeding.
Conservation Tillage--
This term can refer to a wide variety of tillage systems. In the present
definition of the Soil Conservation Society of America, it is any tillage
system which reduces loss of soil and water compared to unridged clean tillage.
Reduction of soil and water loss is accomplished in most cases by protecting
the surface with crop residues.
Contour Tillage--
Contour tillage is a traditional SWCP in which field operations, such as
plowing, planting, cultivating, and harvesting, are performed following the
natural field contour. Because of its'ineffectiveness on long or steep slopes,
contouring is often combined with other practices such as strip cropping and
terracing (Jamison et^ al. 1968).
Graded Rows--
Where poor drainage or the likelihood of extreme storms are a problem,
rows on a slight gradient may be used rather than contouring to prevent break-
ing over of rows. Rows, therefore, should have just sufficient gradient to
prevent the accumulation of excess water.
Contour Listing and Ridge Planting--
Neither practice is commonly used at the present time, but they have
potential for reducing runoff and soil loss. Contour listing is accomplished
by an implement called a lister, which is a double plow, the shares of which
throw the soil in opposite directions, leaving the field with a series of
alternate ridges and furrows. Row crops may be seeded in the bottoms of the
furrows or on the top of the ridge as they are opened up. Ridge planting
refers to a practice in which crops are planted on preformed ridges. Crop
residues are moved into the furrows between rows to help control runoff and
erosion. Ridges are generally on the contour or at a slight gradient.
Contour Strip Cropping--
The practice of growing crops in comparatively narrow strips of land on
which the farming operations are performed on the contour is called strip
cropping. Strips of grass, close growing crops, or fallow are alternated
with those in cultivated crops. Thus, contour strip cropping implies both
contouring and crop rotation.
Sod-based Rotations--
Growing different crops in recurring succession is called crop rotation
and is one of the oldest methods of maintaining soil fertility, dating back
at least to the Middle Ages in Europe. A typical rotation in much of the
country has been a few years of a row crop, followed by one year of small grain
required as a nurse crop for the establishment of a sod crop. However,
recently introduced herbicides and decreased demand for straw have encouraged
16
-------
many farmers to directly seed their hay crops. Both of these alternatives
are included in this definition. In this project, the sod crop was assumed
to last for at least 3 years and to consist predominately of alfalfa. In
practice, other legumes and/or grasses may be used as sod crop while the
duration will be determined by the particular economic situation and the type
of crop.
Cover Crops--
A cover crop is a close growing crop grown primarily for the purpose of
protecting and improving soil between periods of regular crop production.
The most common crops are rye and wheat which are planted shortly before or
soon after harvest and plowed down or chemically killed in the spring. The
typical situation analyzed in this report is a cover crop following corn
silage or soybean harvest, as neither crop leaves sufficient residue to
protect the soil against erosion.
Structural Practices
Terraces--
Terraces are not the same in each part of the country. This is reflected
in the definition of a terrace as an embankment or combination of an embank-
ment and channel constructed across a slope to control erosion by diverting
or storing surface runoff instead of permitting it to flow uninterrupted
down the slope. Terraces or terrace systems may be classified by their
alignment, gradient, outlet, and cross-section. Alignment may be parallel or
non-parallel. Gradient may be level, uniformly graded or variable graded.
Grade is often incorporated to permit paralleling the terraces. Outlets may
be soil infiltration only, vegetated waterways, tile outlets or a combination
thereof. Cross-section may be narrow base, broad base, bench, steep back
slope, flat channel or channel.
Diversion Terrace--
Diversions differ from terraces in that they consist of individually
designed channels across a hillside. They may be used to protect bottom land
from hillside runoff, placed above a terrace system, or used in connection
with strip cropping to shorten the length of the slope so that the strips can
more effectively control erosion. Finally, they also can offer a last-chance
possibility for controlling erosion in critical zones: for example, by
diverting water away from the beginning of an existing or potential gully.
Grassed Waterways--
Grassed waterways are natural or constructed broad and shallow channels
that are planted with erosion resistant grasses. They are used as outlets
for terraces, graded rows or contouring.
Filter or Buffer Strips--
Strips of permanent vegetation which are placed above farm ponds,
diversion terraces or other natural or man-made drainage channels or at the
edge of field to retard flow are called filter strips or buffer strips.
They cause deposition of transported material and thereby reduce sediment flow.
Artificial Drainage--
Drainage is the removal of excess surface water or groundwater from land
17
-------
by means of surface or subsurface drains. Drainage is usually considered as a
soil and water conservation practice, although in the true sense of the
definition, it is not, as it conserves neither soil nor water.
Extent of Use of SWCPs
Although accurate information on the extent of use of SWCPs is somewhat
difficult to obtain, estimates of the area associated with certain practices
have been furnished by Dr. R. M. Gray of the U.S. Soil Conservation Service
(Gray, 1977, personal communication). These have been summarized for a
number of Land Resource Areas (LRAs) shown in Figure 4-1 and are given in
Table 4-2. The term "Crop Residue Use" is not included in the definitions of
the previous section. Crop residue use implies leaving at least 2 MT/ha of
crop residue on the field over the critical erosion period. Conservation
tillage means leaving all or part of crop residues on the field throughout the
year. Thus, these two practices are mutually exclusive.
Crop-residue use is by far the most widely used practice, accounting for
over 60% of land in practices in the LRAs in Table 4-2. Approximately 13% is
in conservation tillage and 18% in terraces.
Contour farming and strip-cropping are fairly minor in most regions, but
account for 24% and 20%, respectively, of all land in practices in the North-
east (LRAs 101, 140, 145).
There is considerable variation in land in terraces. In LRA 78, 50% of
the land is in practices and nearly 40% of the total cropland is in terraces.
For several areas (101, 131, 1403 145, 153), terraced areas amount to under 1%
of land in practices. In the corn belt areas (102, 108, 111), about 4% of
land in practices is terraced.
Although the LRAs included in Table 4-2 are not necessarily representa-
tive of the United States east of the Rocky Mountains, they do represent a
variety of agricultural areas and can be assumed to typify patterns of SWCP
use in that part of the country.
PHYSICAL PRINCIPLES OF SWCPs
As their name implies, soil and water conservation practices are pri-
marily intended to conserve soil and water. In humid areas, conservation of
soil is often the primary goal, whereas in dry areas, conservation of water
is traditionally given more emphasis. However, since water and soil move
together, controlling the movement of one generally affects the movement 0f
the other. This section discusses the physical means by which SWCPs help
control soil and water movement from cropland.
The physical effects of soil and water conservation practices are
primarily based on control of runoff, rain splash energy and soil structure.
These effects are discussed below and summarized in Table 4-3.
Control of Runoff
Reduction of Runoff Velocity--
18
-------
145
153
153
FIGURE 4-1. LAND RESOURCE AREAS AS REFERRED TO IN TABLE 4-1.
-------
ro
o
TABLE 4-2. ACREAGE (1000 HECTARES) OF LAND UNDER SOIL AND WATER
CONSERVATION PRACTICES AND CONVENTIONAL TILLAGE IN
CERTAIN LAND RESOURCE AREAS
2 LRR*
SKCP^ LRA3
MT
CRU
CF
SC
TER
TER, CF
TER, MT
TER, CRU
SC, CF
SC, MT
SC, CRU
CF, MT
CF, CRU
TER, MT, CF
TER, CRU, CF
SC, MT, CF
SC, CRU, CF
Total Practices
Conventional
Tillage
Total
% of Total
Cropland in
Practices
78
57
1085
202
3
0
715
0
0
7
*
3
13
227
43
853
*
1
3210
922
4132
78
H
80
5
911
21
3
33
123
*
83
1
*
4
1
31
2
274
*
1
1494
392
1886
79
L
101
43
118
46
35
0
*
0
0
4
3
8
4
11
*
*
*
1
274
485
759
36
00
131
75
2430
15
4
*
3
*
*
*
*
6
1
25
*
4
0
0
2564
1691
4255
60
134
56
1208
159
1
0
17
0
0
*
*
4
4
163
1
19
*
*
1633
888
2521
65
P
136
65
262
105
28
0
91
0
0
10
4
16
16
61
15
61
1
5
740
419
1159
64
137
4
101
17
1
2
21
*
1
1
*
1
2
26
1
26
4
1
206
79
285
72
T
153
— T
458
4
*
0
1
0
0
*
*
*
*
6
*
1
0
it
544
454
998
55
R
140
26
62
97
85
0
*
0
0
13
3
8
4
8
0
*
1
1
30S
684
992
31
145
9*
4
1
*
0
1t
0
0
0
0
*
0
0
0*
0
0
0
7
43
50
14
102
409
1734
221
78
0
162
0
0
8
11
52
42
127
20
40
1
5
2910
2697
5607
52
108
899
1875
315
33
0
55
0
0
9
12
32
109
278
19
50
3
10
3699
2815
6514
57
111
856
2765
28
28
0
22
0
0
*
S
7
0
IS
3
65
*
*
3744
3019
6763
55
Percent of
total area
Total in Practices
2567
13013
1231
300
35
1211
1
84
55
41
142
205
978
106
1334
9
26
21333
12
61
6
1
0
6
0
0
0
0
1
1
5
0
6
0
0
100%
i
Land resource regions
H = Central Great Plains
L - Great Lakes
00 = Mississippi Delta
P « South Atlantic and Gulf Slope
T = Atlantic and Gulf Coast
R « Northeast
M « Midwest
Soil and Water Conservation Practice
MT = Minimum Tillage
CRU « Crop Residue
CF = Contour Farming
SC = Strip Cropping
TER = Terraces
Land Resource Area
See Figure 4.1 for location
-------
TABLE 4-3.
No-till
Conservation Tillage
Contouring
Graded Rows
Strip Cropping
Sod-based Rotations
Cover Crops
Listing
Ridge Planting
Terraces
Diversions
Grassed Waterways
Filter Strips
Drainage
Decrease
Runoff
Velocity
+
+
0
+
+
+b
+c
+ d
+
+
+
+
+g
0
Increase
Surface
Storage
0
0
+
0
+
0
0
d
d
+e
0
0
+g
0
Increase
Soil Decrease Improve
Moisture Splash Soil
Storage Energy Structure
+ +
+ + +
ooo
ooo
+
+
0 H- -f
0 0
+ 0
ooo
0 0
+f .f
+g +g +g
+ 00
a+ The effect is as indicated
0 There is no effect
f
g
- The effect is opposite of the indicated
During the period of sod crop only
During winter and early spring only
Only if plowing on the contour
On level raised bench terraces and tile-outlet terraces with raised
outlets only
Only in the waterway
Effective only at the edge of the unit source area
21
-------
The velocity of moving water will be reduced whenever the total energy
available for water movement is decreased. This may be accomplished by
forcing the water to move laterally rather than straight down the slope, by
reducing the slope of the land through landforming, or by increasing surface
roughness which dissipates the water's kinetic energy. Surface roughness is
effectively increased by reducing secondary tillage operations, increasing
water-stable aggregates on the soil surface or by the use of a mulch cover.
Decreasing runoff velocity helps reduce both surface runoff volume and
soil loss. A slower flow rate allows water to remain on the field for a longer
period permitting increased infiltration.
Increase in Surface Storage--
Any obstruction in the flow path (e.g., ridges of soil or vegetation)
which allows water to pool will increase surface storage. The trapped water
is removed from the total surface runoff volume, which results in a-decreased
runoff velocity and a reduced sediment carrying capacity.
Trial results of measurements of increased surface storage must be
evaluated with care, particularly if results are given in percentages. For
example, suppose a certain practice increases a field's surface storage
capacity from 0.5 to 1 cm of water. For a 2-cm rainstorm, the runoff would
drop from 1.5 to 1.0 cm, a 33% decrease. But for a 10-cm storm, the change
from 9.5 to 9.0 cm would be less than 6%. Also large rainstorms may wash
over and erode the obstructions. Thus, in some cases, for large storms,
increased surface storage may not save any more soil or water than if straight
row cropping was used QVischmeier and Smith, 1965).
Increased Conductivity and Moisture Storage--
Some practices increase the soil macropores connecting to the soil sur-
face, which can greatly increase infiltration and conductivity. Soil mois-
ture storage can be increased by either draining or evaporating moisture in the
soil profile.
Reduction of the Splash-Energy of Falling Rain—
Rainfall striking bare soil can result in sealing of the soil surface.
In many cases a surface crust is a limiting factor for water infiltration
(Hillel and Gardner, 1969). Dissipating raindrop energy by use of a plant
canopy or a mulch will greatly reduce the surface sealing effect (Duley, 1939)
thus increasing infiltration and decreasing runoff volume.
Improvement of Soil Structure--
A change in bulk density, porosity, and percent of water-stable aggregates
in soil all affect its credibility and its capacity for water infiltration.
Lower bulk density and higher porosity increase the infiltration rate of soil
water. Formation of surface crust is less extensive in soils having a high
percentage of water-stable aggregates (Schwab et_ al. 1966). Practices which
reduce tillage operations or increase soil orgarficTlnatter will generally cause
improvements in soil structure.
22
-------
PHYSICAL PRINCIPLES OF WATER AND POLLUTANT MOVEMENT
Interactions between pollutant movement, climate and SWCPs are so
numerous that resource limitations preclude experimental study of all of
them (Dean and Mulkey, 1978). Interest in practices for control of agricul-
tural nonpoint sources is recent and few studies of these interactions have
been done to date (U.S. Comptroller General, 1977). However, a relatively
large body of research is available on the effect of SWCPs on water and sedi-
ment movement. This knowledge, together with what is known about pollutant
movement, is used in this report to predict the effect of SWCPs on loss of
pesticides, nutrients and other chemicals from agricultural land. The phys-
ical principles of water and pollutant movement are discussed here to give
the reader an insight in the basic processes which are involved, while the
movement of soil is discussed in Section 5.
Water Movement
When rain and melt water reach the surface of the ground, they encounter
a "filter" that is of great importance in determining the path by which the
water reaches the stream channel. The paths taken by the water (Figure 4-2)
determine in many cases the type and load of pollutants leaving the field, and
hence, the strategies required for decreasing the pollutant load.
OVERLAND FLOW
INTERFLOW
INFILTRATION
DEEP PERCOLATION
BASEFLOW
SUBSURFACE
FLOW
FIGURE 4-2. PATHWAYS BY WHICH WATER MOVES POLLUTANTS FROM CROPLAND.
23
-------
When the rate of rainfall or snowmelt exceeds the infiltration rate of
the soil and the storage capacity of the land-surface depressions, excess
water becomes overland flow. Overland runoff will contain dissolved materials
such as nutrients or pesticides and carry a sediment load which in its turn
may carry adsorbed substances. The quality and quantity of the dissolved
chemicals depend on the soil, vegetation, organic residue management, fertil-
izer and pesticide management, and soil conservation practice.
After water infiltrates the soil, it may be stored there or move out of
the field. If the soil is deep and of uniform permeability, the subsurface
water percolates to the zone of saturation and then moves with the ground
water flow. Because rates of groundwater flow are generally slow and the
underground paths long, most of the water following such paths continues to
flow between rainstorms. When the groundwater resurfaces in a stream channel
it is termed "base flow."
If at some shallow depth in the soil, percolating water encounters an
impeding horizon, a portion of the water will be diverted horizontally and
will reach the stream by a much shorter route. Because of the shorter route
and high permeability of top soil and weathered rocks (relative to unweathered
rock), and generally greater gradients of hydraulic potential in these upper
sloping horizons, water following this path reaches the edge of the field
much more quickly than the groundwater flow described above. This water is
classified as subsurface storm flow, or interflow.
In some fields, vertical and horizontal flow may cause the soil to
become saturated throughout its depth. When this happens, some of the water
moving by the shallow subsurface path emerges from the soil surface and
becomes overland flow.
Movement of Pollutants
In this section, only the behavior of pesticides and nutrients are
considered. The reader is referred to Sections for a discussion of the
movement of soil. The physical and chemical properties of pollutants deter-
mine their path and mode of travel out of a field. Pollutants can be dis-
solved in the soil water, be adsorbed to soil particles or be in solid phase.
Water transport of a substance includes movement in its dissolved phase
in the surface and subsurface runoff and transport of solid and adsorbed
phases in the surface runoff. The volume of runoff water is generally much
greater than the quantity of sediment yielded from a watershed. Therefore,
even if the concentration of a substance is greater in the sediment than in
the water, the quantity of the substances moving off a field dissolved in
the runoff might be greater than that adsorbed to the sediment (Mulkey and
Fa/co, 1977; Baker et_ a^. 1977, 1978). Both the absolute amounts of a
substance carried in the aqueous, adsorbed and solid phases and its relative
mobility in the water and sediment are important. Runoff water and the
dissolved or aqueous constituents will move farther in a shorter time than
sediment and the adsorbed and solid pollutants which are subject to redeposi-
tion. The possibility of degradation and transformation is greater during
the longer transport time required for sediments.
24
-------
Grouping Pollutants by Adsorption Properties--
There are many ways to categorize pollutants (solubility, persistence,
etc.), but in studying pollutant losses in soil and water, it has been
proven useful to classify pollutants, other than sediment, according to
their relative concentrations in water and on soil particles as indicated by
adsorption-desorption isotherms (White and Beckett, 1964; Ryden et al. 1973).
These may be nonlinear and multiple valued for each combination of chemical
and adsorbent. However, for the purpose of comparing the adsorption of
substances an adsorption partition coefficient, ks,for a given solution
concentration can be calculated as the ratio of amount adsorbed to that in
solution, i.e.,
k _ concentration of substance adsorbed to soil particles (ppm; mg/kg)
s concentration of substance in solution (ppm; mg/Jl)
Typical partition coefficients obtained from published isotherms and runoff
studies are given for pesticides and nutrients in Table 4-4 and 4-5, respec-
tively. Some of the members of the pesticide groups exhibit high ks values,
especially with montmorillonite clay. Values with kaolinite clay are lower.
TABLE 4-4. COMPARATIVE GROUPING OF SELECTED PESTICIDES BASED ON ADSORPTION
PARTITION COEFFICIENTS OBSERVED IN OVERLAND FLOW WITHIN ONE
MONTH AFTER APPLICATION.
Group I
ks ~ 1000
paraquat^
DDTe
toxaphene*
Group II
ks ~ 5
trifluralin^
atrazinec>£
cyanazine^'S
2-4-D2
diphenamid6
alachlor^
Group III
ks ~ 0.5
chloramben
dicamba1-
3.
dichlobenil,
fluometuron
Bailey et al .
Baker et al .
1974
1978
s*
Baker and Johnson, 1977
dDavidson et auU 1975
6Haan, 1971
£Helling, 1971
gSmith et al. 1978
25
-------
TABLE 4-5. COMPARATIVE GROUPING OF SELECTED NUTRIENTS BASED ON ADSORPTION
PARTITION COEFFICIENTS
Group I Group II Group III
ks ~ 1000 ks ~ 5 ks ~ 0.0.5
Organic Nitrogen Soluble Inorganic Nitrate
Phosphorus
Ammonium
Solid Phase
Phosphorus
Many substances are strongly adsorbed to soil and have adsorption par-
tition coefficients of 1000 or more. Another group of potential pollutants,
including most currently used pesticides, have adsorption partition coeffi-
cients ranging from 2 to 20. Therefore, for the purposes of this study, the
following groupings of pollutants are useful.
I. ks - 1000, strongly adsorbed and solid phase pollutants
(e.g., solid phase organic nitrogen, DDT, paraquat)
II. kg ~ 5, moderately adsorbed pollutants (e.g., most pesticides)
III. kg ~ 0-0.5, nonadsorbed or soluble pollutants (e.g., nitrate)
Losses of Strongly Adsorbed and Solid Substances (Group I)--
Losses of strongly adsorbed and solid substances in baseflow and inter-
flow are usually small, while losses in overland flow can be high and often
are related to the amount of sediment in the runoff and the amount of the
substance in the soil. The concentration of the different nutrients vary
tremendously for different soils. Brady (1974) , for instance, gives a range
of 0.02-0.50 mg/liter for nitrogen.
Losses of organic matter in eroded soil can be as high as 1.2 tons/ha
(Barrows and Kilmer, 1963). Since organic matter is concentrated in the soil
surface and has low density, it is among the first components to be removed
(Holt j?t_ a.1^. 1970). Thus, sediment contains more organic matter and associa-
ted nutrients than the surface soil from which it originates (Massey and
Jackson, 1952; Rogers, 1941; Bruce ejt al_. 1975; and McElroy ert al_, 1976).
Jacobs (1972, quoted by Casler and Jacobs, 1975) and White (1953,
quoted by Barrows and Kilmer, 1963) showed that total phosphorus (P) and
nitrogen (N) losses are related to the amount of soil carried off the land.
When soil loss is high, P and N losses have been positively correlated with
soil loss (Duley and Miller, 1923; Timmons e^t aJU 1968), and the concentra-
tion of P and N in the sediment and the original soil are usually similar
(Massey and Jackson, 1952). But, if the concentration of sediment in the
runoff is low, the nutrient concentrations in the sediment can be several
times that in the original soil. When soil loss is low, as is often the
case with grasslands, there is often no correlation between sediment load
and level of nutrient loss.
26
-------
Paraquat and diquat are known to be strongly adsorbed to soil constit-
uents (Weber and Weed, 1968; Knight and Tomlinson 1967; Burns et al. 1973;
Weed and Weber, 1969). Nearly all paraquat and diquat losses are, therefore,
associated with sediment (Smith et^ al_. 1978). Nicholson et_ al_. (1966) found
that at the outlet of a 400 sq. mile watershed, essentialTy~ail DDT was
associated with suspended sediment.
The above substances are transported primarily on soil in runoff. There-
fore, if a SWCP reduces sediment, it should also influence transport of these
substances. SWCPs which deal only with water movement and do not affect
sediment movement will not significantly affect losses of these strongly
adsorbed substances.
In general, phosphorus is probably best categorized as a strongly adsorbed
nutrient. But because it has so many forms, it cannot be as clearly categor-
ized by adsorptive properties as other potential pollutants such as pesticides.
There are other complicating issues with regard to phosphorus as well. The
origin of phosphorus leaving a field is often difficult to trace. Phosphorus
is taken up by the plant and is built into its basic structure. This phos-
phorus can wash off a field without ever being part of the soil or soil
solution.
In the spring, summer and fall, phosphorus losses are mainly associated
with the sediment. However, in the winter, at least in the colder climates,
phosphorus losses in the dissolved phase tend to be high (Harms et al. 1974).
Presumably the cell walls of plant residue freeze and rupture allowing phos-
phorus to leach out (Martin et^ al_. 1970; White, 1973). Thus, residue manage-
ment may decrease soil loss but increase the soluble phosphorus losses.
McDowell(personal communication, 1978) found that although the soluble
phosphorus losses increased due to residue leaching, the total phosphorus
load from test fields was lower because of the decrease in soil loss.
Phosphorus has been found in the subsurface flow. Bouldin et_ al.(1975)
estimated that this source accounted for one-third of the inorganic phosphorus
load from a New York watershed. The phosphorus concentration in the subsur-
face flow seems little affected by agricultural practices. Thomas and
Crutchfield (1974) found that the amounts of soluble phosphates in the
drainage water from a field were essentially the same as published by
McHargue and Peter in 1921 despite the fact that fertilizer use had increased
from zero up to considerable amounts.
Losses of Moderately Adsorbed Substances (Group II)--
A distinction should be made between the loss of moderately adsorbed
substances in overland flow, interflow, and deep percolation. The behavior
of substances in overland flow is completely different than that in subsurface
flow and the two flow processes are discussed separately.
Although some studies have indicated that losses of moderately adsorbed
substances in overland flow are related to both soil loss and runoff water,
most evidence shows that they are essentially related to runoff water. The
following studies illustrate this point.
27
-------
In research on the pesticides atrazine and GS13529 applied at rates of
2.2 and 4.5 kg/ha on Hagerstown silty clay loam on a 14 percent slope with
corn planted, it was found that over eight times more atrazine was lost in
runoff than with sediment at the low rate of application, and over four
times at the high rate (Hall et al_. 1970) . Kearney (1972) found much higher
losses of 2,4-D, 2,4,5-T and picloram in runoff water than on sediments.
In Iowa, Ritter et_ a.L (1974) found that most atrazine was lost in the
water fraction. Baker~ancF Johnson (1977) showed that changing tillage
practices on a Kenyon soil resulted in a tenfold decrease in soil loss but
the atrazine load in the runoff almost doubled even though atrazine appli-
cation rates were not changed. Runoff water loss remained nearly constant.
In Coshocton, Ohio, a strong correlation was found between amount of runoff
and total pesticide loss (Triplett et_al. 1978). Studies in Georgia showed
that initial amounts of rainfall abstraction and infiltration were inversely
related to pesticide loss. However, there was also a positive correlation
of soil loss with total pesticide loss (Bailey et_ al. 1974).
Contrary to movement by surface flow, transport of adsorbed substances
by water passing through the soil matrix is relatively slow, and an equi-
librium exists between the substances dissolved in the soil water and
adsorbed to the soil. Freed and Hague (1974) found that none of the pesti-
cides they tested moved downward more than 50 cm in a loam soil at 25°C
under simulated annual rainfall of 150 cm. They looked at a number of
pesticides typical of those currently applied to agricultural fields. The
worst pesticide pollution of subsurface flow water derives from pesticides
that are very weakly adsorbed such as aldicarb (Porter, personal communi-
cation, 1977) or that are very slow to degrade.
Studies have indicated that the leaching of strongly and moderately ad-
sorbed substances from the upper horizons does not result in increased con-
centrations in the groundwater. Simonson (1970) argued that because of the
soil formation process in the groundwater zone, losses of nutrients in base
flow have been substantial both before and since the appearance of man on
earth. Many nutrients dissolved in base flow are, then, of geological origin.
Losses of Non-adsorbed Substances (Group (III)--
Losses of soluble substances in surface runoff are small. Soluble
substances are found in very low quantities from small plots where direct
overland flow accounts for all surface runoff. With larger areas, subsurface
flow, which can transport significant amounts of nitrates and other soluble
substances, may emerge as surface flow and contribute to the pollutant load
in surface flow (Beasley, 1976; Chichester, 1976). Contributions of nitrates
to surface waters by base flow and interflow are at least five times as high
as direct overland flow contributions in humid regions (Jackson et^ al. 1973;
Campbell, 1976; Booram and Asmussen, 1976). The subsurface flow will contain
lower amounts of soluble nutrients when the field is covered with an actively
growing crop. The plants will absorb most of the nutrients and percolation
will be decreased due to evaporation (Chichester, 1977,Kilmer, 1978 personal
communication).
28
-------
Timing of Substance Movement--
Pesticide losses in surface runoff were found to be highest in the first
three storms after surface application (Pionke and Chesters, 1973; Edwards,
1972). For moderately adsorbed substances (atrazine, trifluralin, 2,4-D),
the dissolved losses were found to be highest in the first storm after
application. The concentration of substances in the sediment decreased with
time after application, but much less rapidly than did the concentration of
the dissolved substances (Smith et aK 1978). Baker and Johnson (1977) and
Baldwin et al,U975) concurred with the conclusion that weakly adsorbed
pollutants in runoff water decrease rapidly with time. Caro et aL (1973)
reported a threefold increase in concentration of carbofuran la weakly
adsorbed pesticide) in runoff in the second storm after application over
that from the first storm. However, the probable cause of this anomaly was
that the pesticide granules were not completely dissolved during the first
storm.
A strongly adsorbed pesticide can be found for years after application
in the overland flow, depending on its susceptibility to degradation. Harrold
and Edwards (1970) and Nicholson et aL (1966) among others, showed this to be
true for DDT, while studies at Watkinsville confirmed it for paraquat (Smith
et_al_. 1978)
EFFECTS OF SWCPs ON POLLUTANT LOSSES
No-Tillage
Since the introduction of the nonresidual herbicides, no-tillage has
become more popular each year. Runoff from no-till fields is in most cases
significantly lower than from conventionally tilled fields. During five
years of monitoring in Ohio, surface runoff volume averaged 1 mm per season
for no-till plots while averaging 11 mm per season on conventionally tilled
plots (Harrold et al. 1970). Thirty-six hours after a 2.6-cm storm, soil
moisture to a depth~of 105 cm was 2.5 cm greater on no-till plots than on
conventional plots, indicating significantly higher water infiltration under
no-till (Blevins et^ aK 1977). McDowell (1978, personal communication) found
a 90% reduction in~overland flow for no-till corn versus conventionally
tilled corn. The reduced runoff rate is a direct outcome of the higher
infiltration rate (Trichell_et_al. 1968). The higher water transport rate
between the surface and subsurface may also cause nitrate ions to infil-
trate deeper under no-till than under conventional till (MacMahon and Thomas,
1976; Steenhuis, 1977; Tyler and Thomas, 1977).
No-till soils have a higher conductivity due to both reduced organic matter
breakdown and the greater number of pores with ports to the surface. Only
channels which are open to the surface can carry tension-free water from the
surface to the subsurface (Ehlers, 1973; Dixon and Peterson, 1971). Such
ports are formed by earthworms and roots. Increased earthworm activity has
been observed in soils under a mulch cover (Graff, 1969; Tetioa et_ al_, 1950,
Schwerdle, 1969; Ehlers, 1973). If left undisturbed, these channels greatly
increased conductivity.
No-till soils are loosened only locally and superficially; yet they have
to bear the normal load of traffic in the field. Thus, no-till topsoils tend
29
-------
to be more densely compacted than plowed topsoils (Free, 1970; Harrold et al.
1970; Baeumer et al. 1973). This higher density does not decrease infiltra-
tion rate since~the increase in the number of surface connected pores out-
weighs the effect of higher density under no-till. Baeumer and Bakermans
(1973) observed that no-till does not result in denser soil if heavy clay is
involved.
Shallow soils with an impeding layer at shallow depths form an exception
to the rule that no-till decreases runoff compared to conventional tillage
(Smith et_ al_. 1974). Runoff starts on these shallow soils when all soil pores
are filTed with water and there is no more place for water to infiltrate. On
these soils, total pore space and not the number of surface connected pores,
therefore, determines the runoff volume. Moreover, the effect of increased
bulk density under no-till is not offset by the effect of increased earthworm
activity; the bulk density is higher and so is the runoff volume.
Numerous studies (Moldenhauer et^ aK 1971; Harrold et^ aK 1970; Meyer
and Mannering, 1968; Shanholtz and Lillard, 1969; McDowell and Grissinger,
1976) show that no-till is one of the most effective ways to reduce soil
loss from a field. Accumulation of organic matter near the surface of the
untilled soil causes significantly higher stability of soil aggregates
(Blevins et_ a^. 1977). This, as well as the absorption of energy by the
impact of falling raindrops and the impedance to water flow by surface trash,
reduces runoff volume and velocity. Since lime, organic matter, nutrients and
pesticides are applied on the surface, the concentration of nutrients and
pesticides in the surface soil is higher in no-till than in conventionally
tilled soil because in the first system, these elements are not incorporated
into the soil.
Without plowing, surface-applied substances (except soluble substances)
are very slowly incorporated into the soil. Sediment, originating from the
enriched top layer, has a higher concentration of strongly adsorbed sub-
stances such as paraquat, DDT and total N and total P than when the chemicals
were incorporated in the soil (McDowell, 1978 personal communication; Smith
et aK 1974; Smith et_ a^. 1978). This enrichment is offset by the dramatic
reduction in soil loss and runoff such that the overall result is a decrease
in total (both aqueous and adsorbed and solid phase) losses of the strongly
adsorbed substances.
Losses of plant nutrients (soluble phosphorus and nitrates) in runoff
may be higher under no-till than under conventional till due to increased
leaching of dead plant residues on the surface of the soil especially during
the winter if freezing temperatures break down plant cells, releasing nutrients
(Barisas et_ al_. 1978; Johnson et_ al_. 1977; Smith e^ al_. 1974; Timmons et al.
1968). In northern areas where snow melt runoff forms a major portion~~of~~the
yearly runoff, phosphorus losses in the aqueous phase may outweigh the losses
in the particulate phase (Timmons et_ al. 1970; Harms et_ al. 1974; Ryden et
al. 1973).
The higher infiltration rate and lower runoff velocity both cause re-
duced runoff and erosion losses from no-till plots. Losses of strongly ad-
sorbed and solid phase pollutants, such as total phosphorus and organic ni-
trogen will, therefore, be significantly reduced. It is uncertain whether
30
-------
no-till can reduce los'ses of weakly adsorbed pesticides and plant nutrients
(such as dissolved phosphorus) significantly because of the increased plant
residue and sometimes higher pesticide use.
Conservation Tillage
Conservation tillage is generally somewhat more adaptable and more suited
to colder and wetter soils than no-till (Stewart et_ aK 1975). However, since
part of the crop residue is buried with conservation tillage, the benefit of
the remaining mulch on reduction of soil and runoff water losses will be less
than with no-till, but greater than with conventional tillage. Conservation
tillage will result in a moderate reduction in runoff volume due to increased
hydraulic conductivity. Soil loss is reduced by reduction in runoff volume
and velocity, and protection from splashing raindrops of the soil by mulch
cover.
The reduction in runoff volume for conservation tillage has been shown
by many workers. Onstad and Olson (1970) monitored two small watersheds in
South Dakota for moisture conservation. A 7.5-cm rainfall produced no sur-
face runoff with conservation tillage, but 2.0 cm of runoff with conventional
tillage. Burwell and Larson (1969), and Siemens and Oschwald (1976) showed
that various conservation tillage methods increase the time before runoff
starts. Laflen et^ aj^, (1978) showed that runoff amounts decreased as residue
cover increased. Harrold (1960 as quoted by Baver et^ al. 1972) found that
for a three-year average, the relativerrunoff losses from conventional tillage
were twice as high as those from a plow-plant system, a conservation tillage
system that deletes secondary tillage. Edwards (1972) found that increasing
tillage causes more runoff and less infiltration of water. Studies by Holt
_et_ al. (1968) and Falayi and Bouma (1975) showed that with moldboard plowing
alone, the initial infiltration and abstraction that takes place before
runoff starts are twice as large as when the soil is cultivated. When
moldboard plowing is combined with a secondary tillage operation such as
disking and harrowing compared to a plow-plant system, initial infiltra-
tion and abstraction are greatly decreased. A reduction of runoff volume for
conservation tillage is less likely to occur on the shallow stony soils in
the Northeast and claypan soils in the Midwest. In New York for example,
Free and Bay (1969) found that the runoff for conservation tillage was some-
what higher than for conventional tillage.
For both shallow and deep loess soils, increased surface residue slows
runoff water velocity. In Indiana, Meyer and Mannering (1968) found that
under steady-state conditions, runoff velocity decreased from 8 meters per
minute to 2 meters per minute if the mulch rate was increased from 0 to 2.5
tons per hectare. The reduced runoff velocity together with the protective
effect of mulch against raindrop impact decreased soil loss by a factor of
5 or higher depending on the amount of mulch on the surface (Amemiya, 1970;
Edwards, 1972; Free and Bay, 1969; Siemens and Oschwald, 1975; and Laflen
et al. 1978).
Nutrient losses from conservation tillage should be less than from no-
till because some mixing of fertilizer and soil occurs under a conservation
tillage system. Still, the dissolved plant nutrient concentration from a
31
-------
field covered with plant material can be higher than for conventional
tillage as shown by Johnson et_ aK (1977) and Barisas _et_ aJ_. (1978).
Losses cf organic phosphorus and organic nitrogen, which are associated
with the sediment phase, decrease with increase in cover (Barisas et al.
1978; Johnson et_ al_. 1977) Conservation tillage increases available
phosphorus concentration in runoff and sediment, due to leaching of plant
residues but at the same time, decreases runoff volume and sediment loss.
Thus, two independent processes determine the direction of the change in
loss of available phosphorus amounts of residue, and runoff and soil loss will
have an influence on the positive or negative effects conservation
tillage on nutrient losses.
The major carrier for strongly adsorbed pesticides is sediment. Thus,
tillage practices such as conservation tillage decrease loss of strongly
adsorbed pesticides in overland runoff. Baker et^ al. (1978) verified that
conservation tillage decreased phosphorus losses as well as losses of
strongly adsorbed insecticides.
Moderately adsorbed pesticides are transported primarily with the water,
and so reduction in runoff caused by conservation tillage will decrease
pesticide losses. However, the reduction in runoff is sometimes offset by
an increase in concentration of the pesticide in the runoff as shown by Baker
and co-workers (1977, 1978) for atrazine and DCBN. The pesticides presumably
leach out of the mulch layer. Thus conservation tillage is most effective
in reducing loss of soil, organic matter, and strongly adsorbed pesticides
to surface waters. It also has some potential to decrease losses of weakly
adsorbed pesticides, but this is dependent on how much pesticide concentra-
tion is increased by the amount of pesticides which is leached out of the
mulch layer.
Contouring, Ridge Planting, Listing, Graded Rows, Strip Cropping
These practices are grouped together because they are all based on the
same principle of creating barriers perpendicular to the natural flow path
of water. The primary results are an increase in volume of depression
storage and an increase in flow path length. The net results are decreased
runoff volume and water velocity. The principal difference between practices
will be discussed in relation to their abilities to control water movement
and pollutant movement.
Water Movement
Contouring results in very low soil ridges. Weathering over the growing
season or intense storms which cause water to wash over the contours can
decrease the effectiveness of contouring in reducing runoff volume (Wischmeier
and Smith, 1965; Jamison et^ aJL 1968). If the slope is too long, contours
tend to be ineffective. Contouring in combination with other SWCPs is,
therefore, recommended.
On highly credible soil in southern Mississippi, contoured plots had
only 45 percent as much runoff as straight row cropping (Saxton and Spomer,
1968). On the clay pan soil, contouring was effective in reducing runoff in
32
-------
dry years only. Jamison et al.(1968) and Stallings (1945a) summarized and
compared older studies on contouring versus straight row in various parts
of the country. Reduction in yearly runoff ranged from 20 to 80 percent
depending on climate and soil type. The success of contour farming in
increasing infiltration of water is greater on permeable soils than on
clayey soils as shown by Baver et^ ah (1972).
Compared with contouring, graded rows have less surface storage. Thus,
reduction in runoff volume, especially on the heavier soils should be small.
This very small or no reduction in runoff was observed by Moldenhauer et al.
(1971).
Contour listing and ridge planting are advocated in the less humid areas
of the Midwest. These practices have higher barriers than contouring and are
also more resistant to breaking due to the residue left on the land. Ridge
planting and contour listing have greater volumes of depression storage and,
therefore, greater potential for conserving water than contouring does. Ritter
(1971) reported a 61 percent decrease in runoff volume for a ridge watershed
versus contouring. Laflen et^ ah (1978) and Baker et_ al^ (1978) reported that
for three hours of simulated rain of 6.4 cm per hour, the percentage of rain-
fall resulting in runoff for conventional and ridge planting were 70 and 54
respectively. For the same region but larger watersheds, the four year
average runoff for conventional tillage was 13% and for ridge planting, 10% ,
of the natural rainfall.
Strip cropping is based on the principle that checking the downhill flow
of water will reduce its capacity both to pick up soil particles and to carry
them in suspension (Bennett, 1939). Flowpath and depression storage is not
greatly affected by strip cropping and runoff volume will not be affected.
This is confirmed by results summarized by Stallings (1945b) from various
parts of the United States. Only in New York on a shallow soil did strip
cropping reduce runoff, probably due to increased evaporation and consequently,
greater effective water-free pore space.
Soil Loss
Contouring, ridge planting, contour listing and strip cropping are all
effective in reducing erosion as long as ridges are not broken by runoff
(Stewart et_ al. 1976) .
Data by Stallings (1945a) showed that contour farming is much more
effective in reducing soil loss than runoff volume.
Jamison et al.(1968) showed that contouring is ineffective in reducing
soil loss on long and/or steep slopes unless combined with other SWCPs,
because flow concentrates under these conditions. Graded rows have the same
effect as contour farming for small storms, but might be more effective for
large storms. Large storms cause contour rows to break over. Ridge planting
and contour listing are the most effective in keeping the soil on the land.
Not only are the ridges less liable to break, but also the velocity of the
water is reduced by mulch in the furrows.
33
-------
Although strip cropping has only minimal effect on runoff volume, it is
effective in reducing soil movement. Stalling (1945a) reported soil losses for
strip cropping as averaging one quarter of that of similar areas not in
strip cropping. The most effective system of strip cropping employs a meadow
strip between the tilled strips (Baver et^ al^ 1972).
Pollutant Movement
Losses of strongly adsorbed substances will be reduced by all these
practices as long as they retard the soil movement. Contouring would not
be a recommended practice on long and steep slopes, as the contours tend -to
break and become ineffective under these conditions. They might do very
well in combination with other practices and on short and moderately sloping
land. Ridge planting, contouring and strip cropping are probably the most
effective in reducing loss of moderately and strongly adsorbed substances
(Baker et^ al. 1978). Losses of moderately adsorbed substances such as
atrazine and DCBN were reduced by ridge planting (Ritter, 1971; Baker et al.
1977, 1978). This was expected because ridge planting, like contour listing
reduces runoff volume. Strip cropping and graded rows have only a minor
impact on loss of this group of moderately adsorbed pesticides. Contouring
might be effective if applied on short slopes. The soluble pollutants are
not affected greatly by any of these practices.
Sod-Based Rotations
Rotating row crops with sod crops improves soil structure relative to
continuous row-cropping. Plowing the sod under helps increase soil organic
matter, while the dense root system of a sod crop increases soil porosity.
Rotation studies with cotton, sorghum, a number of vegetables, and clover
showed a slight increase in organic matter for those.rotations containing
clover (Gerard et al. 1962). Other soil properties were not significantly
affected. Because of the large amount of sand in the soil, no aggregation
occurred.
It is well established that surface runoff from a sod crop will generally
be considerably less than runoff from a row crop (Bennett, 1939). This effect
is due mainly to the increase in soil porosity. The benefit can continue
for the first few years of row crop after the sod crop is plowed under.
Experiments at the University of Missouri showed that surface runoff and soil
losses in corn following sod were only one-third and one-fifth, respectively,
of the amounts from continuous corn (Miller and Krusekopf, 1932). Other
tests at Missouri showed continuous corn had twice the surface runoff and
several times as much soil loss as corn in a corn-wheat-clover rotation
(Bennett, 1939). Plant population densities of corn have increased con-
siderably since the 1930's, so the differences in runoff and soil loss be-
tween the two practices has probably decreased, but will still be considerable.
On a black-clay soil in Texas, a gradual decrease in soil organic matter was
noted when continuous sorghum replaced ten previous seasons of rotations
(Adams, 1974). In Georgia on Cecil sandy loam with 7 percent slopes and a
three-year rotation of oats-lespedeza, volunteer lespedeza and cotton, the
turned-under residues of the lespedeza and oats effectively reduced runoff
and erosion during the following cotton year (Hendrickson and Barnett, 1963).
34
-------
In addition to affecting soil and water movement, rotations have a
potential to limit pollution by their effect on soil fertility and pests and
diseases in the crops. The increased organic matter and the natural elimina-
tion of pests and diseases obviate much of the fertilizer and pesticide usage.
Considering all these benefits, rotation appears to be an excellent tool to
combat edge of field losses of nutrients and pesticides. One exception
might be formed by rotations which include legumes. Recent research has
shown that nitrogen losses can be greatly increased in this system (Kilmer,
1978, personal communication).
Cover Crops
This practice has its greatest effect during the non-growing season. It
reduces direct runoff and soil loss in the fall, winter, and spring,(Uhland
and Hendrickson, 1946; Stewart et al_. 197Sa). Cover crops may decrease
leaching of nitrates to the groundwater because of plant uptake. Studies
quoted by Stewart _et_ al,(1976) show that non-legume crops are especially
effective.
Cover crops might decrease the loss of strongly adsorbed pesticides
which usually degrade slowly and are still available in the soil during
the winter (Smith et^ al. 1978). Most of the moderately adsorbed pesticides
have short half-lives and by the time the cover crop becomes effective in
reducing runoff, pesticide concentration in the water is usually negligible
and the effect on total pesticide loss is minimal. Studies by Smith et al.
(1978) showed this to be true for atrazine, trifluralin and 2,4-D in
Watkinsville, Georgia.
Terraces
Installation of terraces results in decreased velocity of water flow as
a consequence of the flatter slope along the terra'ce as well as the increased
flow path length due to lateral movement through water outlets (Spomer et al.
1971; Laflen et_ al_. 1972). Terraces may be divided into three groups on the
basis of functionT level, graded, and impoundment. Level terraces are used
to conserve moisture, whereas the graded terraces are used for orderly dis-
posal of surplus water during times of excess rainfall. Impoundment terraces
cause an increase in surface detention storage and, thus, a lower peak
runoff (Hanway and Laflen, 1974).
Terraces without a water storage facility are not very effective in
controlling runoff volume on the less permeable soils (Stallings, 1945b;
Baver et al. 1972). Simulation techniques indicated that for impoundment
terraces", ~30% of the water may infiltrate due to ponding (Laflen et al. 1972).
Terraces are more effective in reducing erosion that in reducing runoff.
Stallings (1945b) reported up to a 100% reduction in soil loss for an experi-
ment in the eastern United States. More recently, Carter et _al_ (1968) found
that soil losses were reduced by 50% due to terracing on highly erodable
loamy soils of the Southern Mississippi Valley. On poorly managed soils
averaging 135 MT/ha annual soil loss, terracing reduced losses to 2.3 MT/ha
(Saxton and Spomer, 1968).
35
-------
Because terraces retain soil on the land, they considerably reduce loss-
es of strongly adsorbed substances such as paraquat and total phosphorus
(Smith e^t al. 1978). However, since terracing does not reduce overland flow
greatly, its impact on losses of moderately adsorbed substances is less pro-
nounced, especially on the less permeable soils. Movement of moderately ad-
sorbed pollutants is affected more by the time interval between application
and first rainfall than by terracing (Smith et^ a^. 1978).
In the case of impoundment terraces, where both runoff volume and sedi-
ment loss are lowered, losses of both strongly and moderately adsorbed pollu-
tants are significantly reduced (Hanway and Laflen, 1974). The reduced run-
off volume increased the subsurface flow of water. This subsurface water,
having percolated through the soil, had a higher inorganic nitrogen content
than the overland runoff (Hanway and Laflen, 1974). Impoundment terraces,
therefore, actually increase the loss of soluble inorganic nitrogen to the
groundwater below the terraced field.
Grassed Waterways and Diversions
Grassed waterways and diversions collect water and transport it to a
point downhill. They are typically designed in such a way that the soil in
the waterway does not erode nor does the sediment (from the upland area) set-
tle out.
The construction of a grassed waterway is usually necessitated as a re-
sult of the implementation of a SWCP or of the occurrence of a natural fea-
ture which concentrates runoff water and necessitates a runoff disposal sys-
tem. Diversions collect overland flow over a wide area of land when no ero-
sion control measures are employed. Both diversions and waterways concen- •
trate the flow of the water, giving the water less chance to infiltrate.
Therefore, these two practices increase runoff volume rather than decrease
it.
Waterways prevent formation of gullies in the field (Bennett, 1939).
Soil loss from gullies may be high and consist mainly of subsoil. Subsoil
is usually low in organic matter and contains little available phosphorus
(Ryden et aJL 1973) and no pesticides. Little research has been done on the
effect of grassed waterways on agricultural land, but recent researchers
have looked at the use of grassed waterways for treating runoff water from
feed and barnlots (Dickey et a^. 1977; Edwards et^ al. 1970; Bendixen et al.
1969). Excellent results were obtained in reducing the total pollutant load
of feed and barnlots to surface waters because most of the water and dis-
solved pollutants were infiltrated before the end of the waterway. The
impression one might obtain from such studies, that pollutant load from
agricultural land would also be reduced, is probably incorrect because of
the greater volume of runoff from the agricultural fields in the grassed
waterway. This greater volume cannot infiltrate. A study by Asmussen et^ al.
(1977) where 2, 4-D was effectively reduced by a grassed waterway should be
viewed in the same light. Runoff from their test plots in the grassed
waterway was much less than what could be expected under normal farm con- '
ditions.
36
-------
Until more research is done, it seems at present time not justified to
conclude that grassed waterways or diversions control any pollutant other
than sediment originating from gully erosion.
Buffer Strips
The concept of using a grass buffer strip to remove polluting substances
from agricultural runoff is a recent development. Historically, pasture
strips, meadow strips, hay or small grain strips were planted as a SWCP.
Research on buffer strips for treating agricultural runoff is limited. There-
fore, an analogy between buffer strips and strip cropping is made. In this
analogy, grassed buffer strips are considered a strip cropping scheme where
only the bottom strip is retained. Since the effect of strip cropping on
runoff volume was minimal, buffer strips should also have no remarkable
effect on runoff. Thus, the effect on moderately adsorbed pollutants is
minimal. If formation of channels can be prevented in the buffer zones,
sediment and associated pollutants may settle out before they reach the
stream. Studies by Doyle et_ al. (1974, 1977) indicate that buffer strips of
4 to 15 m might reduce the'pollutant load on small plots treated with man-
ure. It is uncertain whether these studies can be extrapolated to larg-
er areas. Therefore, more research is required before buffer strips can
be recommended as a SWCP for reducing pollutant load to streams.
CONCLUSIONS AND SUMMARY
This section analyzed SWCPs as to their effect on three types of pollu-
tants: strongly adsorbed, moderately adsorbed, and soluble. Table 4-6 is
a summary of the effects of SWCPs on pollutant movement. Gneralizations
have been made, so that for any particular set of environmental conditions
and pollutant characteristics, the effectiveness of a practice might be
different from that indicated in the table.
The amounts of strongly adsorbed and solid phase pollutants will be
reduced by those practices which reduce sediment loss. Concentrations of
moderately adsorbed substances are lower in the water than on the sediment.
However, the volume of runoff water removed is much greater than the amount
of sediment. Therefore, moderately adsorbed substances are carried mostly
with-the water. SWCPs which control sediment have only a slight effect on
total loss of this group of substances. SWCPs which control total amount
of runoff volume should reduce loss of moderately adsorbed substances. Loss
of dissolved substances in surface runoff is usually lower than in subsurface
flow. Reduced runoff might increase the subsurface flow and thus increase
the losses of soluble substances to the interflow and groundwater. If
reduced runoff is obtained by increased evaporation and crop uptake, the losses
of non-adsorbed nutrients will be decreased.
ACKNOWLEDGEMENTS
The authors are especially grateful to the following persons for their
help in preparing and reviewing this section: Erick Smith, Lee Jacobowitz,
Hanneke DeLancey, Rick Schwartz, Mathew Lorber, Sue Roedel and Marian Ogden.
37
-------
TABLE 4-6. EFFECTIVENESS OF SOIL AND WATER CONSERVATION PRACTICES IN CONTROLLING POLLUTANTS
1
Control of
SWCP
Strongly Adsorbed
Soil, Organic N, Organic P
Paraquat
Available Phosphorus
Moderately
Adsorbed
Non-Adsorbed
CO
oo
Effective
Not Effective
or Very
Slightly
Effective
no tillage
conservation tillage
ridge planting
sod-based rotations
cover crops2
sod-based rotations
ridge planting
sod-based rota-
tions
Moderately
Effective
Slightly
Effective
graded rows
contour listing
terraces
contour farming
filter strips
diversions
graded rows
terraces
no tillage
conservation tillage
contour farming
ridge planting
contour listing
cover crops
diversions
filter strips
contour listing
no tillage
conservation
tillage
contour farming
terraces
sod-based
rotations
contour farming
ridge planting
contour listing
grassed waterway
drainage
•7
grassed waterway
drainage
graded rows
no tillage
cover crops conservation
diversions tillage
grassed waterway graded rows
filter strips cover crops
drainage diversions
terraces
grassed waterway
filter strips
drainage
These evaluations have been made for general situations. The site specific effectiveness of a
particular practice as it fits into a total farming system might be different than indicated
"Not effective for pesticides .
In combination with other practices, grassed waterways are effective.
-------
SECTION 5
THE EFFECTS OF SOIL AND WATER
CONSERVATION PRACTICES ON SEDIMENT
Michael F. Walter, Tanuno S. Steenhuis and Hanneke P. DeLancey
Soil can be transported by overland runoff and if it reaches a surface
water body it can become a water pollutant. Soil is a potential water
pollutant because (1) as sediment its mass volume often reduces stream and
lake capacities; (2) it can cause high levels of turbidity and settle out on
stream bottoms upsetting the stream or lake ecosystem; and (3) sediment can
be associated with adsorbed substances which are transported on soil particles,
The association of substances with sediment is important because some SWCPs
are more effective at reducing the sediment losses than they are at reducing
total runoff.
One recent national estimate of natural levels of sediment moving into
surface water was 30 percent of the total sediment load (USDA-SCS, 1978).
According to this source, approximately 50 percent of all sediment originated
from cropland. Other major sources of erosion include that from streambanks,
road ditches, construction sites, urban areas, and forests. The erosion from
some of these other sources, notably streambanks and road ditches, have very
high rates of sediment delivery. Table 5-1 illustrates the relative distri-
bution of sediment from various sources. While in this project, analysis was
principally centered on losses of eroded soil from fields, the relationships
between cropland erosion and stream sediment levels cannot be ignored. Back-
ground levels of sediment and suspended soil particles are an important
consideration in evaluating benefits of sediment or suspended soil particle
control from cropland because in some watersheds these naturally occurring
levels will limit potential water uses.
Consideration of soil loss in runoff is important not only because soil
itself is a potential water pollutant but also because some pesticides and
nutrients are associated with soil losses. Substances adsorbed to soil move
with it in the runoff. Therefore, control of soil loss may directly control
losses of adsorbed substances. The interrelatiqnsip between the character
of sediment that is associated with adsorbed substances and that controlled
39
-------
TABLE 5-1. RELATIVE SEDIMENT LOADS FROM VARIOUS SOURCES (AFTER USDA-SCS, 1978)
Geologic 30 percent
Cropland - Sheet and Rill 50 percent
Grassland and Forests 10 percent
Roadside, Streambank, Construction and Mining 10 percent
by various SWCPs can provide a means to estimate the effectiveness of a
specific SWCP to control adsorbed substances.
In humid regions, SWCPs have traditionally been designed to control soil
erosion. Since some form of soil erosion is the first step in the production
of sediment, one could logically expect SWCPs to potentially control soil loss
to surface water from areas where the practices are used. One could also
extend that reasoning to the control of the movement of soil-associated
substances to surface water. But to do so requires the assumption that
reduction in gross soil erosion leads directly to similar reductions of
sediment and sediment-adsorbed substances reaching the surface waters. This
line of reasoning has been the starting point for much of the evaluation of the
effects of SWCPs on sediment and substances adsorbed to soil in sediment.
However, the factors used to evaluate the effectiveness of SWCPs for erosion
control are often different than those needed for analysis of effects on sedi-
ment and sediment-associated substances. For example, in addition to the
average quantity of soil that moves, information is needed on when it moves,
how far it moves, from where it originates and what its character is (e.g.,
adsorptive properties). These and other factors are considered in evaluating
the potential effects of SWCP on sediment yield and sediment associated
substances.
Sediment Related Problems
Estimated annual sediment yield in the United States is 1 to 2 billion
metric tons (ASCE, 1977; USDA-SCS, 1978). However, sediment load estimates
must be made for a specific watershed to be meaningful in terms of water
quality. Most watersheds do not have such data although sediment loads to
a number of reservoirs are published (Dendy and Champion, 1978).
Reduction in channel capacity and reservoir storage resulting in increased
flooding and reduced water supplies is perhaps the most serious detrimental
consequence of sediment. A survey in the late 1940!s indicated that more than
33 percent of the water supply reservoirs in midwestern states would have to
be replaced in less than 50 years due to sedimentation (Beasley, 1972). Many
reservoirs are filling at a rate of 5 percent of their capacity per year.
Reservoirs with capacities of 123,000 m or less are filling at a rate of 2.7
percent annually. The median storage depletion for all U.S. reservoirs is
1.5 percent per year (McDowell and Grissinger, .1976 after Dendy et al. 1973).
40
-------
One survey of several hundred reservoirs showed that the rate of sedimentation
™ smal1/eSerVOirS 1S mch MSher than in largg reservoirs (Beasley, 1972).
The added cost of providing storage in reservoirs for sedimentation can be
significant. Thirty percent or more of the total storage capacity of a
reservoir can be designed for sedimentation. Sediment removal from streams,
™fV01oS and harbors costs $250,000,000 annually by one estimate (ASCE,
1S/7J. Reservoirs which fill with sediment must often be replaced at an even
greater cost than initially designed for because of the need to go to second-
ary sites.
Flood damage which is a direct result of reduced stream capacity due to
sediment from agricultural land is difficult to identify. Little quantitative
information seems to exist on this topic. Presumably flood damage is increased
due to channel clogging by sediment. Estimated losses in the U.S. from sedi-
ment and associated flood water damages are one billion dollars annually
(Nelson, 1968 after McDowell and Grissinger, 1976). Other sediment related
problems include reduced recreational value of water, increased maintenance
cost for navigable channels, and drainage problems where the sediment is
deposited.
Most of the sediment related problems discussed thus far in this section
are a result of sediment accumulation over a relatively long period of time.
However, other problems have more short-term impacts. Soil pollutes by muddy-
ing the water, inhibiting photosynthesis and increasing oxygen demand (Hartman
et^aj^. 1977; Mulkey and Falco, 1977). Sparks (1977) lists fish kills as an
important problem that results from, sedimentation. Silt can clog the gills
of adult fish and settle on the spawning beds, destroying the fish eggs. Sparks
also reports that sediment can reduce the dissolved oxygen levels of streams
which will adversely affect aquatic life. Table 5^2 lists some of the major
impacts of suspended soil in freshwater ecosystems. Suspended material, such
as clay or organic matter, can increase the cost of obtaining a suitable water
TABLE 5-2. CLASSIFICATION OF SUSPENDED SOLIDS AND THEIR PROBABLE MAJOR
IMPACTS ON FRESHWATER ECOSYSTEMS (MULKEY AND FALCO, 1977 AFTER
SORENSEN.ET AL. 1977)
Biochemical, Chemical
Sediment Type and Physical Effects Biological Effects
Clay, Silt, Sand Sedimentation, erosion and Respiratory interfer-
physical abrasion, turbidity ence, habitat destruc-
(light penetration, habitat tion, light limitation
change)
Natural Organic Matter Sedimentation, dissolved Food sources, dissolved
oxygen utilization oxygen effects
supply for some users. All domestic water suppliers and some industrial users
of water require that the water be relatively free of soil particles. Indus-
tries dealing with food processing must treat sediment-laden water before use.
41
-------
The Environmental Protection Agency has proposed that sediment loads do not
exceed 80 mg/1 for the protection of aquatic life (EPA, 1973). The above
listed problems are presumably from high suspended soil concentrations, which
are the result of single events or a series of events over a relatively short
period. Where such problems exist, analysis is necessary on a storm event
basis.
Soil as a Carrier of Water Pollutants
Those compounds that are carried by soil are usually associated with the
colloidal fractions of the sediment. The smaller and lighter the soil
particle to which substances are adsorbed, the greater the transportability of
adsorbed pollutants per unit soil weight.
Most soil particles have a negative charge at the surface, and will adsorb
cations. Adsorption of a particular cation will depend on its concentration
and type. The, greater the concentration of a particular cation, the more of
it will be adsorbed. A cation will be adsorbed preferentially if it has a high
valence and/or small size. Organic matter is perhaps even more important as
a substance carrier, but the adsorption mechanisms are not necessarily the
same as they are for clay.
Nutrients are, besides potentially being adsorbed to soil, an intrinsic
part of the soil. Even granite rock is composed of about 0.3 percent P2 0$
(Black, 1958). For cropping purposes the total quantity of a nutrient in
soil is not a particularly useful measurement (Black, 1958). Only the
quantity of a nutrient that is available to a crop is generally of interest.
Whether nutrients in sediment cause eutrophication response in a manner
analogous to field crops is not clear. Early research showed a correlation
between total phosphorus loading and increased levels of eutrophication
(Vollenweider, 1968). Some researchers who have studied the question more
recently argue that the real cause of increased eutrophication is soluble or
available nutrients (Lee et al. 1978). Very likely the character of the
receiving water (e.g., deep versus shallow lake, low versus high pH) will
influence the degree of eutrophication resulting from one form of a nutrient
to another. But an understanding of the origin of nutrients causing a water
pollution problem is critical to an effective solution, because total nutrient
loading is often best controlled by reducing sediment losses while available
nutrients may not be. Available forms of nutrients are often partly adsorbed
to soil particles and are partly in solution. Their control may require run-
off controls or better source (e.g., fertilizer) management. Although sediment
is typically made up of about 1 kg nitrogen and 0.8 kg of phosphorus per ton
(ASCE, 1977), the portions of those nutrients in an available form may be
much less, so sediment control may reduce total nutrient losses without really
controlling available nutrient forms. Sediment is often higher in clay,
silt and organic matter than the soil from which it originated (Massey et al.
1953) . This phenomenon and the fact that polluting substances have a greater
affinity for the smaller particles cause the erosion process to remove the
polluting substances preferentially. These two factors; the adsorption
behavior of the polluting substances and the sediment characteristics are
therefore very important in the evaluation of the effectiveness of SWCPs on
42
-------
nonpoint source pollution control. Adsorption behavior is discussed in
Sections 6 and 7 and sediment characteristics are considered later in this
section.
Sediments also appear to have a dampening influence on nitrogen and
phosphorus concentrations in surface waters (Holt et al. 1970). Glymph (1975)
has reported phosphorus adsorption from sewage effluent by sediment. Holt
e£ al^. (1970) concluded that adsorption of anions such as phosphate could be
an important reaction which occurs in natural colloidal suspensions. These
researchers based their conclusion in part on work by Rich (1968) and Chao
et a*• (1965) . Sediments have been shown to remove cesium-137, a major radio-
active contaminant of natural waters (Lomenick and Tamura, 1965). Frink (1969)
found that potassium was fixed by sediment in lakes. Holt et^ a\_. (1970)
refer to adsorption of lindane and parathion on the clay-organic" complex as
"chemical scavenging" of uncharged pollutants in natural waters.
Factors Affecting Erosion and Sediment Transport
Since soil erosion is the first step in producing sediment it is fair to
assume that if erosion is controlled sediment will be also. Erosion is very
localized so that its control must be field oriented, but sediment can be
controlled anywhere between source and sink. In many, if not most instances,
erosion control will be the mechanism used to control sediment. But the
careful selection of which control practices to use and which eroding soil to
control will be necessary for cost-effective sediment control. This selection
process requires consideration of the area location relative to sediment
polluted water bodies and the type of erosion process producing the pollutant.
The sediment-causing process includes soil detachment, transport, and
deposition, and each of these processes has its preference for particular
particle characteristics. Soil detachment can occur from either raindrop
impact or shear forces of flowing water. Raindrop impact, however, is the
primary mechanism responsible for soil detachment at the field surface (Baver
et al. 1972; Foster and Meyer, 1977). A number of naturally occurring
pFoperties of soil make some more susceptible to detachment. Primary clay
particles are relatively less easily detached than sand particles. Particles
in the range of 150-250um (fine sand) are most easily detached by raindrop
splash. Eckern and Muckenhirn (1947), Ellison (1947), Bisal (1960), Bubenzer
and Jones (1971), Garriels e_t al_. (1974) , and Ghadiri and Payne (1977) among
others have researched the relationship between rainfall and soil detachment.
All concluded that rainfall kinetic energy and drop velocity were important
parameters. In general, the higher intensity storms have much higher soil
detachment potential than the less intensive storms, even if the storm volume
is the same. Free (1960) found that increased rain intensity increased the
splash losses of an aggregated soil more than the splash losses from a sandy
soil, indicating that a greater percentage of sand is detached at lower
intensities. Laws (1941) related drop velocity to drop size and fall height.
This relationship has been widely used since soil splash is roughly equal to
the product of the drop diameter times the velocity squared (Ghadiri and Payne,
1977) Laws and Parsons (1943) have shown that the mean drop size increases
with rainfall intensity which leads directly to the conclusion that the higher
43
-------
the rainfall intensity the greater will be the kinetic energy of that event.
Another feature of splashing raindrops is that they break down the aggre-
gates into smaller aggregates and primary particles, 'which are more vulnerable
to transport than the original aggregates. Detachment by shear forces occurs
mainly in areas where water is concentrated (e.g., rills) rather than over a
broad area (Foster and Meyer, 1975) . Flowing water will detach soil particles
if the detachment capacity exceeds the detachment resistance of the soil.
As with raindrop splash, fine sand particles are most easily detached. In
contrast to splashing raindrops, flowing water does not break down aggregates
into smaller units. Hjulstrom (1935) found that the critical flow velocity
for detachment of quartz spheres of clay particle size was about 200 cm/sec,
a velocity well above that expected in overland sheet flow. Fine sand size
particles on the other hand were detached at less than 20 cm/sec. Foster and
Meyer (1977) also describe soil detachment by flowing water in undercutting,
sloughing of sidewells, and cleanout of the dumped material.
The second phase of the sediment causing process is the transport of
detached soil particles to a water body. Raindrop splash on steep slopes can
transport significant amounts of soil downslope, but overland and channel flow
are the primary mechanisms for soil particle transport (Young and Wiersma,
1973). Soil particles or aggregates detached by raindrop impact are trans-
ported by shallow overland flow to rills. Rainfall significantly increases
the transport capacity of shallow overland flow by continuously lifting
particles up into the runoff flow (Foster and Meyer, 1977) . These particles
originate from a very thin layer at the surface and tend to be smaller, due to
selective sorting and aggregate breakdown, than the soil in situ. Once in
defined rills the flow will detach some additional soil but these particles
will be similar to the original soil with larger aggregates than those detached
by splash (Foster and Meyer, 1977). Once the sediment carrying capacity of
the rill flow is exceeded, deposition will begin to occur. Deposition usually
occurs due to a change in the carrying capacity of the flow as a result of
the decreased flow velocity. Frequently deposition, especially of large
particles, occurs long before the runoff reaches streams or lakes (Foster and
Meyer, 1977; Meyer and Harmon, 1977).
The transport of detached soil needs to be put into a time frame to be
meaningful. Soil particles are often detached and transported for only a
short distance in any one storm but after many storms ultimately reach streams.
Moore et al. (1977) refer to this phenomena as the "play fairs" law, implying
that sooner or later sediment deposited in concentrated flow channels will
reach streams or lakes. Trimble (1975) has reported that sediment
deposited in stream channels will move downstream as upstream loading
decreases. The idea that sediment deposition increases flow energy, resulting
in increased detachment and transport by the runoff, has been widely discussed.
McDowell and Grissinger (1976) reported that upland control measures that
decrease erosion more than runoff can cause downstream channel instability
problems. They discuss some of the theoretical reasons for this phenomenon.
For situations where dams have removed sediment load,increased streambank
erosion below the dam has occurred (Vanoni, 1977; Graf, 1971).
44
-------
Sediment is deposited in such a way that larger particles fall out first
(Foster and Muggins, 1977). The smallest particles of sediment load, the
colloidal component made up of fine clay and organic matter, may stay in
suspension for a very long time. So to control the losses of these particles,
reduction of their detachment is usually the best mechanism to use.
Rainfall and resulting runoff drive the sediment causing processes.
Variations in rainfall and runoff will affect sediment yield. Rainfall can be
characterized by its intensity, duration and frequency. SWCPs have no effect
on rainfall frequency or duration but these two factors will influence the
selection of a SWCP for a particular location. High intensity but relatively
short duration thunderstorms, for example, occur more frequently in the
midwest than in the northeast of the United States. While SWCPs do not actu-
ally change rainfall intensity, they can significantly reduce the energy of the
rainfall impact on the soil surface.
Sediment Characteristics
As described above sediment deposited in streams and lakes differs in
character from the original soil. Regardless of the source or detachment
mechanism, sediment is usually made up of a higher percentage of smaller
particles and aggregates than the original soil. This tendency is less
when rills and gullies develop in which bigger particles can be attached
and transported. As smaller particles have a greater adsorbing capacity,
adsorbed substances will be more associated with sediment originating from
splash or interrill erosion than with that from rill and especially from
gully erosion. This is even more nearly true when chemicals are surface
applied; a great part of the sediment which originates from gully erosion
comes from deeper layers in the soil profile where the chemicals did not
occur. If the chemicals are incorporated into the soil, their concentration is
approximate, even through the top layer of the soil profile. In this case, only
a higher content of small soil particles will increase the amount of adsorbed
chemicals in the sediment. To quantify the increased percentage of a specific
fraction in the sediment, the term "enrichment ratio" (ER) is used. This
term indicates the ratio between the particle fraction in the sediment and
that in the original soil. The enrichment ratio for clay (ERc) is usually
greater than one.
Although very little is published on organic matter enrichment (ER ),
the transport of organic matter is expected to be much the same as that Of
clay due to its characteristic low density. It may be that the detachment
processes are the same as well. Slater and Carleton (1942) ran two experi-
ments without vegetative coyer and found ERC equalled 1.11 for both and the
ERom equalled 1.12 in both cases. These results indicate..that clay and
organic matter are detached and transported in similar ways.
One of the complications in arriving at ERom is that organic debris can
be mixed with the sediment. Values of ERom were 20 to 140 percent greater
than ER in work reported by Slater and Carleton (1942) for plots with
vegetative cover. Residue, leaves, stems, etc., in the sediment will increase
45
-------
the ER . Relative to adsorbed substances these forms of organic matter
probab?y do not have the same character as clay or soil organic material.
Assuming that substances are not adsorbed to organic debris, which may not be
so for pesticides, one could reason that practices that rely on vegetative
cover or residue control mechanisms, and thus cause low ERC and ERom (where
ERom does not include organic debris), will be most effective, other things
being equal, in controlling adsorbed pollutants.
One can generalize that the enrichment ratio for clay or organic matter
increases with an increase in soil detachment by raindrop splash relative to
that by flow. These enrichment ratios will also increase as transport energy
decreases. One shortcoming of the enrichment ratio concept is that it does
not indicate the actual amount of clay in the sediment load. A low ERC for
sediment originating from a clayey soil may very well mean a much greater
amount of clay in the sediment than a high ERC from a sandy soil. A high ERC
for a small sediment load may entail less clay than a low ER for a big
sediment load.
Regional Variations in Erosion
Regions with extended periods of subfreezing temperatures are less
susceptible to sediment-producing processes during the winter than warmer
areas (Wischmeier, personal communication, 1978) . The temporal distribution
of intense storms as well as their magnitude varies widely over the humid
United States. This is reflected in the extreme regional variation in the
rainfall erosivity (R of the USLE) as shown in Figure 5^1. Rainfall erosivity
is particularly high in the southern United States. To reach the same level
of annual sediment control from two identical fields near, say Ely, Minnesota
and Bogalusa, Mississippi, the field near Bogalusa would require five times
as intensive controls as that near Ely.
Temporal Variations in Sediment Yield
The combination of erosion index and crop development reflects the seas-
onal variability in soil loss potential for a particular field. Table 5-3
illustrates the effect of different monthly .distributions of the erosion
index and crop development for three locations. This table indicates the
effect of the distribution of the erosion index and not the quantitative
erosivity for the locations. The potential for soil erosion is greatest
during the summer months at all three locations but the range between months
with high and low erosion potential is most extreme in Iowa.
Although the combination of rainfall patterns and crop management must be
used to have an overall view of variations in seasonal erosion potential,
some information can be drawn from rainfall patterns alone. Wischmeier and
Smith (1965) give rainfall distributions for all the regions covered by the
EPA project. Examples for five locations are given in Table 5-4. Although
the values do not reflect erosivity due to snow melt some conclusions can be
tentatively drawn from this table alone. Control measures for use in Iowa or
Minnesota would be designed for maximum effectiveness in the summer, Louisiana,
46
-------
300
/ /A400
^100
300
FIGURE 5-1. AVERAGE ANNUAL VALUES OF THE RAINFALL FACTOR, R (AFTER WISCHMEIER AND SMITH, 1965)
-------
TABLE 5-3. VARIATION OF SOIL LOSS POTENTIAL OVER THE YEAR AS A RESULT OF CROP
DEVELOPMENT AND EROSIVITY DISTRIBUTION FOR 3 LOCATIONS WHEN PLANTED
WITH CONTINUOUS CORN, SPRING PLOWED AND RESIDUES LEFT
J
F
M
A
M
J
J
A
S
0
N
D
Watkinsville,
GA
6.05
6.05
8.25
8.25
17.6
20.35
28.33
9.35
4.13
3.03
4.13
1.93
Aurora ,
NY
.00
.48
.48
.72
3.48
7.68
9.68
6.88
2.16
.8
.96
.48
Ames,
IA
.00
.48
.96
1.92
10.56
24.64
12.32
7.04
3.52
1.60
0.80
.00
TABLE 5-4. PERCENTAGE OF AVERAGE ANNUAL RAINFALL EROSIVITY INDEX VALUE EACH
MONTH (WISCHMEIER, PERSONAL COMMUNICATION, 1978)
City § State
Aurora ,
New York
Rochester,
Minnesota
Ames,
Iowa
Lafayette,
Louisiana
Watkinsville,
•"Georgia
J
0
0
0
6
6
F
2
1
1
7
6
M
2
1
2
8
8
A
3
3
4
12
8
M
10
10
12
11
10
J
16
23
29
11
12
J
22
24
17
12
19
A
21
22
17
8
11
S
13
11
11
6
7
0
5
3
5
5
5
N
4
1
2
8
4
D
2
1
0
6
4
on the other hand, appears to have fairly constant erosivity index. Potential
sediment problems could occur in the winter nearly as easily as any other time
so practices should be effective year round.
48
-------
From the previous discussion characterizing the sediment causing process
one would expect that the occurrence of soil loss would vary over the year.
This is illustrated by data from Georgia given in Table 5-5 for the ten events
resulting in the highest sediment yield from a 1.3-hectare watershed planted
in continuous corn from 1973 to 1975. During this three-year period all of
the ten highest sediment producing storms occurred during either May, June or
July. Spomer et^ al. (1976) found that this pattern also occurred in the upper
Midwest. Also consistent with work of Spomer et_ a^. is the fact that a few
TABLE 5-5. COMPARISON OF PEAK SEDIMENT YIELDING STORMS IN 1973-1975* P-2
WATERSHED, GEORGIA
Storm dataRainfallRunoffEl*Sediment YieldConcentration
(cm) (cm) (foottons/ha) (kg/ha) (mg/1)
May 28, 1973
June 11, 1975
June 6, 1973
June 27, 1974
May 23, 1973
July 27, 1974
June 13, 1973
July 24, 1975
July 8, 1974
July 13, 1975
10.9
7.1
3.1
10.8
1.9
7.2
2.7
4.3
4.8
2.7
7.4
4.0
2.2
4.2
0.6
4.5
0.7
2.4
1.9
0.8
119.8
123.2
38.2
207.3
10.8
288.1
11.2
40.6
37.5
46.6
8240
4122
1108
966
716
661
530
427
422
320
11,140
10,305
5,036
2,300
11,930
1,468
7,571
1,779
2,221
4,000
Calculated with USLE method.
+Runoff and sediment yield from all storm events in these 3 years are given in
Appendix C.
storms produced most of the sediment yield over the three-year period. The ten
biggest events produced a total soil loss of 17,500 kg/ha while the 62 smaller
storms accounted for only 860 kg/ha soil loss. Since May, June and July account
for only 41 percent of the total average annual erosivity index in Georgia
(see Table 5-4) one would not expect the ten highest events to be concentrated
in these three months. However, this watershed was planted to continuous corn
and the effect of crop and residue cover was minimal during May and June. The
combination of these two factors made these months highly vulnerable for soil
loss. Although the developing canopy cut the erosion potential considerably in
July, the erosion potential was still great in this month due to the very high
erosion index.
The storm of May 28, 1973, was the largest recorded in the three years at
Watkinsville, but only 0.1 cm more rain fell in that storm than in the one on
June 27, 1974. However, the sediment yield was almost an order of magnitude
greater for the first storm. Total runoff was greater for the first storm
49
-------
(7.4 cm vs. 4.2 cm). The kinetic energy (El) value of the second storm was
actually greater than the first. The reasons for the May 28th storm producing
so much more sediment than the June 27th storm were probably many, but a prime
factor was the method of tillage. In 1973 a moldboard plow was used while in
1974 the land was chisel plowed. The 1973 storm apparently resulted in greater
soil detachment due to reduced residue cover which caused increased raindrop
impact and greater surface sealing which in turn reduced infiltration and
increased runoff. An additional factor was that by June 27th some crop canopy
had developed which reduced the soil loss potential. This point is perhaps
better illustrated by comparing the similar storms of June 6 and July 8, 1973.
The sediment yield was over 60 percent less for the second storm for the
probable reason that the crop canopy was better developed by the July storm.
Predictive Models
Table 5-5 illustrates that no single parameter can be used to predict
sediment yield from a single storm. Wischmeier and Smith (1965) tried to
correlate a great many factors with total erosion and finally determined that
their proposed method of determining El values was best. The El parameter
used in the USLE is dependent only on rainfall characteristics. The USLE was
not intended to be used for single storm predictions so an anomaly that occurs
for some storms is that soil loss is predicted by the USLE even when no
surface runoff leaves the watershed. For sediment predictions, particularly
if the receiving water is a substantial distance from the eroded area, trans-
port energy estimation is quite important. Therefore, models have been
developed that include runoff volume and/or peak flow in an attempt to better
predict transport of sediment (See Appendix C).
The three models that were used are defined as the USLE Model, the
Williams Model, (Williams, 1975) and the Onstad-Foster Model (Onstad and Foster,
1975). The single storm El value was used in the USLE Model. The Williams
Model uses the same form as the USLE except the R value is replaced by a
single event energy term that is a function of total and peak runoffs. The
Onstad-Foster Model has an energy term that includes both the USLE single
event El value and total peak runoff. In most situations rainfall data
are available but runoff data are seldom readily available so total and peak
runoff must be calculated. Total and peak runoffs were calculated for the use
in the energy terms as shown in Appendix A.
Total and peak runoff were calculated for the Williams and Onstad-
Foster Models. The predictions for total and peak runoff were not very
accurate. This, of course, had its influence on the error in the predictions
of the soil loss with the Williams and Onstad-Foster models. Therefore, a
comparison was made between the predictions of these models with predicted and
with observed values for total and peak runoff. The results are presented
in Appendix C. All storms for which an energy term could be calculated
resulted in predicted sediment yields. Therefore, the USLE and Onstad-
Foster models predicted sediment yields for some storms that had no observed
or predicted runoff. The energy term in the Williams Model is zero if
either total or peak runoff input are zero. Soil loss as predicted
50
-------
by the three estimating techniques is compared with observed soil loss in
Appendix C, Table C-l. Except for the largest observed event, the USLE
predicted higher soil losses than the Onstad-Foster Model which was consist-
ently higher in its predictions than the Williams model. The USLE over
predicts losses from five of the ten largest events listed in Table C-l.
It overestimates losses on all of the other 62 events that occurred during
1973, 1974 and 1975.
Sediment Source Controls
The options for source control of soil loss are limited compared to those
for many other potential water pollutants. Fertilizer and pesticide appli-
cations can be reduced or eliminated but soil is a part of the natural land-
scape. However, since some soils due to soil character, topography, etc., are
potentially more susceptible than others to producing sediment, one control
mechanism might be to use these less intensively.
Land use changes to less intensive cropping will by and large reduce
sediment losses. But in many situations this alternative is not attractive.
There are ways, however, to reduce sediment yield significantly without
reducing the intensity of production. Sediment yield can be reduced by
selectively increasing cropping intensity on less susceptible fields while
decreasing it on those most susceptible to producing sediment. White and
Partenheimer (1978) have published results of a study in Pennsylvania that
indicated that this selection process might be done on a watershed basis,
but most likely it would be more easily implemented on a farm by farm basis.
That is, adjustments in field rotations or use within a farm unit while
maintaining total production would be easier to implement than adjustments
between farms. Ranking of fields as to potential sources of sediment can be
done by use of the field's soil erodibility factors (K), slope length factors
(LS), practice factors (P) and sediment delivery ratios (SDR). This approach
is illustrated for an example farm in New York State in Table 5-6. This
information illustrates some of the advantages of analysis on a larger than
field unit for more efficient sediment controls.
Even within a field one can to some degree select the origin of sediment
by choice of SWCP for sediment control. Those practices that control rain-
drop impact as mentioned earlier will protect soil at the very surface over
the entire field. However, they might have very little influence on rill or
gully erosion which tends to cover a more limited area but produces a sediment
source deeper in the soil profile. The relative volume of sediment originating
from interrill and rill erosion as compared to that which has its source in
gullies is difficult, if not impossible to quantify. But correct identifi-
cation of the problem source is critical in selecting effective practices.
EVALUATION OF SOIL AND WATER CONSERVATION PRACTICES
SWCP Control Mechanisms
Most SWCPs are designed to control either (1) detachment and transport or
(2) only transport of soil particles. Depending on the particular pollution
51
-------
problem resulting from sediment or suspended soil particles, choice of a SWCP
based on its control mechanism might or might not be important. If sedimen-
tation is reducing reservoir capacity for example, it makes little difference
what the character of the sediment is or where it came from. The important
element in this case is how much sedimentation occurs. In another situation
turbidity might be the critical problem; in which case control of fine
particles is the goal and will require a mechanism based on detachment control.
TABLE 5-6. FIELD MANAGEMENT FOR REDUCING SOIL LOSS ON AN EXAMPLE FARM IN NEW
YORK STATE
Field Size
Identification (Ha)
1 23.9
2 4.9
3 3.7
4 1.6
5 7.1
6 1.0
7 1.6
8 3.6
9 2.6
System of
Option 1*
c **
s
C . H.
g4 4
C „ H,
g4 4
C H
g4 4
C
s
Cs4H4
C
g
C . H.
g4 4
C . H.
g4 4
Crop Rotation
Option 2*
C
s
C
g
C
s
C A H,
s4 4
C , H.
g4 4
C . H.
g4 4
C . H.
g4 4
C . H.
s4 4
C . H.
s4 4
TOTALS
Average
Option
Q****
19.1
5.9
2.6
314.5
17.0
22.6
10.1
7.0
398.8
Annual Sediment Yield***
1 Option 2
(MT/field)
0
35.3
34.4
5.8
54.7
7.7
12.3
22.3
15.3
187.8
*0ption 1 and 2 have the same acreage in C , C and H
**C - Corn Silage Continuous
5 ***Based on USLE and SDR (Walter et al.,
C
- Corn Grain 4 Years
H - Hay 4 Years
C - Corn Grain Continuous
g
C . - Corn Silage 4 Years
1978)
****Slope essentially equal to zero so
erosion predicted to be zero
Table 5-7 lists some general pollutant forms and the potential for their
control by detachment or transport mechanisms. Choice of control mechanism
may depend partially on climate, soil character, topography and especially
field crop management.
Mechanisms for soil detachment control are often more effective for clay
soils because clay is more easily controlled at its source than in transport.
For rainfall splash control a broad area is typically covered by vegetative
52
-------
TABLE 5-7. POTENTIAL EFFECTIVENESS OF DETACHMENT AND TRANSPORT CONTROLS ON
SEVERAL CATEGORIES OP SOURCES OR CAUSES OF POLLUTION
Control Mechanism
Source or Cause
of Pollution Rainfall Transport Flow Transport
Clay High Low
Sand Medium High
Aggregates High Medium
Adsorbed Substances High Low
Organic Matter High Low
Suspended Loads High Low
Bed Loads Low High
Sediment Yield High High
canopy or crop residue. The requirement for broad coverage is sometimes in
conflict with the farm operations. Sod, for example, will prevent soil
detachment almost completely but many farmers have only limited use for sod.
These SWCPs are effective in reducing soil detachment by intercepting rain-
drops and dissipating their energy. The amount of reduction in raindrop
impact energy depends on canopy height and density when vegetative cover is
used to reduce detachment. Figure 5-2 indicates that, for SWCP that include
crop canopies, the erosivity as given by the USLE would be reduced 80 percent
or more when the ground is completely covered. Schottman (1978) found that
crop canopies tend to break up raindrops so that, for intense storms, effec-
tiveness is likely even greater than shown in Figure 5-2. At least for most
agricultural canopies, control of soil detachment is approximately directly
related to the percent of ground area which is covered by the canopy. If
the canopy is formed by sod forming crops, the soil erodibility is also
reduced, as a result of a sod-residual effect (Wischmeier, personal communi-
cation, 1978).
Crop residues might be considered an extreme situation for crop canopies,
as the canopy is at the soil surface. Comparable percent ground cover
residues or "mulches" are more effective at reducing erosion than crop
canopies as can be seen by comparing Figures 5-2 and 5-3. Residue reduces
both soil transport, and detachment by flow and raindrop impact.
To this point the discussion of crop canopies and residue has centered
on their effectiveness in reducing raindrop impact. However, ground cover has
other positive effects on the sediment-causing processes as well. Ground
cover decreases surface sealing and increases surface roughness which increases
infiltration and tends to slow overland runoff and reduce detachment by flow.
53
-------
I-
o
Ul
1.00
BJ -80
fc
O
< .60
O
cr
£.40
CC
I,
.* AVERAGE FALL HEIGHT
OF DROPS FROM CANOPY
(FEET)
0 20 40 60 80 100
% GROUND COVER BY CANOPY
FIGURE 5-2. EFFECT OF CROP CANOPY ON EFFECTIVE El
(ADAPTED FROM STEWART ET AL. 1975).
1.0 r
0.8
oc
0 0.6
O
0.4
0.2
0 20 40 60 80 100
% OF SURFACE COVERED BY MULCH
FIGURE 5-3. EFFECT OF PLANT-RESIDUE MULCH ON SOIL
LOSS (ADAPTED FROM STEWART ET AL. 1975)
54
-------
Runoff is the primary vehicle for transporting soil particles once they
are detached although under special situations raindrop splash will also
transport particles downslope. Transport controls include techniques of
reducing both runoff velocity and volume. In general, but not always, when
either is reduced so is the other.
The mechanisms used to reduce runoff velocity or volume can be either
macro or micro in scale. For example, graded terraces are constructed at the
base of a field strip. The primary purpose of graded terraces is to collect
and concentrate runoff so that it can be carried off on a gentle slope in a
designed channel. The mechanism to reduce slope is to increase flow path on
a macro scale. Graded rows have essentially the same purpose (i.e., increase
flow path) but act on a much smaller broader scale. Both of these SWCPs will
also increase temporary surface storage in the terrace channels or crop rows
which can lead to deposition, increased infiltration and many other inter-
related processes. Although few data are available on comparison of
terraces vs. contour tillage or graded rows, one might expect some differences
due to their areal distribution. Terraces do nothing or very little to prevent
interrill erosion but do reduce flow detachment and influence transport of soil
off a field. Contours do not reduce soil detachment by raindrop impact
either but they do cause deposition to occur much sooner in the process. The
whole point of controls that operate on soil particle transport is to cause
sedimentation to occur on the field. Presumably the sooner sedimentation
occurs the better.
SWCP Effectiveness
The best information where available, to judge effectiveness of practices
at reducing soil erosion is that used to quantify C and P factors on the USLE.
Table 5-8 lists some common cropping systems together with C values, the factor
used to reflect the effect of crop cover on annual soil erosion. Wischmeier
and Smith (1965) and Beasley (1972) list 128 different cropping systems and
Beasley gives a method for calculating a C value for any system. Published
values of the P factor are given in Table 5-9. Most SWCPs affect both soil
detachment and transport but each generally is primarily effective at one or
the other control mechanisms. The categorization of SWCP by control mechanism
is similar, though not the same as to division of cultural and structural
practices given in Section 4.
Soil Detachment Control Practices
On an average annual basis one can easily see the effectiveness of vege-
tative cover from Table 5^-8. By simply planting corn, even continuous corn
that is fall plowed with residue removed, the expected average soil erosion
is reduced 40 percent as compared to fallow conditions. Inclusion of a small
grain into a two-year rotation cuts the potential average erosion in more than
half again. A rotation of corn-wheat-meadow reduces soil erosion by an order
of magnitude when compared to fallow conditions. Continuous meadow essenti-
ally eliminates soil losses. The effectiveness of some SWCPs are functions of
55
-------
TABLE 5-8. SELECTED C VALUES FOR COMMON PRACTICES WITH CONVENTIONAL TILLAGE
(AFTER STEWART ET AL. 1975)
Rotation Range of Values for Crop Factor
(C factor of USLE)
Corn - Residue Removed, Fall Plow
Corn - Residue Removed, Spring Plow
Corn - Residue Left, Fall Plow
Corn - Residue Left, Spring Plow
C-0*
C-C-C-W-M - Residue Left
C-C-W-M-M - Residue Left
C-C-W-M - Residue Left
C-W-M - Residue Left
Cotton
Grass
Alfalfa
Soybeans
W-M
W-Fallow
Fallow
0.54
0.50
0.42
0.38
0.17
0.17
0.09
0.07
0.06
0.34
0.00
0.02
0.48
0.05
0.38
1.00
- 0.62
-0.59
- 0.52
- 0.48
- 0.25
- 0.23
- 0.14
- 0.11
- 0.10
- 0.49
- 0.01
- 0.54
C-Corn, 0-Oats, W-Wheat and M-Meadow
how much residue is left on the surface. This is true for practices such as
chisel or no-till plowing. Table SrlO lists a few practices and the typical
percent ground covered with residue when they are used. Baker et al. (1978")
concluded that percent of soil covered by crop residue explained between 78
and 89 percent of the variance in erosion among six different tillage systems.
Some judgments as to effectiveness of SWCPs to prevent soil detachment
are possible by relating the time-dependent rainfall erosivity and crop
cover with critical periods of pollution by suspended solids. Two primary
considerations in evaluating suspended solids are 1) the time during the year
when sediment concentrations are highest, and 2) the time period when water
quality problems related to suspended solids occur. The expected time of
highest erosion potential due to soil splash detachment can be predicted by
combinations of the time dependent erosion index distribution and the crop
cover. Sediment produced by splash erosion is more likely to stay in
suspension than that from other types of erosion. Critical pollution periods
are a function of the nature of the problems. For example, settleable solids
56
-------
TABLE 5-9. VALUES OF SUPPORT PRACTICE FACTOR, P (AFTER STEWART ET AL. 1975
AND WISCHMEIER AND SMITH, 1965)
1.
2.
3.
Practice
Contouring
Contour -Terrac e s
Contour Listing
or
Ridge Planting
or
Contour Strip
Croppinga
Land Slope (percent)
1-2 2-7 7-12 12-18 18-24
0.6 0.5 0.6 0.8 0.9
0.6/n*" 0.5/n 0.6/n 0.8/n 0.9v^n
0.3 0.25 0.3 0.4 0.45
aReconunended Strip Widths 31 31-27 27-23 23-18 18-15
(in meters)
Slope-Length Limits 120 120-90 90-25 25-20 20-18
(in meters)
n* is the number of equal length intervals the field is divided into.
TABLE 5-10. THE EFFECT OF RESIDUE LEFT ON SURFACE BY SELECTED TILLAGE
PRACTICES ON RUNOFF AND EROSION (AFTER BAKER ET AL. 1978)
Tillage Practices
Conventional
Buffalo Till
Chisel
Disk
Ridge-Plant
Coulter (no -till)
% Residue
Remaining on Surface
5
16
19
30
35
56
Runoff *
(mm)
65
39
58
49
37
39
Erosion *
(MT/ha)
15.4
6.2
10.2
5.4
2.0
0.9
*Averaged result of three simulated storm events on three soils.
57
-------
destroy fish eggs during the incubation period but turbidity can foul
drinking water at any time. Crop and residue cover when used as part of a
sediment control system differ from some SWCPs in that their effectiveness
is very seasonally dependent. This variability is reflected in the C value
of the USLE.
Wischmeier (personal communication, 1978) has divided crop effectiveness
for erosion control (primarily soil detachment, see Stewart et_ al. 1975)
into six stages. Crop cover factors can be determined for each of the follow-
ing six periods:
Period F : Inversion plowing to secondary tillage.
Period SB : Secondary tillage for seedbed preparation until the crop has
developed 10% canopy cover.
Period 1 : End of SB period until crop has developed a 50% canopy
cover.
Period 2 : End of period 1 until canopy cover reaches 75%.
Period 3 : End of period 2 until crop harvest.
Period 4 : Harvest to plowing or new seeding.
The actual periods of the year covered by each of these crop stages depend on
the location and particular cultural practices. The monthly crop cover values
for the USLE when combined with the monthly percent of average erosion index
result in products that reflect seasonal soil detachment potential. Table
5-11 illustrates these seasonal detachment potentials for four crop management
systems for corn: spring and fall plowed with residue left, spring plowed
with residue removed, and minimum tillage in Watkinsville, Georgia. Crops
are generally planted in a rotation so that•the variability is even more
pronounced.
To determine the average annual erosion potential a procedure similar to
that shown in Table 5-11 is used except that "crop-stage periods" are used
instead of months. Cover factors such as those given in Table 5-8 are
general values. The actual values used by SCS are specific to a location and
include rainfall patterns. If a table for C values for monthly or crop stage
periods can be found for a location of interest they can be used directly to
evaluate time effectiveness of some practices. However, these values are
not readily accessible so they need to be calculated as shown in Wischmeier
and Smith (1965) and Beasley (1972) .
The annual average cover factor for the year of corn grown using systems
B (spring plow, residue removed), C (fall plow, residue left), D (spring plow,
residue left), and E (minimum tillage) of Table 5-11 is 0.59, 0.49, 0.40, and
0.24, respectively. But the important point relative to single event
controls is that certain practices work better than others at any given time.
That is,the cover factor is not uniformly reduced or increased from one
58
-------
TABLE 5-11. RELATIVE EFFECTIVENESS OF SELECTED PRACTICES AT VARIOUS TIMES OF THE YEAR IN WATKINSVILLE,
GEORGIA, FOR CONTINUOUS CORN
Ul
VD
A.
B.
C.
D.
E.
Fraction of Annual
Erosion index
Residue Removed-
Spring Plow*
A X B
Residue Left-
Fall Plowt
A X C
Residue Left-
Spring Plow*
A X D
Minimum TillA
AXE
J
.06
.69*
.041
.57s
.034
.37*
.022
.30a
.018
F
.06
.69
.041
.57
.034
.37
.022
.30
.018
M
.08
.69
.055
.57
.046
.37
.030
.30
.024
A
.08
.69
.055
.57
.046
.37
.030
.30
.024
M
.10
.76
.076
.68
.068
.64
.064
.21
.021
J
.12
.71
.085
.65
.078
.62
.074
.21
.025
J
.19
.64
.122
.54
.103
.54
.103
.20
.038
A
.11
.33
.036
.31
.034
.31
.034
.17
.019
S
.or
.27
.019
.21
.015
.21
.015
.11
.008
0
.05
.27
.014
.21
.011
.21
.011
.21
.011
N
.04
.69
.028
.39
.016
.37
.015
.30
.012
D
.04
.69
.028
.57
.023
.37
.015
.30
.012
*Assumes: plowed May 1, planted May 15, harvested October 31.
tAssumes: plowed November 15, planted May 15, harvested October 31.
AAssumes: planted May 15.
*a
Soil loss ratio as defined in Wischmeier and Smith, 1965.
-------
practice to another for all months. In the example of Table 5-11, the minimum
tillage system is most effective at precisely the period when erosion rates are
highest from June to September. The practice of leaving residue in the field is
about as effective as minimum tillage in the winter months but not during the
critical summer period.
SWCPs that are relatively ineffective in late spring and early summer are
likely to be not very effective for sediment control, at least not for
continuous corn cropping. Crop canopies and residue cover are very seasonally
dependent, while most structural practices such as terraces are not.
Whenever crop rotations are used,certain years can be expected to have
greater potential for soil loss than others. For example, the probability of
soil loss increases each year after sod if corn is planted in a sod-based rota-
tion. This is illustrated in Figure 5-4. A sediment control program needs
to recognize these critical years within a rotation and a management system
should be developed that intensifies control practices during critical years
where possible. For example, tillage practices which leave more residue might
be used during the third and fourth years of corn following sod.
Tables 5-12 and 5-13 list the probability,as determined by a 25-year
simulation with Williams' model, of daily soil loss in tons per hectare for
similar fields in Aurora, New York, and Watkinsville, Georgia, respectively.
The lower probabilities of soil losses in New York as compared to Georgia are
primarily due to climatic conditions. Because of the nature of the factors in
the model no soil losses are predicted from December through March in New
York even though some small losses could be expected. However, soil losses in
New York are small enough in the winter so that winter controls (e.g., spring
versus fall plowing) would not be as effective as in Georgia. No daily soil
losses greater than 5 tons per hectare were predicted for the 25 years
simulated in New York but 20 such events could have been expected in Georgia.
Soil Transport Control Practices
Many of the practices listed in Section 4 function by reducing transport
capacity of runoff. The support-practice factor (P) of the USLE reflects
effectiveness of SWCPs to reduce soil transport. Of the six factors
included in the USLE, the P factor is quite often considered to most reflect
the effect of a SWCP. Actually, P values are available for only a few SWCPs
and they are based on "limited field data" in comparison to the other USLE
factors (Stewart et al. 1975). Table 5-9 lists published P values for
various SWCPs.
The effectiveness of transport control of a SWCP is dependent on the
magnitude of the runoff event. Unfortunately, since every situation is a
little different it is difficult to quantify how effective a particular
practice is under situations of a large storm event. However, this is an
important point because, as was shown in Table 5-5, relatively few storms
produced most soil movement from the 1.3-hectare fields in Watkinsville,
Georgia, and similar results were found for the Black Creek Study in
Indiana (Lake and Morrison, 1978).
60
-------
CTi
O
rt
u,
fn
0)
>
O
CJ
X
g
+j
O
ex
1/5
tfl
O
O
CO
r.
O base conditions and third
and later years corn after
hay
• first year corn after hay
A second year corn after hay
J J A S 0 N D
MONTHS OF THE YEAR
FIGURE 5-4. RELATIVE POTENTIAL FOR SOIL LOSS OVER THE YEAR FOR CONTINUOUS CORN AND FOR
CORN AFTER HAY IN NEW YORK.
-------
en
to
TABLE 5-12. PERCENT PROBABILITY PER MONTH OF THE OCCURRENCE OF A SOIL LOSS PRODUCING STORM EVENT IN
AURORA, NEW YORK, WHEN CORN IS SPRING PLOWED AS WITH WILLIAMS' MODEL IN A 25-YEAR
SIMULATION
Soil Loss (Tons/Hectare)
0.01 0.50 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00
Month Zero to to to to to to to to to to to to to to
0.49 0.99 1.99 2.99 3.99 4.99 5.99 6.99 7.99 8.99 9.99 10.99 11.99 More
1 100.0 --------- 1 .
2 100.0 -_-__--_-_-
3 100.0 -----------
4 99.8-- ---------
5 100.0 -----------
6 99.3 0.1 - 0.1 0.2 - - - 0.1 - - -
7 99.8- - 0.1 -
8 99.6 0.3 ----------
9 99.70.1- - 0.1 -
10 99.6 0.1 0.1 0.1 --------
11 100.0 __--------_
12 100.0 -----------
-------
TABLE 5-13. PERCENT PROBABILITY PER MONTH OF THE OCCURRENCE OF A SOIL LOSS PRODUCING STORM EVENT IN
WATKINSVILLE. GEORGIA, WHEN CORN IS SPRING PLOWED AS PREDICTED WITH WILLIAMS' MODEL IN
A 25-YEAR SIMULATION
en
u>
Month
1
2
3
4
5
6
7
8
9
10
11
12
Zero
92.1
92.7
91.8
94.4
92.9
96.2
97.2
97.9
97.7
98.7
94.9
93.8
Soil
0.01
to
0.49
5.2
5.3
3.8
2.4
1.4
1.0
0.7
0.9
1.4
0.7
3.0
3.6
Loss (Tons/Hectare)
0.50 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00
to to to to to to to to to to to to to
0.99 1.99 2.99 3.99 4.99 5.99 6.99 7.99 8.99 9.99 10.99 11.99 More
2.0
1.1
1.9
1.8
1.4
1.0
0.7
0.6
0.2
-
1.2
1.8
0.1
0.5
1.6
1.2
1.5
0.9
0.2
0.3
0.4
0.5
0.4
0.6
0.2 0.1 ----- -
0.1 -------
0.2 0.2 0.1
0.1 - - - - - -
1.1 0.3 0.2 0.2 0.1 0,1 €.2 - - - 0.1
0.1 0.1 0.1 ----- - - 0.2
0.5 0.1 0.1 - - - - 0.1 -
0.1 -------
0.1 -------
--------
0.4 -------
0.1 -------
-------
Soil transport control by SWCP is reduced in effectiveness as slope length
and storm size increase. The practices are designed to control or intercept
relatively shallow overland flow. In the case of contouring, contour listing
or ridge planting flow will concentrate if the slope length is too long or the
storm event large and will eventually break over the ridges, potentially
creating an even worse problem. Recommended spacing for terrace or strip crop
intervals (Table 5-9) is based on preventing concentrated flow that could
produce deep rills or gullies. As the magnitude of a runoff event increases,
the effectiveness of a SWCP to control soil transport decreases. For example,
the relative effects of increased surface roughness or infiltration on runoff
reduction are less for large storms.
Grass buffer strips are placed between streams and upland agricultural
land for the purpose of removing pollutants from overland flow. The principal
control mechanism of grassed buffer strips is to reduce runoff velocity so that
particles settle out. Infiltration might also be increased so that runoff will
be reduced. Some of the factors which might influence buffer strip effective-
ness are width of strip relative to runoff volume, density of grass cover,
detention time of flow in buffer strip and perhaps most importantly, uniformity
of flow over the buffer strip.
The effectiveness of terraces is not time dependent as most crop canopy or
residue inclusive SWCPs are. However, the effectiveness of ridge planting, con-
tour strip cropping, and contouring may vary with time. Wischmeier (personal
communication, 1978) shows ridge planting most effective in crop stage periods
SB and 1 (seedbed and establishment, respectively). These periods typically
span May, June and July, and as illustrated in Figure 5-4, these are at least
for continuous corn, potentially very vulnerable periods for soil loss.
Following the categorization of Section 4, the cultural and structural
practices are compared in Table 5-14 relative to their effectiveness of
control of each mechanism. The table should be used with care because
effectiveness of a practice is site specific.
Selected SWCP Effect on Soil Loss
Four different management systems were analyzed for a 25-year period in
Georgia for continuous corn. The field physical features were those of water-
shed P2 in Watkinsville, Georgia. In all cases the field was tilled on the
contour. The four systems analyzed were 1) spring moldboard plow, 2) fall
moldboard plow, 3) spring chisel plow, and 4) no-till. Assumptions were that
fall moldboard plowing was done November 15, spring moldboard and chisel
plow on May 1, emergence date was May 15, the crop was harvested on November
1st and had a yield of 4 metric tons/ha while 2900 kg residue was left per ha.
The predicted probabilities of a storm event capable of producing runoff
are given in Table 5-15. Probabilities of surface runoff are highest in winter
and lowest in late summer and fall. Table 5-16 lists the probabilities of
a soil loss producing storm event on a monthly basis as predicted with three
different models (the USLE, the Onstad-Foster and the Williams Models). In
64
-------
TABLE 5-14. QUALITATIVE EFFECTIVENESS OF SELECTED SWCP TO CONTROL VARIOUS
STAGES OF SEDIMENTATION PROCESS
Raindrop Runoff Runoff Seasonal Effectiveness*
Practice Type Detachment Detachment Transport F SB 1 2 3 4
Cultural
No Tillage
Cons. Tillage
Contour Farming
Graded Rows
Contour Strip
Cropping
Sod Based Rotations
Cover Crops
Contour Listing
Ridge Planting
Structural
Terraces
Diversions
Grassed Waterways
Filter Strips
Artificial Drainage
H**
M
N
N
M
L-H
L-H
N
M
N
N
N
N
N
M
L
L-M
L-M
M-H
L-H
L
M7
H7
H8
H8
H8
L
N
M H H H M M
M M M M M L
L-M1
L-M1
M-H2
L-H3
L M L L M H4
M
M H H M M M
H5
L
L
L-M6
L
*Crop stage periods (Wischmeier, personal communication, 1978).
**H - High, M - Medium, L - Low, N - No Significance or none.
1. Effectiveness depends on slope and slope length, see Table 5-9.
2i See Table 5-9.
3. H for years when in sod, L for other years.
4, Level of effectiveness depends on degree of canopy development, see
Figure 5-2.
5. Effectiveness depends on the reduction in transport energy which will
depend on outlet design.
6. Effectiveness depends on size and character of filter strip and uniformity
of overland runoff.
7. Controls detachment in rills.
8. Controls detachment in gulleys.
9. Greatest effectiveness during years with sod.
65
-------
TABLE 5-15. PERCENT PROBABILITY OF DAILY RUNOFF FROM THE P2 WATERSHED IN WATKINSVILLE, GEORGIA
OVER A 25YEAR PERIOD FOR BASE CONDITIONS. PREDICTIONS WITH SCS METHOD
CTi
Month
1
2
3
4
5
6
7
8
9
10
11
12
Zero
86.4
84.1
87.2
92.4
92.6
96.2
97.2
97.8
97.3
98.4
91.7
88.3
0.01
to
0.24
6.4
8.9
6.1
3.7
3.4
2.4
1.8
1.6
1.8
1.0
4.1
6.5
0.25
to
0.49
2.7
2.8
1.2
0.6
1.1
0.5
0.2
0.2
0.1
-
2.0
1.1
0
0
1
2
2
1
1
0
0
0
0
0
0
1
.50
to
.99
.4
.1
.0
.2
.4
.2
.5
.2
.2
.1
.9
.6
1.00
to
1.49
1.8
0.7
0.7
0.8
0.3
0.2
-
-
0.1
0.2
0.1
0.7
1
1
0
0
0
0
0
0
0
0
.50
to
.99
.5
.5
.6
.6
.3
-
-
-
.1
-
.4
.2
2
2
0
0
0
0
0
0
0
0
.00 2.50 3.00 3.50 4.00 4.50 5.00
to to to to to to to
.49 2.99 3.49 3.99 4.49 4.99 5.49
.2 0.2 -
.2 - 0.1 - 0.2 -
.2 0.3 0.1 0.2 0.1 0.2
.1 - 0.2 0.1
.1 0.2 -
-------
0.1
_______
.1 ------
0.1
.2 0.2 0.1
.6 0.2 - 0.1 0.1
5.50 6.00
to to
5.99 More
0.1
-
0.2 0.1
-
0.1
0.2
-
-
-
-
-
__ _
-------
en
TABLE 5-16. PERCENT PROBABILITY OF A STORM EVENT PRODUCING SEDIMENT (GREATER THAN 0.1 T/HA) FOR
FOUR DIFFERENT PRACTICES AS PREDICTED WITH 3 MODELS ON THE P2 WATERSHED IN WATKINSVILLE,
GEORGIA
Month
1
2
3
4
5
6
7
8
9
10
11
12
SP
10
10
12
9
11
9
10
8
7
5
7
9
.5
.4
.4
.1
.3
.5
.1
.0
.0
.6
.9
.9
USLE MODIFIED
FP CP
10.9
11.2
12.4
9.1
11.3
9.5
10.1
8.0
7.1
5.7
8.2
9.9
10.9
10.1
12.3
9.1
11.1
9.2
10.1
7.9
6.3
5.1
7.9
9.7
NOT*
9.7
9.2
11.9
8.8
9.6
7.2
9.0
7.4
4.8
3.5
7.5
9.1
ONSTAD -FOSTER
SP FP CP
8
8
10
7
11
9
10
7
5
4
6
6
.3
.1
.2
.8
.3
.2
.1
.9
.6
.3
.4
.6
8.4
8.3
10.2
8.4
11.1
9.2
10.1
7.9
5.9
4.7
6.6
7.0
8.0
8.0
9.9
7.5
9.7
7.5
9.2
8.1
4.7
3.0
6.4
6.5
NOT
7.9
7.3
9.2
7.1
7.8
5.2
5.0
4.7
3.2
2.1
6.3
6.5
SP
7.9
7.3
8.7
5.6
7.1
3.8
2.8
2.1
2.3
1.3
5.1
6.2
WILLIAMS
FP CP
7.9
7.7
8.8
5.6
7.0
3.8
2.8
2.1
2.6
1.3
5.1
6.2
7.7
6.8
7.9
5'. 4
6.8
3.8
2.5
1.9
2.2
1.2
5.0
6.2
NOT
7.5
6.1
7.8
5.1
6.1
3.0
2.2
1.7
1.8
1.1
5.0
5.9
"SP- Springboard Moldboard Plow, FP-Fall Moldboard Plow, CP-Chisel Plow, NOT-No Till
-------
all cases the Williams Model predicts less soil loss than the other two. It
also indicates more variation in the probability of soil loss over the year.
This greater variation in soil loss probabilities is not surprising since
the Williams Model is more dependent on runoff than either of the other two.
The monthly runoff probabilities are quite variable.
Probabilities of a soil loss producing storm event from the spring mold-
board plowed system are lowest in September, October and November. The
Williams Model indicates that it is quite low in June and July as well but
the other two do not. None of the models predict much difference in'fall vs.
spring plowing relative to probability of a soil loss producing storm event.
The USLE and Onstad-Foster Models suggest a very slight increase in
soil loss probability from about December to May; the Williams Model does not.
On the other hand all of the models predict reduced probability of soil loss
relative to moldboard plowing when fields are chisel plowed. They predict
even greater reductions when no till is used.
The probability of an event causing more than one ton of soil loss per
hectare is given in Table 5-17. This table indicates when severe soil loss
is likely to occur. For these larger events the Williams Model seems to
agree with the others. Apparently the small storm events are included by
the USLE and the Onstad-Foster Models but not by the Williams Model. The
point was made earlier that the USLE and Onstad-Foster Models can predict
soil losses for runoff events even if no runoff is predicted.
Under the spring plow system the most severe soil loss can be expected
in May, June and July according to the USLE Model. This was precisely the
conclusion reached by analyzing three years of sediment data from the P2
watershed (see Table 5-5). The Onstad-Foster and Williams Models support
this conclusion also but they indicate a somewhat more uniform distribution
throughout the year. April is also predicted to be a somewhat troublesome
month.
When comparing spring versus fall moldboard plowing for severe soil
loss events (i.e., greater than 1 t/ha) all the models predict increased
vulnerability from December through April with the fall plowing. In January,
for example, the three models predict from 3 to 5 percent chance of a one
ton per hectare soil loss event with spring plowing as compared to 9 to 18
percent chance with fall plowing. Over the whole year the probability of
a severe soil loss increases about 25 percent when the field is plowed in
the fall instead of in the spring.
Chisel plowing greatly reduces the potential for severe soil loss events
throughout the year. This is particularly true from April to September when
the probability of severe events is 50 to 60 percent less than when they are
spring moldboard plowed.
No till plowing reduces soil loss 70 to 80 percent over spring mold-
board plowing. It is most effective in May through September but substan-
tially reduces soil loss from severe events in all months.
68
-------
<£>
TABLE 5-17 PERCENT PROBABILITY OF A STORM EVENT RESULTING IN SOIL LOSS GREATER THAN ONE TON PER
HECTARE FOR 4 DIFFERENT PRACTICES AS PREDICTED WITH THREE MODELS ON THE P2 WATERSHED
IN WATKINSVILLE, GEORGIA
Month
1
2
3
4
5
6
7
8
9
10
11
12
USLE MODIFIED
SP FP CP
0.3
0.3
2.0
1.6
3.9
3.9
4.4
2.2
1.1
0.8
0.6
0.6
0.9
1.2
3.6
2.6
3.9
4.0
4.6
2.2
1.3
0.8
0.8
1.5
0.3
0.1
1.6
0.9
1.8
0.9
1.4
0.9
0.4
0.5
0.3
0.3
NOT*
0.3
0.1
1.0
0.7
0.9
0.2
0.4
0.3
0.3
0.0
0.4
0.3
ONSTAD-FOSTER
SP FP CP
0.5
0.9
2.2
1.4
3.0
1.9
2.1
1.1
0.4
0.4
0.6
1.0
1.4
1.4
3.4
2.6
2.8
2.1
2.0
1.1
0.5
0.4
0.9
1.9
0.4
0.3
1.7
1.0
1.8
0.4
0.7
0.3
0.3
0.3
0.6
0.6
NOT
0.3
0.2
1.5
0.6
0.7
0.2
0.1
0.1
0.1
0.0
0.5
0.5
SP
0.4
0.6
2.1
1.3
3.8
1.4
0.9
0.4
0.5
0.5
0.8
0.8
WILLIAMS
FP CP
1.8
1.4
3.5
2.9
3.5
1.5
1.0
0.6
0.5
0.5
1.1
1.9
0.4
0.3
1.7
1.1
1.7
0.4
0.3
0.1
0.1
0.1
0.6
0.6
NOT
0.3
0.1
1.5
0.2
1.2
0.2
0.1
0.0
0.1
0.0
0.5
0.3
*SP-Spring Moldboard Plow, FP-Fall Moldboard Plow, CP-Chisel Plow, NOT-No Till
-------
This analysis suggests that small soil losses are going to occur under
all four of the systems. Even when one goes from spring to fall moldboard
plowing to no till, only modest reduction can be expected in the probability
of soil loss producing storm events. But the probability of severe soil loss
events is greatly reduced by practices such as chisel plowing or no till
cultivation. In other words, the number of soil loss producing storm events
stays approximately the same, but the amount of soil which is lost is greatly
reduced. As compared to fall plowing? spring moldboard plowing reduces soil
loss potential in the winter. Chisel plowing is most effective in the spring
and summer, but is better than either spring or fall moldboard plowing in all
months. No till reduces the potential for serious soil losses substantially
over the other three systems from March through November and is comparable
to chisel plowing in the winter months.
SUMMARY AND CONCLUSIONS
The sediment causing processes include soil detachment and transport.
The bulk of on-field soil detachment is a result of rainfall impact. Both
primary particles and aggregates can be detached. Sand and silt are more
easily detached than clay. Aggregates tend to be broken down by continuous
raindrop impact. Soil detachment by rainfall occurs in a very thin layer at
the soil surface. Soil detachment by runoff can occur if shear forces of flow
are great enough or if flow undercuts the soil. Detachment by flow is
localized in areas where runoff concentrates and generally comes from deeper
in the soil profile than that detached by rainfall. Aggregates detached by
runoff tend to be larger than those detached by rainfall.
Soil is transported by runoff. Clay and organic matter are more easily
transported than coarser particles such as sand. The energy of runoff is
reflected in its velocity and volume.
Sediment characteristically has smaller and lighter particles than the
original soil. Sediment deposition in streams and lakes is not the only
form of water pollution from soil; soil can also be in suspension or move as
a bed load. The suspended load which can cause turbidity, reduced photo-
synthesis, etc., will be made up predominately of clay and organic matter.
Suspended load is relatively variable over the year and is typically the
result of a single or series of intense storm events. If the suspended
load originated as soil detached by rainfall, it is more likely to have
adsorbed chemicals than if it is made up of soil detached by runoff,
especially if the chemicals were surface applied. Bed load typically
consists of coarse material. The bed load does not fluctuate as much as
the suspended load and is a minor element in terms of soil adsorbed
substances.
All practices designed to control erosion will potentially reduce
sediment losses. Their effectiveness at erosion control is reflected in
the cover and supporting practice factors of the USLE. The effectiveness
of SWCPs at sediment control will depend on field location and to a lesser
degree on the sediment producing storm characteristics. Soil and water
70
-------
conservation practices that control soil detachment as a result of rainfall
impact (e.g., crop canopies and residue cover) will not be as sensitive to
storm intensity as SWCPs based on transport control (e.g., contour tillage
and terraces). The effectiveness of some SWCPs that control transport decrease
as the magnitude of runoff increases (e.g., contour tillage and lister planting).
Raindrop detachment controls are typically crop canopies or residue
covers. These controls are more effective at preventing movement of adsorbed
substances than transport control SWCP. Detachment of soil by runoff is
controlled by intercepting or slowing overland runoff by increased surface
roughness before a flow becomes concentrated. Transport control of soil is
accomplished by reducing runoff velocity or volume. Transport controls rely
on soil settling out of runoff so that they are more effective at removing
coarse than fine particles.
The potential for a storm event to produce sediment varies regionally
and over the year for a given region. A few very intense storms can account
for most of the sediment yield from a watershed.
Certain practices are more effective during one time of the year than
another. This is particularly true of vegetative canopies and residue covers.
Seasonal practice effectiveness should be matched with seasonal sediment
production potential. A comparison of four tillage systems - fall moldboard
plowing, spring moldboard plowing, chisel plowing and no-till in Watskinsville,
Georgia, showed that none of the practices greatly reduced the probability of
the number of soil loss producing storms, but the probability of severe
events (i.e., greater than 1 ton/ha sediment yield) was significantly different.
As compared to spring plowing, fall plowing resulted in about 25 percent more
severe events mostly occurring in the winter; chisel plowing reduced the
probability by 40 to 50 percent; and no-till cultivation reduced the probability
of severe soil loss by 70 to 80 percent.
71
-------
SECTION 6
EFFECTS OF SOIL AND WATER CONSERVATION PRACTICES
ON EDGE-OF-FIELD NUTRIENT LOSSES
Douglas A. Haith
This section presents the results of modelling studies which were used to
quantify the effects of certain SWCPs on edge-of-field losses of nutrients.
The rationale for the use of mathematical models is two-fold. Although as
noted in Section 4 qualitative conclusions regarding the effects of SWCPs on
nutrient losses are possible, relatively few field data are available
which would permit a quantitative assessment. An objective of these modelling
studies was to provide this quantification for certain practices and locations.
A second objective was to develop a general methodology for evaluating SWCPs
which would be applicable to a variety of crops, locations and practices.
The importance of such a methodology should not be underestimated.
Nutrient losses from croplands are site-specific, depending on weather, soils,
nutrient management and cropping practices. No SWCP can have a uniform
nutrient control effectiveness (e.g., 10% reduction in nitrogen losses) for
all locations and crops. In the absence of very extensive field studies,
models are essentially the only means of generating site-specific information.
There is a variety of information which is of interest in a quantitative
assessment of SWCPs. Edge-of-field losses of nitrogen (N) and phosphorus (P)
dissolved in runoff and in percolation waters from a field are of concern
as are the solid-phase losses of these nutrients associated with eroded soil.
This information can be reported in several ways, such as losses for specific
time periods or storm events and average monthly and yearly losses. The most
straightforward evaluation of SWCPs is a comparison of their effectiveness in
controlling the mean or average annual nutrient loss from croplands. These
losses are reported in this section, based on either ten or twenty-five year
simulations.
An evaluation of SWCPs in terms of their ability to control mean annual
nutrient losses is not completely adequate, however. Since nonpoint source
pollution is influenced by weather conditions, nutrient losses in runoff and
percolation can be highly variable. Thus losses may be very high in certain
years and quite low in other years. This variation can be presented as
frequency distributions of annual nutrient losses. At least conceptually,
such distributions can be used to evaluate the risks (probabilities) of large
72
-------
nutrient losses associated with various SWCPs. For example, two alternative
SWCPs may both reduce mean annual dissolved P loss in runoff by approximately
the same amount, but one of the practices may result in a much larger variance
of annual losses than the other. It then could be inferred that the risk
of P losses significantly larger than the mean is greater with the high
variance SWCP. In many instances this would indicate that the alternative
SWCP which has a comparable mean, but smaller variance is preferable. Accord-
ingly SWCPs are evaluated in this section both in terms of mean annual losses
and frequency distributions of annual losses.
This section also discusses the simulation models used in the estimation
of nutrient losses, model calibration or validation and the long-term (10-25
yr) simulation of SWCPs for four locations (central New York, central Iowa,
northeast Georgia, and central Michigan). These locations were chosen primar-
ily to reflect the effects of climatic variability. While results should not
be extrapolated directly to other locations, they are indicative of the differ-
ent levels of effectiveness that may be expected from SWCPs. The SWCPs
analyzed were conservation (minimum) tillage, contouring, terracing, sod-based
rotations, and no-tillage.
Although this report is primarily an evaluation of SWCPs, it must be
recognized that other agricultural management practices may be suitable for
control of cropland nutrient losses. In particular, nutrient losses may be
closely related to the quantities and timing of fertilizer applications.
While a systematic evaluation of fertilizer management was not attempted in
the project, the simulation models were used to estimate the effects of
changes in N fertilizer applications on losses of the nutrient in runoff and
percolation. This work is summarized in Appendix D.
SIMULATION MODELS
Two general simulation models were used in the analysis of nutrient
losses. The first of these is the Agricultural Runoff Management (ARM) model
which was developed by Hydrocomp, Inc. for the U.S. Environmental Protection
Agency (Donigian and Crawford, 1976; Donigian et^ al. 1977). The second model
was developed at Cornell for a related research project (Tubbs and Haith, 1977)
and will be referred to as CNS (Cornell Nutrient Simulation) for convenience. The
general features of the two models are outlined in this section, while more
detailed descriptions are given in appendices A (CNS) and F (ARM).
The two simulation models differ significantly in structure and data needs.
The CNS model is based on the U.S. Soil Conservation Service (SCS) curve num-
ber runoff equation and the universal soil loss equation. The impacts of
SWCPs on CNS model parameters can be obtained from readily available sources,
and model calibration is unnecessary. The model provides a general computa-
tional tool which incorporates the most commonly used runoff and soil loss
estimating procedures and can be applied to arbitrary crops, soils and loca-
tions in the absence of field water quality data. In contrast the ARM model
attempts a less empirical approach to the estimation of runoff and soil loss
and is in principle more general than either the curve number or universal
soil loss equations. This generality is achieved through the use of calibra-
tion parameters and the model can only be applied to field situations which
73
-------
have been monitored for runoff. Since ARM has been calibrated on only two
experimental fields (one each in Georgia and Michigan), the model is presently
of limited value in the analysis of regional differences in effectiveness of
SWCPs.
As might be expe'cted from the above discussion, the evaluations of SWCPs
reported in this section are based primarily on the CNS model. However,
certain practices, most notably no-tillage, cannot be studied using CNS,
largely because runoff parameters (curve numbers) for this condition are un-
known. The ARM model provided a means of studying no-tillage and comparing
its effects with other SWCPs such as contouring and terracing. Also, since
the CNS model is based on the SCS curve number equation, it predicts only
total or direct runoff and does not separate a runoff hydrograph into surface
and subsurface flow components (overland flow and interflow). Conversely,
the ARM model estimates both runoff forms and was used to study the effects
of certain SWCPs on nutrient losses in both surface and subsurface runoff.
Agricultural Runoff Management (ARM) Model
In the current application, the ARM model computes water, soil and
nutrient losses from a field in surface runoff and interflow at 15-min inter-
vals during a runoff event. During other times, soil nutrient levels are up-
dated every 6 hrs. Vertical movement and transformations of water and
nutrients in the soil profile are simulated in surface, upper, lower and
groundwater zones or layers. The depths of the surface and upper soil zones
are specified by model input parameters and are generally 0.3-0.8 cm and
8-15 cm, respectively. The upper zone depth corresponds to the depth of soil-
incorporated minerals which are available for loss in overland flow (surface
runoff). Nitrogen and phosphorus in the soil is modelled using first order
kinetics and single-valued adsorption isotherms. Model details are provided
in Appendix F.
ARM Model Calibration
Characteristics of the two watersheds for which ARM has been calibrated
are listed in Table 6-1. Details of the calibration study are given in
Donigiari et_ aL (1977). Calibration results are summarized in Table 6-2 for
the various water, soil and nutrient losses predicted by the model. In
Table 6-2 "runoff" includes both surface runoff and interflow. In general,
the results in Table 6-2 indicate that it is possible to calibrate the ARM
model in such a way that it simulates dissolved nutrient losses with consider-
able accuracy. However, the losses of nutrients in solution are simulated
much better than those associated with the sediment or eroded soil. The
Georgia results are generally better than those for Michigan.
Cornell Nutrient Simulation (CNS) Model
The CNS model is described in detail in Appendix A. This model is a
modification of an earlier model developed by Tubbs and Haith (1977). The
model consists of three basic components: daily water balance, a daily soil
loss (erosion) calculation, and monthly soil N and P inventories. The
hydrologic model is based on the Soil Conservation Service's curve number
74
-------
TABLE 6-1. WATERSHEDS USED IN ARM MODEL SIMULATIONS CDonigian et al. 19771
Watershed
Location
Soil Type
Area (ha)
Crop
Supporting Practices
Fertilizer Applications
Nitrogen (kg/ha)
Phosphorus (kg/ha)
Calibration Period
P2
Watkinsville, GA
Cecil Sandy Loam
1.3
Corn
Contoured
273
54
5/74-9/75
exluding
2/75, 5/75
P6
East Lansing, MI
Spinks Loamy Sand
0.8
Corn
Contoured
460
224
6/74-9/75
excluding
5/75
runoff equation, with suitable extensions to handle snowmelt. A daily soil
moisture inventory is maintained for the top 30 cm of soil which is assumed
to be homogeneous. Percolation is computed as any excess of soil water over
field capacity in the 30-cm layer. Although soil water content is not modelled
explicitly for depths below 30 cm, it is assumed that plant evapotranspiration
needs in excess of the available water in the 30-cm surface layer will be
met by soil water from lower depths. Soil erosion is estimated for each rainy
day (exluding snowmelt periods in northern climates) using Onstad and Foster's
(1975) modification of the universal soil loss equation. Daily runoff,
percolation and erosion values are summed for each month and these monthly
water and soil losses are inputs to the nutrient calculations.
Monthly inventories or mass balances are computed for soil N and P in the
30-cm surface layer. Although portions of plant nutrient needs may be supplied
from nutrients (particularly N) at depths below 30 cm, N and P losses will be
largely influenced by the levels of these nutrients in the surface layer.
Separate organic and inorganic N inventories are maintained, but it is assumed
that all soil inorganic N is converted to nitrate and hence moves readily in
runoff and percolation waters. Runoff losses of solid-phase N are assumed to
be organic nitrogen associated with eroded soil. Mineralization of organic N
is considered temperature limited. Denitrification, ammonia volatilization
from the soil and ammonium fixation are not modelled explicitly, but 25% of
fertilizer N applications are subtracted .to partially account for these .loss
mechanisms. Soil P is divided into available and fixed inorganic forms. The
former is the pool of available or extractable P which can enter readily into
soil solution. The partitioning of available P into dissolved and adsorbed
75
-------
TABLE 6-2. COMPARISONS OF ARM MODEL SIMULATION RESULTS WITH OBSERVED LOSSES
OF WATER, SEDIMENT AND NUTRIENTS (Donigian et §j. 1977)
~~~ Watershed P2 Watershed P6
(Georgia) (Michigan)
Predicted Observed Predicted Observed
Total Runoff (cm)
Sediment (MT/ha)
N03-N (kg/ha) in Runoff
29
8.6
2.5
26
7.1
1.5
26
7.1
6.3
27
16.8
2.2
NH4-N (kg/ha) in Runoff
Dissolved
Solid-Phase
Total N (kg/ha) in Runoff
2.8
2.8
2.2
2.3
2.0
0.2
20
. O
Dissolved
Solid-Phase
Dissolved (PO
Solid-Phase P
)-P (kg/ha)
(kg/ha)
5.
14.
0.
10.
2
3
4
1
7.
9.
0.
5.
7
7
3
5
8.
15.
2.
8.
3
5
3
1
7
44
1
23
.9
.0
.4
.3
constituents is based on a linear equilibrium isotherm which is a function of
soil clay content and pH. In calculating losses of dissolved N and P, all in-
organic N and dissolved available P in the 30-cm layer are considered available
for loss in runoff. This assumption is necessary since the model cannot distin-
guish between subsurface and surface runoff. In situations where runoff con-
sists entirely of overland flow, this assumption can lead to over-estimates of
dissolved nutrient forms in runoff.
The CNS model has some disadvantages compared to ARM since it can only
determine monthly nutrient losses and cannot provide estimates of losses of
such nutrient forms as ammonium and organic N in solution. However, it does
have several distinct advantages. The foremost feature of the model is that
it is completely deterministic; i.e., all model parameters are based on the
physical and climatic characteristics of the field being modelled and no
calibration is needed. As a result, the CNS model can be applied to fields
which have not been monitored for runoff and water quality data.
Validation of the CNS Model
A number of small watersheds (fields) were modelled in New York and
Georgia to evaluate the accuracy of the model. Data for the validation were
76
-------
taken from two previous studies sponsored by the U.S. Environmental Protection
Agency (Smith et^ al_. 1978; Klausner et^ al . 1976).
The four Georgia watersheds (all at Watkinsville) are described in Table
6-3. All watersheds have a Cecil sandy loam soil. Although the watersheds
were moldboard plowed at the beginning of the study, subsequent tillage
practices on all watersheds were minimized (generally discing and harrowing).
Watersheds P3 and P4 had major conservation practices (terracing, winter cover
crops) as compared with PI and P2 which, except for the final year on PI, had
continuous row cropping with contouring only. The watersheds thus provided
a test of the CNS model's capability to predict differences in the effects of
conservation practices. Watersheds P2 and P4 were monitored for runoff,
sediment and nutrient losses, while only runoff and sediment were measured on
PI and P3.
TABLE 6-3. SMALL WATERSHEDS AT WATKINSVILLE, GEORGIA,
USED FOR. CNS MODEL VALIDATION
Watershed
Area (ha)
PI
2.7
P2
1.3
P3
1.3
P4
1.4
Supporting Practice Contouring Contouring
Contouring, Contouring,
Terracing, Terracing,
Winter Cover Winter Cover
Crop
1972 Growing Season Soybeans
Winter Cover None
1973 Growing Season Soybeans
Winter Cover None
1974 Growing Season Soybeans
Winter Cover Barley
1975 Growing Season Sorghum
a
"a
Corn
None
Corn
None
Corn
Soybeans
Rye
Soybeans
Rye
Soybeans
Barley
Soybeans
a
a
Corn
Rye
Corn
Rye
Corn
— —
Watersheds P2 and P4 were not monitored in 1972.
Data for nutrient inputs, cropping dates and initial soil nutrient levels
were obtained from Smith £t al_. (1978). Soil moisture parameters were obtained
from soil surveys. Curve numbers and soil loss factors for the runoff and
erosion submodels were obtained from standard sources (Ogrosky and Mockus,
1964; Wischmeier and Smith, 1978; Stewart eit al^. 1976). Model input parameters
for the validation runs are summarized in Appendix A.
Validation results for the Georgia watersheds are given in Table 6-4.
Model predictions and observations can be compared in several ways. Consider-
ing first the absolute prediction errors, it can be seen that model accuracy
77
-------
varies among watersheds. Considering the possible errors in both field
measurements and model parameters, runoff and sediment predictions are quite
accurate for watersheds PI and P3. Sediment losses are substantially over-
predicted for P2 and P4 and the runoff prediction is too small for P4. The
large discrepancy between predicted and observed sediment losses for P4 may
be due to sediment deposition in terrace channels and a grassed waterway.
Such deposition was assumed not to occur in the model. Except for dissolved
P losses from P4, all other nutrient losses are over-predicted. The errors
in predicted solid-phase nutrient losses are due to the over-estimation of
sediment losses, while the overprediction of dissolved N losses in runoff is
likely due to the modelling assumption that all inorganic N in the top 30 cm
of soil is available for runoff loss.
The significance of errors in predicted water, soil and nutrient losses
depends on the purposes for which the predictions are made. In the current
application, the ability of the simulation model to estimate the relative
effectiveness of SWCPs is of most concern. Referring to Table 6-3, it can be
seen that watersheds PI and P3 have similar growing season crops, but
different supporting practices. Although the two watersheds are not otherwise
identical, it is reasonable to assume that supporting practices are the major
factor accounting for differences in losses from the watersheds. A similar
comparison is possible for watersheds P2 and P4. Observed and predicted runoff,
sediment and nutrient losses from the two watershed pairs (PI, P3 and P2, P4)
are indicative of the relative effectiveness of contouring Con PI and P2) ver-
sus terracing and winter cover (on P3 and P4). If the predicted reductions
in losses correspond to observations, this can provide at least a partial
testing of the CNS model's ability to evaluate SWCPs.
Table 6-5 compares the predicted and observed reductions in water, sedi-
ment and nutrient losses achieved by the conservation practices on P3 and P4.
These comparisons are generally encouraging. Predicted and observed reductions
are all in the same direction (positive) and in most cases are of comparable
magnitude. The significant exceptions are the effects of terracing on runoff
and solid-phase N and dissolved P losses from watershed P4. These errors are
mainly due to the underprediction of runoff and overprediction of sediment
loss on P4. It appears that runoff and sediment loss for this watershed is
not adequately accounted for by the parameters of the curve number runoff
equation and the universal soil loss equation. Given the relatively shortness
of the validation time period, this discrepancy is not surprising. A second
factor may account for the small reduction in observed dissolved P losses in
runoff. As noted in Section 4, the leaching of P from crop stubble and
residues may be a significant source of this nutrient in runoff, and such
leaching is not included in the simulation nodel.
A second validation of the CNS model was based on data from 24 field plots
(each approximately 0.3 ha) at Aurora, N.Y. (Klausner, et_ aJ^. 1976). The 24
plots represented three manure application rates, two management levels and
three times of manure application. The lowest application rates were
replicated, bringing the total to 24 plots. Straight-row corn was planted on
all plots and the two management levels were "good" (retention of crop residues
after harvest) and "poor" (removal of residues). The three manure application
rates were 35,100 and 200 MT/ha of dairy manure per year. The predominant
78
-------
TABLE 6-4. COMPARISONS OF CNS PREDICTIONS WITH OBSERVED LOSSES OF WATER, SEDIMENT AND NUTRIENTS FOR
FOUR WATERSHEDS AT WATKINSVILLE. GEORGIA
Dissolved N Solid-Phase N Dissolved P Solid-Phase
Runoff Sediment in Runoff in Runoff in Runoff P in Runoff
(cm) (MT/ha) (kg/ha) (kg/ha) (kg/ha) (kg/ha)
Watershed Pl&
Predicted
Observed
Watershed P2b
Predicted
Observed
Watershed P3a
Predicted
Observed
Watershed P4b
Predicted
Observed
52 62
62 64
36 32 22.0 16.6 0.8 7.8
48 19 9.6 9.2 0.6 5.8
36 13
48 12
20 18 8.3 10.6 0.2 3.1
42 6 5.1 3.6 0.5 1.6
a/
July, 1972 - Sept., 1975; No nutrient observations available
"/Runoff and sediment data for May, 1973 - Sept., 1975; Nutrient data for May, 1974 - Sept. 1975
-------
TABLE 6-5^ COMPARISON BETWEEN CNS PREDICTED AND OBSERVED EFFECTS OF SWCPs
% Reduction% Reduction
from PI to P3 from P2 to P4
Predicted Observed Predicted Observed
Runoff 31 23
Sediment 79 81
Dissolved N in Runoff
Solid-Phase N in Runoff
Dissolved P in Runoff
So lid- Phase P in Runoff
44
44
62
36
75
60
14
68
47
61
17
72
soil on the plots was a moderately to poorly drained Lima-Kendaia silt loam.
Runoff was collected in grass-lined interception ditches and percolation was
measured by tile drainage on twelve of the plots.
Data for model validation was obtained from Klausner et al. (1976)
and soil surveys. Additional, unpublished data for 1974 was provided by
Stuart D. Klausner, Department of Agronomy, Cornell University. As with
the Watkinsville validations, curve numbers for runoff calculations were
obtained from Ogrosky and Mockus (1964). Model input data are summarized
in Appendix A.
The Aurora validations provided additional tests of the CNS model
which were not possible using the Watkinsville watersheds. In particular,
the percolation and snowmelt components of the model could be te.sted at
Aurora. In addition, since the Aurora study involved a large number of
plots, modeJ predictions and observed losses could be averaged in various
ways among plots. The averaging is useful, since a single field or water-
shed is likely to include anomalies which are not accounted for in a simu-
lation model.
The results of the Aurora validations are given in Table 6-6. Because of
deposition in the grassed collection ditches, sediment loss data could not be
considered reliable and hence were not used in validations. The predicted and
observed losses are averaged over management and manure application rates.
As with the Watkinsville results, runoff is under-predicted, but with signifi-
cantly less error. Inorganic N losses are again over-predicted, but errors
are again much less significant that at Watkinsville. This is likely due to
the presence of greater portions of interflow in the Aurora runoff. The only
significant errors in the validation are in the predictions of runoff for the
poorly managed plots and of available P losses in runoff. The latter, however,
is consistently 30-40% of the predicted value, indicating errors in estimates
80
-------
TABLE 6-6. COMPARISON OF CNS PREDICTED AND OBSERVED LOSSES OF WATER, INORGANIC N
AND AVAILABLE P FOR 24 0.3-ha FIELD PLOTS AT AURORA, N.Y. (1972-1974)
oo
Plot Type
We 11 -Managed
Predicted
Observed
Poorly Managed
Predicted
Observed
35 MT/ha
Predicted
Observed
100 MT/ha
Predicted
Observed
200 MT/ha
Predicted
Observed
Overall Average
Predicted
Observed
Runoff
— — — — 1
5.1
5.7
6.3
13.2
5.7
6.1
5.7
16.1
5.7
9.4
5.7
9.4
Percolation
C' TO / V Y* 1 — — — — .
^ m/ y r j
45.2
32.2
43.4
45.3
44.3
41.8
a
44.3
35.6
44.3
38.8
Dissolved N
iTi T?1 in c\ ~P"P
4.8
3.9
6.5
5.5
4.3
2.7
5.8
8.4
8.2
5.0
5.6
4.7
Dissolved N
in P(aT*r*ol a f" "i nn
55.7
45.6
57.2
70.4
40.0
43.5
a
72.9
72.6
56.4
58.0
Dissolved P in
Rnnn-f f
0.22
0.55
0.30
0.83
0.14
0.31
0.28
0.91
0.49
1.23
0.26
0.69
percolation data available.
-------
of available P in the soil or manure or an error in the computed value of the
P adsorption coefficient. Both percolation and N losses in percolation are
adequately predicted. The relative differences in losses observed among plots
are also seen in the simulation predictions.
EVALUATION OF CONTOURING, TERRACES, SOD-BASED ROTATIONS AND CONSERVATION
TILLAGE IN NEW YORK, IOWA AND GEORGIA
The effects of contouring, terraces, sod-based rotations and conservation
tillage on water, soil and nutrient losses were simulated for Aurora, N.Y.
(central New York), Tama, IA (central Iowa) and Watkinsville, GA (northeast
Georgia). Nutrient, soil and water losses were estimated by 25 one-year
(Jan. - Dec.) runs of the CNS model, using meteorological data for 1952-1976.
Soil nutrient levels were reinitialized each year so that the results from
each one-year run can be considered independent samples (assuming weather in-
dependence from year to year).
All SWCPs are compared with "conventional" management which is defined
as continuous grain corn grown in straight rows with moldboard plowing and
no supporting conservation practices. Soil and crop characteristics for con-
ventional management are given in Table 6-7. Nutrient inputs and crop uptake
for sod-based rotations are shown in Table 6-8.
The parameters given in Tables 6-7 and 6-8 were determined mainly from
soil surveys and university extension staff in the relevant states (at Cornell,
Iowa State and University of Georgia) and represent current cropping practice
and soil properties at the respective locations. It should be noted that
the properties of the New York soil would indicate greater susceptibility to
runoff and erosion than the Iowa and Georgia soils. Average annual precipi-
tation at the New York, Iowa and Georgia locations for the 25-year simulation
period (1952-1976) was 88, 80 and 129 cm respectively. Soil characteristics
are not changed for any SWCP. The only changes in crop characteristics for
contouring and terracing (which is assumed to include contours) are appropriate
adjustments in runoff curve numbers. The curve numbers for the average ante-
cedent moisture conditions in New York, Iowa and Georgia are 84, 79 and 79 for
contouring and 80, 74, 74 for terracing, respectively.. Since the Iowa soil
is assumed to be bare following fall plowing (except for snow), the curve
number for fallow (86 for average antecedent moisture) is used during winter for
conventional management, contouring, and terracing. This implies that con-
touring and terracing do not reduce runoff from bare soil during the winter.
Conservation tillage consists of chisel or shallow disc plowing in the
spring at each location. Runoff curve numbers were chosen corresponding to
"good" hydrologic condition (85, 78 and 78 for average antecedent moisture in
New York, Iowa and Georgia, respectively). Sod-based rotations are assumed to
be four years of corn followed by four years of alfalfa (and vice-versa) at all
locations. Nutrient inputs and removals for these rotations are given in
•
Table 6-8. For the purposes of curve number selection, it is assumed that
the soil is in good hydrologic condition at all times. Fallow curve numbers
were used in the nongrowing season for second, third and fourth year corn in
Iowa.
82
-------
TABLE 6-7. SOIL AND CROP CHARACTERISTICS FOR STUDY LOCATIONS
New York
Iowa
Georgia
I. Soil Characteristics
Type
Hydrologic Group
Erodibility
Area (ha)
Slope (%)
Slope Length (m)
Bulk Density (g/cm )
Percent Clay
PH
Field Capacity (em per 30 cm)
Wilting Point (em per 30 cm)
Saturation (cm per 30 cm)
Organic N in top 10 cm (kg/ha)
Organic N Mineralization Rate
(% per yr)
Initial Inorganic N in top
30 cm (kg/ha)
Phosphorus in top 30 cm (kg/ha)
Available P
Total P
Lima-Kendaia TamaSilty Cecil Sandy
Silt Loam Clay Loam Loam
c
0.30
2
5
160
1.31
10
6.8
9,8
4.9
15.0
2500
B
0.28
2
3.5
160
1.22
28
6.7
13.2
6.6
16.4
2450
B
0.28
2
4
160
1.65
7
7.0
8.4
4.2
11.7
600
15
45
3100
15
140
2300
15
115
700
83
-------
(cont'cT
TABLE 6-7. SOIL AND CROP CHARACTERISTICS FOR STUDY LOCATIONS
New York
Iowa
Georgia
II. Crop Characteristics
(Conventional Management)
Type
Emergence
Harvest
Yield (kg/ha)
Plowing
Curve Number for Average
Antecedent Moisture
Fertilizer N (kg/ha)
Fertilizer P (kg/ha)
Time of Fertilizer
Application
Crop Uptake of N (kg/ha)
Crop Uptake of P (kg/ha)
Addition of Organic N
from Crop Residues
Continuous Continuous Continuous
Corn
June 1
Oct. 31
5020
Spring
Moldboard
88
90
35
Spring
120
20
40
Corn
May 15
Nov. 10
7850
Corn
May 15
Nov. 1
4080
Fall Spring
Moldboard Moldboard
81
200
80
Spring
185
30
65
81
140
35
Spring
100
20
35
The modelling of the effects of the legume (alfalfa) on the soil nitrogen
budget requires a special set of assumptions. It is assumed that during the sod
years the nitrogen fixed by the legume is not available for runoff or percola-
tion losses. During the corn years, it is assumed that an inorganic N*carry-
over from the legume would be available for crop growth and leaching losses
(runoff and percolation). The assumed carryovers in the corn years are given
in Table 6-8. The carryovers N is considered to be uniformly available from
May through October, and was modelled as discrete inorganic N inputs during
each of these months. The practical implication of these assumptions is that
predicted dissolved N losses in runoff and percolation are very low for the
legume years of the rotation.
The limited field data which are'available appears to support the pre-
dictions for runoff, but perhaps not for percolation. Moderate levels of per-
84
-------
TABLE 6-8. ANNUAL NUTRIENT INPUTS AND REMOVAL FOR ROTATIONS
Inorganic N
Year/Crop
New York
1 . Corn
2. "
3.
4.
5. Hay
6-8. "
Iowa
1. Corn
2. "
3. "
4. "
5. Hay
6-8. "
Georgia
1. Corn
2. "
3. "
4. "
5. Hay
6-8. Hay
Fertilizer
0
25
55
90
0
0
0
100
150
200
0
0
25
80
110
140
0
0
Legume
Carryover
130
65
35
0
--
—
230
115
60
0
--
--
115
60
30
0
--
--
Organic N
Crop
Uptake
Vcr/Via
Kg/ na-
120
"
II
II
110
190
185
n
"
n
165
280
100
n
n
"
95
160
Added in
Residues
40
n
ii
n
--
—
65
n
it
n
—
--
35
"
"
"
--
--
Available P
Fertilizer Crop
Uptake
35 20
n it
n n
n it
25 10
" 15
80 30
n n
n it
n ii
" 15
" 25
35 20
n n
n ii
n it
" 10
" 15
85
-------
eolation N losses have been observed for grass-legume mixtures and shallow-
rooted legumes, particularly after being turned under in the fall (Kilmer, 1974)
In the present modelling application, the sod crop is deep-rooted (alfalfa)
and the release of fixed N is explicitly modelled as a carry over effect in
the rotation years following the legume. Hence low predicted percolation N
losses are not necessarily inconsistent with field data. Nevertheless, the
uncertainty in the predictions must be recognized, since they have not been
validated.
Cover factors (C in the universal soil loss equation) were selected for
the appropriate crop stage from Wischmeier and Smith (1978). The conservation
practice factor (P) for terraces is based in division of the slope length into
2 terraces. As suggested in Stewart £t _aL (1975)'the resulting factors are
multiplied by 0.2 to provide estimates of sediment loss. This assumes that
80% of the eroded soil is deposited in terrace channels. With all practices,
it is assumed that crop residues are not removed after harvest.
Mean Annual Water and Sediment Losses
Mean annual runoff, percolation and sediment losses obtained from the
25-year simulations are shown in Table 6-9. The effects of contouring,
terracing and sod rotations in controlling sediment are similar for the three
locations, but conservation tillage effects are more site-specific. The
effects of all SWCPs on water movement are highly site-specific. For example,
terraces reduce runoff by 60, 30 and 40% at the New York, Iowa and Georgia
sites, respectively. The high runoff reductions for the poorly drained
New York soil suggests that SWCPs may be relatively more effective on soils
which are highly susceptible to runoff. All practices except sod rotations
are significantly more effective in controlling sediment than runoff. One
effect of runoff reduction is increased percolation, although the increase
is not always equal to the runoff reduction. In Iowa, runoff reduction may
make more water available in the upper soil layer for evapotranspiration and
hence only a portion of the additional infiltration percolates.
Average Annual Nutrient Losses
Mean annual nutrient losses obtained from the simulation runs are shown
in Table 6-10. Probably the most significant aspect of the table is that
the four SWCPs reduce every category of nutrient loss with the exception of
N in percolation. Moreover the percentage increases in percolation N are
small compared to reductions in runoff N. One of the practices, sod-based
rotations, achieves substantial reductions in all loss categories, since it
reduces runoff, soil loss and fertilizer applications.
In comparison with conventional management, conservation tillage is less
effective at reducing solid-phase nutrient losses in New York and Georgia
than in Iowa. This is because the soil is protected with residues during the
dormant season with all practices at the former locations, while corn residues
are plowed under in the fall with all practices except for conservation tillage
in Iowa.
86
-------
TABLE 6-9. MEAN ANNUAL RUNOFF, PERCOLATION AND SEDIMENT LOSSES FOR SELECTED SWCPs
Practice
New York
Conventional
Contouring
Terracing
Sod Rotation
Conservation Tillage
Iowa
Conventional
Contouring
Terracing
Sod Rotation
Conservation Tillage
Georgia
Conventional
Contouring
Terracing
Sod Rotation
Conservation Tillage
Runoff
Mean
(cm)
10
6
4
3
8
7
6
5
3
5
19
13
11
13
16
%a
Change
-40
-60
-70
-20
-15
-30
-55
-30
-30
-40
-30
-15
Percolation
Mean %
(cm) Change
39
43 +10
45 +15
42 +10
41 +5
25
26 +5
26 +5
27 +10
27 +10
58
64 +10
66 +15
64 +10
61 +5
Sediment
Mean
(MT/ha)
20
7
1
6
9
17
8
1
7
5
41
17
2
16
24
Change
-65
-95
-70
-55
-55
-95
-60
-70
-60
-95
-60
-40
87
-------
TABLE 6-10. MEAN ANNUAL NUTRIENT LOSSES FOR SELECTED SWCPs
co
00
Dissolved N
in
Runoff
Dissolved
P
in Runoff
Practice
New York
Conventional
Contouring
Terracing
Sod Rotation
Conservation Tillage
Iowa
Conventional
Contouring
Terracing
Sod Rotation
Conservation Tillage
Georgia
Conventional
Contouring
Terracing
Sod Rotation
Conservation Tillage
Mean
(kg/ha)
10
5
3
2
7
16
12
11
4
11
14
9
7
6
12
%a
Change
__.
-50
-70
-80
-30
...
-25
-30
-75
-30
._.
-35
-50
-55
-15
Mean
(kg/ha)
0.15
0.08
0.06
0.04
0.11
0.24
0.19
0.17
0.11
0.16
0.39
0.28
0.23
0.28
0.32
%
Change
-45
-60
-75
-25
_ _ _
-20
-30
-55
-35
_ — —
-30
-40
-30
-20
Dissolved
N in
Solid-Phase
N in Runoff
So lid- Phase
in
Runoff
Percolation
Mean
(kg/ha)
41
45
47
30
44
59
64
65
30
65
51
56
59
34
54
%
Change
.._
+10
+15
-25
+ 5
...
+10
+10
-50
+10
— _ —
+10
+ 15
-35
+ 5
Mean
(kg/ha)
93
34
4
29
42
82
37
5
33
24
41
17
2
16
24
%
Change
...
-65
-95
-70
-55
...
-55
-95
-60
-70
— _ —
-60
-95
-60
-40
Mean
(kg/ha)
33
12
1
10
15
23
10
1
9
7
17
7
1
6
10
%
Change
...
-65
-95
-70
-55
_> _ _
-55
-95
-60
-70
_ _ _
-60
-95
-65
-40
Rounded to nearest 5%
-------
The major nutrient losses are dissolved N in percolation and solid-phase
N and P in runoff. The solid-phase losses are sharply reduced by all SWCPs,
following the reduction of sediment loss. The reductions in dissolved N
and P in runoff from most SWCPs are significantly less than reductions in
particulate P and N. The exceptions are sod-based rotations which are equally
effective at controlling losses of nutrients associated with sediment and those
which are in solution with the runoff.
The reductions in runoff losses of dissolved nitrogen are almost exactly
balanced by increases in percolation N for contouring, terracing and conserva-
tion tillage. The significance of these percolation losses is conjectural.
If percolation waters subsequently emerge as base flow in streams, and the
leached nitrogen is conserved without dilution in the groundwater, these
three practices may not reduce stream nitrogen loadings in the long run.
However, it is unlikely that all the N leached from the top 30 cm of soil (as
predicted by the simulation model) will reach either groundwater aquifers or
streams. Substantial portions of the dissolved N which moves below 30 cm can
and will be taken up by plant roots, particularly corn and alfalfa roots.
Therefore, the 5-15% increases in percolation N losses probably should not be
considered significant.
Yearly Variations in Nutrient Losses
As noted in the introduction to this chapter, the effects of SWCPs can
be expected to vary from year to year, primarily due to weather conditions.
These variations can be considerable, as indicated by the simulation results
shown in Table 6-11. Annual losses of dissolved available P and N in runoff
are compared or the first 15 years of CNS model runs for Iowa. As is apparent,
the effectiveness of contouring in controlling these nutrient losses is highly
variable (0-69% for N, 0-50% for P). Moreover the relative effectiveness
of P and N control is not consistent (e.g., in 1953, N in controlled much
more than P, but in 1957 both are controlled equally).
These results suggest that SWCPs cannot be evaluated based only on two
or three years of data. This is equally true whether the nutrient loss data
are obtained from field measurements or simulation models. In either case,
studies must be of sufficient duration to produce the full expected range
of possible nutrient losses.
While mean annual losses are probably the most meaningful results which
can be obtained from the 25-yr simulations, other statistical parameters can
provide insight to the relative effectiveness of SWCPs. One of these para-
meters is the coefficient of variation of annual nutrient losses (standard
deviation of annual losses divided by mean annual losses). These coefficients
are summarized in Table 6-12. High coefficients indicate that the nutrient
losses associated with a practice vary substantially from year to year. It
can be noted from Table 6-12 that all the SWCPs studied tend to increase the
relative variability of dissolved nutrients in runoff, primarily due to the
practices' non-linear effects on runoff. Small runoff events are often pre-
vented completely by SWCPs while much larger events are only partially con-
trolled. This increase in a variability is not seen in solid-phase nutrient
losses, since rainfall erosivity remains the same for all practices. Dissolved
89
-------
TABLE 6-11. ESTIMATED EFFECTIVENESS OF CONTOURING IN CONTROLLING NUTRIENT
LOSSES OVER 15 YEARS IN IOWA
Year
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
Dissolved N in Runoff
Conventional Contoured %
(kg/ha) (kg/ha) Reduction
4.8
5.6
21.4
16.2
1.2
21.0
12.6
17.3
31.4
13.8
8.2
28.6
26.9
19.5
24.9
1.5
3.9
13.6
11.7
1.2
16.3
8.2
12.2
26.3
9.0
3.9
24.3
22.4
15.7
17.1
69
30
36
28
0
22
35
29
16
35
52
15
17
19
31
Dissolved
Conventional
(kg/ha)
0.12
0.13
0.43
0.15
0.01
0.27
0.14
0.23
0.41
0.27
0.08
0.35
0.28
0.29
0.26
P in Runoff
Contoured
(kg/ha)
0.06
0.12
0.35
0.12
0.01
0.21
0.09
0.18
0.35
0.20
0.04
0.30
0.25
0.25
0.19
%
Reduction
50
8
19
20
0
22
36
22
15
26
50
14
11
14
27
-------
N losses in percolation are much less variable from year to year, and except
for sod-based rotations SWCPs have little effect on coefficients of variation.
With the exception of dissolved P the variation in annual nutrient losses
associated with sod-based rotations significantly exceeds that of other SWCPs.
This is of course due to the large differences in dissolved N and sediment
losses between the sod and corn years in the rotation. Thus the practice
alternates (sod) years of very effective nutrient control with (corn) years
in which nutrient losses are only moderately lower than those associated with
conventional management. Although nutrient losses averaged over the eight-year
rotation are relatively low, sod-based rotations are of limited effectiveness
during the latter two years of the four-year corn sequence.
The results shown in Table 6-12 suggest that SWCPs are relatively less
effective at controlling dissolved nutrient losses in extreme or wet years
(i.e., years 'characterized by large runoff events) than they are at controlling
average annual losses. This conclusion can be demonstrated by examining
the probability distribution functions (cumulative probabilities) of annual
losses. As an example, these functions are shown in Figures 6-1, 6-2, 6-3 for
losses of dissolved P associated with three practices. If a wet or extreme
year is defined as a year in which nutrient losses are exceeded only one year
in ten, then the extreme losses can be approximated as the 90% values from
the probability distributions. These losses are summarized in Table 6-13,
which compares the effectiveness of two SWCPs in controlling average and
one-in-ten year losses of dissolved P in runoff. It can be seen that both
contouring and sod-based rotations offer less reduction of the one-in-ten
year dissolved P losses than of average annual losses, although the differences
in percentage reductions are relatively small for contouring.
The probability distributions can be used to compare practices in several
additional ways. One evaluation is a comparison of the probabilities that
losses will be kept below a certain level. For example, in New York, with
conventional management, there is only a 33% probability that dissolved P
losses in any year will not exceed 0.10 kg/ha. The probability increases to
69% and 92%, respectively, for contouring and sod rotations. The converse
to these probabilities is an assessment of risk. Thus the risk of dissolved
P loss greater than 0.10 kg/ha in any year decreases from 67% to 8% with a
sod rotation.
EVALUATION OF REDUCED TILLAGE, CONTOURING, TERRACING AND NO-TILLAGE FOR
CALIBRATED GEORGIA AND MICHIGAN WATERSHEDS
The ARM simulation model was used to estimate the effects of tillage
management practices on nutrient losses. Unlike the CNS model, ARM contains
several parameters that are directly related to tillage. Since the ARM
calibrations were for reduced or minimum tillage conditions, the moldboard
plowing of the previous section's "conventional management" could not be dup-
licated. Evaluations of the following four conditions were possible.
1. Reduced Tillage - (Base Condition) Row crops in straight rows parallel
to the land slope. Tillage of the soil is in the form of discing and
is done in the spring to prepare the soil for planting. Afer har-
91
-------
TABLE 6-12. COEFFICIENTS OF VARIATION OF ANNUAL NUTRIENT LOSSES
to
Practice
New York
Conventional
Contouring
Terracing
Sod Rotation
Conservation Tillage
Iowa
Conventional
Contouring
Terracing
Sod Rotation
Conservation Tillage
Georgia
Conventional
Contouring
Terracing
Sod Rotation
Conservation Tillage
Dissolved N
in Runoff
0.5
0.7
0.8
1.1
0.6
0.5
0.6
0.7
1.4
0.7
0.5
0.7
0.8
1.0
0.6
Dissolved P
in Runoff
0.5
0.6
0.7
0.8
0.5
0.5
0.5
0.5
0.7
0.6
0.5
0.7
0.8
0.7
0.6
Dissolved N
in Percolation
.1 n V rr /T"i n -r- VT*
K&/ Ua^Y-l
0.2
0.1
0.1
0.6
0.1
0.3
0.3
0.3
0.8
0.2
0.2
0.2
0.2
0.6
0.2
Solid-Phase N
in Runoff
0.6
0.6
0.6
1.1
0.6
0.5
0.5
0.5
1.4
0.5
0,6
0.6
0.6
1.3
0.5
Solid-Phase P
in Runoff
0.6
0.6
0.6
1.1
0.6
0.5
0.5
0.5
1.4
0.5
0.6
0.6
0.6
1.4
0.5
-------
0.05
0.10 0.15 0.20 0.25
DISSOLVED PHOSPHORUS (kg/ ha/ yr)
0.30
FIGURE 6-1. PROBABILITY DISTRIBUTION FUNCTIONS FOR DISSOLVED PHOSPHORUS
IN RUNOFF - NEW YORK .
-------
O.I 0.2 0.3
DISSOLVED PHOSPHORUS (kg/ha/yr)
0.4
FIGURE 6-2. PROBABILITY DISTRIBUTION FUNCTIONS FOR DISSOLVED AVAILABLE
PHOSPHORUS IN RUNOFF - IOWA-
-------
95
90
80
70
Ii 60--
CD
§ 50
ct
Q.
UJ
H
13
O
4O
30
20 ••
10--
SOO
ROTATION
CONTOURING
\
0.9
DISSOLVED PHOSPHORUS ( kg/ha/yr )
FIGURE 6-3. PROBABILITY DISTRIBUTION FUNCTIONS FOR DISSOLVED AVAILABLE
PHOSPHORUS IN RUNOFF - GEORGIA
-------
TABLE 6-13. COMPARISON OF THE EFFECTIVENESS OF CONTOURING AND SOD-BASED
ROTATIONS IN CONTROLLING AVERAGE AND EXTREME LOSSES OF DISSOLVED
P IN RUNOFF . .
Dissolved P in Runoff
Practice
Losses Exceeded
One Year in Ten
(kg/ha-yr)
Average
% Annual
Change Losses
(kg/ha-yr)
%
Change
New York
Conventional
Contouring
Sod Rotation
Iowa
0.26
0.16
0.10
-40
-60
0.15
0.08
0.04
-45
-75
Conventional
Contouring
Sod Rotation
Georgia
Conventional
Contouring
Sod Rotation
0.38
0.32
0.25
0.78
0.64
0.67
-15
-35
-20
-15
0.24
0.19
0.11
0.39
0.28
0.28
-20
-55
-30
-30
Rounded to nearest 5%
vest, the crop residue is removed from the field and the soil is
fallow through the winter. This situation is defined as "base condi-
tions" as described in Appendix F.
2. Contours - Identical to reduced tillage with the exception that rows
are perpendicular to land slope.
3. Terraces - Identical to contours with the watershed divided into two
terraces.
4- No-Tillage - Differs from reduced tillage in that there is no tillage
prior to spring planting and the crop residue is left on the field
after harvest.
Reduced tillage is essentially minimum or conservation tillage without
crop residues. The remaining conditions impose additional supporting practices:
contouring, terraces, tillage elimination and residue retention. Each of the
four alternatives was simulated using 10-year ARM model runs for watersheds
P2 and P6 in Georgia and Michigan, respectively (Table 6-1). These simulations
96
-------
are described in detail in Appendix F. The results, which are summarized in
Tables 6-14 and 6-15, indicate the changes in nutrient losses which can be
achieved by contouring, terracing and no-tillage as compared to tillage re-
duction alone. It should be noted that these results are not directly com-
parable to those obtained from the CNS simulations described in the preceding
section. Different watersheds and locations were simulated using CNS and the
standards of comparison for the two simulations are different ("conventional
management," implying moldboard plowing for CNS and reduced tillage or spring
discing for ARM).
The effects of contouring, terracing and no-tillage on runoff and sedi-
ment losses as predicted by the ARM model, are compared with reduced tillage
in Table 6-14. Runoff predictions are divided into two components, overland
flow and interflow. Contouring and no-tillage provide only minimal decreases
in runoff compared to the reduced tillage base conditions, but substantial
reductions in runoff are achieved by the addition of terraces. All practices
significantly reduce sediment loss, although the relative effectiveness of
the SWCPs is different for the Georgia and Michigan watersheds. For watershed
P6 (Michigan), the no-tillage practice is much more effective than countouring
or terraces, while in Georgia watershed P2 the effects of all practices are
somewhat similar.
The effects of the SWCPs on nutrient losses are shown in Table 6-15. The
practices all offer substantial control of solid-phase nutrients which travel
with the sediment. The most striking aspects of the simulations are the
minimal or nonexistant reductions in losses of dissolved nutrients in runoff
with contouring, terracing and no-tillage when compared to the effects of
reduced tillage alone. The increase in dissolved nutrients with no-tillage
can be partially explained by the nutrient additions from residues. The
absence of significant reductions in dissolved N losses in runoff may be due
to increased interflow, which leaches nitrate from the soil.
A general conclusion which can be made from these results is that the
addition of either contouring or terraces to tillage reductions is substantially
more effective at reducing particulate nutrient losses associated with sedi-
ment than dissolved losses in runoff. No-tillage will also reduce particulate
nutrient losses, but reduces neither dissolved N and available P in runoff.
COMPARISON OF CNS AND ARM SIMULATION PREDICTIONS
The CNS and ARM simulation results reported in the previous two portions
of this section cannot be directly compared since the locations, practices
and simulation periods were different for the two models. In an attempt to
evaluate the consistency of results produced by the two models, the CNS model
was used to simulate the same practices which were modelled by ARM (reduced
tillage, contouring, terracing and no-tillage) for the Georgia watershed P2.
Unfortunately, there are limitations on CNS's ability to model these practices.
The selection of curve numbers was somewhat arbitrary. The soil was assumed
to be in poor hydrologic condition for reduced tillage, contouring and
terracing. Fallow curve numbers for straight row cropping were used between
harvest and planting for these three SWCPs, since comparable numbers are un-
known for contouring and terracing. Curve numbers for good hydrologic condi-
97
-------
TABLE 6-14. MEAN ANNUAL RUNOFF AND SEDIMENT LOSSES FOR SELECTED SWCPs AS PREDICTED BY ARM MODEL
<£>
00
Practice
Watershed P2
(Georgia)
Reduced Tillage
Contouring^
Terracing"
No-Tillagec
Watershed P6
(Michigan)
Reduced Tillage
Contouring^
Terracing^
No-Tillage0
Runoff
Overland
Flow (cm)
18.6
16.4
11.1
17.1
16.5
15.4
11.2
15.6
Interflow
(cm)
1.3
1.4
1.7
1.5
2.6
2.7
3.2
2.7
Total
(cm)
19.9
17.8
12.8
18.6
19.1
18.1
14.4
18.3
% Change
in Total a
__-
-10
-35
- 5
_.
- 5
-25
- 5
Sediment Loss
Mean
(MT/ha)
6.3
4.0
2.8
3.3
3.5
2.8
2.4
1.0
% Change
_.
-35
-55
-50
— — .—
-20
-30
-70
a/
, .rounded to nearest 5%
/Spring discing, residues
C/Residues left
removed
•
-------
TABLE 6-15. MEAN ANNUAL NUTRIENT LOSSES FOR SELECTED SWCPs AS PREDICTED BY ARM MODEL
<£>
VO
Practice
Watershed P2
(Georgia)
Reduced Tillage
Contouring^
Terracing^
No-Tillage0
Watershed P6
(Michigan)
Reduced Tillage
Contouring^
Terracingb
No-Tillage0
Dissolved N
in
(kg/ha)
5.0
5.1
5.9
5.6
16.9
15.7
16.4
18.1
Runoff
3,
Change
_._
0
+20
+10
— — —
- 5
- 5
+ 5
Dissolved P
in Runoff
(kg/ha)
0.48
0.47
0.46
0.56
6.6
6.0
6.0
6.7
%
Change
- 0
- 5
+ 15
_ — _
-10
-10
0
Solid-Phase N
in Runoff (Organic N
+ Adsorbed NH4~N)
(kg/ha)
8.5
6.4
5.0
7.5
4.3
3.7
3.2
2.2
Change
-25
-40
-10
.__
-15
-25
-50
So lid- Phase
P04-P in Runoff
(kg/ha)
2.5
1.9
1.4
1.6
1.6
1.3
1.2
0.6
Change
-25
-45
-35
-20
-25
-65
b/
c/
Rounded to nearest 5%
Spring discing, residues removed
Residues left
-------
tions were used year-round for no-tillage. Simulations were based on the same
ten years of weather data which were used in the ARM simulations. The results
of the CNS simulations are compared in Tables 6-16 and 6-17 with those obtained
previously from ARM.
Runoff and Sediment Predictions (Table 6-16)
Runoff
The magnitudes of runoff predictions are comparable for both'models but
the relative effects of terracing and no-tillage are reversed. The small per-
cent change for terracing predicted by CNS is a consequence of using straight-
row curve numbers during the non-growing season. This is probably incorrect.
Conversely, the minimal reduction in runoff predicted by the ARM for no-tillage
is inconsistent with the literature results reported in Section 4.
Sediment
Sediment predictions differ by a factor of three, even though sediment
predictions for the watershed were well calibrated for ARM (Table 6-2) and
validated for CNS (Table 6-4). However, the sediment losses predicted by
ARM appear much too low, particularly when compared with observed losses
(Table 6-4). The discrepancy is inexplicable. However, the predicted relative
effects of SWCPs are very similar for both models.
Nutrient Predictions (Table 6-17)
Solid-Phase N and P
The predicted N and P losses from CNS are substantially greater than
those obtained from ARM except for no-tillage. The percent reductions in
these losses due to the SWCPs are somewhat greater with CNS, since the solid-
phase losses are assumed to be proportional to sediment. Based on the earlier
calibration and validation results, the CNS model predicts particulate nutri-
ent losses with better accuracy that ARM.
Dissolved P
Predictions of dissolved P losses are very similar for reduced tillage,
contouring and terracing. The no-tillage results differ substantially.
Since, as noted earlier, the CNS model does not consider the phosphorus dyna-
mics associated with plant residues, and the ARM model has not been calibrated
for no-tillage, the accuracy of predictigns of dissolved P losses for no-
tillage is questionable for both models.
Dissolved N
Perhaps the most interesting and significant differences in the two sets
of model predictions are in losses of dissolved N in runoff. The magnitudes
of N losses are very different, but this is due to the CNS model's consistent
over-prediction of these losses. Given the potential errors in both models,
percentage changes for contouring cannot be considered different. However,
ARM predicts that terraces and no-tillage will increase dissolved N in runoff
while CNS predicts reductions. These differences are important, since they
lead to different conclusions regarding the effectiveness of these SWCPs.
Most fundamentally, the CNS results imply that reductions in runoff decrease
dissolved N losses, but ARM predicts that runoff reductions increase these
losses.
100
-------
TABLE 6-16. COMPARISON OF ARM AND CNS RUNOFF AND SEDIMENT PREDICTIONS - WATERSHED P2
Total Runoff
Reduced Tillage
Contouring
Terracing
No- Till age°
(cm/yr)
19.9
17.8
12.8
18.6
ARM
o
% Change
-10
-35
- 5
(cm/yr)
22.2
21.4
20.8
14.6
CNS
% Change
- 5
- 5
-35
Sediment
ARM
(MT/ha-yr) %
6.3
4.0
2.8
3.3
Change
-40
-60
-50
Loss
CNS
(MT/ha-yr)
24.4
14.8
9.6
8.0
% Change
-40
-60
-65
a/
b/
c/
Rounded to nearest 5%
Spring discing, residues removed
Residues left
-------
TABLE 6-17. COMPARISON OF ARM AND CNS NUTRIENT LOSS PREDICTIONS - WATERSHED P2
o
to
Solid-Phase N in Runoff
Reduced Tillage
Contouring
Terracing
No-Tillage0
ARM
( kg/ha- yr) %
8.5
6.4
5.0
7.5
o
Change
-25
-40
-10
CNS
(kg/ha-yr) %
16.1
9.6
6.5
5.4
Change
.._
-40
-60
-65
Dissolved P in Runoff
Reduced Tillage
Contouring
Terracing
No-Tillage0
ARM
(kg/ha-yr) %
0.48
0.47
0.46
0.56
Change
_.
0
- 5
+ 15
CNS
(kg/ha-yr) %
0.57
0.55
0.52
0.38
Change
_.
- 5
-10
-35
Solid-Phase P in Runoff
ARM
(kg/ha-yr) %
2.5
1.9
1.4
1.6
CNS
Change (kg/ha-yr) %
9.0
-25 5.3
-45 3.5
-35 3.0
Change
_..
-40
-60
-65
Dissolved N in Runoff
ARM
(kg/ha-yr) %
5.0
5.1
5.9
5.6
CNS
Change (kg/ha-yr) %
14.6
+ 0 13'. 3
+20 10.8
+10 10.3
Change
._.
-10
-25
-30
a/
b/
c/
Rounded to nearest 5%
Spring discing, residues removed
residues left
-------
There may be several reasons for these differences, but the most signifi-
cant may be the handling of subsurface flows in the two models. With ARM,
a portion of the subsurface drainage water appears as interflow in runoff.
Increases in interflow, as with terracing and no-tillage (Table 6-14) will
leach additional soil N which is reflected in increases in runoff N. Con-
versely, interflow is not considered in the CNS model, and runoff reductions
are reflected in comparable increases in percolation. The effect of this
is shown in Table 6-18. Considering the total dissolved N loss in runoff
and percolation water, it can be seen that this total is not reduced by
contouring, terracing or no-tillage compared to reduced tillage.
TABLE 6-18. DISSOLVED N LOSSES FROM WATERSHED P2 PREDICTED BY CNS MODEL
Runoff
Dissolved N Loss
Percolation
Total
(kg/ha-yr) % (kg/ha-yr) %
(kg/ha-yr)
Change
Change
Change
Reduced Tillage
Contouring
Terracing
No- Till age
14.6
13.3
10.8
10.3
...
-10
-25
-30
47.4
48.7
51.3
53.6
...
+ 5
+ 10
+ 15
62.0
62.0
62.1
63.9
___
0
0
+ 5
a/
Rounded to nearest 5%
Summary of Model Comparison
In some ways this comparison of the two simulation models is an academic
excercise. The ARM model has not been validated and the CNS model is not well-
suited for predicting the effects of alternative tillage practices. Never-
theless, the excercise is informative. Although it does not appear that any
meaningful comparison of no-tillage is possible, the results for reduced
tillage, contouring and terracing yield several conclusions.
1. As indicated by the magnitudes of runoff, sediment and nutrient
losses, the two models produce significantly different results.
2. In spite of the absolute differences noted above, the relative
effects of adding contouring or terracing to a reduced tillage
base condition are consistent with both models for sediment,
particulate nutrients and dissolved P. Thus both models predict-
that for this watershed contouring or terraces will substantially
reduce losses of sediment and particulate nutrients but will have
negligible effect on dissolved P losses.
3. The substantial difference in model predictions of the effect of
terracing on dissolved N losses is due to assumptions concerning
subsurface water movement. The partioning of subsurface drainage
into percolation and interflow can significantly affect the
quantities of dissolved N in runoff.
103
-------
SUMMARY AND CONCLUSIONS
The results of these.simulation studies must be interpreted with care.
The ranking of SWCPs based on relatively small differences in predicted
nutrient losses is clearly inappropriate. The basic value of the modelling
results is the demonstration of consistent patterns of results. The con-
clusions presented in this section are based mainly on predictions of nutrient
losses obtained from the 25-year runs of the Cornell Nutrient Simulation (CNS)
model for comparable 2-hectare fields in New York, Iowa and Georgia. Although
the CNS model was found to have predictive errors, validation results indica-
ted that it is generally capable of estimating the relative effects of manage-
ment practices. The conclusions apply to corn grain cropping under five
conditions: conventional management ( moIdboard plowing, straight-row cropping),
contouring, terraces, sod-based rotations and conservation tillage. SWCPs
were all compared with conventional management. The corn stalk residues are
left on the field under all conditions. Sod-based rotations are four con-
secutive years each of corn and alfalfa. Conservation tillage is shallow
disc or chisel spring plowing.
The principal conclusions are as follows:
1. All SWCPs reduce all forms of nutrient loss except for dissolved N •
in percolation.
2. The effects of contouring, terracing and sod-based rotations on solid-
phase nutrient losses are similar for the three locations. The ef-
fects of these practices on dissolved N and P in runoff and dissolved
N in percolation varies significantly with location and can be con-
sidered site-specific. Conservation tillage is site-specific for all
nutrient forms.
3. Sod-based rotations are unique in their ability to significantly re-
duce losses of dissolved and solid-phase nutrients in runoff and
dissolved nutrients in percolation. Contouring, terracing and con-
servation tillage are substantially more effective at reducing run-
off losses of nutrients in the solid-phase form than in the dissolved
form. These three practices also increase percolation losses of
dissolved N.
4. The effectiveness of conservation tillage depends on the prevailing
management practice which it replaces. In Iowa, conventional manage-
ment was defined as fall moldboard plowing, which leaves the soil
bare during the non-growing period. Conservation tillage provides a
residue cover during this period and is hence fairly effective in
reducing dissolved N and P losses in runoff. Conversely, since
spring plowing was considered typical in New York and Georgia, resi-
due cover was equivalent for both conventional management and con-
servation tillage and hence the SWCP was relatively less effective
for these two locations.
5. Losses of all nutrient forms vary significantly from year to year.
More importantly, the effects-of SWCPs show similar yearly variations.
104
-------
ACKNOWLEDGMENTS
Computer program runs and data collection for CNS modelling studies
were carried out by Mathew N. Lorber and Lawrence J. Tubbs.
105
-------
SECTION 7
SIMULATION OF THE ACTION OF SOIL AND WATER CONSERVATION
PRACTICES IN CONTROLLING PESTICIDES
Tammo S. Steenhuis
Various studies have examined the effects of certain SWCPs on losses of
certain pesticides. Triplett et^ ah (1978) found that losses of atrazine and
cyanazine were less under no-tiTlThan under conventional tillage. Baker and
co-workers (1977, 1978) compared ridge planting, contour listing, and conven-
tional tillage as to their control of fonofos, cyanazine, and atrazine. Other
studies have compared: terraced fields with grassed waterways versus con-
touring for six pesticides (Smith et_ a]U 1978); ridge planting versus conven-
tional tillage for propachlor and diazinon losses (Ritter, 1971); different
rotations for pesticide losses (Epstein and Grant, 1968); and no-till versus
conservation tillage for pesticide losses (Harvey et^ al. 1976). More specific
information can be found in Sections 4 and 9.
These studies, all excellent by themselves, do not give a complete
picture of the interactions between SWCPs and pesticide loss. The studies
cover only a small subset of pesticides in current use. Some findings
appear to contradict each other, apparently because of climatic and/or soil
type differences. Considering the multitude of SWCPs and pesticides, the
only practical method of conducting a comprehensive study of SWCP control of
pesticide movement is mathematical simulation (Dean and Mulkey, 1978).
The actions of SWCPs on fields have been simulated to determine the
effectiveness of SWCPs in reducing pesticide loss in runoff and percolating
waters. Pesticides with different characteristics were simulated for this
study. In this Section, the results of these simulations are presented.
t
SIMULATION MODELS
Special Considerations for Pesticide Models
Modeling of pesticide losses differs from the nutrient and sediment
modeling presented in the previous two sections. First, nutrient, sediment
and pesticide characteristics differ widely. Nutrients and sediment are
continuously available for loss in runoff water. Concentrations of these
elements in the runoff water, therefore, depend on soil and climatic condi-
tions and not on the time of year per se. Pesticide losses, on the other
106
-------
hand, are greatest when rain occurs soon after application, and decrease
thereafter (Ritter et al_. 1974; Baker £t al. 1978). Thus, the effective-
ness of SWCPs can be quite different for pesticides, nutrients and sediment.
Second, pesticides usually are degraded before they are leached out of the
root zone (Freed and Hague, 1974). Losses in base and interflow are, there-
fore, of much less concern for pesticides than for nutrients.
Third, pesticides, especially those which are spray applied and not
incorporated, have a non-uniform distribution in the soil (Smith ^t al. 1978).
Therefore, there are high and low concentration areas. The pesticide"
chemistry in the high concentration areas is not well understood, and may
be different than when the concentration is uniform throughout the soil
(Donigian et^ a 1. 1977).
Previous Models
Modeling of pesticide loss in runoff water from agricultural land has
been attempted by Crawford e_t ah (1973), Donigian et_ aL. (1976, 1977),
Adams and Kurisu (1976), Frere et_ aL (1975), Bruce et_ al_. (1975), and
McElroy et al- (1976). All but the latter authors have developed continuous,
iterative simulation models that are capable of modeling intermediately and
strongly adsorbed pesticides. McElroy et^ aL, (1976) have developed a simple
loading formula that only applies to strongly adsorbed pesticides.
Two Different Models
In Section 6, the reasons for using more than one simulation model were
explained. In summary, models are no better than the assumptions on which
they are based. Which assumptions are best is not known at the present time.
It is hoped that the use of different models can prevent results from being
strongly biased.
Two models were used in this analysis of the pesticide losses. The
first was the Agricultural Runoff Management (ARM) Model developed by
Hydroc'omp, Inc., for the U.S. Environmental Protection Agency. The second
was a simulation model developed at Cornell for the specific purpose of
evaluating the effectiveness of SWCPs to control pesticide losses.
The ARM Model--
The ARM pesticide model, which is described in detail in Appendix F
computes water, soil and pesticide losses in surface runoff and interflow,
and downward displacement of the applied pesticides. The hydrologic and
sediment portions of the ARM pesticide model are identical to the counter-
parts in the ARM nutrient model.
The sub-model which handles the movement and degradation of pesticides
combines water and sediment flows with a single or multi-valued Freundlich
isotherm to determine the amounts of pesticide in solution and in sediment.
For modeling of methyl parathion, new algorithms based on the work of Gunther
(1977) and Brown (1978) were added by Hydrocomp to represent pesticide
storage, decay, and washoff from the crop cover. Attenuation of pesticides
through volatilization and degradation is simulated by a first order stepwise
degradation function that allows different degradation rates for separate
time periods. 107
-------
ARM Model Calibration--
The ARM model has been calibrated for two watersheds. The first is
located near Athens, Georgia, and has the code name of P2. It is one of four
plots located on the same type of soil but with different SWCP and cropping
patterns. The second calibration watershed, P6, is in Michigan, maintained
by the Department of Agricultural Engineering at Michigan State University.
Data on pesticide and nutrient loss were obtained under sponsorship of the
EPA laboratory in Athens. Characteristics of watersheds are presented in
Section 6. Data on pesticide applications are given in Table 7-1.
TABLE 7-1. PESTICIDE APPLICATIONS AT WATERSHEDS P2 IN WATKINSVILLE,
GEORGIA. AND P6 IN EAST LANSING. MICHIGAN
Cropping
Season
1973
1973
1974
1974
1975
Watershed
P2
(Watkinsville)
P6
(Michigan)
P2
P6
P2
Applied
Chemical
paraquat
atrazine
paraquat
atrazine
paraquat
atrazine
paraquat
atrazine
paraquat
atrazine
paraquat
atrazine
Application
Rate (g/ha)
1120
3360
1000
920
1120
3360
1120
4480
1120
2240
1120
1680
Date
5-11-73
5-11-73
6-09-73
11-05-73
4-29-74
4-29-74
5-22-74
5-22-74
11-08-74
11-08-74
5-21-75
5-21-75
Tables 7-2 and 7-3 compare the observed annual water, soil and pesticide
losses with those predicted by the model. The simulation results were
obtained from Donigian et^ al. (1977). Based on the limited data in Table 7-3
the ARM model does not appear to predict pesticide losses very well for
Michigan. The pesticide prediction losses for Georgia (Table 7-2) seem to
be more accurate. At both locations, annual runoff was predicted more
accurately than were sediment and pesticide losses.
The Cornell Pesticide Model (CPM)--
The Cornell pesticide model consists of five sub-models (Figure 7-1).
This approach was chosen so that model components could be changed easily
and so that experimental results could be substituted for input data. Model
components are discussed below and presented in detail in Appendix B.
The sub-model TEMPMELT takes daily weather records as input data and
then computes the rainfall or change in snowpack, if any. The snowmelt
108
-------
TABLE 7-2. OBSERVED AND ARM-PREDICTED RUNOFF SEDIMENT AND
Year
1973
1974
1975
observed
ARM
observed
ARM
observed
ARM
Runoff
(cm)
15.9
15.1
12.1
13.8
16.0
19.3
Sediment
(kg/ha)
11,400
6,900
2,200
4,400
5,400
5,800
Atrazine
(g/ha)
64 ,
43a,28b
5a>
10 ,
S3 7b
0,1
Paraquat
(g/ha)
169
271
79
242
183
242
t)ased on a multi-valued adsorption/desorption isotherm
based on a single adsorption/desorption isotherm
TABLE 7-3. OBSERVED AND ARM-PREDICTED RUNOFF SEDIMENT AND PESTICIDE
LOSSES FOR THE P6 WATERSHED IN EAST LANSING, MICHIGAN
Year
1973
1974
observed
ARM
observed
ARM
Runoff
(cm)
10.9
12.3
17.1
20.5
Sediment
(kg/ha)
1,859
4,163
5,638
5,687
Atrazine
(g/ha)
177a,196b
52 K
.a .b
4 , 4
Paraquat
(g/ha)
11
40
228
286
oased on a multi-valued adsorption/desorption isotherm
based on a single adsorption/desorption isotherm
formula is based on that of the U.S. Army Corps of Engineers (1960). Soil
temperature at 10 cm depth is predicted as a linear combination of the
running average of the previous 20 days' air temperatures (adjusted when
there was snow on the ground). The linear combination was then regressed
with the observed soil temperature for the years such data were available.
Hydrology--The hydrology sub-model HYDRO assumes, for well-drained, deep
soils, that soil water moves readily under gravitational force when its water
content is above field capacity. Thus, field capacity is defined here as the
moisture content below which water movement is small. It is further assumed
that water does not move downward when water content is below field capacity.
For shallow soils or where the water table is near the surface, the water
cannot move downward but must flow laterally. The number of days that the
soil requires to change from saturation to field capacity can be specified.
109
-------
TEMPMELT computes soil
temperature and snow-
pack
HYDRO computes surface
runoff and zone
moisture
SED computes
sediment loss
PEST computes pesticide
loss and downward
displacement
STATS performs
statistical analysis
FIGURE 7-1. FLOW CHART FOR CORNELL PESTICIDE MODEL.
110
-------
The structures of the four-layered root-zone model used is shown in
Figure 7-2. Zone 1 is the first 5 cm of topsoil. Zone 2 extends from 5 cm
depth to the bottom of the active rooting zone. Zone 3 represents the soil
below the active root zone and above the maximum depth of rooting. Zone 4
is the remaining soil depth to the impermeable layer. The above assumptions
are essentially equivalent to those used by Woolhiser (Stewart et_ al_. 1976).
Evapotranspiration for growing plants takes place out of Zones 1 and 2
and is assumed to occur at the potential rate if the soil has adequate water
(i.e. greater or equal to field capacity). Evapotranspiration is assumed to
be zero when the soil is at the wilting point. Actual evaporation rates are,
therefore, less than potential as the soil dries. Evaporation during the
non-growing season is set at the maximum potential rate when the soil is wet,
and decreases with the square root of time thereafter (Black et al. 1969).
The hydrology model has three options to partition rainfall into infil-
tration and runoff components when the soil temperature is above freezing.
The first option is based on the SCS runoff equation (Soil Conservation
Service, 1964; Stewart et^ aj^. 1976 ). Boundaries for the antecedent moisture
groups are similar to those of the CNS model presented in Section 6. The
second option uses infiltration rates based on the location of the wetting
front. This method was originally proposed by Green and Ampt (1911) and
later modified by Mein and Larson (1971) and Bouwer (1976). This option
requires additional input data for rainfall distribution during the storm.
Finally for shallow soils, a simple method is used that assumes that all
water infiltrates until all pores are filled and hereafter all water runs
off. This phenomenon has been observed to occur on soils with fragipans
(MacVicar, 1978, Dunne, 1978).
When soil temperature is below freezing, the infiltration model used in
the USDAHL74 watershed model (Holtan et^ al. 1975) was modified such that it
could be used in the SCS equation.
This hydrology model is similar in many respects to the simple model of
Stewart et^ al. (1976) and the model described in Section 6. The model does
not incorporate the relationships between potential gradient, hydraulic con-
ductivity and water flux in porous media, but instead uses the concept of
"field, capacity". With the inclusions of the option for shallow soil and
allowance for lateral drainage, approximations and assumptions included in
the model should not lead to serious errors.
Sediment--The sediment model (SED) uses the Universal Soil Loss Equation
with the kinetic energy term computed by either the Onstad and Foster (1975),
Williams and Berndt (1977) or Wischmeier and Smith (1965) method. The sedi-
ment models are the same as those described in Section 5. The precision and
accuracy of the models are also described in Section 5.
Pesticides—The pesticide model consists of three parts. The first
part is the prediction of pesticide losses in runoff water. The key assump-
tion of this part is that there exists a certain "mixing" layer at the soil
surface where the concentrations of pesticide in the percolating water,
111
-------
I-1
10
UUKhACt
5,
!§•
25-
FH 35.
1)
45-
55-
65-
~T K
f O •
^
1
C
ZONE 1
•
p
ZONE 3
•
MB
*
MAXIMUM RO
IMPERMEABLE
OTING DEPTH
ZONE
2
ZONE 3
ZONE 4
LAYER
) 25 36 43 5
A
60 HARVEST
DAYS AFTER PLANTING
FIGURE 7-2. SOIL ZONE BOUNDARIES ,
-------
runoff water, and soil water are equal. Theory for predicting the concen-
tration of pesticide in the runoff water is presented elsewhere (Steenhuis
and Walter, 1978).
The second part of the model simulates downward displacement of pesti-
cides. It is based on a simple model developed, used and tested success-
fully by the following authors: Gardner (1965), Bower et^ al_. (1957), and
Rao et al. (1976). The displacement of the midpoint of the pesticide band
can be found by multiplying a retardation factor by the amount of water
passing the band. The value of the retardation factor depends on field
capacity of the soil and the adsorption partition coefficient.
The third part of the pesticide sub-model handles the attenuation of
pesticide through volatilization and degradation between rainstorms. It is
similar in form to that used in the ARM model.
Statistics—The statistical program which was developed as part of this
project, summarizes the results produced by the modeling program.
Calibration and Validation—Most of the sub-models of the Cornell
pesticide model were developed and tested for accuracy in following modeling
studies. The soil temperature, runoff from shallow soils, and pesticide
runoff sub-models are exceptions to the above. Predicted and observed soil
temperatures at 10 cm depth for the years 1972 and 1973 are shown in Figure
7-3. A good agreement can be noted.
The SCS and Green and Ampt equations used in the hydrology model are in
wide use for runoff and infiltration prediction in non-frozen soils. The
SCS runoff equation has been tested many times and found to give reasonable
results (Stewart et_ aK 1976; James et_ al^. 1977; Tubbs and Haith, 1977;
Soil Conservation Service, 1964). Results presented in Section 6 show that
the SCS equation predicted yearly runoff losses somewhat better in New York
than in Georgia.
The Green and Ampt equation has recently been reported to predict
runoff well (Li et^ al^. 1976; Chu, 1977; Idike et_ a^. 1977). Simulation
results of the CPM model in Table 7-4 show that both sub-models varied at
times from the observed losses.
For New York, the SCS runoff equation was not used, but instead, the
sub-model for shallow soils which assumes that water removal is
a function of lateral movement and evapotranspiration. Draining from com-
plete saturation to field capacity was assumed to occur in 20 days. This
corresponds approximately to the time that small tributaries of rivers are
flowing after the soil is saturated in New York.
The model for shallow soils was verified for Aurora. Observed values
for runoff on Aurora plots with a hard pan at 40 cm depth were compared with
runoff values predicted by the shallow soil runoff model for the years 1971-
1974 (Figure 7-4). A reasonable agreement was obtained for the first two
years between predicted and observed values of runoff for well managed plots
113
-------
1
ta
a.
A actual temperature
O predicted temperature
650
700
750
DAYS
FIGURE 7-3. SOIL TEMPERATURE AT AURORA, N.Y., JAN. 2, 1972 THROUGH Dec. 27, 1973 AT 10 cm.
-------
on shallow soil. The shallow soils model's runoff predictions were as good
as those of the SCS curve number equation.
Three soil loss equations were compared in Section 5. It was shown that
the Onstad Foster equation (Onstad and Foster, 1975) gave more accurate
monthly values for sediment loss than did either the Universal Soil Loss
Equation (Wischmeier and Smith e£ al. 1965) or the Williams equation
(Williams and Berndt, 1977). The disadvantage of the Onstad Foster equation
is that sediment loss can theoretically occur on days when there is no runoff.
Thus, the Onstad Foster equation is preferable only when annual or monthly
losses are desired, as in the 'nutrient simulation of Section 6. The CPM
model predicts daily losses of pesticides. Therefore, the CPM predicts soil
loss with the Williams model, which restricts its predictions to days when
runoff occurs.
The pesticide sub-model has three components, two of which have been
validated extensively. The first-order degxadation model has been tested by
12
10-
8
RUNOFF
(CM)
6
4
2
0
1
vXXXXXXXXI
i
xxxxx^^
| | OBSERVED
VTA PREDICTED
77% iZX
1970-1971 1972 1973 1974
YEAR
FIGURE 7-4. OBSERVED AND PREDICTED LOSSES FOR AURORA, NEW YORK.
115
-------
TABLE 7-4. OBSERVED AND PREDICTED RUNOFF LOSSES FOR P2
WATERSHED IN WATKINSVILLE, GEORGIA
JAN
FEE
MARCH
APRIL
MAY
JUNE
JULY
AUG
SEPT
OCT
NOV
DEC
Year
Total
Actual
8.0
3.6
2.0
0
1.1
0
0
1.2
15.9
Green
SCS Ampt
1973
8.
0.
0.
0
0.
0
0.
4.
14.
7
1
2
2
1
9
2
5.5
3.5
1.1
0
1.8
0
0
0.9
12.8
Actual
DI in rt
- — — — — KUIIO
0
0.4
0
0.7
0.8
4.2
4.6
1.2
0
0
0
0.2
12.1
SCS
1974
ff (era
0
2
0
0
1
0
0
0
0
0
0
1
8
.4
.1
.1
.7
.9
.9
.2
.8
.1
.2
Green
Ampt
1
tj —
0
0
0
0.5
1.8
5.1
5.1
2.7
0
0
0
0.1
15.3
Green
Actual SCS Ampt
1975
0.2 1.7 0
1.0 2.3 0.3
4.8 7.1 2.8
1.7 3.2 2.2
0.9 1.1 0.9
4.1 1.7 3.7
3.2 0.1 1.3
0.1 0.1 0.9
16.0 17.3 12.1
Donigian e^t al. (1977) and found to be superior to other methods. The for-
mulas which determine leaching of pesticides were shown to predict the down-
ward movement of substances reasonably well (Rao e£ aK 1976). The pesti-
cide runoff algorithm is based on parameters that have a physical basis and
thus can be measured. An exception is the mixing depth, for which no easy
measurable technique exists. A value of 0.9 cm for mixing depth was
obtained by Steenhuis and Walter (1978). Their computational procedure
incorporated values of alachlor and cyanazine losses measured by Baker et al
(1978). '
Data observed by Bailey e£ aL (1974) were used to validate the pesticide
runoff algorithm. Predicted and observed losses were compared by Steenhuis
and Walter (1978). A summary of these results is given in Table 7-5. Although
observed and predicted losses agreed rather well, the predicted losses were
somewhat lower in most cases.
116
-------
The fonofos concentration in the runoff water measured by Baker et al.
(1978) was used to test the sensitivity of the pesticide model to SWCPs".
Predicted and observed losses are shown in Figure 7-5 as a percentage of the
conventional tillage loss. Trends in pesticide loss under different practices
were predicted satisfactorily, as indicated by a correlation coefficient of
73% when observed losses were regressed with predicted losses. However, as
shown in Figure 7-5, the observed losses under conventional tillage were
lower than for either buffalo till or chisel plowing, while the reverse was
true for the predicted losses.
COMPARISON OF THE ARM AND CPM MODELS - 1972-1975
To compare the ARM and CPM predictions of runoff and pesticide loss,
both models were run using input data from watershed P2. The ARM model
first was calibrated to the watershed. Input data for the ARM simulations
are given in Donigian et^ aL (1977).
The CPM used the Green and Ampt sub-model to compute runoff. Input for
the Green and Ampt equations was obtained from the literature: wetting front
suction data from Li et al. (1976) and saturated hydraulic conductivity from
England (1970) . The CPM~~sediment loss model was based on that of Williams
(Williams and Berndt, 1977). Most of the CPM adsorption isotherm for atra-
zine was made from observed concentrations of the chemical in runoff and
sediment (Smith et_aJU 1978). Outside the range of the observations, the
ARM adsorption isotherm was used (Donigian et_ al_. 1977). The paraquat
isotherm was the same as in Donigian et_ a_L (1977) adjusted for the difference
in mixing depth. Input to the CPM is summarized in Appendix B.
TABLE 7-5. PESTICIDE LOSSES FROM GEORGIA PLOTS, OBSERVED
BY BAILEY £1 AL.(1974) AND PREDICTED BY THE CPM
Predicted
Plot 2
Plot 5
Plot 6
Plot 7
Atrazine
DCBN
Atrazine
DCBN
Atrazine
DCBN
Atrazine
DCBN
total in
runoff
*
*
8
7
9
7
8
5
total in
sediment
*
*
1
1
1
1
1
2
Observed
total in
runoff
5
2
11
8
10
6
7
2
total in
sediment
2
2
2
2
3
3
2
* No estimate could be obtained for adsorption partition coefficient
** Trace
117
-------
Comparison of Results
Simulated runoff losses by ARM and CPM both appear not to be very
accurate when compared to observations (Table 7-6). The major shortcoming
of both models is that they assume homogeneous fields and overland flow
produced equally over the whole watershed.
OBSERVED
UJ
< T40-
_,
p
_l 120-
z
0
r~
z 100-
UJ
>
z
o
O PQ~
o
u.
u.
g 60-
ac
~ 40-
co
CO
o
H 20-
z
1 1 1
UJ
O
cr
UJ
U.
-i £
< n
o
H
UJ
Z
o
o
•J
r
i
UJ
(O
X
o
_
i
u
c
, n
• f 1
' ^
— b.
E
l__l
J
^
^ °
: L
: H
_
•1
c
c
WITH
c
j
i^
j
3
>
^
PREDICTED
| PESTICIDE TRANSPORTED
WITH SEDIMENT
PESTICIDE TRANSPORTED
z
o
H
Z
LJ
20 40 60
RESIDUE COVER, %
C
0
> -
> i-
o
1
CD
">,
4
.
ISIHDi-
^
(J
^H
C
'
I
>0
f
m
)
\ U
C
c
Q
J
>
3
»
s
H
O
0
I 1
40 60
FIGURE 7-5. PESTICIDE LOSS IN RUNOFF AS
PERCENT OF LOSS WITH CONVENTIONAL TILLAGE.
118
-------
TABLE 7-6. ARM AND CPM PREDICTIONS OF RUNOFF FOR THE P2
JAN
FEE
MAR
APR
MAY
JUNE
JULY
AUG
SEPT
OCT
NOV
DEC
Year
Total
Observed
8.0
3.6
2.0
0
1.1
0
0
1.2
15.9
ARM*
1973
7.0
2.5
0.8
0
1.8
0
0
3.0
15.1
CPM
5.5
3.5
1.1
0
1.8
0
0
0.9
12.8
Observed
Runoff
0
0.4
0
0.7
0.8
4.2
4.6
1.2
0
0
0
0.2
12.1
ARM
1974
(s
cmj -•
0.5
0.8
0
0.2
1.8
4.1
4.4
1.9
0
0
0
0.1
13.8
CPM
0
0
0
0.5
1.8
5.1
5.1
2.7
0
0
0
0.1
15.3
Observed ARM CPM
1975
0.2 0.3 0
1.0 2.0 0.3
4.8 7.9 2.8
1.7 2.8 2.2
0.9 2.3 0.9
4.1 3.0 3.7
3.2 1.0 1.3
0.1 0 0.9
16.0 19.3 12.1
*Donigian et^ al. 1977
Simulated atrazine and paraquat losses in runoff water and sediment are
compared with observed values in Table 7-7. Annual atrazine losses were pre-
dicted better by the ARM than by the CPM model. Paraquat losses were some-
what better simulated by the CPM than by the ARM model.
The CPM is cheaper to run than is the ARM model. A 25-year simulation
by the CPM uses less than 1 minute of computer time on an IBM 370/168, at a
cost of less than 15 dollars. A 25-year simulation by the ARM cost at least
10 times as much. Furthermore, once soil loss and runoff have been simulated,
the effect of pesticide characteristics on pesticide loss can be found within
20 seconds of computer time using the CPM. ,
The ARM model can predict pesticide losses at any time during the storm.
The CPM does not have this capability. The CPM variables have a physical
basis. Therefore, the CPM input data are available in the literature whereas
the ARM must be calibrated for each watershed. For this reason, the CPM is
more appropriate for the studies of regional variations of the effects of
SWCPs on pesticide losses.
119
-------
TABLE 7-7. ARM AND CPM PREDICTIONS OF PESTICIDE LOSSES (GRAMS)
FOR THE P2 WATERSHED IN WATKINSVILLE, GEORGIA
Atrazine
Sediment
Actual CPM1 ARM2
Atrazine
Solution
Actual CPM1 ARM2
Paraquat
Total
Actual CPM1 ARM2
1975
May
June
July
Aug
Sept
Total
1974
May
June
July
Aug
Sept
Total
1975
10
1
0
0
_^
11
1
0
0
0
0
18
1
0
0
_0
19
1
0
0
0
0
2
0
0
0
0
0
0
0
0
0
65
5
1
0
_0
71
6
1
0
0
0
63
0
0
0
_0
63
4
0
0
0
0
50
4
0
0
jO
54
0
0
0
0
1
188
23
7
0
0
218
23
47
27
5
0
102
predicted by the Cornell Pesticide Model
2
predicted by the Agricultural Runoff Model by Hydrocomp
based on non-single valued adsorption isotherm
*too high because of over-prediction of runoff by CPM
129
18
35
0
12
50
103
98
40
0
294
29
14
0
13
194 350
91
164
42
15
0
291 312
May
June
July
Aug
Sept
Total
1
1
0
0
0
2
1
0
0
0
0
1
0
1
0
0
0
1
1
10
0
0
0
11
37*
26
0
0
0
63
1
8
0
0
0
9
20
191
25
0
0
236
37
200
50
0
13
300
34
237
27
15
0
313
EVALUATION OF THE EFFECTS OF SWCPs ON PESTICIDE LOSSES
The ARM and CPM were used to evaluate the effects of management practices
on pesticide losses. Atrazine, paraquat, and methyl parathion losses were
modeled using the ARM for a 10-year period with the same management practices
and locations as in the nutrient simulations described in Section 6 and
Appendix F. Four groups of pesticides were modeled using the CPM for 25-year
periods at locations in Georgia, Iowa and New York. Two management practices
120
-------
•were evaluated: a) straight row planted corn, and b) corn on terraces
combined with contour planting.
The ARM Comparisons
The ARM model, which has been calibrated and tested on the P2 (Georgia)
and P6 (Michigan) watersheds, was used to model three pesticides for the 10-
year period 1966 through 1975 at these two locations. The management prac-
tices were: a) base condition (straight row corn, disking in spring and no
plant residues during the winter), b) contours (rows perpendicular to the
land slope), c) terraces combined with contours, and d) no tillage (crop
residue left on field after harvest, and no tillage prior to spring planting)
The ARM model is not directly sensitive to SWCPs. Therefore, input
parameter changes were based on either the SCS runoff equation, the Universal
Soil Loss Equation or Hydrocomp's experience of the model's behavior on other
watersheds. The estimation procedure for these input parameters is presented
in Appendix F.
Atrazine, paraquat, and methyl parathion were chosen because they are
representative of broad groupings of pesticides. Atrazine, as modeled by
Hydrocomp, is an intermediately adsorbed pesticide with a relatively short
half-life (15 days). Paraquat is a strongly adsorbed pesticide with high
persistence (half-life of 360 days). Methyl parathion, which was applied
to the canopy six times per year, is intermediately adsorbed and has an
intermediate half-life (about 40 days). Application days and rates are
given in Table 7-8.
TABLE 7-8. PESTICIDE APPLICATIONS ON WATERSHEDS P2 and P6
FOR ARM MODEL SIMULATIONS
Atrazine Paraquat Methyl Parathion
Date Rate Date Rate Date Rate
Applied (kg/ha) Applied (kg/ha) Applied (kg/ha)
P2 May 9 2.9 May 9 2.5 June 15
June 29
July 13
July 27
August 10
August 24
1.1
1.1
1.1
1.1
1.1
1.1
P6 May 20 2.8 May 20 1.5
ARM Simulation Results and Discussions
Table 7-9 summarizes the mean annual runoff and pesticide losses (in
sediment and in solution) predicted by Hydrocomp, Inc. for each of the mana-
gement practices on the P2 and P6 watersheds. Since application rates for
the two watersheds were not equal, results are also given in percent of total
121
-------
pesticide applied. The results show that the annual runoff pesticide losses
in Georgia and Michigan were decreased when any of the three SWCPs were incor-
porated. Runoff was reduced by 6 to 11% for contours and by 25 to 36% for
terraces (Table 7-10). There were lesser reductions (6%) for no tillage.
Simulated sediment losses for the management practices reflect the reductions
in surface runoff. For both locations, the percent change in sediment loss
from base condition for contouring was 21 to 37%, for terraces 32 to 55%; and
for no tillage 48 to 71%. These changes are comparable to those reported by
others as noted in the following paragraphs.
Annual runoff losses for corn planted on the contour in experiments in
Illinois, Missouri, New Jersey and Ohio (Stallings, 1945) were approximately
TABLE 7-9. P2 AND P6 MEAN ANNUAL RUNOFF AND POLLUTANT LOSSES UNDER
DIFFERENT MANAGEMENT PRACTICES MODELED BY THE ARM MODEL
FOR A 10-YEAR PERIOD
Base
Condition
Minimum
Tillage
Contours
Terraces and
Contours
Runoff*(cm)
Sediment Loss
(mt/ha)
PESTICIDE LOSSES
Agrazine
in solution
(g/ha)
as % of applied
on sediment
(g/ha)
as % of applied
Methyl Parathion
in solution
(g/ha)
P2
19.9
6.3
53.1
1.8
0.4
0.0
283
P6
19.1
3.5
77.4
2.8
1.8
0.0
as % of applied 4.3
on sediment
(g/ha) 23
as % of applied 0.4
Paraquat**
total (g/ha) 341 115
as % of applied 13.6 7.7
P2
18.6
3.3
46.9
1.6
0.3
0.0
246
4.2
18
0.3
188
7.5
P6 P2 P6
18.3 17.7 17.9
10. 4.0 2.8
75.7 40.6 61.7
2.7 1.4 2.2
1.1 0.3 1.3
0.0 0.0 0.0
241
3.6
15
0.2
36 225 93
2.4 9.0 6.2
P2 P6
12.8 14.4
2.8 2.4
18.8
0.6
0.2
0.0
145
2.2
10
0.2
162
6.5
28.9
1.0
0.9
0.0
81
5.4
* Overland flow plus interflow
**Paraquat losses in the solution phase are assumed negligible
122
-------
TABLE 7-10. P2 AND P6 MEAN ANNUAL RUNOFF AND POLLUTANT LOSSES UNDER
DIFFERENT SWCPs AS PERCENT CHANGE FROM BASE CONDITION LOSSES
MODELED BY ARM FOR A 10-YEAR PERIOD
Runoff*
Sediment loss
Pesticide Losses
Minimum
Tillage
P2 P6
- 6 - 4
-48 -71
Contours
P2 P6
-11 - 6
-37 -21
Terraces and
Contours
P2 P6
-36 -25
-55 -32
Atrazine
in solution -12
in sediment -17
combined -12
Methyl Parathion
in solution -13
in sediment -34
combined -15
Paraquat -45
- 2
-40
- 3
-68
-24
-32
-24
-14
-45
-17
-34
-20
-31
-20
-19
-65
-61
-64
-49
-66
-50
-52
-63
-53
-62
-29
* Overland flow plus interflow
30% lower than losses for corn planted in straight row. Contours at these
same locations reduced soil loss by about 50%. More recent studies (Stewart
et_ aJ_. 1976) showed reductions similar to those quoted by Stallings (1945).
The effect of terraces on runoff and soil loss depends mainly on the type
of terrace and the permeability of the soil (Baver et^ ajL 1972). Studies
quoted by Stallings (1945) and in Stewart et_ aL (1976) showed that terraces
were more effective in reducing soil loss than runoff. Data presented by
Stallings (1945) show that terraces in most cases reduced runoff 20 to 50%
as compared to runoff from unterraced fields in Arkansas, Iowa, Missouri,
Oklahoma and Texas. However, runoff was approximately 20% greater on the
terraced than on the unterraced fields in Ohio and Wisconsin. Soil loss for
all locations was reduced, with reductions ranging from 35% on a 3% slope in
Iowa to 98% in Missouri on an 11% slope.
The reductions in runoff and soil loss predicted by the ARM simulation
for terraces and contours, although somewhat small, agree reasonably well
with the above observations. However, reductions in runoff and sediment
loss under no tillage seem to be underestimated. Jamison et^ aL (1968) and
Free and Bay (1969) reported reduction of less than 15% for runoff. However,
their results were for a soil with'a fragipan, and are not applicable to the
123
-------
soils of Georgia and Michigan. Most other studies, such as those by Harrold
et_aL (1970), and Shanholtz and Lillard (1968), indicate reductions in runoff
surpassing 40%. No-tillage reductions in sediment loss are most times greater
than 90% (Harrold and Edwards, 1972; Harrold ejt aK 1970).
In the ARM model, pesticide losses are directly related to runoff and
sediment losses. Therefore, if runoff and sediment losses are not predicted
correctly, neither will be the pesticide losses. Predictions of pesticide
losses under no-tillage should, therefore, be looked at critically. Simulated
average annual overland losses of methyl parathion for the base condition are
4.6% of total amount applied. The few experimental studies currently available
indicate that runoff and sediment losses of methyl parathion are minor and
amount to a maximum of 0.1% of the applied material (Sanborn et^ al_, 1976,
Pesticide manual, 1971; Von Rumker, 1975). The ARM model seems, based
these few studies, to overpredict methyl parathion losses. Cause for this might
be the difference in drift and volatilization between actual and simulated
conditions. Drift and volatilization losses are high in the field (Hague
and Freed, 1974, Quinby et_ al^ 1958) but neglected in the ARM model. Another
cause for the higher simulated losses is that the degradation rate in ARM is
lower than observed under field conditions. Lichtenstein and Schulz (1964)
found that only 8% of the 5 pounds per acre of methyl parathion remained
15 days after application. King and McCarty (1968) estimating the theoretical
half-life have shown similar results. ARM model assumed half-life for
methyl parathion degradation of 49 days.
Methyl parathion and atrazine are mainly lost as solutes transported
in the runoff water (Table 7-9). Therefore, total reductions of pesticide
losses are more related to the reductions in runoff water than in the
sediment (Table 7-10).
ARM simulation results show that contours combined with terraces in
Georgia and Michigan are the most effective in reducing the losses of inter-
mediately adsorbed pesticides such as atrazine or methyl parathion. Contours
alone are not as effective in reducing losses of these pesticides as terraces
combined with contours are.
Paraquat is strongly and irreversibly adsorbed to soil particles and,
therefore, paraquat loss corresponds to sediment loss. Small variations
are caused by the quantity of fines transported in the runoff--the fines
are the relatively strongest adsorbing particles. Therefore, minimum tillage,
which has its greatest effect on sediment transport, is the most effective
practice for reducing paraquat losses.
Three principal conclusions can be drawn from the ARM simulations:
1) The changes in loss caused by SWCPs are simulated for the strongly adsorbed
pesticides such as paraquat and the intermediately adsorbed pesticides such
as methyl parathion and atrazine; 2) Absolute quantities of paraquat are more
reduced than either atrazine or methyl parathion; 3) Average annual pesticide
losses form a small percentage of the total amount of applied pesticide.
124
-------
The CPM Simulation's 25-Year Runs
In this section, the effect of two SWCPs and four groups of pesticides
on soil, water and pesticide losses are evaluated at three locations over a
25-year period by the CPM model. The two SWCPs, terraces and straight row,
represented the good and poor extremes in terms of preventing pollutant
losses in overland flow. The pesticide classification was chosen such that
most pesticides currently on the market were represented.
The CPM simulations were made for the same locations and years (1952-
1976) as for the CNS simulation in Section 6 and the erosion studies in
Section 5. Relevant input parameters were kept approximately equal to those
of the two preceding sections with the exception of slope, which was kept constant
at 2% to facilitate the evaluation of climatic effect of pesticide loss. Soil
and crop characteristics and pesticide application dates are summarized in
Table 7-11.
Local climatic data were input for Watkinsville,. Georgia; Ames, Iowa;
and Aurora, New York. Runoff and infiltration simulations in Georgia and
Iowa were based on the SCS curve number method. For New York, the sub-
program for prediction of runoff on shallow soils with a hard pan was used.
Erosion was modeled with the Williams model (Williams and Berndt, 1977) for
reasons given before.
Simulations were done for four groups of pesticides with similar pro-
perties:
Group A weakly adsorbed and easily degradable
Group B intermediately adsorbed and easily degradable
Group C intermediately adsorbed and intermediately persistent
Group D strongly adsorbed and very persistent
Although many classifications are possible, it was thought that the above class-
ification was the most suitable for evaluating pesticide loss in runoff. To
demonstrate grouping according to persistence and adsorption strength, selected
pesticides are classified in Table 7-12.
In the CPM simulation studies, all pesticides were applied on May 10 at
a rate of 3 kg active ingredient per ha. In a seperate simulation, the Group
B pesticides were also applied on June 10 at 3 kg per ha.
Complete simulations were made for straight-row-pianted corn in poor
hydrological condition and for corn planted on the contour combined with
terraces and a good hydrological condition. These two cases are the worst
and best, respectively, in terms of preventing loss.
Water and Sediment Losses--
Simulation results are presented as average annual values and as monthly
frequencies. The average annual value is the cumulative 25-year value divided
by 25. Monthly frequencies are the 25-year totals of the number of times an
event occurs in each month of the year.
125
-------
TABLE 7-11. SOIL AND CROP CHARACTERISTICS FOR STUDY LOCATIONS
Georgia
Iowa
New York
I. Soil Characteristics
Cecil Sandy
Loam
Tama Silt
Loam
Kendaia Silt
Loam
Hydrologic group B
Erodibility 0.30
Area (ha) 1.3
Slope (%) 2
Slope length (m) 180
Bulk density (g/cm3) 1.65
Field capacity (cm3 • cnr3) 0.28
Saturation (cm3 • cnr3) 0.39
Wilting point (cm3 • cm-3) 0.14
Depth to impermeable layer (cm) n.a.
II. Crop and Field Data
1
2
180
B
0.30
3
1.22
0.44
0.55
0.22
n.a.
n.a.*
0.30
1.3
2
180
1.31
0.33
0.50
0.16
40
crop
plowing time
pesticide applications
emergence data
harvest date
CN fallow season (AMC = 2)
CN growing season (AMC = 2)
maximum rooting depth (cm)
continuous
corn
spring plow
May 10**
June 1
Nov. 1
86
81
75
continuous
corn
spring plow
May 10**
May 15
Nov. 10
86
81
135
continuous
corn
spring plow
May 10**
June 1
Nov. 1
n.a.
n.a.
40
* not applicable
** June 10 pesticide application also modeled for group B pesticides
Precipitation Analysis--
The average annual precipitation at the Georgia, Iowa, and New York
locations for the 25-year simulation period (1952-1976) was 129, 80 and
88 cm, respectively. The frequency analysis for rainfall in these locations
is presented in Table 7-13.
Climates in Iowa and Georgia are characterized by heavy thunderstorms in
late spring. There are relatively few heavy rainfalls in the summer. Highest
rainfall intensities occur, thus, at the time that Aost herbicides are applied
and the soil is not protected by plant cover. The climate in New York is
characterized by many days of precipitation, but the rains are usually of low
intensity. Only in summer do heavy rainfalls occur. As opposed to Georgia,
winter precipitation in Iowa and New York is often in the form of snow. Thus,
Georgia receives the most rain and the most severe storms.
126
-------
TABLE 7-12. CLASSIFICATION OF SELECTED PESTICIDES*
Group A Group B Group C Group D
weakly adsorbed intermediately intermediately strongly adsorbed and
and easily adsorbed and adsorbed and very persistent
degradable easily intermediately
degradable degradable
Chlorambenb AtrazineC'd AtrazineC'd Paraquat f
Carbarylee TrifluralinC'e Toxaphene
Malathion
Methyl Parathion^
Phorate"
Propachlor
Trifluralinc'e
adsorption characteristics from Stewart et^ aL (1975)
persistent data from authors as indicated
b Sanborn
-------
TABLE 7-13. FREQUENCY ANALYSIS FOR RAINFALL IN GEORGIA, IOWA, AND NEW YORK
Probability of
a dry day
Georgia Iowa
N.Y.
Probability of daily rainfall occurring in excess of
0 cm/day
Georgia
Iowa
1 cm/day
N.Y. Georgia
Iowa
N.Y.
5 cm/day
Georgia
Iowa
N.Y.
10 cm/day
Georgia
Iowa
N.Y.
January
February
March
April
May
June
July
August
September
October
November
December
63.6
64.7
64.9
71.3
70.8
69.2
63.7
72.2
76.9
80.7
73.4
66.8
81.0
70.3
71.7
67.2
64.0
65.4
70.1
73.6
70.5
77.1
80.1
81.3
54.0
53.1
54.7
53.6
52.9
61.0
64.7
63.7
63.8
59.6
54.4
51.0
36.4
35.3
35.1
28.7
29.2
30.8
36.3
37.7
33.1
19.3
26.6
33.2
19.0
21.7
28.3
32.8
36.0
34.6
29.9
27.4
29.5
22.9
19.9
17.7
46.0
46.1
45.3
46.4
47.1
39.0
35.3
36.3
36.2
40.4
46.6
49.0
12.7
14.5
14.4
11.1
12.3
10.3
11.5
9.7
7.8
6.8
10.2
11.2
1.5
1.4
4.5
9.6
11.8
12.7
8.4
7.8
8.5
5.8
3.4
1.8
3.6
4.0
5.6
7.4
9.8
8.1
8.3
8.4
7.0
7.5
6.9
6.4
0.4
0.5
1.7
0.8
1.2
1.2
0.6
0.9
0.6
0.6
0.4
0.8
0.0
0.1
0.1
0.3
0.6
1.3
0.5
0.7
0.3
0.2
0.0
0.0
0.0
0.2
0.0
0.0
0.0
0.4
0.2
0.4
0.2
0.3
0.0
0.2
0.0
0.0
0.0
0.0
0.2
0.2
0.2
0.0
0.1
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.1
0.0
0.1
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.0
0.0
0.0
0.0
0.0
-------
TABLE 7-14. MEAN ANNUAL RUNOFF AND SEDIMENT LOSSES
FOR TWO MANAGEMENT PRACTICES
cm
Runoff
% rainfall
change
Sediment
% change
mean
MT/ha
Georgia
straight row 17.1
terraces 15.8
13
12
- 7
18.0
3.0
-83
Iowa
straight row 6.2
terraces 4.8
8
6
-22
6.2
0.6
-90
New York
straight row 2.3
terraces 2.3
3
3
0.8
0.2
-75
Local climatic conditions had a much greater effect on magnitude and
time of runoff events than did management practices (Tables 7-15 and 7-16).
On the shallow fragipan soil in New York, soil management practices did not
reduce runoff at all, while reductions in Iowa and Georgia were small (7-22%
reductions when terraces were installed). The differences in runoff
due to location was much more significant -- there was 87% less runoff
in New York than in Georgia.
Soil Loss Analysis--
Average annual soil loss followed the same pattern as the average annual
runoff losses (Table 7-14). However, SWCPs do have the potential to correct
for unfavorable local climatic conditions and control soil loss. With
terraces, not only was average soil loss reduced by at least 75% for all
locations, but soil loss did not exceed 5 tons per hectare for any storm
(Table 7-16). The frequencies of occurrences of soil loss of all magnitudes
ware less with terraces.
Runoff and soil loss for other management practices are presented in
Sections • 5 and 6. All the SWCPs controlled soil loss better than they con-
trolled runoff. This is an important conclusion for explaining the inter-
action of pesticide characteristics and SWCPs in the following section.
Pesticide Losses--
Mean annual pesticide losses are given in Table 7-17 for the four groups
of pesticides applied on May 10 of each year, and in Table 7-18 for the pest-
icides in Group B applied on May 10 and June 10. For the weakly and intermediately
129
-------
TABLE 7-15. MONTHLY RUNOFF FREQUENCY ANALYSIS
Ul
o
Number of
runoff events
0 -
Georgia
straight
row terraces
January
February
March
April
May
June
July ;
August
September
October
November
December
82
98
74
42
47
24
20
17
17
9
53
73
82
98
74
42
46
13
7
7
5
5
53
73
«
in a 25 year period
1 cm/day
Iowa
straight
row
19
31
52
56
29
37
11
15
15
5
14
15
terraces
19
31
52
56
10
14
5
6 .
2
2
14
15
which are
New York*
straight row
and terraces
2
6
7
2
-
2
1
3
1
2
0
5
1-5 cm/day
Georgia
straight
row terraces
22
14
22
15
9
2
i >
-
3
3
9
17
22
14
22
15
8
-
1
-
1
1
9
17
1
In New York, runoff is the same for both terraces and straight row.
-------
TABLE 7-15. MONTHLY RUNOFF FREQUENCY ANALYSIS (con't)
1
Iowa
straight
row
2
'"V
Z,
5
13
6
3
1
1
1
-
1
2
- 5
Number of runoff
cm/day
New York A
straight row
terraces and terraces
2
2
5
13
3
1
-
2
1
-
1
2
1
5
2
-
-
2
-
-
1
1
-
4
events in a 25 year period which are
Greater than 5 cm/day
Georgia Iowa New York-1
straight straight straight row
row terraces row terraces and terraces
11 - -
11
33 - -
_ _
11 11
22 - -
1
1 _
- - - -
1
_ _
_ _ _
In New York, runoff is the same for both terraces and straight row.
-------
TABLE 7-16. MONTHLY SOIL LOSS FREQUENCY ANALYSIS
CJ
to
Number of days that soil
0.1-1 MT/ha day
January
February
March
April
May
June
July
August
September
October
November
December
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
tsrraces
straight row
terraces
straight row
terraces
Georgia
57
33
46
25
45
44
32
26
22
39
16
6
12
4
12
4
13
5
6
4
32
23
42
31
Iowa New York
__
—
-.
--
34
13
22
26
9 3
8 1
4 1
14 1
4 1
10
1 2
4
2
7
4
--
loss is
1-5 MT/ha day 5-10 MT/ha day Greater than 10 MT/ha day
Georgia
4
5
18
--
9
1
27
7
10
2
8
4
4
4
6
—
Iowa New York Georgia Iowa New York Georgia Iowa New York
--
--
-.
__
4 i
13 62-- 11--
2
11 1 -- -- -- 2
--
3 i i
3
12-- —
2
_.
--
-------
TABLE 7-17. AVERAGE ANNUAL PESTICIDE LOSSES APPLIED EACH YEAR ON MAY 10
H
OJ
00
straight row
Georgia Iowa New York
Group A (low k*, t} **=15 days)
in solution (g/ha)
as percent of applied
in sediment (g/ha)
as percent of applied
Group B° (intermediate k,
tj, = 15 days)
in solution (g/ha)
as percent of applied
in sediment (g/ha)
as percent of applied
Group Cf (intermediate k,
tj__ - 100 days)
in solution (g/ha)
as percent of applied
in solution (g/ha)
as percent of applied
Group tf°(high k, t^ = 360 days)
total (g/ha)
as percent of applied
11.6
0.4
4.1
0.1
26.3
0.9
9.0
0.3
45.4
1.5
10.8
0.4
536.2
17.9
19.9
0.7
2.6
0.1
40.0
1.3
4.9
0.2
54.5
1.8
7.8
0.3
313.6
10.4
j***
T
T
T
T
T
T
T
0.7
T
0.5
T
48.8
1.6
straight row
Georgia Iowa New York
11.0 6.3
0.4 0.2
0.7 0.2
T T
26.0 16.0
0.9 0.5
0.4 0.4
T T
42.4 19.7
1.4 0.7
0.6 0.6
T T
139.6 64.4
4.7 2.2
T
T
T
T
T
T
T
T
0.7
T
0.1
T
26.7
0.9
k = adsorption partition coefficient
**tl/2 = half-life
= Trace, less than 0.05
***
adsorption isotherm equal to atrazine
in appendix B
00 adsorption isotherm equal to paraquat
in appendix B
-------
TABLE 7-18. AVERAGE ANNUAL PESTICIDE LOSSES (IN GROUP B) FOR TWO DAYS OF APPLICATION
UJ
straight row
Georgia Iowa New York
Applied May 10
in solution (g/ha)
as % of applied
in sediment (g/ha)
as % of applied
Applied June 10
in solution (g/ha)
as % of applied
in sediment (g/ha)
as % of applied
26.3
0.9
9.0
0.3
3.8
0.1
2.0
0.1
40.0
1.3
4.9
0.2
8.6
0.3
1.6
0.1
T*
T
T
T
1.0
T
0.4
T
Georgia
26.0
0.9
,0.4
T
0.5
T
T
T
terraces
Iowa New York
16.0
0.5
0.4
T
1.6.
0.1
T
T
T
T
T
T
1.0
T
T
T
* T = trace, less than 0.05
-------
adsorbed pesticides, most loss occurred in the solution phase. Pesticide
losses for both Iowa and Georgia were significantly higher than for the
shallow fragipan soil in New York. Pesticide loss from New York soils with
the exception of the most strongly adsorbed persistent pesticides, does not appear
to be an environmental problem. Pesticide losses for the June 10 application
are much lower than for the May 10 application for both Georgia and Iowa;
conversely, a slight increase was noted for New York.
For the three locations, Tables 7-19, 7-20, and 7-21 show the number of
days with a pesticide loss in a 25-year period. For example in Georgia for
corn planted straight row, there were for Group A, 40 days in May over a
25-year period that pesticides could be detected in the overland flow. Of
these 40 days, 27 days had an amount of pesticide lost below 0.3% of the total
amount applied and 13 days an amount between 0.3 and 4% of that applied.
For Iowa and Georgia, the May 10 application is at the beginning of the
period of heavy thunderstorms and consequent high runoff events (see previous
section). The June 10 application falls towards the end of this period.
Tables 7-19 and 7-20 clearly show this. The number of runoff events with
measurable pesticide losses is reduced significantly for the June 10 appli-
cation (of pesticides in Group B) when compared with the May 10 application
of the same pesticide.
For New York, the large runoff events occur in summer when heavy thunder-
storms occur (Table 7-15). Again, the closer the date of pesticide application
to the period of intense rainfall, the greater the chance of pesticide loss.
This is illustrated in Table 7-21.
Adsorption partition coefficient and half-life greatly affected pesticide
loss. The higher the half-life and adsorption partition coefficient, the
greater was the pesticide loss. For straight row corn, as much as 18% of the
applied Group D pesticide was lost in Georgia. The longer the half-life and
the stronger the adsorption of the pesticide, the greater were the losses which
occured in the month of application.
In Georgia, terraces were not very effective in reducing the quantity of
pesticides transported in the runoff water if the date of application was
May 10. In Iowa, as opposed to Georgia, terraces are much more effective in
reducing the dissolved component of the pesticide loss. The main reason for
this difference is that the soil has a lower moisture content in May in Iowa
than in Georgia. Low evaporation and relatively large rainfall quantities in
the winter in Georgia cause this high moisture content. The water infiltrated
before runoff occurs (or initial abstraction) computed with the SCS curve
number is much greater for dry conditions than for wet conditions. The more
water that infiltrates before runoff occurs, the more pesticide is leached
out of the surface zone and the less is lost in surface water. For all
locations, terraces greatly reduce the portion of pesticide transported with
the sediment. The decrease in loss of strongly adsorbed pesticides is almost
as great as the decrease of sediment (Table 7-16).
Pesticides of Groups A, B, and C are mostly transported in the water
phase. Therefore, the effect of SWCPs on runoff is the determining factor
135
-------
TABLE 7-19. FREQUENCY OF NUMBER OF PESTICIDE LOSSES IN A 25-YEAR PERIOD
.UNDER TWO MANAGEMENT PRACTICES IN GEORGIA
w
-------
TABLE 7-20. MONTHLY FREQUENCY OF PESTICIDE LOSSES IN A 25-YEAR PERIOD UNDER
TWO MANAGEMENT PRACTICES IN IOWA
H
U>
Group A
appl. May 10
0 0.3 4
to to and
0.3 4 more
percent of applied
per day
January straight row
terraces
February straight row
terraces
March straight row
terraces
April straight row
terraces
May straight row 18* 4 2
terraces 10 3
June straight row 32
terraces 7
July straight row 4
terraces
August straight
terraces
September straight row
terraces
October straight row
terraces
November straight row
terraces
December straight row
terraces
Group B Group B Group C
appl. May 10 appl. June 10 appl. May 10
0 0.3 4 0 0.3 4 0 0.3 4
to to and to to and to to and
0.3 4 more 0.3 4 more 0.3 4 more
percent of applied percent of applied percent of applied
per day per day per day
7
8
16 S 3 11 10 4
931 941
37 21 5 1 35 3
11 91 IS 0
7 10 12
2 4
8 17
6
11
4
4
0
7
S
3
3
Group D
appl.
0
to
0.3
percent
Mav 10
0.3 4
to and
4 more
of applied
per day
6
14
9
9
9
12
27
25
7
7
9
11
5
4
8
6
12
2
3
2
7
10
11
14
15
14 10
9 1
28 4
4
6 1
1
8 1
2
3 0
1
2
S
3
2
1
•number of daily pesticide losses in this category seen in 25 years
-------
TABLE 7-21. MONTHLY FREQUENCY OF PESTICIDE LOSS EVENTS FOR 25 YEARS UNDER TWO
MANAGEMENT PRACTICES IN MEW YORK
Group A Group B Group B
appl. May 10 appl. May 10 appl. June 10
percent of applied percent of applied percent of applied
0.0-0.3 0.31-4.0 above 4.0 0.0-0.3 0.31-4.0 above 4.0 0.0-0.3 0.31-4.0 above 4.0
January
February
March
April
M
w May
oo
June
July
August
September
October
November
December
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row 1 4 31
terraces 1 4 31
straight row 1
terraces 1
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
-------
TABLE 7-21. MONTHLY FREQUENCY OF PESTICIDE LOSS EVENTS FOR 25 YEARS
UNDER TWO MANAGEMENT PRACTICES IN NEW YORK fcon't")
VD
Group C Group D
appl. May 10 appl. May 10
percent of applied percent of applied
0.0-0.3 0.31-4.0 above 4.0 0.0-0.3 0.31-4.0 above 4.0
January
February
March
April
May
June
July
August
September
October
November
December
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
straight row
terraces
2
2
1 5 1
1 5
10
10
1
41 113
41 1 3 1
1 1
1 1
3 2 1
2 3
2 2
2 1 1
3 2 1 1
3 2 1
4 5
4 5
-------
for controlling pesticide loss. For pesticides in Group D, the effect of
SWCPs on sediment movement determines pesticide pollution control.
Table 7-22, derived from data in Table 7-17, indicates the percent
change in pesticide loss in overland flow in Georgia due to a different
pesticide management practice or date of application as compared to the loss
of a Group D pesticide applied on May 10 to corn planted either straight row
or on terraces. Thus, if a farmer used a Group A pesticide on a terraced
corn field, he would obtain a 98% reduction in pesticide loss over a 25 year
period as compared to using a Group D pesticide with straight row planting.
Table 7-22 also indicates that if a pest could be controlled by a Group B or a
Group D pesticide, by applying the former on June 10 as opposed to the latter
on May 10, a reduction in loss of pesticide of 99% over a 25 year period for
corn planted in straight row would occur.
Other conclusions which can be drawn from Table 7-22 are: 1) substituting
a pesticide out of Group A, B, or C for a pesticide out of Group D is more
effective in reducing pesticide loss than is implementation of any SWCP; 2) to
TABLE 7-22. PERCENT CHANGE IN PESTICIDE LOSS FOR DIFFERENT PESTICIDE
GROUPS, TILLAGE PRACTICES, AND APPLICATION DATES AS COMPARED TO
GROUP D PESTICIDES APPLIED ON MAY 10!
Group D (appl. May 10)
straight row terraces
Group A (appl. May 10)
straight row -97 -89
terraces -98 -92
Group B (appl. May 10)
straight row -93 -75
terraces -95 -81
Group B (appl. June 10)
straight row -99 -96
terraces -100 -100
Group C (appl. May 10)
straight row -90 -60
terraces -92 -69
Group D (appl. May 10)
straight row +74
terraces -74
140
-------
prevent losses, pesticides should not be applied during a period of high
runoff and sediment losses; 3) any combination of the following practices is
more effective than is implementing one individually: application of pesticide
with a) a shorter half-life, b) a lower adsorption coefficient, c) at a date
when probability of runoff is small, or d) implementing management practices
which reduce soil and runoff losses. These^conclusions are in agreement with
findings of a simulation study by Dean and Mulkey (1978).
Similar conclusions can be drawn for Iowa and New York. However, since
pesticide losses are not great to begin with in New York (Table 7-15) imple-
menting a SWCP to control pesticide pollution may be of limited value. The
conclusions from the simulation studies (Table 7-22) should be considered
carefully since, they were not verified by field data other than for the base
condition.
Pesticide Leaching--
This section describes the sub-model used to simulate the downward move-
ment of pesticide bands over 25 years for two SWCPs at three locations.
The CPM simulates pesticide loss in percolation water as well as in over-
land flow. As the pesticide band moves downward through the soil, the chemical
degrades. Therefore, the model estimates both the depth of the midpoint of
the band and the pesticide concentration at this depth.
At some depth in the profile, the pesticide will have degraded to the
extent that the original compound can no longer be identified. For the
purpose of this section, the lower limit of detection is taken as 0.01% of
the amount applied. The average and maximum depth of pesticide bands are
shown in Table 7-23.
Several factors affect pesticide leaching: adsorption coefficient --
the stro'nger the binding, the slower the downward movement, half-life -- the
longer the half-life the longer the pesticide remains in the soil, and quantity
of water percolating through the soil -- the greater the percolation, the
faster the band moves. This effect was seen in Table 7-23. Pesticides of
Group A could be detected at a lower depth than Group B because Group A
adsorption strength was less. Both Group A and Group B pesticides had the
same degradation half-life. Group B and Group C pesticides only differed in
half-life (Group C's half-life was approximately seven times as great for
Group B). Pesticides of Group C could, therefore, be found deeper in the
root zone than Group B.
SWCPs which decrease runoff promote infiltration of water, causing
deeper displacement of the pesticide bands. This appears to present a pro-
blem, because once a pesticide moves below the root zone, there is little
bacteriological activity, and pesticide degradation is greatly hampered.
Terraces were found to increase the depth of the pesticide band only slightly.
Pesticides did not reach the bottom of the root zone in any simulation
(Table 7-23).
In conclusion, there appears to be little danger of pesticide pollution
of groundwater in soils where the water table is lower than the root zone.
141
-------
K>
TABLE 7-23. AVERAGE AND MAXIMUM DEPTH AT WHICH THE AMOUNT OF PESTICIDE IS LESS THAN
0.01% OF THAT APPLIED
Georgia*
straight row terraces
Group A
average
maximum
Group B
average
maximum
Group C
average
maximum
Group D
average
maximum
depth
depth
depth
depth
depth
depth
depth
depth
(cm)
(cm)
(cm)
(cmi)
(cm)
(cm)
(cm)
(cm)
18
28
6
8
26
29
5
5
19
29
6
8
27
31
5
5
straight row
14
21
4
6
16
23
4
4
Iowa+
terraces
15
23
4
6
16
23
4
4
New York*
straight row terraces
16
27
6
8
21
28
4
5
16
27
6
8
21
28
4
5
+ maximum plant rooting depth for Georgia, Iowa and New York were 75, 135 and 40 cm, respectively
-------
COMPARISON OF PREDICTIONS OF SWCPs
Three models have been used throughout this report. The Agricultural
Runoff Management (ARM) Model was used to predict runoff, soil loss, and
nutrient and pesticide losses over a ten-year period in Georgia and Michigan.
The Cornell Nutrient Simulation Model (CNS) predicted runoff, soil loss, and
nutrient losses in overland and percolating waters over a 25-year period in
Georgia, Iowa and New York. Finally, the Cornell Pesticide Model CCPM)
simulated runoff, soil loss, and pesticide losses in overland and drain water
over a 25 year period at the same three locations.
Two management practices were input to all three models: straight row
and terraces. Other things in common for all models were runoff and soil
loss predictions. Moreover, pesticides in Groups B and D applied on May 10
in the CPM closely resemble the atrazine and paraquat used in the ARM model.
The differences in model predictions can be seen in a summary of results
(Table 7-24).
The CPM and CNS models predicted different values for runoff in Georgia.
The difference was not expected because both models employed the same run-
off formula. One important difference was in the value chosen for antece-
dent moisture condition (one of the equation variables). This parameter
depends on soil moisture content, which the computer continuously budgets.
The moisture budgets of the two models differ in zone configuration,
evapotranspiratiori prediction methods, and value of soil moisture content
at the beginning of each year.
For Iowa, predicted amounts of runoff are very similar. As opposed to
Georgia, antecedent moisture conditions predictions were similar. Percent
changes for terraces compared to straight row were all in the same direction.
As expected, in New York, the CPM and CNS predictions differed substantially.
The CNS model used the SCS curve number for the runoff predictions, while the
CPM employed the sub-program which predicted runoff on shallow fragipan soils.
The CPM model is valid for soils with good structure, many stones, and hardpans
within 50 cm of the surface. The CNS model simulates a field with a deeper
,hardpan and having a more poorly drained soil. Both of these soil types can
be found in New York.
The three models used fairly different methods to predict erosion. The
ARM model used the Negev sediment algorithm (Negev, 1967) and assumed chisel
tillage. The CNS model used the USLE with the energy term computed by the
Onstad-Foster method (Onstad and Foster, 1975). In New York and Georgia,
spring plow is assumed, while for Iowa, fall plow is simulated. The CPM
model also employs the USLE, but the energy term is computed by the Williams
method (Williams and Berndt, 1977). Spring plowing is assumed at all three
locations.
Considering all these differences, closeness of the CNS and CPM models'
predictions of percent change in soil loss using terraces is somewhat sur-
prising. The ARM model predicted that the SWCP had a smaller effect than was
predicted by either the CNS or CPM model.
143
-------
TABLE 7-24. COMPARISON OF AVERAGE ANNUAL LOSSES FOR ARM, CNS, AND CPM
MODELS FOR SELECTED PRACTICES
Runoff
straight row (cm)
terraces (cm)
% change
Soil Loss
straight row (cm)
terraces (tons/ha)
% change
Pesticides
Group B
straight row
% of applied
terraces
% of applied
% change
Group D
straight row
% of applied
terraces
% of applied
% of change
, Georgia
CNS CPM^
22.2 17.1
20.8 15.8
-6 -7
24.4 18
9.6 3
-6 -83
1.8
1.0
-44
13.6
6.5
-52
, Iowa
ARM CNS CPM
19 7 6.2
14.4 5 4.8
-25 -29 -22
3.5 17 6.2
2.4 1 0.6
-32 -94 -90
1.2
0.9
-25
17.9
10.4
-74
New York
CNS CPM
10 2.3
4 2.3
-60 0
20 0.8
1 0.2
-95 -74
Cornell nutrient simulation
Cornell pesticide model
5
Agricultural runoff management model
-------
The average annual losses for atrazine and paraquat in overland flow as
predicted by the ARM and CPM are again remarkably similar. The percent change
in pesticide loss achieved by using the SWCP was not quite the same for these
twoTnodels, but they were in the same direction.
SUMMARY
The action of SWCPs on unit source areas has been simulated to determine
the effectiveness of SWCPs in reducing pesticide loss in runoff and percolating
waters. The effects of different pesticide characteristics also were simulated.
Two models were used for the simulation studies, a) the Agricultural
Runoff Management (ARM) Model, and b) the Cornell Pesticide Model (CPM) which
was developed for this project. Both models had approximately the same pre-
cision in predicting pesticide loss.
Simulation runs were made to evaluate the interaction of SWCPs, climate
and pesticide characteristics at several locations in the United States for
a 10 or 25 year period. The selection of management practices was limited by
the capability of the simulation models. Straight row, contouring combined
with terraces, sodbased rotations, and minimum tillage were modeled by either
one or both models.
The pesticides modeled were grouped as follows:
a) weakly adsorbed and easily degradable
b) intermediately adsorbed and easily degradable
c) intermediately adsorbed and intermediately persistent
d) strongly adsorbed and very persistent
CONCLUSIONS
1. Effects of SWCPs on pesticide loss are not only site specific, but
also depend on pesticide characteristics.
2. Greatest pesticide runoff losses occurred in areas with heavy
thunderstorms. Therefore, losses in Georgia are much higher
than for New York, where they were almost negligible.
3. In all case studies, yearly average pesticide losses were lower
for those pesticides with low adsorption coefficients and short
half-lives than for these pesticides which were more strongly
adsorbed and more persistent.
4. SWCPs are more effective in controlling strongly adsorbed pesticides
than in controlling the less well adsorbed pesticides.
5. Under normal agricultural practices, for the pesticides modeled, it
is not likely that pesticides will be found in water percolating
out of the root zone.
145
-------
ACKNOWLEDGEMENTS
The author expresses his gratitude to the many persons who helped with
the preparation of this section: Mrs. Hanneke DeLancey for her help with
the validation of the model, Mr. Lee Jacobowitz for the editing of the
original manuscript; Mrs. Marian Harris for collecting data; Mr. H.D. Murray-
Rust, Mr. Larry Tubbs, and Mr. Orson Yancey for their help with the computer
programming; Mr. Matt Lorber for collecting climatological data; and Ms. Sue
Roedel for the typing.
146
-------
SECTION 8
COST-EFFECTIVENESS OF SOIL AND
WATER CONSERVATION PRACTICES
FOR IMPROVEMENT OF WATER QUALITY
Erick E. Smith, Earl A. Lang,
George L. Casler and Roger W. Hexem
The principal function of soil and water conservation practices has
traditionally been to reduce soil erosion on cropland. Recent concern over
non-point source pollution from cropland has prompted an interest in examin-
ing the effectiveness of SWCPs for improving water quality. In areas where
sediment contributes to the degradation of water quality, it has been assumed
that practices which reduce soil erosion in a field have the effect of
causing less sediment to be available for transport to water. The principal
focus in this section is on the cost-effectiveness of SWCPs for control of
soil erosion and the transport of sediment to water. Although other substances,
primarily pesticides and nutrients, are also potential pollutants, the direct
causal relationship between SWCPs and the availability of these substances
to water bodies is less clear. Thus investigation of these substances is
limited to edge-of-field losses.
Several studies have dealt with the relationship between the cost of
implementing SWCPs and their potential benefits to water quality. Studies
using a watershed as the basis for analysis have been done by Seitz et al.
(1978) in Illinois, Alt et al. (1978) in Iowa, and Taylor et al. (1978) in
Texas. Each of these studies compared costs of SWCPs (implementation and
maintenance) with assumed benefits. These included both the yield benefits
of retaining topsoil and water quality benefits. A number of studies have
analyzed costs associated with various farm conservation plans. These
studies were conducted in Iowa (McGrann _e£ al_. 1978), Pennsylvania (White
_e£ al_. 1978) and Wisconsin (Sharp et aJL 1978). In general, the approach
in these studies has been to focus on practices tor controlling soil erosion
and the reduction in soil erosion was then assumed to result in improved
water quality. Only the Wisconsin study evaluated sediment separately from
soil erosion and this was done after the fact. That is, farm plans for
control of soil erosion were developed, then their effect on sediment reaching
waterways was determined. Direct evaluation of the relationship between SWCP
implementation and water quality is difficult for a number of reasons
including: 1) No generally accepted values exist for the amount of sediment
from a field, farm or watershed which should be allowed in a body of water.
2) Existing physical models for predicting sediment transport (Section 5) are
not widely accepted and generally involve fairly detailed simulations which
147
-------
make the evaluation of a wide variety of options quite difficult. Thus, at
this point, little distinction exists in the literature between methods of
implementing SWCPs for control of soil erosion and methods of implementation
for the purpose of reducing sediment delivered to streams.
A recent study by Meta Systems (1978) evaluated individual SWCPs at the
farm level on the Black Creek Watershed in Indiana. Each practice considered
was assumed to be implemented on the whole farm. Estimates of farm cost,
soil erosion levels, and sediment delivery to the stream as well as loss
estimates for nutrients and pesticides were made. Although useful for com-
paring individual practices, this study did not compare actual farm plans,
where combinations of practices which vary from field to field would typically
be used.
A primary objective of this section is to separate and compare the issue
of soil erosion from the issue of sediment and its transport to waterways.
Soil erosion is undesirable principally because of long-term adverse effects
on soil productivity while sediment delivery to waterways is undesirable be-
cause of the effects on water quality and sedimentation rates in reservoirs
and lakes.
The study evaluates costs and effects of a number of SWCPs and combin-
ations of practices when implemented on individual farms. Costs are inter-
preted as the loss in net farm income caused by imposing a practice or system
of practices on a farm to meet some particular goal such as a reduction in
soil loss to some predetermined level. Effectiveness of SWCPs is judged pri-
marily by estimated reductions in sediment reaching a waterway. However,
discussions on their effects on soil erosion and on other pollutants are also
included. Reducing sediment load depends on both the practice or practices
being used, the soil complex, and the particular location of practices
relative to the stream. Areas which have a high potential for sediment loss
to streams may not have the greatest soil erosion problems. Thus the evalua-
tion of cost-effectiveness of SWCPs for reducing sediment, load to streams
must consider transport as well as gross soil erosion.
By judging the effectiveness of practices primarily in terms of the
reduction of sediment delivered to waterways, the benefits of reduced levels
of soil erosion are ignored. It would be expected that when SWCPs are actually
implemented on a farm, both types of effectiveness would be considered.
Although this section concentrates on the difference between these types of
effectiveness, it is not necessarily implied that the goals are mutually
exclusive.
Since the emphasis is on examining single practices and systems of
practices, in relation to individual situations, the farm was chosen as the
unit of analysis. This allows evaluations of the costs of practices as they
affect farm operations and thus farm income. Also effectiveness of a practice
can be related to the physical properties of the field in which it is
implemented.
148
-------
DEFINITIONS
Most of the terms used in this section have been defined in Section 4.
Some additional terms specific to this section are defined below.
Base System. This is the particular cropping system for a farm which
is used as the standard for determining both cost and effectiveness
of SWCPs on that farm. In general it would designate the most profit-
able cropping system for the farm without regard to soil erosion or
water quality.
Waterway. A waterway is any place where channelized flow would normal-
ly occur, including streams, road ditches, sod waterways, diversion
ditches, tile outlets, etc. However, for terraces, the waterway
begins at the terrace outlet rather than the terrace channel.
Sediment delivery. The average annual amount of sediment reaching
a waterway. Units are metric tons per hectare.
Cost of implementing SWCPs. The cost of implementing an SWCP or a
system of practices on a farm is the change in expected annual net
farm income after implementation as compared to annual income with the
base system. Costs generally are expressed in dollars per hectare.
Effectiveness of SWCPs. The effectiveness of an SWCP or combination
of practices for control of a pollutant is the expected change in the
level of the pollutant reaching a waterway after implementation as
compared to the base system. A decrease in the level is considered
a positive effect. Units are either kilograms of metric tons per
hectare.
Cost-effectiveness of SWCPs. The cost-effectiveness of a practice
or system of practices is the effectiveness divided by the cost.
Unless otherwise specified, cost effectiveness will always be used
in relation to changes in sediment delivery. Units are metric tons
per dollar.
Marginal cost of sediment control. At any given level of sediment
delivery on a farm, the marginal cost is the additional cost required
to reduce sediment delivery by one ton.
OBJECTIVES
1). Summarize factors which affect the cost of SWCPs.
2). Summarize factors which affect the effectiveness of SWCPs for
reducing sediment delivery to waterways.
3). Develop some general principles on the cost-effectiveness of SWCPs
for reducing sediment delivery.
149
-------
4). For a few example farms, construct farm plans oriented towards
reducing sediment delivery by certain given amounts. Compare
these plans with plans which are constructed with an orientation
towards reducing soil erosion.
5). Develop representative cost curves which would be facing a farmer
attempting to reduce sediment delivery.
6). Analyze how the farm plans for controlling sediment delivery might
affect levels of losses of other pollutants.
7). Discuss how current cost-sharing programs might influence the
choice of practices used in farm plans for reducing sediment
delivery.
METHODOLOGY
Costs
An initial evaluation of the changes in farm costs associated with
implementation of SWCPs is done through a general review of the literature.
A summarization of this review associates practices with directional changes
in individual items in a crop budget.
The magnitudes of these changes are illustrated for an example situation,
using a basic crop budget for corn grain from Iowa State University (James
McGrann, Personal Communication, 1978). A number of selected SWCPs are
budgeted, to illustrate their effects on the items in this basic crop budget.
Details on this budgeting process are given in Appendix E.
Effectiveness for Control of Sediment Delivery
To determine the level of effectiveness for selected SWCPs in any given
field, both the level of soil erosion and the sediment delivery ratio (SDR)
for each practice must be known.
Soil Erosion
Soil erosion was calculated using the Universal Soil Loss Equation
(USLE) (Wischmeier and Smith, 1965). Soil erosion is calculated by the
equation A=RKLSCP where:
A = Average annual gross soil erosion
R = Rainfall erosivity factor for the area
K = Soil credibility factor for the particular soil
L = Slope length factor
S = Slope gradient factor
150
-------
C = Cropping practice factor
P = Conservation practice factor
All values for factors used in the equation were from standard sources (either
Wischraeier and Smith, 1965 or state SCS offices) except the "C" values for
New York and Iowa which were individually calculated, based on recently re-
vised tables (Wischmeier and Smith, 1978).
Soil erosion for terraces included all soil moved down the slope to the
terrace channel. Thus terraces affect soil erosion directly only by a change
in the L factor, and indirectly as a result of the implied contouring.
This is relatively important point since for tile-outlet terraces, only about
10-20% of soil reaching the channel actually leaves the field through the
tile (Stewart et al. 1976). Although the sediment deposited in the terrace
channel does not actually leave the field, it is eroded from its original
location and thus has potential long-term detrimental effects on top soil
maintenance. Spreading this material back over the field is difficult and
almost never done (Earnest Hintz, Personal Communication, 1978).
For crops such as corn silage which leave little or no crop residue on
the field, method of tillage has little effect on soil erosion (Walter
Wischmeier, Personal Communication, 1978). Thus the "C" factors for corn
silage under conventional tillage, conservation tillage, and no-till are
nearly identical.
Estimates of soil erosion by area, soil, and practice and more details
on methodology are contained in Appendix E.
Sediment Delivery
In this section, sediment delivery represents the estimated tons of
sediment from cropland entering a waterway. Sediment delivery for an area
is equal to a calculated sediment delivery ratio (SDR) times the estimated
soil erosion as calculated from the Universal Soil Loss Equation. The SDR
for an area is a function of the distance from the center of the area to a
waterway. For the purpose of this study, a waterway is considered to be an
actual stream, a terrace outlet, a diversion ditch, or any other channeled
waterway. For tile-outlet terraces, the SDR is multiplied by 0.2 to account
for settlement in the terrace channel. Details on the methodology and cal-
culation of SDR are in Appendix E.
It is recognized that in many cases, not all sediment entering a water-
way will affect water quality. Some redeposition may occur in the waterway,
particularly in grassed waterways and diversion ditches. Also, in some
cases, tile outlets and waterways may discharge into woodlands or similar
areas which may filter out much sediment. Since these effects are site-
specific, they should be taken into account when designing water quality
plans for actual farms. The results of this study only apply directly
to situations where all sediment entering waterways has a potential effect
on water quality.
151
-------
Of course, other problems are associated with this method for calcu-
lation of sediment delivery. Conceptually, the greatest difficulty is
that the model implies that the SDR for any area is independent of con-
ditions between that area and the stream. Thus the effect of, say, hav-
ing a field of sod crop rather than row crop below the field being eval-
uated is nullified. In particular, grass buffer strips placed along
waterways are not evaluated in this analysis. Although claims have been
made that buffer stips are effective sediment traps, little verification
exists, and their effectiveness is questionable when the width of the
strip is considerably less than the total slope length (Section 5).
Despite these shortcomings, the method does seem useful. It should pro-
vide a reasonably good way for comparing the relative importance of individual
fields or areas of a field relative to their location on the watershed. It
also provides a means for comparing relative effectiveness of individual and
combinations of SWCPs. Finally, and perhaps more importantly, it may provide
a methodology which, when combined with reasonable judgement, could be used
in the field for determination of critical areas for control of sediment
delivery.
Although the estimates are made in terms of actual tonnage reaching a
waterway, it should be emphasized that these may not be good representations
of actual amounts reaching the stream. Rather, it is hoped that the relative
effects of distance and practice are accurately portrayed. Thus results of
the analysis would be better viewed in terms of percent decrease in sediment
delivery from the base conditions, rather than as actual amounts of sediment
reaching a waterway.
Initial Evaluation of Factors GoverningCost-Effectiveness
To develop some insight into the factors governing the cost-effectiveness
of SWCPs, certain individual practices are investigated on a field basis.
Soil erosion levels are initially calculated for a hypothetical field with a
4% slope gradient and a 120 m slope. The "R" (rainfall erosivity) factor
for the USLE is assumed to be 200, while the "K" (soil credibility) factor
is assumed to be 0.3. These values would fall in a moderately erosive range
common for much of the central and eastern United States.
Soil erosion levels are calculated for continuous corn grain with no
SWCPs and for each of the practices for which costs budgets were determined
in the section on costs.
Initially, the SDR for the field is assumed to be 1.0. This would occur
if the field were directly adjacent to a waterway such that all eroded soil
entered the waterway. Under these conditions, sediment delivery and effect-
iveness for individual practices are calculated and compared.
In order to determine the effects of varying SDRs and levels of soil
erosion, the effectiveness of the practices is recalculated after first lower-
ing the SDR to 0.3, then by raising the levels of soil erosion by a factor of
four. Although these changes are somewhat arbitrary-, they serve the purpose
of illustrating the relationships between the effectiveness of SWCPs and
different types of field situations.
152
-------
Example Farms and Farm Plans for Control of Sediment
Delivery and Soil Erosion ~~~
Three areas representing a variety of types of agriculture in the eastern
half of the United States were chosen for study. A major soil association
is identified for each area. The areas with their representative soil
associations are: 1) Eastern Iowa, Tama-Muscatine Soil Association; 2) Central
New York, Honeoye-Lima Association; and 3) Eastern Texas, Houston Black-Heiden
Association. The example farms are not intended to be "typical" of the area in
physical layout of soils, but are idealized versions of each soil association.
In each case, the farm is constructed as a rectangle with a stream flowing in
the center. The sides of the farm parallel to the stream are considered to
be the boundaries of the drainage area such that no water flows onto the
farm from outside these areas (See Figure 8^1). Individual soil series (or
combinations) are then laid out in strips parallel to the stream with areas
approximately proportional to their fraction of the soil association. Each
soil series is assumed to have a uniform slope which is typical of slopes
for that series. All land on the farms is assumed to be tillable.
Through contact with state SCS offices and local Extension Service
personnel, a list of SWCPs common to each area was determined. These included
various tillage practices, rotations, structural practices, and contouring.
Because of a lack of data, some practices which may be applicable for the areas
studied were not included in the analysis. For example, contour ridge planting
appears to be a promising practice for much of Iowa and other cash-grain areas.
At present, however, adequate data for budgeting this practice are apparently
not available.
Individual crop budgets for each example farm were constructed for both
base conditions (no practices) and for each SWCP considered. Fertilizer
and pesticide applications rates for each soil and crop budget were based
on recommendations made by the Extension Service. Both pesticide and ferti-
lizer (nitrogen) rates are dependent on the crop rotation as well as the in-
dividual crop. Available budgets have variable machinery costs and total
tractor hours. In certain cases where these figures were not available for
all tillage practices, variable machinery costs are calculated from engineering
records. Labor requirements are assumed to be 10% greater than tractor hours.
Estimated yields are based on the particular soils and fertilization
rates. Effects of SWCPs on yields are variable and depend on the. area. For
example, in New York it is assumed that neither tillage practice nor crop
rotation has an effect on corn yields (Fred Swader, Personal Communication,
1978), while in Iowa a decrease in yields is assumed for no-till while corn
following hay or soybeans shows a yield increase (William Shrader, Personal
Communication, 1978). In both New York and Iowa, no yield change was assumed
for conservation tillage. For these two cases, the budgeting process
indicated that conservation tillage was more profitable than conventional
tillage. Thus conservation tillage was considered to be the basic tillage
system. Conservation and conventional tillage are compared in Table 8-14.
Although moldboard plowing is still the predominant primary tillage implement
in these areas, the use of other tillage implements appears to be spreading
153
-------
rapidly. Chisel plows and offset discs are now used on nearly 45% of Iowa
cropland (James McGrann, Personal Communication, 1978).
Although the farms as laid out for this study would not necessarily
have higher costs for contouring, it is assumed that any operations done on
the contour require 10% additional machinery and labor time. All fieldwork
on terraced fields or fields with diversions is assumed to be done on the
contour.
Terraces are included both at SCS recommended spacings and at wider
spacings in order to evaluate a range of terrace systems. Although it is
not possible to determine all associated costs of wider terrace spacings, it
is assumed they would be technically feasible if adequate maintenance of
terrace channels is carried out.
Costs for structural practices are amortized costs over a 45-year period.
Although this period is somewhat arbitrary, it is within the life expectency
for structures in both Iowa and New York. Maintenance costs for terraces and
diversions depends on spacing, with wider spacing requiring more maintenaince
per unit length of structure.
In each area a cost-share system which paralleled as closely as possible
the existing cost-share program was developed. For structural practices
where cost-sharing is generally based on a percent of initial construction
cost, the cost-share money was subtracted from construction costs before
calculating the annual cost. For practices such as strip-cropping, where
cost-sharing is often available as a one-time payment based on field area,
this payment was amortized over 20 years. Cost-sharing for sod-based crop-
ping systems is generally based on a percent of establishment costs. Thus
this could be simply subtracted from the crop budgets.
Detailed explanation of the methods used for determining costs and
returns is included in Appendix E.
Development of Farm Plans and Cost Curves
The framework of this part of the analysis is a linear programming (LP)
model developed for each example farm. The model is set up to maximize
income subject to certain restrictions. These include seasonal labor restric-
tions, land area, barn capacity, etc. The options include the growing of
a variety of crops in combination with the selected SWCPs.
After initially running the model with no restrictions on soil erosion
or sediment delivery, the base system for the farm is obtained. The model
is then run sequentially with incremental decreases in allowable sediment
delivery down to a level of approximately one metric ton per hectare. From
these results, farm plans are constructed, representing 50% and 90% reductions
in sediment delivery from the base system.
The procedure is repeated using soil erosion as a restraint. Equivalent
farm plans are constructed for 50% and 90% reductions in soil erosion. Further
analysis and comparison of the results yields information on relative effect-
iveness of practices and their locations relative to the stream for soil
erosion control versus sediment control.
154
-------
It should be noted that the restraints on sediment delivery were for the
average over the whole farm, while soil erosion restraints were on individual
fields. Since the major justification for reducing sediment delivery is to
improve water quality, the point of origin of the sediment is not important.
Thus only whole farm average loss was considered. The justification for
controlling soil erosion, however, is to maintain topsoil, thus restrictions
on a field basis were considered to be appropriate. This difference in the
form of restraints allows more flexibility in developing farm plans for
controlling sediment delivery than for controlling soil erosion.
After adding cost-sharing equivalent to current ASCS policies, the
model was run again with sediment delivery as the restraint. It is expected
that the farmer cost of controlling soil erosion or sediment delivery will
be less with cost-sharing. Further analysis should yield methods by which
cost-sharing could be made more efficient, particularly in terms of whether
the goal of the cost-sharing program is water-quality (control of sediment
delivery") or long-term soil productivity (control of soil erosion).
RESULTS
The results are presented in four sections. First, the effects of
SWCPs on farm costs and income are investigated. A brief survey of related
literature points out the types of effects others have observed. Then a
common crop, corn grain, is budgeted for several SWCPs to show the type and
magnitude of costs associated with SWCP implementation.
Second, the effectiveness of these practices for controlling sediment
loss to waterways is determined for a few hypothetical field situations.
Combining this with the cost-budgets already developed, cost-effectiveness
values are calculated and discussed.
Three example farms are then evaluated, using an LP model to determine
optimal farm plans for controlling sediment loss as compared to soil erosion.
Potential effects of cost-sharing programs on these farm plans are also
discussed.
An analysis of the relationship between levels of soil erosion, redeposition,
and resultant net losses of top-soil for the New York example farm is also
presented.
The effects of reduced levels of sediment delivery on nutrient losses is
analyzed for the Iowa farm, using results from the CNS model (Section 6).
Finally, the effects of SWCPs on losses of herbicides from the Iowa farm
are examined. These results are compared to the reduction in losses that
could be achieved by the use of alternative herbicides.and by incorporation.
Farm Costs and Benefits of SWCPs
Short-term effects include outlays for structural and maintenance costs,
changes in input levels and changes in yields. Items which generally remain
155
-------
unaffected include seed, phosphorus, potassium, lime, harvesting and selling
costs, and taxes and insurance. Table 8-1 lists the major cost items likely
to change with implementation SWCPs and the likely direction of change for
each SWCP. The following review of the literature indicates the reasoning
behind the effects noted in Table 8-1.
No-tillage
A number of studies on yield effects of no-tillage corn exist. Using
data from two and three year trials, Fink and Wesley (1974) found no signifi-
cant difference in corn yields between no-tillage and conventional tillage
in Illinois. The one exception was in 1970 when heavy rains occurred shortly
TABLE 8-1. POSSIBLE EFFECTS OF SWCPs ON YIELD AND PRODUCTION INPUTS*
No- tillage
Minimum Till
Rotations
Cover Crops
Contouring
Graded Rows
Diversions and
Grassed Waterways
Strip Cropping
Terraces
Ridge Planting
Listing
Timing of Field
Operations
Yield N
I 0
I 0
0,+
I 0,-
0,+ 0
0 0
0,+ 0
0,+ 0,-
0,+ 0
I 0
I 0
0,- 0,-
Pesti-
cides
+
0,+
-
0,+
0
0
0
0
0
0,+
o,+
0
Equip- Const. (J
ment Labor Land Maint.
0 0
0 0
0,+ I 0 0
+ + 0 0
0,+ 0,+ 0,+ 0,+
0,+ 0,+ 0,+ 0,+
00+ +
0,+ 0,+ 0 0,+
0,+ 0,+ 0,+ +
+ 0,+ 0,+ 0,+
+ 0,+ 0,+ 0,+
0,+ 0,+ 0 0
if + : Increase
- : Decrease
0 : No significant effect
I : Effect may go either way, depending on the situation
156
-------
after planting, causing flooding of the no-tillage slots and the young corn.
This reduced the no-tillage plant population; yield was 22% below that for
the conventionally tilled plots.
In West Virginia, Bennett et al. (1976) found significantly higher corn
yields for no-tillage over conventional tillage following sod. Further
evidence that no-tillage following sod helps improve yield relative to conven-
tional tillage was reported by Van Doren and Triplett (1969). In studies
conducted in Ohio from 1960-1967 on a clay-loam to clay textured soil, no-
tillage yields were 10% below conventional tillage following a row-crop, but
less than 4% below on corn following sod. In silt-loam textured soils, corn
yields for no-tillage and conventional tillage were approximately equal
following a row-crop, but following sod, no-tillage yields were nearly 13%
higher than yields with conventional tillage.
A summary of no-tillage yield potentials in Ohio (Forster et al. 1976)
indicated yield increases in the neighborhood of 10% on well-drained soils
(relative to conventional tillage), approximately equal yields on moderately
well-drained and somewhat poorly drained soils, and yield decreases in the
order of 10% on poorly and very poorly drained soils.
In general, reports indicate somewhat better results for no-tillage in
areas of moderate climate. Underwood (1976) reports 36% higher corn yields
for no-tillage over conventional tillage in Virginia while Swader (1970) and
Peters (1970) seem to indicate that at best no-tillage yields will equal
conventional-till yields in New York and New England. In colder climates,
cool spring soil temperatures appear to be a limiting factor in no-tillage
corn yields. Tillage buries mulch and aerates soil to facilitate warming.
Poorly drained soils warm more slowly than well-drained soils, adding to this
problem. Also, since no-tillage corn makes more efficient use of water
(Underbill, 1976), well-to excessively drained soils benefit from no-tillage,
particularly in dry seasons. Thus spring soil temperature and soil droughti-
ness are two of the key limiting factors in determining yield effects of
no-tillage.
Data on soybean yields under no-tillage are less prevalent. A number of
results reported by Clapp (1972) indicate no-tillage soybean yields are compar-
able to conventionally tilled yields under most situations and may be better
in cases where inadequate soil moisture is a problem.
Pest control is another key factor in determining no-tillage yields.
Weeds, insects, rodents, birds, and disease can all, in certain situations,
be greater problems with no-tillage systems (Phillips et al.. 1973). Thus
pesticide expenses are generally higher. A contact herbicide, usually
paraquat, must be applied when weed size is too great to be controlled by
systemic herbicides. There is some evidence that increased use of systemic
herbicides, such as atrazine may be necessary for optimal yield levels
(Bennett et a!. 1976).
Insect problems can also be a greater problem with no-tillage. Army-
worms, sod webworms, cutworms, slugs, and root aphids may all cause crop
157
-------
damages (Phillips, 1973). All of these factors add to a generally increased
level of pesticide expenditure for no-tillage over conventional tillage.
Although initial conversion to a no-tillage planter requires an in-
creased machinery investment, in the long run this should be offset by selling,
not replacing, or possible replacing on a smaller scale tillage equipment
such as plows and discs. The situation and attitude of the individual
farmer primarily determines which of these courses is followed. Clapp (1972)
reports savings of about $9-25/hectare in machinery expenses for soybeans.
Fuel savings of up to 73% have been reported for growing corn under no-
tillage as compared to conventional tillage. Often the increased costs for
pesticides and the decreased machinery and labor costs nearly balance.
Estimates from Indiana showed a net savings of only $14-25/ha for no-tillage
(Doster et ai- 1973). Thus, yield differences appear to be the most
crucial factor in determining the profitability of no-tillage practices.
Estimated budgets for continuous corn from Ohio State University (Forster
et. al. 1976) show increased profits of no-tillage over conventional of
$29-96/ha on soils where no-tillage yields were greater than conventional -
till yields, but decreased profits ranging from $54-103 on soils with lower
no-tillage yield potential.
Conservation Tillage
Conservation tillage is generally more suited to colder and wetter soils
than no-till (Stewart et aJL 1975). Chisel plowing or discing as a primary
tillage technique buries roughly 50% of surface plant growth and crop residue,
and helps aerate and warm the soil, thus partially overcoming some no-tillage
limitations. On moderately well-drained to somewhat poorly drained soils in
Ohio, conservation tillage, no-tillage and conventional tillage corn yields
were approximately equal. On poorly and very poorly drained soils, conser-
vation tillage yields were slightly better than no-tillage and in some
cases even better than with conventional tillage (Forster etal_. 1976).
In a three-year trial on a moderately well-drained silt loam in Illinois, a
chisel-plow tillage system produced average yields about 8% over conventional-
till and no-tillage yields. (Fink et al. 1974).
As would be expected, earlier comments on increased expenses for pesticides
and savings on machinery and labor inputs for no-tillage operations are some-
what modified with conservation tillage. Although many weeds are buried by
chisel and disc tillage, increased weed problems as compared with conventional
tillage are still expected (Erbach e_t al. 1974). Many of the insect pro-
blems of no-tillage are also present with minimum-tillage but at a reduced
level (Musiek et al _. 1973).
Machinery inputs are less than with conventional tillage since chisel
or disc plowing is faster and uses less energy than moldboard plowing. Com-
pared to conventional tillage systems, savings of 36% for a chisel-plow system
(James McDivett, Personal Communication, 1977) and 46% for a disc tillage
system (Wittmuss _et al_. 1975) have been estimated.
The moderating effects of conservation tillage on net profits as compared
to no-tillage are shown in budget estimates by Forster et al. (1976). On
158
-------
two soils, one moderately well-drained and the other somewhat poorly drained,
no-tillage profits averaged approximately $38/ha above conventional tillage,
while conservation tillage averaged only about $21/ha greater. However, on
poorly and very poorly-drained soils, no-tillage average $78/ha less than
conventional while conservation tillage averaged $10/ha less.
Sod-based Rotations
The effect of this practice on yields is represented by the difference
between yields from row crops grown in sod-based rotations, particularly corn,
soybeans, and cotton, as compared to yields from growing these crops con-
tinuously. Sod crops may help improve soil structure, organic matter content,
and infiltration. The extent to which this is true depends on the soil type
and level of management. High clay soils, working soils when too wet, and
removing of all crop residue all tend to produce soil conditions which could
be improved by a sod crop (S. D. Klausner, Personal Communication, 1977).
Under any or all of these conditions, yield increases may be expected for the
first one to two years after plowing under sod. Yields of sorghum forage
grown on a clay soil in Texas decreased 40% by the fourth year following
clover. Soil analysis showed that there was a significantly higher percent
of water stable aggregates and organic matter in soils in a 2-year sorghum-
sweetclover rotation than in soils growing continuous sorghum (Adams, 1974).
Other rotation experiments in Texas clay soils showed cotton yields were
considerably higher (19-22%) in cotton rotations with a legume. In rotation
with a non-legume, yield ranges were +4% to -17% relative to continuous
cotton. Average yields for corn grown in rotations were about 12% above
continuous corn (Jetter et al. 1962).
On a fine sandy loam in southern Texas, rotations did not have a signi-
ficant effect on cotton yields. One rotation (cotton, clover, cotton, to-
matoes) did, however, apparently somewhat retard development of cotton root
rot (Gerard et al. 1963). Crop rotations can often be effective in breaking
weed, disease, and insect cycles (Stewart et §1. 1975), particularly for
corn rootworm, which can often be completely controlled by using a sod
rotation (Luckmann, 1977). Generally, this would imply either a lower pesti-
cide requirement or a higher yield or both.
Savings on nitrogen fertilizer may be considerable in a legume-based
rotation. Up to 340 kg/ha nitrogen may be plowed down in a good alfalfa
legume crop, of which 170 kg may be available the first year (S. D. Klausner,
Personal Communication, 1977). Although other estimates have been somewhat
lower, it is generally accepted that legumes can help meet nitrogen require-
ments for row crops (Stewart et aj_. 1975).
Equipment costs for row crops in a rotation should not change, other than
that capital costs for specialized equipment may have to be spread over fewer
hectares. However, if a farmer has specialized in continuous row crops, pur-
chase of forage equipment may be necessary to begin crop rotations, adding
considerably to machinery costs.
Changes in labor requirements depend completely on the relative labor
intensity of the forage crop versus the row crop. Requirements for growing
159
-------
are generally less for hay crops, although harvesting requirements may be
higher, particularly for handling dry hay. Even when total labor requirements
increased, this may be offset by the more even labor distribution over the
growing season.
Cover Crops
Winter cover-crops may have either positive or negative effects on the
yield of the succeeding row crop. If excess soil moisture in spring postpones
soil warming or field operations, cover crops may help dry fields, thereby
allowing timely planting with resultant yield increases. However, under
dry conditions, cover crops may draw needed soil moisture, and consequently
have an adverse effect on yields (Stewart etal. 1975).
If rye, oats, or other non-legumes are used as a cover-crop, nitrogen
requirements for the row-crop are unchanged, but extra N may be needed to
facilitate breakdown of cover crop. In areas where a good legume can be
established over winter, this may reduce nitrogen fertilizer requirements
in the next production period (Mitchell ejt al. 1977).
In general, pesticide requirements are unchanged from those for con-
ventional tillage. However, with conservation or no-tillage, a contact
herbicide may be required to kill the cover crop.
Equipment and labor expenses are increased to cover planting of the
cover crop.
Contouring
Since contouring tends to decrease runoff and keep water on the field,
yields would be expected to increase where normal soil moisture is inadequate,
but be decreased in areas of excess rainfall or poor drainage.
Data reported by U.S.D.A. in 1945 (Stalling, 1945a) from various studies
around the country show an average increase in corn yields of about 17%
with contouring. Of 31 results reported, only one, from Temple, Texas, shows
a yield decrease (28%). The same report shows average soybean yield increases
of 14% and average cotton increases of about 12%. No soybean and only one of
seven cotton trials showed a yield decrease.
Because of the possibility of short or pointed rows, contouring may
require more labor time, particularly in maneuvering multi-row equipment.
Land may also be lost in field corners where .rows are too short to be use-
ful. On some topographies, contouring is not practical (Stewart et. al. 1975).
Graded Rows
All of the cost related to contouring also affect graded rows. Yield
effects should be more moderate since excess water is allowed to runoff
along the row. Machinery efficiency may be higher than with contouring by
grading rows to avoid point rows (Richardson, 1973). Construction of grassed
waterways or other outlets may, however, add construction and maintenance
costs.
160
-------
Diversions and Grassed Waterways
When constructed as an SWCP to help control runoff and erosion, diversions
and grassed waterways do not necessarily affect yields. However, diversions
in particular may help remove excess water allowing more timely field operations
and possible yield increases. Some land is lost from production to the
diversion and waterway. Construction and maintenance costs are quite varied,
depending on the particular situation.
Strip cropping
Since strip cropping always implies rotations and may imply contouring,
yield effects of both should occur. Stallings (1945a) shows no consistent
effect on yields for several crops. However, many of the reported results
were from one year only.
All costs associated with contouring and crop rotations may occur with
strip cropping, i.e. nitrogen and pesticide use may decrease, but machinery
and labor requirements may increase or stay the same.
Terraces
Like some other practices, terraces can have a positive effect on yield
by making more efficient use of water. This would be particularly true
in dry areas where level bench terraces are constructed. Little water con-
servation occurs on gradient terraces unless they are contoured (Stewart,
et al_. 1975). Since terrace construction disturbs topsoil, initial
reductions in yield may occur. However, most terrace construction is on
deep soils, often loess. For these soils, yields associated with terraces
should eventually equal or surpass non-terraced yields (Spomer et al. 1973).
Stallings (1945a) reports results of the effect of both level and
graded terraces on yield. For level terraces, sorghum yields averaged 36%
higher, wheat 55% higher, cotton 42% higher, corn 6% higher and oats 4%
higher. All are as compared to yields from unterraced fields. The lower
benefits for corn and oats are not explained, but are from only one year at
Alcester, South Dakota. The other crops were grown in other states and
results are for longer time intervals. Stallings notes that all these results
are from dry areas (South Dakota, Texas, and Oklahoma) and similar results
would not be expected in humid regions.
Results from graded terraces are more varied. Corn yields, as percent
of yields from unterraced fields, ranged from 73% in Iowa (3 years) to 126%
in Wisconsin (3 years) and 166% in Oklahoma (1 year). Cotton yield increased
from 10% to 30%. A 5-year trial in Wisconsin, showed terraced hay yields
increased 11% over hay grown on non-terraced soils.
Non-parallel terraces, like contouring, may require additional machinery
and labor time. Based on 2 row equipment, Mitchell et_ al_. (1965), esti-
mated that approximately 20% additional field time is required for growing
and harvesting soybeans on non-parallel terraces as opposed to flat-land
farming. Parallel terraces, however, showed a 10% decrease in field time
over non-parallel terraces using four-row equipment for corn production.
Construction of level parallel terraces may save machinery and labor time
161
-------
over non-terraced farming if steep slopes or irregular topography create
odd shaped fields (Spomer et_ al. 1973).
Construction and maintenance is the most costly part of terracing,
particularly for parallel terraces on uneven topography or terraces requiring
grassed back slopes or tile outlets (Stewart et_ al_. 1975).
Ridge Planting and Contour Listing
These practices may increase or decrease yields. Ridge planting is
designed for humid areas with excess rainfall, particularly in spring.
Ridges warm and dry more quickly and have less problems with excess water
during the growing season (Stewart et _al. 1975). Contour listing, on
the other hand, holds water in theTiirrows where the crop is planted,
thereby improving water use efficiency. Ridging should improve yields
under excessive moisture conditions, but may cause decreased yield in dry
periods. Contour listing should have just the opposite effect.
Because of possible difficulties in cultivating, either practice may
require additional herbicides, depending on the particular situation.
Both practices require some specialized equipment to form the ridges
and/or furrows which are generally more expensive than conventional equip-
ment. Like other operations performed on the contour, additional field
time may be required for short or pointed rows.
Timing of Field Operations
In areas where fall plowing and fertilizer application are conventional,
switching these operations to the spring may cause yield decreases because
the planting date may be delayed due to the slower warming and drying of the
soil. This is particularly true on fine textured soils with some drainage
problems. Research in Iowa and Minnesota shows little or no corn yield gain
for fall plowing on well-drained medium textured soils (Swan et^ al_. 1972).
However on fine-texture, moderately well to poorly drained soils in southern
Minnesota, corn yields on fall plowing averaged 10% greater than yields on
spring plowing over a six-year trial. Other results from Illinois (Bateman
et_ al. 1967) and Indiana (Doster et_ al. 1973) show little difference in
corn yields on fall plowing. Since "tKese states are possibly drier
than Minnesota and Iowa, early spring drying of the soil may be less ad-
vantagous and possibly a disadvantage.
If nitrogen fertilizer is applied in the fall, additional amounts may
be needed due to losses by leaching and denitrification (Stewart et al.
1975).
Switching to spring plowing may require additional or larger equipment
and require additional spring labor if planting is to be completed by or
near the optimal date. Doster (1973) indicates two 70-hp tractors, one
4-16" (40 cm ) plow and one 4.3 m disc are adequate for preparing 240
hectares of corn land in central Indiana with fall plowing. On the other
hand, spring plowing requires one 140 hp tractor, one 100-hp tractor, one
6-16" (40 cm ) plow, one 5-16" (40 cm ) plow, and a 7.3 m disc. Total
machinery plus labor costs per hectare were estimated at $40.30 for fall-
plow and $49.90 for spring-plow.
162
-------
Some other factors may be actual costs in SWCPs, but are either not
well documented, not evident in the short run, or not significant. Most
contouring operations require some initial cost in laying out the contours.
This generally is done only once, requires minimal equipment, and often is
done at no cost to the farm by Soil Conservation Service personnel.
Fertilizer savings, particularly of nitrogen and phosphorous are usually
associated with retaining top soil. Presently, fertilizer losses associated
with soil and water losses are generally not reflected in most fertilizer
recommendations. This is partly because the bulk of these losses are with
soil organic matter and are in a form not immediately available to crops.
Thus the ultimate effect of these losses is in long-term soil fertility.
Evaluation of Cost-Effectiveness for Selected SWCPs
As indicated in the previous section, costs for implementation of SWCPs
generally fall within a range of values and depend primarily upon the crop,
the soil, the climate, and the physical properties of the field. Since it
is not practical to evaluate all possible conditions, one example situation
has been chosen to illustrate the magnitudes of cost-changes associated with
some common SWCPs. A budget for continuous, conventionally tilled, straight
row corn grain is used as the base condition against which budgets for the
SWCPs are compared. This budget is presented in Table 8-2. Note that fixed
costs for land, machinery, insurance, etc. are not included. These costs
do not change significantly when implementing SWCPs.
Changes in Variable Costs
Table 8-3 shows the changes in variable costs associated with implement-
ing some selected SWCPs. To facilitate comparisons, the columns in 8-3
correspond to the columns in Table 8-1. "Land" is not included since no
TABLE 8-2. SAMPLE CORN GRAIN BUDGET FOR USE IN
CALCULATING COSTS OF SWCPs
Continuous Corn Grain, No SWCPs
Yield MT/ha 7.9
Gross Income, $/ha 695
Variable Costs, $/ha :
Nitrogen Fertilizer
Pesticides
Equipment
Labor
Construction, Maintenance and
Other
Total Variable Cost
Net Income above Variable Costs $445/ha
163
-------
TABLE 8-3. TYPICAL CHANGES IN VARIABLE COSTS ASSOCIATED WITH
IMPLEMENTATION OF SWCPs AS COMPARED TO CONTINUOUS CORN GRAIN
Values represent expected change in input cost from continuous,
straight-row, conventionally tilled corn grain (Table 8-2)
Nitrogen
SWCP Pert. Pesticides
No-tillage 0 +24
Conservation 0 +5
tillage
Rotation* -26 -30
Contouring 0 0
Diversion -2 - 3
Strip Crop- -26 -30
ping
c
Terrace A 0 0
Terrace B* 0 0
$/ha Const.,
Maint.
Equipment Labor § Other Total
-10 -9 0 +5
- 4 -3 0 -2
-15 -1 -2 -74
+ 6 +3 0 +9
+3 +2 +15 +15
-11 +2 -2 -67
+6 +3 +101 +110
+6 +3 +60 +69
* Six year rotation with three years corn, one year oats, two years hay.
Values are average for the six years.
#
One diversion ditch across center of 120 m slope, with contouring.
Construction costs are amortized over 45 years.
Terrace A has a terrace at 30 ra , 60 m and 90 m, respectively, above
lower edge of field with 120 m slope, with contouring. Construction
costs amortized over 45 years.
Terrace B has one terrace at lower edge of 120 m slope, with contouring.
Contouring costs amortized over 45 years.
164
-------
significant amount of land was assumed to be lost for production with any of
the practices except the diversion ditch. The diversion required 5% of the
land and thus reduced all variable costs by 5% below what they would other-
wise be.
For both no-tillage and conservation tillage decreases in machinery and
labor expense nearly balance increased pesticide expenditures. Thus the most
important factor determining changes in farm income associated with these
practices is their effect on yields.
Contouring also shows only a relatively small change in variable costs.
However, this practice is not applicable in many fields. Also for this
situation, machinery and labor costs for contouring were assumed to increase
by 10% above costs for straight-row tillage. Increases of 3 to 4 times this
much could occur in many situations.
Changing from continuous corn to a rotation (50% corn) causes a large
decrease in variable costs. This is due not only to the lower costs
associated with oats and hay, but also to a savings in nitrogen and rootworm
insecticide costs amounting to $47.00 for the first year of corn following
hay.
Strip cropping costs are simply a combination of rotation costs and
costs for contouring. No additional costs would normally occur.
Three systems of structures are considered. To budget these on a per
hectare basis, a slope length and gradient had to be specified. These were
set at 120 m and 4%, respectively. All field operations with structures are
assumed to be done on the contour, thus costs include costs for contouring.
Construction costs are amortized over 45 years. The diversion system is one
ditch 6 m wide across the center of the field. Since the ditch cannot be
used for crop production, all crop inputs are reduced by 5% (as is yield).
The two terrace systems considered are both parallel tile-outlet terraces.
Terrace A has a horizontal interval of 30 m which would be approximately
equal to SCS specifications. Three terraces are constructed at 30 m , 60 m ,
and 90 m , respectively, above the lower edge of the field. Terrace B is es-
sentially a sediment basin with one terrace running along the lower edge of
the field.
Yield Effects and Total Costs
Some SWCPs can have substantial effects on yields, depending primarily
on the climate, soil texture, and soil drainage characteristics. In order
to take yield effects into account, total costs have been calculated for
these practices for a range of yield levels. These are shown in Table 8-4.
Each practice is listed with one or more yield levels. For each yield level,
total cost is calculated, where total cost is the change in net income from
the base system, conventionally tilled corn grain. A decrease in net income
is shown as a positive cost.
165
-------
TABLE 8-4. EXAMPLES OF TOTAL COSTS ASSOCIATED WITH
IMPLEMENTATION OF SWCPs AS COMPARED TO CONTINUOUS CORN GRAIN
SWCP
No-tillage
Conservation
tillage
g
Rotation
Contour
g
Diversion
Strip Cropping
c
Terrace A
c
Terrace B
Assumed*
Yield
Change
+ 10%
0
-10%
+ 5
0
- 5%
+ 4%*
+ 5%
0
- 5%
+ 4%*
0
0
Corn
Grain
Yield
MT/ha
8.7
7.9
7.1
8.3
7.9'
7.5
(**)
8.3
7.9
7.5
(**)
7.9
7.9
Gross
Income
$/ha
765
695
625
720
695
660
524
720
695
660
524
695
695
Change in
Gross
Income
$/ha
+ 70
0
-70
+35
0
-35
-171
+ 35
0
-35
-171
0
0
Change in
Variable
Costs
(Table 8-3)
$/ha
+5
+5
+5
-2
_2
-2
-81
+9
+9
+ 15
-74
+ 110
+60
Total
Cost
$/ha
-65
+ 5
+ 75
-37
- 2
+33
+90
-26
+9
+50
+97
+ 110
+60
* Base corn grain yield is 7.9 MT/ha (Table 8-2).
#
A positive total cost indicates a decrease in net return compared to
continuous corn grain with no practices. A negative total cost indicates
an increase in net return.
c
Specifications for these practices are in Table 8—3.
* The 4% yield increase for rotations and strip cropping applies only to
corn the first year after hay.
** Rotation and strip cropping yields are: oats, 3.1 MT/ha ; straw, 2.2 MT/ha ;
hay, 19.7 MT/ha ; corn, 8.2 MT/ha first year after hay, 7.9 MT/ha second
and third years.
166
-------
The reduced tillage systems show the greatest variance having substantial
increases in net income (negative total costs) when yields increase, but also
substantial losses in income if yields decrease. Even though the rotation
(and strip cropping) showed a large decrease in variable costs (Table 8-3),
it still is one of the most costly practices due to the low levels of return
on oats and hay as compared to corn.
Except where yield increases are expected, the total cost for many of
these practices is a fairly significant proportion of the net income above
variable costs with the base crop (Table 8-2). The total cost for what
would be considered the standard terrace system, Terrace A, reduces this net
income by about 25%. Thus careful consideration of the expected and desired
effects should be given before undertaking a program of implementation.
Effectiveness of SWCPs
The effectiveness of SWCPs for reducing sediment delivery to waterways
depends on both the initial level of soil erosion and the initial sediment
delivery ratio (SDR). To investigate the relative effectiveness of different
practices and how effectiveness changes for different field situations,
levels of soil erosion and sediment delivery have been calculated for a few
hypothetical field situations. Initially, levels of soil erosion were
calculated for those practices which were budgeted in the previous section
on a field with a 120 m slope and a 4% gradient. These are presented in
Table 8-5 in order of decreasing levels of soil erosion. Contouring is
assumed for the diversion and the terrace systems. Since the Terrace B system
has only one terrace at the lower edge of the field, it acts essentially
as a sediment basin and reduces soil erosion little beyond that already
accomplished by contouring.
Soil erosion levels for practices such as no-tillage and conservation
tillage which depend primarily on crop residue for their effect are sensitive
to both crop yield and amount of residue left on the field. The corn yield
used here was fairly high, over 30% above the national average. Thus soil
erosion levels are low especially for no-tillage. For yields of 5.7 MT/ha ,
approximately average for the country, soil erosion levels with no-tillage
would be about 14 MT/ha , while at yields of 4.4 MT/ha , soil erosion would
rise to nearly 21 MT/ha.
To calculate levels of sediment delivery, the SDR must also be known.
The SDR is a function of the location of the field relative to waterways and
thus is independent of practices on the field except those which cause water-
ways to be built on or near the field. For this example, diversions and ter-
races provide waterways in the field and thus may change the SDR, but only
for that area of the field lying above the structure. Because of redeposition
in the terrace channel, only 20% of sediment reaching the terrace channel
is assumed, to be lost through the tile-outlet. Thus the SDR for the area above
a terrace is 0.2. All sediment reaching a diversion ditch is assumed to have
entered a waterway, thus the SDR for areas above diversions is 1.0.
Sediment delivery is initially calculated for two field situations.
First an SDR of 1.0 is assumed, which would occur for a field lying directly
167
-------
adjacent to a waterway. Next the SDR is assumed to be 0.3, which corresponds
to fields lying roughly in the range of 200 - 300 meters from a waterway.
Levels of sediment delivery and effectiveness for the different practices
are listed in Table 8-5.
To simplify the discussion on the effectiveness of SWCPs, comparison of
the different practices will first be discussed on Field One only (SDR = 1.0),
Then the change in effectiveness for each practice when going to Field Two
(SDR = 0.3) will be examined.
Since the initial SDR is 1.0, sediment delivery for all practices which
do not affect the SDR are the same as the levels of soil erosion. Terraces,
however, reduce the SDR and thus are relatively more effective for control-
TABLE 8-5. LEVELS OF SOIL EROSION AND SEDIMENT DELIVERY
FOR AN EXAMPLE FIELD* AT TWO SEDIMENT DELIVERY RATIOS
Practice
Erosion
Field One
Sediment
Delivery
MT/hectare
SDR = 1.0
Effect-*
iveness
Field Two SDR = 0.3
Sediment
Delivery
Effect-
iveness
#
MT/hectare
Corn Grain, no
SWCPs (Base crop)
Contour
P
Terrace B
Conservation
tillage
Rotation
c
Diversion
No-tillage
Strip Cropping*
P
Terrace A
31.6
15.8
15.5
14.0
13.4
11.2
8.0
7.9
7.5
31.6
15.8
3.1
14.0
13.4
11.2
8.0
7.9
3.0
15.8
28.5
17.6
18.2
20.4
23.6
24.7
28.6
9.5
4.7
3.1
4.2
4.0
7.3
2.4
2.4
1.7
4.8
6.4
5.3
5.5
2.1
7.1
7.1
7.8
* Slope length = 120 m , Gradient = 4%. USLE factors:
R = 200, K = 0.30.
Effectiveness is the reduction in sediment delivery from the base crop
continuous conventionally tilled corn grain.
Practice specifications are in Table 8-3.
168
-------
ling sediment delivery than for controlling soil erosion. Also, even though
the more intense terrace system (Terrace A) reduces soil erosion to about
half the level achieved by the simple terrace system (Terrace B), levels of
sediment delivery are about equal for both systems. Since the intense
terrace system has three terraces across the slope, but none at the lower
edge, the SDR on the lower 25% of the field remains at 1.0, while on the
upper 75%, above the terraces, it is only 0.2. Thus the average SDR over the
field with this terrace system is 0.4. The other system (Terrace B) has only
one terrace, which is located at the lower edge of the field, thus the SDR
for the whole field is only 0.2 with this system, or half of what it is with
Terrace A. Although the effectiveness of Terrace A could be increased by
putting a fourth terrace at the lower edge of the field, this would add
significantly to its cost. This option will be discussed further in the
section on cost-effectiveness.
The diversion ditch also can change the SDR in a field. In this case,
however, the SDR is already 1.0, so no change takes place. Thus, for this
situation levels of soil erosion and sediment delivery are the same for the
diversion ditch.
In general, for any given field situation, the effectiveness of non-
structural practices will be proportional to their respective reductions in
soil erosion. For structural practices which decrease the SDR, effectiveness
will be more than proportional to reductions in soil erosion, while the
effectiveness of practices which increase the SDR will be less than propor-
tional to their decreases in soil erosion.
Table 8-5 shows that the effectiveness of all practices is reduced for
Field Two, with an SDR of 0.3, even though levels of soil erosion have not
changed. For the non-structural practices, this reduction is directly pro-
portional to the difference in the initial (base) levels of sediment delivery
for the two fields. For the terraces and the diversion, the decrease in
effectiveness is more than proportional to the difference in the base levels.
This can be seen most clearly for the diversion ditch. Since the diversion
raises the SDR to 1.0 from its initial level of 0.3 for the area above it,
its effectiveness for reducing sediment delivery on this field is below
that for all other practices. Sediment delivery is actually higher with the
diversion than it would be without it. It is assumed that field operations
are done on the contour when a diversion ditch is constructed. Table 8-5
shows that for fields with initially low SDRs, such as Field Two, more sedi-
ment is lost with the diversion than with contouring alone.
The terrace systems are also comparatively less effective for controlling
sediment delivery than non-structural practices on fields with a low SDR.
Terrace B, for example, is more effective than all the non-structural practices
on Field One, but less effective than either no-till or strip cropping on
Field Two.
Any structural practice which changes the SDR for a field to some fixed
number by bringing a waterway to or near the field will be relatively more
effective on fields having an initially high SDR. A striking illustration
169
-------
of this can be seen in the following example. A third field, Field Three,
with an SDR of 0.3, and with levels of soil erosion four times higher than
on Field One is used to illustrate the importance of the SDR in determining
the effectiveness of structures. Using Terrace A as an example, levels of
soil erosion, sediment delivery, and effectiveness are shown in Field One
and Field Three in Table 8-6. Even though the initial levels of both soil
TABLE 8-6. EFFECTIVENESS OF A TERRACE SYSTEM IN
REDUCING SEDIMENT DELIVERY ON TWO EXAMPLE FIELDS
Field One
SDR = 1.0 MT/ha
Soil Sediment Effect-*
Erosion Delivery iveness
Field Three
SDR = 0.3 MT/ha
Soil Sediment Effect-
Erosion Delivery iveness
Corn Grain
(Base Crop) 31.6 31.6 --- 126.4 37.9
Terrace A* 7.5 3.0 28.6 30.0 6.8 23.2
* Effectiveness is the change in sediment delivery with implementation of
SWCP
#
Terrace A is specified in Table 8-3.
erosion and sediment delivery were higher on Field Three than on Field One,
a higher degree of effectiveness of the terrace system is obtained by implementing
it on Field One because of its initially higher SDR.
Although the above example was hypothetical and the relationship would
not necessarily be the same for other structures or even other terrace systems,
it does illustrate the point that simply looking at levels of soil erosion
does not necessarily indicate those fields or areas for which SWCPs would be
most effective in reducing sediment delivery.
The final point on practice effectiveness deals with the optimal place-
ment of crops on different areas of a farm or field depending upon the SDR
for each area. In general, maximum effectiveness in reducing sediment
delivery is obtained if crops or SWCPs with the lowest associated levels of
soil erosion are placed in areas with the highest SDRs. This is illustrated
in Table 8-7. Using Field Two with an initial SDR of 0.3 as an example,
it was assumed that contouring was implemented with a diversion ditch con-
structed across the center of the field. This reduces average levels of
soil erosion by about 73%. The upper half of the field, above the diver-
sion, now has an SDR of 1.0, while the SDR on the lower half remains at 0.3.
If half the field were to be put into strip cropping which has a relatively
low level of soil erosion, the other half into contoured corn grain which
170
-------
has a considerably higher level of soil erosion, significant differences in
reductions in sediment delivery are obtained depending on where the two crops
are placed, even though placement has no effect on the level of soil erosion.
Table 8-7 shows that placing the strip cropping above the diversion and
contoured corn below (Option Two) increases the reduction in sediment deli-
very (effectiveness) by nearly 70% over reversing this placement, even though
no change in the level of soil erosion occurs.
TABLE 8-7. OPTIMAL PLACEMENT OF CROPS FOR
REDUCING SEDIMENT DELIVERY IN A FIELD WITH A DIVERSION DITCH
Field Two
(initial SDR =0.3)
Soil Erosion Sediment Delivery Effectiveness
MT/ha
Corn Grain 31.6 9.5 —
(Base Crop)
Option One:
Contoured Corn above
diversion, strip crop-
ping below diversion 8.4 6.4 3.1
Option Two:
Strip cropping above
diversion, contoured
corn below diversion 8.4 4.5 5.0
Cost-Effectiveness of SWCPs
When a goal has been set to reduce the levels of sediment reaching water-
ways, and there is a cost associated with reaching the goal, then it is
desirable to have some measure of the efficiency of plans or practices which
may be implemented to achieve the goal. Cost-effectiveness, defined as the
reduction in sediment delivery achieved per dollar of cost, is one measure
of the efficiency of practices, combinations of practices, or total farm
plans for reaching the goal of reduced sediment delivery. In the previous
two sections, costs and levels of effectiveness were determined for a number
of SWCPs under certain specified conditions. In Table 8-8, these costs
and levels of effectiveness are combined to determine cost-effectiveness values
for SWCPs The cost-effectiveness values were calculated for the two field
situations used to investigate effectiveness (Table 8-5), and with the various
levels of change in yields used in calculating costs (Table 8-4). In sit-
uations where yield increases would occur when implementing either no-tillage,
or contouring or where they would not decrease when changing to conservation
tillage, increases in net income (negative costs) occur. Cost-effectiveness
is not defined for these situations since it is assumed they would be adopted
for economic reasons and thus would be part of the base system for a farm.
171
-------
TABLE 8-8. COST-EFFECTIVENESS VALUES FOR
SELECTED SWCPs FOR TWO EXAMPLE FIELDS
Assumed
%
Yield
Practice Change
Contour +5%
0
Terrace B 0
Conservation +5%
tillage 0
-5%
Rotation* +4%
Diversion -5%
Strip Cropping +4%
Terrace A 0
No-till +10
0
-10%
it
Cost*
($/ha)
-26
9
60
-37
-2
+33
+90
+50
+97
110
-65
+5
+ 75
Field One
Effect-
iveness
(MT/ha)
15.8
15.8
28.5
17.6
17.6
17.6
18.2
20.4
24.7
28.6
23.6
23.6
23.6
, SDR = 1.0*
Cost-Effec-
tiveness
(MT/$)
(**)
1.76
.48
(**)
(**)
.53
.20
.41
.25
.26
(**)
4.72
.31
Field Two, SDR = 0.3*
Effect-0
a
iveness
(MT/ha)
4.8
4.8
6.4
5.3
5.3
5.3
5.5
2.1
7.1
7.8
7.1
7.1
7.1
Cost-Ef-
fectiveness
(MT/$)
(**)
.53
.11
(**)
(**)
.16
.06
.04
.07
.07
(**)
1.42
.09
* Field One and Field Two have the same levels of soil erosion and differ
only in their SDR. Field specifications are in Table 8-5.
Cost is defined as the loss in income with the SWCP as compared to contin-
uous corn grain with no SWCPs. See Table 8-4.
c
Effectiveness is defined as the drop in sediment delivery with the SWCP
as compared to continuous corn grain with no SWCPs. Initial levels are
31.6 MT/ha on Field One and 9.5 MT/ha on Field Two. See Table 8-5.
**
Specifications for practices are in Table 8-4.
Cost-effectiveness is not defined for practices which show an increase
in net return (negative cost).
172
-------
Many of the factors affecting the cost-effectiveness of practices were
discussed in the section on costs and the section on effectiveness. Thus
some repetition is involved in analyzing cost-effectiveness. Also, as
previously mentioned, there is a fair amount of variance in both the cost
and the effectiveness of these practices in other field situations. Thus the
values of cost-effectiveness and their rank do not necessarily remain con-
stant for all cases. However, a number of conclusions can be reached, using
Table 8-8 as a guide.
It would appear that the practices with the highest potential levels of
cost-effectiveness are no-tillage, contouring, and conservation tillage.
However, there are many fields where these high levels cannot be achieved.
Yield losses will often occur with no-tillage in areas with poorly drained
or finely textured soils or which typically have cold spring soil temperatures.
Yield losses greatly reduce the cost-effectiveness of no-tillage. Contouring
is only practical in areas with a reasonably even topography. Although
Table 8-8 does not show a particularly high value for the cost-effectiveness
of conservation tillage, it may be adaptable in more situations than either
no-tillage or contouring. Under most conditions, yield levels can be main-
tained at least equal to conventional tillage. Thus, often little or no
cost is associated with this practice.
Switching to a sod-based rotation is one of the least cost-effective
practices shown in Table 8-8. However, this value also can vary considerably.
Hay markets are generally local rather than national, thus prices vary con-
siderably. The price used here was $37/MT, which is based on average
current prices in Iowa. In some areas, particularly the Northeast, prices
have been nearly double that amount in recent years. Also, most dairy and
beef farms grow some hay to obtain a balanced feed ration. The cost of
increasing the amount of hay grown for this type of farm may be less than
for a cash-crop farm, since the hay can be utilized on the farm rather than
being sold.
The cost-effectiveness for strip cropping shown in Table 8-8 is also
quite low. However, it is based on changing from continuous corn grain to
strip cropping. In situations where a sod-crop is already being grown and
thus strip cropping can be implemented with no increase in the area of sod
crops, it will often be one of the most, highly cost-effective practices
available. In this situation, the cost would be approximately equal to
that for contouring, and the effectiveness (and thus the cost-effectiveness)
about 50% greater than for contouring.
The two terrace systems have nearly equal levels of effectiveness,
however, Terrace B having only one terrace along the foot of the slope is
considerably more cost-effective than Terrace A, which has terraces at 30-m
intervals starting 30 m above the foot of the slope. It was noted in the
previous section that the effectiveness of Terrace A system could be in-
creased by adding one more terrace at the foot of the slope. This situation
is evaluated in Table 8-9. There is an increase in costs, a slight increase
in effectiveness and a slight decrease in cost-effectiveness for this system
as compared to Terrace A. However, a fourth terrace system, Terrace D,
evaluated in Table 8-9 has three terraces, like Terrace A, but space at 40-m
173
-------
intervals with one terrace at the lower edge of the field. Terrace A and
Terrace D have the same cost, Terrace A has a lower level of soil erosion,
but Terrace D is most effective for reducing sediment delivery and thus has
the higher cost-effectiveness. Terrace B is still the most cost-effective
structural practice. However, there may be a question on the technical
feasibility of placing just one terrace on such a long slope. It is
possible that supporting practices such as no-tillage or strip cropping would
be required in some field situaitons. More research in this area may be
required.
TABLE 8-9. COST, EFFECTIVENESS, AND
COST-EFFECTIVENESS OF A MORE INTENSE TERRACE SYSTEM
*
Terrace C
#
Terrace D
Cost
($/ha)
145
110
Soil
Erosion
(MT/ha)
7.4
8.7
Field One
Effect-
iveness
(MT/ha)
30.1
30.0
SDR = 1.0
Cost-Effec-
tiveness
(MT/$)
.21
.27
Field Two
Effect-
iveness
(MT/ha)
8.0
7.8
SDR = 0.3
Cost-Ef-
fectiveness
(MT/$)
.06
.08
Four terraces at 30-m intervals starting at the foot of the slope.
Three terraces at 40-m intervals starting at the foot of the slope.
All the practices have lower values of cost-effectiveness on Field Two
than Field One. Except for the terraces and diversions, the levels of cost-
effectiveness for any individual practice on the two fields are proportional
to the initial levels of sediment delivery for the two fields. This relation-
ship is independent of the individual SDR or the level of soil erosion for
either field. However, comparisons of this type are not valid if the cost of
implementing a practice varies considerably between two fields.
This type of relationship also holds for structural practices. However,
as discussed in the section on effectiveness, the cost-effectiveness of
structures depends on both the initial SDR and the initial level of sediment
delivery. Thus if two fields have identical levels of sediment delivery, but
different SDRs and levels of soil erosion, structures will be more cost-
effective on the field with the higher SDR (and thus the lower level of soil
erosion). Non-structural practices would have the same cost-effectiveness on
both fields. Table 8-8 shows a ratio of about 3.3:1 for levels of cost-
effectiveness of each non-structural practice on Field One versus Field Two.
However, Terrace A, Terrace B, and the diversion have ratios of 3.7:1,
4.4:1, and 10.2:1, respectively, reflecting their relatively lower levels
of cost-effectiveness on Field Two which has the lower SDR.
The cost-effectiveness of a combination of practices will generally
be less than for either practice by itself. Costs for combinations are
usually simply the sum of individual costs. Effectiveness, is usually pro-
174
-------
portional to existing levels of sediment delivery. Thus, for instance, if
two practices both reduce sediment delivery by, say, 50%, and both have the
same cost, then in combination, their cost is the sum of the individual
costs or twice the cost of either by itself. The effectiveness, however, is
only 1.5 times the individual effectiveness of either practice. Thus the
cost-effectiveness of the combination is just 75% (1.5/2) of the cost-
effectiveness of either individual practice. Thus combining practices tends
to reduce cost-effectiveness. In cases where two practices are combined which
individually have significantly different values of cost-effectiveness, the
cost-effectiveness of the combination may be higher than for the practice
which originally had the lowest value, but will always be less than the simple
average of the two individual cost-effectiveness values. For example, the
cost-effectiveness of combining no-tillage (with no assumed yield change)
with contouring on Field One (Table 8-8) is 2.06 MT/$, higher than for con-
touring alone, but lower than the average of the two practices (3.49 MT/$).
The result of this relationship is that when high degrees of effectiveness
are desired such that practice combinations are required, costs will in-
crease proportionately more than the decreases in sediment delivery. Another
way of viewing this conclusion is to note that if a certain amount of money
were available to invest in reducing sediment delivery, it would be more
effective to attempt to achieve moderate reductions over a wide area than to
concentrate on large reductions in a smaller area.
Farm Plans for Controlling Sediment Delivery and Soil Erosion on
Three Example Farms
The three example farms represent different types of farming operations
in three parts of the country: New York, Iowa, and Texas. The farms are
laid out as idealized models of a typical soil association in the areas
being studied. Each is a rectangle with a stream flowing through the center.
Individual soils of the soil association are in strips parallel to the
stream, positioned as they would typically occur on a watershed. A list
of the soils for each farm, their slopes, slope lengths, and distances from
the stream are given in Table 8-10. Each soil type was considered an individ-
ual field on which one or more crops could be grown.
The discussion on each example farm concentrates on looking at a farm
plan for reducing sediment delivery by 50%, and then a plan for obtaining a
90% reduction. Because of the importance of practice and crop location in farm
plans for reducing sediment delivery these plans are illustrated as well as
described. The plans are compared to plans for achieving similar reductions
in soil erosion. In general, the cost of reducing soil erosion by a given
percentage is greater than that for achieving the same percentage reduction
in sediment delivery. This is due to a number of factors, including:
1). Restrictions on soil erosion are placed on each individual
soil type, while restrictions on sediment delivery apply
only to the average over the whole farm.
2). Changing the location of individual crops or practices
will often have a greater effect on sediment delivery
than on soil erosion.
175
-------
3). Terraces are more efficient at reducing sediment delivery
than soil erosion, particularly when placed in areas
with high sediment delivery ratios.
Thus in general, there is more flexibility in designing a farm plan
for reducing sediment delivery than in designing a plan for reducing soil
erosion.
New York Example Farm
The New York example farm is a dairy farm with a maximum of 80 cows and
70 crop hectares. Three crops are grown, corn silage and hay to feed the
dairy herd, and corn grain for either feed or sale.
TABLE 8-10. DESCRIPTION OF SAMPLE FARMS
Soil Type
Size
(Ha)
Slope
Gradient
Slope*
Length
(Meters)
Distance
to
Stream
(Meters)
New York: Honeoye - Lima Association
Honeoye- Lima
Lima-Kendaia
Totals
35
35
70
8.0
4.0
250
250
500
250
0
Iowa: Tama - Muscatine Association
Muscatine
Tama
Dinsdale
Garwin
Totals
30
40
15
15
100
1.0
3.5
7.0
1.0
150
200
75
75
500
350
150
75
0
Texas: Houston Black - Heiden Association
Houston Black
Heiden
Trinity
Totals
45
35
20
100
2.0
3.0
0.5
110
90
50
250
140
50
0
— _ _
* Slope length is given for each individual soil, however, soil loss calcula-
tions were made using the total slope length and apportioning soil loss
to individual fields by the method of Foster and Wischmeier (1973) .
Distance from lower edge of soil type to the stream.
176
-------
The base system (no restrictions on soil erosion or sediment delivery)
consists of 35 ha of corn silage on the Lima-Kendaia soil which lies next
to the stream, and 20 ha of corn grain in rotation with 15 ha of hay on the
Honeoye-Lima soils. The base plan also assumes the use of a chisel plow
(conservation tillage). No expected yield difference occurs between
conventional and conservation tillage on these soils, thus conservation
tillage is the most profitable tillage method (see Table 8^14). Farm
plans for obtaining a 50% and a 90% reduction in sediment delivery are shown
in Figure 8-1. Levels of sediment delivery and soil erosion for the base
system and each farm plan are given in Table 8-11. Note that due to re-
deposition, the levels of soil erosion shown in Table 8-11 do not necessarily
indicate the net loss of soil from either field. Net losses of soil for
this example farm are discussed in a later section, tinder the base system,
levels of soil erosion and sediment delivery are extremely high on the lower
field (Lima-Kendaia soil). This field lies on the lower half of a rather
long slope and receives considerable runoff from above. Combined with a
cropping pattern of continuous corn silage, high levels of soil erosion
result. Since this field also lies next to the stream, it has a high SDR.
Thus the Lima-Kendaia soil is the most critical area on this farm for con-
trolling sediment delivery and soil erosion.
To achieve a 50% reduction in sediment delivery (Figure 8-1) all corn
grain production is switched to the lower field (Lima-Kendaia soil). Corn
grain has considerably lower levels of soil erosion than corn silage. A
small area on this soil remains in corn silage, however, it is contoured
and lies below a diversion ditch. The diversion ditch reduces the effective
slope length and thus reduces levels of soil erosion on this area. Since
the diversion ditch increases the SDR on the upper field (Honeoye-Lima), the
area immediately above the ditch is strip-cropped to minimize soil erosion.
Note that if the contoured corn silage in the upper field were above the
diversion ditch, instead of strip cropping, sediment delivery would be
increased, but neither soil erosion nor farm income would be changed.
A 90% reduction.in sediment delivery is achieved by adding a simple ter-
race system and switching to no-tillage for corn grain production. Since
almost all crop residue is removed with corn silage, no-tillage has little
effect on reducing soil erosion and thus is not used on this crop. Crop
location is again a key to obtaining this level of reduction. Continuous
contoured com silage, the most erosive crop in this plan, is grown above
a terrace, which acts to trap sediment, and below either a terrace or a
diversion ditch, which reduce the slope length. Most of the streambank is
protected by a terrace. Where terraces along the streambank are not built,
strip cropping with no-tillage corn grain is implemented. This is the
least erosive cropping system available and thus minimizes soil erosion in
this area which has a high SDR.
On the upper field, contoured no-tillage corn grain is placed above the
diversion. This is a less erosive system than strip cropping with silage,
which is thus grown above the terrace.
The key to the location of all these practices is an attempt to place
the most highly erosive crops in areas with the lowest SDR, and the least
177
-------
-J
00
~: :::::::::::::::::::::::::::::
j:::: I::::::::::::::
"" •• "«:::::::: ""::::::":::
Sediment Delivery, 50$, Reduction
Sediment Delivery, 90% Reduction
Terrace, all land used
Planted on Contour
Diversion, Poor Quality Hay
Corn Silage, on Contour
i°o «"<>„« Corn Grain
* *> O A i
•.$.&.'"£.•': Corn Grain no till, on Contour
ZZ~-- Creek
^^~~ — Soil Type Boundary
FIGURE 8-1. FARM PLANS FOR REDUCING SEDIMENT DELIVERY, NEW YORK.
-------
erosive crops in areas with higher SDRs. Since this generally has little
effect on soil erosion, it is one of the major differences between plans
for reducing sediment delivery and plans for reducing soil erosion. A second
option used in the sediment delivery plans is the placing of a terrace along
the streambank. This also has a minimal effect on soil erosion. Thus, farm
plans for reducing soil erosion must rely on other practices which are often
more expensive. This is particularly true at high levels of reduction.
Table 8-11 shows that obtaining a 90% reduction in soil erosion costs about
four times as much as obtaining an equivalent percentage reduction in sediment
delivery. The major expense in the farm plan for a 90% soil erosion reduction
is a system of eight diversion ditches running the full length of the farm
(parallel to the stream) at 60-m intervals.
There is not a great difference in the farm plans for reducing soil
erosion and sediment delivery at a reduction level of 50%. The diversion
ditch is about 50% longer for the soil erosion plan and most grain is grown
by the no-tillage method. These differences account for the slightly greater
cost associated with the soil erosion plan (Table 8-11).
Iowa Example Farm
The Iowa example farm is a 100-hectare hog-cash crop farm. Corn grain
could be grown either to feed hogs or to sell while soybeans, oats, and hay
were grown only as cash crops. Soybeans could be grown only in rotation
with'corn. As in the New York case, conservation tillage was the most pro-
fitable tillage system and thus was used in the base system (see Table 8rl4) .
Two options available in New York were not used in Iowa. Yield losses of
around 5% are expected with no-tillage, thus this practice is not common
TABLE 8-11. NEW YORK: FARM PLANS FOR REDUCING
SEDIMENT'DELIVERY AND SOIL EROSION
Base
System
Sediment Delivery (MT/ha)
Honeoye-Lima 5.1
Lima-Kendaia 47.1
Farm Total 26.1
Soil Erosion (MT/ha)
Honeoye—Lima
Lima-Kendaia
Farm Total
Cost ($/ha)
16.9
60.2
38.5
Plans for Sediment Delivery
50% 90%
Reduction Reduction
8.5
17.5
13.0
24.8
22.3
23.6
$5.13
2.7
2.3
2.5
7.4
11.7
9.6
$13.01
Plans for Soil Erosion
50% 90%
Reduction Reduction
7.1
14.9
11.0
19.0
19.0
19.0
$7.21
3.4
4.0
3.7
4.0
4.0
4.0
$52.77
179
-------
and was not included. Diversion ditches are also seldom used in this area
of Iowa and were not considered. Both fall and spring tillage were considered.
It is fairly difficult to accurately determine the economic costs of spring
tillage, although they certainly exist. To account for these costs, an
arbitrary additional labor charge of $0.50/ha was assigned to all spring
tillage work. The farm plans for Iowa shown in Figure 8-2 do not distinguish
between fall and spring tillage. However, as restrictions on sediment deli-
very become greater, more tillage is shifted to the spring. For the base
plan, over 90% of tillage is in the fall, at a 50% reduction, nearly one
half of tillage work is in the spring, and at 90% reduction level, about
60% of tillage occurs in the spring.
The base system in Iowa consisted of 20 ha of continuous corn grain and
80 ha of the corn-soybeans rotation. The continuous corn is grown on the
Tama soil, while the more erosive cropping pattern, corn-soybeans is grown
on the other soils, including the Dinsdale and Garwin which lie closest to
the stream. Levels of sediment delivery and soil erosion for the base system
and the farm plans are shown in Table 8-12.
The farm plan for reducing sediment delivery by 50% utilizes a two-
terrace system covering about one third of the farm. One terrace runs along
the lower-edge of the Muscatine soil, the other along the lower edge of the
Tama soil. A single terrace placed on a field reduces the SDR for the area
above the terrace to 0.2 but does not affect the level of soil erosion on
that area (except for the contouring effect). Conversely, the terrace reduces
soil erosion on the area below the terrace by reducing the runoff from above,
but does not affect the SDR on this area. Thus levels of soil erosion on
the areas of Dinsdale and Garwin below the terrace system are reduced. Since
the Dinsdale soil has the highest levels of sediment delivery in the base
plan, it is the most critical area to control. The area of Dinsdale below
the terrace remains in a corn-soybeans rotation but is contoured and spring-
plowed. The area of Dinsdale not below the terrace is switched from fall-
plowed corn-soybeans to spring-plowed continuous corn, reducing soil erosion
levels in this area. The total area of continuous corn is increased to
about 24 ha with this plan.
The plan for a 90% reduction in sediment delivery has the same terrace
system on about 50% of the farm. On the remaining 50% a two-terrace system
is implemented with one terrace across the Tama soil and one terrace at the
foot of the Dinsdale soil. This latter system is more efficient at reducing
sediment delivery than the former terrace system, but is more expensive to
maintain due to the longer slope lengths involved.
All of the Muscatine and Tama soils are in a corn-soybeans rotation. The
Dinsdale soil is completely in continuous corn while the area of Garwin below
the first terrace system is in continuous corn. These areas have the highest
potential for soil erosion and sediment delivery. The total area of continu-
ous corn remains at 24 hectares for this plan, with the remainder in the corn-
soybeans rotation.
The farm plan for a 50% soil erosion reduction uses the same terrace
system as the plan for reducing sediment delivery by 50%. However, in order
180
-------
oo
>J^*&'"
Sediment Delivery, 50% Reduction
Sediment Delivery, 90% Reduction
Terrace
Corn, Soybeans
Corn, Soybeans, on
Contour
Corn
jjjjjjjjjjijjjji Corn> on Contour
-^_— -- Creek
— " Soil Type Boundary
FIGURE 8-2. FARM PLANS FOR REDUCING SEDIMENT DELIVERY, IOWA*
-------
to reduce soil erosion on the Dinsdale soil, this area is completely in a
corn-oats-hay rotation. This results in the increased cost over the sediment
plan.
To obtain a 90% reduction in soil erosion, over 70% of the farm has a
system of four terraces and nearly 25% is in a corn-oats-meadow rotation.
Thus costs are greatly increased over the plan for reducing sediment delivery
by 90% (Table 8-12).
If farm plans are designed to reduce soil erosion, crop placement may
be highly unfavorable for controlling sediment delivery. The Iowa plan for a
90% reduction in soil erosion is a good example of this. All of the Garwin
soil, which lies next to the stream and has an SDR of 1.0, is in the highly
erosive corn-soybeans rotation, while most of the hay rotation is in areas
directly above terraces, which have a low SDR.
Although Table 8-12 clearly shows that for the Iowa example farm, it
is less costly to achieve reductions in sediment delivery than to achieve
percentage equivalent reductions in soil erosion, it also shows that each
soil erosion plan reduces sediment delivery proportionally more than it re-
duces soil erosion. When soil erosion is reduced by 50%, sediment delivery
actually decreases by 74% (to 5.0 MT/ha). The cost is $29.56/ha. However,
TABLE 8-12. IOWA: FARM PLANS FOR REDUCING
SEDIMENT DELIVERY AND SOIL EROSION
Plans for Sediment Delivery Plans for Soil Erosion
Base 50% 90% 50% 90%
Plan Reduction Reduction Reduction Reduction
Sediment Delivery (MT/ha)
Muscatine 1.5 0.8 0.6 0.9 0.6
Tama 8.5 5.8 1.7 5.6 6.6
Dinsdale 68.3 27.5 2.9 8.1 1.4
Garwin 9.1 6.9 2.1 7.5 1.4
TOTAL FARM 15.5 7.8 1.5 5.0 1.1
Soil Erosion (MT/ha)
Muscatine 5.0 3.0 3.0 4.4* 3.0
Tama 20.8 15.9 8.4 15.0 3.0
Dinsdale 128.9 51.1 10.3 15.0 3.0
Garwin 9.1 6.9 2.9 7.5 3.0
TOTAL FARM 30.5 15.9 6.3 10.7* 3.0
Cost ($/ha) $13.13 $42.40 $29.56- $93.01
* Although soil erosion was only restricted to 15 MT/ha, practices required
to reduce the Tama and Dinsdale soils to this level caused erosion levels
to be considerably less on the Muscatine and Garwin soils.
182
-------
a farm plan designed to reduce sediment delivery by an equivalent amount
(to 5.0 MT/ha) costs only $22.14/ha or 75% of the cost of the soil erosion
plan. Likewise reducing soil erosion by 90% reduces sediment delivery to
1.1'MT/ha and costs $93/ha. This level of sediment delivery can be achieved
through a sediment delivery plan at a cost of only $48 or about 53% of the cost
of the soil erosion plan.
Texas Example Farm
The example farm for Texas is cash crop-hay operation consisting of
100 ha of tillable land plus an unspecified but fixed amount of additional
permanent pasture ground assumed to be unsuitable for crops. Since the
market for hay is fairly limited in this area of Texas, the area of hay
was fixed for all plans at about 19% of total crop acres (19 ha). Other
crops considered in this farm were cotton, wheat, and sorghum.
Methods for maintaining cotton yields under either conservation tillage
or no-tillage have not been fully developed. Also, the soils in this area
are predominately fine-textured clays, which decreases their suitability
for these reduced tillage practices. Thus neither conservation tillage nor
no-tillage was considered as an option for this farm.
Although no acreage restrictions exist for the crops grown in this area,
cotton, wheat, and sorghum have a base acreage which is established by ASCS
for each farm. If a farm exceeds the base, it is not eligible for price
supports or government commodity loans. Since many farmers participate in
the program significant areas of sorghum and wheat are grown even though
cotton provides the highest net return per hectare. These base allotments
vary from farm to farm and were not considered on the example farm. Although
this has the effect of distorting the actual levels of income, soil erosion,
and sediment delivery, it does not have a large effect on the evaluation
of SWCPs, particularly terraces, for the example farm.
In this area of Texas, the topography is typically complex with few
long slope lengths. The total slope length for all three soils combined is
only 250 m , thus the total farm width is 500 m. Therefore the 100 ha
example farm is 2,000 m long. Since these dimensions are not conducive
to illustrating farm plans, the plans are not drawn to scale, rather the
farm is shown with a proportionately expanded width.
Under the base system, 81 ha of cotton and 19 ha of hay are grown, with
all the hay being produced on the Trinity soil. Although this soil has the
lowest slope gradient (0.5%), the other soils are relatively more efficient
in producing cotton than hay.
A 50% reduction in sediment delivery is obtained by using a two terrace
system which covers about 40% of the farm (Figure 8-3). One terrace is next
to the stream, the other approximately halfway up the slope at the lower
edge of the Houston-Black soil. All of the terraced area is in contoured
cotton. Hay production is moved to the Heiden soil, the most erosive area
of the farm.
183
-------
CD
Sediment Delivery, 50% Reduction
— Terrace
..- Hay
IVheat, on contour
w Cotton, IVheat, Wheat,
s& on contour
iHouston Black-
- Trinity-
--,Trinity
iHeiden:
ipuston Black:
Sediment Delivery, 90% Reduction
Cotton
jjjHHgjjgjj: Cotton, on contour
~--~ Creek
_ — Soil Type Boundary
FIGURE 8-3. FARM PLANS FOR REDUCING SEDIMENT DELIVERY, TEXAS.
-------
This simple two-terrace system was not adequate to obtain a 901 reduct-
ion in sediment delivery, thus a system with shorter terrace intervals is
required, having 12 terraces across the farm, six on each side of the stream
(Figure 8-3). None of the terrace options available for this farm had both
reduced terrace intervals and a terrace along the streambank. Therefore
to obtain the 90% reduction in sediment delivery, hay production was shifted
to the Trinity soil to minimize soil erosion in the area having the highest
SDR. Even with the more intense terrace system, cotton production is
severely cut back with this plan. Nearly 40 ha of wheat are grown replacing
acreage which was in cotton on both the base plan and the 50% reduction plan.
A 50% reduction in soil erosion is obtained by a single terrace system
over nearly half the farm. As in the sediment delivery plan, hay is shifted
to the Heiden soil at the 50% reduction level. The additional cost of this
plan over the plan for reducing sediment delivery by 50% is primarily due to
contouring on all the Houston-Black soil.
To achieve a 90% reduction in soil erosion, a terrace system covers
about 90 ha , with the remaining 10 ha removed from crop production. The
total cotton area is reduced to about 10 ha. These latter two factors account
Qor -the substantially greater cost of this plan over all the other plans for
the Texas farm.
TABLE 8-13. TEXAS: FARM PLANS FOR REDUCING
SEDIMENT DELIVERY AND SOIL EROSION
Farm Plans For:
Base
Plan
50% Reduct-
ion in
Sediment
Delivery
90% Reduct-
ion in
Sediment
Delivery
50% Reduct-
ion in
Soil
Erosion
90% Reduct-
ion in
Soil
Erosion
Sediment Delivery (MT/ha)
Houston-Black 15.2 11.1
Heiden 40.5 5.4
Trinity 4.2 20.6
TOTAL FARM 21.8 11.0
Soil Erosion (MT/ha)
Houston-Black 40.6 34.1
Heiden 83.5 19.8
Trinity 4.2 25.4
TOTAL FARM 48.3 27.4
Cost ($/ha) $10.19
3.0
1.1
1.2
2.0
15.2
6.7
3.6
9.9
$52.81
7.5
11.4
24.0
12.2
24.0
24.0
24.0
24.0
$13.34
1.0
1.0
5.0
1.8
5.0
5.0
5.0
5.0
$120.02
185
-------
Further Discussion of Farm Plans on the Example Farms
Thus far, primary exmphasis has been on the types of differences that
occur between plans oriented towards reducing sediment delivery versus those
oriented towards reducing soil erosion. Some practices are equally effective
for both types of farm plans. In particular, the four practices listed in
the previous section as having the highest potential levels of cost-effective-
ness occur in the farm plans for each farm on which they were considered
applicable. Contouring was the most extensively used practice occuring on
all plans on all the farms. Conservation tillage was not applicable in Texas,
but was the primary tillage practice in both Iowa and New York. No-tillage
was only considered on the New York farm and was the primary practice for
growing corn grain on all the farm plans except the plan for reducing sedi-
ment loss by 50%. Strip cropping was also used in all farm plans where sod-
based rotations were part of the cropping system (only continuous hay was
considered in Texas).
On the other hand, sod-based rotations were shown to have a fairly low
level of cost-effectiveness. On the New York farm, a certain amount of hay
was required to feed the dairy herd, and was part of the base system. However,
no significant increase in hay area occurred for any of the farm plans
investigated. In Iowa, small areas of rotation hay occurred at the 90%
reduction level. However, this was one of the major factors in creating the
greatly increased costs at this level of reduction. No variance in hay area
was allowed on the Texas farm, thus this practice was not applicable.
In general, the differences in the types of practices used on the three
farms is more dependent on the particular soils and type of farming operation
than on the area of the country. For example, strip cropping was used only
on the New York example farm. However, farms with cattle enterprises in Texas,
Iowa, or other areas would normally grow hay crops, and could readily employ
strip cropping. Likewise, conservation tillage could be used in many areas
of Texas, even though it was not used on the Texas example farm. Thus,
evaluation of practices generally must be done on an individual farm basis
rather than on a regional basis.
Conservation Tillage versus Conventional Tillage on the Example Farms
As previously noted, conservation tillage was assumed to be the normal
tillage practice on the Iowa and New York example farms. These practices
are compared for the base plan on these two farms in Table 8-14.
For both farms, income, (above selected variable costs) is reduced by
about $5/ha when moldboard plowing is used instead of chisel plowing. In
both cases, the two tillage systems were assumed to produce equal yields,
thus the difference in income is due only to differences in field time and
machinery maintainence costs.
Levels of soil erosion and sediment delivery with conventional tillage
on the Iowa farm are nearly double the levels with conservation tillage. In
New York, however, conventional tillage increases soil erosion by only about
30% and sediment delivery by about 16%. With the base plan in New York, all
186
-------
the Lima-Kendaia soil is in continuous corn silage. There is little dif-
erence « 4%) in levels of soil erosion between conservation and conventional
tillage on crops where crop residues are removed, such as corn silage. Thus
the switch in tillage practices has a small effect on soil erosion levels.
Since the Lima-Kendaia soil has the higher SDR of the two soils on the farm,
the differences in sediment delivery are even less than the differences in
soil erosion.
Cost Associated with Reducing Sediment De1ivery
There are several methods of looking at costs of soil and water conser-
vation practices. The average cost per hectare is a useful method of
measuring the potential effect on farm income. This is the cost shown in
Tables 8-11, 8-12, and 8-13. However, average cost per hectare does not
necessarily relate costs to a particular goal. If the goal is to reduce
the amount of sediment reaching a stream, then a more meaningful cost is the
cost per unit of reduction in sediment delivery. This would be particularly
true if it were also possible to quantify the benefits of reducing sediment
delivery. In theory, the optimal level of reduction in sediment delivery
occurs where the marginal cost of reducing sediment delivery equals the
marginal benefit. At any given level of reduction in sediment delivery,
marginal cost is defined as the additional cost of keeping one additional
metric ton of sediment out of a waterway, while marginal benefit is the
additional benefit which would be received by restraining that one additional
ton. No attempt was made in this study to estimate water quality benefits
TABLE 8-14. CONSERVATION TILLAGE* VERSUS
CONVENTIONAL TILLAGE* ON THE IOWA AND NEW YORK EXAMPLE FARMS
Income above Se- Soil Sediment
lected Variable Erosion Delivery
Base Farm costs** MT/ha MT/ha
Plan $/ha
Iowa:
Conservation 705 30.5 15.5
Conventional 699 61.8 31.1
New York:
Conservation 1083 38.5 26.1
Conventional 1078 50.3 30.4
* Conservation tillage is chisel plowing with one secondary tillage operation.
Conventional tillage is moldboard plowing followed by two secondary tillage
operations.
** Income is total farm income and thus includes income from the hog enter-
prise in Iowa and the dairy enterprise in New York.
187
-------
of reduced sediment delivery to streams. One should keep in mind that while
marginal costs of reducing sediment delivery increase as sediment delivery
is restricted, marginal benefits are likely concurrently to decrease.
For each of the example farms, marginal cost and cost-effectiveness curves
were plotted. The curves on all the farms had a common pattern. At low
levels of reduction in sediment delivery, cost-effectiveness values were high,
and up to a point, dropped rather slowly as levels of reduction increased.
On each farm, however, there was a point where cost-effectivness began to drop
sharply. This occurred between the 70% and 90% levels of reduction. The
marginal cost curve had a reverse shape, starting low, increasing gradually
at first up to a certain point where it suddenly rose rapidly. Marginal
costs also tend to rise in a step-wise manner. As sediment delivery is re-
duced, marginal cost increases in jumps with fairly level areas in between.
Comparing these points of rapid increase with the farm plans, it was seen
that they occurred with the introduction of major new practices. After a new
practice is introduced on part of the farm, marginal cost remains nearly
level until that practice is used on the whole farm, then a rapid jump occurs
with the introduction of a new more intensive practice. This pattern can be
clearly seen in the cost curves for Texas, shown in Figure 8-r4. The marginal
cost curve is at first nearly level with a slight jump at a 55% reduction. It
rises rapidly above the 75% reduction level. Reductions below the 75% level
are achieved with non-structural practices, including contouring and crop
placement combined with a simple terrace system. This system is shown in
the farm plan for a 50% reduction in Figure 8-3. At the 75% reduction level,
the whole farm is in this simple terrace system; thus to achieve greater
reductions, new, less cost-effective practices must be introduced. In this
case three things happen together which have the combined effect of rapidly
increasing marginal cost. They are introduction of a more intense terrace
system, substituting wheat for cotton, and taking some land out of production.
The existence of these points where marginal costs suddenly rise rapidly
is partly a function of the number and type of practices available on a farm.
For the New York farm, for example, which had more options available, the
point of rapid rise occurred at a higher level (about 85%) and was somewhat
less abrupt. It is clear that to the extent that critical point exists
for a farm, maximum reductions per unit of cost will occur if levels of re-
duction are below this point. Thus one general conclusion is that it is
more cost-effective to achieve moderate reductions in sediment delivery
over several farms than to attempt large reductions on only a few farms, pro-
viding all farms have roughly equivalent initial levels of sediment delivery.
The Relationship between Soil Erosion and Conservation
The farm plans for controlling either soil erosion or sediment delivery
discussed earlier did not take into account the long run impact of reduced
soil loss on crop yields. Estimation of such long run yield effects has been
attempted by some researchers by relating yields to topsoil depth (e.g., Seitz
et al_* 1978). Future topsoil depth has been estimated by substracting
erosion calculated from the USLE from initial topsoil. This probably over-
estimates the reduction in average topsoil depth in a field because no ad-
justment is made for topsoil redeposited either within the same field or on a
188
-------
field between the field in question and the stream. The level of knowledge
in this area is inadequate to make definite statements about the effects of
redeposition on topsoil depth and future crop yields. As stated earlier, there
are no generally accepted methods of calculating sediment delivery ratios
There is less agreement on where the sediment not delivered to stream is
redeposited. This section attempts to shed some light on the subject by use
of examples based on the New York, Iowa and Texas farms.
I 4.00 r
2.00
0.00
8.00
-te-
in
O
O
- 6.00
o
c
o>
o
5
4.00
2.00
I.OOh
Cost Effectiveness
Marginal cost
i
i.40
1.20
1.00 o
o
.80
.60
.40
.20
.1 0
m
o>
3
(/I
20 40 60 80 100
% Reduction in Sediment Delivery
FIGURE s-4. COST-EFFECTIVENESS AND MARGINAL COST FOR REDUCING SEDIMENT
DELIVERY ON THE TEXAS EXAMPLE FARM
189
-------
Soil erosion, as defined in Section 4, is the detachment and transport
of soil by water. Soil erosion levels given for the farm plans for the
example farms (Tables 8-11, 8-12, and 8-13) represent the expected total soil
erosion on those farms over a year as calculated from the USLE. When, as
is true for most fields, the SDR is less than 1, soil is redeposited and at
least part of this redeposition is within cropped fields. This could be in
depressions within the field or on fields over which runoff flows before it
reaches the stream.
On the example farms, all slopes were assumed to be convex on those soil
types away from the stream and concave on the soils adjacent to the stream.
If it is assumed that all redeposition occurs on the concave areas, then
net soil loss from these areas is considerably less than the soil erosion
levels shown in Tables 8-11, 8-12, and 8-13. In Table 8^15, net soil loss
for each soil type is calculated for the base plan for each example farm.
This calculation is based on the assumption that all eroded soil which does
not leave the farm as sediment delivery is redeposited on the soil type which
lies next to the stream. Thus net soil loss from the farm is the same as
the level of sediment delivery. This would not be true, however, if terraces
or other types of sediment traps are on the farm, since, in this case, some
of the redeposition of sediment occurs in these structures instead of on the
lower fields.
TABLE 8-15. SOIL EROSION AND NET SOIL LOSS
FROM THE EXAMPLE FARMS UNDER THE BASE PLANS (Ml/ha)
New York
Soil Erosion
Net Soil Loss
Iowa
Soil Erosion
Net Soil Loss
Texas
Honeoye-Lima
16.9
16.9
Muscatine
5.0
5.0
Houston-Black
Tama
20.8
19.6
Lima-Kendaia
60.2
25.3
Dinsdale
128.9
68.3
Garwin
9.1
-27.2
Total Farm
38.5
26.1
Total Farm
30.5
15.5
Heiden
Trinity Total Farm
Soil Erosion
Net Soil Loss
40.6
40.6
83.5
83.5
4.2
-128.5
48.3
21.8
Net soil loss is considerably less than USLE soil erosion on the Lima-
Kendaia, Garwin, and Trinity soils. On the latter two, redeposition actually
exceeds the level of soil erosion, thus net soil loss is negative.
The net soil loss levels shown in Table 8-15 are based on the calculated
SDRs and thus may not be precise. However, they do show that levels of gross
190
-------
soil erosion, particularly on complex slopes, are not necessarily the best
indicator for the conservation of topsoil and thus for estimating long run
yield effects of reducing soil erosion.
Cost-Sharing
Cost-sharing with farmers is one method of encouraging adoption of SWCPs.
The Agricultural Stabilization and Conservation Service (ASCS) has a continu-
ing program of cost-sharing with farmers for adoption of practices principally
designed for preventing soil erosion. Only practices which meet fairly exact
standards of both ASCS and the Soil Conservation Service (SCS) are cost-shared.
Recently, money has been authorized for cost-sharing with farmers on practices
specifically aimed at reducing non-point source pollution from agricultural
land.1 Since there can be a significant difference in the effectiveness of
SWCPs for reducing sediment load compared to the effectiveness for reducing
soil erosion, it would be expected that the structure of the new cost-
sharing program would be somewhat different from the traditional ASCS program.
The goal of any cost-sharing program should be to encourage farmers to
meet the program's objectives at the least cost to both the farmer and the
cost-sharing agency. If cost-sharing is not available on the least costly
system of practices or it is not proportional to the actual costs of
practices to the farmers, farmers may adopt practices which are more costly
and/or less effective for reaching the water-quality goals. However, because of
monitoring problems and variations between farms, it may be difficult to cost-
share on certain practices, particularly those which are non-structural,
require annual on-site checks to ensure compliance, and/or do not have
specific costs associated with their implementation. Practices such as
contouring, minimum tillage, no-tillage, and crop rotations would generally
fall into this category. Also, for practices which influence farm costs
primarily by their effect on yield (primarily no-till and conservation tillage),
it may not be possible to determine a level of cost-sharing which is actually
proportional to farm costs, since these can vary from field to field, and in
some cases, are insignificant or even negative.
All of the farm plans presented in the previous section were derived
assuming no cost-sharing funds were available. However, cost-sharing funds
generally are available for many practices through the Agricultural Stabiliza-
tion and Conservation Service (ASCS). Practices which are eligible for
cost-sharing must meet the criteria of both the ASCS and SCS. To determine
how present cost-sharing programs affect farm plans for controlling sediment
delivery, the LP model for each example farm was rerun with current cost
share levels included for eligible practices. Levels of cost-sharing were
taken from current ASCS programs for a representative county in each state.
The counties were Tompkins County, New York; Grundy County, Iowa; and
Collin County, Texas.
Cost-sharing under the ASCS program generally falls into three cate-
gories: a fixed percentage of initial installation costs for structures,
Section 35. Clean Water Act of 1977.
191
-------
a fixed percentage of establishment costs for sod crops, and a single per
hectare payment for other eligible practices such as strip cropping,
minimum tillage and/or contouring. Eligible practices in the third category-
vary significantly from area to area and from year to year, and with the
possible exception of strip cropping, are not available at all in most areas.
Cost-sharing rates for the three counties are given in Table 8-16.
Although cost-sharing is currently available at 90% for structural
practices in Tompkins County, New York, for most years and for most counties
in the state, rates of 60%-75% would be more typical. Still, rates for New
York are higher than for the other areas.
For both Texas and Iowa, availability of cost-sharing funds had no
significant effect on practices used in farm plans for sediment reduction.
The terrace system with conventional (SCS) spacing was assumed to be the
only system eligible for cost-sharing. For both Texas and Iowa, this system
had a low level of cost-effectiveness and did not occur in farm plans except
at extremely high levels of restraint. Although cost-sharing caused a
slight increase in the areas with this terrace system, it had no significant
overall effect.
Cost-sharing could potentially have brought significantly more hectares
of hay into the Texas farm plans, if hay had not been restrained to a maximum
of 19% of total area. In Iowa, however, adding hay area had a low cost-
effectiveness, which was not significantly affected by cost-sharing on hay
TABLE 8-16. COST-SHARING RATES FOR
ELIGIBLE SWCPS ON THE EXAMPLE FARM
*,#
Terrace and Diversion Systems
Conservation Tillage
e
Strip Cropping
Permanent Sod Cover
New York
90%
--
$75
75%
Iowa
60%
$12
$10
60%
Texas
50%
—
—
50%
* Percent of initial construction cost
#
Cost-sharing for terraces available only for terrace intervals meeting
SCS specifications
P
* One-time per hectare payment for initial implementation
* Percent of establishment cost
192
-------
establishment. Even though cost-sharing did not affect optimal farm plans
on these two example farms, there was no significant amount of cost-sharing
money available for practices in the farm plans with less than about an 80%
reduction in sediment delivery. Thus the current cost-sharing program would
not offer a significant incentive to farmers where only moderate decreases
in sediment delivery were desired.
For the New York example farm, cost-sharing played a more significant
role. More practices are eligible for cost-sharing, and cost-sharing rates
are higher. Thus availability of cost-sharing money changed the optimal
farm plans at all levels of reduction, primarily by causing substitutions
of more extensive terraces and diversions for contouring and no-till to
occur. Although cost-sharing results in lower farm costs, the total cost
is increased. Table 8-17 shows that total cost, which includes actual
farm cost plus cost-sharing funds, nearly doubles when cost-sharing is avail-
able at the 90% reduction level and increases by about 20% at the 50% level.
The cost increases with cost-sharing are due primarily to a lack of
funding for certain practices which may be highly cost-effective. These would
include practices such as conservation tillage, no-till, contouring, and
terrace systems with wider intervals than provided for by current specifica-
tions. Providing cost-sharing for non-structural practices may require some
sort of annual monitoring. This would certainly have the effect cjf in-
creasing costs for the responsible agency which would partially offset the
benefits of these practices. Further analysis in this area would be required
to fully evaluate these alternatives.
Technical evaluation of the feasibility of the wider terrace intervals
used on the example farms was not attempted. Thus in some cases, these ter-
race systems may not be possible since runoff and/or sedimentation may exceed
their capacity. This would be particularly true if continuous highly erosive,
row crops are grown. However, the possibility of combining wider terrace
spacings with other practices such as conservation tillage, no-till or strip-
cropping would certainly be feasible under many conditions and would often
be more cost-effective than a more intense terrace system.
TABLE 8-17, EFFECT OF COST-SHARING ON COSTS
FOR REDUCING SEDIMENT DELIVERY, NEW YORK ($/haJ
% Reduction
50% 90%
with without with without
Cost-sharing Cost-sharing Cost-sharing Cost-sharing
Farm Cost 1.00 5.10 19.70 27.40
Cost-share 5.20 —- 52-40
Total Cost 6.20 5.10 52.10 27.40
193
-------
Potential Effects of Farm Plans for Reducing Sediment Delivery on
Losses of Other Pollutants
Farm plans for reducing sediment delivery may not have similar effects
on the losses of other potential pollutants. In this section, results from
Sections 4, 6, and 7 are used to evaluate some possible effects of farm plans
for reducing sediment delivery on the losses of nutrients and pesticides.
First, results from Section 4 are used to describe the types of expected
changes that would occur with the farm plans. Then the CMS model is used to
predict nutrient losses for the Iowa example farm. Finally the CPM model
is used to predict losses of atrazine and several alternative herbicides
for continuous corn with several SWCPs on the Iowa farm.
Tables 4-1 and 4-2 list pollutants in categories according to their
degree of adsorption to soil particles. Table 4-4 lists individual SWCPs
as "Effective", "Slightly-effective", or "Not effective" in controlling
pollutants in each of the adsorption categories. Table 4-4 is used as the
basis for much of the following discussion. "Strongly adsorbed" substances
move primarily with sediment and are concentrated on the fine particles.
Thus in general, the effectiveness of a practice for controlling strongly
adsorbed substances will be highly correlated to its effectiveness in con-
trolling sediment delivery. However, practices, such as terraces, which
achieve their effect primarily by providing for redeposition of eroded
sediment will be relatively less effective than practices which are effective
due to reducing splash energy or increasing infiltration. Thus terraces will
reduce losses proportionately less than they reduce sediment delivery.
The non-structural practices, however, generally increase surface cover
and thus reduce splash erosion. For these practices, particularly no-till,
conservation tillage, and sod-based rotations, reductions in strongly
adsorbed substances should be nearly proportional to reductions in sediment
delivery. In general, farm plans designed to reduce sediment delivery
should be effective in reducing losses of strongly adsorbed substances, but
this reduction will usually be less than proportional to the reduction in
sediment delivery.
"Moderately adsorbed" substances move primarily with overland flow, thus
the effectiveness of SWCPs in reducing losses depends primarily on reducing
surface runoff. Sod-based rotations is the only practice evaluated in this
chapter which is listed as effective in Table 4-4. However, sod-based re-
ductions were found to generally have a low level of cost-effectiveness for
reducing sediment delivery and did not occur (as an increase over the base
system) extensively in any of the farm plans. Contour fanning, conservation
tillage, no-tillage, and terrace did occur commonly in the farm plans and
were found to have potentially high levels of cost-effectiveness. These four
practices are listed as slightly effective in controlling moderately adsorbed
substances. Thus farm plans for reducing sediment delivery should reduce
levels of moderately adsorbed substances, but probably considerably less
than the reductions in sediment delivery.
Only contouring and sod-based rotations are listed as slightly effective
in controlling non-adsorbed pollutants, all other practices evaluated in
this chapter are listed as "Not-effective." Thus it would not be expected
194
-------
that farm plans for reducing sediment delivery would have any significant
effect on losses of these substances.
Levels of Nutrient Loss on the Iowa Example Farm
Results from the CNS model (Section 6) were used to estimate losses
of nutrients from the Iowa farm under the base plan and with the farm plans
for achieving a 50% and a 90% reduction in sediment delivery. These results
are presented in Tables 8-18, 8-19, and 8-20.
Solid-phase Nutrients. The change in losses of solid-phase nutrients
as sediment delivery is reduced depends on both the level of sediment delivery
and the enrichment ratio (Section 6). The CNS model assumes the enrichment
is constant and independent of the level of soil erosion or sediment delivery.
TABLE 8-18, LOSSES* OF SOLID-PHASE NUTRIENTS
FOR THE IOWA EXAMPLE FARM
Base 50% Reduction 90% Reduction
Plan in Sediment Delivery in Sediment Delivery
#
Constant Enrichment Ratio
Enrichment Ratio 1.25 1.25 1.25
Nitrogen 9.7 4.8 .97
Phosphorous 13.2 6.6 1.3
% Reduction -- 50% 90%
from Base Plan
Changing Enrichment
Enrichment Ratio
Nitrogen
Phosphorous
% Reduction
from Base Plan
Ratio *
1.25
9.7
13.2
—
1.5
5.8
7.7
40%
2.25
1.7
2.4
82%
Average annual losses in kg/ha.
The losses are based on results from the CNS model (Section 6). However,
this model assumes a constant enrichment ratio of 2.0. Results were
adjusted for an enrichment ratio of 1.25 for the sake of comparison.
From Casler and Jacobs (1975).
195
-------
TABLE 8-19. AVERAGE ANNUAL LOSSES OF DISSOLVED NUTRIENTS
WITH TWO CROP ROTATIONS AND SELECTED SWCPS
FROM THE IOWA MUSCATINE SOIL
Soil
Erosion
(MT/ha)
Corn-Soybeans Rotation
Conservation tillage 7.2
plus Contouring 3.3
plus Terracing 2.0
Dissolved
N in Runoff
(kg/ha)
4.7
3.7
3.2
Dissolved
P in Runoff
(kg/ha)
0.17
0.13
0.10
Lea'ched
Nitrogen
(kg /ha)
31.0
32.2
33.2
Continuous Corn
Conservation tillage
plus Contouring
plus Terracing
% 5.0
2.3
1.5
11.0
8.8
7.2
0.16
0.13
0.11
65.3
68.1
69.8
If this is true, reductions in losses of solid-phase nutrients will be pro-
portional to reductions in sediment delivery. However, if the enrichment
ratio increases as levels of sediment delivery decrease, then reductions in
losses of solid-phase nutrients will be proportionately less than reductions
in sediment delivery. There appears to be considerable debate on the re-
lationship between the enrichment ratio and levels of soil erosion. This
relationship may depend on both the properties of the soil and the types of
SWCPs implemented, and at present, is not known. Casler and Jacobs (1975)
developed a curve relating the enrichment ratio for phosphorous with the level
of soil erosion. This curve was based on research reported by Massey and
Jackson, (1952). Table 8-18 shows levels of losses of solid-phase nutrients
for the Iowa farm using both the constant enrichment ratio as in the CNS
model, and the variable enrichment ratio, developed by Casler and Jacobs.
It appears that reducing sediment delivery will definitely reduce losses
of solid-phase nutrients. However, the amount of reduction depends on the
presently unknown relationship between the enrichment ratio and the levels of
soil erosion and sediment delivery.
Dissolved Nutrients. Losses of dissolved nutrients depend, among
other things on both the quantity of surface runoff water and the amounts
of fertilizer applied. Thus effects of individual practices, where rates of
fertilization do not change may not be the same as the effects of implement-
ing farm plans which change the crop mix and thus rates of fertilizer applica-
tion.
196
-------
Since soybeans require no nitrogen fertilizer, twice as much nitrogen
fertilizer is applied with*continuous corn as with a corn-soybeans rotation.
Using the Muscatine soil from the Iowa example farm, losses of soluble
nutrients for several SWCPs with continuous corn and with a corn-soybeans
rotation are given in Table 8-19.
Total dissolved nitrogen losses with continuous corn are roughly
double the losses with com-rsoybeans while losses of dissolved phosphorous
are about the same under either cropping system. Losses of both nitrogen
and phosphorous in runoff are decreased by both contouring and terracing,
but the reductions are proportionately less than the reductions in soil
erosion. Also levels of leached nitrogen increase as dissolved nitrogen
.losses decrease such that there is little change in the total dissolved nit-
rogen leaving the field.
Table 8-20 shows the edge-of-field losses of dissolved nutrients for
the Iowa farm plans for reducing sediment delivery. It should be noted
that these are edge-of-field losses and not estimates of stream loading. The
transport of these dissolved substances was not investigated. Note also
that dissolved nutrient losses are not reduced by the same percentage as the
reductions in sediment delivery.
The combinations of practices used in the Iowa farm plans tend to slightly
increase nitrogen losses and to decrease losses of phosphorous by small
amounts. The magnitude of the changes, however, is within the expected
accuracy of the CNS model, thus the changes are not considered to be signifi-
cant. In particular, the model does not include the effects of phosphorous
TABLE 8-20. EDGE-OF-FIELD LOSSES* OF
DISSOLVED NUTRIENTS FOR THE IOWA EXAMPLE FARM
FARM PLANS~~ —-
Base 50% Reduction in 90% Reduction in
Plan Sediment Delivery Sediment Delivery
Dissolved Inorganic
Nitrogen in Runoff 6.9 6.5 7.2
Dissolved Phosphorous
in Runoff 0.20 0.18 0.15
Leached Inorganic
Nitrogen 37.7 39.4 41.1
* Average annual losses kg/ha
leaching from crop residues on losses of soluble phosphorous. As levels of
sediment delivery are decreased on the Iowa farm, there is a shift from pre-
dominately fall tillage to predominately spring tillage. Thus more crop
197
-------
residues are left on the field over the winter and early spring months as
levels of sediment delivery are reduced. This may have the potential for
raising the losses of dissolved phosphorous.
The increase in levels of nitrogen losses as sediment delivery is re-
duced is primarily due to an increase in corn grain acreage relative to
soybean acreage. For the situation of the Iowa example farm, the change
in levels of, dissolved nitrogen losses (runoff and leached) are small since
the change in corn acreage is fairly small (about 10%). However, if there
were a large shift from soybeans to corn on a farm, losses of dissolved
nitrogen could increase significantly.
Evaluation of Alternative Methods for Reducing Herbicide Losses
from Continuous Corn
In Section 7, the effect of individual SWCPs on losses of pesticides
was investigated using the CPM model. One of the conclusions from that
section was that sizeable reductions in losses of pesticides in runoff may
often be achieved with SWCPs. However, in situations where short-lived
pesticides may be substituted for more persistent pesticides, losses in run-
off may also be reduced, and in some situations, these reductions will be
greater than these achieved with SWCPs. In Section 9, other methods of
reducing pesticide losses from cropland are discussed. These include use of
alternative application methods, use of alternative pesticides, and inte-
grated pest management practices.
Only a few of the techniques mentioned in Sections 7 and 9 may be
applicable in any given cropping situation. For example, few integrated pest
management practices are currently available for reducing herbicide use on
corn. However, other methods of reducing losses of herbicides do exist.
These include SWCPs, changing the type of herbicide used, and alternative
application methods.
Many different herbicides and combinations are currently available for
use on corn. Currently the single most common method of weed control in
Iowa is a pre-emergence surface-spray application of atrazine and alachlor,
usually in combination with at least one cultivation (Richard Fawcett,
Personal Communication, 1978). This is the herbicide combination which
was used for determining costs on the Iowa example farm.
Atrazine, the most widely used herbicide is more persistent than many
other herbicides and is commonly found as a contaminant in surface waters in
corn growing areas (Morley, 1977; Richard et_ al. 1975). Thus decreasing
herbicide losses in runoff could potentially have water quality benefits
in these areas.
In.this section, several techniques for reducing herbicide losses in
runoff are investigated. These include use of alternative herbicides
(cyanazine plus alachlor and butylate plus cyanazine instead of atrazine
plus alachlor), alternative application methods (surface applied versus
incorporated) and SWCPs (conservation tillage, contouring, and terracing).
Properties and application methods for the three herbicide combinations
198
-------
are given in Table 8-21. Note that the cyanazine plus alachlor (Cy-al)
combination has the same application procedure as atrazine plus alachlor
(at-al) but cyanazine is less persistent in soil than atrazine. Butylate
plus cyanazine (bu-cy) has the same persistence as cy-al but is applied
earlier and incorporated. It should also be noted that since incorporation
requires thorough mixing of the top 15 cm of soil (two secondary tillage
operations after application), it is not compatible with either conservation
tillage or no-tillage.
TABLE 8-21. PROPERTIES AND TYPICAL APPLICATION METHODS
FOR THREE COMMON CORN HERBICIDE COMBINATIONS
Common Trade Names
Atrazine
plus
Alachlor
(at-al)
Aatrex
plus
Lasso
Cyanazine
plus
Alachlor
(cy-al)
Bladex
plus
Lasso
Butylate
plus
Cyanazine
(bu-cy)
Sutan+
plus
Bladex
Estimated Half-life in soil
(days)
Adsorption Group
(Table 4.3)
Application Rate
Kg/ha a.i.#
Application Method
Application Time
Cost
$/ha
Available tillage
practices
50*
B
1.25
2.5
Surface
Spray
Pre-Emer-
gence
(May 15)
$30.00
All
14
B
1.25
2.5
Surface
Spray
Pre-Emer-
gence
(May 15)
$32.00
All
14
B
4.0
1.25
Incorporation
Pre-PIant
(May 1)
$45.00
Conventional
only
* Estimated half-life for atrazine only. Half-life for alachlor is 14 days.
* Application rates are in kilograms of active ingredients (a.i.). Upper
number refers to rate for first herbicide listed in each combination.
199
-------
The base condition shown in Table 8-22 is straight^row conservation
tillage using atrazine plus alachlor. As discussed earlier, conservation
tillage is the most economical tillage system for growing corn in this area
of Iowa. Under conventional tillage, herbicide losses are over 25% greater.
The two SWCPs evaluated, contouring and contouring plus terracing, reduce
herbicide losses by 43% and 72%, respectively, when the herbicide is not
changed. Switching from the at-al combination to the cy-al combination
achieves only a fairly small (12%) reduction in herbicide loss. This re-
duction is due to the reduced half-life of cyanazine compared to atrazine •
(Table 8-21).
The largest reduction is brought about by switching to the incorporated
bu-cy combination. Although not evaluated here, even larger reductions
could be achieved with contouring and terracing. However, incorporation
TABLE 8-22. ESTIMATES OF HERBICIDE LOSSES IN RUNOFF
FROM CONTINUOUS CORN ON THE TAMA SOIL FROM THE IOWA EXAMPLE FARM
Average An-
rtual Loss
g/ha-yr
Effectiveness*
g/ha-yr
%
Reduction
Level of
Soil Erosion
MF /ha-yr
II
Base Condition
22.1
Atrazine-Alachlor (at-al)
SR-Conv
SR-Cons
Cont-Cons
Terr-Cons
27.9
22.1
12.7
6.2
Cyanazine-Alachlor (cy-al)
SR-Conv
SR-Cons
Cont-Cons
Terr-Cons
24.7
19.5
11.2
5.4
Butylate-Cyanazine (bu-cy)
S
SR-Conv
4.3
(*)
9.4
16.1
2.6
10.9
16.7
17.8
43
72
12
49
76
81
15.0
37.1
15.0
7.5
5.3
37.1
15.0
7.5
5.3
37.1
Effectiveness is the reduction in losses from the base condition.
The base condition is the least-cost method of growing corn. In
this case it is straight-row conservation tillage using the atrazine
plus alachlor herbicide combination.
Abreviations for SWCPs are: SR - straight-row; Conv - conventional
tillage; Cons - conservation tillage; Cont - contouring; Terr - con-
touring plus terracing.
Base Condition.
200
-------
requires conventional tillage and thus brings about a significant increase
in the level of soil erosion. Thus in areas where conservation tillage is
a feasible soil conservation practice, incorporation of herbicides as a
means for controlling losses of herbicides may require implementation of al-
ternative soil erosion control practices which would be less cost-effective
than conservation tillage. However, in areas where conservation tillage
is not economically feasible, such as the area of the Texas example farm,
incorporation may be an effective means of reducing herbicide losses.
Costs for these alternative practices are shown in Table 8-23 as the
level of cost increase expected above the cost of the base condition. Cost-
TABLE 8-23. COST AND COST-EFFECTIVENESS Op VARIOUS
PRACTICES FOR CONTROL OF HERBICIDE LOSSES FROM THE IOWA TAMA SOIL
Costs* $/ha
Tillage SWCP Herbicide
Costs Cost Cost
Total
Cost
Cost-
Effectiveness
g/$
Atrazine-Alachlor (at-al)
SR-Conv 8 5.00 — — 5.00
SR-Cons CO CO CO CO CO
Cont-Cons -- 2.22 -- 2.22 4.2
Terr-Cons -- 66.67 -- 66.67 0.2
Cyanazine-Alachlor Ccy-al}
SR-Conv ' 5.00 -- 2.00 7.00
SR-Cons - -- 2.00 2.00 1.3
Cont-Cons - 2.22 2.00 4.22 2.6
Terr-Cons — 66.67 2.00 68.67 0.2
Butylate-Cyanazine (bu-cy)
SR-Conv 9.00"" - 15.00 24.00 0.7
**
* Costs represent the increase in costs from the base condition, which is
straight-row conservation tillage using atrazine plus alachlor.
* Cost-effectiveness is the level of effectiveness, from Table 8-22
divided by the total cost.
^ Abbreviations are listed in Table 8-21.
* Base condition.
"* Incorporation requires one additional secondary tillage operation,
thus increases the tillage cost above that for conventional tillage
alone.
201
-------
effectiveness, which is the reduction in grams of annual pesticide loss
per dollar of cost is also tabulated. It should be remembered that these
costs can vary considerably, particularly the cost for contouring and the
relative prices of the herbicides. Thus the cost-effectiveness values
are only indicative and will vary from field to field and from year to year.
Given the cost assumption for contouring used here, it is the single
most cost-effective practice for reducing herbicide loss. Contouring and
switching to the cy-al combination results in a reduction of nearly 50% in
herbicide loss at a fairly small cost per hectare. However, to achieve
greater reductions either incorporation or a terrace system must be im-
plemented at a significantly higher level of cost and reduced cost-effectiveness.
If any practice is to be implemented solely for the purpose of reducing
pesticide losses in runoff, the implications of these cost-effectiveness
values should be considered. The maximum value in Table 8-23 is 4.2 for
contouring. Thus by implementing contouring, 4.2 grams of herbicide are
prevented from leaving the field per dollar spent. To achieve an 80%
reduction (switching to the incorporated bu-cy combination) only 0.7 grams
are retained per dollar spent. These values are based on edge-of-field loss
and not stream loading, thus the potential exists for degradation and de-
position (adsorption to deposited soil) before these chemicals reach a water-
way. Thus cost-effectiveness values for these alternatives may well be less
than one gram per dollar for many situations. Therefore it would seem that
careful consideration of benefits should be undertaken before implementing
on-field practices for reducing pesticide losses in runoff.
CONCLUSIONS
1.) Since sediment delivery is the product of soil erosion times the
sediment delivery ratio (SDR), SWCPs may affect sediment delivery
by either changing the level of soil erosion and/or changing the
SDR. With the possible exception of grassed buffer strips, non-
structural practices act solely by reducing soil erosion, while
structural practices, such as terraces, diversions, sod waterways,
etc., generally affect both soil erosion and the SDR.
2.) When comparing the effectiveness of different practices within
the same field, the effectiveness of non-structural practices
will be directly proportional to their effects on soil erosion.
Structural practices such as terraces which trap sediment will
usually (but not always) reduce the SDR. Thus their effectiveness
for reducing sediment delivery is proportionately greater than
their reduction in soil erosion. Structures such as diversion
ditches and grassed waterways provide channelized flow from the
field to the stream, thus often increase the SDR. In this case,
their effectiveness for controlling sediment delivery will be
proportionately less than their reductions in soil erosion. In
certain cases, these practices may actually increase levels of
sediment delivery in fields having an initially low SDR.
202
-------
3.) The cost-effectiveness of practices on a field, can vary con-
siderably depending on effectiveness, as noted above, and costs
which can vary according to type of farm, climate, and physical
properties of the field. Thus actual rankings of cost-effectiveness
of practices must be determined individually for different types
of situations. Four practices were found to have particularly
high potential levels of cost-effectiveness. These are conservation
tillage, no-till, contouring, and strip-cropping. However, cost-
effectiveness for the two reduced tillage systems drops consid-
erably in areas where yield decreases occur with these practices.
Strip-cropping is cost-effective only if sod crops are a part
of the basic cropping system. If additional sod crops must be
grown, the cost-effectiveness of strip-cropping is often quite
low.
Conservation tillage in particular, appears to have potential
for widespread adoption with little resistance from farmers. Til-
lage is generally more rapid than with a moldboard plow, and yields
can usually be maintained or even increased. Although moldboard
plowing is still done on most cropland, the use of conservation
tillage appears to be spreading fairly rapidly. Nearly 45% of
Iowa cropland now receives primary tillage with either a tandem
disc or a chisel plow (James McGrann, Personal Communication,
1978). It should be noted, however, that the use of these imple-
ments does not necessarily imply conservation tillage. In some
cases several trips over the field are made which buries almost
all residue, greatly reducing their effectiveness.
Conservation tillage could be implemented on much if
not most cropland in the country with little, if any, cost,
other than perhaps a program to convince farmers to reduce
their number of field operations.
4). The maximum cost-effectiveness for a terrace system is achieved
with one terrace at the lower edge of the field, acting es-
sentially as a sediment basin. If more terraces are constructed
to reduce the terrace interval, effectiveness will be increased
slightly, but less than the proportional increase in costs.
Thus the smaller the terrace interval, the lower the cost-effect-
iveness of the system.
5). The cost-effectiveness of a farm plan for reducing sediment
delivery is highly dependent on the location of practices on the
farm as well as the actual practices implemented. If the costs
for practices are approximately the same on all fields on a
farm, the cost-effectiveness of a non-structural practice on
individual fields will be directly proportional to the base
level of sediment delivery on the individual fields. Thus a
higher cost-effectiveness may be achieved by placing a
practice on a field with only a moderate level of soil erosion,
but with a high SDR, than on another field with high levels of
203
-------
soil erosion but with a low SDR. The cost-effectiveness of
structural practices which change the SDR is proportional to
both the initial level of sediment delivery and the initial
SDR. Thus if two fields have initially identical levels of
sediment delivery, but the first has a high level of soil
erosion and a low SDR, while the second has a relatively
lower level of soil erosion but a higher SDR, a structural
practice will have a higher cost-effectiveness on the second
field than on the first.
When a structural practice is implemented on a field,
the area above the structure will have a new SDR, while the
SDR for the area below the structure remains essentially un-
changed. A significant increase in cost-effectiveness for
the field can be obtained by placing the most highly erosive
crops in the area with the lowest SDR, even though this may
have no effect on levels of soil erosion.
6). In general, the higher the degree of effectiveness desired, the
lower the resultant cost-effectiveness. This is due to a
number of factors including:
i) Terrace systems are often required to obtain
high reductions in sediment delivery, but are
not highly cost-effective. Cost-effectiveness
of terrace systems declines as terrace intervals
become smaller.
ii) Combinations of practices are generally
less cost-effective than individual practices.
The marginal cost of reducing sediment delivery increases slowly
at low levels of reduction, but often turns sharply upward at
high levels of reduction. The point where marginal cost begins
to rapidly increase coincides with the implementation or practices
with low levels of cost-effectiveness. Among these practices
are intensive terrace systems, substituting sod-crops for row
crops, and removing land from prodcution.
7). Optimal farm plans developed for reducing sediment delivery
relied heavily on optimal location of practices, and on terrace
systems which reduced the SDR. Since these options reduce the
sediment delivery proportionately more than soil erosion, it was
found that sediment delivery from a farm could usually be reduced
by a given amount at less cost than that required to obtain
proportional reductions in soil erosion. The primary difference
between farm plans for reducing sediment delivery and plans for
reducing soil erosion was the choice of fields where practices
were implemented and the location of terraces within fields.
8). Current ASCS cost-sharing programs are generally concentrated on
structural practices, particularly terraces and permanent sod
204
-------
cover crops. The terrace spacings required by ASCS and SCS are
generally considerably less than those found to be optimal in the
farm plans. Thus for most cases, little or no cost-sharing money
was available for those practices which were most cost-effective
for reducing sediment delivery. The net effect of current cost-
sharing plans could therefore be that farmers will either not
implement any practices or that they will implement practices
which have a total cost greater than it would be if other, more
efficient practices were adopted. In general, most cost-sharing
on non-structural (cultural) practices, especially contouring
and reduced tillages systems would be helpful. Also, where
technically feasible, cost-sharing on large interval terrace
systems would increase the effectiveness of the cost-sharing
program.
9). The relative effectiveness of SWCPs for reducing losses of other
potential pollutants depends on the classification of the sub-
stance as defined in Tables 4-1 and 4-2. In general, farm plans
to reduce sediment delivery will reduce losses of these substances
less than proportional to the reductions in sediment delivery.
Control of solid-phase nutrients and substances classified as
strongly adsorbed should be only slightly less than control of
sediment delivery; losses of substances classified as moderately
adsorbed should be reduced with reductions in sediment delivery,
but only slightly. Losses of dissolved nitrogen and non-adsorbed
substances will often not be affected by farm plans for reducing
sediment delivery, unless these plans include practices, such as
sod-based rotations, which reduce required levels of inputs.
10). Although SWCPs and/or incorporation will often be effective
in reducing losses of herbicides in runoff, the cost-effectiveness
of all practices evaluated was fairly low. All practices reduced
edge-of-field losses of herbicides by less than 5 grams per year
for each dollar of cost. Thus careful consideration of benefits
should be given before implementing these practices solely to
reduce herbicide losses.
205
-------
SECTION 9
THE EFFECTIVENESS OF SWCPs IN COMPARISON WITH OTHER METHODS
FOR REDUCING PESTICIDE POLLUTION
Christine A. Shoemaker and Marion 0. Harris
The development and successful use of organic pesticides in the 1940's
resulted in a new approach to pest control, an approach which seemingly mark-
ed the beginning of man's triumph over the traditional enemies of agriculture.
The convenience and cost effectiveness of the new chemical methods greatly
simplified the task of pest control for the farmer and at the same time
conferred on consumers the benefit of high quality, inexpensive agricultural
goods. In. the intervening thirty years the use of pesticides has continued
to increase in both agricultural and nonagricultural areas. At present,
about one half billion kilograms of pesticides are used annually in the
United States, about half of which are used in agriculture. Use patterns of
the major agricultural insecticides and herbicides are listed for various
crops in Tables 9-1 and 9-2.
Unfortunately, the use of pesticides in agriculture has created certain
environmental problems in both target and non-target areas. Target agricul-
tural areas have suffered the consequences of pest resistence and resurgence
and have witnessed the creation of secondary pests and the decimation of
beneficial plant and animal species. Agricultural pesticides transported
away from target areas have disrupted neighboring as well as distant eco-
systems.
Although the movement of pesticides from application sites to surface
waters is not entirely understood, the major transport mechanisms are: over-
land flow, interflow (to ground and surface waters), direct spills or dumping,
drift directly into surface waters, and rainout of air-borne pesticides bound
on soil particles or in a vapor stage (Leonard et al. 1976). Pesticides
reaching surface waters may continue to move on through other components of
the ecosystem (leaving surface waters through evaporation or food chains) or
they may be strongly absorbed or retained until degradation occurs.
SWCPs are only one of many strategies being considered for the control
of pesticide pollution of surface waters. Other strategies include improved
operational and disposal techniques, changes or improvements in application
procedures, and increasing the efficiency of pesticide use through improved
pest management techniques. Each of these strategies will obviously have a
206
-------
TABLE 9-1. USE OF INSECTICIDES ON MAJOR CROPS IN THE U.S., 1971
Organo-
phosphates
Methyl
Parathion
Parathion
Diazinon
Phorate
Disulfoton
Malathion
Other Organo-
phosphates
Total Organo-
phosphates
Carbamates
Bux
Carbaryl
Carbofuran
Other Carbamates
Total Carbamates
Organochlorines
Toxaphene
Other Organo-
chlorines
Total Organo-
chlorines
Other Insecti-
cides
Total
Corn
7
604
905
1210
142
52
497
3417
1625
750
1219
-
3594
83
4483
4566
839
12416
1000
Cotton
10449
1163
-
45
102
305
1286
13352
552
-
35
587
12778
6594
19372
37
33348
kg of Active Ingredients
Soybeans
1004
27
-
64
1
40
1
1137
612
-
-
612
693
113
806
58
2613
Alfalfa
62
112
69
10
103
182
186
724
47
-
25
72
8
229
237
20
1053
Total Use On
Major Crops
12528
4259
1427
1899
1840
1232
6373
29558
1639
7542
1298
506
10985
14940
13185
28125
29049
97717
Source: Adrilenas (1974)
1 Total use in U.S. on the following crops: corn, cotton, soybeans, alfalfa,
peanuts, tobacco, wheat, other grains, other field crops, other hay and
pasture, Irish potatoes, other vegetables, citrus, apples, other fruits
and nuts, nursery and greenhouse crops.
207
-------
TABLE 9-2. USE OF HERBICIDES ON MAJOR CROPS IN THE U.S., 1971
Herbicide
2,4-D
Propachlor
Alachlor
Butylate
Atrazine
Simazine
Arsenicals
Fluonieturon
Trifluralin
Alanap
Vernolate
Dinitno Group
Chloramben
Nitralin
Other
Herbicides
Total
1000
Corn Cotton
4156 2
9682
3800 2
2645
23636
418
3440
1515
2065
2
-
7 174
20
227
6651 1505
51015 8932
kg of Active Ingredients
Soybeans Alfalfa
101 105
•
214 16
2867
-
8
56 8
22
-
2710
1344
657
1638 10
4245
975
1796 156
16633 295
Total1
15115
10786
6706
2689
26007
783
3562
1515
5194
1515
1698
3269
4343
1230
83048
167460
Source: Adrilenas (1974).
Total use in U.S. on the following crops: corn, cotton, soybeans, alfalfa,
wheat, sorghum, rice, other grains, peanuts, sugarbeets, other field crops,
pasture and rangeland, Irish potatoes, other vegetables, citrus, apples,
all other fruits and nuts, summer fallow, nursery and greenhouse crops.
208
-------
different effect on the amount of pesticide available for transport away from
the target area via drift, runoff, leaching, volatilization, and improper
disposal. SWCPs will, in most cases, only affect the fraction of pesticide
which is being lost in runoff (solid phase and dissolved) whereas pest con-
trol techniques like scouting will bring about an overall reduction in the
amount of pesticide being applied, thereby reducing pesticide losses occur-
ring through all transport routes.
The purpose of the study described in Section 9 is to compare the
effectiveness of SWCPs with the effectiveness of other strategies in reducing
the amount of toxic pesticide reaching the environment. The following items
have been critical in this analysis:
1) a delineation of the relationship between pesticide pollution
control strategies and the amount of pesticide available for
transport via drift, runoff, volatilization, leaching, and
disposal;
2) a quantification of the amounts of applied pesticides lost by
means of each of these transport mechanisms;
3) a determination of the environmental hazards (toxicity and
persistence) associated with pesticide losses occurring through
each transport mechanism;
4) a determination of any secondary benefits or hazards associated
with the use of each strategy.
Evaluations of the proposed pollution control strategies are hindered
by the tremendous ranges in toxicity, persistence, and mobility exhibited by
the numerous pesticides presently available to farmers. To reduce this
problem to a manageable size, we have examined the effectiveness of the
various strategies in reducing the losses of five major pesticides from agri-
cultural lands. The five materials examined are toxaphene, methyl parathion,
carbofuran, paraquat, and atrazine. The first three represent each of the
major categories of insecticides: organochlorines, organophosphates, and
carbamates, respectively. Paraquat and atrazine represent bipyridylium and
triazine herbicides. The five pesticides studied are also major weed and
insect control agents in corn and cotton, two crops whose production requires
a major portion of all pesticides used in U.S. agriculture (Tables 9-1 and
9-2). Fifty percent of the insecticides and eleven percent of the herbicides
used in agriculture are used in cotton production. The use of toxaphene and
methyl parathion constitutes more than two-thirds of the insecticides used in
cotton. The production of corn consumes eighteen percent of the insecticides
and fifty percent of the herbicides used in agriculture. Atrazine use makes
up for nearly one half of the herbicides used in corn and was identified in
1971 as the most heavily used herbicide in the U.S. (Adrilenas, 1974).
Paraquat is another herbicide used in corn production and was studied mainly
because of its importance in no till production systems. Carbofuran is a
relatively new insecticide used primarily for the control of soil insect
209
-------
pests of corn. Carbofuran and other soil insecticides account for most of
the insecticides used in corn production.
In order to relate the conclusions obtained for the five representative
pesticides to other pesticides, the following sections include a general dis-
cussion of runoff and leaching, air-borne losses, losses due to improper
operational and disposal techniques, toxicity, persistence, SWCPs, increasing
efficiency in the use of pesticides, and improving operational and disposal
techniques. These sections are followed by a more detailed discussion inte-
grating the previous material in an analysis of the representative pesticides.
Readers wishing to skim this discussion are advised to read the introductory
section entitled, Representative Pesticides, the summary sections at the end
of the representative pesticide discussions, and the final conclusions section
which includes Table 9-11.
LOSSES OF PESTICIDES FROM TARGET AREAS
Runoff and Leaching
The small percentage of toxic material transported in runoff water and
sediment limits the usefulness of SWCPs for reducing pesticide pollution in
comparison with other control procedures. Wauchope (1978), in his extensive
review of the literature on pesticide runoff losses, concluded that losses
are generally 0.5% or less of the amount applied, with larger losses indicat-
ing the occurrence of a large runoff-producing rainfall event within one to
two weeks after the pesticide application. Wauchope estimated losses for
organochlorines to be about one percent of the applied material. He con-
cluded that the major reason for the higher losses is the persistence of
organochlorine insecticides in soil. Wauchope also noted that losses of
herbicides applied as a wettable powder may be as high as five percent of the
applied material, depending on weather conditions and the slope of the treated
field. The results of field and modelling studies on runoff are shown in
Table 9-3.
The effectiveness of SWCPs in reducing these runoff losses depends in
large part upon the chemical and physical characteristics of the pesticide.
Since mobility phenomena occur most readily, and in some cases exclusively
with dissolved pesticides in the soil water, the balance between solid phase
pesticides and dissolved pesticides is the single most important factor in
determining the fate of pesticides in soil (Helling et^ al.1971). Adsorption
is often described by the adsorption partition coefficient (Ka) which is
defined as the adsorbed concentration on the soil divided by the concentration
in the soil water. Under laboratory conditions this value is typically de-
termined by agitating pesticide-free soil with a pesticide in solution,
allowing sufficient time for equilibrium, and measuring the concentrations.
Under field conditions, the same situation presumably occurs after application
as the pesticide goes into solution in the soil water, and eventually
reaches equilibrium with the soil. Desorption occurs when an equilibrium
soil-water-pesticide solution is diluted with soil, water or both. This
would typically happen during and after a precipitation event occurring
210
-------
TABLE 9-3. RUNOFF STUDIES ON VARIOUS PESTICIDES
Pesticide
Carbaryl3"
Carbofuran
Diazinon
Methyl Parathion
Q
Toxaphene
Alachlor
to g
i-- Atrazine
f
Cyanazine
£
Metribuzin
Paraquat
Propachlor
Propachlor
Simazine**
Trifluralin
Chemical Class
Carbamate
Carbaraate
Organophosphate
Organophosphate
Organochlorine
Acetanilide
s-Triazine
s-Triazine
s-Triazine
Bipyridylium
Chloroacetamide
Chloroacetamide
s-Triazine
Nitroaniline
Application Rate
(kg Al/ha)
5. 031'4,7
5.411.1*.7
1.121.S7
6.7 !,6,8
26. 82,8
.2.241'5'9
1.2-3.361.1*.7
2.241,5,9
0.561,5,9
1.121,5,9
2.241.5,9
6.7M,9
1.12-4.481>'*»7
1.383>6»7
Average Slope (%)
10
9
10-15
2
3
2-9
8-22'
2-9
2-9
2-9
2-9
10-15
8-22
0,2
Runoff
(% of Applied)
0.15
0.9
insignificant
4.8
0.36
0.48
0.02-5.7
0.96
0.72
3.34
0.21
3.1
0.04-5.40
0.04
-------
TABLE 9-3. FOOTNOTES
1 silty loam soil
2 loamy sand soil
3 silty clay soil
^ measurements taken in corn
5 measurements taken in fields with a corn-soybean rotation
6 measurements taken in fields with cotton-soybean rotation
7 application incorporated into soil
8 foliar application
9 broadcast spray application
a Caro et al_. (1974)
b Caro et_ al. (1973)
C Ritter et_ al_. (1974)
See Appendix F
e Bradley et al. (1972)
£ Baker et al. (1972)
g Triplett et_ al_. (1978)
h Willis et al. (1975)
sometime after application of pesticides. The equilibrium ration of concen-
trations after the dilution is called the desorption partition coefficient
Q(d]. A plot of pesticide concentration on soil vs. concentration in water
yields the adsorption-desorption isotherm. This isotherm is generally not
single-valued (Davidson £t al_. 1972), that is, Kd i= Ka for the same soil water
concentration under most conditions. Ka and Kd are also dependent on organic
carbon content, clay type and content, pH, soil structure, and temperature,
as well as the particular pesticide. Thus, adsorption-desorption phenomena
are difficult to predict, and-are one of the major problem areap in deter-
mining pollutant -losses from cropland.
Pesticides having strong adsorption characteristics will exist primarily
212
-------
in the solid-phase when located in the soil environment. Paraquat, diquat,
and DDT are considered to be in this class of pesticides. Upon contact with
soil clay minerals, paraquat and diquat are almost completely and irrever-
sibly adsorbed (Calderbank, 1968). Helling et al.(1971), in their classifi-
cation of the soil mobility of numerous pesticides, listed DDT as the least
mobile. Monitoring studies in the South (Nicholson e^ al. 1968) have found
significant amounts of DDT in sediment, but only insignificant amounts in
river water. Toxaphene is similar to DDT in having a low water solubility,
and strong adsorption characteristics. Although some studies have indicated
that toxaphene is also nearly 100 percent adsorbed fLaFleur, 1974)
and highly immobile in soil (Helling et al. 1971), Nicholson has found that
toxaphene moves more readily in water than DDT and occurs more frequently
in water samples in monitoring studies (personal communication, 1977).
Most other pesticides will exist in both the solid-phase and dissolved
states when in the soil environment. Trifluralin, parathion, methyl para-
thion, phorate, and diazinon are strongly adsorbed (Scott and Phillips, 1972;
Saltzman and Yaron, 1971; Meyers ejb aL 1969), while carbofuran, carbaryl,
atrazine, propachlor, and 2,4-D show more moderate adsorption characteristics
(Jamet et_ aL 1974; Leenheer and Ahlrichs, 1971; Meyers et_ al. 1969; Colbert et^
al.1975; Ritter ejt a_L 1974; Newman and Thomas, 1949; Helling ejt al, 1971).
Although no studies have been found on chloramben adsorption, it is appar-
ently one of the most weakly adsorbed pesticides. Helling et_ al_. (1971)
classify it as one of the most highly mobile pesticides in soil, while Hilton
et al. (1974), state that it "moves readily in sandy soils or following heavy
rains."
Pesticide adsorption and desorption is thus a complicated phenomena. A
few pesticides are almost 100 percent adsorbed in the soil while
others exhibit no significant degree of adsorption. However, the majority of
pesticides are in an intermediate class. Because the degree of adsorption
for these pesticides depends on the particular pesticide, climatic conditions,
soil conditions, and equilibrium time, accurate prediction of the behavior of
each individual pesticide in soil may be difficult, and even impossible in
certain cases.
Table 9-4 contains several indices which give information concerning
the adsorption characteristics of selected pesticides. Some indication of
the degree of pesticide adsorption to soil particles is given by a transport
mode index taken from Stewart £t al.(1975). This index indicates how the
pesticide will generally move from the treated field if runoff occurs, that
is, whether concentrations of pesticides will be higher in water (dissolved),
sediment (solid-phase) or sediment and water equally. It must be remembered
that this index only predicts which fraction of the runoff will carry the
highest concentrations of pesticides. Since the volume of water lost via
runoff is always larger than the volume of sediment lost, a higher percent of
the applied pesticide will leave in the dissolved state than in the solid-
phase state, in spite of the fact that residue concentrations may be higher
in the soil fraction. Paraquat, diquat, toxaphene, and DDT are exceptions to
this general rule because of their aforementioned adsorption characteristics.
213
-------
TABLE 9-4. PESTICIDE - WATER - SOIL - AIR INTERACTIONS*
Pesticide
Carbaryl
Carbofuran
Diazinon
Malathion
Methyl Parathion
Parathion
to
*> Phorate
Toxaphene
Alachlor
Atrazine
Chloramben
2,4-D
Paraquat
Propachlor
Trifluralin
o
Transport
Mode
SW
W
SW
W
SW
S
SW
S
SW
SW
W
W
S
W
S
Water ,
Solubility
(ppm)
40
700d
40d
145
60d
25
50
3
242
70
700d
900h
very soluble
580d
24
Leaching
Index
2.0
2.0e
2.0
2.0-3.0
2.0
2.0
1.0
1.0-2.0
2.0-3.0h
3.0-4.0h
2.0
1.0j
1.0-2.01
1.0-2.0
c
Vaporization
Index
3.0-4.0
o£
3.0
2.0
4.0
3.0
2. OS
4.0
3.0
2.01
very low
1.0
v
nonvolatile
2.0
-------
FOOTNOTES FOR TABLES 9-4, 9-5, and 9-6.
a
Stewart et^ al. (1975) where runoff losses occur from treated fields,
W = chemical moves primarily with water,
S = chemical moves primarily with sediment,
SW = chemical moves with both sediment and water.
Gunther £t al.(1968) except where noted.
r*
Haque and Freed (1974) estimated leaching and vaporization rates from best
available information on loam soil at 25°C under annual rainfall of 150 cm.
Data from their table except where noted. Leaching Index number indicates
the approximate number of centimeters the material moved through the soil
profile.
1 = < 10 cm
2 = < 20 cm
3 = > 35 cm
4 = > 50 cm
Vaporization Index number indicates the kilograms vaporized from one
hectare of soil surface during one year.
1 = < 0.1 kg/ha-year
2 = 0.2 to 3.0 kg/ha-year or more
3 = 3.5 to 6.5 kg/ha-year or more
4 = 7 to 14 kg/ha-year or more
Leonard et al.(1976).
e U.S.E.P.A. (1976).
f Harris and Miles (1975) classified insecticides as volatile and non-
volatile in soil based on measurements of fumigant activity against
insects.
g Sanborn
-------
FOOTNOTES FOR TABLES 9-4, 9-5, and 9-6 (CONTINUED).
m Pimentel (1971) except where noted.
n Hilton £t al. (1974).
° von Rumker et^ al_. (1974).
P Eichelburger and Lichtenberg (1971).
q Mathur et_ a]_. (1976) .
r F. M. C. Corporation, unpublished data from U.S.E.P.A. (1976).
S Lichenstein and Schulz (1964) • However, in later work they found that
one half of the remaining methyl parathion persisted as bound residues.
Work later proved that these bound residues were not excluded from
environmental interaction.
t Guyer e^t al. (1971).
U Terriere £t al. (1966).
V Haque and Freed (1974).
W Yeo (1967).
Pesticide adsorption characteristics are not the only factors influenc-
ing the amount of pesticide leaving agricultural fields in runoff. Other
factors are: (1) pesticide properties such as solubility, persistence and
volatility; (2) climatic characteristics such as radiation, temperature, and
magnitude and timing of runoff-producing precipitation; (3) watershed
characteristics such as slope, roughness, management practices, and proximity
to water courses; (4) soil characteristics such as texture, pH, antecedent
moisture status, hydraulic properties, and organic content; and (5) miscel-
laneous factors such as timing, mode, placement, formulation, and rate of
the pesticide application (Dean and Mulkey, 1979; Stewart et al. 1975;
Davidson et^ aL 1978). Recent work by Bovey et_ al. (1978) indicates that
concentrations of pesticides in runoff water may be reduced if subsurface
applications of pesticides are made instead of surface applications.
Because of the difficulties and high costs associated with empirical
studies measuring the amount of pesticide found in runoff under varying
weather conditions, mathematical models of pesticide movement have become an
important tool for estimating pesticide losses. The adsorption coefficient
and degradation time of a pesticide are important aspects of these models.
This type of mathematical modelling is discussed in earlier sections and in
216
-------
Appendix F. The results for paraquat, methyl parathion, and atrazine are
discussed later in this chapter. Results from several mathematical runoff
models are given in Table 9-3.
Very little is known about the extent and environmental significance of
the pollution caused by the runoff of pesticides. Incidents of surface water
contamination by the organochlorine insecticides are still occurring, and
several lakes in the Midwest have been closed to fishing because of accumula-
tions of toxic materials (Luckmann, personal communication, 1978). It is not
known whether this contamination is due to the continuing runoff and erosion
of soils carrying these persistent pesticides, though many researchers have
noted that concentrations of organochlorine residues are higher in areas of
high sediment losses and turbidity of surface waters (Glooschenko et^ al.
1976; Truhlar and Reed, 1976; Schulze at al. 1973; Miles, 1976). A~study by
Richard et^ al. (1975). revealed the presence of pesticide residues in the
Mississippi River at New Orleans and in every major watershed in the State of
Iowa. They found fluctuations in pesticide concentrations in surface water
to be consistent with predictions made by the pesticide runoff model of
Bailey et al. (1974).
The rate at which pesticides leach through the soil profile is influen-
ced by many of the same factors influencing runoff losses: adsorption/
desorption characteristics of the pesticide, chemical reactions, solubility,
rate of application, antecedent soil moisture, soil structure, and flow
velocity (Sanborn et^ al, 1977). Such pesticides as 2,4-D, atrazine, cyanazine,
dicamba, dimethoate, chloramben, dinoseb, monuron, and methoxychlor all
exhibit a greater propensity for movement in the soil. However, Gerakis and
Sficas (1974) have concluded that under normal conditions, extensive leach-
ing, and subsequent contamination of groundwater is unlikely.
Leaching will be more of a problem in areas of pesticide disposal, areas
with water tables not far beyond the root zone of crops, and in areas with
sandy soils containing little organic matter or clay to bind pesticides as
they percolate through the soil profile. Some of the reported groundwater
contamination by pesticides may occur when solid-phase pesticides are washed
down the deep cracks which can appear when heavy rains follow periods of
drought (Phillips and Feltner, 1972; Willis and Hamilton, 1973).
Several reports have been made on the presence of pesticides in ground-
water. Atrazine and monuron have been found in groundwater during periods of
application, and in wet years (Leh, 1968). A study in Iowa revealed pesti-
cide (atrazine, DDE, dieldrin) contamination in all waters originating from
shallow wells located in the alluvial plains of contaminated rivers. Current
treatment processes for municipal waters did not significantly reduce con-
centrations of these pesticide residues (Richard et al.1975).
The most widely documented cases of pesticide contamination of surface
waters have concerned the organochlorines. It is not known whether the
paucity of pollution reports attributed to runoff and leaching of other
pesticides is indicative of field conditions or of the state of the analytical
217
-------
art. In either case, one cannot entirely dismiss the possibility that the
runoff and leaching of pesticides from agricultural lands could bring about
a low level of pollution in surface and ground waters. This pollution could
produce little understood chronic and sublethal toxicity effects which would
be hard, if not impossible to monitor (Livingston, 1977).
Air-Borne Losses of Pesticides
In most cases, SWCPs are only effective in controlling the transport
of pesticides which have already reached the soil. However, much of the
pesticide applied never reaches the soil because it is carried away in the
form of drift or because it volatilizes from the leaves of the crop. Pesti-
cides may also volatilize from the soil before they can be washed away by
runoff or may be carried on particulate matter eroded by wind. In some cases,
the amount lost by drift and volatilization is over 50 percent of the
material applied. In such cases, the amount of air-borne pesticide leaving
the target area completely dwarfs that being carried in runoff. In other
cases, especially for pesticides incorporated into the soil at the time of
plowing, the air-borne losses are negligible.
Though it is generally assumed that substantial amounts of pesticides
enter the atmosphere by drift, volatilization and wind erosion, little is
actually known about the fate of these contaminants once air-borne. There is
some indication that pesticides do not remain in the atmosphere for long
periods of time, though the rate at which they return to earth depends on
degradation patterns during transport, concentration, deposition rate of
aerosol particles, and the turbulent exchange coefficient (Duttweiler and
Malakov, 1977). Decomposition apparently occurs with air-borne pesticides
both in the vapor state and when they are deposited on surfaces (Crosby,
1978). Pesticides not undergoing decomposition may be precipitated in rain
or may diffuse slowly into the stratosphere as vapor (Duttweiler and Malakov,
1977).
Work by Cohen and Pinkerton (1966) showed that dieldrin and atrazine
residues are associated with suspended solids in rainfall. These data gave
the first qualitative evidence that pesticides could enter the atmosphere
adsorbed on soil particles, and thus could be transported and redeposited on
the earth's surface quite distant from the original area of application.
Using reported residue levels of DDT in rainwater samples, Pearce et al.
(1978) estimated that about 5000 kilograms of toxic material were
deposited by rainfall in the wedge-shaped area covering the Gulf of the St.
Lawrence from July to October in 1968. Indeed, Goldberg et^ aL. (1971) have
suggested that this aerial fallout may greatly exceed river discharge in its
contributions of pesticides to the oceans of the world. The oceans may in
turn serve as vast reservoirs, constantly discharging pesticides back into
the atmosphere, and thereby serving as a "continual source of replenishment
for atmospheric pollution" (Edwards, 1977).
Drift losses of pesticides can be substantial if applicators do not
strictly adhere to the operational guidelines intended to prevent most
218
-------
off-target losses. Aerial applications in windy conditions may cause drift
damage to surrounding crops, thus one would assume that liable spray appli-
cators would attempt to prevent such contamination. Whether such care is
taken to avoid the contamination of surface waters is not known. Factors
influencing drift losses have been covered in extensive reviews by von Rumker
et ad.(1975) and Plimmer (1976). We will only briefly discuss their results
here.
Wind velocity, pump pressure, nozzle angle, time of application,
nozzle type, droplet size, formulation and nozzle height above the ground all
influence the percent of applied pesticide ending up in drift losses (Ware
et al,. 1969; Ware et al. 1975; Ware et aL 1972; Byass and Lake, 1977). The
impact of these factors on pesticide drift losses have been summarized in the
extremely informative general drift loss tables presented by von Rumker ert
al, (1975). Droplet size is extremely important to drift losses. Droplets"
ranging from 10 to 50 ym in diameter are likely to drift up to several
kilometers, whereas droplets greater than 100 pm in diameter rarely
drift except in very windy conditions (Gerakis and Sficas, 1974). Brazzel
et_ aL (1968) have suggested that more progress in the control of drift could
be made by the precise control of the droplet spectrum than by any other
means. Others have pointed out that smaller droplets provide better plant
coverage and better insect control (Smith et^ al_. 1975). Obviously, a compro-
mise must be made between application strategies providing optimal pest
control and those strategies bringing about minimal drift losses.
The formulation of the pesticide also has a large impact on drift loss-
es. In the past, dust applications of pesticides presented great risks of
drift losses and thus, have been largely replaced by spray and granular
formulations. In the case of DDT, it was found that dust applications placed
fourteen times more material in downwind drift than was found with spray
applications (Gerakis and Sficas, 1974). The use of granular formulations
greatly reduces drift potential.
The way in which the formulation is applied also influences how much of
the pesticide will reach the target area. The potential for drift losses is
greatly increased as the distance from nozzle to the target increases. Byass
and Lake (1977) have estimated that amounts of pesticides lost in drift can
increase by a factor of five when boom heights are increased to levels within
the range often used in ground applications of herbicides. Surface incorpor-
ation of pesticides allows for the most accurate, though expensive, placement
of toxic materials. Aerial applications present the greatest potential for
drift problems, resulting in four to five times more drift than occurs from
high clearance ground sprayers (Ware e^ al_. 1969). Whatever the application
procedure, insecticide applications tend to create greater drift losses than
herbicide applications as treatment is at high pressure with small droplet
sizes, as compared to lower pressure herbicide applications, resulting in
larger droplet sizes. Byass and Lake (1977) have shown, however, that even
small drift losses of herbicides can be hazardous. In the case of picloram,
drift losses comprising of more than 0.01% of the applied material resulted
in significant damage to sensitive plant species located downwind of the
219
-------
spraying operation. Thus, in evaluating the importance of drift losses, one
must consider the toxic quality of the drift as well as the quantity pro-
duced.
Volatilization can be a major cause of pesticide disappearance from
plants and soil. It can contribute to the short term toxicity of many insec-
ticides (methyl parathion), the low efficiency of some herbicides (triflur-
alin), and environmental contamination by those pesticides not readily de-
grading in air (Spencer and Cliath, 1977; Quinby et^ a]_. 1958). Gerakis and
Sficas (1974) have estimated that about one half of the pesticides applied
to field crops enter the atmosphere through evaporation from soil and plant
surfaces. The rate at which this volatilization occurs depends to a certain
extent on the pesticide's vapor pressure, water solubility, concentration,
and strength of adsorption. However, environmental factors such as cover
crops, soil incorporation, air flow rate, composition of the soil, and pene-
tration into plant surfaces greatly modify the effects of the pesticide's
physical characteristics, and give rise to wide ranges of volatility in field
situations (Harns and Lichtenstein, 1961; Farmer et_ aJL 1973; Lichtenstein
et al. 1970; Spencer and Cliath, 1977). Soil temperatures reaching 50-60°C
further encourage volatilization losses of pesticides from agricultural
fields (Kearney et^ aJL 1964; Swoboda et^ aJ_. 1971). The rate of volatilization
of pesticides buried in the soil is much slower and depends on the rate of
pesticide desorption from soil, the rate of diffusion to the soil surface,
and the mass flow of water to the soil surface (Spencer and Cliath, 1973;
Farmer et_ al_. 1973). Farmer and Letey (1974) have further discussed the
effects of soil incorporation on volatility, and have developed several models
for the prediction of volatility losses.
Direct measurement of volatilization under field conditions is quite
difficult, thus indications of volatilization losses generally come from
either laboratory studies or indirectly from air quality sampling. Lichten-
stein et^ al. (1970) have studied the relative volatilities of various organo-
phosphate and organochlorine pesticides and found that approximately 0.96%
of the parathion, 0.63% of the diazinon, 16.3% of the fonofos, and less than
0.01% of the azinphos methyl volatilized from soil incubated at 30°C, where-
as 26.1%, 11.5%, 1.32% and 0.78% of aldrin, lindane, dieldrin, and DDT
volatilized, respectively, in the same period of time. Haque and Freed
(1974) have predicted higher vaporization rates for both parathion and
diazinon.
Parathion and methyl parathion have been found in air samples taken in
Florida (parathion and methyl parathion) and Alabama and Mississippi (methyl
parathion alone), albeit less frequently, and at lower levels than the
organochlorines (Stanley et_ al. 1971). The pervasiveness of organophosphorus
residues was further illustrated by the results of another study wherein the
air in the homes of people working with pesticides (farmers and formulators)
was tested for methyl parathion and parathion. For the pesticide formulators,
25 percent of the indoor samples, and six percent of the outdoor
samples contained methyl parathion, while none was found at the farmers'
homes. For parathion, 85 percent of the inside sites and 65
220
-------
percent of the outdoor sites at formulators' homes showed residues, while
approximately 50 percent of the indoor sites and 33 percent of
the outdoor sites at farmers' homes showed contamination. The researchers
commented on these surprising results by noting that the volatility and short
environmental half life of parathion would not lead one to expect its
occurrence in samples of house dust and inside air (Tessari and Spencer
1971). F
Few studies on carbamate volatilization are available. Harns and
Lichtenstein (1961) found no evidence of carbaryl volatilization in tests
for toxicity to house flies. However, Haque and Freed (1974) give carbaryl a
vaporization index of three to four (see Table 9-4), compared to two for
malathion and three for diazinon and parathion. Thus carbaryl losses from
soil apparently can occur. Stanley et^ al_. (1971) did not look specifically
for any carbamates in atmospheric samples tested, but did check unidentified
peaks (gas chromatograph), and mentioned no findings of carbamates.
Volatilization rates of the herbicides propachlor, chloramben, and
paraquat are either unknown or considered insignificant, while rates for
2,4-D may vary considerably, depending on the formulation (Hilton et_ al. 1974).
Atrazine losses through volatilization are generally not large, but can be
significant if high temperatures and prolonged sunlight follow application
prior to precipitation (Hilton et^aj.. 1974; Harris et^ al. 1968). Trifluralin
is subject to both volatilization and photodecomposition, and must be soil
incorporated within eight hours after application. Results from Watkinsville,
Georgia, show vapor losses of approximately 3.5% of applied trifluralin during
application, and a total of 25.9% of the applied amount lost through photo-
decomposition and volatilization over the growing season (White et al. 1977).
Indices and qualitative data concerning pesticide vaporization are
shown in Table 9-4. A vaporization index developed by Haque and Freed (1974)
estimates the annual vapor losses of selected pesticides from a loam soil
at 25°C, under an annual rainfall of 150 centimeters. According to their
index, BHC, heptachlor, parathion, diazinon, and alachlor may have vapor
losses of up to 6.5 kilograms per hectare, while carbaryl, methyl parathion
and toxaphene may vaporize at a rate of 7 to 14 kilograms per hectare.
Very little is known about the environmental significance of air-borne
losses of pesticides from agricultural lands. Although air samples taken
over agricultural and nonagricultural areas have indicated the presence of
several insecticides, many researchers have dismissed the low residue levels
as insignificant. However, the reduced winter hardiness of non-target plants
associated with herbicide use, and the mental health problems observed with
exposure to organophosphate insecticides may produce environmental hazards
which are so insidious as to escape detection. The dangers and presence of
herbicide drift have been attested to in a study done by Hibbs (1976), who
showed that the decline of the hackberry fCeltis occidentallis L.) in north-
west Iowa could be attributed to herbicide drift from corn growing regions.
Byass and Lake (1977) have shown that sensitive plant species located down-
wind of spraying operations can be damaged even when applications are done by
221
-------
ground equipment in the lightest winds encountered. Unfortunately, the
transient nature of such drift losses makes it difficult to trace the resul-
tant contamination back through its transport mechanism to the source.
Further evaluations of drift and volatilization losses will probably be
limited until more data are gathered on the degradation and rate of fallout
of air-borne pesticides.
Pesticide Losses Due to Improper Operational and Disposal Techniques
Improper operational and disposal techniques are a serious cause of
pesticide pollution of surface waters in agricultural areas. The filling of
spray tanks and subsequent disposal of containers often takes place beside
ponds or streams, thus, spills from overfilled spray tanks, spills from
containers, and the absence of check valves on intake hoses may all result in
direct contamination of aquatic systems. In some cases, diluted pesticides
remaining in the spray tank after crop treatment are dumped on land or down
storm sewers. Disposal problems have also occurred in nonagricultural sec-
tors because of negligence, and the lack of available techniques for disposal
of the large quantities of surplus pesticides handled by commerical appli-
cators, chemical manufacturers, and registered pesticide disposal services.
The preponderance of disposal-related fish kills in western New York reflects
the hazards involved in such procedures and indicates the need for programs
and research to help establish safer and more convenient means of preventing
such contamination (Walter et al. unpublished).
Several investigations have been done to determine whether pesticides
and their containers are being properly disposed of in farming areas. In a
study currently being done at Southern Illinois University, 600 five-galIon,
used pesticide containers were sampled at one farm disposal site. It was
found that less than 40 percent of the sampled containers were rinsed and
less than 20 percent were bottom punctured. An average of more than
2.5 ounces of pesticide was found in the nearly 370 cans which were not
rinsed. Other used cans of trifluralin, alachlor, 2,4-D, and butylate con-
tained more than 1 quart to 3.5 gallons, 4.25 gallons, 1 gallon and 1 quart,
respectively, of undiluted pesticides. Another study at Oregon State Univer-
sity has shown that on the average, nearly 6 ounces of pesticide is left un-
rinsed in five-gallon containers (Luckmann et^ al. 1978).
ENVIRONMENTAL IMPACT OF PESTICIDE LOSSES
Toxicity
It is important to consider not only the amounts of material leaving
target areas but also their impact on non-target organisms. Most researchers
have attempted to evaluate these environmental effects through acute toxicity
bioassays. However, while these laboratory tests have been valuable for a
comparison of the relative toxicities of various pesticides, their results
are often of limited significance without verification from field tests
(Livingston, 1977). These lab experiments, by their very nature, leave out
many complex environmental factors which can influence the uptake and
potential effects of pesticides. In the case of aquatic systems, pesticide
222
-------
adsorption to sediments and vegetation could greatly reduce toxicity, while
sublethal and secondary effects, concerning destruction of an important food
species, could end up being as important as the direct toxic action itself
(Holden, 1973; Hurlbert, 1975). Potential destruction of food species is
particularly important in studies of herbicides, for they, by the very nature
of their action, are likely to affect the lower trophic levels of food chains
(von Runker £t al_ 1975), and could conceivably have complex secondary effects
related to the presence of decaying vegetation and altered water quality
(Livingston, 1977). Another factor, which acts to further complicate any
evaluation of environmental impact, is the possible toxic synergism of pesti-
cides with other pesticides and pollutants present in the aquatic environ-
ment (Lichtenstein et al. 1969; Lowe et_ al, 1970).
Livingston (1977), after having considered the limitations of laboratory
data, suggested that "if there is one major need in pesticide research today,
it is the completion of the natural sequence of hypotheses based on experi-
mental work through field testing and ecological analysis." Hurlbert (1975)
in turn, has criticized present attempts at experimental field investigations
and suggests that if their conclusions are to be taken seriously in the
future, proper consideration must be given to adequate controls and replica-
tions. Whatever the approach, it is evident that more research must be done
to ascertain the effects of pesticide use on both terrestial and aquatic
systems, for as von Runker et^ aL (1975) put it, "it is not known whether or not
currently practiced monitoring and observation methods would detect such
effects prior to the occurrence of massive ecological damage."
Having stated the limitations of such data, we present lethal dose and
lethal concentration ranges in Table 9-5. A number of insecticides are
classified as highly toxic. These include aldicarb, parathion, methyl para-
thion, carbofuran, phorate, diazinon, and toxaphene, the latter three being
extremely toxic to fish. At a toxicity of 0.65 mg/kg for aldicarb, less than
0.2_g would be fatal to an average size (70kg) human. Thus special care is
required for handling these materials, particularly since most are toxic by
oral, inhalation, and dermal routes. Herbicides, with the exception of para-
quat, are relatively nontoxic to mammals. Effects on fish, with the exception
of trifluralin, are also minor when compared to insecticidal toxicity, how-
ever as mentioned previously, sublethal and secondary effects may adversely
influence fish populations. Other materials such as simazine, diuron and
linuron are extremely toxic to algae (Sanborn ejt al. 1977).
Persistent chemicals are sometimes stored in the cells of organisms
and are passed on to predators in higher trophic levels in a food chain.
Organochlorine insecticides such as DDT have been found in much higher con-
centrations in the bodies of members of higher trophic levels than in the
bodies of the lower organisms which originally ingested the material
(Metcalf et^ aL 1971) . This phenomenon is known as biomagnification.
Few studies have been done on biomagnification of organophosphates and
carbamates. Miller et al. (1966) showed short term concentration of parathion
and diazinon in mummldiog and mussels. Concentrations in the animals were as
223
-------
TABLE 9-5. TOXICITY OF VARIOUS PESTICIDES
m
Pesticide
Carbaryl
Carbofuran
Diazinon
Ma lath ion
Methyl Parathion
Parathion
Phorate
Toxaphene
Alachlor
Atrazine
Chloramben
2,4-D
Paraquat
Propachlor
Trifluralin
Mammals LD mg/kg
200-540
8-14b
108
480-1500
6-32
4-56
3.7
15-240
1800b
1750-3080
3500-5620b
100-1000
150
71 On
> 2000
Fish LCrn ppm
1.5-2.0
0.21b
0.052-1.45
0.02-12.9
2.75-9.0
0.125-2.7
0.005-1.0
0.002-0.05
2.3b
0.55-12.6
7.0b
1-1000
45-840
735-5000n
0.001-0.540
Birds LD,..,, mg/kg
1000-3000
0.24-5.0
3.5-4.3
1485-5000
7.5-10.0
0.12-24
6.2-7.1
10-316
> 2000
> 1000
970-4048h
> 2000
Footnotes defined on pages 215 and 216.
224
-------
high as 24 times the water concentration, but dropped considerably within six
days. Some organophosphates, such as parathion have been found in small con-
centrations in sea animals (Deubert and Gray, 1976). However, because of
their low fat solubility, rapid degradation in biological organisms, and
other chemical and physical properties, it is unlikely that organophosphorus
insecticides biomagnify in food chains (USEPA, 1975). Sangha (1972), re-
porting on carbamates in a model ecosystem, found propoxur and carbofuran to
"be degradable, but had persistent residual products, whereas mobam and aldi-
carb exhibited low biodegradability. Mobam was concentrated from 9 ppb in
water to 15400 ppb (parts per billion) in mosquito larvae, and 110 ppb in
fish, while aldicarb had a concentration of 7.4 ppb in water, 17000 ppb
in mosquito larvae, and 2320 ppb in fish.
Biomagnification has generally not been a problem with most herbicides
(Livingston, 1977). Metcalf and Sanborn (1975) have tested alachlor,
atrazine, dicamba, cyanazine, 2,4-D, propachlor, and metribuzin in a model
ecosystem, and have concluded that their use will not result in biomagnifi-
cation in the aquatic food chain. They also tested Counter, which showed a
low level of ecological magnification in aquatic systems. Mauck et al.(1976)
have found that simazine residues can persist for over a year in aquatic
fauna, with biomagnification occurring in invertebrates. However, they found
no significant harmful effects over a two-year period of studying pond fauna.
An exception to the general rule of herbicide nonaccumulation is trifluralin,
which has been shown to magnify in aquatic food chains (Sanborn et al, 1977).
Problems have arisen with herbicides and the organophosphate and
carbamate insecticides in spite of their general low levels of biomagnifi-
cation. Adverse effects have occurred at levels below the established
acutely "safe" level. Leptophos has been shown to produce delayed neutro-
toxic effects on chickens and was responsible for the deaths of 1500 water
buffalo in Egypt after inadvertent exposure. Recent evidence that DBCP, a
pesticide ingredient previously thought to be safe, can cause sterility and
possibly cancer in humans at long-term, low-level exposures has brought into
question the adequacy of the whole procedure for testing pesticide safety
(Stevens, 1977). Further research on neurological and other types of low-
exposure chronic effects, particularly from organophosphorus insecticides,
has also been called for (Metcalf and McKelvey, 1976).
Persistence
The term persistence, as applied to pesticides, has come to have several
meanings perhaps implicit in the literature, but nonetheless confusing for
the layman. One meaning commonly used by researchers defines persistence by
the presence or absence of insecticidal or herbicidal activity. Highly ad-
sorbed pesticides losing toxicity on adsorption to soil particles would, by
this definition, be described as nonpersistent. A second definition might
lead other researchers to studies of the period of time necessary for a
complete disappearance of the original material, be its toxicity continued or
discontinued. Highly adsorbed pesticides would be persistent according to
this definition. Concern about toxic degradation products has led to the
225
-------
third, and what is perhaps the most inclusive of all persistence definitions,
that is, persistence as that period of time necessary for a complete degra-
dation of the pesticide and its metabolites into harmless products. Research-
ers using this definition have often expanded what was simply a determination
of soil and water persistence, to a consideration of persistence in the total
environment, realizing that pesticides disappearing from soil through vola-
tilization may go on to cause pollution problems through subsequent atmos-
pheric cycling. Unfortunately, research using this last definition has been
limited by inadequate analytical techniques, and by a lack of data concern-
ing the persistence of pesticides and their metabolites in soil, water and
air.
Until recently, pesticides were generally classified as persistent or
nonpersistent. The chlorinated hydrocarbons, which are generally considered
to be persistent, have degradation rates often measured in years, while the
"nonpersistent" pesticides generally have degradation rates measured in weeks
or months. Recent evidence of the existence of "bound residues" of pesti-
cides existing in soil, makes these classifications considerably less dis-
tinct. This appears to be particularly true for the organophosphorus insect-
icides, previously considered non-persistent (Davidson, personal communica-
tion, 1977). Lichtenstein et al. (1977) tested an agricultural loam soil
for methyl parathion and fonofos residues and found that after 28 days,
seven percent of applied methyl parathion and 47 percent of applied fonofos
existed as extractable residues. However, an additional 43 percent of
the methyl parathion and 35 percent of the fonofos existed as bound residues
(measurement was by radio-carbon techniques, thus percentages include both
the pesticide and possible metabolites). Similar results have been re-
ported for parathion (Katan et^ a^. 1976). Lichtenstein concludes: "In
view of the above findings, the expressions 'disappearance' and 'persis-
tence' of pesticides so widely used during the last two decades should be
reassessed to consider the bound products." The possible existence of bound
residues should be kept in mind when interpreting the results from the
literature quoted below.
Persistence data for various selected herbicides and insecticides is
given in Table 9-6. When looking at this data, it must be remembered that
each figure represents one set of interactions between the pesticide, agro-
nomic practices and biological conditions. A more detailed review of per-
sistence can be found in the work done by Sanborn et al. (1977).
The persistence of the organochlorine insecticides is attested to by
the pervasive nature of their residues in agricultural soils throughout the
country. Some of the crops grown on these soils continue to be contaminated
by pesticide residues and metabolites in spite of the general decline in
organochlorine use in recent years (Moore et al. 1977). The contamination
resulting from these persistent residues is~notf isolated in soils and crops;
wind and water erosion of agricultural fields provide ready transport, intro-
ducing toxic materials to aquatic and terrestial ecosystems far from the area
of original application. Moore et^ al. (1977) have found dieldrin residues
in soybeans grown on fields previously treated with aldrin and in soybeans
from untreated fields. They suggest that the dieldrin contamination in un-
treated fields is due to the transport of contaminated soil particles by air
226
-------
TABLE 9-6. PERSISTENCE OF VARIOUS PESTICIDES
Pesticide
Carbaryl
Carbofuran
Diazinon
Malathion
Methyl Parathion
Parathion
Phorate
Toxaphene
Alachlor
Atrazine
Chloramben
2,,4-D
Paraquat
Propachlor
Trifluralin
In Soil
3 weeks
16 weeks^
8-16 weeks
< 2 weeks0
4 weeks
1-834 weeks
1-104 weeks
5 years
8-12 weeks0
16-48 weeks
4-8 weeks
1-4 weeks
m
48 weeks or longer
w
3-4 weeks
12-16 weeks
In Water
1-2 weeks in river water"
1 week
2-4 weeks in river water17
2-4 weeks in river water*
4-8 weeks in river water*
1-5 years
half-life 6 hours
9 days
1/2 hour
w
Footnotes defined on pages 215 and 216.
227
-------
or water. Menzie (1972) lists the half-lives of several organochlorine in-
secticides in soils as three to ten years for DDT, one to seven years for
dieldrin, seven to twelve years for heptachlor, two to four years for chlor-
dane, and ten years for toxaphene.
Much research has been done on the persistence of organophosphate in-
secticides. Lichtenstein and Schulz (1964) investigated parathion, methyl
parathion, malathion, and the following parathion metabolites: para-oxon
(the main oxidation product), aminoparathion, p-nitrophenol, and p-aminophenol.
They found that five percent of the parathion remained two and one half
months after application, whereas, malathion degraded to the same percent of
the applied material in four days. Para-oxon degradation in soil was rela-
tively rapid with 50 percent of the material being lost within 5-1/2 hours,
and 90 percent gone 24 hours after it was applied. Another experiment
showed p-nitrophenol to be a major degradation product of para-oxon; as
para-oxon residues decreased over a 24-hour period, p-nitrophenol residues
increased proportionately. The p-nitrophenol disappeared completely within
sixteen days, amino-parathion within three days, and p-aminophenol almost
immediately. Stewart et^ aJ^. (1971) have found residues persisting for up to
sixteen years when parathion was applied at high application rates (35 kg/ha).
Diazinon disappears relatively rapidly in most soils. Getzin and Rosefield
(1968) found that in a moist Sultan silt loam, 45 percent of the
applied diazinon had degraded in four weeks. Diazinon can be more persistent,
though problems are often limited to those situations when the material is
sprayed on a cold, dry, alkaline soil which has never been treated with the
insecticide before (Sanborn et al. 1977). Experiments carried out on phorate
on the same soil showed that only 20 percent of the initial concentration
remained after four weeks (Getzin and Shanks, 1970). However, a review by
Sanborn et^ al.(1977) reported studies wherein phorate's persistence in soil
extended up to 104 weeks.
Herbicides range from the relatively persistent to those disappearing
within weeks. Horowitz et^ al_, (1974) rated trifluralin as moderately persis-
tent, compared with ten triazine, diazine, and substituted urea herbicides.
At a high concentration rate of application (7.2 ppm), trifluralin continued
to cause oat injury up to five months in the glasshouse. In the field, tri-
fluralin remained phytotoxic for six months, except at the highest rate of
application (14.7 kg/ha), which showed residual activity after nine months.
The use of simazine on fruit and vegetable crops has created some soil
residue problems for susceptible crops following treated crops, as has the
use of atrazine on field and sweet corn (Sweet, personal communication, 1978).
The persistence of alachlor is similarly marked in months, though its use
presents no problem to sensitive crops grown in the following season (von
Rumker ert al_. 1974) .
Propachlor, 2,4-D, and chloramben all seem to have a somewhat lower
persistence. Most of the 2,4-D applied to soil disappeared within a month
(Audus, 1952; Newman and Thomas, 1949), while approximately 30 percent of
applied propachlor remained in the soil 21 days after application
(Ritter et al, 1974). Although no studies on chloramben degradation were
228
-------
found, Hilton et_ aL (1974) report average persistence of six to eight weeks
in the field. Paraquat's complex soil adsorption behavior creates a situ-
ation wherein it is difficult to label it as either persistent or nonpersis-
tent. This phenomenon will be further discussed in the section on paraquat.
As can be seen from Table 9-6, persistence data for pesticides in water
are scanty. This information is extremely important to any evaluation of the
importance of nonpoint sources of pesticide pollution from agriculture. The
behavior of the toxic material in aquatic systems can be such that it may be
imperative or entirely pointless to try to prevent small concentrations from
entering surface waters.
Numerous cases of contamination of aquatic environments by organo-
chlorines have been documented, an occurrence which might be expected given
the persistence of these materials. Eichelburger and Lichtenberg (1971), in
their persistence studies using river water, found that 100 percent
of the heptachlor epoxide, endrin, dieldrin, DDT and DDE remained after
eight weeks. Wolfe et_ al. (1977) studied the hydrolytic degradation of
methoxychlor at pH's and temperatures commonly found in aquatic systems, and
found its half-life to be approximately one year. Organochlorines generally
degrade slowly in aerobic aquatic environments, but are aided in their
decomposition by microorganisms in anaerobic aquatic environments (Luckman
et alf 1978).
The few data available on other pesticides indicate a more rapid
degradation in water than is found with the organochlorines. The rapid
hydrolysis of organophosphate insecticides in aquatic systems (eight to
twelve days) greatly reduces any hazards associated with their use (Pionke
and Chesters, 1973; Muirhead-Thomson, 1971). Parathion does not hydrolize as
rapidly as the other organophosphates, but undergoes microbial degradation
in both anaerobic and aerobic aquatic environments (Graetz et_ aj^. 1970).
Studies by Harris and Miles (1975) have found residues of parathion, diazinon,
and ethion in waters draining Bradford Marsh, an agricultural area which
received as many as 19 applications of parathion in one season for the
control of onion maggot. The degradation of the carbamate insecticides
depends on the pH of the solution, with degradation occurring in less than
four weeks in slightly alkaline river water (Eichelburger and Lichtenberg,
1971). Yu and associates (1974) have reported that carbofuran hydrolizes
rapidly in water. The degradation of herbicides is influenced by temperature
and the availability of sediments containing active microbial populations
(Luckman e_t al. 1978).
STRATEGIES FOR REDUCING POLLUTANT LOSSES
The Effectiveness of SWCPs in Reducing Pesticide Transport
The effectiveness of SWCPs in reducing the runoff losses of pesticides
which are moderately adsorbed to soil particles has been addressed in other
sections of this report. As our intent in Section 9 is to compare the effec-
tiveness of SWCPs with other methods for reducing losses of pesticides, we
229
-------
will only summarize the findings of Sections 4 and 7 which are pertinent to
our discussion.
a. Most pesticides used in agriculture are moderately adsorbed to
soil particles and are transported primarily in the dissolved
phase rather than the solid phase of runoff. The pesticide
residues which are transported in the solid phase are primarily
associated with fine clay particles. As a result, SWCPs are not
as effective in reducing the movement of these pesticides as
they are in reducing the movement of sediment.
b. The persistence of a pesticide has a major impact on runoff
losses. In Section 7 it is reported that substitution with a
less persistent pesticide will reduce runoff losses more than
any of the SWCPs examined.
c. In most cases, SWCPs do not increase the fraction of applied
pesticide which is lost in surface runoff.
Before one can conclude that SWCPs are beneficial for the control of tfie
pollution caused by the runoff of moderately adsorbed pesticides, several
factors must be considered. The first is that some SWCPs such as no-
tillage are associated with an increase in pesticide usage (Section 8).
Because of these increased application rates, the amount of pesticide trans-
ported in runoff from no-tillage fields can be larger than the amounts trans-
ported in runoff from conventional tillage. Secondly, we must consider the
possibility of increased leaching of pesticides with SWCPs. Although labor-
atory, field and modelling studies (Section 7] have generally dismissed the
possibility of significant leaching losses, monitoring studies have found
pesticides in groundwater and sliallow wells (LaFleur et^ al. 1973; Leh, 1968;
Richard et_ aj^ 1975). The increased percolation of water brought about by
SWCPs could make this pollution problem more serious.
The use of SWCPs for reducing non-point source pollution of surface
waters by pesticides is more effective in the case of the strongly adsorbed,
persistent organochlorine insecticides and bipyridylium herbicides. These
materials are transported mainly in the sediment fraction of runoff, and
degrade slowly in surface waters. It has been surmised that SWCPs would keep
these toxic materials in agricultural soils where they would degrade without
harming species living outside of the target area. However, several factors
make these arguments for the implementation of SWCPs to control runoff losses
of strongly adsorbed pesticides less compelling. First of all, most of the
organochlorines have been taken off the market, leaving farmers with moder-
ately adsorbed pesticides which are transported primarily in the dissolved
phase rather than the solid phase of runoff. The use of the SWCP strategy
for the strongly adsorbed pesticides is thus limited to pesticides remaining
on the market such as toxaphene and paraquat, and the residues of the
formerly used organochlorine insecticides remaining in the soil environment.
Case studies on paraquat and toxaphene which follow this discussion describe
the effectiveness of SWCPs for reducing losses of these materials to
230
-------
non-target ecosystems. The question of organochl-orine residues remains and
poses some difficult problems. Is it worthwhile to implement SWCPs for the
purpose of reducing runoff losses of pesticide materials whose residues re-
main in the soil but whose use has been much restricted since the earlv
1970's?
Several facts throw some light on this question. As
mentioned previously in Section 9, runoff is not the solitary or primary
transport route for organochlorine residues. Many of these residues leave
agricultural fields through volatilization, drift, and wind erosion of con-
taminated soil particles. These air-borne losses may be degraded in the
atmosphere or redeposited on the earth. Pearce et_ al_. (1978) have estimated
that about 5000 kilograms of DDT were deposited by rainfall during a
four-month period in the wedge-shaped area covering the Gulf of St. Lawrence.
Moore et al. (1977) have found dieldrin residues in untreated fields, and
have related the contamination to volatilization, drift, and wind erosion
losses of dieldrin as well as runoff from treated fields. A second point to
be made about the soil residues of banned organochlorines concerns their
degradation in soil. Lichtenstein et al. (1971) investigated the soil residues
of aldrin, heptachlor and their metabolites or formulation impurities in
cultivated and non-cultivated fields, and found that the cultivated fields
had 76 to 82 percent fewer residues. Runoff losses appar- •
ently did not occur (Lichtenstein, personal communication, 1979). Thus the
authors attributed the reduced amounts of residues to the breakdown and
volatilization of toxic materials. If this prolonged persistence in non-
cultivated fields does indeed occur, the reductions of runoff losses gained
by such SWCPs as no-till and conservation tillage might be negated by the pro-
longed persistence of organochlorine residues. As discussed in Section 7,
reduced persistence can decrease runoff losses more than SWCPs. A final
point to be made about these organochlorine residues is that they have been
present in soils for several years now and have to a certain degree already
been carried to streams and lakes by runoff or entered the atmosphere by
volatilization. Levels of organochlorine residues in wildlife continue to
decline (Butler and Schutzmann, 1978), thus it is questionable whether it
would be worthwhile to implement SWCPs for strongly adsorbed pesticides at
this late date.
The remainder of the material in Section 9 focuses on the comparison of
the effectiveness of SWCPs with the effectiveness of other practices for re-
ducing pesticide transport away from target areas. In many cases, it will be
shown that there are alternatives which are more effective than SWCPs in re-
ducing pesticide pollution. However, it must be noted that many SWCPs imple-
mented for nutrient and sediment control will have additional benefits in
reducing pesticide losses. In some cases SWCPs will serve a dual role as
pollution control strategies and as pest control strategies. Sod-based
rotations and strip-cropping will, in certain situations, reduce the amounts
of pesticide being applied to the crop and furthermore reduce any pesticide
runoff losses that might occur. Shelter belts serve a similar dual function
by reducing drift and wind erosion losses of pesticides from agricultural
lands. SWCPs which do not increase pesticide use can also be. used in combi-
nation with practices like scouting which reduce the number of pesticide
231
-------
applications needed for crop production.
Unfortunately, little is known about the impacts of various SWCPs on
weed, insect and disease populations. For example, SWCPs such as no-till and
conservation tillage markedly alter the crop ecosystem and can, thus, have a
significant impact on pest population dynamics. As was pointed out earlier,
these alterations require increased usage of herbicides and allow for a build-
up of crop residues in fields. There is some concern that insect and disease
problems will be exacerbated by the continuous presence of this protective
crop cover. Similarly, other SWCPs could potentially have an impact on the
survival of insect and pathogen pests because of the increases in soil mois-
ture accompanying reductions in runoff losses. Until further research is
done in these areas, we cannot be sure that the implementation of SWCPs will
not increase pest problems in some cases and thus also increase the amounts
of pesticide applied and transported from fields.
Increasing Efficiency in the Use of Pesticides
Because of the pest resistance and hazards to non-target organisms
associated with pesticide use, extensive efforts have been made to develop
effective pest control programs which require reduced amounts of pesticides.
These programs increase efficiency of pesticide use by implementing one or
more of the following strategies:
(1) pest monitoring to improve the timing of pesticide applications
relative to the pest population,
(2) biological or cultural pest control methods to supplement or
replace the control given by pesticides,
(3) changing application procedures or equipment to improve the
placement of pesticides,,
Strategies (1) and (2) have been developed and implemented primarily for the
management of insect and disease populations, whereas, strategy (3) deals
with placement problems which are common to the application of insecticides,
fungicides, and herbicides. As there has been little integration in the
research and development of the first two strategies with the third strategy,
we will discuss their potential contributions to pollution control separately.
Population monitoring indicates when disease and insect populations
reach densities which warrant a pesticide treatment. This density is called
an "economic threshold" or "the density at which control measures should be
applied to prevent an increasing pest population from reaching the economic
injury level" (Stern et^ al^ 1959). The economic injury level is the pest
density, p, such that,
C = Y(p) - Y'(p)
where C = cost of the pesticide application
Y = net economic return from the harvested crop given that the
232
-------
pest density is p and a pesticide application is made
Y! (p) = net economic return from the harvested crop given that the
pest density is p and a pesticide application is not made.
These carefully timed treatments can be contrasted with the formerly used,
fixed-schedule spray treatments which occurred at regular intervals throughout
the growing season, whether or not the pest population was present at injuri-
ous levels. Pest population monitoring has in most cases reduced the number
of pesticide treatments applied in comparison with these fixed-schedule spray
programs. A national survey of entomologists indicated that if the population
monitoring technique of scouting was implemented in all areas where it was
cost effective QL.£., where scouting causes no net reduction in farm income),
the national use of insecticides in cotton would be reduced by approximately
20 percent (Pimentel e* al. in press).
Cultural methods of pest management alter the environment of the pest
in order to reduce the potential for pest damage and the need for pesticide.
Examples of cultural methods of control are changes in planting and harvest
dates and the destruction of crop residues serving as overwintering sites for
insect pests. Pests which cannot complete their life cycles in the absence
of a continuous crop of their host plant are often successfully controlled by
crop rotations.
Biological control methods take advantage of the pest control provided
by the various natural enemies of insect and weed pest populations. In some
cases, predators or pathogens are released repeatedly throughout the growing
season to provide short term control similar to the control given by pesti-
cides. The most widely used of these short-lived biological agents is
Bacillus thuringiensis which is used for the control of a range of lepidop-
terous insects. Much effort has also been directed at establishing more
permanent biological controls by encouraging the growth of predator or para-
site species which can maintain population levels large enough to contain
pest populations for many generations. One of the most successful examples
of biological control is the suppression of olive parlatoria scale by two
hymenopterous parasites. Prior to the introduction and establishment of
these parasites, olive parlatoria scale was the major insect pest of olives.
It is estimated that these parasite introductions have saved olive growers
over seven million dollars in insecticide costs (Huffaker et_ al^. 1976). In
addition to the introduction of exotic parasite species, attempts have been
made to protect native populations of beneficial insects by changes in pesti-
cide formulation, dosage or application timing. Beneficial insects can also
be protected by habitat management practices such as the strip-cropping of
alfalfa.
Resistant plant varieties represent another important means of reducing
pesticide use in agriculture. Wheat varieties resistant to Hessian fly and
corn varieties resistant to European corn-borer have been available for years
and have been tremendously important in reducing the need for insecticides in
wheat and corn production. Cotton varieties such as frego bract cotton with
resistance to boll weevil, are currently being developed for commercial
233
-------
release. The use of resistant plant varieties is more widespread for disease
control than it is for the control of insect damage.
Recent efforts in insect pest control have concentrated on the integra-
tion of biological controls, cultural controls, resistant plant varieties, and
pesticide treatments timed by population monitoring. Approaches which attenpt
to coordinate several methods of control are referred to as "integrated pest
management programs" or "IPM". The most effective combinations of control
methods vary for different crop species and localities. For example, in
Texas cotton pests are controlled by an IPM program which consists of a
combination of scouting, short season cotton, and a "diapause control" pro-
gram for destroying overwintering sites. "Diapause control" requires the
destruction of crop remains in crop fields and adjacent vacant fields follow-
ed by insecticide treatments to kill overwintering boll weevils. Such a
program involving scouting and sanitation has been implemented on thousands
of hectares of cotton in the lower Rio Grande Valley region. Frisbie has
estimated that the implementation of this program has allowed farmers to re-
duce insecticide applications from twelve treatments to nine treatments per
year (reported for 1972-1974 in Pimentel et_ al. in press). Because of differ-
ences in climate and pest complexes, IPM programs for cotton in the West and
in the Southeast are quite different from the programs used in Texas, but in
all areas there are IPM programs which require reduced amounts of insecticide
at no loss in income to the grower.
To determine the national impact of the implementation of integrated
pest management techniques in cotton production, Pimentel et al. (in press)
interviewed 32 insect pest specialists working in cotton producing
regions. The results of this survey indicate that the implementation of IPM
techniques in cotton would reduce insecticide use by an average of 40
percent without decreasing farm income (Table 9-7). Lacewell et_ aL (1976)
studied an IPM program in Texas which utilized narrow row, short season
cotton and found that the program required 27 percent fewer insec-
ticides and herbicides. In addition, they found that production costs, in
terms of dollars spent per kilogram of lint produced, were reduced from
$1.04 for typical Frio County production systems ito $0.593 using the IPM
program. Energy use on a per hectare basis was reduced 33 per-
cent, while energy use per kilogram of lint produced was reduced by 56
percent.
Integrated pest management programs are also being developed for insect
pests of corn. The major focus of this research has been on the control of
the insects which cause the most damage to corn, that is, the corn rootworm
and other soil inhabiting insects. Scouting has a great potential for re-
ducing the amounts of insecticide needed to control these soil inhabiting
pests. Luckman reported that at least 50 percent of the fields scouted
in his field studies in Illinois had corn "rootworm population densities below
the economic threshold, and thus were not treated with a soil insecticide
(personal communication, 1978). Before scouting programs were implemented,
fields were routinely treated every year to control the corn rootworm. Since
the cost of scouting is less than half the cost of an insecticide treatment,
234
-------
to
U)
Ul
TABLE 9-7: EFFECT OF NATIONWIDE IMPLEMENTATION OF COTTON INTEGRATED CONTROL PROGRAMS ON INSECT
CONTROL COSTS AND ON INSECTICIDE USE (From Pimente.1 et al. in press)
1.
2.
3.
Insect
Control
Alternative
Scouting Only
Most Economical
IPM Currently
Available
Most Economical
IPM Available
in 5-10 years
Savings
Insect
Control
Costs
(Million $)
26
81
136
Insecticide
Chlorinated
Hydrocarbons
82
61
41
Use as a Percentage of Current
Organo-
phosphates Carbamates
72 45
54 98
40 88
Use1
Total
77
59
42
Using insect control practices and cotton acreages reported in 1972-1974 it was estimated that
insect control costs were $253 million per year and the use of chlorinated hydrocarbons, organo-
phosphates and carbamates in cotton was 46, 31, 1.3 million kilograms, respectively.
-------
scouting is a cost-effective means of reducing insecticide use in these
areas.
Cultural control methods have also proven to be an effective way of
reducing the need for chemical control of the corn rootworm. Rotation of
corn crops with other crops disrupts the corn rootworm1s life cycle, and
prevents any rootworm damage from occurring in the corn crop following the
rotation crop. When rotations involving more than one year of corn are used,
rootworm damage may occur during the second or subsequent years following a
rotation. Sod-based rotations can thus be seen as both a SWCP and a pest
management technique. Unfortunately, in some areas the use of crop rotation
is quite costly for the farmer because the net income from the alternate crop
(.£.£., alfalfa or soybeans) is considerably lower than the net income for
corn.
In summary, recent advances in the integration of pest control methods
have resulted in cost-effective programs which reduce the need for insecti-
cides in cotton, corn, and other crops. Wider implementation of these pro-
grams promises substantial reductions in insecticide use and corresponding
reductions in air-borne, runoff, and leaching losses of insecticides. The
development of integrated pest management programs has been slower in the
case of weeds, although expanded research efforts in this area are currently
underway. It is impossible to predict the impact of this IPM research on the
future of herbicide use in U.S. agriculture.
Given the large losses of pesticides to non-target areas, it is indeed
regrettable that more has not been accomplished to improve the placement of
pesticides relative to the pest species, von Rumker et_ ai. (1974) estimated
that less than one percent of the insecticide applied by aerial equipment is
absorbed by insects through contact, inhalation, and ingestion. The rel-
atively low price of pesticide has reduced the incentive for research and
development in this area. Himel (1974) described present systems of crop
spraying as the most inefficient industrial process ever practiced.
In spite of the lack of research on new spraying technologies, there
are several simple techniques which are presently available to improve the
placement of pesticides. As discussed in the section Air-borne Losses,
proper care and adjustment of spray equipment could vastly improve this situ-
ation. Pumps, agitation systems, and pressure regulators should all be in
proper working condition, and nozzle tips should be regularly checked for
wear. Tests have shown that wettable powder formulations quickly wear down
nozzle tips creating situations wherein application delivery rates increase
by twelve percent after spraying just 50 acres (Fruit Grower, 1978). Chem-
ical rate, tractor speed, nozzle size, line pressure, height" of the boom and
band width all influence the placement of the pesticide and thus will greatly
affect the effectiveness of the material. Proper timing of spray procedures
is also extremely important to any improvement in the placement of pesticides.
Spraying in the early morning or at night will substantially reduce drift
losses to non-target areas (Smith et^ aJL 1974; Ware et_ al. 1972). The use of
adjuvants to increase the viscosity of spray solutions will further reduce
236
-------
drift losses. Richardson (1974) tested two adjuvants and obtained 63
to 98 percent reductions in drift losses.
While the development of new spray equipment has been slow in coming,
several researchers have reported on promising new technologies. Ware et al.
(1975) have shown that under identical weather and application conditions,
aircraft using Raindrop™ nozzles deposited 25 percent more spray on
target, and reduced off-target drift by 50 percent when compared with type
D flooding nozzles. McWhorter (1977) has reported on the use of the re-
circulating sprayer for post-emergence control of johnsongrass, redroot pig-
weed, and hemp sesbania in soybeans, and has found a spray recovery of 60 to
90 percent, which is then recirculated for use. Work at the University of
Minnesota has been done to evaluate the roller applicator which has a
frame-mounted, carpet-covered roller replacing the header of a self pro-
pelled swather. This roller improves the placement of pesticides because
the active material is fed onto a roller which does not drip pesticide on
the crop. The unit has been tested using Roundup to treat quackgrass and
reed canary grass in blue grass seed fields, however, the principle may
have an application in many crops (Sein, 1978).
Improving Operational and Disposal Techniques
Extension personnel have for many years made available information per-
taining to the proper techniques for filling spray tanks and disposing of
concentrated and dilute pesticides and their containers. This information
has been received with varying degrees of interest on the part of farmers.
Pesticide experts working in the field have- noted disposal and filling tech-
niques ranging from very poor to excellent and have called for an expansion
of education programs to further acquaint farmers with proper techniques and
hazards resulting from improper disposal. Deposits on pesticide containers
might give farmers further incentive to properly rinse and return containers
to dealers for recycling. If compliance with voluntary programs is not ade-
quate, expanded regulations concerning disposal should be considered.
While much of the responsibility for proper disposal and operational
techniques lies with the farmer, large improvements in the situation are
somewhat dependent on the development of new technologies. For example,
although some surplus pesticides and empty containers may be returned to the
manufacturer for reuse, others will have to be disposed of in a manner which
protects the environment and complies with various Federal and State laws.
Unfortunately, the technology for such disposal is not highly developed.
Incineration is perhaps the most promising method for disposal (Sanborn et
al. 1977) but is presently limited in its applicability by high costs and
the small number of companies offering such a service. Burial and soil in-
jection of pesticides have also been considered as means of disposal with
sites being in areas where public health, terrestrial and aquatic systems will
not be adversely affected. Sanborn et al. (1977) have studied the feasibility
of disposing of large quantities of selected pesticides in the soil and
have concluded that such disposal is feasible for EPTC, trifluralin, methoxy-
chlor, carbaryl, and malathion but not for atrazine, paraquat, diuron,
237
-------
linuron, and endosulfan. Their general conclusion was that not enough infor-
mation concerning degradation and metabolites exists to allow for concentra-
ted soil disposal of most pesticides.
It is our belief that research concerning the disposal of pesticide
containers should be encouraged. Techniques like the inoculation of soils
with bacterial strains utilizing pesticide compounds have a potential appli-
cation in both disposal and polluted areas. Changes in application equipment
might also work to minimize the problems of losses due to improper filling
and disposal techniques. One such method is the closed-system filling of
spray tanks wherein the unopened pesticide container is inserted into an
attachment on the sprayer. After the pesticide has flowed out of the con-
tainer, the container is rinsed and punctured, ready for recycling or burial
in approved areas. This system, which is mandatory in California, insures
proper disposal of containers while reducing poisoning risks to the appli-
cator. Tank rinsate disposal problems might also be solved by equipment
changes. At present, a new application procedure is being developed which
circumvents this problem. Instead of mixing an entire tank of dilute pesti-
cide before application begins, new spray equipment mixes the pesticide con-
centrate with water as it comes through the spray nozzle. By using this
technique farmers avoid the problem of disposing of dilute spray materials
not used on the crop. At the end of spraying procedures, unused pesticides
are in their original concentrated form, ready to be stored and eventually
reused. Given the high cost of many pesticides (atrazine costs over $50 for
five gallons), the farmer has an incentive to save the unused concentrate.
As the problem of disposal is similar for all pesticides, we will not
discuss the significance of these losses specifically for the case study
pesticides.
REPRESENTATIVE PESTICIDES
The chemical characteristics of pesticides themselves as well as in
the methods of application and availability of alternative pest control
programs vary so much that it is difficult to make general state-
ments about the relative erfectiveness of SWCP and other methods. Hence, we
have selected five pesticides to examine in detail. These pesticides are
representative of major classes of pesticides and thus allow tentative con-
clusions to be made concerning the importance of SWCPs for their close
chemical relatives.
Pesticides move into the environment through runoff, drift, volatil-
ization, and possibly leaching. Each of the methods we have discussed affects
the amounts carried by each of these pathways in different ways. Improved
pest control programs like IPM which reduce the amount of pesticide applied,
result in lower amounts of pesticide being carried in all pathways. SWCPs
in most cases, reduce only the amount carried in runoff. Adjustments in
spray procedures or equipment are effective for reducing drift losses, but
by increasing pesticide efficiency, could' reduce the amount of pesticide
applied to the field and thus, could reduce losses by runoff and leaching.
238
-------
The following sections discuss for each of the representative pesticides
the effectiveness of SWCPs and methods for increasing the efficiency of pesti-
cide use. Since the methods have varying effects on the amounts carried in
different pathways, the possible environmental impact associated with material
in each pathway is discussed to the extent possible given the limited data
available.
The benefits of each of the methods for pollution control are limited
by the maximum amounts of pesticide carried in each of the transport pathways.
Table 9-8 gives a summary of literature reports on amounts of each of the
representative pesticides which were transported through runoff, drift and
volatilization. For the foliar insecticides, toxaphene and methyl parathion,
losses in runoff are insignificant in .comparison with the amount of material
lost in drift and volatilization. Very little carbofuran is carried by run-
off, and that which is carried in -runoff degrades very rapidly in aquatic
systems. Paraquat losses in sediment are much higher than for drift and
volatilization; however, as is discussed below, the paraquat carried in
sediment does not appear to pose a serious environmental hazard.
Methyl Parathion
Methyl parathion is a broad spectrum organophosphate insecticide charac-
terized by high toxicity and a low level of persistence. Rapid disappear-
ance in soil limits methyl parathion's insecticidal control to those pests
attacking the foliar portions of the crop, while rapid disappearance from
treated plant surfaces necessitates repeated applications throughout the
growing season. Eighty-four percent of the methyl parathion used in U.S.
agriculture in 1971 was applied to cotton (Table 9-1) where it replaced the
DDT formerly used in combination with toxaphene for the control of the boll
weevil, and the bo11worm complex.
Methyl Parathion: Runoff and Leaching--
The potential for runoff losses of methyl parathion from agricultural
lands is limited by factors occurring both before and after the material
reaches the soil environment. Substantial drift losses and volatilization
and degradation from plant and soil surfaces reduce the amount of material
available for runoff, while physical characteristics of the material limit
its mobility in and from soil. Sheets et_ aL (1972) studied the runoff of
methyl parathion from cotton in Rocky Mount and Lewiston, North Carolina, over
a six-month period and found that losses ranged from 0.008 to 0.25 percent of
the 13.4 kg/ha applied to the four plots studied.
Beyerlein and Donigian (Appendix F) have done the only modelling study
on the runoff of methyl parathion from cotton agricultural lands. The model
estimated losses of 4.6 percent of the material applied, with 4.2 percent of
the losses occurring in the dissolved phase and 0.4 percent occurring in the
solid phase of the runoff. These predictions are an order of magnitude higher
than those measured in the empirical study mentioned above, and unfortunately
could not be validated because of the lack of field data. The higher runoff
noted in this study may be partially explained by the exclusion of drift
losses and volatilization losses except as a function of degradation.
239
-------
TABLE 9-8. PESTICIDE LOSSES AS A PERCENTAGE OF THE MATERIAL APPLIED
Atrazine
Chemical
Class s-Triazine
Type Application broadcast or
band applied by
ground or air
equipment
to Application 2.2 - 4.4
g Rate
(kg AI /hectare)
Drift losses negligible to
40%*
Volatilization 11%
losses
Runoff losses average 5%
(see Table
9-9)
Paraquat
Bipyridylium
spray application
by ground or air
equipment
.27 - 1.1
negligible to 40I1
insignificant
1.28 - 22%10
Carbofuran
Carbamate
soil treatment by ground
equipment
Broadcast application of
granules and foliar spray
treatment by air or ground
equipment
1.68 - 3.36
negligible (soil treat-
ment) to 40% (foliar spray)
insignificant
0.5 - 2.0%11
-------
TABLE 9-8. PESTICIDE LOSSES AS A PERCENTAGE OF THE MATERIAL APPLIED (Continued)
Chemical
Class
Methyl
Parathion
Organophosphate
Toxaphene
Organochlorine
Type Application
foliar application by
ground or air spray
equipment
foliar application by
ground or air spray
equipment
Application
Rate
(kg Al/hectare)
0.13 - 3.36
(3-10 applications/
season)
1.1 - 3.36
(1-12 applications/
season)
Drift Losses
50 - 60%'
50%
Volatilization
33%
24%"
Runoff Losses
0.05 - 4.1
,12
0.36%
13
-------
FOOTNOTES FOR TABLE 9-8.
von Rumker et^ a^. (1975). A general percentage for all herbicides based
on application procedures.
negligible = broadcast surface application of spray by ground equip-
ment.
40% = broadcast surface application of spray by air equipment
based on estimated losses from U.S. corn crop. Baker
and Johnson (1977) found that atrazine had 48-51% drift
losses using ground equipment under windy conditions
(windspeed 20 km/hour).
von Rumker et al. (1975).
negligible = broadcast application of granules by ground equipment.
40% = foliar application of low-volume spray by air equipment.
From table on likelihood of pesticide drift during crop
treatment in agriculture by method of application.
Adair et^ al^. (1971). Test of aerial application at an altitude of 1.5
meters. 40-50% of applied pesticide found on ground in area which in-
cluded the target field plus the area 800 meters downwind from the field.
4
Ware et_ aJL (1970). Studied drift of aerially applied toxaphene with both
spray and dust formulations at wind speed of 4.8-6.4 km/hour. Actual
amount of applied toxaphene reaching plants and the ground was 47.7% for
the spray applications, 14% for the dust formulation.
Foy (1963). 11% evaporated from the soil surface in two days at 40°C and
an air velocity of 2 liters/minute.
6 Calderbank and Slade (1976).
von Rumker e?t_ al. (1974).
Q
Haque and Freed (1974), estimated volatilization losses from a loam soil
at 25°C under annual rainfall of 150 cm = 7 - 14 kg/ha-year or more. Here
we assume application rate 3.36 kg/ha, 10 times/season = 33.6 kg/ha-year,
volatilization average = 11 kg/ha-year.
242
-------
FOOTNOTES FOR TABLE 9-8 (Continued)
9
Nash et JLK (1977). Volatilization losses estimated in agroecosystem
chamber where toxaphene was applied at a rate of 2.0 - 2.7 kg/ha to
cotton plants at weekly intervals for 6 weeks.
Baker et^ al . (1979). Found runoff losses 1.28 to 3.34%. Smith ert aJ_.
(in press) found runoff losses as high as 22% on sediment.
Caro et_ al_. (1973). Study on two watersheds planted with maize.
Amount Applied % of Applied
Year kg Al/ha Type Application Carbofuran Lost
1971 5.31 broadcast and disced into 0.9
7.6 cm depth
1971 4.08 band application - in furrow 0.5
treatment
1972 3.05 band application 1.9
12
von Rumker et al. (1975) . Estimated typical runoff losses of methyl
parathion to~~be~~0.05% of the applied material. Beyerlein and Donigian
(Appendix F) , in a 10-year simulation study based on data from Georgia,
found losses of 4.8% in sediment and water runoff.
13 Bradley £t al_. (1972). In a 6-month study on DDT and toxaphene runoff
from cottoiTplots found toxaphene runoff losses of 0.36%, 75% of which
was associated with the sediment fraction.
243
-------
Furthermore, Beyerlein and Donigian included a rather long half-life of
45 days for methyl parathion located in soil; half-life ranges of three to
11 days have been given for methyl parathion by other sources (USEPA,
1975b). The inclusion of this extended half-life in the model will allow
for prolonged periods of methyl parathion's exposure to raindrop splash
and surface runoff, and will inevitably produce higher estimates of runoff
losses. Therefore, given these limitations and the lack of field data for
model validation, it seems reasonable to assume that the empirically ob-
served losses of less than 0.25 percent more accurately approximate typical
runoff losses of methyl parathion from cotton.
Very little is known about the leaching losses of methyl parathion from
agricultural soils, although one can surmise that losses are limited by
the same factors which limit runoff losses. Li and Fleck (1972) have re-
ported the presence of small amounts of the material (in the parts per
trillion range) in surface and subsurface drain effluents. Haque and Freed
(1974) have estimated that leaching will move methyl parathion less than
20 centimeters through the soil profile under an annual rainfall of 150
centimeters.
Methyl Parathion: Air-Borne Losses—
Substantial air-borne losses of methyl parathion from target areas
may occur both during and after application procedures (Table 9-8). Haque
and Freed (1974) estimated annual volatilization losses of 7-14 kg/ha or
more from a silt loam soil under an annual rainfall of 150 centimeters.
Given an application rate of 3.3 kg/ha repeated ten times throughout the
growing season, volatilization losses of methyl parathion would comprise
approximately 33 percent of the applied material. Volatilization losses
from plant surfaces probably occur at a more rapid rate, and can be so large
as to reduce methyl parathion's half-life to one half hour in cotton grow-
ing regions of Mississippi (Quinby et^ al. 1958). Drift losses of methyl
parathion are also quite large as much of this material is sprayed on the
crop by air equipment. In one test of drift from an aerial application at
an altitude of 1.52 meters, only 40 to 50 percent of the applied methyl
parathion was found in a sample area which included the target field and the
area 800 meters downwind from the crop (Adair et_.al. 1971). von Rumker et al.
(1975) have reviewed the literature and predict~~siinilar drift losses for
any insecticide applied by aerial equipment.
The environmental significance of methyl parathion's air-borne losses
is not known, although there are indications that drift into areas of human
habitation could have insidious effects on public health (Gershon and Shaw,
1961). Residues of methyl parathion found in air samples taken oven Alabama,
Florida, and Mississippi (Stanley et_ aJL 1971; Arthur et^ al_. 1976) prove
that such air-borne losses do not simply disappear but persist in the atmos-
phere until they are degraded or redeposited on the earth's surface with
dust particles or rainfall. Studies establishing the concentrations of
methyl parathion in rainfall over agricultural areas would help determine
whether the redepositing of this material contributes significant amounts
of methyl parathion to aquatic and terrestrial ecosystems.
244
-------
Methyl Parathion: Toxicity--
Methyl parathion's extreme toxicity to aquatic insects, and moderate
toxicity to fish and lower aquatic fauna (Table 9-5) have allowed its success-
ful use for mosquito control in surface waters (USEPA. 1975b) Apperson et
al. (1976) reported no lasting effects on zooplankton when methyl parathion"
was applied to Clear Lake for gnat control, and concluded that, in this case,
direct applications of the material to the lake "achieve no perceivable per-
manent alteration of the lake ecosystem." However. Eisler (1970) has found
that the toxicity of methyl parathion to fish may be increased in aquatic
systems having increased oxygen levels. Holden (1973) has further shown that
sublethal concentrations of methyl parathion affect reproduction by inducing
abortion in fish.
Effects on terrestrial non-target species appear to be of a more severe
nature, von Rumker ejt aJ_. (1974) report high levels of toxicity to soil fauna,
although soil microflora may actually benefit from lower dosages of methyl
parathion because of their ability to use the insecticide as an energy source
(Naumann, 1970). Methyl parathion is highly toxic to insect predators and
parasites (Wilkison et_ aJ^. 1975; Edwards and Thompson, 1973) and has been re-
sponsible for numerous honey bee kills. Honey bee mortality has been espe-
cially high in those cases where methyl parathion was applied in a micro-
encapsulated form to increase its residual activity (Burgett and Fisher,
1977). Methyl parathion is highly toxic to mammals and birds, and has caused
bird kills in treated areas (U.S. Dept. of Interior, 1966; Bejer-Peterson et_
al^. 1972) .
Mammals exposed to sub-lethal doses of methyl parathion over a period
of time will suffer from symptoms occurring as a result of progressive inhi-
bition of cholinesterase (von Rumker £t a^. 1974). Gershon and Shaw (1961)
studied the case histories of 16 people suffering from psychiatric dis-
orders who had also been repeatedly exposed to organophosphate materials over
periods of 1 1/2 to 10 years. In a follow-up of four cases, the authors
noted that severe impairment of memory, difficulty in concentration and re-
actions of a schizophrenic and depressive nature persisted for six months
after the cessation of exposure to organophosphate insecticides. Normal
behavior was observed 12 months after exposure ceased.
The potential for biomagnification of methyl parathion is lessened by
its low fat solubility, high toxicity, and rapid degradation in plants,
animals, microorganisms, and aquatic systems (Sanborn et_ al_. 1977; von Rumker
£t a^. 1974; USEPA, 1975b) . However, residues have been found in sea animals
(USETA, 1975b) and Apperson ejt al_. (1976) have suggested a further investiga-
tion of the possible bioaccumulation of methyl parathion in bluegill and
sunfish.
Methyl Parathion: Persistence--
Methyl parathion, like other organophosphates, is considered to be a
relatively nonpersistent insecticide, degradable by nonbiological factors and
biological organisms (Table 9-6). Though often more toxic than the parent
compound, the by-products of these degradation pathways are not considered to
be of great environmental importance as extreme instability causes their
245
-------
rapid disappearance from terrestrial and aquatic ecosystems (von Rumker et al.
1974). Methyl parathion has a short residual life on treated plants. In mid-
summer, applied concentrations of 106 ppm may fall to 3.9 ppm 96 hours after
application (Ware et_ al_. 1972). None of the studies conducted on methyl
parathion1s disappearance from treated plants stated whether the believed
mechanism of dissipation was by degradation or volatilization. Degradation
of the material entering the plant does not appear to be as rapid, for
methyl parathion is commonly found in food and feed samples examined for
residues (Duggan and Duggan, 1973).
The supposed nonpersistence of methyl parathion in aquatic systems has
led some researchers to dismiss the possibility of serious contamination
being caused by this material. However, a review of the literature indicates
that methyl parathion residues are moderately persistent in aquatic ecosystems.
Apperson et^ al. (1976) in their study on Clear Lake found that there was a
carry-over of methyl parathion residues between insecticide treatments
occurring at intervals of 20 days. The absence of residues in sediment
samples suggested that degradation took place before or upon penetration of
the bottom mud. Eichelburger and Lichtenberg (1971) found that methyl para-
thion degraded completely in bottled river water within eight weeks, and
data from the USEPA (1975b) show persistence rates of less than one month in
soil-water and less than four months in lake water.
Methyl parathion's degradation in soils generally occurs at a rapid
rate, though the process can be slower when the material is applied at heavier
dosages and in those areas where soils are alkaline or contain high propor-
tions of dry sand or clay (Nayshteyn et_ al_. 1973; Sanborn et^ al_. 1977).
Lichtenstein and Shulz (1964) applied 5.6 kg/ha of methyl parathion to a
Carrington silt loam and found that only eight percent of the insecticide
remained 15 days after treatment. Studies by King and McCarty (1968)
estimating the theoretical half-life have shown similar results. Methyl
parathion's persistence in soil has been reassessed in recent persistence
studies using ^^carbon labelled insecticides (Katan et_ al. 1976; Lichtenstein
et^ a_K 1977). In 1977, Lichtenstein et^ al_. published data showing that only
seven percent of the applied methyl parathion was extractable 28 days after
soil treatment, with an additional 43 percent existing in the form of non-
extractable bound residues.
Questions raised about the nature and potential toxicity of these bound
residues resulted in a study by Fuhremann and Lichtenstein (1978) which inves-
tigated the release and potential pickup of the residues by earthworms and
oat plants. Data showed that bound 14carkon residues of methyl parathion were
incorporated into the tissues of earthworms and oat plants exposed to the
soil-sand-bound residue mixture. In the case of the earthworms, 58 to 66
percent of the adsorbed ^carbon residue was converted to a strongly bound
form within the bodies of the worms. The remainder of the 14carbon residues
was converted to a soluble form. In the case of the oat plants, 82 to 95
percent of the 14carbon residues were extractable as benzene or water soluble
J-4Carbon compounds. The authors concluded that bound residues are not exclu-
ded from environmental interactions, and suggested that further research be
done to determine the importance of soil-bound residues in biological systems.
246
-------
Methyl Parathion: The Effectiveness of SWCPs in Reducing Transport--
Methyl parathion's small runoff losses coupled with its relatively
rapid degradation in water and moderate toxicity to aquatic organisms indicate
that only minimal benefits will be gained if SWCPs are implemented to control
pollution resulting from the use of methyl parathion. The only results which
might temper these conclusions are given by Beyerlein and Donigian. in their
simulation study on methyl parathion (Appendix F). Their model predicts that
methyl parathion's losses in runoff would comprise 4.6 percent, 3.9 percent,
3.8 percent, and 2.3 percent of the amount originally applied under conven-
tional tillage, minimum tillage, contours, and contours with terraces,
respectively. Thus, according to this model, runoff losses of methyl para-
thion could be reduced by 50 percent using contours with terraces. If one
uses the larger runoff losses obtained by Beyerlein and Donigian, 50 per-
cent reductions in runoff will prevent 156 g/ha from leaving cotton fields;
if one uses the field results of Sheets et^ aL (1972), 50 percent reduc-
tions will only prevent 0.55 to 16.5 g/ha from leaving cotton fields. As was
pointed out earlier, Beyerlein and Donigian did not include air-borne losses
and were not able to validate their model with field data. Without additional
empirical information to substantiate these much higher runoff losses, it is
difficult to build a strong case for the benefits of SWCPs for controlling
methyl parathion's runoff losses.
Methyl Parathion: Increasing Efficiency in Use--
The substantial drift losses of methyl parathion ( 50 to 60 per-
cent) noted in Table 9-8 indicate the need for increasing the accuracy of
application procedures. Available information indicates that these drift
losses could be substantially reduced by switching from aerial application to
application with ground equipment. Although no field studies have been done
on drift losses of methyl parathion with ground applications, a general report
on pesticide drift losses indicates that drift losses can be reduced from an
average of 40 percent of the material to five percent if foliar applica-
tions are made with ground equipment instead of aerial equipment (von Rumker
et^ al. 1975). Ground applications thus have the advantage of increasing the
accuracy of pesticide placement within the target area. In Texas, cotton
field entomologist Ray Frisbie reported that ground applications cost about
the same as aerial applications (within +_ $2.50/hectare) and that, in most
cases, the advantages of increased accuracy obtained with ground applications
outweigh the advantages of aerial applications (personal communication, 1979).
However, in some areas such a substitution of application procedures
may be difficult or impossible to implement as aerial applications have the
advantage of timeliness. When and if insect infestations are noted in a
field, aerial equipment can be used to apply pesticides to large acreages in
a short period of time. Ground applications of pesticides take more time
and are hindered by muddy ground conditions resulting from heavy rain or
irrigation. When and if such conditions make it impossible to substitute
ground equipment for aerial application, it may be possible to reduce drift
losses to hon-target areas by making changes in the aerial equipment itself.
Indeed, Ware et al. (1975) have shown that the use of Raindrop M nozzles can
reduce off-target drift by 50 percent when compared to type D flooding
nozzles.
In the case of methyl parathion, which is used primarily in cotton
247
-------
production [Table 9-1), integrated pest management programs represent a more
comprehensive approach to increased efficiency of use and reduced losses to
non-target areas. The survey by Pimentel et_ aL (in press) discussed in
Table 9-7 indicates that if IPM techniques were implemented in all cotton
producing regions where they are cost-effective (i_.e_., where IPM causes no
net reduction in farm income), the national use of organophosphate insecti-
cides in cotton production would be reduced by 46 percent. Casey et
al. (1975) have studied the impact of cotton IPM programs on the use of methyl
parathion and other insecticides in Texas, and reported that methyl parathion
use dropped from 7.49 kg/ha to 0.24 kg/ha in the Trinity River region. The
reduction in use was not as substantial in the Brazos River region (10.55
kg/ha to 5.38 kg/ha) but included 4.57 kg/ha used for a fall diapause program.
Casey et al. estimated that if the new pest management strategy was imple-
mented~Tor~all cotton hectares in the Trinity River and Brazos River regions,
methyl parathion's annual use would be reduced from 849,724 kg to 355,678 kg.
Such substantial reductions in pesticide use will inevitably reduce the
amounts of methyl parathion reaching non-target areas via air-borne, runoff,
and leaching losses. Within agricultural areas, reduced methyl parathion use
will cause increased survival of beneficial insects, and fewer opportunities
for improper disposal of pesticide containers and tank rinsate. In addition
to the economic and environmental benefits gained by reductions in methyl
parathion's use, Lacewell et_ aL (1976) have found that the cotton production
system studied by Casey ert al. (1975) required 56 percent less energy
(kcal) to produce a kilogram of lint.
Methyl Parathion: Summary and Conclusions—
A review of the literature indicates that the potential for reducing
transport of methyl parathion is much greater with methods which increase
efficiency of use than with SWCPs. Empirical studies indicate that less than
0.26 percent of the methyl parathion applied to cotton leaves the field in
runoff. Terraces with contours appear to be the most effective SWCP for the
control of this runoff and reduce losses from 0.2 percent of the material
applied to 0.13 percent. Drift of methyl parathion with commonly used aerial
application procedures may produce pesticide losses to non-target areas which
are over a hundred times larger than losses in runoff. These drift losses
can be reduced substantially by switching from aerial to ground applications
of the material. Information on the changes in costs and pest control effec-
tiveness associated with a switch to ground applications is not available for
most cotton producing regions. However, pest control experts in the cotton
growing regions of Texas have indicated that such a change is not only feas-
ible economically but is also desirable in terms of improving the effective-
ness of control practices.
It has been argued that the methyl parathion carried in drift is less
environmentally hazardous than the methyl parathion carried in runoff and
that runoff control is therefore more important than drift control. A review
of the literature does not support this conjecture. Methyl parathion is
generally more toxic to terrestrial organisms and may have insidious effects
on human mental health. Toxic effects will also be extended to aquatic
species as a certain percent of the drifting material will settle on lakes,
248
-------
rivers, streams and drainage ditches. The lower toxicity of methyl parathion
to aquatic species is well illustrated by the fact that the material has
been registered for application directly to surface waters for mosquito con-
trol.
The most promising strategy currently available for reducing the trans-
port of methyl parathion away from target areas involves improvements in the
timing of spray operations and the use of biological and cultural control
methods to supplement the pest control obtained with pesticides. Integrated
pest management programs in cotton have reduced methyl parathion use by
approximately 40 percent at no net cost to the farmer and have resulted in
substantial reductions in energy use. This reduction in use would be expec-
ted to result in similar reductions in runoff, drift, and volatilization
losses. Hence the impact of IPM programs on runoff losses is considerably
greater than the impact of the less expensive SWCPs such as conservation
tillage and contour plowing. The more expensive but most effective SWCP,
terraces with contours, will reduce runoff losses as much as IPM programs,
but it does not have the added benefits of reduced energy use, reduced drift
and volatilization losses, and reduced hazards to farmworkers and non-target
species living in agricultural areas.
SWCPs which have been implemented for erosion control can have a small
beneficial effect in reducing the runoff of methyl parathion. As long as
SWCPs do not require an increased rate of methyl parathion application, it
does not appear that the implementation of SWCPs will increase the movement
of methyl parathion into the environment. In addition, most SWCPs can be
used in conjunction with other methods which increase the efficiency of pesti-
cide use. However, if a decision must be made regarding the allocation of
limited resources specifically for the reduction of methyl parathion pollu-
tion, currently available information indicates that SWCP should be given a
lower priority than increasing the efficiency of pesticide use by wider imple-
mentation of IPM and drift control programs.
Toxaphene
Pollock and Kilgore (1978), in their extensive review of the literature
concerning toxaphene, have stated that "toxaphene remains the most heavily
used and perhaps, the least understood insecticide presently available ."
Although the technical product has been shown to contain nearly 200
polychlorinated camphenes (Holmstead _ejt al^. 1974), its chemical composition
remains largely undetermined. The lack of information about the metabolites
and photoproducts of the numerous compounds making up toxaphene makes it
difficult to ascertain the environmental hazards associated with its use.
Toxaphene: Runoff and Leaching--
Runoff losses of toxaphene apparently can occur but are limited by
drift and volatilization losses which occur before the material is exposed
to raindrop splash and surface runoff. The potential for runoff losses is
further reduced by toxaphene?s low solubility in water and strong attraction
for soil particles. These characteristics place toxaphene in the solid phase
of runoff and thus limit its transport to the small amounts of sediment found
249
-------
in runoff leaving agricultural fields.
Field studies on losses of toxaphene in runoff support these conjectures
and report losses which make up only a small percent of the applied material.
Bradley et_ al. (1972) studied the runoff of toxaphene from cotton in North
Carolina~~and found that losses ranged from 0.078 to 0.72 percent of the 26.9
kg/ha applied to the eight plots under observation. Of this material, 75
percent was associated with the solid phase, with the remainder found in the
dissolved phase of the runoff. Willis et^ al. (1976) have obtained similar
results in their study of toxaphene runoff from larger cotton plots in Miss-
issippi.
While these losses appear to be small when seen as a percentage of the
material applied, they take on greater significance when it is observed that
0.078 to 0.72 percent of the 26.9 kg/ha applied results in losses of 21 to
200 g/ha of toxaphene. Given large tracts of cotton acreage, these losses
could result in serious contamination problems in surface waters located in
cotton growing areas. Bradley et_ al. (1972) did find that significant con-
centrations of toxaphene appeared in a 0.2 hectare pond located in the ex-
perimental watershed. Concentrations increased from levels of less than 1
ppb before the cotton was treated to 65 ppb at midseason, and equaled or
exceeded the 96 hour median tolerance for bluegill on all but one occasion
during the growing season. The authors attributed this toxaphene pollution
to runoff but did not study possible residue inputs from drift losses and
contaminated rainfall.
Leaching losses of toxaphene through agricultural soils are limited by
the same factors limiting runoff losses. Haque and Freed (1974) estimated
from the best available information that toxaphene will move less than 10 cm
through the soil profile with an annual rainfall of 150 centimeters. Field
studies have indicated a slightly higher propensity for leaching. Swoboda
et^ al. (1971) found that after ten years, 90 to 95 percent of the toxaphene
recovered from a Houston black clay was concentrated in the top 30 centimeters
of the soil. LaFleur et_ al_. (1973) applied 100 kg/ha of toxaphene directly
to the soil surface to study the leaching behavior of "sterilant concentra-
tions" of the material. In two months, toxaphene leached through approxi-
mately 300 cm of soil, and entered and contaminated groundwater for the
entire year of observation. The direct application to soil, and the high
rates of toxaphene used in the experiment are dissimilar to conditions of
toxaphene use in the field, but may indicate the leaching hazards associ-
ated with spills or disposal of the material.
Toxaphene: Air-borne Losses--
Applications of toxaphene are predominantly made by aerial equipment,
and can thus result in substantial drift losses to non-target areas (Table
9-8). Ware et_ al^. (1970) studied drift losses of aerially applied toxaphene
in windspeeds of 5.6 km/hr and found that only 47 percent of the spray-
formulated material and 14 percent of the dust-formulated material were
deposited on cotton plants. Gerhadt and Witt (1965) found similar increases
in drift losses with dust formulations and reported that aerial applications
of dust-formulated toxaphene produced downwind drift deposits which were six
250
-------
times greater than those obtained with spray formulations. Dust formu-
lations of toxaphene have been largely abandoned because of associated
drift losses.
Losses of toxaphene to the atmosphere continue to occur after the
material has reached the soil surface and treated crops. Nash et al.
(1977) found that 24 percent of the applied toxaphene volatilized~~from
soil and cotton plant surfaces in a model ecosystem. The authors sug-
gested that volatilization losses in the field might be larger because
cotton plants are more widely spaced, and thus allow more of the vola-
tilization losses from the soil to escape to the atmosphere. Swoboda
ejt al. (1971) observed the rapid disappearance of toxaphene from a heavy
clay soil and attributed it to large volatilization losses encouraged by
daily soil temperatures of 60°C. Haque and Freed (1974) used available
information to quantify annual vaporization losses of pesticides from
soil, and estimated that 7 to 14 kg/ha of toxaphene would vaporize from
a loam soil at 25°C.
The ecological importance of these atmospheric pollutants has not
been ascertained, although air-borne toxaphene residues have been shown
to have a half-life of approximately 15.1 days (Nash et al. 1977). Res-
idue levels of 280, 170, 44, 86, and 220 ppt in five out of eight rainwater
samples collected in Maryland indicate that some of the air-borne toxaphene
is redeposited on the earth's surface (Munson, 1976). As mentioned pre-
viously, Goldberg et^ al. (1971) have suggested that aerial fallout such
as this may greatly exceed river discharge in its contributions of pesticides
to the oceans of the world. Toxaphene in rainfall and aerial fallout may
also contribute to the agricultural soils. Further studies should be
undertaken to determine how much toxaphene is actually being deposited
by rainfall and aerial transport on aquatic and terrestrial ecosystems.
Toxaphene: Toxicity--
In general, there seems to be a serious lack of data concerning the
effects of toxaphene on aquatic flora and fauna. In aquatic systems,
isolation of the toxic component, or toxic group of components, is compli-
cated by a lack of information concerning changes in the components' rela-
tive concentrations over time as a result of chemical reaction, microbial
degradation, and preferential sorption by flora and fauna (Veith and Lee,
1971). However, whatever the mechanism or components of toxicity may be,
toxaphene in toto is extremely toxic to lower aquatic organisms, having
inhibitory effects on algae, plankton, and protozoa at concentrations as
low as 0.1 ppm (Thompson, 1973; Butler, 1977; Sanborn et a],. 1977). Small-
er concentrations (1 ppb), while not toxic themselves, will have synergistic
sublethal effects on oysters when in the presence of 1 ppb of both DDT
and parathion (Lowe e£ al. 1970).
Although some fish species have developed resistance, toxaphene is
in most cases extremely toxic to fish (Table 9-5), causing mortality at con-
centrations in parts per billion (Sanbom et al. 1977). The hazards posed
by this degree of toxicity are prolonged by persistence rates of up to 12
months in water (Edwards, 1977) and have been responsible for numerous fish
251
-------
kills, and a general decline in the diversity of aquatic ecosystems located
in cotton producing areas (Johnson, 1968; Grzenda et^ a\_. 1964; Keiser et al.
1973; Reimold, 1974; Reimold e* a.^. 1973). Mehrle and Mayer (1977) have
reviewed the literature concerning the sublethal effects of pesticides on
fish and found that toxaphene inhibits collagen synthesis thereby depressing
bone development in young fish. Other studies have indicated that sublethal
concentrations of toxaphene may cause behavioral changes which jeopardize the
survival of fish by affecting their ability to learn or react properly to
external stimuli (Holden, 1973; Warner £t al_. 1966).
Less is known about the effects of toxaphene on terrestrial organisms.
Martin e_t aL. (1959) have shown that high levels of agricultural use of toxa-
phene can inhibit bacteria and fungi populations in soil. Normal dosages of
toxaphene are highly toxic to insects including insect parasitoids and
predators (Wilkison et^ aJL 1975); however in some areas, target insects have
developed resistance. Harris (1972) noted that the LD5Q for a toxaphene-DDT
mixture (2:1) varied throughout the state of Mississippi and ranged from 10
to 21 yg per larvae for the bollworm, and from 34 to 1000 yg per larvae for
the budworm. Toxaphene is of intermediate toxicity to birds and mammals
(Pollack and Kilgore, 1978) but has been responsible for two bird kills re-
ported by Rudd (1966) and Keith (1966). While it has been reported that
mammals metabolize most of the discovered components and that birds do not
seem to accumulate residues in their fat tissue, not all of toxaphene's fat
soluble metabolites have been identified, and analytical standards have not
been available for those components which have been identified (Casida et_ al.
1974; Holden, 1973).
Toxaphene is similar to other organochlorines in that it is both
stable and highly fat soluble, two conditions conducive to biomagnification.
A number of studies have indicated that biomagnification of toxaphene does
occur, although generally at lower concentrations, and with lower persistence
than most other organochlorine insecticides (Guyer et^ al. 1971) . Sanborn et_
al.(1976) measured ^carbon toxaphene biomagnification in an aquatic-
terrestrial model ecosystem, and found that concentrations in algae, snails,
mosquito, and fish were 6902, 9600, 890, and 4247 times higher than the con-
centration found in the water, respectively.
Toxaphene: Persistence--
Like most organochlorines, toxaphene has a degradation rate measured in
years (Table 9-6) rather than the weeks and months which describe the degra-
dation rates of other pesticides. Although it is known that toxaphene is
degraded by biological as well as non-biological factors (von Rumker et al.
1974), analytical problems arising from its complex structure complicaTe~any
assay of degradation rates and pathways. Toxaphene's persistence on plants
is greatly reduced by volatilization losses to the atmosphere. The effect of
wind on volatilization losses has been shown by Mistric and Gaines (1953) who
found that toxaphene residues had 85 percent reductions in effective-
ness when subjected to a simulated wind, with nine percent reductions seen in
the absence of the wind. When sprayed on kale, most of the insecticide
disappeared within three days without precipitation (Klein and Link, 1967).
The toxaphene which escapes volatilization through adsorption into plant
252
-------
tissues, persists at low levels as a contaminant in cotton and various root
crops (Nash, 1974).
Toxaphene's strong adherence to cotton leaves, and large volatilization
losses from plant and soil surfaces allow only small amounts of the insecti-
cide to reach or remain on the soil surface. The degradation of remaining
residues is generally quite slow, but depends on the pH, temperature, and
organic matter content of soils (Fowkes et_ al^, 1960; Weed and Weber, 1974).
Nash and Harris (1973) studied the persistence of toxaphene on a Congaree
sandy loam, and found that 49 percent of the material remained 16 years after
application. Randolph et_ al. (1.960) found that 50 percent of the applied
material degraded in one year in soil conditions more conducive to degradation
processes.
Studies on toxaphene's behavior in surface waters report instances of
rapid disappearance on the one hand and persistence of up to nine years on
the other (Guyer ejt al. 1971) . Some of these discrepancies may be due to
toxaphene's chemical complexity and reversible sorption to aquatic flora and
fauna (Pollock and Kilgore, 1978). Williams and Bidleman (1978) studied the
degradation of toxaphene in an anaerobic salt marsh and found that residues
were quickly degraded in sterile and non-sterile sediment to compounds having
gas chromatographic retention times shorter than those of standard toxaphene
components. Lee £t_ al. (1977) found "weathered" toxaphene components to be
less toxic to bluegills and concluded that some of the more toxic components
had indeed degraded.
Several mechanisms for the "disappearance" of toxaphene from aquatic
systems have been suggested. MacKay and Wolkoff (1973) predicted high evapor-
ation rates and resultant short half-lives in water for the organochlorines
because of their high activity coefficients while in solution and concluded
that "evaporation from rivers and lakes may thus represent a major mechanism
of the transport of pollutants through the environment." Hughes and Lee
(1968) noted the great adsorption capacity of suspended sediments and sug-
gested that sorption activities might play a significant role in the removal
of toxaphene from natural water systems. A further evaluation of the de-
toxification role of lake sediments was made by Veith and Lee (1971) who
found that toxaphene residue concentrations in the top five centimeters of
the sediment increased for 190 days after application and then decreased by
a factor of two every four months. Leaching of residues through the aquatic
sediment did not appear to contribute to this disappearance as residues were
not found at levels lower than 20 centimeters. This discovery of the
detoxification role of sediments cast light on several older studies which
showed that detoxification occurred with greater ease in well-mixed, shallow
lakes (Terriere et al_. 1966) and in surface waters containing large amounts of
suspended sediments (Stringer and McMynn, 1958).
Veith and Lee (1971) have shown that the adsorption of toxaphene
residues to suspended and settled sediment is irreversible. However, Reimold
and Durant (1974) have shown that these apparently irreversibly adsorbed
residues are not excluded from biological interactions. Their study of the
pesticide residue levels resulting from the dredging of a toxaphene contami-
nated stream showed concentrations of 200 ppm in mummichog (Endulus
253
-------
heteroclitus) and 7.5 ppm in salt marsh cordgrass. The toxaphene found in
the marsh grass was apparently translocated by the plant from contaminated
sediments into plant tissue.
Toxaphene's persistent nature and heavy use in agriculture would lead
one to expect numerous reported incidents of residues in plants, animals,
soil and water; however this generally has not been the case. Toxaphene is
rarely found in water samples (Pollack and Kilgore, 1978) although a few
incidents of pollution have been reported. Barthel et_ al.(1969) have re-
ported that toxaphene is a contaminant of the lower Mississippi and its
tributaries. Nicholson et al. (1964) found toxaphene in all samples taken
from a stream draining a 1000 square kilometer watershed which con-
tained approximately 6060 hectares of cotton, and they reported toxaphene
residues in all water samples taken before and after treatment by a water-
treatment plant. Traces of two heptachloronorbornenes, compounds which could
be degradation products of toxaphene, have been found in New Orleans drinking
water at levels of 0.04 to 0.06 ppb (USEPA, 1974). Toxaphene has also been
found in rainwater samples in Maryland (Munson, 1976) and in air samples
taken over Mississippi (Stanley et_ a^U 1971; Arthur et_ a_l_, 1976) and areas
as far away as Bermuda (Bidleman and Olney, 1975).
The 1971 National Soil Monitoring program sampled soils in 37
states and found toxaphene residues at only 6.2 percent of the sites sampled
(Carey et^ al. 1974). However, the national results of this survey are some-
what deceiving since cotton-growing states such as South Carolina, Louisiana,
and Mississippi had toxaphene residues in 76.5 percent, 30.8 percent and 71
percent, respectively, of the soil samples tested. Analytical techniques
insensitive to toxaphene concentrations lower than 10 ppm undoubtedly also
contributed to the small percent of toxaphene contamination reported in the
national results.
Pollock and Kilgore (1978) reviewed the small amount of data that have
been gathered on toxaphene's persistence and residues and concluded that
the paucity of pollution reports resulted from inadequate analytical tech-
niques and potential interference from other organochlorine insecticides
present in monitoring samples. Pollock and Kilgore (1978) predicted greater
success in uncovering toxaphene residues in future years because of better
analytical methods, greater awareness of chemical analysts, and the declin-
ing use of other organochlorines whose presence interfered with attempts to
monitor toxaphene.
Toxaphene: The Effectiveness of SWCPs in Reducing Transport--
It is quite difficult to evaluate the effectiveness of SWCPs in reduc-
ing toxaphene losses to non-target areas. On the one hand, persistent
toxaphene residues seem to be ubiquitous in the soils of cotton-growing
areas and thus have the potential for entering surface waters via runoff.
On the other hand, runoff losses are dwarfed by drift and volatilization
losses occurring during and after the application of toxaphene to the crop.
The contributions of these air-borne losses to surface water pollution are
not known, thus SWCPs could be implemented only to find that a good deal of
the contamination resulted from the fallout of air-borne losses. Furthermore,
254
-------
Lichtenstein et al. (1971) found that the organochlorines, aldrin and hepta-
chlor and their metabolites persisted for longer periods in fields with con-
tinuous crop cover than in fields which were annually cultivated. Thus, it
is possible that SWCPs involving reduced cultivation may similarly prolong
toxaphene's persistence in soil by limiting volatilization losses and degra-
dation processes. This prolonged persistence could counter the benefits
gained by SWCPs by extending the period of toxaphene's exposure to raindrop
splash and surface runoff. As discussed in Section 7, reduced persistence of
a pesticide may have a greater impact than any SWCP in reducing runoff losses
of pesticides from agricultural lands. Hence, in some cases, prolonged
persistence may counter the reductions in runoff gained by the implementation
of certain non-structural SWCPs.
An evaluation of the effectiveness of SWCPs for reducing toxaphene's
losses is further limited by the absence of studies determining the effects
of various SWCPs on the runoff losses of toxaphene. As toxaphene does travel
primarily with the sediment suspended in runoff (75 percent in the solid
phase), one can extrapolate/estimate the effects of SWCPs by looking at
another strongly adsorbed pesticide, paraquat. Beyerlein and Donigian's model-
ling results (Appendix F) indicate that SWCPs reduce paraquat losses by a
maximum of 52 percent. One would not expect toxaphene losses in runoff
to be reduced by more than 52 percent which amounts to approximately
78 g/ha.
Toxaphene: Increasing Efficiency in Use--
Since the primary use of toxaphene is in combination with methyl para-
thion for the control of cotton insect-pests, programs to increase the
efficiency of its use are similar to those mentioned in the section on methyl
parathion. Drift losses are large as the material is mostly applied by
aerial equipment. Reductions in these losses can be made either by changing
to application by ground equipment or by improving the accuracy and placement
of toxaphene by aerial equipment. As in the case of methyl parathion, inte-
grated pest management programs have allowed farmers to reduce the number of
toxaphene treatments needed for the control of the insect-pest complex of
cotton. Pimentel et^ al. (in press) interviewed numerous entomologists
throughout cotton growing regions in 1974 and found that the use of organ-
ochlorines would have been reduced by 39 percent if IPM techniques available
at that time had been implemented in all areas where they were cost effec-
tive. Casey et_ aL, (1975) monitored the impact of an IPM program on pesti-
cide use in the Brazos River and Trinity River regions of Texas and found
that the traditional annual applications of toxaphene (1.41 kg/ha and 1.91
kg/ha) were entirely eliminated. The authors estimated that if this IPM
program could have been implemented on all cotton hectares in the Trinity
River and Brazos River regions, 133,367 kilograms of toxaphene would have
been eliminated from cotton production. This elimination of toxaphene use
was achieved without increases in the use of other insecticides and without
extra cost-to farmers. In fact, net returns on a per hectare basis increased
by $55.87 in the Brazos River region and by $85.50 in the Trinity River
region. Annual insecticide use on a per hectare basis was reduced from
14.51 kg/ha to 7.19 kg/ha in the Brazos River region and from 12.28 kg/ha to
6.22 kg/ha in the Trinity River Region. Such IPM programs thus provide the
ultimate means of reducing any pollution associated with toxaphene's use in
255
-------
cotton production.
Toxaphene: Summary and Conclusions--
Both air-borne and runoff losses of toxaphene can contribute to the
pollution of surface waters; however, approximately three-fourths of the
toxaphene carried in runoff is adsorbed to suspended sediment and is thus
not immediately available to biological organisms. Air-borne losses settled
on surface water and dissolved toxaphene in runoff are readily available
to biological organisms and thus have an immediate impact on aquatic ecosystems.
Currently available information indicates that the amounts of toxaphene
transported by drift and volatilization average over 60 percent of the
material applied (52% + 0.24 [47%]). Hence air-borne losses of toxaphene
are two orders of magnitude greater than the amounts of toxaphene transported
in runoff. Unfortunately, an evaluation of the relative impact of air-borne
and runoff losses on water quality is not possible because studies have not
been done to measure actual imputs of air-borne and runoff pollutants to
surface waters. However, given the much larger air-borne losses and the fact
that most toxaphene carried with runoff is adsorbed to sediment, it is reason-
able to assume that air-borne losses of toxaphene are at least as hazardous
to aquatic systems as the toxaphene contributed by runoff.
In the absence of field or modelling studies on the impact of SWCPs on
the runoff of toxaphene, evaluations of the effectiveness of SWCPs for the
control of toxaphene pollution must rely on extrapolations from data on the
effectiveness of SWCPs in controlling the runoff of other pesticides from
agricultural fields. Given the results for other related pesticides, it is
unlikely that any SWCP would reduce toxaphene transport in runoff by more
than 50 percent. Empirical studies have generally found no more than 0.5
percent of the applied toxaphene in runoff, thus the maximum impact of SWCPs
on toxaphene's transport would involve a reduction of only 0.25 percent of
the material applied.
The substantial drift losses which occur with the most commonly used
application procedures (53 percent) can be reduced significantly by changes
in the aerial equipment itself or by changes from aerial applications to
ground applications. Little published information exists to indicate whether
in all cotton growing areas such a substitution is economically feasible
or possible in terms of the goals of pest management. However, pest manage-
ment specialists in Texas have indicated that this substitution is not only
feasible economically, but also desirable in that drift losses significantly
reduce the effectiveness of expensive pesticide treatments.
Increasing efficiency of pesticide use through the implementation of
integrated pest management programs is a widely available, inexpensive means
of substantially reducing the amount of toxaphene reaching non-target areas.
The results of a national survey indicated that IPM would reduce organo-
chlorine use by about 39 percent using currently available methods and by
59 percent using methods which will be widely available in the next
five years. Hence, the impact of currently available IPM programs on
toxaphene's runoff losses would probably be almost as great as any of the
SWCPs. The difference in runoff losses between the most effective SWCP
256
-------
studied are available IPM programs is on the order of only 0.05 percent
of the material applied ([0.5-0.4][0.5%]). IPM methods available in the
near future would be expected to be more effective than SWCPs in reducing
runoff losses. IPM programs have additional advantages over SWCPs in that
they have allowed farmers to cut down on crop production costs and energy
needs while reducing toxaphene's drift and volatilization losses. These
reductions are substantial, amounting to approximately 25 percent
of the material applied to cotton. It should be noted that reducing applica-
tion rates is the only feasible way of decreasing the sizeable volatilization
losses of toxaphene.
In conclusion, SWCPs, changes in application procedures, and IPM are
all effective means of reducing toxaphene's transport away from target
areas. However, IPM programs combine reductions in runoff and air-borne
losses with reductions in energy needs and crop production costs and thus
appear to be the most cost effective means of reducing toxaphene contamination
in both target and non-target areas. IPM programs should therefore be given
a higher priority than SWCPs or changes in application procedures when allo-
cating limited resources for the control of toxaphene pollution. However,
given the large amounts of air-borne losses, changes in application procedures
should be encouraged in conjunction with IPM programs where feasible.
Carbofuran
Concern about the environmental hazards associated with the use of
persistent pesticides encouraged the development of several insecticides of
a nonpersistent nature which are now considered to be safer materials in
spite of their greater toxicity to non-target animals. One of these recently
introduced insecticides is carbofuran, a broad spectrum contact and systemic
carbamate which has been especially successful in granular, soil treatments
for the control of corn rootworm. Carbofuran is applied as a foliar spray
to a much lesser degree and can be used to control various insect pests
of alfalfa, corn, peanuts, potatoes, rice, tobacco, peppers, and sugarcane.
Carbofuran: Runoff and Leaching--
Although carbofuran's physical and chemical characteristics indicate
a strong potential for runoff losses, empirical studies have generally re-
ported losses of less than two percent of the applied material. Unusually
large single event losses of up to 14 percent have been reported in
areas with claypan soils and large losses of runoff water. After reviewing
studies on runoff, von Rumker et_ al. (1975) estimated that runoff losses will
comprise no more than one percent of the carbofuran applied to the U.S. corn
crop.
Caro et al, (1973) studied carbofuran in runoff from a silty loam soil with
soil treatment? by broadcast and in-furrow applications and reported losses
of 0.5 to 2.0 percent of the original material. Although the carbofuran
applied as an in-furrow treatment persisted for a longer period of time
(117 days versus 46 days for the broadcast material), runoff losses were
less for a given volume of runoff when in-furrow applications were used.
Runoff losses were transported primarily in the dissolved phase of the runoff
and were higher after the granules had dissolved and released the active
ingredient. Smith et al. (1974) have studied loss of carbofuran
257
-------
from conventionally tilled corn grown on claypan soils and found larger
runoff losses than those reported from silty loam soils. Carbofuran
was applied at a rate of 1.12 kg Al/ha and suffered single event runoff losses
which ranged from zero to 10.73 g/ha in 1971, 32.4 to 56.6 g/ha in 1972, and
zero to 161.5 g/ha in 1973. Maximum single event losses in each year in
terms of a percentage of the material applied consisted of 0.96 percent in
1971, 5.05 percent in 1972, and 14.42 percent in 1973; maximum concentrations
in water ranged from 298 to 600 ppb in the three years of the study.
The FMC Corporation also conducted runoff studies (reported in USEPA,
1976], and found carbofuran residues of 1 ppm in a neighboring pond when
heavy rainfall followed four days after corn treatment with 4.48 kg Al/ha.
This contamination was reduced to "negligible" concentrations by the stime
the next sample was taken 16 days later. As the 96-hour LC5Q for ten
fish species range from 0.08 to 1.18 ppm (Pimentel, 1971), these runoff
losses constitute a serious threat to non-target species.
Leaching losses are limited although carbofuran's physical and chemical
characteristics might lead one to expect otherwise. Leaching studies con-
ducted by the FMC Corporation (USEPA, 1976) in Nebraska and Georgia show
that the carbofuran remained primarily in the top 15 centimeters of
soils rich in clay and organic matter. Rapid degradation in soils will in
most cases limit both runoff and leaching losses of carbofuran from treated
fields.
Carbofuran: Airborne Losses—
The studies thus far conducted on carbofuran's air-borne losses from
agricultural fields indicate that losses are generally quite small, since
the material is primarily applied as granules for soil treatment (Table 9-8).
von Rumker et^ al. (1975) reviewed the data on drift losses and estimated
drift losses to be much larger with foliar spray treatments than with
granular treatments of the soil by air or ground equipment. Drift losses
of foliar spray treatments can range from one to ten percent with application
by a tractor boom sprayer, to 10 to 60 percent with application by an air-
craft boom sprayer; losses with granular treatments are negligible because
most granules are made to be 250 pm or greater in size. It is possible that
larger drift losses occur when granules are applied in windy conditions.
Baker et aJ^. (1977) applied 1.12 kg/ha of fonofos as granules in 20 km/hr
winds on three different occasions and found only 0.84 kg/ha, Q.63 kg/ha, and
0.45 kg/ha of fonofos in the soil immediately after application. Losses to
the atmosphere through volatilization from treated soils and crops are small
or perhaps nonexistent (von Rumker et^ al. 1974). Harris and Miles (1975)
classified carbofuran as a non-volatile insecticide based on measurements
of fumigant activity against insects (Table 9-8).
Carbofuran: Toxicity—
Carbofuran is quite toxic to aquatic organisms at low concentrations.
Sangha (1972) and Yu et_ aL (1974) studied carbofuran's effect on lower
aquatic organisms in a model ecosystem and found that fresh water clams,
fresh water crabs, frogs, snails and water fleas were killed when sorghum
258
-------
plants in the terrestial part of the ecosystem were treated with a normal
application rate of 1.12 kg Al/ha. In spite of carbofuran's observed rapid
hydrolysis in water (Yu et al_. 1974), successful re-establishment of test
organisms did not occur until the twentieth day after carbofuran's appli-
cation. Carbofuran is also quite toxic to fish and produces mortality in
50 percent of the population for ten fish species after 96 hours of expos-
ure to concentrations of 0.08 to 1.18 ppm (USEPA, 1976). Sanborn (1974)
used the model ecosystem to construct a food chain which included sorghum,
frogs, saltmarsh caterpillar larvae, snails, freshwater clams, freshwater
crabs, water fleas, green filamentous algae, mosquito larvae, and mosquito
fish. At the end of 33 days, organisms were analysed for 14C labelled
carbofuran; carbofuran did not accumulate in any of the test organisms, and
Sanborn concluded that the material was highly biodegradable and posed no
hazards in terms of biomagnification.
Carbofuran is a direct inhibitor of acetylcholinesterase and thus ex-
hibits high toxicity to both avian and mammalian species (Table 9-5). Toxic
effects on birds have resulted in numerous bird kills (USEPA, 1976), one of
which was cited as perhaps the worst single incident of a wildlife kill in-
volving pesticides in the United States. In this particular incident, 2400
wild ducks died after exposure to the carbofuran which had been applied for
the control of Egyptian alfalfa weevil and pea aphids (Fukuto, 1976). This
high level of toxicity greatly limits carbofuran's use as an insecticide,
and has prompted the development of related carbamate insecticides with
lower degrees of mammalian and avian toxicity. Bioaccumulation of carbo-
furan has not been noted in birds or mammals and is deemed highly unlikely
because of the insecticide's rapid degradation in biological systems, and
low fat solubility (USEPA, 1976).
Carbofuran's toxic effects on beneficial anthropods are more severe
with foliar spray applications than they are with granular soil applications.
Bee kills and significant decreases in earthworm populations have been ob-
served when these organisms came into contact with the flowable formulation
(Tomlin, 1975; USEPA, 1976). Soil incorporation of carbofuran should reduce
toxic effects on various insect predators in many cases; however, Brown and
Shanks (1976) have shown that populations of two predators of the two spot-
ted mite on carbofuran-treated plants showed significant mortality. Since
carbofuran is a systemic insecticide and the presence of the prey had lit-
tle impact on predator mortality levels, the authors concluded that the
mortality resulted from an ingestion of the poisoned sap by the predator
species.
Soil flora are the only group of terrestrial organisms which are not
adversely affected by the use of carbofuran. Hubbell et al_. (1973) studied
the toxicity of pesticide combinations in the field and found that normal
application rates of carbofuran brought about only a marginal reduction in
bacterial and fungal populations. Higher than normal application rates
caused a 25 percent reduction in nitrification during the eight weeks
which followed application. In other studies carbofuran has proved to be
actually beneficial to soil microorganisms, with normal application rates
resulting in 100 to 300 percent increases in bacterial and actinomycetal
populations (Mather e£ al. 1976).
259
-------
Carbofuran: Persistence--
Carbofuran is a relatively nonpersistent insecticide (Table 9-6) whose
degradation occurs through both biological and non-biological pathways.
Hydrolysis, hydroxylation, oxidation, and conjugation metabolic pathways all
contribute to the degradation of carbofuran in plants, insects, and other
animals (Getzin, 1973). Persistence studies on carbofuran in corn have shown
residue levels to be below established tolerances both at the silage stage
and at harvest (Caro e* al_. 1973).
Soil'degradation rates of four months (Table 9-6) have given carbofuran
the label of a moderately persistent pesticide although its degradation de-
pends greatly on such factors as manner of application, soil pH, and micro-
bial populations. Caro et^ al. (1973) found persistence to be higher when
carbofuran was applied in the furrow to an acidic, heavy textured soil. Half-
life in this case was 117 days vs.the 46-day half-life of the material when
it was broadcast on a more neutral soil. Getzin (1973) showed a similar
relationship of persistence to soil pH and found that carbofuran's half-life
extended to 54 weeks in organic muck soils. Getzin also pointed out
the importance of microbial degradation of carbofuran; Williams et al. (1976)
noted that high microbial activity can be so great as to reduce the insecti-
cidal effectiveness of the compound. Work by Read (1969) indicates that car-
bofuran may be more persistent than is commonly assumed, band applications
retained all but six percent of their larval toxicity 90 days after
treatment. Williams and Brown (1976) have found similar evidence of slow de-
gradation in British Columbia with a build-up of residues occurring after two
successive years of treatment.
Studies in model ecosystems have shown that carbofuran hydrolyzes
rapidly in aquatic systems (Yu et_ al. 1974) and degrades to water soluble
materials which do not persist or remain at high levels (Sanborn, 1974).
Applications to flooded rice fields showed maximum concentrations 14
hours after application whereupon residues quickly dissipated, having a
half-life of one day or less (USEPA, 1976).
Carbofuran: The Effectiveness of SWCPs in Reducing Transport--
Although very few studies have been done on the effects of SWCPs on
carbofuran1s runoff, one can estimate the effects of their implementation
by looking at the data on other pesticides. Atrazine, for instance, has
been used in numerous studies on the effects of SWCPs and shares several
physical and chemical characteristics with carbofuran. Although atrazine
persists for longer periods of time in the soil (16 to 48 weeks
vs. 16 weeks for carbofuran), carbofuran is more soluble (700 ppm vs.
70 ppm for atrazine) and thus is transported more readily in runoff. Studies
on atrazine's runoff from agricultural fields report losses that generally
fall within the range of 1 to 5 percent of the applied material; Caro et al.
(1973) reported annual losses of 0.5 to 2.0 percent of the carbofuran applied
to corn grown on silty loam soils, and Smith et^ al. (1974) reported single
runoff event losses of up to 14 percent of the~~carbofuran applied to
corn grown on claypan soils. If the data on the use of SWCPs to reduce
atrazine runoff are used, it will be predicted that SWCP could indeed help
to reduce the runoff losses of carbofuran (see Tables 9-9 and 9-10 in atra-
zine section). Stripcropping, ridged corn, and conservation tillage systems
260
-------
brought about reductions of two percent to 79 percent in atrazine's
runoff. The modelling studies of Beyerlein and Donigian (Appendix F) indicate
that no tillage reduces atrazine losses by twelve percent, contours by twenty-
four percent, and terraces with contours by 64 percent. Reductions
in carbofuran's runoff losses with no tillage, contours, and terraces
with contours may be smaller than the reported reductions for atrazine simply
because carbofuran is almost entirely transported in the dissolved phase
of runoff. Beyerlein and Donigian reported reductions of six percent,
11 percent and 36 percent in total runoff water with no tillage, contours,
and terraces with contours, respectively; reductions in carbofuran's runoff
losses would probably be between the reported reductions in atrazine runoff
and water runoff. However, there may be cases where SWCPs do not aid in the
reduction of carbofuran's runoff losses. It has been shown by Smith et aL
(1974) that on claypan soils the runoff losses of atrazine and carbofuran"
from corn grown under conventional tillage will be equal to, or smaller than
the runoff losses from corn grown under conservation and no tillage production
systems.
Thus it appears that on claypan soils, carbofuran's runoff losses are
large enough to warrant the implementation of runoff control measures; how-
ever in the case of claypan soils, the soil and water conservation practices
of conservation tillage and no-tillage have no effect or a negative effect
on the runoff losses of carbofuran. The impacts of other SWCPs on carbofuran's
runoff losses from claypan soils have not been studied. Runoff losses of
carbofuran from silty loam soils are generally not large enough to warrant
the implementation of control measures although in this case SWCPs could
potentially bring about substantial reductions in runoff losses.
Carbofuran: Increasing Efficiency in Use—
Increasing the efficiency of use is not as easily accomplished in the
case of carbofuran as it is in the cases of toxaphene and methyl parathion.
Because carbofuran is used primarily as a granular formulation for the control
of the corn rootworm, drift losses are small and will not usually contribute
to the pollution of non-target areas. Simple changes in application procedures
thus will not reduce pollution due to carbofuran's use in agriculture.
The cultural control method of crop rotation is an effective means of
reducing the need for soil insecticides like carbofuran but is limited in
its applicability. Corn rootworms require the continuous presence of corn.
host plants for the completion of their life cycles, thus annual rotation of
corn with another crop eliminates the corn rootworm and the need for soil
insecticide treatments for its control. Rotations involving more than one
year of corn will not provide complete protection from corn rootworms in
years following the first year, although populations are greatly reduced
when compared to rootworm populations found in continuous corn. Unfortunately,
rotations often mean reduced profits for farmers and thus are not used as
widely as might be expected. Corn-soybean rotations, are profitable in much
of the Midwest; however in this case, the increased sediment and water losses
associated with corn-soybean rotations (Section 4) create a conflict between
the goals of plant protection and the goals of soil and water conservation.
There is strong evidence indicating that the number of treatments used
261
-------
for the control of corn rootworm could be substantially reduced even without
the use of rotations. Having observed that many corn farmers accept soil
insecticides as a necessary production input, Turpin and Thieme (1977)
monitored rootworm populations in 234 untreated Indiana cornfields and found
corn rootworm larvae present in only 34 percent of the test fields.
Rootworm populations only reached economic injury levels in ten percent of
the infested fields, thus the authors concluded that "prophylactic use of
soil insecticides on corn is seldom profitable in Indiana." Integrated
pest management programs are now being developed and implemented to help
farmers determine when it is necessary to use insecticides like carbofuran
for the control of corn rootworm; however, these IPM programs are not as well
developed or as accessible as they in the case of cotton. Pilot IPM programs
involving scouting and a corn variety resistant to European corn borer are
being implemented with much success throughout much of the corn belt. Luckmann
(personal communication, 1979) studied these IPM programs in Illinois and
found that fewer than 50 percent of the scouted fields had rootworm
populations that required insecticide treatments. In former years without
scouting, all corn fields would have received an insecticide treatment for
corn rootworm whether or not injurious populations were present. Since
insecticide treatments cost more than twice as much as scouting, it appears
from Luckmann's studies that reductions of up to 50 percent in the use of
carbofuran can be gained at no net cost to the farmer.
Carbofuran: Summary and Conclusions--
In contrast to the other two insecticides studied, most of the transport
of carbofuran away from target areas occurs with runoff. The major transport
routes of these insecticides differ primarily because of differences in
formulation and application procedures. Instead of being sprayed by aerial
equipment as methyl parathion and toxaphene are, carbofuran is applied
primarily in a granular formulation by ground equipment and consequently
suffers only small losses through drift. The amount of carbofuran leaving
agricultural fields in runoff generally consists of less than two percent
of the material applied; methyl parathion and toxaphene have annual runoff
losses of 0.5 percent or less of the applied material. However, because
carbofuran is applied only once a year and the other insecticides are applied
as much as twelve times in one growing season, the annual runoff losses of
carbofuran are relatively small and generally consist of less than 30 g/ha.
The effects of these concentrations on aquatic organisms are severe but
relatively ephemeral as carbofuran has a half-life of only one day in water.
Thus, given the small amounts of carbofuran transported in runoff and the
rapid degradation of the material in water, carbofuran does not appear to
be as hazardous as toxaphene to aquatic ecosystems.
Strategies for reducing pollution are somewhat different in the case
of carbofuran simply because the air-borne losses of carbofuran cannot be
substantially reduced as they can in the cases of toxaphene and methyl
parathion. Any residues of carbofuran found in surface waters will probably
have resulted from runoff losses or improper disposal of tank rinsate or
carbofuran containers; thus reductions in carbofuran pollution will only
come about as a result of programs which improve disposal techniques, reduce
runoff losses or decrease the amount of carbofuran applied. Because almost
all carbofuran in runoff is in the dissolved phase, the only SWCPs which
262
-------
will significantly reduce carbofuran's transport are those which reduce the
amount of runoff water leaving target areas. Although very few studies have
been done on the impacts of SWCPs on carbofuran's runoff losses, data on
other pesticides sharing solubility and persistence characteristics with
carbofuran indicate that terraces with contours could reduce runoff losses
by 60 percent. As discussed in Section 8, terraces with contours vary in
costs throughout the country but are quite expensive in most cases. Less
expensive SWCPs such as conservation or minimum tillage would be expected
to reduce runoff losses of carbofuran to a much lesser extent. Simple ex-
changes in application procedures may prove to be an easier method of reduc-
ing the runoff losses of carbofuran; one study on runoff has indicated that
runoff losses will be smaller for a given volume of runoff when band (in-
furrow) applications instead of broadcast applications of carbofuran are made.
Further studies should be conducted to establish the impact and cost effect-
iveness of such changes in application procedures.
Another strategy for reducing carbofuran losses involves the implementa-
tion of pest management techniques which reduce the need for carbofuran
applications. For example, the use of carbofuran for corn rootworm control
(its primary use) can be completely eliminated if corn is rotated with a
crop which does not serve as a host for corn rootworm larvae. In most of
the major corn and soybean producing regions, the corn-soybean rotation pro-
duces substantial economic returns while eliminating the need for soil in-
secticide treatments for the rootworm. Unfortunately, sediment losses from
soybean fields are quite high; thus if such a rotation scheme resulted in
higher acreages of soybeans, it would also contribute to increased sediment
pollution. On farms where both corn and soybeans are grown, rotating corn
and soybean fields without increasing soybean acreages would not necessarily
increase erosion. Sod-based rotations reduce sediment pollution as well as
the need for soil insecticides like carbofuran but are not as profitable as
continous corn in most areas. Another IPM alternative is the use of scout-
ing programs to monitor pest population levels. This IPM method shows con-
siderable promise for reducing insecticide application rates and frequency
of application. In pilot programs on limited acreages in Indiana and
Illinois, applications of soil insecticide were reduced by over 50 percent
when fields were scouted to determine whether densities of corn rootworm
were large enough to warrant an insecticide treatment. With such a reduc-
tion in application rates, the cost of scouting is more than offset by the
savings in insecticide costs. It should be noted that scouting does not
reduce insect densities but does help determine whether insecticide treat-
ment is necessary. Hence in those areas where climate and cropping practices
are such that high population levels of corn rootworm occur every year,
scouting is not likely to significantly reduce insecticide use. Entomologists
working in corn growing areas have estimated that such areas comprise only a
small fraction of the total corn acreage currently treated with insecticides
for the control of corn rootworm.
In conclusion, both SWCPs and integrated pest management programs can
significantly reduce carbofuran losses in runoff. Both strategies also have
additional benefits:
263
-------
1) SWCPs may reduce sediment and nutrient pollution while conserving
valuable topsoil and water;
2) the implementation of IPM programs and the resulting decrease in
the use of carbofuran would be expected to also reduce the potential
for pollution due to improper disposal and for accidents involving
farm workers and non-target species living within and outside of
target areas. Crop production costs and energy use may also be re-
duced if IPM programs lead to reductions in pesticide use.
Detailed information on the effectiveness or costs of SWCPs and IPM is not
available; however in most areas of carbofuran's use, it appears that IPM
is more cost-effective than any of the SWCPs in reducing carbofuran's
transport away from agricultural areas. In the remaining areas, the
implementation of SWCPs for the control of carbofuran's runoff losses
should be preceeded by a careful analysis of the environmental hazards
associated with the small and ephemeral concentrations of carbofuran trans-
ported to surface waters with runoff.
Atrazine
Atrazine, a s-triazine herbicide, accounts for almost 50 percent of
all herbicides used in corn production and is, as stated previously, the
most heavily used herbicide in the United States (Adrilenas, 1974).. Its
high degree of selectivity allows for applications to the soil both before
and after the emergence of the crop and greatly reduces the need for time-
consuming tillage and cultivation operations.
Atrazine: Runoff and Leaching--
The quantity and concentration of atrazine found in runoff depends on
many factors among which are: intensity of rainfall (Baker and Johnson, 1977;
White e^ al_. 1967), site characteristics (Bailey et_ a±. 1974), land manage-
ment and conservation practices (Hall 1974; Ritter et^ al. 1974; Smith
et aj_. 1974; Baker and Johnson, 1977; Triplett et_ al. 1978), application rates
(Hall et al. 1972) and timing of first runoff producing storm relative to
application (White et ad. 1967; Hall 1974; Ritter et al, 1974; Smith et al.
1974). This last factor is probably the most important determinant of the
actual quantitites of atrazine that will leave a field via runoff. Baker
and Johnson (1977) found that seasonal losses were less than five percent in
years when the first runoff producing storms occurred two weeks or more after
application. However, in one year in which a storm took place 24 hours after
application, the losses were 16 percent of the applied atrazine.
The runoff of atrazine from agricultural lands has been studied inten-
sively under a variety of geographic and experimental conditions, von Rumker
et^ al. (1975) have reviewed these numerous studies and have estimated that
of the 24 million kilograms of atrazine applied to the 1971 U.S. corn crop,
165,000 to 945,000 kilograms may have been lost in the dissolved and solid
phase of runoff. The data from several field and modelling runoff studies
are summarized in Table 9-9. These studies vary in terms of the reported
264
-------
TABLE 9-9. ATRAZINE LOSSES IN RUNOFF
Amount
Applied Type
(kg Al/ha) Application
3.361 surface with
simulated
o
rainfall
2.22 surfaceb
1 . 68 * incorporated
,0l with simulated
1.68 . „ , ,c
rainfall
3.361
3.361
2.22 surfaced
4.52
3.363 surface6
2 . 242 surface
1.12-3.362 incorporated8
1 Duration of experiment was
2 Duration of experiment was
3 Duration of experiment was
? White et al. (1967).
Hall et al. (1972).
J! Bailey et al. (1974).
Hall (1974).
~ Ritter et al. (1974).
Baker and Johnson (1977).
g Triplett £t al. (1978)..
Average
Slope (%)
6.5
14
2.2
3.6
2.5
5.7
14
10-15
12-18
8-22
1 hour on
Loss in Runoff Concentration in
(% Application) Runoff Water (ppm)
2 0.16-8.08
(2 - 7.3)
2.5 0.0-0.8
6.44 0.0-3.3
12.47 0.0-7.9
13.3 0.0-11.1
10.18 0.0-4.0
5.0 0.0-2.3
4.8 0.05-4.6
2.7-16.0 1.17-4.91
>5
0.02-5.7 0.10-0.48
fallow lands
5 months on corn crop.
58 days on
corn crop.
265
-------
quantities and concentrations of atrazine found in runoff, but all note
that atrazine concentrations are much higher in the sediment portion of
the runoff. Baker and Johnson (1977) found atrazine concentrations in
sediment runoff were five times as large as those in the water runoff, but
pointed out that in terms of total quantity lost, more atrazine is associated
with the water portion as water losses are much larger than sediment losses.
Baker and Johnson go on to say that because the major herbicide loss occurs
with water, erosion control practices will not prevent the bulk of atrazine
runoff losses.
Researchers have differed in their evaluations of the ecological signi-
ficance of atrazine's runoff losses. Some have conjectured that runoff waters
from non-treated areas and surface waters will dilute atrazine-laden waters
and reduce hazardous herbicide concentrations to levels below acute toxicity
(White et_ al_. 1967; Hall, 1974; Triplett e^ al_. 1978). This argument ignores
the possibilities of sublethal effects on non-target species and disruptions
in food webs caused by reduced densities of algal and microbial populations.
Such possible sublethal and secondary effects should be further investigated
for this ignorance, coupled with the lack of data on atrazine's degradation
in water, makes it difficult to evaluate the environmental impact of
atrazine's runoff losses.
The rate at which atrazine leaches through agricultural soils varies
greatly according to the following factors: plant adsorption (Hall and
Hartwig, 1978), reversible and irreversible adsorption (Bailey and White,
1970; Sanborn £t al_. 1977; Hall and Hartwig, 1978), degradation rates (Hall
and Hartwig, 1978; Harris et_ aJ^ 1969), soil composition and moisture
(Burnside £t aJ, 1971; BaiTey~and White, 1970; Sanborn e_t a]_, 1977); cropping
practices, and nutrient levels (von Stryk and Bolton, 1977). Leaching losses
would be expected to be greatest from light soils with small adsorption and/or
degradation capacities (Sanborn et^ al. 1977).
Different combinations of the above factors have yielded data showing
a considerable range of leaching depths for atrazine residues. Burnside
et al. (1963, 1971) have found atrazine residues well below the plow layer,
and have cited leaching as a major factor in the removal of the herbicide
from agricultural areas. The possible hazards from atrazine leaching losses
are well illustrated by Leh's work (1968) in which groundwater was found to
be polluted by atrazine in wet years and during periods of application.
Richard et_ al. (1975) also detected atrazine in all of the wells sampled in
the alluvial plains of Iowa rivers contaminated with atrazine. Other studies
have shown that the bulk of applied atrazine reacts and dissipates in the
plow layer with only small amounts reaching lower levels in the soil pro-
file (Hall and Hartwig, 1978; Baker and Johnson, 1977; Schwab et_ al. 1973;
von Stryk and Bolton, 1977). von Stryk and Bolton (1977) studied the
leaching of atrazine into tile drains under rotation corn and continuous
corn with higher fertility levels and found that although losses were higher
in the continuous corn, both cropping systems had relatively small losses
of 0.25 percent and 0.8 percent, respectively.
Atrazine: Air-borne Losses--
The magnitude of atrazine's drift losses depends on weather conditions
266
-------
and application procedures (Table 9-8). von Rumker et aK (1975) have re-
viewed the available literature on drift losses of herbicides and have con-
structed general drift loss tables for any herbicide based on the applica-
tion procedure. According to their table, atrazine losses through drift
could range from negligible for ground equipment applications, to 40 percent
for application by air equipment. A field study by Baker and Johnson (1977)
shows that significant losses of atrazine can occur using ground equipment
under windy conditions. Immediately after application measurements were
made of the amounts of herbicide deposited on the soil and on filter paper.
Results showed that with winds of 20 km/hr during application, losses of
atrazine range from 48 to 51 percent of the applied material. Unfortunately,
it is not known how often farmers apply pesticides under such windy conditions
or how large drift losses are in wind speeds more typical for spraying opera-
tions.
Volatilization losses can also be quite high when atrazine is surface
applied to warm, dry soil, and is not incorporated via cultivation, rainfall
or irrigation. Kearney et^ al. (1964) measured losses under such conditions
and reported that with temperatures of 35°C, losses were as high as 40 per-
cent during a 72 hour period. With a temperature increase to 45°C, they
reported losses of nearly 80 percent of the material originally applied.
Foy's (1963) experiments pointed to a similar relationship between volatility
and temperature, with temperatures of 40°C and 60°C bringing losses of 11
percent and 95 percent, respectively.
The environmental fate and significance of atrazine1 s air-borne losses
are not well understood. Atrazine drift reaching non-target plants could
result in damage similar to the decline of the hackberry seen in towns
receiving herbicide drift from corn growing regions in Iowa (Hibbs, 1976).
Other air-borne losses from drift, volatilization, and wind erosion will
be redeposited on the earth's surface with rainfall and the settling out of
dust particles. Atrazine residues have been found to be associated with
suspended solids in rainfall at concentrations of 0.1 ppb (Cohen and
Pinkerton, 1966). Further studies establishing the persistence and concen-
trations of atrazine in rainfall would be extremely beneficial for future
evaluations of the importance of controlling air-borne losses.
Atrazine: Toxicity--
Knowing the mode of action of a herbicide helps one to predict which
non-target species will be adversely affected by its use. Atra/ine, which
kills weeds by inhibiting photosynthesis, would be expected to have deleter-
ious effects on phototrophic organisms living in surface waters and the soil.
These phototrophic organisms are often important links in food webs, thus
their demise may cause reductions in the populations of organisms (such as
fish) dependent on them for a food source. A disruption of both plant and
animal populations may also occur as a result of the habitat changes brought
about by atrazine use.
Generally, those conditions which allow for the effective herbicidal
action of atrazine are identical to those conditions which make for the
greatest inhibition of soil algal populations. Atrazine-treated soils rich in
organic matter apparently absorb and inactivate larger amounts of the
267
-------
herbicide, and thus escape a severe inhibition of their algal populations
(Kaiser et_ al. 1970). Atrazine is also removed from the sphere of biological
interactions~in highly acidic soils. The hydrogen ions prevalent in these
soils associate with the herbicide to form positively charged substances
that are readily adsorbed by soil particles (Weber et^ al_, 1975). Kaiser et al.
(1970) have found that while small dosages of atrazine may actually stimulate
microflora, large applications inhibit growth. They go on to predict that
the decreases in soil organic matter and weed populations brought about by
herbicide use will further inhibit soil microbial populations.
Atrazine has also been shown to affect arthropod population levels.
Springtails, mayflies, caddisflies and leeches have all been adversely
affected by low levels of atrazine (Thompson, 1973). Lichtenstein et al.
(1973) have shown that atrazine can have a synergistic effect on the toxicity
of insecticides. In experiments with parathion, the addition of normally
non-toxic atrazine increased insect mortality from eight to 50 percent.
Such synergistic effects must be taken into consideration when evaluating
the toxicity of various insecticides to insect predators and parasites.
Atrazine is less toxic to higher organisms. It is relatively non-toxic
to birds and mammals (Table 9-5), the latter having efficient excretion and
degradation systems for the herbicide (von Rumker, et_ al. 1974; Esser et_ al,
1975). Atrazine is considered to be moderately toxic to fish (Table 9r5)
with lethal concentrations for 50 percent of the population ranging from
0.55 - 12.6 ppm, levels quite within the range of concentrations seen in
surface runoff. Although little work has been done on the more insidious
sublethal effects on terrestrial and aquatic organisms, the aforementioned
work by Plewa and Gentile (1976) showing atrazine to be mutagenic to yeast
does not allow for complacency. Yoder et al. (1973) have further shown that
custom applicators of herbicides (including atrazine) have up to four times
as many chromosome aberrations in their blood lymphocyte cultures as
compared to a control population.
Haque et^ al. (1977) have used a bioaccumulation ratio to evaluate the
biomagnification of atrazine in aquatic food chains. This ratio compares
the concentration of the chemical in the organism to the concentration of the
chemical in the water. Values obtained for atrazine were two to 15 for
snails, ten to 83 for algae and three to ten for fish. Metcalf et a^. (1971)
conducted biomagnification studies in a terrestrial-aquatic modeT~ecosystem
and concluded that magnification is unlikely to occur in aquatic food chains.
Atrazine: Persistence--
Atrazine is generally applied to the soil surface, although incorporation
increases its activity and prolongs its effectiveness. Once in the soil,
atrazine is reversibly adsorbed to soil particles, with desorption occurring
readily, although at a much slower rate than that of adsorption. This steady
desorption releases atrazine into the soil solution and allows excellent
continuing control of weeds but also creates residue problems in the soil
environment. At times, atrazine residues have accumulated in high concen-
trations and have caused damage to susceptible crops planted in following
268
-------
years. In many other crop situations, factors such as soil temperature
(Haque and Freed, 1974), pH, moisture (Weber and Weed, 1974), crop cover
(Hiltbold, 1974), rate of application (Libik and Romanowski, 1976) and
herbicide depth in the soil (Harris et_ al. 1969) interact to create conditions
that are more optimal for degradation of atrazine. In any case, atrazine has
proven to be a fairly stable substance in the soil with persistence rates
of four months to one year (Haque and Freed, 1974). In the 1971 National
Soils Monitoring Program, Carey et al. (1974) found that, of the 213 soil samples
taken in areas treated with atrazine during the growing season, 71.4 percent
contained residues of the material. LeBaron (1970) has suggested that atrazine
residue problems could be reduced by plowing and thorough mixing of the soil
following crop harvest. LeBaron also indicates that soils could be de-
toxified with the use of charcoal but adds that this technique is not yet
feasible because of unresolved economic and application problems.
Surprisingly enough, an extensive review of the literature uncovered little
information on the persistence of atrazine in water although studies showed
that degradation does take place via photodecomposition and hydroxylation
(Jordan et_ aK 1970). Khan (1978) has shown that the chemical hydrolysis of
atrazine in solution is pH dependent, proceeding much slower at those pHs
associated_ with soil solutions and surface waters and more rapidly in waters
originating from coniferous forests. Studies of atrazine and its hydrolysis
product, hydroxyatrazine, under conditions of limited aeration have shown
that the total breakdown of atrazine will be slower in submerged sediments
and faster in aerated soils (Goswami and Green, 1971).
Data have been published which establishes the presence, be it persistent
or transient, of atrazine in aquatic systems. In a study by Morley (1977),
atrazine was found in four out of six watersheds, with mean concentrations
peaking at 5 ppm in June and declining to 2 ppb in July and 0.5 ppb for
the remainder of the year. In another study by Richard et^ al. (1975) samples
were taken in Iowa to ascertain the extent and the degree of contamination
of surface, sub-surface and finished drinking waters by dissolved DDE, dieldrin,
and atrazine. Of the three pesticides monitored, atrazine appeared in the
highest concentrations, and was found in several shallow wells and in
water which had been ostensibly decontaminated by filtration through activated
carbon beds. The contaminated shallow wells were all located in the alluvial
plains of contaminated rivers. It is difficult to determine the importance
of these residues without degradation data, for if degradation is fairly rapid,
the transitory presence of the herbicide may not warrent extensive pollution
control plans to prevent it from entering surface waters.
Atrazine: The Effectiveness of SWCPs in Reducing Transport--
Several studies have examined the impact of SWCPs on atrazine»s
runoff losses from agricultural fields (Table 9-10). In the majority of
cases the impact of a SWCP was significant; however, in areas with claypan
soils the amounts of atrazine measured in runoff from fields with con-
servation tillage exceeded the atrazine levels measured in runoff from
fields with conservation tillage (Smith et al.. 1974). Stripcropping
provided the most effective control and reduced atrazine's runoff losses by
79 percent (Hall, 1974). To incorporate the effects of changing weather
conditions, a simulation model of atrazine transport was developed by
269
-------
TABLE 9-10. EFFECT OF SWCPs ON ATRAZINE LOSSES IN RUNOFF
to
^]
O
2
Tillage Systems
Compared
conventional Ia
vs.
stripcropping
surface contoured
vs.
ridged corn
conventional II c
vs.
no-till 1
conventional III
vs.
till plant
Application
Rate
(Al/ha)
2 . 2 kg/ha3
4.5 kg/ha
3.36 kg/ha4
3.25 kg/ha5
2.24 kg/ha3
Runoff Losses
Under Runoff Losses Under
Conventional Tillage Conservation Practice
(g/ha
77
49
275
537
17.
19.
41.
80.
20
and % of
g/ha =
g/ha =
g/ha =
g/ha =
3 g/ha =
7 g/ha =
6 g/ha =
6 g/ha =
g/ha =
applied)
3.5%
1.1%
8.2%
16.0%
.53%
.61%
1.28%
2.48%
.89%
(g/ha)
10
17
131
91
32.5
76.8
52.3
67.9
18
g/ha
g/ha
g/ha
g/ha
g/ha
g/ha
g/ha
g/ha
g/ha
conventional III
vs.
ridge plant
2.24 kg/ha'
20 g/ha =
.89%
40 g/ha
conventional IV
vs.
no-till 2
f
conventional V
vs.
minimum tillage
1.12 kg/ha3
2.91 kg/ha6
2.80 kg/ha6
0.35 g/ha =
53.51 g/ha =
79.24 g/ha =
.03%
1.8%
2.8%
0.27 g/ha
47.24 g/ha
76.81 g/ha
-------
TABLE 9-10. (Continued)
f
Tillage Systems'
Compared
f
conventional V
vs.
contour tillage
f
conventional V
vs.
terraces with
contours
, Application Runoff Losses Under
Rate Conventional Tillage
(Al/ha) (g/ha and % of applied)
2.91 kg/ha6 53.51 g/ha =1.8%
2.80 kg/ha6 79.24 g/ha =2.8%
2.91 kg/ha6 53.51 g/ha = 1.8%
2.80 kg/ha6 79.24 g/ha =2.8%
Runoff Losses Under
Conservation Practice
(g/ha)
40.88 g/ha
62.97 g/ha
18.96 g/ha
29.76 g/ha
Applied as 80%
2
wettable powder.
Descriptions of tillage systems:
Conventional I = rototilling to 15 cm and soil smoothed by hand
Stripcropping = inclusion of oatstrip adjacent to corn at slope base.
Conventional II = land prepared by turning with a moldboard plow, discing, harrow-
ing plus cultivation of row crops.
No-till 1 = crop residues from previous years are chopped and remain on the
surface. Seed is planted through these residues with no addi-
tional land preparation. Weeds were controlled by use of
herbicides.
Conventional III = spring plowing, discing and surface planting. 3% residue cover.
Till Plant
Ridge Plant
= buffalo till planting in which previous years' ridges were
split at plow planting time. 20% residue cover.
= technique which retained the same range from year to year.
45% residue cover.
-------
FOOTNOTES FOR TABLE 9-10. (Continued)
to
-J
10
4
5
6
a
b
c
d
e
f
Conventional IV = plowed to depth 20 cm, disced and harrowed. One post planting
tillage of rotary hoeing or cultivation was used.
No-till 2
= corn planted in residues from previous crop with planter that
disturbed band of soil 7-10 cm wide. Weeds controlled by
herbicides.
Conventional IV = disced and planted in straight rows parallel to the slope.
Crop removal in the fall.
Minimum Tillage = no till prior to spring planting. Planting done with crop
residue remaining from fall, rows parallel to slope.
Contour tillage = disc in spring prior to planting, rows perpendicular to slope.
Terrace with
Contours
5-month field study on corn.
58-day field study on corn.
3-month field study on corn.
10-year simulation study on corn.
Hall (1974).
Ritter e£ aL (1974).
Smith et al. (1974).
Baker and Johnson (1977).
Triplett ejt aL (1978).
Appendix F.
Residues removed after harvest.
= disc in spring prior to planting, watershed area of 1.29 hectares
is divided into 2 terraces with contoured rows perpendicular to
slope.
-------
Hydrocomp, Inc. (Appendix F). The model was computed over a long period of
time and a range of weather conditions. The atrazine losses estimated by
the simulation model indicate that less than 1.84 percent of the material
applied is carried by runoff from a field with conventional tillage. The
use of SWCPs reduced this loss to 0,65 percent using terraces with
contours. Less expensive SWCPs such as no tillage and contour tillage bring
about smaller reductions in atrazine's runoff losses.
Atrazine: Improving Efficiency in Use—
Finding methods to increase efficiency in the use of herbicides is in
many ways a more difficult task than it is with insecticides. High costs
and crop injury resulting from the overuse of herbicides have worked together
in a fortuitous manner to prevent farmers from the sometimes wasteful use of
pesticides seen in the case of cotton insect control programs. However,
although weed control has not constituted a large portion of IPM research and
implementation, there is some indication that improvements could be made in
the use of herbicides. The careful monitoring of weed population sizes,
growth stage, and species type would help farmers make decisions concerning
the type and rate of herbicide needed and whether treatment is
necessary. Research establishing what size weed population can be tolerated
in a field without significant damage to crops might persuade farmers to
abandon the "clean" field as a goal in weed control.
Research to determine when the weed species is most susceptible to
chemical treatment could also potentially reduce the amounts of herbicides
needed for weed control. Recent work has shown that weather conditions can
influence the effectiveness of the triazine herbicide metribuzin in that
the tolerance of both tomatoes and jimsonweed declines following cloudy days.
Fortunately, the decline in tolerance is greater in the jimsonweed. In this
case, a reduction of 40 percent in the application rate of metribuzin still
provided effective weed control after three days of cloudy weather (Seim, 1978),
Crosby (1975) has noted that the herbicidal powers of atrazine are similarly
inactivated in sunlight. The development of application equipment with
improved placement and conservation of herbicidal materials will further
decrease the amounts of pesticides available for transport to surface waters
(McWhorter, 1977; Seim, 1978).
It is only quite recently that attempts have been made to investigate
the use of integrated pest control methods to suppress weed populations in
field crops. The success of biological control of Klamath weed by an insect
herbivore indicates that there is a potential for integrated pest managment
for control of agricultural weeds as well. However, at the present time it
is not possible to predict how effective such programs will be or when they
can be implemented.
Unfortunately, the use of atrazine can interfere with integrated pest
management programs for insect control. Harris and Miles (1975) have noted
that the use of persistent herbicides such as atrazine encourages the farmer
to grow com year after year because other crops such as soybeans may be
damaged by residues remaining in the soil the following year. Unfortunately,
the corn rootworm thrives under such continuous corn programs; increases in
its population levels have necessitated the widespread use of insecticides
273
-------
for rootworn control. The cultural control method of crop rotation, which
formerly controlled the rootworn, thus cannot be utilized in agricultural
areas producing rotation crops susceptible to damage by atrazine residues.
Atrazine: Summary and Conclusions—
The effectiveness of SWCPs for reducing pesticide transport in runoff
has been studied more extensively for atrazine than it has been for any of
the other four representative pesticides. The results of these studies
indicate that atrazine losses in runoff with conventional tillage are gener-
ally rather small (one to five percent ) and that SWCPs can reduce these
losses. Stripcropping and terracing with contours appear to be the most
effective SWCPs for this purpose (Table 9-10). However, it is possible
that the use of a SWCP to decrease the amount of atrazine in surface runoff
will be accompanied by an increase in atrazine's leaching losses. Modelling
studies have predicted that atrazine will not leach into groundwater, but
in empirical studies atrazine has been detected in groundwater. Thus, it is
not clear that SWCPs would actually decrease water contamination by atrazine
in all areas.
Since pest control techniques requireing less atrazine use are not cur-
rently available, the only other pollution control strategies beside SWCPs
are control of drift and volatilization or the replacement of chemical weed
control with cultivation. The studies measuring atrazine losses from vola-
tilization and drift reported losses ranging from negligible to greater than
40 percent. The procedures for reducing air-borne losses include incorpora-
tion of atrazine into the soil and elimination of aerial applications.
As with the other pesticides, the question exists whether air-borne los-
ses are as environmentally hazardous as runoff. The air-borne losses of atra-
zine are smaller than they are in the cases of methyl parathion and toxaphene
primarily because weed treatments are generally made with high volume large
droplet applications by ground equipment. Unfortunately, there were not
enough data over a range of wind and temperature conditions to determine the
exact magnitude of air-borne losses; however, available evidence suggest that
drift and volatilization losses combined probably exceed the one to five per-
cent of the material commonly transported by runoff. The environmental impact
of air-borne and water-borne atrazine also depend upon atrazine's persistence
in water and air. Although several studies have shown that atrazine is quite
persistent in soil, little information is available on its persistence in
water or air.
In conclusion, it appears that SWCPs are among the most effective means
of reducing potentially hazardous concentrations of atrazine in runoff. How-
ever, before wide scale implementation of SWCPs occurs, studies on aerial in-
puts of atrazine to terrestrial and aquatic ecosystems should be done to es-
tablish whether it would be more cost effective to reduce environmental pol-
ution from atrazine use by reducing drift and volatilization losses. Simple
changes in spraying equipment or changes in atrazine's placement in the soil
may also prove to be more cost effective means of reducing atrazine's runoff
losses. Furthermore, empirical studies on a range of soil types should de-
termine whether SWCPs which increase percolation will increase the leaching
losses of atrazine.
274
-------
Paraquat
Paraquat is a contact bipyridylium herbicide whose chemical properties
and non-selective mode of action make it highly effective as a desiccant
used in the harvest of cotton, potato haulm, and sugar cane, and as a herbi-
cide for grass and broadleaved weed species in both aquatic and agricultural
ecosystems. Rapid action and a non-residual nature make it well adapted
for use in no tillage and conservation tillage corn production systems,
where it has been used extensively to destroy weeds before the planting of
the crop or before crop emergence.
Paraquat: Runoff and Leaching--
Although substantial amounts of paraquat may leave agricultural fields
with runoff, the physical and chemical characteristics of the herbicide
greatly reduce the hazards commonly associated with these losses. When
paraquat is applied to a field, a portion of the material decomposes or
is adsorbed into plant tissues. The paraquat which reaches the soil is
irreversibly adsorbed to clay particles and reversibly adsorbed to organic
matter. In the weeks following application, the paraquat initially adsorbed
to organic matter is redistributed to the clay particles, which are believed
to be the final repository of paraquat residues. If a runoff producing
storm occurs during this period of redistribution, unattached paraquat
could potentially leave fields in runoff water; however, this is apparently
not the case for paraquat is rarely found in runoff water.
Researchers conducting field and modelling studies on paraquat's runoff
have reported both large and small losses from agricultural fields. Smith
®1 al. (1978] found solid phase runoff losses which comprised 22
percent of the applied material; however, because paraquat is only applied
to foliage when used in agriculture and the authors applied the material
directly the soil, Smith et_ al. stated that these losses do not represent the
range of losses expected with normal agricultural use. Studies in Iowa have
found smaller losses of 1.28 percent of the applied material (Baker et al,
1979). Modelling studies based on data from the Georgia watershed studied
by Smith et_ al. (1978) predicted losses of 340.6 g/ha of the 2520 g/ha of
paraquat applied to the soil (Appendix F). As discussed below, there is no
indication that soil-bound residues of paraquat will desorb from soil or
enter into biological interactions upon reaching aquatic systems (Calderbank
and Slade, 1976).
Paraquat: Air-borne Losses--
Although some aerial application of paraquat does occur, fields are
generally treated with high volume, large droplet applications with ground
equipment which presumably do not produce hazardous amounts of drift.
However, recent studies indicate that small but serious drift losses of
herbicides can occur even when applications are made by ground equipment in
light winds. Byass and Lake (1977) studied drift losses of paraquat resulting
from normal ground equipment application procedures used in Great Britain and
correlated percent drift losses with damage to crops located downwind of
spraying operations. They found that drift losses which comprised more
than 0.1 percent of the applied paraquat caused significant damage to
275
-------
bean plants located downwind of the treated area; the downwind area at
risk for bean plants was up to one meter in a light wind and up to 20
or 30 meters when paraquat was applied in the highest wind speed in which
spraying would normally be attempted. Hazards from paraquat drift may
be extended for hundreds of meters if crop or non-target species located
downwind are more sensitive than bean plants are to paraquat. Aerial
applications of paraquat will result in higher drift losses and serious
hazards to non-target areas. Although no studies have been done specific-
ally on paraquat's drift losses with aerial applications, von Rumker et al.
(1975) constructed general drift loss tables based on application procedures
and estimated that ten to 40 percent of the herbicides applied to the U.S.
corn crop by aerial equipment will drift over 3040 meters from the target
area. Volatilization does not contribute to the air-borne losses of para-
quat from agricultural areas (Calderbank and Slade, 1976).
Paraquat: Toxicity--
Hazards to non-target species are greatly reduced because paraquat
loses its toxicity when adsorbed to clay particles and retains little
toxicity when adsorbed to soil organic matter. Paraquat generally does not
have adverse effects on most soil organisms or microbial activities important
to soil fertility; however, reductions in springtail populations and residues
in earthworms have been reported (Calderbank and Slade, 1976; Sanborn et al.
1977; Thompson, 1973). Thompson (1973) has concluded that these toxic
effects are not drastic and points out that residues in earthworms are
excreted when the worms are placed in untreated soil. Although bound
paraquat residues may be innocuous, the material is quite toxic in the
period between spraying and adsorption to soil particles. Paraquat is
extremely toxic to entomophagous mites and insects and has been considered
to be too toxic to allow it's use in apple pest management programs (Rock and
Yeargan, 1973). The material is relatively non-toxic to birds and moderately
toxic to mammals (Table 9-5) but poses hazards for spray applicators because
of the irreversible, proliferative changes brought about in lung tissue
after inhalation (Vettorazzi, 1977).
The incidence of paraquat contamination of aquatic systems via runoff
losses is greatly reduced because paraquat is so strongly adsorbed by eroded
soil particles. Even when paraquat is directly applied to surface waters
for weed control, paraquat is given but a small chance to adversely affect
aquatic animals. The herbicide is rapidly adsorbed by weeds and by the
hydrosoil, and may remain in aqueous solution for as little as one half hour
(Yeo, 1967). Once the paraquat has reached the hydrosoil, desorption would
not be expected to occur (Calderbank and Slade, 1976). However, in the
brief period between application and adsorption, paraquat will have severe
effects on bacteria, freshwater algae, and freshwater protozoa at concentra-
tions of 0.6 ppm (Sanborn et_ al. 1977). Bioconcentration has not been a
problem in aquatic ecosystems and levels used for aquatic weed control are
well below toxic levels for fish (Calderbank and Slade, 1976).
Paraquat: Persistence—
Paraquat's "persistence" can be perceived in several different ways
depending on which general definition of persistence one has in mind. If
276
-------
persistence is taken to mean the length of time required for the disappearance
of the original form of the pesticide, then paraquat will be seen as extremely
persistent with some researchers asserting that degradation of bound residues
does not occur (Fryer et a].. 1975). Perhaps more appropriate to the bipyridy-
lium herbicides, is the definition of persistence as that period of time where-
in^the pesticide maintains its insecticidal or herbicidal activity. By
this definition, paraquat is seen as a non-persistent pesticide since its
phytotoxicity is lost on adsorption to the clay particles so abundant in
agricultural soils (Riley et^ a.^. 1976).
The phytotoxicity of paraquat depends on the strong adsorption capacity
of the soil and the type of clay minerals present. Weber and Scott (1966)
and Weber et_ aL (1969) found that given comparable application rates, phyto-
toxicity was greater in soils dominated by vermiculite and kaolinite clays
than in soils dominated by montmorillonite. This presumably occurred
because montmorillonite has a strong adsorption capacity which completely
deactivates paraquat residues, whereas vermiculite and kaolinite cause
reductions in paraquat's phytotoxicity but release the herbicide in small
amounts. This slow release of paraquat residues results in concentrations
which can injure cucumber seedlings grown in model soil systems. Tucker
et al. (1967) leached soils with concentrated salt solutions and found ratios
of irreversibly bound-paraquat to reversibly bound-paraquat of 1:4 in loam
soils, 1:27 in sandy soils, and 1:107 in muck soils. Riley et_ ad_. (1976)
have concluded that bound paraquat is not available to living organisms, thus
whatever damage is seen is due to free paraquat in the soil solution.
Numerous studies have been done on the abilities of agricultural soils
to adsorb and deactivate applied paraquat after continued use. Some re-
searchers have shown that would take 50 to 100 years to saturate the strong
adsorption capacity of the top 2.54 cm of the least adsorptive soil consid-
ered (Boon, 1964; Riley et_ a^. 1976). Riley et_ ad. (1976) have calculated
that if 100 percent of the active ingredients of a 1.0 kg/ha application
reached the soil surface and was uniformly incorporated in the top 15
centimeters of the soil, concentrations of 0.5 yg paraquat/gram soil would
result. They also have reported the results of studies done by the Chevron
Chemical Company in the United States and the 1C I Plant Protection Division
in the United Kingdom; these studies indicated that almost all agricultural
soils can bind from 50 to 500 yg paraquat per gram of soil and that even
the least adsorptive soils (sandy soils and peat soils) can bind at least
50 yg paraquat per gram of soil. Calderbank and Slade (1976) have stated
that desorption of these bound residues would not be expected to occur
either in agricultural soils or from bottom mud.
Residue problems may occur in soils containing only a small percent of
deactivating clays. Juo and Oginni (1978) have pointed out that although
farmers in temperate regions have had no phytotoxicity persistence problems
with paraquat, soils in the humid tropics contain smaller amounts of deacti-
vating montmorilloinite clays. Damanakis et al. (1970) have indicated that
residue problems could also occur in organic soils where paraquat can be
made available to plants in amounts as low as one twenty-fifth of the strong
adsorption capacity. Calderbank and Slade (1976) have reported residual
277
-------
activity of paraquat in organic soils for up to two weeks after application
and have suggested that this period indicates the amount of time necessary
for the paraquat initially adsorbed on the organic matter to redistribute
itself onto the deactivating clay minerals. Weber and Best (1972) have also
shown that paraquat is more phytotoxic in soils that have been heavily limed
as calcium cations compete with paraquat cations for adsorption sites on
solid colloids, leaving more "free" paraquat in the soil solution for plant
uptake. Because these phytotoxic effects lasted only several weeks, it
was hypothesized that all the paraquat was eventually adsorbed by the soil
particles with strong adsorption capacities.
Degradation of paraquat is greatly hindered by its strong adsorption
on clay particles; however, a certain amount of the material does degrade on
plant surfaces and in the soil solution. Slade (1966) has shown that before
its rapid adsorption into plant tissues, some of the applied paraquat
photodecomposes on plant surfaces, forming two photolysis products which
rapidly degrade soon after formation (Riley et_ aL. 1976). When paraquat reaches
soil, microbial and chemical degradation are greatly hindered by the binding
of residues to clay particles (Weber and Weed, 1974; Burns and Hayes, 1974;
Calderbank and Slade, 1976; Hance, 1967).
Degradation of residues in water is in most cases, only a laboratory
phenomenon as any paraquat reaching surface waters is strongly adsorbed by
weeds, suspended soil particles, and bottom mud (Way et_ a^. 1971; Juo and
Oginni, 1978). Indeed, Coates et a_L (1965) have shown that 35 weeks
are required for a complete disappearance of paraquat in water, while dis-
appearance from water with sediment and from water with sediment and plants
takes six to eight weeks and three to four weeks, respectively. The rapid
destruction of plant tissue caused by paraquat does not allow time for
metabolic degradation in the plants (Slade, 1966; Funderbank and Lawrence,
1964) thus residues eventually make their way to the bottom mud (Calderbank
and Slade, 1976). Photodecomposition of paraquat residues does not occur in
aqueous solution (Calderbank and Slade, 1976).
Paraquat: The Effectiveness of SWCPs in Reducing Transport--
Paraquat 's strong adsorption to soil particles indicates that SWCPs will
be quite effective at reducing paraquat!s runoff losses. Field studies
measuring the effectiveness of SWCPs at reducing paraquat's transport are
not available; however, Beyerlein and Donigian (Appendix F) have used the
field data of Smith et^ aL (1978) to model the effectiveness of no tillage
practices, contours, and terraces with contours in reducing paraquat's runoff
losses. The model predicted runoff losses with conventional tillage to be
fourteen percent of the applied material. Runoff losses with no tillage,
contours, and terraces with contours were predicted to be seven percent, nine
percent, and six percent, respectively, of the material applied. It should
be noted that paraquat's primary use is in providing the additional weed
control necessary in no tillage corn. In many cases, paraquat applications
can be eliminated by converting to conventional tillage. In such cases,
conventional tillage is the most effective means of reducing paraquat losses.
278
-------
Although SWCPs can substantially reduce the amounts of paraquat
transported in runoff, it is questionable whether these runoff losses are
hazardous to aquatic systems. As discussed previously, paraquat is carried
only on the sediment fraction of runoff and apparently does not desorb from
clay particles reaching surface waters. Unless later studies indicate that
paraquat runoff losses can be hazardous to aquatic ecosystems, the implementa-
tion of SWCPs specifically for the control of paraquat transport does not
appear to be a cost effective means of improving water quality.
Paraquat: Increasing Efficiency in Use--
Although the paraquat leaving agricultural fields in runoff does not
appear to pose hazards to aquatic ecosystems, losses due to drift or improper
disposal of paraquat containers may enter and contaminate surface waters until
residues are adsorbed by sediment or aquatic plants. When and if this occurs,
increasing the accuracy of placement and improving disposal techniques may
prove to be the only pollution control strategies available short of elimina-
ting paraquat use by a return to conventional tillage. Methods for improving
disposal techniques have been discussed in an earlier general section on dis-
posal. Strategies for improving the efficiency of paraquat's use in agri-
culture would be similar to those strategies discussed in the section on
atrazine, and would include substitution with ground applications in areas
where paraquat is applied by aerial equipment.
Paraquat: Summary and Conclusions--
Before considering the allocation of funds for the implementation of
SWCPs for reducing paraquat transport in runoff, the extent and importance
of the pollution caused by paraquat in runoff should be determined. On
the basis of currently available information, it appears that paraquat in
runoff does not cause serious pollution problems. Researchers have reported
that paraquat in runoff is completely and irreversibly adsorbed to clay
particles and hence is not available to fauna and flora. Thus, although
losses of paraquat on eroded soil particles are relatively large, these
losses appear to be environmentally innocuous, and little incentive exists
for controlling them.
However, there may be unusual situations in which contamination problems
will arise. Soils containing high proportions of organic matter and sand
adsorb paraquat in a rather weak manner and may exhibit phytotoxicity until
the residues are redistributed onto deactivating clay particles. If these
soil particles are carried in runoff to aquatic ecosystems, desorption may
occur and cause contamination if deactivating clays are not suspended in
the water or found in the bottom mud.
The most effe.ctive way to reduce paraquat contamination is to eliminate
its'use. Since paraquat's primary use is in no till corn production, para-
quat's use can be eliminated by changing to an alternative tillage system.
Such a change in tillage would not only eliminate the paraquat in runoff _
but would also eliminate the apparently more serious problems associated with
the exposure of farm workers and non-target species to drift losses.
Although only a small fraction of the paraquat used in agriculture is
applied by aerial equipment, the drift losses which result from aerial
279
-------
applications appear to pose a more serious hazard to farm workers and to
aquatic ecosystems in these areas than do the bound paraquat residues trans-
ported in runoff. Hence, in areas where aerial applications are used, their
replacement with ground applications would be more effective than any SWCP
in reducing paraquat pollution.
With practices such as no tillage which increase application rates,
SWPCs are very effective at reducing paraquat losses in runoff. They also
do not appear to have any negative impacts such as increasing paraquatfs
leaching losses. Hence, the implementation of SWCPs for the control of
water and sediment losses would not appear to increase paraquat contamina-
tion of surface waters. However, for most soil conditions, it does not ap-
pear that the hazards posed by paraquat1s runoff losses would justify allo-
cation of resources for the implementation of SWCPs specifically for the
control of paraquat's transport away from agricultural fields.
RESEARCH NEEDS
The preceding discussions of the five representative pesticides have
been based upon currently available information. In some cases this inform-
ation was not very extensive. Thus, it was often necessary to extrapolate
from data on related pesticides in order to draw conclusions about pollution
control strategies.
Research in the following areas would be beneficial in evaluations of
these five pesticides and would undoubtedly influence decisions regarding
the control of their residues in aquatic systems.
1. Methyl Parathion
a. Additional studies measuring methyl parathion1s runoff losses
should be initiated to reconcile results from field studies and
results from the simulation study done by Hydrocomp (Appendix F).
b. The environmental and health hazards associated with bound resi-
dues of methyl parathion should be evaluated to determine the
benefits of reducing the transport of bound residues.
2. Toxaphene
a. Studies on toxaphene's rate of desorption from runoff-trans-
ported sediment would help planners to decide what degree of
transport control is necessary.
b. Studies on the fate, persistence and chemical and physical
properties of toxaphene's active moieties would allow moni-
toring agencies to better evaluate the impact of toxaphene
use on terrestrial and aquatic ecosystems.
3. Carbofuran
a. Field studies should be initiated for a range of soil types to
280
-------
measure the impact of SWCPs on the amount of carbofuran carried
in runoff.
4. Atrazine
a. If SWCPs are being considered for the control of atrazine runoff,
research should be done to determine the impact of increased
vegetative cover (associated with minimum tillage techniques) on
pesticide degradation rates in soil.
b. If SWCPs prove to increase levels of pesticide residues in
agricultural soils, studies should be done to determine the
impact of these residues on soil microorganisms, earthworms,
and beneficial arthropod species.
c. The mutagenicity of atrazine should be further investigated in
order to evaluate the hazards associated with the use of the
material in agriculture.
5. Paraquat
a. The spatial distribution of paraquat use should be investigated
to determine if there are areas of heavy use where the clay
content of the soils is so unusually low that paraquat residues
would not be totally adsorbed.
b. Studies on the ultimate fate of paraquat in soil and in aquatic
systems should be carried out to determine if a significant
degree of desorption ever occurs.
In addition, research in the following areas would allow environmental
quality planners to make more complete evaluations of the hazards associated
with pesticide use and would further enable these planners to make more
accurate decisions concerning pollution control methods:
1. the amounts of pesticide loss via drift and volatilization from
agricultural fields under a range of climatic conditions and appli-
cation procedures,
2. the rate at which air-borne losses of pesticides are redeposited
on the earth's surface,
3. the rate at which pesticide degradation occurs in air and water,
4. the sublethal and secondary effects of pesticides and their toxic
synergistic effects with other pollutants,
5. the percent of the reported pollution incidents concerning pesti-
cides which occur as a result of improper disposal techniques,and
6. the effectiveness of changing application procedures to reduce
losses of pesticides from agricultural fields.
281
-------
SUMMARY AND CONCLUSIONS
In most cases SWCPs will reduce the amount of pesticide transported by
surface runoff. The exceptions to this rule are those SWCPs which require
increases in pesticide use which are larger than the reductions obtained in
pesticide runoff losses. A summary of the runoff losses of pesticides and
the effectiveness of two SWCPs in reducing these losses is given in Table 9-11.
The table indicates that the amount of pesticide carried in runoff usually
consists of only a small fraction of the total amount of material applied.
The small fraction which is carried in runoff can be effectively reduced
by terraces with contours because terracing reduces the amount of runoff
water as well as the amount of sediment transport. Less expensive, non-
structural SWCPs appear to be less effective in reducing the transport of
moderately adsorbed pesticides but are quite effective in reducing the trans-
port of the few strongly adsorbed pesticides currently in use. It is possible
that in some circumstances SWCPs might increase the amounts of pesticide
leaving target areas through leaching; however, in most cases these losses
would be relatively minor in comparison with reductions in runoff losses.
Pesticides differ from sediment in that their primary transport route
away from target areas is often by drift or volatilization. In cases where
pesticides are applied by aerial equipment, the amount of pesticide trans-
ported to non-target areas by air may be as much as several hundred times as
large as the amount transported by runoff. Unfortunately, an evaluation of
the relative impact of air-borne and runoff losses on water quality is not
possible because studies have not been conducted to measure actual inputs of
air-borne and water-borne pollutants to surface waters. However, in those
cases where air-borne losses are almost 100 times larger than runoff losses,
the impact of alternative pollution control methods on air-borne losses as
well as on water borne losses should be considered.
In the cases of methyl parathion and toxaphene, which are applied pri-
marily by aerial equipment, the amounts of toxic material lost in drift can
be as much as two orders of magnitude greater than the amounts lost in run-
off. Because of these large air-borne losses to non-target areas, changing
application procedures by replacing aerial applications with ground appli-
cations would be more effective at reducing methyl parathion and toxaphene
transport than would SWCPs. However, improvements in pesticide efficiency
through the implementation of IPM programs seem to be the most cost-effec-
tive means of preventing the contamination of non-target areas by toxaphene
and methyl parathion since both air-borne and water-borne losses are reduced
with improved pest control methods that usually have no net cost to the
farmer.
The most effective method for stopping paraquat transport is to elim-
inate its use. In many cases, this can be done by replacing no-till corn
with an alternative tillage practice. SWCPs are also quite effective at
controlling the amount of paraquat lost with eroded sediment; however, because
this material is so tightly bound under most soil conditions, paraquat leav-
ing agricultural fields in runoff does not appear to pose any dangers to
aquatic systems and thus does not need to be controlled. On the other hand,
the air-borne losses of paraquat are very hazardous and should be controlled
282
-------
TABLE 9-11. REDUCTIONS IN RUNOFF LOSSES WITH VARIOUS POLLUTION CONTROL STRATEGIES
Pesticide
Methyl
Parathion
Toxaphene
Carbofuran
Atrazine
Paraquat
Type of Study
Field
Modelling
Field
Modelling
Field
Modelling
Field
Modelling
Field
Modelling
Application
Rate
13400 g/ha
6720 g/ha
26900 g/ha
•^1120 g/ha
•x-2910 g/ha
2910 g/ha
1120 g/ha
2520 g/ha
Runoff
Losses
0.008 - 0.25%
Cl.l - 33.5 g/ha)
4.6%
(310.7 g/ha)
0.078 - 0.72%
(21 - 193.7 g/ha)
0.05 - 2.0%
(0.56 - 22 g/ha)
1.0 - 16%
(29.1 - 465.6 g/ha)
1.8%
(53.5 g/ha)
1.28%
(14.3 g/ha)
13.5%
(340.6 g/ha)
Runoff Losses With SWCP
Contours
0.007 - 0.21% 2
(0.91 - 27.8 g/ha)
3.8%
(256.5 g/ha)
•vO.052 - 0.48% 3
(-X.14 - 127.9 g/ha)
T-0.04 - 1.6% ••
(0.45 - 17.6 g/ha)
0.76 - 12.16% 2
(23.12 - 353.9 g/ha)
1.4%
(40.9 g/ha)
0.85% 2
(9.4 g/ha)
8.9%
(225 g/ha)
Terraces With Contours
0.004 - 0.125% 2
(0.55 - 16.75 g/ha)
2.3%
154.7 g/ha
•x.0.037 - 0.35% 3
(10.1-93 g/ha)
n-0.02 - 0.8% *
(M>.22 - 8.8 g/ha)
0.36 - 5.8 2
(10.5 - 167.6 g/ha)
0.65%
(19.0 g/ha)
0.62% 2
(6.9 g/ha)
6.4%
(162. S g/ha)
Runoff Losses With
IPM Programs l
0.0043 - 0.135%
(0.58 - 18.1 g/ha)
0.048 -0.44%
(12.91 - 118.4 g/ha)
0.025 - 1.0%
(0.28 - 11 g/ha)
00
U)
'Assumes that reductions in runoff losses are equal to reductions in rate of application.
2Estimates of SWCP reductions in pesticide runoff losses made by using modelling results for SWCPs and runoff losses of pesticides in empirical studies.
3 Estimates based on reductions in paraquat' s runoff losses shown in modelling st-jdy of Beyerlein and Donigian (Appendix F) and runoff losses of to.xaphene
seen in empirical study.
''Estimates based on runoff losses, of carbofuran seen in empirical study and modelling results of Beyerlein and Donigian (Appendix F for atrazine and
water losses in runoff.
Beyerlein and Donigian study found:
•reductions of atrazine in runoff with contours » -24%
•reductions in water runoff losses with contours = -11%
•reductions of atrazine in runoff with terraces with contours - -64%
•reductions in water runoff losses with terraces with contours = -36%
Thus, '.e estimated that carbofuran w^uld have 20% reductions in runoff losses with contours, and 60% reductions when terraces with contours ar= used
-------
especially in those areas where paraquat is applied by aerial equipment.
Atrazine differs from paraquat in that its losses in runoff are avail-
able to biological organisms, thus the control of these losses can be bene-
ficial to the environment. Programs to increase the efficiency of atrazine
use have not been developed at this point in time; hence, SWCPs like strip-
cropping and terraces with contours may prove to be the best pollution control
methods available. Substitution with less persistent herbicidal materials
will also reduce the runoff losses of toxic materials from agricultural fields
(see Section 8). As mentioned previously, the losses of atrazine are quite
small except when the material is applied by ground equipment in windy
conditions and when the material is applied by aerial equipment. Volatiliza-
tion losses can be larger than either the drift or runoff losses of atrazine;
however, little is known about the relative hazards that drift, volatiliza-
tion, and runoff losses of atrazine pose for water quality.
Because carbofuran is usually applied in a granular formulation with
ground equipment and because its volatilization losses are insignificant,
carbofuran has the lowest air-borne losses of all the pesticides examined.
As a result, the major transport mode of carbofuran is runoff; thus, SWCPs
may have a significant impact on the total amount of carbofuran leaving crop
fields. The only field data on SWCPs and their impact on carbofuran's
runoff losses showed that the runoff losses of carbofuran from no-till and
conservation tillage corn grown on claypan soils were actually slightly
higher than the runoff losses observed with conventional tillage. However,
the Hydrocomp model (Appendix F") predicts that the use of terraces and con-
tours would result in a substantial reduction in the movement of atrazine, a
pesticide which would be expected to behave in a manner similar to carbofuran.
Although terraces with contours are probably effective at preventing runoff
of carbofuran, they are also usually quite expensive in comparison with other
methods for reducing carbofuran1s losses in runoff. The most effective of
these alternative methods for reducing transport eliminates the need for
carbofuran applications by rotating corn with an appropriate crop such as
soybeans. In many areas such a rotation is quite profitable. In those
areas where it is not, the use of insect population monitoring techniques
to determine when applications are needed for corn rootworm shows consider-
able promise for substantially reducing the amount of carbofuran applied.
Crop rotation and scouting appear to be as effective as terraces and contours
in reducing carbofuran transport and are usually much less expensive.
The results of the analysis of alternatives for each of the represen-
tative pesticides are summarized in Table 9-12. The qualitative ratings
are based upon the impact of each method on total transport of the material.
In order to make these calculations it was necessary to assume that a reduc-
tion in application rate would be accompanied by a proportional reduction in
runoff and air-borne losses from the target area. From Table 9-12 and the
preceding discussion it appears that atrazine is the only one of the rep-
resentative pesticides for which SWCPs might possibly be the most cost-
effective means of preventing environmental contamination.
Our analysis of the advantages of various strategies for reducing en-
vironmental contamination from pesticides has been based on an analysis of
284
-------
TABLE 9-12. SUMMARY EVALUATION
co
Reduction of Toxic Material
Entering the Environment^
Atrazine
Carbofuran
Methyl
Parathion
Paraquat
Toxaphene
Overall
Environmental
Rating1
6
12.0
9.7
8.0
10.0
With SWCP With IPM
poor —
very poor fair to
to poor excellent
very poor excellent
to poor
3
very poor
very poor excellent
With Changes in
Application
Procedures
very poor to
fair
very poor
fair to good
very poor to
fair
fair to good
The numbers represent the value of an index of environmental hazard developed by Metcalf fl975). The
index is based upon both toxicity and persistence data. The values for the insecticides are directly
from Metcalf (1975). The values for the herbicides were computed by the same method. The values
for insecticides in Metcalf's study ranged from a low of 4.3 for Trichlorfon to the most hazardous
rating of 14.0 for Endrin.
2
The evaluation is based on the average reduction as a percentage of the material originally applied.
Very poor corresponds to a reduction of < 2%, poor to a reduction of < 5%, fair to a reduction of
< 15%, good to a reduction of between 15% and 25% and excellent to a reduction of over 25%.
Low rating is because the paraquat in sediment appears to lose its toxicity.
-------
the amounts and toxicities of the materials transported by air and water.
However, there is considerable uncertainty about both the amounts of material
transported and their toxic impact on the environment. The uncertainty about
the amounts transported is due not only to the lack of extensive monitoring
but also to the limitations of the monitoring techniques themselves. For
example, using 14 carbon labelled insecticide - Lichtenstein et_ al. (1977)
discovered bound residues of methyl parathion that currently available moni-
toring techniques cannot detect. Similarly, the impact of pesticides on the
aquatic and terrestrial environments is very difficult to assess: adsorption
and desorption phenomena constantly change concentrations of toxic materials
in water and soil, and sublethal, secondary and synergistic effects of pesti-
cides on public health and aquatic and terrestrial ecosystems are not well
understood. An example of these sublethal effects are the psychiatric dis-
orders associated with exposure to organophosphorus insecticides. Studies
establishing the mutagenicity of certain pesticides are equally disturbing.
Thus, given the limitations of pollution monitoring techniques and a
multitude of unanswered questions about the sublethal, secondary and
synergistic toxicities of the pesticides presently used in agriculture, it
appears that pollution control methods which reduce the use of pesticides
have a considerable advantage over methods which reduce the movement of
pesticides that have already been released into the environment. Methods
which decrease the usage of pesticides greatly reduce the potential for pol-
lution events: less pesticide is available for the occurrence of runoff
losses, air-borne transport, human and wildlife poisonings, accidental spills,
and improper disposal. In certain cases, integrated pest management programs
which reduce pesticide use may also reduce the energy needs and production
costs of crop production. For those situations where such a reduction in
pesticide use is not possible, prophylactic methods like SWCPs may prove to
be the best way of containing toxic pesticide materials within agricultural
areas. However, there is a great potential for improving the efficiency of
pesticide use in U.S. agriculture thereby reducing the amounts of pesticide
required for pest control; thus, emphasis should be placed on the development
and implementation of programs which utilize biological and cultural pest
control methods and improve the placement, timing and equipment used in pesti-
cide application.
286
-------
REFERENCES
Adair, H.M., F.A. Harris, M.V. Kennedy, M.L. Laster, and E.D. Threadgill. 1971,
Drift of Methyl Parathion Aerially Applied Low Volume and Ultra Low Volume.
Journal of Economic Entomology 64(3):718-721.
Adams, J.E. 1974. Residual Effects of Crop Rotations on Water Intake,
Soil Loss, and Sorghum Yield. Agronomy Journal 66:299-304.
Adams, R.T. and P.M. Kurisu. 1976. Simulation of Pesticide Movement
on Small Agricultural Watersheds. EPA-600/3-76-066. U.S. Environmental
Protection Agency, Athens, Georgia.
Adrilenas, P.A. 1974. Farmers Use of Pesticides in 1971 - Quantities,
U.S. Department of Agriculture, Agricultural Economic Report No. 252.
Washington, D.C.
Alt, K., J.A. Miranowski, and E.O. Heady. 1979. Social Costs and
Effectiveness of Alternative Nonpoint Pollution Control Policies. In:
R.C. Loehr, D.A. Haith, M.F. Walter and C. Martin, eds. Best Management
Practices for Agriculture and Silviculture, Ann Arbor Science, Ann Arbor,
Michigan, pp. 321-328.
Amemiya, M. 1970. Land and Water Management for Minimizing Sediment. In:
T.L. Willrich and G.E. Smith, eds. Agricultural Practices and Water Quality.
The Iowa State University Press, Ames, Iowa.
American Society of Civil Engineering. 1977. Quality Aspects of Agricultural
Runoff and Drainage. In: Proceedings by the Task Committee on Agricultural
Runoff and Drainage of the Water Quality of the Irrigation and Drainage
Division. Irrigation and Drainage Division. American Society of Civil
Engineers. Vol. 103, No. IR4, pp. 475-495.
Apperson, C.S., R. Elston and W. Castle. 1976. Biological Effects and
Persistence of Methyl Parathion in Clear Lake, California. Environmental
Entomology 5(6):1116-1120.
Arthur, R.D., J.D. Cain and B.F. Barrentine. 1976. Atmosphere Levels of
Pesticides in the Mississippi Delta. Bulletin of Environmental Contamination
and Toxicology 15(2):129-134.
Asmussen, L.E., A.W. White, Jr., D.W. Hauser and J.M. Sheridan. 1977.
Reduction of 2,4-D Load in Surface Runoff Down a Grassed Waterway. Journal
of Environmental Quality 6:159-163.
287
-------
Audus, L.J. 1952. The Decomposition of 2,4-dichlorophenoxyacetic Acid
in the Soil. Journal of the Science of Food and Agriculture 3:268-274.
Baeumer, K. and W.A.P. Bakermans. 1973. Zero-tillage. Advances in
Agronomy 25:77-123.
Bailey, G.W., A.P. Barnett, W.R. Payne, and C.N. Smith. 1974. Herbicide
Runoff From Four Coastal Plain Soil Types. EPA-660/2-74-017. U.S.
Environmental Protection Agency, Washington, D.C.
Bailey, G.W. and J.L. White. 1970. Factors Influencing the Adsorption,
Desorption and Movement of Pesticides in Soil. Residue Reviews 32:29-92.
Bailey, G.W., R.R. Swank, and H.P. Nicholson. 1974. Predicting Pesticide
Runoff from Agricultural Land: A Conceptual Model. Journal of Environ-
mental Quality 3(2):95-102.
Baker, J.L. and H.P. Johnson. 1977. Tillage System Effects on Runoff
Water Quality: Pesticides. ASAE Paper No. 77-2504B. American Society
of Agricultural Engineers, St. Joseph, Michigan.
Baker, J.L., H.P. Johnson, M.A. Borcherding, and W.R. Payne. 1979. Nutrient
and Pesticide Movement From Field to Stream: A Field Study. In: R.C. Loehr,
D.A. Haith, M.F. Walter and C. Martin, eds. Best Management Practices for
Agriculture and Silviculture. Ann Arbor Science, Ann Arbor, MI. pp. 213-246.
Baker, J.L., J.M. Laflen and H.P. Johnson. 1978. Effects of Tillage Systems
on Runoff Losses of Pesticides. A Rainfall Simulation Study. Transactions
of the American Society of Agricultural Engineers 21:886-892.
Baldwin, F.L., P.W. Santelmann and J.M. Davidson. 1975. Movement of
Fluometuron Across and Through the Soil. Journal of Environmental Quality
4:101-194.
Barisas, S.G., J.L. Baker, H.P. Johnson and J.M. Laflen. 1978. Effect of
Tillage Systems on Runoff Losses in Nutrients, A Rainfall Simulation Study.
Transactions of the American Society of Agricultural Engineers 21:893-897.
Barrows, H.L. and V.J. Kilmer. 1963. Plant Nutrient Losses From Soils by
Water Erosion. Advances in Agronomy 14:303-316.
Barthel, W.F., J.C. Hawthorne, J.H. Ford, G.C. Bolton, L.L. McDowell,
E.H. Grissinger, and D.A. Parsons. 1969. Pesticide Residues in Sediments
of the Lower Mississippi River and Its Tributaries. Pesticides Monitoring
Journal 3:8-66.
Bateman, H.P. and T.D. Hinesly. 1967. Should You Fall-Plow or Spring-Plow?
Circular 972, Cooperative Extension Service, University of Illinois, Urbana,
Illinois.
Baver, L.D., W.H. Garner and W.R. Gardner. 1972. Soil Physics. 4th ed.
John Wiley and Sons, Inc., New York.
288
-------
Beasley, R.B. 1972. Erosion and Sediment Pollution Control. Iowa State
University Press, Ames, Iowa.
Beasley, R.S. 1976. Contribution of Subsurface Flow From the Upper Slopes
of Forested Watersheds to Channel Flow. Soil Science Society of America
Journal 40:955-957.
Bejer-Petersen, B., R.R. Hermansen and M. Weihe. 1972. On the Effects of
Insecticide Spraying in Forests on Birds Living in Nest Boxes. Dansk
Ornithologisk Forening 66(1-2):30-50.
Bendizen, T.W., R.D. Hill, F.T. Dubyne and G.G. Robic. 1969. Cannery
Wastes Treatment by Spray Irrigation Runoff. Journal of the Water Pollution
Control Federation 41:385-391.
Bennett, H.H. 1939. Soil Conservation. McGraw Hill, New York.
Bennett, O.L., E.L. Mathias, and C.B. Sperew. 1976. Double Cropping for
Hay and No-Tillage Corn Production as Affected by Sod Species With Rates
of Atrazine and Nitrogen. Agronomy Journal 68:250-254.
Bidleman, T.F. and C.E. Olney. 1975. Long Range Transport of Toxaphene
Insecticide in the Atmosphere of the Western North Atlantic. Nature
257:475-477.
Biggar, J.W. and R.B. Corey. 1969. Agricultural Drainage and Eutrophi-
cation. In: Eutrophication: Causes, Consequences, Correctives. National
Academy of Sciences, Washington, D.C. pp. 404-445
Bisal, F. 1960. The Effect of Raindrop Size and Impact Velocity on Sand
Splash. Canadian Journal of Soil Science 48(2):242-245.
Black, C.A. 1958. Soil-Plant Relationships. John Wiley and Sons, Inc.,
New York.
Black, T.A., W.R. Gardner and G.W. Thurtell. 1969. Prediction of Evapora-
tion, Drainage and Soil Water Storage for a Bare Soil. Soil Science Society
of America Proceedings 33:655-660.
Blevins, R.L., G.W. Thomas and P.L. Cornelius. 1977. Influence of No-tillage
and Nitrogen Fertilization on Certain Soil Properties After 5 Years of
Continuous Corn. Agronomy Journal 69:383-386.
Boon, W.R. 1964. The Chemistry and Mode of Action of the Bipyridylium
Herbicides Diquat and Paraquat. Outlook on Agriculture 4:4,163.
Booram, C.V., Jr. and L.E. Asmussen. 1976. A Technique for Evaluating
Chemical Movement From Land Application of Agricultural Chemicals. ASAE
Paper No. 76-2068. American Society of Agricultural Engineerings, St.
Joseph, Michigan.
289
-------
Bouldin, D.R., A.H. Johnson and D.A. Lauer. 1975. The Influence of Human
Activity on the Export of Phosphorus and Nitrate from Fall Creek. In:
K.S. Porter, ed. Nitrogen and Phosphorus: Food Production, Waste and
the Environment. Ann Arbor Science. Ann Arbor, Michigan, pp. 61-122.
Bouwer, H. 1976. Infiltration Into Increasingly Permeable Soils. American
Society of Civil Engineers Irrigation and Drainage Division Journal 102(1):
127-136.
Bovey, R.W., E. Burnett, R.E. Meyer, C. Richardson and A. Loh. 1978a.
Persistence of Tebuthiuron in Surface Runoff Water, Soil, and Vegetation
in the Texas Blacklands Prairie. Journal of Environmental Quality 7(2):
233-236.
Bovey, R.W., C. Richardson, E. Burnett, M.G. Merkle, and R.E. Meyer. 1978b.
Loss of Spray and Pelleted Picloram in Surface Runoff Water. Journal of
Environmental Quality 7(2):178-180.
Bower, C.A., W.R. Gardner and J.O. Goertzen. 1957. Dynamics of Cation
Exchange in Soil Columns. Soil Science Society of America Proceedings
21:20-24.
Bradley, J.R., T.J. Sheets and M.D. Jackson. 1972. DDT and Toxaphene
Movement in Surface Water From Cotton Plots. Journal of Environmental
Quality 1(1):102-105.
Brady, N.C. 1974. The Nature and Properties of Soils. 8th Edition,
MacMillen, New York.
Brazzel, J.R., W.W. Watson, J.S. Hursh and M.H. Adair. 1968. The Relative
Efficiency of Aerial Application of Ultra-Low-Volume and Emulsifiable
Concentrate Formulations of Insecticides. Journal of Economic Entomology
61(2):408-413.
Brown, G.C. and C.H. Shanks, Jr. 1976. Mortality of Two-Spotted Spider
Mite Predators Caused by the Systemic Insecticide, Carbofuran. Environmental
Entomology 5(6):1155-1159.
Brown, H. 1976. Infiltration Into Increasingly Permeable Soils. Journal
of Irrigation and Drainage Division. Proceedings of the American Society of
Civil Engineers 102:127-136.
Bruce, R.R., L.A. Harper, R.A. Leonard, W.M. Snyder and A.W. Thomas. 1975.
A Model for Runoff of Pesticides From Small Upland Watersheds. Journal
of Environmental Quality 4:541-548.
Bubenzer, G.D. and B.A. Jones, Jr. 1971. Drop Size and Impact on the
Detachment of Soils Under Simulated Rainfall. Transactions of the
American Society of Agricultural Engineers 14(4):625-628.
Burgett, M. and G. Fisher. 1977. The Contamination of Foraging Honey
Bees and Pollen With Penncap-M. American Bee Journal 117:626-627.
290
-------
Bums, I.G. and M.H.B. Hayes. 1974. Some Physico-Chemical Principles In-
volved in the Adsorption of the Organic Cation Paraquat by Soil Humic
Materials. Residue Reviews 52:117-148.
Burns, I.G., M.H.B. Hayes and M. Stacey. 1973. Some Physio-Chemical
Interactions of Paraquat With Soil Organic Materials and Model Compounds.
Weed Research 13:67-90.
Burnside, O.C. 1974. Prevention and Detoxification of Pesticide Residues
in Soil. In: W.D. Guenzi, ed. Pesticides in Soil and Water. Soil Science
Society of America, Inc. Madison, Wisconsin, pp. 387-412.
Burnside, O.C., C.R. Fenster, and G.A. Wicks. 1963. Dissipation and
Leaching of Monuron, Simazine, and Atrazine in Nebraska Soils. Weeds 11:
209-213.
Burnside, O.C., C.R. Fenster and G.A. Wicks. 1971. Soil Persistence of
Repeated Annual Applications of Atrazine. Weed Science 19:290-293.
Burwell, R.E. and W.E. Larson. 1969. Infiltration as Influenced by Tillage
Induced Random Roughness and Pore Space. Soil Science Society of America
Proceedings 33:449-452.
Butler, G. 1977. Algae and Pesticides. Residue Reviews 66:19-62.
Butler, P.A. and R.L. Schutzmann. 1978. Residues of Pesticides and PCBs
in Estuarine Fish, 1972-76: National Pesticide Monitoring Program. Pesti-
cides Monitoring Journal 12 (2):51-59.
Byass, J.B. and J.R. Lake. 1977. Spray Drift From a Tractor-Powered Field
Sprayer. Pesticide Science 8:117-126.
Calderbank, A. 1968. The Biphyridylium Herbicides. In: R.L. Metcalf, ed.
Advances in Pest Control Research. Interscience Publishers, John Wiley and
Sons, New York 8:127-236.
Calderbank, A. and P. Slade. 1976. Diquat and Paraquat. In: P.C. Kearney
and D.D. Kaufman, eds. Herbicides: Chemistry, Degradation, and Mode of
Action, Volume 2. Marcel Dekker, Inc., New York. pp. 501-540.
Campbell, J.K. 1973. Selecting Field Machinery. Agricultural Engineering
Extension Bulletin No. 395. Cornell University, Ithaca, New York.
Campbell, K.L. 1976. Nutrient Transport in Runoff From Sandy Soils.
ASAE Paper No. 76-2547. American Society of Agricultural Engineers, St.
Joseph, Michigan.
Carey, A.E., J.A. Gowan, H. Tai, W.G. Mitchell and G.B. Wiersma. Pesticide
Residue Levels in Soils and Crops in 1971 - National Soils Monitoring
Program (III). Draft Report, Washington, D.C.
291
-------
Caro, J.H. 1976. Pesticides in Agricultural Runoff. In: Control of
Water Pollution From Cropland, Volume 2, B. Stewart Co-ordinator, EPA-600/2-
75-026a. U.S. Environmental Protection Agency Report, U.S. Department of
Agriculture, Washington, D.C.
Caro, J.H., H.P. Freeman and B.C. Turner. 1974. Persistence in Soil and
Losses in Runoff of Soil-Incorporated Carbaryl in a Small Watershed.
Journal Agriculture and Food Chemistry 22(5):860-863.
Caro, J.H., H.P. Freeman, D.E. Glotfelty, N.C. Turner and W.M. Edwards.
1973. Dissipation of Soil-Incorporated Carbofuran in the Field. Journal
of Agricultural Food Chemistry 21:1010-1015.
Carter, C.E., C.W. Doty and B.R. Carrell. 1968. Runoff and Erosion
Characteristics of the Brown Loam Soils. Agricultural Engineering 49:296.
Casey, J.E., R.D. Lacewell and W. Sterling. 1975. An Example of Economically
Feasible Opportunities for Reducing Pesticide Use in Commercial Agriculture.
Journal of Environmental Quality 4(l):60-64.
Casida, J.E., R.L. Holmstead, S. Khalifa, J.R. Knox and T. Ohsawa. 1974-
Toxaphene Insecticide: A Complex Biodegradable Mixture. Science 183:520-
521.
Casler, G.L. and J.J. Jacobs. 1975. Economic Analysis of Reducing Phosphorus
Losses From Agricultural Production. In: K.S. Porter, ed. Nitrogen and
Phosphorus: Food Production, Waste and the Environment. Ann Arbor
Science, Ann Arbor, Michigan, pp. 169-215.
Chao, T.T., M.E. Harward and S.C. Fang. 1965. Exchange Reactions Between
Hydroxyl and Sulfate Ions in Soil. Soil Science 99:104-108. Referenced by
Holt et^ al. In: T.L. Willrich and G.E. Smith, ed., 1970, Agricultural
Practices and Water Quality. Iowa State University Press, Ames, Iowa.
pp. 11-34.
Chichester, F.W., J.O. Legg and G. Stanford. 1975. Relative Mineralization
Rates of Indegous and Recently Incorporated N-_ Labelled Nitrogen. Soil
Science 120(6):455-460. lb
Chichester, F.W. 1976. The Impact of Fertilizer Use and Crop Management
on Nitrogen Content of Subsurface Water Draining From Upland Agricultural
Watersheds. Journal of Environmental Quality 5:413-416.
Chichester, F.W. 1977. Effects of Increased Fertilizer Rates on Nitrogen
Content of Runoff and Percolate from Monolith Lysimeters. Journal of
Environmental Quality 6(2):211-217.
Chu, S.T. 1977. Modeling Infiltration During a Variable Rain. ASAE
Paper No. 77-2063. American Society of Agricultural Engineers, St. Joseph,
Michigan.
292
-------
Clapp, J.G. 1972. No-Tillage Soybean Production. North Carolina Agri-
cultural Extension Service Circular 537, Raleigh, N.C.
Coates, G.E., H.H. Funderburk, J.M. Lawrence and D.E. Davis. 1965. Studies
on Translocation, Degradation and Factors Affecting the Persistence of
Diquat and Paraquat. Proceedings of the Southern Weed Conference. 18:614.
Cohen, J. M. and C. Pinkerton. 1966. Widespread Translocation of Pesticides
by Air Transport and Rain-out. In: Robert Gould, ed. Organic Pesticides
in the Environment. Advances in Chemistry Series. American Chemical Society,
Washington, D.C. 60:163-176.
Colbert, F.O., V.V. Volk and A.P. Appleby. 1975. Sorption of Atrazine,
Terbutryu and GS-14254 on Natural and Lime-Amended Soils. Weed Science
23:390-395.
Council on Environmental Quality. 1975. Environmental Quality.
Superintendents of Documents. U.S. Government Printing Office, Washington,
D.C.
Crawford, N.H. and A.S. Donigian. 1973. Pesticide Transport and Runoff
Model for Agricultural Lands. EPA-660/2-74-013. U.S. Environmental
Protection Agency, Washington, D.C.
Crosby, D.G. 1975. Herbicide Photodecomposition. In: P.C. Kearney and
D.D. Kaufman, eds. Herbicides: Chemistry, Degradation, and Mode of Action.
Marcel Dekker, Inc., New York 2:835-890.
Crosby, D.G. 1978. Atmospheric Transport and Transformation of Pesticides.
EPA-9-78-003, U.S. Environmental Protection Agency, Athens, Georgia.
Damanakis, M. , D.S.H. Drennan, J.D. Fryer and K. Holly. 1970. The Adsorp-
tion and Mobility of Paraquat on Different Soils and Soil Constituents.
Weed Research 1.0:264-277.
Damanakis, M., D.S.H. Drennan, J.D. Fryer and K. Holly. 1970. The Toxicity
of Paraquat to a Range of Species Following Uptake by the Roots. Weed
Research 10:278-283.
Damanakis, M., D.S.H. Drennan, J.D. Fryer and K. Holly. 1970. Availability
to Plants of Paraquat Adsorbed on Soil or Sprayed on Vegetation. Weed
Research 10:305-315.
Davidson, J.M., G.H. Brusewitz, D.R. Baker and A.L. Wood. 1975. Use of
Soil Parameters for Describing Pesticide Movement Through Soils. EPA 660/2-
75-009. U.S. Environmental Protection Agency, Corvallis, Oregon.
Davidson, J.M., R.S. Mansell and D.R. Baker. 1972. Herbicide Distributions
Within a Soil-Profile and Their Dependence Upon Adsorption-Desorption.
Proceedings of the Soil and Crop Science Society of Florida 32:36-41.
293
-------
Davidson, J.M., Li-Tse Ou, and P.S.C. Rao. 1978. Adsorption, Movement,
and Biological Degradation of High Concentrations of Selected Pesticides
in Soils. In: D.W. Shultz, ed. Land Disposal of Hazardous Wastes.
EPA-600/9-78^016, U.S. Environmental Protection Agency, Washington, D.C.
Dean, J.D. and L.A. Mulkey. 1978. Interactive Effects of Pesticides and
Selective Conservation Practices on Runoff Losses - A Simulitude Study.
In: R.C. Loehr, D.A. Haith, M.F. Walter and C. Martin, eds. Best Manage-
ment Practices for Agriculture and Silviculture. Ann Arbor Science,
Ann Arbor, Michigan, pp. 715-734.
Dendy, F.E. and W.A. Champion. 1978. Sediment Deposition in United States
Reservoirs. United States Department of Agriculture. Miscellaneous
Publications No. 1362. U.S. Department of Agriculture, Agricultural
Research Service, Oxford, Mississippi.
Dendy, F.E., W.A. Champion and R.B. Wilson. 1973. Reservoir Sedimentation
Surveys in the United States. Geophysical Monograph Series. American
Geophysical Union 17:349-357.
Deubert, K.H. and R.S. Gray. 1976. Parathion Residues in Environmental
Samples From Untreated Areas. Bulletin of Environmental Contamination
and Toxicology 15:613-616.
Dickey, E.G., D.H. Vanderholm, J.A. Jackobs and S.L. Spahr. 1977. Vegeta-
tive Filter Treatment of Feedlot Runoff. ASAE Paper No. 77-4581. American
Society of Agricultural Engineers, St. Joseph, Michigan.
Dixon, R.M. and A.E. Peterson. 1971. Water Infiltration Control: A
Channel System Concept. Soil Science Society of America Proceedings
35:968-973.
Donigian, A.S. and N.H. Crawford. 1976. Modelling Pesticides and Nutrients
on Agricultural Lands. EPA-600/2-76-043, U.S. Environmental Protection
Agency, Athens, Georgia.
Donigian, A.S., D.C. Beyerlein, H.H. Davis and N.H. Crawford. 1977.
Agricultural Runoff Management Model Version II: Refinement and Testing.
EPA-COO/3-77-098, U.S. Environmental Protection Agency, Athens, Georgia.
Doster, D.H., and J.A. Phillips. 1973. Costs, Inputs and Returns: Humid
and Subhumid Areas. Conservation Tillage, Proceedings of a National
Conference, Soil Conservation Society of America, Ankeny, Iowa.
Doyle, R.C., D.C. Wolf and D.F. Bezdicek. 1974. Effect of Forest Buffer
Strips in Improving the Water Quality of Manure Polluted Runoff. In:
Proceedings of the Third International Symposium on Livestock Wastes".
Managing Livestock Wastes. ASAE Publication PROC-275, American Society
of Agricultural Engineers, St. Joseph, Michigan.
294
-------
Doyle, R.C., G.C. Stanton and D.C.- Wolf. 1977. Effectiveness of Forest
and Grass Buffer Strips in Improving the Water Quality of Manure Polluted
Runoff. ASAE Paper No. 77-2501. American Society of Agricultural Engineers,
St. Joseph, Michigan.
Duggan, R.E. and M.B. Duggan. 1973. Pesticide Residues in Food. In_:
C.A. Edwards, ed., Environmental Pollution by Pesticides. Plenum Press,
London, pp. 334-364.
Duley, F.L. and M.F. Miller. 1923. Erosion and Surface Runoff Under
Different Soil Conditions. Missouri Agricultural Experiment Station,
Bulletin No. 203.
Duley, F.L. 1939. Surface Factors Affecting the Rate of Intake of Water
by Soils. Soil Science Society of America Proceedings 4:60-64.
Dunne, T. 1978. Field Studies of Hillslope Flow Processes, jta: J.M.
Kirkby, ed. Hillslope Hydrology. John Wiley and Sons, New York.
Duttweiler, D.W. and S.G. Malakhov. 1977. U.S.A. - U.S.S.R. Symposium
on Environmental Transport and Transformation of Pesticides. Journal
of Agricultural and Food Chemistry 25(5):975-978.
Eckern, P.C., Jr. and R.F. Muckenhirn. 1974. Waterdrop Impact as a
Force in Transporting Sand. Soil Science Society of America Proceedings
12:441-44.
Edwards, C.A. 1977. Nature and Origins of Pollution of Aquatic Systems
by Pesticides. In: Mohammed A.W. Khan, ed. Pesticides in Aquatic
Environments. Plenum Press, New York. pp. 11-38.
Edwards, O.A. and A.R. Thompson. 1973. Pesticides and Soil Fauna.
Residue Reviews 45:1-79.
Edwards, W.M. 1972. Agricultural Chemical Pollution as Affected by Reduced
Tillage Systems. Proceedings No Tillage Systems Symposium, Ohio State
University. Columbus, Ohio. pp. 30-40.
Edwards, W.M., F.W. Chichester and L.L. Harrold. 1970. Management of
Barnlot Runoff to Improve Downstream Water Quality. Proceedings of the
International Symposium Livestock Wastes. ASAE Publication PROC-271.
American Society of Agricultural Engineers, St. Joseph, Michigan.
Eggleston, K.O., E.K. Israelsen and J.P. Riley. 1971. Hybrid Computer
Simulation of the Accumulation and Melt Processes in a Snowpack. Utah
State University, Logan, Utah.
Eichelburger, J.W. and J.J. Lichtenberg. 1971. Persistence of Pesticides
in River Water. Environmental Science and Technology 5:541-544.
295
-------
Eisler, R. 1970. Acute Toxicities of Organochlorine and Organophosphorus
Insecticides to Estuarine Fishes. Technical Paper. U.S. Fisheries and
Wildlife Service 46:1.
Ellison, W.D. 1947. Soil Erosion Studies. Agricultural Engineering 28:
145-146, 197-201, 245-248, 297-300, 349-351, 353, 402-405, 408, 442-444,
450.
Enfield, C.G. and B.E. Bledsoe. 1975. Kinetic Model for Orthophosphate
Reactions in Mineral Soils. EPA-600/2-75-022. U.S. Environmental
Protection Agency, Corvallis, Oregon.
England, C.G. 1970. Land Capability: A Hydrologic Response Unit in
Agricultural Watersheds. ARS 41-172. Agricultural Research Service,
U.S. Department of Agriculture, Washington, D.C.
Epstein, E. and W.J. Grant. 1968. Chlorinated Insecticides in Runoff
Water as Affected by Crop Rotation. Soil Science Society of America
Proceedings 32:423-426.
Erbach, D.C. and W.G. Lovely. 1974. Weed Control With Conservation Tillage
Production of Corn and Soybeans. Journal Soil and Water Conservation 29:
46-47.
Esser, H.O., G. Dupuis, E. Ebert, G. Marco and C. Vogel. 1975. s-Triazines.
In: P.C. Kearney and D.D. Kaufman, eds. Herbicides: Chemistry, Degradation,
and Mode of Action, Volume 1. Marcel Dekker, Inc., New York. pp. 129-208.
Falayi, 0. and J. Bouma. 1975. Relationships Between the Hydraulic Con-
ductivity of Surface Crusts and Soil Management in a Typical Hapludalf.
Soil Science Society of America Proceedings 39:957-963.
Farm Chemicals Handbook. 1973. Meister Publishing Company, Willoughby, Ohio.
Farmer, W.J. and J. Letey. 1974. Volatilization Losses of Pesticides
from Soils. EPA-670/2-74-054, U.S. Environmental Protection Agency,
Washington, D.C.
Farmer, W.J., K. Igue and W.F. Spencer. 1973. Effect of Bulk Density on
the Diffusion and Volatilization of Dieldria from Soil. Journal of Environ-
mental Quality 2(1):107:109.
Fink, R.J. and Dean Wesley. 1974. Corn Yield as Affected by Fertilization
and Tillage System Agronomy Journal 66:70-71.
Forster, D.L., N. Rask, S.W. Bone and B.W. Schurle. 1976. Reduced Tillage
Systems for Conservation and Profitability. ESS 532. Department of
Agricultural Economics and Rural Sociology, Ohio State University,
Columbus, Ohio.
Foster, G.R. and L.F. Huggins. 1977. Deposition of Sediment by Overland
Flow on Concave Slopes. In: Soil Erosion: Prediction and Control. Soil
Conservation Society of America. Akeny, Iowa. pp. 167-180.
296
-------
Foster, G.R. and L.D. Meyer. 1975. Mathematical Simulation of Upland
Erosion Using Fundamental Erosion Mechanics. Inj Present and
Prospective Erosion Technology for Predicting Sediment Yields and Sources.
ARS-S-40. U.S. Department for Agriculture, Agricultural Research Service,
Washington, D.C. pp. 190-207.
Foster, G.R. and L.D. Meyer. 1977. Soil Erosion and Sedimentation by
Water - an Overview. Proceedings of the National Symposium on Soil Erosion
and Sedimentation by Water. ASAE Publication 4-77. American Society of
Agricultural Engineers, Chicago, Illinois.
Foster, G.R. and W.H. Wischmeier. 1973. Irregular Slopes and the Universal
Soil Loss Equation. ASAE Paper No. 73227. American Society of Agricultural
Engineers, St. Joseph, Michigan.
Fowkes, P.M., H.A. Benesi, L.B. Ryland, W.M. Sawyer, K.D. Detling, E.S.
Loeffler, F.B. Folckemer and M.R. Johnson. 1960. Clay-Catalyzed Decomposi-
tion of Insecticides. Journal of Agricultural and Food Chemistry 8:203-210.
Foy, C.L. 1963. Volatility of Various Herbicides Under Lab Conditions.
Research Progress Report: Western Weed Control Conference 81-82.
Free, G.R. 1960. Erosion Characteristics of Rainfall. Agricultural Engin-
eering 41(7):447-449, 450.
Free, G.R. 1970. Minimum Tillage for Corn Production. Cornell University
Agricultural Experiment Station. Bulletin No. 1030. Ithaca, New York.
Free, G.R. and C.E. Bay. 1969. Tillage and Slope Effects on Runoff and
Erosion. Transactions of American Society of Agricultural Engineers
12(2):209-211 and 215.
Freed, V.H. and R. Hague. 1974. Behavior of Pesticides in the Environment.
Residue Reviews 52:89-116.
Frere, M.H., C.A. Onstad and H.N. Holtan. 1975. ACTMO: An Agricultural
Chemical Transport Model. ARS-H-3. Agricultural Research Service,
U.S. Department of Agriculture, Washington, D.C.
Frink, C.R. 1969. Chemical and Mineralogical Characteristics of Eutrophic
Lake Sediments. Soil Science Society of America Proceedings 33:369-372.
Fryer, J.D., R.J. Hance and J-W. Ludwig. 1975. Long Term Persistence of
Paraquat in a Sandy Loam Soil. Weed Research 15:189-194.
Fuhremann, T.W. and E.P. Lichtenstein. 1978. Release of Soil-Bound Methyl
[14C] Parathion Residues and Their Uptake by Earthworms and Oat Plants.
Journal of Agricultural and Food Chemistry 26(3):605-610.
Fukuto, T.R. 1974. Carbamate Insecticides. In: R.L. Metcalf and
J.J. McKelvey, eds. The Future for Insecticides. Volume 6. John Wiley
and Sons, New York. pp. 313-348.
297
-------
Funderburk, H.H. and G.A. Bozarth. 1967. Review of the Metabolism and
Decomposition of Diquat and Paraquat. Journal of Agricultural and Food
Chemistry 15:563-567.
Funderburk, H.H. and J.M. Lawrence. 1964. Mode of Action and Metabolism
of Diquat and Paraquat. Weeds 12:259-264.
Gardner, W.R. 1965. Movement of Nitrogen in Soil In: W.V. Bartholomew
and F.E. Clark, eds. Soil Nitrogen. Agronomy, Volume 10. American
Society of Agronomy, Madison, Wisconsin.
Garriels, D., M. deBoodt and D. Minijauw. 1974. Dune and Stabilization
with Synthetic Soil Conditioners: A Laboratory Experiment of Splash
Erosion. Journal of Soil Science 118(5):332-338.
Gentile, J.M., E. D. Wagner and M.J. Plewa. 1977. The Detection of Weak
Recombinogenic Activities in the Herbicides Alachlor and Propachlor Using
a PIant-Activation Bioassay. Mutation Research 48:113-116.
Gerakis, P.A. and A.G. Sficas. 1974. The Presence and Cycling of Pesticides
in the Ecosphere. Residue Reviews 52:69-88.
Gerard, C.J., C.A. Burleson, W.R. Cowley, M.E. Bloodworth and S.H. Khan.
1962. Effect of Selected Cropping Systems on Cotton Production and the
Physicochemical Properties of a Coarse-Textured Soil. Publication MP-624,
The Agricultural and Mechanical College of Texas, Texas Agricultural
Experiment Station, College Station, Texas.
Gerhardt, P.O. and J.M. Witt. 1965. Pesticide Residues and Problems
Resulting from Drift from Aerial Applications of Dusts and Sprays.
In: Proceedings of the 12th International Congress of Entomology, (1964).
London.
Gershon, S. and F.H. Shaw. 1961. Psychiatric Sequelae of Chronic
Exposure to Organophosphorus Insecticides. Lancet 1-2:1371-1374.
Getzin, L.W. 1973. Persistence and Degradation of Carbofuran Soil.
Environmental Entomology 2(3):461-467.
Getzin, L.W. and I. Rosefield. 1968. Organophosphorus Insecticide
Degradation by Heat-Labile Substances in Soil. Journal of Agricultural
and Food Chemistry 16:598-601.
Getzin, L.W. and C.H. Shanks, Jr. 1970. Persistence, Degradation and
Bioactivity of Phorate and Its Oxidative Analogues in Soil. Journal of
Economic Entomology 63:52-58.
Ghadiri, H. and D. Payne. 1977. Raindrop Impact Stress and the Breakdown
of Soil Crumbs. Journal of Soil Science 28:247-258.
298
-------
Glooschenko, W.A., W.M.J. Strachan and R.C.J. Sampson. 1976. Distribution
of Pesticides and Pol/chlorinated Biphenyls in Water, Sediments, and Seston
of the Upper Great Lakes - 1974. Pesticides Monitoring Journal 10(2):61-67.
Glyraph, L.M. 1975. Evolving Emphasis in Sediment Yield Predictions.
Present and Prospective Technology for Predicting Sediment Yields and
Sources. ARS-S-40. U.S. Department of Agriculture, Agricultural
Research Service, Washington, D.C. pp. 1-4.
Goldberg, E.D., P. Butler, P. Meier, D. Menzel, G. Paulik, R. Risebrough
and L.F. Stickel. 1971. Chlorinated Hydrocarbons in the Marine Environment.
National Academy of Sciences, Washington, D.C.
Goswami, P. and R.E. Green. 1971. Microbial Degradation of the Herbicide
Atrazine and Its 2-Hydroxy Analog in Submerged Soils. Environmental Science
and Technology 5:426-429.
Graetz, D.A., G. Chesters, T.C. Daniel, L.W. Newland and G.B. Lee- 1970.
Parathion Degradation in Lake Sediments. Journal of the Water Pollution
Control Federation 42:R76-R94.
Graff, 0. von. 1969. Regenwurmtatigkeit in Ackerboden unter Verschiedenem
Bedeckungsmaterial, Gemessen an der Losungsablage. Pedo biologigia 9:120-127.
Graf, W.H. 1971. Hydraulics of Sediment Transport. McGraw Hill, New York.
Green, W.H. and G.A. Ampt. 1911. Studies on Soil Physics I: The Flow of
Air and Water Through Soils. Journal of Agricultural Science 4(l):l-24.
Grzenda, A.R., G.J. Lauer and H.P. Nicholson. 1964. Water Pollution by
Insecticides in an Agricultural River Basin II: The Zooplankton, Bottom
Fauna and Fish. Limnology and Oceanography 9:318.
Gunther, F.A. 1977. The Citrus Reentry Problem: Research on its Cause
and Effects and Approaches to its Minimization. Residue Reviews. Volume
67. Springer-Verlag, New York.
Gunther, F.A., W.E. Westlake and P.S. Jaglan. 1968. Reported Solubilities of
738 Pesticide Chemicals in Water. Residue Reviews 20:1-148.
Guyer, G.E., P.L. Adkisson, K. Dubois, C. Menzie, H. Page Nicholson and
G. Zweig. 1971. Toxaphene Status Report. Special Report to the Hazardous
Materials Advisory Committee, Office of Pesticide Programs, U.S. Environmental
Protection Agency., Washington, D.C.
Haan, C.T. 1971. Movement of Pesticides by Runoff and Erosion. Transaction
of American Society of Agricultural Engineers 4:445-447, 449.
Hague, R. and V.H. Freed. 1974. Behavior of Pesticides in the Environment:
Environmental Chemodynamics. Residue Reviews 52:89-116.
299
-------
Haith, D.A. 1973. Optimal Control of Nitrogen Losses From Land Disposal
Areas. Journal of Environmental Engineering Division, American Society of
Civil Engineers 99(EE6):923-937.
Haith, D.A., A. Koenig and D.P. Loucks. 1977. Preliminary Design of
Wastewater Land Application Systems. Journal Water Pollution Control
Federation 49(12):3271-3279.
Hall, J.K. 1974. Erosional Losses of s-Triazine Herbicides. Journal of
Environmental Quality 3(2):174-180.
Hall, J.K. and N.L. Hartwig. 1978. Atrazine Mobility in Two Soils Under
Conventional Tillage. Journal of Environmental Quality 7(l):63-68.
Hall, J.K., M. Pawlus, and E.R. Higgins. 1972. Losses of Atrazine in
Runoff Water and Soil Sediment. Journal of Environmental Quality 1(2):
172-176.
Hall, R.F., R.H. Dowdy and D.R. Timmons. 1970. Chemistry of Sediments
in Water. In: T.L. Willrich and G.E. Smith, eds. Agricultural Practices
and Water Quality. Iowa State University Press, Ames, Iowa.
Hamon, W.R. 1961. Estimating Potential Evapotranspiration. Journal
Hydrology Division, American Society of Civil Engineers 87(HY3):107-120.
Hance, R.J. 1967. Decomposition of Herbicides in the Soil by Non-Biological
Chemical Processes. Journal of the Science of Food and Agriculture 18:
544-547.
Hanway, J.J. and J.M. Laflen. 1974. Nutrient Losses from Tile Outlet
Terraces. Journal of Environmental Quality 3:351-356.
Haque, R. and V.H. Freed. 1974. Behavior of Pesticides in the Environment:
Environmental Chemodynamics. Residue Reviews 52:89-116.
Haque, R., P.C. Kearney and V.H. Freed. 1977. Dynamics of Pesticides
in Aquatic Environments. In; Mohammed Khan, ed. Pesticides in Aquatic
Environments. Plenum Press, New York. pp. 39-52.
Harms, L.L., J.N. Dornbush and J.R. Anderson. 1974. Physical and Chemical
Quality of Agricultural Land Runoff. Journal of the Water Pollution
Control Federation 46:2460-2470.
Harns, C.R. and E.P. Lichtenstein. 1961. Factors Effecting the Volatili-
zation of Insecticidal Residues from Soil. Journal of Economic Entomology
54:1038-1045.
Harris, C.I., D.D. Kaufman, T.J. Sheets, R.G. Nash and P.C. Kearney. 1968.
Behavior and Fate of s-Triazines in Soils. In: R.L. Metcalf, ed. Advances
in Pest Control Research, Volume 8. Interscience, New York. pp. 1-56.
300
-------
Harris, C.I., E.A. Woolson and B.F. Hummer. 1969. Dissipation of Herbicides
at Three Soil Depths. Weed Science 17:27-30.
Harris, C.R. and J.R.W. Miles. 1975. Pesticide Residues in the Great
Lakes Region of Canada. Residue Reviews 57:27-80.
Harris, F.A. 1972. Resistance to Methyl Parathion and Toxaphene - DDT
in Bollworm and Tobacco Budworm from Cotton in Mississippi. Journal of
Economic Entomology 65(4):1193-1194.
Harrold, L.L. 1960. The Watershed Hydrology of Plow Plant Corn. Journal
of Soil and Water Conservation 15(4):183-184.
Harrold, L.L. and W.M. Edwards. 1970. Watershed Studies of Agricultural
Pollution. Ohio Report 55(4):85-86.
Harrold, L.L. and W.M. Edwards. 1972. A Severe Rainstorm Test of No-till
Corn. Journal of Soil and Water Conservation 27:1-4.
Harrold, L.L., G.B. Triplett, Jr. and W.W. Edwards. 1970. No-tillage corn:
Characteristics of the System. Agricultural Engineering 51:128-131.
Hartman, J.P., M.P. Wanielesta and G.T. Baragona. 1977. Prediction of
Soil Loss in Nonpoint-Source Pollution Studies. In: Soil Erosion.
Prediction and Control. Soil Conservation Society of America, Ankeny,
IA. pp. 298-302.
Harvey, R.G., A.E. Peterson, R.L. Higgens and H.W. Paulson. 1976. Influence
of Tillage and Planting Practice on Erosion and Atrazine Runoff. Abstracts
No. 10. Weed Science Society of America.
Helling, C.S. 1971. Pesticide Mobility in Soils III. Influence of Soil
Properties. Soil Science Society of America Proceedings 35:743-748.
Helling, C.E., P.C. Kearney and M. Alexander. 1971. Behavior of Pesticides
in Soils. Advances in Agronomy 23:147-240.
Hendrickson, B.H. and A.P. Barnett. 1963. Runoff and Erosion Control Studies
on Cecil Soil in the Southern Piedmont. U.S. Department of Agriculture,
Agricultural Research Service Technical Bulletin No. 1281. Washington, D.C.
Herschfield, D.M. 1961. Rainfall Frequency Atlas of the United States
for Durations from 30 Minutes to 24 Hours and Return Periods from 1 to
100 Years. U.S. Weather Bureau Technical Report 40. U.S. Government Printing
Office. Washington, D.C.
Hibbs, Robert H. 1976. Decline of Hackberry Attributed to Ambient Herbicide
Drift. Proceeding of the Iowa Academy of Science 82(3-4):187-190.
Hillel, D. and W.R. Gardner. 1969. Steady Infiltration Into Crust Topped
Profiles. Soil Science 108:137-142.
301
-------
Hiltbold, A.E. 1974. Persistence of Pesticides in Soil. In: W.D. Guenzi,
ed. Pesticides in Soil and Water. Soil Science of America, Inc., Madison,
Wisconsin, pp. 203-222.
Hiltibran, R.C., D.L. Underwood and J.S. Fickle. 1972. Fate of Diquat
in the Aquatic Environment. Research Reports University of Illinois,
Volume 52, Water Resource Center. Urbana-Champaign, Illinois.
Hilton, J.L., R.W. Bovey, H.M. Hull, W.R. Mullison and R.E. Talbert. 1974.
Herbicide Handbook of the Weed Science Society of America, Champaign,
Illinois.
Himel, C.M. 1974. Analytical Methodology in ULV. In: Pesticide
Application by ULV Methods. British Crop Protection Council Monograph
No. 11.
HjulstrOm, F. 1935. Studies of the Morphological Activity of Rivers as
Illustrated by the River Fyris. Bulletin XXV, Geological Institute of
University of Upsala, Sweden.
Holden, A.V. 1973. Effects of Pesticides on Fish. In: C.A. Edwards, ed.
Environmental Pollution by Pesticides. Plenum Press, London, pp. 213-253.
Holmstead, R.L., S. Khalifa and J.E. Casida. 1974. Toxaphene Composition
Analyzed by Combined Gas Chromatography-Chemical lonization Mass Spectro-
metry. Journal of Agricultural and Food Chemistry 22(6):939-944.
Holt, R.F., W.B. Voorhuis and R.R. Allmaras. 1968. Nutrient Relationships
in Fertilizer Placement as Affected by Tillage. Tillage for Greater Crop
Production. ASAE Publication PROC-168:26-29. American Society of
Agricultural Engineers, St. Joseph, Michigan.
Holt, R.F., R.H. Dowdy and D.R. Timmons. 1970. Chemistry of Sediment in
Water. In: T.L. Willrich and G.E. Smith, eds. Agricultural Practices
and Water Quality. The Iowa State University Press, Ames, Iowa. pp. 21-34.
Holtan, H.N., G.J. Stiltner, W.H. Henson and N.C. Lopez. 1975. USDAHL-74.
Revised Model of Watershed Hydrology. USDA-ARS Technical Bulletin No. 1518.
Agricultural Research Service, U.S. Department of Agriculture. Washington,
D.C.
Horowitz, M., N. Hilin and T. Blumenfeld. 1974. Behavior and Persistence
of Trifluralin in Soil. Weed Research 14:213-220.
Hubbell, D.H., D.F. Rothwell, W.B. Wheeler, W.B. Tappan and P.M. Rhoads.
1973. Microbial Effects and Persistence of Some Pesticide Combinations
in Soil. Journal of Environmental Quality 2(l):96-99.
Hughes, R.A. and G.F. Lee. 1968. Report to the Wisconson Conservation
Division, Department of Natural Resources. Mineo, Wisconsin.
302
-------
Hurlbert, S. 1975. Secondary Effects of Pesticides on Aquatic Ecosystems.
Residue Reviews 57:81-148.
Idike, P., C.L. Larson, D.C. Slack and R.A. Young. 1977. Experimental
Evaluation of Two Infiltration Equations. ASAE Paper No. 77-2558.
American Society of Agricultural Engineers. St. Joseph, Michigan.
Jackson, W.A., L.E. Asmussen, E.W. Hauser and A.W. Shite. 1973. Nitrate
in Surface and Subsurface Flow From a Small Agricultural Watershed. Journal
of Environmental Quality 2(1):480-482.
Jacobs, J.J. 1972. Economics of Water Quality Management Exemplified by
Specific Pollutants in Agricultural Runoff. Ph.D. Thesis. Iowa State
University. Ames, Iowa.
James, L.G., M.F. Walter, and R.E. Muck. 1977. Evaluation of Several
Levels of Hydrological Models on Small Watersheds. ASAE Paper No. 77-2050.
American Society of Agricultural Engineers. St. Joseph, Michigan.
Jamison, V.C., D.D. Smith and J.F. Thornton. 1968. Soil and Water Research
on a Claypan Soil. USDA Technical Bulletin 1379. U.S. Department of
Agriculture. Washington, D.C.
Jetter, B.E., J.C. Smith and E.L. Whiteley. 1962. Influence of Cropping
Systems on Cotton and Corn Yields on the Gulf Coast Prairie. Publication
B-993. The Agricultural and Mechanical College of Texas, Texas Agricultural
Experiment Station. College Station, Texas.
Johnson, D.W. 1968. Pesticides and Fishes - A Review of Selected Literature.
Transactions of the American Fisheries Society 97:398.
Johnson, H.P., J.L. Baker, W.D. Shrader and J.M. Laflen. 1977. Tillage
System Effects on Small Watershed Runoff: Nutrients and Pesticides.
ASAE Paper No. 77-2504. American Society of Agricultural Engineers.
St. Joseph, Michigan.
Jordan, L.S., W.J. Farmer, J.R. Goodin, and B.E. Day. 1970. Nonbiological
Detoxification of the s-Triazine Herbicides. Residue Reviews 32:267-286.
Juo, A.S.R. and 0.0. Oginni. 1978. Adsorption and Desorption of Paraquat
in Acid Tropical Soils. Journal of Environmental Quality 7(1):9-12.
Kaiser, P., J. Pochon and R. Cassini. 1970. Effect of Triazine Herbicides
on Soil Microorganisms. Residue Reviews 32:211-234.
Katan, J., T.W. Fuhremann, and E.P. Lichtenstein. 1976. Binding of [ C]
Parathion in Soil: A Reassessment of Pesticide Persistence. Science
193:891-894.
Kearney, P.C. 1972. Movement of Herbicides Off, Into and Through Soils.
Pesticide Study Series No- 1.0071. U.S. Environmental Protection Agency,
Office of Water Program, Division of Applied Technology, Rural Waste
Branch. Washington, D.C.
303
-------
Kearney, P.C., T.J. Sheets, and J.W. Smith. 1964. Volatility of Seven
s-Triazines. Weeds 12:83-86.
Keiser, R.K., A.A.A. Julia and M.S. Rafael. 1973. Pesticide Levels
in Estuarine and Marine Fish and Invertebrates from the Guatemalan Pacific
Coast. Bulletin of Marine Science 23:905.
Keith, J.O. 1966. Insecticide Contaminations in Wetland Habitats and
Their Effects on Fish Eating Birds. Journal of Applied Ecology 3:71.
Khan, S. 1978. Kinetics of Hydrolysis of Atrazine in Aqueous Fulvic
Acid Solution. Pesticide Science 9:39-43.
Kilmer, V.J. 1974. Nutrient Losses From Grasslands Through Leaching
and Runoff. In: D.A. Mays, ed. Forage Fertilization. American Society
of Agronomy. Madison, Wisconsin, pp. 341-362.
King, P.H. and P.L. McCarty. 1968. A Chromatographic Model for Predicting
Pesticide Migration in Soils. Soil Science 106:248-261.
Klausner, S.D., P.J. Zwerman and D.R. Coote. 1976. Design Parameters
for the Land Application of Dairy Manure. EPA-600/2076-187. U.S.
Environmental Protection Agency. Athens, Georgia.
Klein, A.K. and J.D. Link. 1967. Field Weathering of Toxaphene and
Chlordane. Journal of the Association of Official Analytical Chemists
50:586-591.
Knight, B.A.G. and T.E. Tomlinson. 1967. The Interaction of Paraquat
With Mineral Soils. Journal of Soil Science 18(2):233-243.
Knoblauch, W.A., R.A. Milligan and M.L. Woodell. 1978. An Economic Analysis
of New York Dairy Farm Enterprises. A.E. Res. 78-1. Cornell University.
Ithaca, New York.
Lacewell, R.D., J.M. Sprott, G.A. Miles, J.K. Walker and J.R. Gannaway.
1976. Cotton Grown With an Integrated Production System. Transactions
of the American Society of Agricultural Engineers 19(5):815-818.
Laflen, J.M., H.P. Johnson and R.CiReeve. 1972. Soil Loss From Tile Outlet
Terraces. Journal of Soil and Water Conservation 27:74-77.
Laflen, J.M., J.L. Baker, R.O. Harting, W.F. Buchele and H.P. Johnson. 1978.
Soil and Water Loss From Conservation Tillage Systems. Transactions of
the American Society of Agricultural Engineers 21:881-886.
LaFleur, K.S. 1974. Toxaphene-Soil-Solvent Interactions. Soil Science
117:205-210.
LaFleur, K.S., G.A. Wojeck and W.R. McCaskill. 1973. Movement of Toxa-
phene and Fluometuron Through Dunbar Soil to Underlying Ground Water.
Journal of Environmental Quality 2(4):515-518.
304
-------
Lake, J. and J. Morrison. 1977. Environmental Impact of Land Use on
Water Quality. EPA/9-0-77-007-B. U.S. Environmental Protection Agency
Chicago, Illinois.
Lauer, D.A., D.R. Bouldin and S.D. Klausner. 1976. Ammonia Volatilization
From Dairy Manure Spread on the Soil Surface. Journal Environmental
Quality 5(2):134-141.
Laws, J.O. 1941. Measurements of the Fall Velocity of Water Drops and
Raindrops. Transactions of American Geophysical Union 22:709-721.
Laws, J.O. and D.A. Parsons. 1943. The Relation of Raindrop Size to
Intensity. Transactions of American Geophysical Union 24:452-459.
LeBaron, H.M. 1970. Ways and Means to Influence the Activity and the
Persistence of Triazine Herbicides in Soils. Residue Reviews 32:311-354.
Lee, G.F., R.A. Hughes and G.D. Veith. 1977. Evidence for Partial Degrada-
tion of Toxaphene in the Aquatic Environment. Water, Air and Soil
Pollution 8:479-484.
Lee, G.F., W. Post and R.A. Jones. 1978. Eutrophication of Water Bodies:
Insights for an Age-old Problem. Environmental Science and Technology
12:900-908.
Leenheer, J.A. and J.L. Alrichs. 1971. A Kinetic and Equilibrium Study of
the Adsorption of Carbaryl and Parathion Upon Soil Organic Matter Surface.
Soil Science Society Proceedings 35:700-705.
Leh, H.O. 1968. Vertical Movement of Herbicides in Soil With Special Con-
sideration of Contamination of Underground Water. Nachrichtenblatt des
Deutschen Pflanzenschutzdienstes 20:99-106. (Chemical Abstracts 70:5655W, 1969)
Leonard, R.A., .G.W. Bailey and R.R. Swank, Jr. 1976. Transport, Detoxifi-
cation, Fate and Effect of Pesticides in Soil and Water Environments. Land
Application of Waste Materials. Soil Conservation Society of America.
Ankeny, Iowa. pp. 48-79.
Li, E.A., V.O. Shanholtz, D.N. Contractor and J.C. Carr. 1976. Generating
Precipitation Excess Based on Readily Determinable Soil Vegetative Character-
istics. ASAE Paper No. 76-2005. American Society of Agricultural Engineers.
St. Joseph, Michigan.
Li, M. and R.A. Fleck. 1972. The Effect of Agricultural Pesticides in the
Aquatic Environment, Irrigated Croplands, San Joaquin Valley. Pesticide
Study Series 6. Environmental Protection Agency, Office of Water Programs,
Applied Technology Division, Rural Waste Branch. Washington, D.C. TS-00-72-
05.
Libik, A.W. and R.R. Romanowski. 1976. Soil Persistence of Atrazine and
Cyanazine. Weed Science 24(6):627-629.
305
-------
Liechtenstein, E.P., J. Katan and B.N. Anderegg. 1977. Binding of "Per-
sistent" and Nonpersistent" i4C-Labeled Insecticides in an Agricultural Soil.
Journal of Agricultural and Food Chemistry 25(1):43-47.
Lichtenstein, E.P. and K.R. Schulz. 1964. The Effect of Moisture and Micro-
organisms on the Persistence and Metabolism of Some Organophosphorus Insecti-
cides in Soils With Special Emphasis on Parathion. Journal of Economic
Entomology 57:618-627.
Lichtenstein, E.P. and K.R. Schulz. 1970. Volatilization of Insecticides
From Various Substrates. Journal of Agricultural and Food Chemistry 18:
814-818.
Lichtenstein, E.P., K.R. Schulz and T.W. Fuhremann. 1971. Effects of a Cover
Crop Versus Soil Cultivation on the Fate and Vertical Distribution of In-
secticide Residues in Soil 7 to 11 Years After Soil Treatment. Pesticides
Monitoring Journal 5(2):218-222.
Lichtenstein, E.P., K.R. Schulz, T.W. Fuhremann and T.T. Liang. 1969.
Biological Interaction Between Platicizers and Insecticides. Journal of
Economic Entomology 62(4):761-765.
Lichtenstein, E.P., T.T. Liang, B.N. Anderegg. 1973. Synergism of Insecti-
cides by Herbicides. Science 181:847-849.
Livingston, Robert J. 1977. Review of Current Literature Concerning the
Acute and Chronic Effects of Pesticides on Aquatic Organisms. Chemical
Rubber Company, Critical Reviews in Environmental Control 7:325-351.
Lomenick, T.F. and T. Tamura. 1965. Naturally Occuring Fixation of Cesium-
137 on Sediments of Locus Trine Origin. Soil Science Society of America
Proceedings 29:383-387.
Lowe, J.I., P.O. Wilson, A.J. Rick, A.J. Wilson and J. Alfred, Jr. 1970.
Chronic Exposure of Oysters to DDT, Toxaphene, and Parathion. Proceedings
of the National Shellfish Association 61:71-79.
Luckmann, W. 1977. Insect Control in Corn-Practices and Prospects.
Pest Control Strategies: Understanding and Action. Conference held at
Cornell University, Ithaca, New York. (In Press).
Luckmann, W.H., R. Barganz, A. Brigham, P. Challand, M. Conlin. H. Dodd,
C. Erb, W. Hadley, M. Shurtleff, P. Hermsen, J. Hogancamp, G. Kapusta,
E. King, J. Kirk, M. Levin, G. Meadows, J.C. Cole and H. Seymour. 1978.
Final Report of the Subcommittee on Pesticides of the State of Illinois
Task Force on Agriculture Nonpoint Sources of Pollution. Urbana, Illinois.
MacKay, D. and A.W. Wolkoff. 1973. Rate of Evaporation of Low Solubility
Contaminants From Water Bodies to the Atmosphere. Environmental Science and
Technology 7(7):611-614.
306
-------
MacMahon, M.A. and G.W. Thomas. 1976. Anion Leaching in Two Kentucky Soils
Under Conventional Tillage and a Killed-Sod Mulch. Agronomy Journal 68(3):
437-442.
MacVicar, T.K. 1978. Solid State Transducer for Recording Piezometer
Systems. M.S. Thesis, Department of Agricultural Engineering, Cornell
University. Ithaca, New York.
Martin, J.P., R.B. Harding, G.H. Cannell and L.D. Anderson. 1959. Influence
of Five Annual Field Applications of Organic Insecticides on Soil Biological
and Physical Properties. Soil Science 87:334-338.
Martin, W.P., W.E. Fenster and L.D. Hanson. 1970. Fertilizer Management
for Pollution Control. In: T.L. Willrich and G.E. Smith, eds. Agricultural
Practices and Water Quality. The Iowa State University Press. Ames, Iowa.
pp. 142-148.
Massey, H.F,' and M.L. Jackson. 1952. Selective Erosion of Soil Fertility
Constituents. Soil Science Society of America Proceedings 16:353-356.
Massey, H.F..M.L. Jackson and O.E. Hays. 1953. Fertility Erosion on Two
Wisconsin Soils. Agronomy Journal 45:543-547.
Mathur, S.P., H.A. Hamilton, R. Greenhalgh, K.A. MacMillan and S.U. Khan. 1976.
Effect on Microorganisms and Persistence of Field Applied Carbofuran and
Dyfonate in a Humic Mesisol. Canadian Journal of Soil Science 56:89-96.
Mauck, W.L., F.L. Mayer, Jr., and D.D. Holz. 1976. Simazine Residue Dynamics
in Small Ponds. Bulletin of Environmental Contamination and Toxicology
16:1-8.
McDowell, L.L. and E.H. Grissinger. 1967. Pollutant Sources of Routing in
Watershed Programs. Proceedings of the 21st Annual Meeting Soil Conservation
Society of America. Ankeny, Iowa. pp. 147-161.
McDowell, L.L. and E.H. Grissinger. 1976. Erosion and Water Quality.
Proceedings of the 23rd National Watershed Congress. Biloxi, Mississippi.
pp. 40-56.
McElroy, A.D., S.Y. Chen, J.W. Nebgen, A. Aleti and F.W. Bennett. 1976.
Loading Functions for Assessment of Water Pollution From Nonpoint Sources.
EPA 600/2-76-151. U.S. Environmental Protection Agency. Kansas City, Missouri.
McGrann, J. and J. Meyer. 1978. Farm Level Economic Evaluation of Erosion
Control and Reduced Chemical Use in Iowa. J[n: R.C. Loehr, D.A. Haith,
M.F. Walter and C. Martin, eds. Best Management Practices for Agriculture
and Silviculture. Ann Arbor Science. Ann Arbor, Michigan, pp.
McHargue, J.J. and A.M. Peter. 1921. The Removal of Mineral Plant Food by
Natural Drainage Waters. Kentucky Agricultural Experiment Station, Bulletin
No. 237.
307
-------
McWhorter, G.G. 1977. Weed Control in Soybeans With Glyphosate Applied in
the Recirculating Sprayer. Weed Science 25(2):135-141.
Mehrle, P.M. and F.L. Mayer. 1977. Bone Development and Growth of Fish as
Affected by Toxaphene. In: I.H. Suffet, ed. Fate of Pollutants in the Air
and Water Environments, Volume 2. John Wiley and Sons, New York. pp. 301-316.
Mein, R.G. and C.L. Larson. 1971. Modeling the Infiltration Component
of Rainfall-Runoff Process. Water Resources Research Center, Bulletin 43.
University of Minnesota Graduate School. St. Paul, Minnesota.
Menzie, C.M. 1972. Fate of Pesticides in the Environment. Annual Review
of Entomology 17:199-222.
Meta Systems Inc. 1978. Water Quality Impact and Socio-Economic Aspects
of Reducing Nonpoint Source Pollution From Agriculture. Preliminary Draft.
U.S. Environmental Protection Agency. Athens, Georgia.
Metcalf, R.L. 1975. Insecticides in Pest Management. In: R.L. Metcalf
and W. Luckmann, eds. Introduction to Insect Pest Management. John Wiley and
Sons, New York. pp. 235-273.
Metcalf, R.L. and J. McKelvey. 1976. Discussion. In: R.L. Metcalf and
J.J. McKelvey, eds. The Future for Insecticides: Needs and Prospects.
Advances in Environmental Science and Technology. John Wiley and Sons,
New York 6:309-312.
Metcalf, R.L. and J.R. Sanborn. 1975. Pesticides and Environmental Quality
in Illinois. Illinois Natural History Survey Bulletin 31(9).
Metcalf, R.L., G.K. Sangha and I.P. Kapoor. 1971. Model Ecosystem for the
Evaluation of Pesticide Biodegradability and Ecological Magnification.
Environmental Science and Technology 5:709-713.
Meyer, L.D. and W.H. Wischmeier. 1969. Mathematical Simulation of the Process
of Soil Erosion by Water. Transactions of the American Society of Agri-
cultural Engineers 12(6):754-758, 762.
Meyer, L.D., G.R. Foster and M.J.M. ROmkens. 1975. Soil Eroded by Water
From Upland Slopes. Present and Prospective Technology for Predicting
Sediment Yields and Sources. ARS-S-40. U.S. Department of Agriculture,
Agricultural Research Service. Washington, D.C. pp. 177-189.
Meyer, L.D. and J.V. Mannering. 1968. Tillage and Land Modification for
Water Erosion Control. ASAE Publication PROC. 168:58-62. American Society
of Agricultural Engineers. St. Joseph, Michigan.
Meyer, L.D. and W.C. Harmon. 1977. Rainfall Simulator for Evaluating Erosion
Ratio and Sediment Sizes From Row Side Slopes. ASAE Paper No. 77-2025.
American Society of Agricultural Engineers. St. Joseph, Michigan.
308
-------
Meyers, N.L., J.L. Ahlrichs and J.L. White. 1969. Adsorption of Insecticides
on Pond Sediments and Watershed Soils. Indiana Academy of Science Proceedings
79:432-437. *
f
Miles, J.R.W. 1976. Insecticide Residues on Stream Sediments in Ontario,
Canada. Pesticides Monitoring Journal 10(3):87-91.
Miller, C.W. .^.M. Zucherman and A.J. Change. 1966. Water Translocation
of Diazinon-C and Parathion-C14 off a Model Cranberry Bog and Subsequent
Occurrence in Fish and Mussels. Transactions of the American Fisheries
Society 95:345-349.
Miller, M.F. and H.H. Krusekopf. 1932. 'The Influence of Systems of Cropping
and Methods of Culture on Surface Runoff and Soil Erosion. Missouri
Agricultural Experiment Station Research Bulletin No. 178. University of
Missouri. Columbia, Missouri.
Miller, M.W. and G.G. Berg. 1969. Chemical Fallout. Current Research on
Persistent Pesticides. Charles C. Thomas, Springfield, Illinois.
Misteric, W.J. and J.C. Gaines. 1953. Effect of Wind and Other Factors
on the Toxicity of Certain Insecticides. Journal of Economic Entomology
46:341-349.
Mitchell, J.K. and C.E. Beer. 1965. Effect of Land Slope and Terrace
Systems on Machine Efficiencies. Transactions of the American Society of
Agricultural Engineering 8:235-237.
Mitchell, W.H. and M.R. Teel. 1977. Winter-Annual Cover Crops for No-
Tillage Corn Production. Agronomy Journal 69:569-573.
Mockus, V. 1972. Estimation of Direct Runoff From Storm Rainfall. National
Engineering Handbook. Sec. 4, Hydrology. U.S. Soil Conservation Service.
Washington, D.C.
Moldenhauer, W.C., W.G. Lovely, N.P. Swanson and H.D. Currence. 1971.
Effect of Row Grades and Tillage Systems on Soil and Water Losses. Journal
of Soil and Water Conservation 26(5):193-195.
Moore, J.A., C.A. Onstad, M.A. Otterby, H.L. Person and D.B. Thompson. 1977.
Preliminary Identification of Literature Models and Data for Evaluating
Rural Nonpoint Nutrient, Sediment and Pathogen Sources. Agricultural Engin-
eering Department, University of Minnesota. U.S. Department of Agriculture.
North Central Soil Conservation Research Center. Morris, Minnesota.
Moore, S., H.B. Petty, W.N. Bruce, R. Randell and D.E. Kuhlman. 1977.
Dieldrin Residues in Soybeans in Illinois, 1965, 1966, 1967, 1971, and 1974.
Pesticides Monitoring Journal 11(2):94-98.
Morley, H.V. 1977. Dynamics of Pesticides in Aquatic Environments. In_:
Mohammed Khan, ed. Pesticides in Aquatic Environments. Plenum Press,
New York. pp. 53-76.
309
-------
Muirhead-Thomson, R.C. 1971. Pesticides and Fresh Water Fauna. Academic
Press, London.
Mulkey, L.A. and J.W. Falco. 1977. Sedimentation and Erosion Control Impli-
cation for Water Quality Management. Proceedings of the National Symposium
on Soil Erosion and Sedimentation of Water, December 1977. ASAE Paper No.
4-77. American Society of Agricultural Engineers. St. Joseph, Michigan.
Munson, T.O. 1976. A Note on Toxaphene in Environmental Samples From the
Chesapeake Bay Region. Bulletin of Environmental Contamination and Toxicology
16:491.
MusgYave, G.W. and H.N. Holtan. 1964. Infiltration. In: V.T. Chow, ed.
Handbook of Applied Hydrology. Chapter 12. McGraw-Hill, New York.
Musiek, G.J. and H.B. Petty. 1973. Insect Control in Conservation Tillage
Systems. Conservation Tillage. Proceedings of a National Conference.
Soil Conservation Society of America. Ankeny, Iowa.
Nash, R.G. 1974. Plant Uptake of Insecticides, Fungicides, and Fumigants
from Soils. In: W.D. Guenzi, ed. Pesticides in Soil and Water. Soil
Science Society of America, Inc., Madison, Wisconsin, pp. 257-314.
Nash, R.G., M.L. Beall, Jr. and W.G. Harris. 1977. Toxaphene and 1,1,1,
-trichloro-2,2-bis(p-chlorophenyl)ethane (DDT) Losses From Cotton in an
Agroecosystem Chamber. Journal Agricultural and Food Chemistry 25(2):336-341.
Nash, R.G. and W.G. Harris. 1973. Chlorinated Hydrocarbon Insecticide
Residues in Crops and Soils. Journal of Environmental Quality 2:269-273.
r
Naumann, K. 1970. Zur Dynamik der Bodenmikroflora nach Anwendung von
Pflanzenschutzmitteln, II. Die Reaktion verschiedener physiologisher Gruppen
von Bodenbakterien auf den Einsatz von Parathion-methyl im Freiland. Zentral-
blatt fOr Bakteriologie, Parasitenkunde, Infektions-Krankheiten und Hygiene
124:755-765.
Nayshteyn, S.Y., V.A. Zhulinskaya and Y.M. Yuroskaya. 1973. The Stability
of Certain Phosphororgenic Pesticides in the Soil. Gigiena i Sanitariia
38(7): 42-45.
Negev, M.A. 1967. Sediment Model on a Digital Computer. Department of
Civil Engineering, Stanford University, Technical Report No. 76. Stanford,
California.
Nelson, G. 1968. United States Resources - Our Air, Land and Water. Food
for Billions. Special Publication No. 11. American Society of Agronomy.
Madison, Wisconsin, ppp. 27-30.
Newman, A.S. and J.R. Thomas. 1949. Decompostion of 2,4-dichlorophenosy
Acetic Acid in Soil and Liquid Media. Soil Science Society Proceedings
14:160-164.
310
-------
Nicholson, H.P., A.R. Grzenda and J.I. Teasley. 1966. Water Pollution
by Insecticides - A Six and One-Half Years Study of a Watershed. Proceedings
of the Symposium on Agricultural Waste Waters. Report No. 10. Water
Resources Center, University of California. Davis, California.
Nicholson, H.P., A.R. Grzenda, G.J. Lauer, W.S. Cox and J.I. Teasley. 1964.
Water Pollution by Insecticides in an Agricultural River Basin I. Occurrence
of Insecticides in River and Treated Municipal Water. Limnology and
Oceanography 9(3):310-317.
Ogrosky, H.O. and V. Mockus. 1964. Hydrology of Agricultural Lands, In:
V.T. Chow, ed. Handbook of Applied Hydrology, Chapter 21. McGraw Hill,
New York.
Onstad, C.A. and G.R. Foster. 1975. Erosion Modelling on a Watershed.
Transactions of the American Society of Agricultural Engineers 18(2):288-292.
Onstad, C.A. and T.C. Olson. 1970. Water Budget Accounting on Two Corn-
Cropped Watersheds. Journal of Soil and Water Conservation 25:150-152.
Paris, D.F. and D.L. Lewis. 1973. Chemical and Microbial Degradation of
Ten Selected Pesticides in Aquatic Systems. Residue Reviews 45:95-123.
Parker, C. 1978. Crop Budgets for Texas Blackland Region. Department of
Agricultural Economics, Texas A§M University. College Station, Texas:
Pearce, P.A., L.M. Reynolds and D.B. Peakall. 1978. DDT Residues in Rain-
water in New Brunswick and Estimate of Aerial Transport of DDT into the
Gulf of St. Lawrence, 1967-68. Pesticides Monitoring Journal 11(4):119-204.
Pesticide Manual. 1972. In: Hubert Martin, ed. British Crop Protection
Council, London.
Peters, R.A. 1970. Status of No-Tillage Corn in New England. Northeastern
No-Tillage Conference. Albany, New York.
Phillips, S.H., and H.M. Young, Jr. 1973. No-Tillage Farming. Reiman
Associates. Milwaukee, Wisconsin.
Phillips, W.M. and K.C. Feltner. 1972. Persistence and Movement of Picloram
in Two Kansas Soils. Weed Science 20:110-116.
Pimentel, D. 1971. Ecological Effects of Pesticides on Non-Target Species.
Executive Office of the President, Office of Science and Technology, U.S.
Government Printing Office. Washington, D.C.
Pimentel, D., C. Shoemaker, E.L. LaDue, R.B. Rovinsky and N.P. Russell.
Alternatives for Reducing Insecticides on Cotton and Corn: Economic and
Environment Impact. U.S. Environmental Protection Agency. Washington, D.C.
(In Press).
311
-------
Pionke, H.B. and G. Chesters. 1973. Pesticide-Sediment-Water Interactions.
Journal of Environmental Quality 2(l):29-45.
Plewa, M.J. and J.M. Gentile. 1976. Mutagenicity of Atrazine: A Maize-
Microbe Bioassay. Mutation Research 38:287-292.
Plimmer, J.R. 1976. Volatility. In: P.G. Kearney and D.D. Kaufman, eds.
Herbicides: Chemistry, Degradation and Mode of Action. Marcel Dekker, Inc.,
New York. pp. 891-934.
Pollack, G.A. and W.W. Kilgore. 1978. Toxaphene. Residue Reviews 69:87-140.
Quinby, G.E., K.C. Walker and W.F. Durham. 1958. Public Health Hazards
Involved in the Use of Organic Phosphorus Insecticides in Cotton Culture
in the Delta Area of Mississippi. Journal of Economic Entomology 51:831-838.
Randolph, N.M., R.D. Chisholm, L. Koblitsky and J.C. Gaines. 1960. Insecti-
cide Residues in Certain Texas Soils. Texas Agricultural Experiment Station
MP-477 (Chemistry Abstracts 55:486a, 1961).
Rao, P.S.C., J.M. Davidson and L.C. Hammond. 1976. Estimation of Non-Reactive
and Reactive Solute Front Locations in Soils. In: W.H. Fuller, ed. Re-
sidual Management by Land Disposal. Proceedings of the Hazardous Waste Re-
search Symposium. EPA 600/9-76-015. Solid and Hazardous Waste Research
Division, U.S. Environmental Protection Agency, Washington, D.C. pp. 235-242.
Read, D.C. 1969. Persistence of Some Newer Insecticides in Mineral Soils
Measured by Bioassay. Journal of Economic Entomology 62:1336-1342.
\
Reimold, R.J. 1974. Toxaphene Interactions in Estuarine Ecosystems. Georgia
Sea Grant Report 74.
Reimold, R.J., R.C. Adams and C.J. Durant. 1973. Effects of Toxaphene Con-
tamination on Estaurine Ecology. Marine Institute Technical Report No. 73.
University of Georgia. Athens, Georgia.
Reimold, R.J. and C.J. Durant. 1974. Toxaphene Content of Estuarine Fauna
and Flora Before, During and After Dredging Toxaphene-Contaminated Sediments.
Pesticides Monitoring Journal 8:44-49.
Renfro, G.W. 1975. Use of the Erosion Equations and Sediment Delivery Ratios
for Predicting Sediment Yield. Present and Prospective Technology for Pre-
dicting Sediment Yields and Sources. ARS-S-40. U.S. Department of Agriculture,
Agricultural Research Service, Washington, D.C.
Rich, C.I. 1968. Applications of Soil Mineralogy in Soil Chemistry and
Fertility Investigations in Mineralogy in Soil Science and Engineering.
Soil Science Society of America Special Publication 3. pp. 61-90.
312
-------
Richard, J., G. Junk, M. Avery, N. Nehring, J. Fritz and H. Svec. 1975.
Analysis of Various Iowa Waters for Selected Pesticides: Atrazine, DDT
and Dieldrin - 1974. Pesticides Monitroing Journal 9(3):117-123.
Richardson, C.W. 1973. Runoff, Erosion, and Tillage Efficiency on Graded-
Furrow and Terraced Watersheds. Journal Soil and Water Conservation 28:
162-164.
Richardson, R.G. 1974. Control of Spray Drift with Thickening Agents.
Journal of Agricultural Engineering Research 19:227-231.
Riley, D., W. Wilkison and B.U. Tucker. 1976. Biological Unavailability
of Bound Paraquat Residues in Soil, jn; D. Kaufman, ed. Bound and
Congugated Pesticide Residues. American Chemical Society. Washington, D.C.
pp. 301-353.
Ritter, W.F. 1971. Environmental Factors Affecting the Movement of
Atrazine, Propachlor and Diazinon in Ida Silt Loam. Unpublished Ph.D. Thesis.
Department of Agricultural Engineering, Iowa State University. Ames, Iowa.
Ritter, W.F., H.P. Johnson, W.G. Lovely and M. Monan. 1974. Atrazine,
Propachlor and Diazinon Residues on Small Agricultural Watersheds-Runoff
Losses, Persistence and Movement. Environmental Science and Technology 8:38-42.
Rock, G.C. and D.R. Yeargan. 1973. Toxicity of Apple Orchard. Herbicides
and Growth - Regulating Chemical to Neoseiulus fallacis and Two-Spotted
Spider Mites. Journal of Economic Entomology 66:1342-1343.
Rogers, H.T. 1941. Plant Nutrient Losses by Erosion From a Corn, Wheat,
Clover Rotation on Dunmore Selt Loam. Soil Science Society of America
Proceedings 6:263-271.
Rudd, R.L. 1966. Pesticides and the Living Landscape. University of
Wisconsin Press. Madison, Wisconsin.
Ryden, J.C., J.K. Syers and R.F. Harris. 1973. Phosphorus in Runoff and
Streams. Advances in Agronomy. Academic Press. New York, New York. pp. 1-45.
Saltzman, S. and B. Yaron. 1971. Parathion Adsorption as Influenced by Soil
Components. Abstracts of the Second International Congress of Pesticide
Chemistry. Israel.
Sanborn, J.R. 1974. The Fate of Select Pesticides in the Aquatic Environ-
ment. Ecological Research Series., EPA-660/3-74-025. U.S. Environmental
Protection Agency. Corvallis, Oregon.
Sanborn, J.R., B. Magnus Francis and R.L. Metcalf. 1977. The Degradation of
Selected Pesticides in Soil: A Review of the Published Literature. EPA-9-77-
022. U.S. Environmental Protection Agency. Washington, D.C.
313
-------
Sanborn, J.R., R.L. Metcalf, W.N. Bruce and P.Y. Lu. 1976. The Fate of
Chlordane and Toxaphene in a Terrestial-Aquatic Model Ecosystem. Environ-
mental Entomology 5(3):533-538.
Sangha, G.K. 1972. Environmental Effects of Carbamate Insecticides as Assayed
in the*Model Ecosystem, a Comparison With DDT. Dissertation Abstracts
International 32(8):4650.
Saxton, K.E. and R.G. Spomer. 1968. Effects of Conservation on the Hydrology
of Loessial Watershed. Transactions of American Society of Agricultural
Engineers 11:848-849.
Schottman, R.W. 1978. Estimation of the Penetration of High-Energy Raindrops
Through a Plant Canopy. PhD Thesis. Department of Agricultural Engineering,
Cornell University. Ithaca, New York.
Schulze, J.A., D.B. Manigold and F.L. Andrews. 1973. Pesticides in Selected
Western Streams 1968-1971. Pesticides Monitoring Journal 7(l):73-85.
Schwab, G.O., E.O. McLean, A.C. Waldron, R.K. White and D.W. Michener. 1973.
Quality of Drainage Water From a Heavy-Textured Soil. Transactions American
Society of Agricultural Engineering 16:1104-1107.
Schwab, G.O., R.K. Frevert, T.W. Edminster and K.K. Barnes. 1966. Soil and
Water Conservation Engineering. John Wiley and Sons, Inc. New York, New
York.
Schwerdle, F. 1969. Untersuchungen zur Populationsdichte von RegenwCirmen
bei HerkOmlicher Bodenbearbeitung und bei Diretsaat. (Research on the Number
of Earthworms With Repeated Tillage and No-Tillage). Zeitschrift fflr
Pflantzenkrankheiten und Planzenshutz 76:635-641.
Scott, H.D. and R.E. Phillips. 1972. Diffusion of Selected Herbicides in
Soils. Soil Science Society Proceedings 36:714-719.
Seim, D. 1978. Better Weed Control at a Better Price. Farm Journal 2:A-4.
Seitz, W.D. and C. Osteen. 1978. Economic Impacts of Policies f.o Control
Erosion and Sedimentation in Illinois and Other Corn Belt States. In:
R.C. Loehr, D.A. Haith, M.F. Walter, C. Martin, eds. Best Management" Practices
for Agriculture and Silviculture. Ann Arbor Science. Ann Arbor, Michigan.
pp. 373-382.
Shanholtz, V.O. and J.H. Lillard. 1968. Hydrologic Aspects of No-Tillage
Versus Conventional Tillage Systems for Corn Production. Water Resources
Research Center. Virginia Polytechnic Institute. Blacksburg, Virginia.
Shanholtz, V.O. and J.H. Lillard. 1969. Tillage System Effects on Water Use
Efficiency. Journal of Soil and Water Conservation 24:186-189.
314
-------
Sharp, B.M.H. and S.J. Berkowitz. 1978. Economic, Institutional and Water
Quality Considerations in the Analysis of Sediment Control Alternatives: A
Case Study. In: R.C. Loehr, D.A. Haith, M.F. Walter, C. Martin, eds. Best
Management Practices for Agriculture and Silviculture. Ann Arbor Science,
Ann Arbor, Michigan, pp. 429-454.
Sheets, T.J., J.R. Bradley, Jr.- and M.D. Jackson. 1972. Contamination of
Surface and Ground Water With Pesticides Applied to Cotton. University of
North Carolina Water Resource Research Institute Report No. 60. Chapel Hill,
North Carolina.
Siemens, J.C .and W.R. Oschwald. 1976. Corn-Soybeans Tillage Systems:
Erosion Control Effects on Crop Production Costs. ASAE Paper No. 76-2552.
American Society of Agricultural Engineers. St. Joseph, Michigan.
Simonson, R.W. 1970. Loss of Nutrient Elements During Soil Formation. In:
Engelstad, P. et_ a^. (ed.). Nutrient mobility in soils: accumulation and"
losses , Soil Science Society of America Special Publication Series No. 4,
pp. 21-45. Madison, Wisconsin.
Slade, P. 1966. The Fate of Paraquat Applied to Plants. Weed Research
6:158-167.
Slater, C.S. and E.A. Carleton. 1942. Variability of Eroded Material.
Journal Agricultural Research 65(4):209-219.
Smid, A.E. and E.G. Beauchamp. 1976. Effects of Temperature and Organic
Matter on Denitrification in Soil. Canadian Journal of Soil Science 56:385-391.
Smith, C.N., R.A. Leonard, G.W. Landale and G.W. Bailey. 1978. Transport
of Agricultural Chemicals From Small Upland Piedmont Watersheds. EPA Pre-
liminary Copy. U.S. Environmental Protection Agency. Athens, Georgia.
Smith, D.B., E.G. Burt and E.P. Lloyd. 1975. Selection of Optimum Spray-
Droplet Sizes for Boll Weevil and Drift Control. Journal of Economic
Entomology 68(3):415-417.
Smith, D.B., C.E. Goering, S.K. Lednc and J.D. McQuigg. 1974. Chemical
Application Decisions Based on Temporal Periods. Transactions of the
American Society of Agricultural Engineers 17(4):620-622.
Smith, E.G., F.D. Whitaker and H.G. Heineman. 1974. Losses of Fertilizers
and Pesticides From Clay Pan Soils. EPA 660/2-74-068. U.S. Environmental
Protection Agency. Athens, Georgia.
Soil Conservation Service. 1972. SCS National Engineering Handbook. Section
4: Hydrology. Parti: Watershed Planning. Soil Conservation Service.
U.S. Department of Agriculture. Washington, D.C.
315
-------
Sorensen, D.L., M.M. McCarthy, E.J. Middlebrooks and D.B. Porcella. 1977.
Suspended and Dissolved Solids Effects on Freshwater Biota: A Review.
EPA-600/3-77-042. U.S. Environmental Protection Agency. Athens, Georgia.
Sparks, R.E. 1977. Effects of Sediment on Aquatic Life. Illinois Natural
History Survey. Havana, Illinois.
Spencer, W.F. and M.M. Cliath. 1973. Pesticide Volatilization as Related
to Water Loss From Soil. Journal of Environmental Quality 2(2):284-289.
Spencer, W.F. and M.M. Cliath. 1977. The Solid-Air Interface: Transfer
of Organic Pollutants Between the Solid-Air Interface. In: I.H. Suffet, ed.
Fate of Pollutants in the Air and Water Environments. John Wiley and Sons,
Inc., New York. pp. 107-126.
Spomer, R.G., H.G. Heineman and R.F. Piest. 1971. Consequences of Historic
Rainfall on Western Iowa Farmland. Water Resource Research 7:524-535.
Spomer, R.G., R.F. Piest and H.G. Heineman. 1976. Soil and Water Conser-
vation With Western Iowa Tillage Systems. Transactions of American Society
of Agricultural Engineers 2:108-112.
Spomer, R.G., W.D. Shrader, P.E. Rosenberry and E.L. Miller. 1973. Level
Terraces With Stabilized Backslopes on Loessial Cropland in the Missouri
Valley: A Cost-Effectiveness Study. Journal Soil and Water Conservation
28:127-131.
Staff. 1978. Give Your Sprayer a Tune-Up. Fruit Grower, pp. 18.
Stallings, J.H. 1945a. Effect of Contour Cultivation on Crop Yield, Runoff
and Erosion Losses. Soil Conservation Service. U.S. Department of Agriculture,
Washington, D.C.
Stallings, J.H. 1945b. Review of Terracing Data on Crop Yield, Runoff, and
Soil Loss. Soil Conservation Service. U.S. Department of Agriculture,
Washington, D.C.
Stallings, J.H. 1945c. Summarization of Strip-Cropping Data on Crop Yield,
Runoff and Soil Loss. Soil Conservation Service, U.S. Department of Agri-
culture, Washington, D.C.
Stanley, C.W., J.E. Barney, M.R. Helton and A.R. Yobs. 1971. Measurement
of Atmospheric Levels of Pesticides. Environmental Science and Technology
5:430-435.
Steenhuis, T.S. 1977. Modeling Nitrogen and Other Nutrient Losses From
Winter Spread Manure. Ph.D. Thesis. Department of Agricultural Engineering,
University of Wisconsin. Madison, Wisconsin.
316
-------
Steenhuis, T.S. and M.F. Walter. 1978. Closed Form Solution for Pesticide
Loss in Runoff Water. ASAE Technical Paper No. 78-2031. American Society
of Agricultural Engineers. St. Joseph, Michigan.
Stern, V.M., R.F. Smith, R, van den Bosch and K.S. Hagen. 1959. The Integra-
ted Control Concept. Hilgardia 29(2):81.
Stevens, W.K. 1977. Sterility Linked to a Pesticide Sharpens Fear on
Chemical Use. The New York Times, 11 September, pp. 1-37.
Stewart, B.A., D.A. Woolhiser, W.H. Wischmeier, J.H. Caro and M.H. Frere.
1975. Control of Water Pollutant From Cropland. Vol. 1, EPA-600/2-75-026a.
U.S. Environmental Protection Agency. Washington, D.C.
Stewart, B.A., D.A. Woolhiser, W.H. Wischmeier, J.H. Caro and M.H. Frere.
1976. Control of Water Pollution From Cropland. Vol. II, EPA-600/2-75-026b.
U.S. Environmental Protection Agency. Washington, D.C.
Stewart, D.K.R., D. Chisolm and M.T.H. Ragab. 1971. Long-Term Persistence
of Parathion in Soil. Nature 229:47.
Stringer, G.E. and R.C. McMynn. 1958. Experiments With Toxaphene as a
Fish Poison. The Canadian Fish Culturist 23:39-48.
Swader, F.N. 1970. No-Plow Corn in New York. Northeastern No-Tillage
Conference, Albany, New York.
Swader, F.N. 1974. The Universal Soil Loss Equation. Agronomy Mimeo
74-19. Cornell University. Ithaca, New York.
Swan, J.B., W.W. Nelson and R.R. Allmaras. 1972. Soil Management by Fall
Tillage for Corn. Extension Folder 264. Agricultural Extension Service,
University of Minnesota. St. Paul, Minnesota.
Swoboda, A., G.W. Thomas, F.B. Cady, R.W. Baird and W.G. Knisel. 1971.
Distribution of DDT and Toxaphene in Houston Black Clay on Three Watersheds.
Environmental Science and Technology 5(2):141-145.
Taschenberg, E.F., G.L. Mack and F.L. Gambell. 1961. DDT and Copper Residues
in a Vineyard Soil. Journal of Agricultural and Food Chemistry 9:207-209.
Taylor, C.R., D.R. Reneau and B.L. Harris. 1978. An Economic Analysis of
Erosion and Sedimentation in Lavon Reservoir Watershed. Bulletin TR-88.
Texas Water Resources Institute. College Station, Texas.
Terriere, L.C., U. Kiigemagi, A.R. Gerlach and R.L. Borovicka. 1966. The
Persistence of Toxaphene in Lake Water and its Uptake by Aquatic Plants and
Animals. Journal Agricultural and Food Chemistry 14:66-69.
Tessari, J.D. and D.L. Spencer. 1971. Air Sampling for Pesticides in the
Human Environment. Association of Official Analytical Chemists 54(6):1376-
1382.
317
-------
Tetioa, S.P., F.L. Duley and T.M. McCalla. 1950. Effect of Stubble Mulching
on Number and Activity of Earthworms. Nebraska Agricultural Experiment
Station Research Bulletin No. 165. Lincoln, Nebraska.
Thomas, G.W. and J.D. Crutchfield. 1974. Nitrate-Nitrogen and Phosphorus
Contents of Streams Draining Small Agricultural Watersheds in Kentucky.
Journal of Environmental Quality 3:46-49.
Thompson, A.R. 1973. Pesticide Residues in Soil Invertebrates. In: C.A.
Edwards, ed. Environmental Pollution by Pesticides. Plenum Press, London.
pp. 87-133.
Timmons, D.R., R.E. Burwell and R.F. Holt. 1968. Loss of Crop Nutrients
Through Runoff. Minnesota Science 24(4):16-18.
Timmons, D.R., R.F. Holt and J.J. Latterell. 1970. Leaching of Crop
Residue as a Source of Nutrients in Surface Runoff. Water Resources Research
6:1367-1375.
Tomlin, A.D. 1975. The Toxicity of Insecticides by Contact and Soil
Treatment to Two Species of Ground Beetles (Coleoptera:Carabidae). The
Canadian Entomologist 107:529:532.
Trichell, D.W., H.L. Martin and M.G. Merkle. 1968. Loss of Herbicides in
Runoff Water. Weed Science 16:447-449.
Trimble, S.W. 1975. A Volumetric Estimate of Man-Induced Soil Erosion
on the Southern Piedmont Plateau. Present and Prospective Technology
for Predicting Sediment Yields and Sources. ARS-S-40. U.S. Department
of Agriculture, Agricultural Research Service. Washington, D.C. pp. 142-154.
Triplett, G.B., Jr., B.J. Conner and W.M. Edwards. 1978. Transport of
Atrazine and Simazine in Runoff From Conventional and No-Tillage Corn.
Journal of Environmental Quality 7(l):77-84.
Truhlar, J.F. and L.A. Reed. 1976. Occurrence of Pesticide Residues
in Four Streams Draining Different Land-Use Areas in Pennsylvania 1969-1971.
Pesticides Monitoring Journal 10(3):101-110.
Tubbs, L.J. and D.A. Haith. 1977. Simulation of Nutrient Losses From
Cropland. ASAE Paper 77-2502. American Society of Agricultural Engineers.
St. Joseph, Michigan.
Tucker, B.V., D.E. Pack and J.N. Ospenson. 1967. Adsorption of Bipyridylium
Herbicides in Soil. Journal of Agricultural and Food Chemistry 15:1005-1008.
Tyler, D.D. and G.W. Thomas. 1977. Lysimeter Measurements of Nitrate and
Chloride Losses From Soil Under Conventional and No-Tillage Corn. Journal
of Environmental Quality 6:63-66.
318
-------
Uhland, R.E. and B.H. Hendrickson. 1946. Evaluation of Cropping Systems
for Soil and Water Conservation in the Southeast. Soil Science Society of
America Proceedings 11:527-530.
Underwood, R.C. 1976. No-Till is the Word. Water Research in Action,
Texas A§M University. College Station, Texas. l(3):l-2
U.S. Army Corps of Engineers. 1960. Runoff From Snowmelt. Manual 1110-2-
1406. Washington, D.C.
U.S. Comptroller General. 1977. National Water Quality Cannot be Obtained
Without More Attention to Pollution From Diffused or Nonpoint Sources.
Report to the Congress by the Comptroller General of the United States.
U.S. General Accounting Office. Washington, D.C. GA1.13:CED-78-6.
U.S. Department of Agriculture - Soil Conservation Service. 1969. Engineering
Field Manual for Conservation Practices. Washington, D.C.
U.S. Department of Agriculture - Soil Conservation Service. 1978. Environ-
mental Assessment Report Rural Clean Water Program. U.S. Department of
Agriculture. Soil Conservation Service. Washington, D.C.
U.S. Department of the Interior. 1966. Wildlife Research; Problems,
Programs and Progress, Pesticide-Wildlife Relations. Circular No. 43.
Fish Wildlife Service, Bureau of Sport Fisheries Wildlife. Washington, D.C.
U.S. Environmental Protection Agency. 1974. Draft Analytical Report - New
Orleans Area Water Supply. U.S. Environmental Protection Agency, Region VI.
Dallas, Texas.
U.S. Environmental Protection Agency. 1973. Proposed Criteria for Water
Quality. Vol. 1. Washington, D.C.
U.S. Environmental Protection Agency. 1975a. Initial Scientific and Miniecon-
omic Review of Malathion. Office of Pesticide Programs, Substitute Chemical
Program Report No. EPA-540/1-75-005. U.S. Environmental Protection Agency.
Washington, D.C.
U.S. Environmental Protection Agency. 1975b. Initial Scientific and Mini-
economic Review of Methyl Parathion, Office of Pesticide Programs. Substi-
tute Chemical Program Report No. EPA-540/1-75-004, U.S. Environmental Protection
Agency. Washington, D.C.
U.S. Environmental Protection Agency. 1975c. Initial Scientific and Mini-
economic Review of Parathion. Office of Pesticide Programs, Substitute
Chemical Program Report No. EPA-540/1-75-001. U.S. Environmental Protection
Agency. Washington, D.C.
U.S. Environmental Protection Agency. 1976. Initial Scientific and Minie-
economic Review of Carbofuran. Office of Pesticide Programs, Substitute
Chemical Program Report No. EPA-540/1-76-009. U.S. Environmental Protection
Agency. Washington, D.C.
319
-------
Vanoni, V.A., ed. 1977. Sedimentation Engineering. American Society of
Civil Engineers. New York, New York.
VanDoren, D.M. and G.B. Triplett. 1969. Mechanism of Com Response to
Cropping Practices Without Tillage. Ohio Agricultural Research and
Development Center, Research Circular No. 169. Wooster, Ohio.
Veith, G.D. and G.F. Lee. 1971. Water Chemistry of Toxaphene-Role of
Lake Sediments. Environmental Science and Technology 5(3):230-234.
Vettorazzi, G. 1977. State of the Art of the Toxicological Evaluation
Carried Out by the Joint FAO/WHO Expert Committee on Pesticide Residues.
III. Miscellaneous Pesticides Used in Agriculture and Public Health.
Residue Reviews 66:137-184.
Vollenweider, R.A. 1968. Scientific Fundamentals of the Eutrophication
of Lakes and Flowing Waters. Technical Report DAS/CSI/68.27. Organization
of Economic Cooperation and Development. Paris, France.
Von Rumker, R., G.K. Kelso, F. Moray and K.A. Lawrence. 1975. A Study of
the Efficiency of the Use of Pesticides in Agriculture. EPA-9/75-025.
U.S. Environmental Protection Agency. Washington, D.C.
Von Rumker, R., E.W. Lawless, A.F. Meiners. 1974. Production, Distribution,
Use and Environmental Impact Potential of Selected Pesticides. EPA-1/74-001,
U.S. Environmental Protection Agency. Washington, D.C.
Von Stryk, F.G. and E.F. Bolton. 1977. Atrazine Residues in Tile-Drain-
Water From Corn Plots as Effected by Cropping Practices and Fertility
Levels. Canadian Journal of Soil Science 57(3):249-53.
Walter, M.F., P.O. Robillard, R. Gilmour and R.W. Hexem. 1978. Best
Management Development for Manure in New York State. ASAE Technical Paper
No. 78-2033, American Society of Agricultural Engineers, St. Joseph, Michigan.
Walter, M., R. Hexem, P. Robillard, R. Gilmour and M. Harris. Best Manage-
ment Practices to Control Non-Point Sources of Water Pollution From Agriculture.
New York State Department of Environmental Conservation. Cornell University,
Ithaca, New York. (In Revision)
Ware, G.W., E.J. Apple, W.P. Cahill, P.O. Gerhardt and K.R. Frost. 1969.
Pesticide Drift II: Mist-Blower vs. Aerial Application of Sprays. Journal
of Economic Entomology 62(4):844-846.
Ware, G.W,, W.P. Cahill and B.J. Estesen. 1975. Pesticide Drift: Aerial
Applications Comparing Conventional Flooding vs. Raindrop Nozzles. Journal
of Economic Entomology 68(3):329-330.
Ware, G.W., W.P. Cahill, P.O. Gerhardt and K.R. Frost. 1970. Pesticide
Drift IV: On-Target Deposits From Aerial Application of Insecticides.
Journal of Economic Entomology 63(4):1982-1985.
320
-------
Ware, G.W., B. Estesen and W.P. Cahill. 1972. Organophosphate Residues on
Cotton in Arizona. Bulletin of Environmental Contamination and Toxicology
8(6):361-362.
Ware, G.W., B. Estesen, W.P. Cahill and N.A. Buck. 1977. DDTR Relocation
From Cotton Cultural Practices. Bulletin 'of Environmental Contamination and
Toxicology 17(3):323-330.
Ware, G.W., B.J. Estesen, W.P. Cahill and K.R. Frost. 1972. Pesticide
Drift VI: Target and Drift Deposits vs. Time of Applications. Journal of
Economic Entomology 65(4):1170-1172.
Ware, G.W., B. Estesen, W.P. Cahill and K.R. Frost. 1972. Pesticide Drift
V: Vertical Drift From Aerial Spray Applications. Journal of Economic
Entomology 65(2):590-592.
Ware, G.W., B.J. Estesen, W.P. Cahill, P.D, Gerhardt and K.R. Frost. 1969.
Pesticide Drift I: High Clearance vs. Aerial Application of Sprays. Journal
of Economic Entomology 62(4):840-843.
Ware, G.W., W.P. Cahill, B.J. Estesen, W.C. Kronland and N.A. Buck. 1975.
Pesticide Drift: Deposit Efficiency From Ground Sprays on Cotton. Journal
of Economic Entomology 68(4):549-550.
Warner, R.E., K.K. Peterson and L. Borgman. 1966. Behavioral Pathology in
Fish: A Quantitative Study of Sub-Lethal Pesticide Toxication. Journal of
Applied Ecology 3:223.
Wauchope, R.D. 1978. The Pesticide Content of Surface Water Draining From
Agricultural Fields - A Review. Journal of Environmental Quality 7(4):459-472.
Way, J.M., J.F. Newman, N.W. Moore and F.W. Knaggs. 1971. Some Ecological
Effects of the Use of Paraquat for the Control of Weeds in Small Lakes.
Journal of Applied Ecology 8:509-532.
Weber, J.B. and J.A. Best. 1972. Activity and Movement of 13 Soil-Applied
Herbicides as Influenced by Soil Reaction. Southern Weed Science Society
Proceedings 25:403-413.
Weber, J.B., R.C. Meek and S.B. Weed. 1969. The Effect of Cation-Exchange
Capacity on the Retention of Diquat and Paraquat by Three Type Clay Minerals
II: Plant Availability. Soil Science Society Proceedings 33:383-385.
Weber, J.B. and D.C. Scott. 1966. Availability of a Cationic Herbicide
Adsorbed on Clay Minerals to Cucumber Seedlings. Science 152:1400-1402.
Weber, J.B. and S.B. Weed. 1968. Adsorption and Desorption of Diquat,
Paraquat and Prometone by Montmorillonite and Kaolinite Clay Minerals. Soil
Science Society of America Proceedings 32:485-487.
321
-------
Keber, J.B. and S.B. Weed. 1974. Effects of Soil on the Biological Activity
of Pesticides. In: W.D. Guenzi, ed. Pesticides in Soil and Water. Soil
Science Society of America, Inc. Madison, Wisconsin, pp. 223-256.
Weber, J.B., S.B. Weed and T.J. Sheets. 1975. Pesticides: How They Move
and React in the Soil. Crops and Soils Magazine 25(1):14-17.
Weed, S.B. and J.B. Weber. 1969. The Effect of Cation Exchange Capacity
on the Retention of Diquat and Paraquat by Three- Layer Type Clay Minerals.
I. Adsorption and Release. Soil Science Society of America Proceedings
33:379-385.
Weed, S.B. and J.B. Weber. 1974. Pesticide-Organic Matter Interactions.
In: W.D. Guenzi, ed. Pesticides in Soil and Water. Soil Science Society
of America, Inc. Madison, Wisconsin, pp. 39-66.
Weed Science Society of America. 1974. Herbicide Handbook. Champaigne,
Illinois.
Whitaker, F.D., H.G. Heinemann and R.E. Russell. 1978. Fertilizing Corn
Adequately With Less Nitrogen. Soil Conservation Society American Journal.
pp. 28-32.
White, A.W., A. P. Barnett, B.C. Wright and J.H. Holladay. 1967. Atrazine
Losses From Fallow Land Caused by Runoff and Erosion. Environmental Science
and Technology 1(9) :740-744.
White, A.W., L.A. Harper, R.A. Leonard, and J.W. Turnbull. 1977. Trifluralin
Volatilization Losses From a Soybean Field. Journal of Environmental Quality
White, E.M. 1973. Water- leachable Nutrients From Frozen or Dried Prairie
Vegetation. Journal of Environmental Quality 2(1) : 104-107.
White, G.B. and E.J. Partenheimer. 1978. The Economic Implications of
Erosion and Sedimentation Control Plans for Selected Pennsylvania Dairy
Farms. In : R.C. Loehr, D.A. Haith, M.F. Walter, C. Martin, ed. Best
Management Practices for Agriculture and Silviculture. Ann Arbor Science,
Ann Arbor, Michigan, pp. 341-358.
White, R.W. and P.H.T. Beckett. 1964. Studies on the Phosphate Potentials of
Soils Part I. The Measurements of Phosphate Potential. Plant and Soil
20:1-15.
Wilkison, J.D., K.D. Biever and C.M. Ignoffo. 1975. Contact Toxicity of
Some Chemical and Biological Pesticides to Several Insect Parasitoids and
Predators. Entomophaga 20(1) : 113-120.
Williams, I.H. and M.J. Brown. 1976. Persistence of Carbofuran Residues
in Some British Columbia Soils. Bulletin of Environmental Contamination and
Toxicology 15 (2) : 242-243.
322
-------
Williams, I.H., H.S. Pepin and M.J. Brown. 1976. Degradation of Carbofuran
by Soil Microorganisms. Bulletin of Environmental Contamination and Toxicology
15(2):244-249.
Williams, J.R. 1975. Sediment Yield Prediction With Universal Equation
Using Runoff Energy Factor. Present and Prospective Technology for Predicting
Sediment Yields and Sources. ARS-S-40. U.S. Department of Agriculture.
Agricultural Research Service. Washington, D.C. pp. 244-252.
Williams J.R. and H.D. Berndt. 1977. Sediment Yield Prediction Based on
Watershed Hydrology. Transactions of the ASAE 20:1100-1104. American
Society of Agricultural Engineers. St. Joseph, Michigan.
Williams, R.R. and T.F. Bidleman. 1978. Toxaphene Degradation in Estuarine
Sediments. Journal Agricultural and Food Chemistry 26(1):280-282.
Willis, G.H. and R.A. Hamilton. 1973. Agricultural Chemicals in Surface
Runoff, Ground Water and Soil I: Endrin. Journal of Environmental Quality
2:463-466.
Willis, G.H., R.L. Rogers and L.M. Southwick. 1975. Losses of Diuron,
Linuron, Fenac and Trifluralin in Surface Drainage Water. Journal of Environ-
mental Quality 4:399-402.
Wischmeier, W.H. and D.D. Smith. 1958. Rainfall Energy and Its Relationship
to Soil Loss. American Geophysical Union Transactions 39:285-291.
Wischmeier, W.H. and D.D. Smith. 1965. Predicting Rainfall Erosion Losses
From Cropland East of the Rocky Mountains. Handbook 282. Agricultural
Research Service, U.S. Department of Agriculture. Washington, D.C.
Wischmeier, W.H. and D.D. Smith. 1978. Predicting Rainfall Erosion
Losses, A Guide to Conservation Planning. Handbook No. 537. U.S. Department
of Agriculture. Washington, D.C.
Wittmuss, H., L. Olson and D.L. Lane. 1975. Energy Requirements for Con-
ventional Versus Minimum Tillage. Journal of Soil and Water Conservation
30:(2)-72.
Wolfe, N.L., R.G. Zepp, D.F. Paris, G. Baughman and R.C. Hollis. 1977.
Methoxychlor and DDT Degradation in Water: Rates and Products. Environmental
Science and Technology 11 (12):1077-1081.
Woolhiser, D.A. and P. Rodorovic. 1974. A Stochastic Model of Sediment Yield
for Emphemeral Streams. Proceedings of the Symposium on Statistical Hydrology.
Miscellaneous Publication No. 1275. U.S. Department of Agriculture. Washing-
ton, D.C.
Yeo, R.R. 1967. Dissipation of Diquat and Paraquat, and Effects on Aquatic
Weeds and Fish. Weeds 15:42-46.
323
-------
Yoder, J., M. Watson and W.W. Benson. 1973. Lymphocyte Chromosome Analysis
of Agricultural Workers During Extensive Occupational Exposure to Pesticides.
Mutation Research 21:335.
Young, R.A. and J.W. Wiersma. 1973. The Role of Rainfall Impact on Soil
Detachment and Transport. Water Resources Research 9:1629-1636.
Yu, C.C., G.M. Booth, D.J. Hansen and J.R. Larsen. 1974. Fate of Carbofuran
in a Model Ecosystem. Journal of Agricultural and Food Chemistry 22(3):431-434.
Yu, C.C., R. Metcalf and G.M. Booth. 1972. Inhibition of Acetylcholinester-
ase From Mammals and Insects by Carbofuran and Its Related Compounds and Their
Toxicities Toward These Animals. Journal 'Of Agricultural and Food Chemistry
20(5):923-926.
324
-------
APPENDIX A
CORNELL NUTRIENT SIMULATION (CNS) MODEL
This appendix summarizes the mathematical details of the simulation
model used to estimate the effects of SWCPs on water, soil and nutrient losses
at locations in New York, Iowa and Georgia. Parameters used in model valida-
tions for New York and Georgia are also presented. The simulation model con-
sists of daily water balance and soil loss components and monthly nutrient
inventories.
WATER BALANCE
The hydrologic component of the simulation model is based on a daily
moisture balance for the top 30 cm of soil:
SWt+l = SWt + Mt + Pt ' ETt ' Et ' Rt - Lt CA'1)
where SW = soil water content of the top 30 cm of soil at the beginning
of day t(cm)
M = snowmelt on day t(cm)
P = precipitation as rain on day t(cm)
ET = evapotranspiration from the soil on day t(cm)
E = direct evaporation from the soil surface on day t(cm)
R = direct runoff, including interflow, on day t(cm)
L = leaching or deep percolation from the 30 cm surface soil
layer on day t(cm)
The soil water content is constrained to be above the wilting point WP, (cm),
and is assumed to drain to field capacity FC, (cm), unless the soil is frozen,
in which case moisture content may reach saturation SAT, (cm).
Evapotranspiration during the growing season is assumed equal to poten-
tial evapotranspiration as estimated by the formula developed by Hamon (1961)
provided soil water content exceeds the wilting point. Evaporation from bare
soil or snowpack is assumed to be temperature-dependent and is determined from
the equation developed by Eggleston et_ aL (1971). Snowmelt is determined
325
-------
using melt equations for clear or rainy days developed by the U.S. Army Corps
of Engineers (1960).
Runoff
During periods of the year when the soil is neither frozen nor covered
with snow, direct runoff is predicted using the U.S. Soil Conservation
Service's curve number equation (Mockus, 1972; Ogrosky and Mockus, 1964):
(P - 0.2S )2
Rt= P*+0.8S* forPt>0.2St (A.2)
and R = 0 for P <_ 0.2S
t t L>
where S is a runoff detention parameter for day t(cm). Values of S are
determined from curve numbers which are defined as functions of soil hydro-
logic group, crop, supporting practice, hydrologic condition and antecedent
moisture condition (AMC). The latter is usually determined using the five-day
antecedent rainfall. Since this simulation model computes soil water content
SW , antecedent moisture conditions can be determined directly from SWt,
provided soil moisture limits can be associated with the appropriate AMC
categories: I(driest), II(average), and III(wettest). These three levels
correspond to decreasing infiltration rates and increasing runoff. The soil
moisture limits used in the simulation model are given in Table A-l. These
limits were developed in such fashion that they are analogous to AMC categories
based on 5-day antecedent rainfall. Since the difference in antecedent rain-
fall for AMC I and AMC III is in the order of 2 cm, it is assumed that the soil
moisture limits should reflect a comparable difference. The determination of
soil moisture for the driest category is based on the knowledge that when
soil moisture falls below field capacity (FC), the soil macropores are open,
and infiltration rates should be high. The AMC I limit is thus arbitrarily
set to one centimeter below FC.
TABLE A-l. RELATIONSHIP BETWEEN SOIL WATER CONTENT
AND ANTECEDENT MOISTURE CONDITION
Soil Water Content, SW.(cm)Antecedent Moisture Condition,
AMC
SWt < FC-1.0
FC+1.0
SW > FC+1.0 III
The curve number equation was not developed to predict snowmelt runoff.
Frozen soil will impede infiltration and percolation and hence increase runoff
326
-------
over summer conditions. Since the curve number equation is one of the few
practical means for estimating runoff, it was adapted to snowmelt in the
current study by modification of the detention parameter St. At least in
humid areas, soil water content is at or exceeds field capacity during most
snowmelt events, and the modification assumes that at SW^ = FC when the soil
is frozen, a modified detention parameter S[ is equal to O.SS(III) , where
S(III)t is the unmodified detention parameter for the wettest AMC.1" Frozen
soil can have a moisture content exceeding field capacity, and when saturation
is reached, any further precipitation or snowmelt will all become runoff.
Assuming a linear relationship between field capacity and saturation (SAT, in
cm), the following equation estimates the modified detention parameter.
Rt
Rt
(Pt -
P
t
= 0
H Mt - 0.2SJ.)2
^ NT -i- 0.8S!
t t
where S' is a detention parameter for frozen soil (cm) . The runoff equation
for frozen soil becomes
for pt + Mt > °'2S; (A'4)
for P_ + M. < 0.2S!
t t t
Whenever the soil is thawed, runoff is predicted using Equation (A. 2) substitu-
ting P + M for P .
L. L C-
Percolation
Percolation, or leaching occurs whenever soil moisture level exceeds field
capacity. Drawing the analogy to five-day antecedent rainfall, the maximum
daily percolation rate is assumed to be equal to (SAT - FC)/5.
SOIL EROSION
Soil loss is estimated from the modification of the universal soil loss
equation proposed by Onstad and Foster (1975).
SL = 2.24WtK(LS)CtP (A.5)
where SL = soil loss on day t(MT/ha)
W = rainstorm kinetic energy
C = cover factor, day t
LS = topographic factor
P = conservation practice factor
327
-------
K = soil credibility factor
The parameters C , LS, P and K are the standard tabulated factors used in the
universal soil loss equation. However, the energy term, Wt is computed for
the rain falling on day t as follows:
15<2754)C2754>
and
Et - [916 + 331 logC^J ^54 (A-7)
where I = the maximum 30-min rainfall intensity(cm/hr)
q = peak runoff(cm/hr)
I = average storm rainfall intensity(cm/hr)
To use the soil loss equations for long simulation periods, it is essen-
tial to estimate I ^, q and I from standard meteorological data (hourly
rainfall). Hence, these values are determined from 1C 60, the maximum 60-min-
ute clock rainfall intensity (cm/hr) and storm duration D (hr). These three
required factors are computed as follows
Average storm rainfall intensity is given by
Tt = Pt/Dt (A.8)
Maximum 60-min intensity I 60, is approximately 1.13 IC.6^ (Hershfield, 1961).
As indicated in Schwab et_ _aL (1966, p. 624),
It3°= 1.58 I/0 (A.9)
Assuming that storm duration is greater than the field's time of concentration,
T (hr), peak runoff is given by
qt = dt - ft) (A. 10)
where I is the average rainfall intensity during the time in which T is
reached (cm/hr) and f is the infiltration rate. c
To compute I , and estimate must be made based on I . The graphical
relationship given in Schwab et^ al. (1966, p. 624) can be approximated as
328
-------
1.34Tc°-62It60
°Tr Tc- 1/3 hr (A. 11)
I c =
T 0.34j 60
c t
T T > 1/3 hr
c
where TC is given by Schwab et^ al. (1966, p. 629) as
1.67(3.28£)°'8(St/2.54 + I)1'67
In equation 12 £ and s are slope length and gradient, respectively, in meters
and percent. As a final step, the infiltration rate f (cm/hr) is estimated
as the minimum rates for the relevant soil category (Musgrave and Holton,
1964): A: 0. 75-1. 15cm/hr, B: 0.40-0.75cm/hr, C: 0.15-0.40cm/hr, D:
0-0.15cm/hr.
The soil loss equation A. 5 is applicable to rainfall erosion only. In
northern climates, erosion associated with snowmelt is neglected.
NUTRIENT BALANCE
While it is apparent that daily soil water levels vary sufficiently that
reasonable runoff predictions can be obtained only from a daily model, soil
nutrient levels can be adequately modelled on a monthly basis. Moreover,
since the dates of fertilizer and manure applications are often uncertain,
there is little point in the 30- fold increase in computations required by
a daily over a monthly nutrient model. The nutrient balance model estimates
monthly losses of dissolved and solid-phase nitrogen (N) in runoff, dissolved
N in percolation, dissolved available (resin - extractable) phosphorus (P) in
runoff and solid-phase P in runoff. It is assumed that all dissolved nutrients
are in the inorganic form and that solid- phase N is organic. Losses of
solid-phase nutrients are assumed to be fixed or adsorbed to soil particles.
The monthly runoff, percolation and soil loss values required for the nutrient
balances are obtained by summing the daily values predicted by the water
balance and erosion components of the simulation model.
Nitrogen
The soil nitrogen budget is described by inventory equations for soil
inorganic and organic nitrogen.
NIn+l = NIn + mnN°n + Mn + PIn ' RIn ' LNn ' CNn + FNn (A. 13)
N°n+l = NOnC1"V " R°n * MN°n (
329
-------
where NI = soil inorganic nitrogen content at the beginning of month in
n (kg/ha)
NO = soil organic nitrogen content at the beginning of month n
n (kg/ha)
m = fraction of soil organic nitrogen mineralized to inorganic
n
nitrogen during month n
MN = inorganic nitrogen added to the soil from manure during month
n n(kg/ha)
PI = precipitation additions of inorganic nitrogen in month n(kg/ha)
RI = inorganic nitrogen lost in runoff during month n (kg/ha)
LN = inorganic nitrogen lost in percolation (leaching) during month
n(kg/ha)
CN = crop uptake of inorganic nitrogen during month n(kg/ha)
FN = fertilizer applications of inorganic nitrogen during month
" n(kg/ha)
RO = organic nitrogen lost in eroded soil during month n(kg/ha)
MNO = addition to soil organic nitrogen content from manure and/or
fresh plant residues during month n(kg/ha)
The inorganic nitrogen budget is applied to the top 30 cm of soil as with the
hydrologic model. The organic nitrogen budget is limited to the top 10 cm,
since the microbial populations responsible for mineralization are most pre-
valent near the soil surface. When applied to legumes, the fertilizer nitrogen
input, FN , must include nitrogen fixation inputs. The various chemical forms
of inorganic nitrogen are lumped together in equation A.13, implying that
oxidation to nitrate is rapid; i.e., most available soil inorganic nitrogen
exists as nitrate.
Equation A. ]3 does not account for gaseous losses of soil inorganic nitrogen
due to denitrification and ammonia volatilization. Such losses can be sub-
stantial, but they are very difficult to model. Although kinetic relationships
can be developed for denitrification and ammonia volatilization, the appro-
priate rate constants for these processes cannot be directly estimated from
soil properties. To partially account for gaseous losses in applications of
the model, 25% of the fertilizer N inputs are substracted.
The soil organic nitrogen, NO , refers to the relatively stable organic
nitrogen tied up in humic and other resistant material. Such nitrogen miner-
alizes slowly (2-3% annually) and must be distinguished from the organic nitro-
gen in fresh manure or plant residues which can mineralize more rapidly.
Nitrogen transformations in manure which has been applied to the soil are very
difficult to describe quantitatively and in this simulation model they are
330
-------
described in a rather arbitrary fashion as follows:
1. Based on the studies of Chichester et al. (1975) it is assumed that
manure organic nitrogen mineralizes at three times the rate of soil
organic nitrogen (i.e., 3mn) during the first year following appli-
cation. In the second year this rate is reduced to 1.5m and there-
after it is assumed that manure organic nitrogen joins tfte soil
organic nitrogen pool (MNO in equation A. 14).
2. The addition of inorganic nitrogen from manure (MNL in equation A.13
is the sum of the manure organic nitrogen mineralized during month n
and the inorganic nitrogen in the manure applied during the month.
The latter is adjusted for ammonia losses in surface spreading.
Based on the work of Lauer et_ al. (1976), it is assumed that such
losses are 85% of the ammonia content of the manure when spread.
Fresh plant residues are assumed to have mineralization properties similar to
manure.
The mineralization rate, m is assumed temperature-limited, and can be
modelled by the Van't Hoff-Arrhenius relationship (Haith, 1973). Within the
typical range of soil temperatures (0-20°C), a linear approximation is possible
(Haith et al. 1977) :
T
m = m ~ for T > 0 (A. 15)
n o IT n —
n
where m = yearly mineralization rate as a fraction of the average soil
0 organic nitrogen content(dmless)
T = average air temperature in month n(°C)
Equation A. 15 apportions the yearly mineralization to the various months on
the basis of degree-months. Mineralization is assumed zero during any month
in which the average air temperature is less than or equal to zero.
Precipitation inputs of inorganic nitrogen, PIn, are relatively small
and may be neglected or modelled as:
PIn =
where np = concentration of inorganic nitrogen in precipitation in
n month n(mg/£)
P = total precipitation in month n(cm)
n
Crop uptake of nitrogen is assumed to follow a sigmoid curve of seasonal crop
growth. The curve is scaled by the growing season and total nitrogen uptake
over the growing season.
331
-------
The most important part of the nutrient model is the prediction of losses
in runoff and percolation. These losses are modelled as the product of
concentration and water volume, where concentration is determined from average
nitrogen level in the soil during the month. The average concentration of
inorganic nitrogen in percolation or leaching water L during month n is given
by NI /(FC + R + L ) where NI is the average soil inorganic nitrogen content
in month n(kg/Ra) and PC is field capacity. Average soil inorganic nitrogen
is:
NI + NI .
NT = — ^ - S±L_ (A. 17)
n z
The inorganic nitrogen is mixed in water volume FC + RR + L^, since percola-
tion occurs only after the soil reaches field capacity. Percolation losses
of inorganic nitrogen are thus given by:
IT
LN = L "
n n FC + R
A similar computation is made for losses of inorganic nitrogen in runoff.
NT
RI = R "
n n SAT + R + L
n n
The soil's saturation water content (SAT) is arbitrarily substituted for field
capacity (FC), in determining nitrogen concentration. This substitution is
made to produce the lower nitrogen concentration in runoff than in percolation
that has been reported elsewhere (Biggar and Corey, 1970) . It is rationalized
by the observation that, when runoff occurs., the near- surface soil is saturated.
This is not, however, the most satisfactory way to make the concentration
adjustment. Lower nitrogen concentrations in runoff than in percolation are
due to the fact that more soil nitrogen is exposed to percolation water than
runoff. Thus, in equation A. 19", NIn should be replaced by some fraction of
NI . The fraction will depend on the distribution of runoff and interflow,
ana there appears to be no practical way to quantify this distribution. The
curve number runoff equation predicts the sum of interflow and surface runoff
and there have been very few field studies in which the two runoff components
were measured separately.
The organic nitrogen lost with eroded soil RO as defined for equation A. 14,
is given by:
SL NO ERN
where SL = average soil loss for month n (kg/ha)
NO = average organic nitrogen in the top 10 cm of soil (kg/ha)
332
-------
ERN - enrichment ratio for organic nitrogen in eroded soil
WT10 = weight of the top 10 cm of soil
The organic nitrogen and soil weight terms in equation 20 are limited to
the top 10 cm since this upper layer is the basis for the organic nitrogen
mass balance (equation A.21). The average organic nitrogen content is given
by:
NO + NO ,
Mn - % ^ (A-21)
The enrichment ratio is typically between 2.0 and 4.0 (McElroy et^ al 1976).
A value of 2.5 Is used in the simulation model.
Phosphorus
The reactions of phosphorus (P) in the soil-water system are substantially
different from nitrogen reactions, and must be modelled according to a differ-
ent set of assumptions. Unfortunately, organic P transformations in the soil
are not well defined. The simulation model considers only inorganic P, and
it is assumed that the breakdown of complex organic P compounds into inorganic
P will not greatly affect the availability of P to crops or increase runoff
losses of P. Furthermore, the phosphorus inventory equations consider only
available inorganic P (orthophosphate-P). Disregarding the transformations
between organic and inorganic P and between fixed and available P essentially
limits the simulation model's ability to predict P losses to the short term
(2-3 years). The two basic inventory equations for available P are:
TP = PA + PD (A.22)
n n n
and
TP = TP + FP + MP - CP - RPD - LPD - RPA (A. 23)
n+1 n n n n n n n
where TP = total available P in the surface soil layer at the beginning
n of month n(kg/ha)
PA = adsorbed available P in the surface soil layer at the
n beginning of month n(kg/ha)
PD = dissolved available P in the soil water at the beginning
n of month n(kg/ha)
FP = fertilizer inorganic phosphorus added to the soil during
n month n(kg/ha)
MP = manure inorganic phosphorus added to the soil during month
n n(kg/ha)
CP = phosphorus removed from the soil by plant uptake (kg/ha)
n during month n(kg/ha)
333
-------
RPD = loss of dissolved available P in runoff during
n month n(kg/ha)
LPD = loss of dissolved available P in percolation during month
n
n(kg/ha)
RPA = loss of adsorbed available P in eroded soil during month
n n(kg/ha)
Crop removal of soil phosphorus, CP , is determined in a fashion similar to
crop nitrogen removals.
The equilibrium between adsorbed and soluble phosphorus is approximated by
a linear isotherm:
= Kp(pdn)
(A.24)
where pa = adsorbed available P concentration in the surface soil
n layer at the beginning of month n(ppm; mg/kg)
pd = dissolved available P concentration in the surface soil
n water at the beginning of month n(ppm; mg/£)
K = P adsorption coefficient
In terms of phosphorus contents in kg/ha, equation A.24 becomes:
K (WT7n)
PA = P 3QJ
n iobsw
n
n
(A.25)
where WT,n = weight of the top 30 cm of soil (kg/ha)
The equilibrium constant K is a function of soil properties. Data from
Enfield and Bledsoe (1975)*were used to develop the regression equations given
in Table A-2. The independent variables used in the equations are pH and
TABLE A-2. PREDICTIVE EQUATIONS FOR PHOSPHORUS ADSORPTION COEFFICIENT
Number of Soils
Number of Soils
Omitted
Regression
19 0
18 1
17 2
K =
P
K =
P
K =
28
1.
5.
.4 +
5 +
1 +
2.
3.0
2.2
5(%C) + 21
|(%C) + 24.
:f%c) + 26.
.6(pH-6)2
4(pH-6)2
4fpH-6)2
0
0
0
.42
.53
.62
334
-------
percent clay (%C), defined as the soil particle percentage with diameters less
than 0.002 mm. Enfield and Bledsoe equilibrated 25 soils at a variety of
initial dissolved P concentrations. Equilibrium times ranged from one to
3000 hours. The regression equations given in Table A-2 were determined from
Enfield and Bledsoe's experimental data for low (0.5-1.0 mg/£) initial P
concentrations and 300-hr equilibrium times. Six soils were analyzed de-
structively and 19 were analyzed nondestructively. For consistency, only re-
sults from the latter were used in the regressions. The three equations given
in Table A-2 are based on elimination of zero, one or two outliers. Although
the equations give similar results, elimination of the two most extreme soils
significantly improves the fraction of variation explained (from 0.42 to 0.62),
and thus the final equation,
K = 5.1 + 2.2(%C) + 26.4(pH-6)2 (A.26)
r
is used in the simulation model.
Runoff and percolation losses of available P are computed similarly to
nitrogen losses:
PD
RPDn = Rn SAT + R + L CA.27)
n n
PD
LPDn = Ln FC * Rn * L (A. 28)
n n
where PD~ = average dissolved available P in the soil during month
n n(kg/ha)
PD + PD
PD = " 2 n+ (A. 29)
The estimate of P percolation losses (LPD ) is somewhat misleading, since the
leached P is likely to be removed from percolation waters by adsorption in the
subsoil well before reaching groundwater aquifers.
Losses of adsorbed available phosphorus RPAn are given by:
SL PA ERP
RPA _ n n (A. 30)
n WT3Q
where WT Q = weight of the top 30 cm of soil (kg/ha)
ERP = enrichment ratio for adsorbed P in eroded soil
PA = average adsorbed available P in the surface soil during
n month n(kg/ha)
335
-------
and
PA + PA
(A. 31)
Total loss of solid phase phosphorus is the sum of both available and fixed P.
SL Erp(PA + PF)
PPn ' " WT^ - ^'^
where PF is the fixed (unavailable) phosphorus in the surface soil layer
(kg/ha). As reported in McElroy et_ al. (1976), the phosphorus enrichment ratio
is similar in magnitude to the ratio for organic nitrogen. A value of ERP =
2.0 is used in the simulation.
VALIDATION DATA
Georgia Validations
Principal data for the Watkinsville validations are given in Table A-3.
Validation results are presented in Table 6-4. Crop curve numbers and cover
factors (C ) were changed according to the crop management changes summarized
by Smith e£ aL (1978). Nutrient inputs and crop yields were obtained from
this same source. Yields were subsequently used to estimate crop nutrient
uptake.
New York Validations
Analogous data for the -Aurora validations are given in Table A-4. Valida-
tion results are presented in Table 6-6. Crop curve numbers and cover factors
were based on continuous corn, with or without residues left during winter
months (corresponding to well and poorly managed plots). Nutrient inputs and
crop yields were obtained from Klausner et_ aL . (1976) and from unpublished data
provided by Stuart D. Klausner, Department of Agronomy, Cornell University.
APPENDIX A REFERENCES
Biggar, J.W. and R.B. Corey. 1969. Agricultural Drainage and Eutrophi-
cation. In: Eutrophication: Causes, Consequences, Correctives. National
Academy of Sciences, Washington, D.C. 00. 404-405
Chichester, F.W. ., J.O. Legg and G. Stanford. 1975. Relative Mineralization
Rates of Indegous and Recently Incorporated N1C Labelled Nitrogen. Soil
and Science 120(6): 455-460. 15
Eggleston, K.O., E.K. Israelson and J.P. Riley. 1971. Hybrid Computer
Simulation of the Accumulation and Melt Processes in a Snowpack. Utah
State University, Logan, Utah.
336
-------
TABLE A-5. PARAMETER VALUES USED FOR WATKINSVILLE, GA. MODEL VALIDATION
Parameter
SAT
FC
WP
£t
m
Value
PI P2 P3 P4
U7 rm
8 A. r-m
47 r»m
Of\ /"*tn i\\ ~Y*
• o t-in/ fir ——— — — —— — — — — — —
„ n ?s
n nx _ __ _
o
WT1() 1.65 106kg/ha
P
£
s
LS
NI.
1
NO,
1
TP1
PF
PH
%C
0.5 0.6 0.5
100m 61m 26m
4% 2% 3%
0.64 0.25 0.27
5 kg/ha
580 "
150 "
750 !'
5.8
17.0%
0.5
26m
3%
0.27
5 kg/ha
790 "
115 "
690 "
6.1
22.5%
Enfield, C.G. and B.E. Bledsoe. 1975. Kinetic Model for Orthophosphate
Reactions in Mineral Soils. EPA-600/2-75-022. U.S. Environmental
Protection Agency, Corvallis, Oregon.
Haith, D.A. 1973. Optimal Control of Nitrogen Losses From Land Disposal
Areas. Journal of Environmental Engineering Division, American Society of
Civil Engineers 99(EE6): 923-937.
337
-------
TABLE A-4. PARAMETER VALUES USED FOR AURORA, N.Y. MODEL VALIDATIONS
ParameterValue
SAT 15.0 cm
FC 9.8 cm
WP 4.9 cm
NI 15 kg/ha
NO 1610-3770 kg/ha
m 0.02
o
WT1Q 1.31 106 kg/ha
TP 16-90 kg/ha
PF 3100 kg/ha
%C 11.1
pH 6.7
Haith, D.A., A. Koenig and D.P. Loucks. 1977. Preliminary Design of
Wastewater Land Application Systems. Journal Water Pollution Control
Federation 49(12): 3271-3279.
Hershfield, D.M. 1961. Rainfall Frequency Atlas of the United States
for Durations from 30 Minutes to 24 Hours and Return Periods from 1 to
100 Years. U.S. Weather Bureau Technical Report 40. U.S. Government Print-
ing Office. Washington, D.C.
Klausner, S.D., P.J. Zwerman and D.R. Coote. 1976. Design Parameters
for the Land Application of Dairy Manure. EPA-600-2076-187. U.S.
Environmental Protection Agency. Athens, Georgia.
Lauer, D.A., D.R. Bouldin and S.D. Klausner. 1976. Ammonia Volatilization
From Dairy Manure Spread on the Soil Surface. Journal Environmental Quality
5(2): 134-141.
McElroy, A.D., S.Y. Chen, J.W. Nebgen, A. Aleti and F.W. Benett. 1976.
Loading Functions for Assessment of Water Pollution From Nonpoint Sources.
EPA 600/2-76-151. U.S. Environmental Protection Agency. Kansas City,
Missouri.
338
-------
Mockus, V. 1972. Estimation of Direct Runoff From Storm Rainfall. National
Engineering Handbook. Sec. 4, Hydrology. U.S. Soil Conservation Service
Washington, D.C.
Masgrave, G.W. and H.N. Holtan. 1964. Infiltration. In: V.T. Chow, ed.
Handbook of Applied Hydrology. Chapter 12. McGraw-Hill, New York.
Ogrosky, H.O. and V. Mockus. 1964. Hydrology of Agricultural Lands. In_:
V.T. Chow, ed. Handbook of Applied Hydrology, Chapter 21. McGraw HilT7
New York.
Onstad, C.A. and G.R. Foster. 1975. Erosion Modelling on a Watershed.
Transactions of the American Society of Agricultural Engineers 18(2):
288-292.
Schwab, G.O., R.K. Frevert, T.W. Edminster and K.K. Barnes. 1966. Soil
and Water Conservation Engineering. John Wiley and Sons, Inc. New York,
New York
Smith, C.N., R.A. Leonard, G.W. Landale and G.W. Bailey. 1978. Transport
of Agricultural Chemicals From Small Upland Piedmont Watersheds. EPA Pre-
liminary Copy. U.S. Environmental Protection Agency. Athens, Georgia.
U.S. Army Corps of Engineers. 1960. Runoff From Snowmelt. Manual 1110-
2-1406. Washington, D.C.
339
-------
APPENDIX B
THE CORNELL PESTICIDE MODEL (CPM)
This appendix summarizes the mathematical details of the simulation model
which evaluates the interaction of water, soil and pesticide characteristics
as they affect pesticide losses in surface and leaching water. Parameters used
as input for model validation are also presented.
The simulation model is event based. Water and pesticide balances, and
soil loss components are updated when a rain day or root zone depth change
occurs. Downward movement of pesticides is calculated only after a signifi-
cant quantity of water has infiltrated into the soil.
The soil profile is divided into four layers in this model (see Figure
7-2). Zone 1 is the first 5 cm of top soil. Zone 2 includes the mass of the
roots and extends from 5 cm depth to the bottom of the active rooting zone.
Zone 3 denotes the soil below the active rooting zone and above the maximum
depth of rooting. Zone 4 is the remaining soil down to the impermeable layer.
Figure B-l shows, the development of the active root zone for corn at
various locations in the United States. Step functions which simulate the
change in the size of the active rooting zone through the year were based on
the curves shown for corn. These functions are shown in Figure B-2. The
variation between locations was due to climate and depth to impermeable layer.
THE SUB-MODELS
The simulation model consists of 5 sub-models, each of which reads in
data from tape and records results on tape (see Figure 7-1). This structure
was chosen so that model components could be changed easily, and so that
experimental findings could be substituted for indirectly determined input
data.
Sub-model TEMPMELT
Event Based' Data Set--
The soil temperature and snowmelt sub-model first reduces the input
daily weather data set to event based data set. Information is filed only
for those days on which 0.01 cm or more of snowmelt or rainfall occurs.
Snowmelt is determined by using melt equations developed by the U.S. Army
Corps of Engineers (1960).
340
-------
Days from planting
10
20
110
20 -
40
60
80
100
120-
14C
160
D corn planted May 1963 at Ames, Iowa
Shaw, R.H., Iowa State University of Science
and Technology. Res. Bull. 520
O corn planted May 1959 at Knoxville, Tennessee
Long, O.H., 1959 Tennessee Agricultural Experiment
Station Bull. 301
A corn planted June 1959 at East Lansing, Michigan
Foth, H.D., 1959 at Michigan Agricultural Experiment
Station Quarterly Bull. 43:2-13
FIGURE B-l- LENGTH OF ROOTS OF CORN.
-------
to
Apri
12
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
-1 30
!0
•L_
»
m
-
•
m
»
m
•»
»
•>
May 30
150
1
1
|,
•••••••••I
L
I
July 30 Sept. 30 Oct. 27 Nov. 11
200 250 300 315
Zone 3
1 •
INew York
Georgia
Iowa
I
FIGURE B-2. LOWER BOUNDARY OF ROOT ZONES IN THE CORNELL PESTICIDE SIMULATIONS
-------
Soil Temperature- -
Next, a preliminary value for soil temperature is calculated as the
running average of the previous 20 days' air temperature. Here, the
observed value of air temperature is used except when there is snow accumu-
lated on the ground; for this condition, the insulating property of snow is
taken into account. On days when the water equivalent of snow exceeds 1.0
cm, air temperature is adjusted empirically as follows:
10IR™TEREQ (B.I)
where ADJAIRTEMP is the adjusted air temperature, °C
AIRTEMP is the actual air temperature, °C
WATEREQ is the water equivalent of snow, cm
To develop a predictive equation, preliminary soil temperatures were then
regressed with soil temperatures observed at 10 cm depth. Observations are
not available for all years of the simulation. Regression equations for
New York and Iowa are:
New York: SOILTEMP = 1.53 + 0.98 • ADJAIRTEMP (r2, = 93%) (B.2)
Iowa: SOILTEMP = 0.52 + 0.84 • ADJAIRTEMP (r = 90%) (B.3)
Hydrological Sub-model
Moisture Budget --
The hydrological sub-model budgets soil moisture for each of the top
three zones in the soil profile as follows:
SW(l)t = SW(l)t_At + Mt H- Pt - ET(l)At - Rt - FLOW12 - INTCEP (B.4)
SW(2)t = SW(2)t_At + FLOW12 - FLOW23 - ET(2)At (B.5)
SW(3) = SW(3)t At + FLOW23 - FLOW34 (B.6)
where SW(J) = soil water content of Zone J at the end of day t, cm
At = the number of days elapsed since the last rain or snowmelt
event
M = snowmelt on day t, cm
P = rainfall on day t, cm
ET(J) = evapotranspiration from Zone J in the period day (t-At)
At through day t, cm
R = runoff on day t, cm
343
-------
FLOWIJ = water flow from Zone I to Zone J, cm
INTCEP = interception of rain water by plants, cm
Soil Water Movement--
In deep well-drained soils with moisture contents above field capacity,
water moves readily under gravitational force from one zone to the next when
the moisture content is above field capacity. For the purposes of simplicity,
it is assumed that water does not move downward when the water content is
below field capacity. Furthermore, when the water content reaches the wilting
point, water is no longer available for plant uptake.
Runoff--
During periods of the year when the soil is neither frozen nor covered
with snow, the partition of rainfall into an infiltration and a runoff com-
ponent may be accomplished by one of three methods. The first method is
based on the SCS Curve Number equation and is similar to the method used in
the Cornell Nutrient Simulation Model (CNS) . Appendix A gives details of
these methods .
The second method is based on the Green and Ampt infiltration equation
(Green and Ampt, 1911). This equation assumes a sharp wetting front, a con-
stant hydraulic conductivity in the wetted zone, and a constant negative
water pressure at the wetting front. The Green and Ampt equation can be
expressed as:
i = k (B.7)
where i = infiltration rate, cm/hour
k = hydraulic conductivity of the soil in wetted zone, cm/hour
hw = suction head at wetting front , cm
z = depth of wetting front, cm
Values for the saturated hydraulic conductivity and condition of wetting
front for the soil groups are given in Table B-l.
Extensions of the Green and Ampt concept by Bouwer (1976) and Mein and
Larson (1971) assume continuous ponding of water on the soil surface. They
are, respecitively,
I = f • z (B.8)
and
t = I • [z-[hw - in C^-5-)]] (B.9)
where I = the amount of water which has infiltrated into a uniform
soil when the wetting front is at depth z, cm
344
-------
f = the difference between the volumetric water content above and
below the wetting front; also called the finable pore space,
cm-5/cm-5
t = the time required for the wetting front to reach depth z under
continuous ponding, hr
TABLE B-l. HYDRAULIC CONDUCTIVITY AND SUCTION AT WETTING POINT
Hydraulic soil group
A
B
C
D
Hydraulic conductivity*
(cm/hr)
0.9
0.6
0.3
0.1
Suction at wetting front*
(cm)
5
22.5
10
7
* England (1970)
**Compiled from data of Li e£ aju (1976)
If rainfall intensity varies, ponding is not continuous. In this case,
the Green and Ampt equation is modified as follows:
1) The rainfall event is divided into periods with constant
intensities
2) For each such period, an effective rainfall intensity is computed
to take into account the water stored in the surface depressions
Thus,
p • t + ST
* per
Peff = til (B.10)
where p __ = effective rainfall intensity, cm/hr
reff
p = rainfall intensity, cm/hr
t = duration of sub-period, hr
per
ST = water stored in depressions,
cm
3) The depth of the wetting front at the end of the period is found by
assuming that all the rain and ponded water has infiltrated during
this period. Thus,
345
-------
z* = z
(t/£)
b eff per
where z* = hypothetical depth of wetting front if all water
would have been infiltrated, cm
z = depth of wetting front at the start of the period, cm
As the wetting front crosses a soil zone boundary, a weighted
average of the f's of all the two zones is substituted for f in
Equation 11.
4) The infiltration rates at the beginning (L) and the end (i*ncj) of
the period are found by using Equation 7:
hw + z,
ib = -j— (B.12)
i* ,
end
5) The computations of surface runoff, water stored in depressions,
and depth of wetting front at the end of the period depend on the
relative magnitudes of i, , i* , and p
5a) When the effective rainfall intensity (peff) is smaller than the
soil infiltration rate at the end of the period (i*enci) > there is
no surface storage at the end of the period, no runoff generated
during the period, and the wetting front will reach the maximum
possible depth for that period, z*. In symbol form,
z = z* (B.14)
R = ° (B.15)
ST = 0 (B>16)
where R is runoff (cm)
5b) If the infitration rate at the beginning of the period (t^) is
less than or equal to the rainfall intensity, then water will pond
or run off. Then the depth of the wetting front at the end of the
period, zen(j, computed using a variation of Equation B.9 derived
from Mein and Larson (1974) :
f hw + z ,
t = f- [(Z , - z. ) - [hw • In (r - — )]] fB 17)
per k ^ end b hw + z, JJJ 1D- •"•'./
= *b - ^nd (B.18)
This equation cannot be solved directly for zen(j. Rather, the
computer program uses an iterative process. "As before, the f is
adjusted when a soil zone boundary is passed.
346
-------
The runoff during the period can be found as:
R = er ' ? - STmax + ST CB.19)
where STmax = the maximum quantity of water which can be stored in
depressions, cm
STb = dePression in storage at beginning of the period, cm
If the runoff computed by Equation B.19 is negative, the surface
storage is
STend = STb = P ' 'per (B'20)
and the runoff is set to 0. Otherwise,
STend = STmax (B-21)
5c) Finally, if the effective rainfall intensity (peff) is smaller than
the infiltration rate at the beginning of the period (i^) but larger
than the infiltration rate at the end of the period (i*end)j then
sometime during the period, the p __ will become equal to the infil-
tration rate .
The wetting front depth when this occurs, z is:
z = hw • k (B.22)
po Pe£f - k
The time elapsed from the beginning of the period to reach z can
be found as
t = f (2b." V) . t (B.23)
P0
and the surface storage at that time is
STPo = STb - V (Peff - P) CB
Hereafter, the surface runoff wetting front, and surface at end of
period can be found with equations B.17 - B.21 with these substitutions:
(B-25)
z = z
zb Zpo
347
-------
where t, = new value of t,
b D
t, = previous value of t,
The third method for finding runoff is used only for shallow soils with
hardpans within 50 cm of the surface. This method assumes that runoff occurs
only after the soil is completely saturated. Water flow in fragipan soils is
mainly horizontal. Therefore, this flow is slower than the flow in deep
drained soils. The maximum flow, "MASFLRATE", rate in fragipan soils is:
MAXFLRATE - (SAT - FC)^ DEPIMP (;
Water does not flow when soil moisture content is below field capacity.
Infiltration in Frozen Soils--
Water infiltration decreases as soil temperature decreases, when the
water in the soil is frozen. When using the SCS curve number equation, the
CPM compensates for this phenomenon by adjusting the curve number according
to the frozen soil infiltration method of Holtan et al. (1975). If the
average daily air temperature remains below 30°F for two weeks, hydraulic
conductivity is decreased by 0.1 inch per hour for each degree below 30°F.
The SCS Curve Number equation is not directly sensitive to hydraulic
conductivity. For a shallow soil with Antecedent Moisture Condition III,
(SCS, 1964) the infiltration rate approximates the saturated hydraulic
conductivity. Saturated hydraulic conductivities for each soil group are
given in Table B-l. By regressing the hydraulic conductivity with the curve
number for each group, the following relationship could be calculated:
CN = 99.55 - 10.72 • k (r2 = 95%) (B.29)
where CN = curve number
k = hydraulic conductivity, cm/hr
Then by decreasing hydraulic conductivity by 0.45 cm/hr for each degree (°C) of
soil temperature below 0°C, the Curve Number for temperatures below freezing
becomes:
CN_ = CN . - 4.8 • SOILTEMP n
fr orig ^
where CN_ = curve number when soil is frozen
rr
CN . = curve number for thawed soil, Antecedent Moisture Condition III
SOILTEMP = soil temperature, °C
This equation uses daily soil temperature, not the 14 day average air
temperature. It was found in Section 7 that in New York, the 20-day average
348
-------
temperature^as^bout^l^^C^lower^than the temperature of a nearly frozen
*> °--s -on as
Evapotranspiration- -
The evapotranspiration model is based on the frequently used assumption
that evapotranspiration will take place at the "potential rate" if thTsoU
is near field capacity The acutal evapotranspiration will be less than
"mioint S011 drleS' ^ fallS t0 zero as the soil approaches
Potential evapotranspiration is computed by the sinusoidal function
EP . Sex . (1 . sin ^^^100, . „, t Epbase
where Ep is potential evaporation, cm/day
Epmax is the maximum yearly potential evaporation, cm/day
JULDATE is the Julian date
Epbase is the minimum yearly potential evaporation, cm/day
In winter, evaporation decreases with the square root of time after
Zone 1 has reached field capacity. If the Zone 1 moisture content is less
than field capacity, evapotranspiration is computed as a function of mois-
ture content.
In the growing season, the plants use water from Zones 1 and 2 for
their transpiration. Assuming that evapotranspiration decreases linearly
with moisture content from field capacity to wilting point, the water loss
due to evaporation from Zones 1 and 2 can be expressed as
LDEPTH(l) + DEPTH(2)JLFC - WPJ (B.32)
• exp |_DEpTH(l) + DEPTH(2) J|_FC - WPJ (B.33)
where ET(J\t = evapotranspiration from Zone J in the period day t-At
through day t, cm
At = number of days elapsed since the last rain or snowmelt
SW(J) . = soil water content of Zone J on day t-At, i.e. the day
" t of the last event, cm
DEPTH (J) = width of Zone (J)
349
-------
3 3
FC = moisture content at field capacity, cm /cm
. . . 3, 3
WP = moisture content at wilting point, cm /cm
Sediment Sub-model--
The sediment sub-model, SED, uses the Universal Soil Loss Equation with
cover factors from the new list composed by Wischmeier (1978) and an energy
factor computed by either the Onstad and Foster (1975), Williams and Berndt
(1977), or the Wischmeier and Smith (1965) method. The sediment models are
described in Appendix C. The method to compute peak runoff for determina-
tion of energy factors is described in Appendix A.
Pesticide Sub-model
The pesticide sub-model consists of three parts: degradation and
volatilization, downward displacement of pesticide bands, and pesticide in
overland flow.
The attenuation of pesticide through volatilization and degradation is
based on a stepwise first order degradation rate, as proposed by Donigian
et al_. (1977) ,
PESTt = PEST • exp [-At • ^_] (B.34)
h
where PEST = quantity of pesticide in the soil at time t, g/ha
At = number of days elapsed since the soil pesticide content
was updated
PEST . = pesticide concentration in soil on day (t-At), i.e. the
day that pesticide concentration was last updated
t^, = half-life for pesticide degradation, days. Two different values
2 for tj, were used: one for the time between application
and first runoff event after application, the other during
the remainder of the year
The downward displacement of pesticides is modeled with a simple model
proposed by, among others, Gardner (1965) and Rao et_ al. (1976). Diffusion
is not considered. Instead, the average displacement of the midpoint of the
pesticide band is computed as:
FLOW
FC • RETARD (B.35)
p • K
RETARD = 1 + -Ljg CB>36)
350
-------
where DISP = displacement of the band, cm
FLOW = quantity of flow past the pesticide band, cm
RETARD = retardation factor (dimensionless)
p = specific weight of soil, g/cm
K = adsorption partition coefficient, cm /g
Pesticide in Overland Flow
The following model used to determine the loss of pesticides in runoff
was developed by Steenhuis and Walter (1978). The key assumption is that the
concentrations of pesticides in the percolating water, runoff, and soil water
are equal in the thin layer of soil, extending 0.9 cm down from the soil sur-
face, designated as the mixing zone.
For a rainstorm of high intensity, the rainfall period is divided into
three segments by the program. In the initial segment, the rainfall intensity
is lower than the soils infiltration capacity, and all rainwater infiltrates.
Pesticides are transported from the mixing zone to the lower zones by per-
colating water. The quantity of pesticide in the mixing zone at the end of
segment is
PEST1 - PESTO - [exp . 1 (B.37)
where PEST1 = amount of pesticide in the mixing layer at the start
of ponding, g • ha"1
PESTO = amount of pesticide in mixing layer just prior to rainfall,
g • ha-1
WIBP = water infiltrated before ponding, cm
HMIX = mixing zone depth, cm
3, 3
SAT = moisture content at saturation, cm /cm
WIBP is obtained by
WIBP = 0.2 • C^T^- - 10) - STmax (B.38)
If the value for WIBP thus calculated is less than the actual rainfall, then
the latter value is used.
In the second and third rain segments, rainfall intensity is greater than
infiltration capacity. Thus, not all rainwater can infiltrate. Before runoff
351
-------
can occur, however, the surface detention storages have to fill up. The
second segment lasts from the first ponding until runoff begins. Pesticides
are diluted in the ponded water as well as carried downward in the percolating
water. The quantity of pesticide in the mixing zone at time when runoff
begins is
WIBP
PEST2 = PEST1
STmax + HMIX • (SAT + p • K)
J
HMIX • (SAT + p • K)
P -WIBP
(B.39)
where PEST2 is the amount of pesticide in mixing zone when runoff began, g/ha
In the third segment, pesticides from the mixing zone are lost both in
runoff and in percolating water. The pesticide concentration in the mixing
zone at the time when runoff stops is
PEST3
LOSSW
LOSSS
= PEST2 •
R •
_
(1
P
w
1 +
[1 +
* pw
• K •
Pw '
(p • K • SED)-R
P WTPP I ^
r — VKXDr T ^- — Cpn
r*V"n f"n 1fi*t
P STmax + HMIX • (SAT + Pg • K) (B'W)
SED • (p • K-l)] • (PEST2 - PEST3)
w
• K • SED) • [(P - WIBP) • (1-SED) + R • SED • p . K)]
SED • LOSSW
K • SED (B.42)
(B.41)
where PESTS = the amount of pesticide in the mixing zone when runoff
stops, g/ha
LOSSW = loss of pesticide in solution in runoff water, g/ha
LOSSS = loss of pesticide adsorbed on sediment, g/ha
SED = sediment concentration in runoff of water, g/g
p = specific weight of water, g/cm
W
Equations (B.37) through (B.42) were based on a linear adsorption isotherm
To simulate a nonlinear adsorption isotherm, a stepwise changing adsorption
partition coefficient was introduced in the program.
Each range of pesticide quantity in the mixing zone is associated with a
particular adsorption partition coefficient. The choice of adsorption
partition coefficient for equations (B..37) through (B.42) depends on the pesti-
cide concentration in the mixing zone.
352
-------
Statistics
The statistical program which was developed for this project summarizes
the results produced by the modeling program. It provides average monthly
values of the variables, together with standard deviation and the maximum
and minimum monthly values recorded during the entire time period. Average
annual values also are provided along with the standard deviation and mini-
mum annual values.
A second program analyzes data produced by the runoff model. It calcu-
lates the total number of days per month on which rainfall, runoff and soil
losses of varying magnitudes were produced, and provides a table showing the
probability of occurrence of events of varying magnitude on a month-by-month
basis.
Both programs can easily be adapted to deal with any of the variables
contained within the modeling program.1
PARAMETER VALUES FOR VALIDATION
The parameter values used in the CPM model for validation on the P2
watershed, Watkinsville, Georgia, are given below.
Plot Characteristics
length flow path
slope
soil group
active root zone depth
crop
area
200 meter
1.75%
B
see Figure 7-4
corn
1.29 ha
Hydrology
saturated hydraulic
conductivity
suction at wetting front
0.5 cm/hour
22.5 cm
Developed by H. D. Murray-Rust. Research Assistant, Department of
Agricultural Engineering, Cornell University.
353
-------
surface storage depression 0.4 cm
interception by plants 0.2 cm
moisture content at
saturation
field capacity
wilting point
maximum potential
evapotranspiration
minimum potential
evapotranspiration
0.39 cm3/cm3
0.20 cm /cm
0.14 cm /cm
0.7 cm/day
0.2 cm/day
Soil Loss
Universal Soil Loss Equation Parameters
K 0.28
P 0.5
LS 0.44
C Factor
Year 1973 1974
Date C Date
*
4-18-73 0.40 4-23-74
5-11-73 0.62 5-29-74
6-09-73 0.54 8-11-74
7-30-73 0.42 9-14-74
8-30-73 0.24
11-02-73 0.33
Pesticide Loss
1975
C Date C
0.20 5-11-75 0.15
0.18 6-19-75 0.13
0.14 8-09-75 0.11
0.28 8-29-75 0.10
10-04-75 0.19
Application dates and rates see Table 7-1
Atrazine
half-life after application 6 days
half-life after first runoff following application 16 days
354
-------
Adsorption Partition Coefficient
cm5/g
1.5
5
20
30
100
1000
10000
100000
Paraquat
half-life
360 days
Adsorption Partition Coefficient
cm /g
100
10000
100000
Atrazine Quantity in Mixing
Zone g/ha
above 400
150-400
91-150
78-91
50-78
20-50
10-20
below 10
Paraquat Quantity in Mixing
Zone g/ha
above 700
10-700
below 10
ADJAIRTEMP
AIRTEMP
CN
DISP
DEPIMP
DEPTH(J)
Ep
Epbase
SYMBOLS USED
adjusted air temp, °C
actual air temp, °C
curve number
displacement of pesticide band, cm
depth of the impermeable layer, cm
width of layer J
potential evaporation, cm
minimum yearly potential evaporation, cm/d
355
-------
Epmax
ET
f
FC
FLOW
FLOWIJ
HMIX
hw
I
i* .
end
INTCEP
ST
po
SW(J)
t
per
po
\
WATEREQ
WIBP
WP
z
maximum yearly potential evaporation
evapotranspiration, cm
fillable pore space
field capacity, cm
quantity of flow past the pesticide band, cm
water flow from Zone I to Zone J, cm
mixing zone depth, cm
suction head at -wetting front, cm
cumulative infiltration, cm
infiltration rate into the soil, cm/hr
infiltration rate in beginning period with
constant intensity, cm/hr
infiltration rate at end of period if all water
would have been infiltrated, cm/hr
interception of rain water by plants, cm
surface storage at time of ponding, cm
soil water content of Zone J, cm
time, days
duration of rain period of equal intensity, hr
time when water starts ponding, hr
half-life, day
water equivalent of snow, cm
water infiltrated before ponding, cm
moisture content at wilting point, cm
depth wetting front, cm
hypothetical depth of wetting front if all water
would have been infiltrated, cm
356
-------
end
7*
zend
z
po
p
-------
RETARD retardation factor, dimensionless
3, 3
SAT saturated moisture content, cm /cm
SED sediment concentration in runoff water, g/g
SOILTEMP soil temperature, °C
ST water stored in depressions, cm
ST, depression storage at beginning of period, cm
STend depression storage at end of period with constant
rain intensity, cm
STmax maximum depression storage, cm
APPENDIX B REFERENCES
Bouwer, H. 1976. Infiltration Into Increasingly Permeable Soils. American
Society of Civil Engineers Irrigation and Drainage Division Journal 102(1):
127-136
Donigian, A.S., D.C. Beyerlein, H.H. Davis and N.H. Crawford. 1977.
Agricultural Runoff Management Model Version II: Refinement and Testing.
EPA-COO/3-77-098, U.S. Environmental Protection Agency, Athens, Georgia.
England, C.G. 1970. Land Capability: A Hydrologic Response Unit in
Agricultural Watersheds. ARS 41-172. Agricultural Research Service, U.S.
Department of Agriculture, Washington, D.C.
Gardner, W.R. 1965. Movement of Nitrogen in Soil.Jn: W.V. Bartholomew and
F.E. Clark, eds. Soil Nitrogen. Agronomy, Volume 10. American Society of
Agronomy, Madison, Wisconsin.
Green, W.H. and G.A. Ampt. 1911. Studies on Soil Physics I: The Flow
of Air and Water Through Soils. Journal of Agricultural Science 4(1):
1-24.
Holtan, H.N., G.J. Stiltner, W.H. Henson and N.C. Lopez. 1975. USDAHL-74
Revised Model of Watershed Hydrology. USDA-ARS Technical Bulletin No. 1518.
Agricultural Research Service, U.S. Department of Agriculture. Washington,
D.C.
Li, E.A., V.O. Shanholtz, D.N. Contractor and J.C. Carr. 1976. Generating
Precipitation Excess Based on Readily Determinable Soil Vegetative Character-
istics. ASAE Paper No. 76-2005. American Society of Agricultural Engineers.
St. Joseph, Michigan.
358
-------
Mein, R.G. and C.L. Larson. 1971. Modeling the Infiltration Component
of Rainfall-Runoff Process. Water Resources Research Center, Bulletin 43.
University of Minnesota Graduate School. St. Paul, Minnesota.
Onstad, C.A. and G.R. Foster. 1975. Erosion Modelling on a Watershed.
Transactions of the American Society of of Agricultural Engineers 18(2):
288-292.
Rao, P.S.C., J.M.Davidson and L.C. Hammond. 1976. Estimation of Non-
Reactive and Reactive Solute Front Locations in Soils. In: W.H. Fuller
ed. Residual Management by Land Disposal. Proceedings of the Hazardous
Waste Research Symposium. EPA 600/9-76-015. Solid and Hazardous Waste
Research Division, U.S. Environmental Protection Agency, Washington,
D.C. pp. 235-242.
Soil Conservation Service. 1972. SCS National Engineering Handbook.
Section 4: Hydrology. Part 1: Watershed Planning. Soil Conservation
Service. U.S. Department of Agriculture. Washington, D.C.
Steenhuis, T.S. and M.F. Walter. 1978. Closed Form Solution for Pesticide
Loss in Runoff Water. ASAE Technical Paper No. 78-2031. American Society
of Agricultural Engineers, St. Joseph, Michigan
U.S. Army Corps of Engineers. 1960. Runoff From Snowmelt. Manual 1110-2-
1406. Washington, D.C.
Williams, J.R. and H.D. Berndt. 1977. Sediment Yield Prediction Based on
Watershed Hydrology. Transactions of the ASAE 20: 1100-1104. American
Society of Agricultural Engineers. St. Joseph, Michigan.
Wischmeier, W.H. and D.D. Smith. 1965. Predicting Rainfall Erosion Losses
From Cropland East of the Rocky Mountains. Handbook 282. Agricultural
Research Service, U.S. Department of Agriculture. Washington, D.C.
Wischmeier, W.H. and D.D. Smith. 1978. Predicting Rainfall Erosion Losses,
A Guide to Conservation Planning. Handbook No. 537. U.S. Department of
Agriculture. Washington, D.C.
359
-------
APPENDIX C
SOIL LOSS PREDICTION MODELS
The primary function of SWCPs in humid regions is control of soil ero-
sion. Techniques for predicting erosion have been primarily empirical in
the past and produce the single output of quantity of erosion in units of
tons/ac or MT/ha. The character of the eroded soil, its original location,
and the distance it moved were not typically of concern. One reason for
this approach was that conservationists were primarily interested in the
soil resource remaining after erosion took place.
As the interest in sediment-related problems increased, attempts were
made to predict sediment yield. The thrust of these attempts has been to
piggy-back some additional functional relationships to account for distance
of transport onto the established erosion prediction techniques. Obviously,
the sediment predicting techniques derived this way have all the limita-
tions inherent to the erosion models.
Perhaps the most serious limitation in the sediment models is that the
character of the eroded soil is important but not included in them. Sedi-
ment character is important first, because its transport properties are
dependent on the sediment particle size and density distributions; second,
because the pollution problems caused by sediment differ depending on the
sediment character; and third, because the amount of a substance adsorbed
on the sediment depends on the adsorptive character of the particles it
consists of as well as their original location in the field.
Since no model was available to deterministically estimate sediment
yield and its characteristics, the approach taken in this project was to
use the models that were available (empirical or partially empirical
models) and to supplement the quantitative values they gave with quali-
tative information concerning sediment character. Sediment character de-
pends on type of erosion, severity of erosion, type of storm, control
mechanisms, etc.
For the transport of strongly adsorbed substances, two considerations
of the sediment transport have to be made:
(a) the amount of sediment that is transported, and
(b) the preferential transport of fine clay and organic material
as compared to coarser silt and sand particles.
360
-------
Soil loss models are categorized as event-based or as average loss over some
time period. Prediction of soil loss on an event basis is important if
concentration of sediment or sediment-carried nutrient of pesticide is
critical as a pollutant. Event-based models may also be used to predict
long-term soil losses.
Average Annual Soil Loss Models
Early methods to predict soil erosion were developed for the purpose of
soil conservation (i.e., to preserve long term soil productivity). A very
extensive data base was collected to develop the most widely used equation
for estimation of erosion from cropland, commonly known as the Universal
Soil Loss Equation (USLE). This equation is described in the popular publi-
cation (currently being revised), Agricultural Handbook No. 282 (Wischmeier
and Smith, 1965) .
Erosion-index distribution curves have been developed by Wischmeier and
Smith (1965) . They give the percent of annual erosion index for each month
of the year. These values can be used directly to determine relative
magnitude of the rainfall factor for various times of the year.
Event-Based Models
At the present time, various models are in use to estimate edge of field
losses of sediment. Some models are still in conceptual form, others are
incomplete, and/or require data not routinely available (Woolhiser and
Todorovic, 1974; Meyer and Wischmeier, 1969). Three models for predicting
event-based soil losses were used in this study. They include the USLE
(Wischmeier and Smith, 1965), Onstad-Foster Model (Onstad and Foster, 1975),
and the Williams Model (Williams, 1975). The primary difference between the
three techniques is the way in which the s±o.rmevent energy term is calculated.
The USLE equation is used on an event basis by replacing the annual rain-
fall erosion index factor (R) by the storm erosion index. The storm erosion
index is the product of the rainfall energy and the maximum 30-minute
rainfall intensity. The rainfall energy term is given in equation 7 of
Appendix A.
Williams (1975) concluded that sediment loss was better correlated with
runoff than with rainfall. He, therefore, tried a number of combinations of
peak flow rate and total runoff for the rainfall factor in USLE. The runoff
parameter which gave the best fit between predicted and measured sediment
yield on 18 watersheds was:
W- 20.91A0'12 (Qxqp)0'56 CD
where Q and qp are total and peak runoff in units of cm and cm/hr, respect-
ively, and A is watershed area in hectares. When the right side of equation
1 is substituted into the USLE, the predicted sediment is in metric tons.
361
-------
Onstad and Foster (1975) presented an erosion-deposition model based on
a modified form of the USLE incorporating hydrologic variables of runoff and
rainfall in the energy term. This model is discussed in Appendix A.
Soil losses for the 72 storm events which occurred in the years 1973 to
1975 on Plot P2 in Watkinsville, Georgia, were predicted with the event-based
USLE, the Onstad-Foster model and.the Williams model using observed runoff and
peak runoff values. Watershed parameters are given in Appendix A. The
observed and predicted values (ordered by order of magnitude of the observed
values) are given in Table Ol. Rainfall records are available or can be
extrapolated to almost any watershed. Runoff records, however, are not
available for most watersheds so total and peak runoff generally need to be
calculated. Techniques used to predict total and peak runoff are given in
Appendix A.
Runoff
Predicted values for total and peak runoff for single storm events were
not as close to observed values as hoped for. Figures C-l and C-2 are
regressions of observed versus predicted total and peak runoff for 72 events
on the P2 watershed in Watkinsville, Georgia. The regression coefficients
(R2) for total and peak runoff were 55 and 50 percent, respectively. For the
72 storms which were evaluated a total of 43.8 cm of runoff was observed while
38.5 was predicted.
If the calculated values of total or peak runoff for a storm were zero,
the energy term of the Williams model was also zero, that of the USLE was
unaffected and the Onstad-Foster energy term was half that of the USLE.
Soil Loss
Since total and peak flow were known for the test watershed the soil
losses predicted by the Onstad-Foster and Williams models were calculated
directly from the field data. In general, runoff data are not available,
so total and peak runoff were also calculated for comparative purposes when
used in these two models.
As shown in Table C-l all three models overpredicted total soil loss for
the 72 storm events when actual runoff and rainfall data were used. The
event-based USLE, Onstad-Foster model and Williams model overpredicted total
soil losses by 133, 112, and 84 percent, respectively. Based on total soil
losses for all storm events the Williams model would appear to give the best
predictions. However, if only the largest storms are considered,the reverse
is true. For the ten largest recorded soil loss events the event-based USLE
gave better total soil results than the Onstad-Foster model or the Williams
model. Even for the ten largest events, all models still overpredict soil
losses. For very small storms when little or no runoff was observed the
Williams model, as expected, gave the best results.
362
-------
TABLE C-l. SOIL LOSS AS OBSERVED AND PREDICTED WITH 3 DIFFERENT MODELS FOR WATERSHED P2 IN WATKINS-
VILLE, GEORGIA (1973-1975). (LOSSES IN KG/HA)
OJ
W
Date
1/8/75
9/17/75
9/12/75
6/8/73
8/27/75
5/3/75
9/6/75
3/31/75
9/30/73 *
11/21/73
11/28/73
2/17/75
12/15/73
3/12/75
12/30/73
3/24/75
1/20/74
2/6/74
2/14/74
2/15/74
2/22/74
3/21/74
12/19/74
12/15/74
3/14/75
1/12/75
5/12/74
1/24/75
8/7/74
Observed
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
USLE
20
290
40
110
280
400
20
170
90
230
50
240
90
210
110
440
80
330
40
80
50
230
90
220
160
150
90
250
120
Actual Q and
Onstad-Foster
10
150
20
50
140
200
10
80
40
110
20
130
40
100
60
220
40
190
20
50
30
110
50
120
80
80
40
140
60
qp
Williams
0
60
0
0
0
0
0
0
0
0
0
30
0
0
10
0
0
70
0
10
20
0
80
10
0
20
0
20
0
Calculated Q
Onstad-Foster
10
170
20
60
140
250
10
100
40
110
20
210
70
170
100
230
40
450
20
40
20
110
80
240
200
140
40
230
60
and qp
Williams
0
120
0
70
0
280
0
80
0
0
0
300
90
210
100
100
0
420
0
0
0
0
80
260
280
170
0
280
0
-------
TABLE C-l. SOIL LOSS AS OBSERVED AND PREDICTED WITH
VILLE, GEORGIA (1973-1975). CONTINUED.
3 DIFFERENT MODELS FOR WATERSHED P2 IN WATKINS-
U)
Date
7/26/74
12/29/74
11/20/74
1/19/75
12/24/74
3/29/74
12/21/73
3/16/75
12/7/73
2/4/75
7/30/73
8/29/74
2/24/75
1/10/75
9/23/75
2/18/75
8/17/74
4/4/74
5/5/74
4/13/74
8/10/74
4/2/75
9/8/73
7/24/74
5/16/75
9/13/73
5/14/75
12/31/73
3/13/75
Observed
0
0
0
0
0
1
1
1
2
2
2
3
3
3
5
6
7
9
10
14
22
22
23
23
30
33
40
54
58
USLE
80
140
180
70
90
340
190
80
600
260
950
80
360
370
70
170
90
470
110
670
490
1810
1790
140
120
100
250
970
2430
Actual Q and
Onstad-Foster
40
80
90
30
40
180
40
50
310
140
500
50
190
190
40
220
50
310
60
420
260
1480
960
80
110
240
140
770
2460
qp
Williams
0
10
0
0
0
30
20
20
30
30
100
40
40
40
20
380
40
260
20
270
90
1250
230
80
70
630
110
630
1650
Calculated
Onstad-Foster
40
130
90
50
80
170
150
60
390
310
470
40
320
300
30
140
50
270
50
340
240
1900
950
70
150
50
120
1550
2910
Q and qp
Williams
0
160
0
70
50
0
210
70
310
390
0
0
370
350
10
190
0
180
0
60
0
1640
370
0
310
0
0
1420
2140
-------
TABLE C-l. SOIL LOSS AS OBSERVED AND PREDICTED WITH
VILLE, GEORGIA (1973-1975). CONTINUED.
3 DIFFERENT MODELS FOR WATERSHED P2 IN WATKINS-
Date
8/16/74
5/23/74
5/7/75
5/31/75
7/13/75
7/8/73
7/24/75
6/13/73
7/27/74
w 5/23/73
g 6/27/74
6/6/73
6/11/75
5/28/73
TOTAL
TOTAL (10
largest
observed
events)
Observed
70
92
110
215
320
422
427
530
661
716
966
1108
4122
8240
18373
17512
USLE
830
1170
750
400
460
2990
380
470
4340
350
3220
2180
1440
5590
42800
21420
Actual Q and
Onstad -Foster
530
710
450
260
370
2720
590
680
3380
560
2780
3020
1620
9010
38600
25730
qp
Williams
330
350
290
210
520
2770
870
1540
2180
1300
2240
4580
1700
8420
33720
30800
Calculated Q
Ons tad-Foster
450
1020
510
240
230
1540
200
240
2410
180
1880
1090
1080
6700
32410
17620
and qp
Williams
200
1060
560
240
0
400
80
0
940
0
1030
0
960
7780
24390
11190
-------
e
o
„
£J 8.0
o
i2
u.
"•*- 6.0
O
(r
o 4.0
LU
t
r~
O
© = I OBSERVATION
2-9= NUMBER of OBSERVATIONS
+ = 10 or MORE OBSERVATIONS O
KZ = 55%
_ PREDICTED^ .19 + .76 OBSERVED
O
00
- o
0 2
LU 2.0o- o
o:
QL
O
4OO O
230 0 0
O+9+40 1 1 1 1
2.0 4.0 6.0 8.0
OBSERVED RUNOFF, cm
10.0
FIGURE C-l. PREDICTED RUNOFF (SCS) VS. OBSERVED RUNOFF.
o 320
U."
fe 240
•z.
cr
160
a
o
cr
LU
80
© = 1 OBSERVATION
2-9= NUMBER OF OBSERVATIONS ®
r2= 50%
_ PREDICTED: 33.87+ .54 OBSERVED
O
O o
o ©
o oo
© o
I o oo ©
62
O 04?-3-©l
0 80 160 240 320 400
OBSERVED PEAK RUNOFF, f/sec
FIGURE C-2. PREDICTED PEAK RUNOFF VS. OBSERVED PEAK RUNOFF.
366
-------
The three models rarely underpredicted soil loss and when they did it was
only for large soil loss events (i.e., greater than 215 kg/ha soil loss). The
event-based USLE underpredicted soil losses on five storm events, while soil
loss was underpredicted only twice by each of the other two models. The
largest difference between observed and predicted results for soil loss seem
to be for a few events, when the models grossly over-estimate soil losses.
For example, on 8 July 1973, 422 kg/ha of sediment was recorded leaving plot
P2 while all three models were in close agreement in estimating soil losses
at nearly 3000 kg/ha.
For the 25-year model simulation referred to in Section 5 only rainfall
data were available so total and peak runoff had to be calculated. Soil loss
predictions for the Onstad-Foster model and the Williams model using calcu-
lated total and peak flow are also given in Table C-l for the storms of 1973
to 1975. Regression analysis showed that the correlation between observed and
predicted soil loss was not much changed when actual instead of predicted
runoff data were used (r2 = 71 and 72% for Williams' model and 71 and 68% for
Onstad-Foster model when observed and predicted runoff data were used
respectively). Total and particularly peak runoffs were underpredicted for
large storm events (see Figures C-l and C~2). This probably compensated
for the occasional large over estimation of soil losses when actual runoff
data were used. This point is also illustrated by the soil loss predicted
with the Williams model for the ten largest observed soil-loss events. The
Williams model very much under estimated soil losses for these selected large
events due to the fact that no runoff was predicted.
Summary and Conclusions
The Williams model is a function only of runoff, the event-based USLE
depends only on rainfall, and the Onstad-Foster relies on both. Runoff data
are typically not available so they must be calculated from rainfall data and
watershed physical characteristics. All three models over-estimated total
soil losses for a three-year period, with the Williams model predicting
closest to observed values. The Williams model worked very well for small
events but was the poorest of the three models for the ten events with
highest observed soil losses. No model was found to be clearly more accurate
than the others- in all cases. Furthermore, since the comparison was made on
only one watershed for a three-year period, extrapolation to other watersheds
needs to be handled with care. Theoretically, inclusion of runoff parameters
in a sediment estimating model seems to be appropriate and advisable in
future model development.
REFERENCES FOR APPENDIX C
Meyer, L.D. and W.H. Wischmeier. 1969. Mathematical Simulation of the
Process of Soil Erosion by Water. Transactions of the American Socity of
Agricultural Engineers 12(6): 754-756-762.
Onstad, C.A. and G.R. Foster. 1975. Erosion Modelling on a Watershed.
Transactions of the American Socity of Agricultural Engineers 18(2):288-292.
367
-------
Williams, J.R. 1975. Sediment Yield Prediction With Universal Equation
Using Runoff Energy Factor. Present and Prospective Technology for Pre-
dicting Sediment Yields and Sources. ARS-S-40. U.S. Department of Agri-
culture. Agricultural Research Service. Washington, D.C. pp. 244-252.
Wishchmeier, W.H. and D.D. Smith. 1965. Predicting Rainfall Erosion Losses
From Cropland East of the Rocky Mountains. Handbook 282. Agricultural
Research Service, U.S. Department of Agriculture. Washington, D.C.
Woolhiser, D.A. and P. Todorovic. 1974. A Stochastic Model of Sediment
Yield for Emphemeral Streams. Proceedings of the Symposium on Statistical
Hydrology. Miscellaneous Publication No. 1275. U.S. Department of Agri-
culture. Washington, D.C.
368
-------
APPENDIX D
CONTROL OF NITROGEN LOSSES BY MANAGEMENT OF FERTILIZER APPLICATIONS
While SWCPs must sometimes be analyzed separately in order to quantify
their effectiveness in reducing potential water pollution, they are in
reality used in concert with a total farming system. Total nutrient loss is
best controlled with a combination of control of soil loss and good manage-
ment of fertilizer (McDowell and Grissinger, 1977).
Control of agricultural chemicals is much more complex once they are
applied to the land than before. Therefore, careful management of chemical
applications is advisable. If fertilizer is applied close to the time of crop
uptake and does not exceed the crop requirement, potential nutrient losses
will be minimized. The most effective application time and technique may not
be the least expensive. Usually, the time and method of fertilization depend
on seasonal fertilizer cost, availability of application equipment and of
farm labor (Whitaker et_ al_., 1978). For example, fall application of nitro-
gen fertilizer is sometimes used to spread the work load throughout the year.
Some farmers prefer to apply more chemical than recommended as a form of
"insurance" that enough will be available.
The opportunities available for source management of applied nitrogen
include application rate and techniques, timing, choice of field and form
of nitrogen fertilizer. When manurial nitrogen is used one can choose appli-
cation rates and field but in many situations the timing and technique of
manure disposal is dictated by the manure handling system.
The obvious source management technique to control nitrogen losses is
reduction in rates of application. Chichester (1977) investigated nitrogen
losses from two rates of applications on corn in Coshocton, Ohio. His results
for inorganic nitrogen losses in percolation and surface runoff are shown in
Table D-l. The lysimeters contained a moderately well-drained Keene silt
loam soil on a six percent slope. Inorganic nitrogen losses in surface run-
off in this experiment were relatively small while percolate losses were very
large. Nitrogen application rates used by Chichester were apparently so high
that nitrogen built up in the soil profile. This is indicated by the compar-
atively high nitrogen losses even in 1973 when no nitrogen was applied.
These results illustrate that excessive nitrogen application rates can result
in large nitrogen losses to percolation.
369
-------
TABLE D-l. INORGANIC NITROGEN LOSSES FROM TWO SYSTEMS OF AMMONIUM APPLICA-
TION TO CORN (ADAPTED FROM Chichester, 19771
Lysimeter
Application
Date Rate
(kg/ha)
Period Inorganic N
of Percolate (kg/ha) Runoff (kg/ha)
Record
A
B
Apr 1971
May 1972
-
May 1974
TOTALS
Apr 1971
May 1972
-
May 1974
TOTALS
336
336
0
178
850
672
336
0
178
1186
1971-72
1972-73
1973-74
1974-75
1971-72
1972-73
1973-74
1974-75
179
244
133
84
640
202
274
165
115
756
1
6
3
7
17
3
10
6
11
28
Whitaker et^ aj^. (1978) reported that for shallow Missouri soils, nitrogen
in runoff increased as rate of application was increased from 15 to 368 kg/ha.
Results from this study are shown in Table D-2. This study was conducted on
plots 3.2 x 27.4 meters on a three percent slope. The soil was underlaid by
a claypan 18 to 45 cm below the soil surface, which restricted water movement.
Metal barriers were placed around the plots down to the claypan to prevent
subsurface flow out of the plots. All flow off the plots was collected at the
bottom of each plot. Losses of nitrogen reported by Whitaker et al. (1978)
were much lower than those reported by Chichester (1977) even at" tFe highest
application rate. The difference could be accounted for by greater denitrifi-
cation or by losses through the claypan in the investigation by Whitaker et^ al.
Maximum yields in the study by Whitaker et_ al. were achieved at nitrogen
application rates of 198 kg/ha. Corn yields at application rates higher than
187 kg/ha did not increase but soluble nitrogen losses in the runoff did,
indicating that fertilization in excess of crop requirements can result in
increased nitrogen losses of a field.
The CNS model was used to evaluate nitrogen losses for continuous corn
crop systems in three locations - New York, Georgia and Iowa - over 25 years.
The physical features of the particular fields modeled are given in Table 6-7.
The "base" management scheme was an application of 75 percent of the total
nitrogen requirements of the corn crop as given in Table 6-7 applied in the
form of nitrate at plant emergence. Mineralization of soil organic nitrogen
was assumed to be 15 to 45 percent of the crop nitrogen needs. The base
management scheme is not necessarily the recommended management system for any
of the locations. Crop yields were assumed to remain constant for all 25 years
modeled.
370
-------
TABLE D-2. SOLUBLE NITROGEN LOSSES FROM CORN PLOTS (Adapted from Whitaker
Application Rate~Losses in Runoffa
(kg/ha) (kg/ha)
_ . . __
98 . 14.7
198 14>9
214 17.0
252 39.6
368 35.!
aThree-year average
Table D-3 shows nitrogen losses for different application rates as com-
pared to the base condition. Inorganic nitrogen losses for the base condition
in June were 36, 31, and 29 kg/ha for New York, Georgia and Iowa, respectively.
If the fertilizer rates were twice those of the base condition, the expected
losses would increase 86 to 93 percent. On the other hand, for the New York
case at least, if the base rate of fertilizer was halved, the expected losses
would be reduced from 36 to 26 kg/ha.
A second technique for good management of nitrogen fertilizer is applica-
tion at the time the plant most needs it. Ideally this would require a split
application. Whitaker et^ al_. (1978) recommend a split application with a small
amount at planting time and the rest 5 or 6 weeks later. The base conditions
of Table D-3 do not include this ideal approach, but timewise it is the best
of the alternatives given in the table. Fall applications typically cause
losses 100 to 148 percent greater than the base condition. The values in
Table D-3 were based on nitrate fertilizers so that the nitogen was available
to leach out of the soil immediately after application. The losses of
inorganic nitrogen in preplant applications are less than those for fall but
are still substantial. The model indicated that fall applications of nitrate
could result in annual nitrogen losses which exceeded the application rate when
the base rate was used. In Iowa this practice resulted in nitrogen losses of
about 70 percent of the amount.applied.
The expected losses of nitrogen increase relatively more in Iowa than in
New York or Georgia when twice the base rates are fall applied. However, a
lower percent of the applied nitrogen is lost in Iowa than in the other two
locations for the fall base application.
As shown in Chapter 6, surface runoff with conventional tillage was
typically 20 to 25 percent of the total of surface runoff and percolation.
Dissolved nitrogen is lost in surface runoff and leachate in about the same
proportions as the water movement. Total inorganic nitrogen losses were
greater in the leachate than in the surface runoff, but yearly variations
over the 25-year simulated period were proportionately much higher in the
surface runoff. For example, nitrogen applied in April in New York at the base
371
-------
rate resulted in a mean and standard deviation for leachate losses of 43 and
4 kg/ha while for the surface runoff the mean and standard deviation were 7
and 3 kg/ha, respectively.
TABLE D-3. NITROGEN LOSSES FOR SELECTED APPLICATION RATES AND TIMES WITH
CNS MODEL, (kg/ha)
Application
Date
Application Rate (kg/ha)
New York
142
122
97
87
67
71
73
57
51
36
(Base) 36
39
34
26
Georgia
Sept
Mar
Apr
June(Emergence) 67
Application Rate (kg/ha)
116 58 (Base)
Sept
Mar
Apr
104
76
67
May(Emergence) 58
62
47
39
31
Application Rate (kg/ha)
202 101 (Base)
Sept
Mar
Apr
126
95
70
May(Emergence) 56
72
52
41
29
Table D-4 illustrates the expected time of nitrogen losses at the three
locations when nitrogen is applied at the base rate. The 25-year simulation
period indicated that when nitrate was applied at plant emergence, losses in
New York would be fairly uniformly distributed in all seasons, while in Iowa
and Georgia, losses could be expected to be greatest in the spring and
smallest in the fall. When nitrogen was applied in March, several months
before the.crop could us it, very high losses occurred in all three locations
immediately following application. September application of nitrate resulted
in high losses in fall and winter in New York and Georgia, while for this
situation, losses in Iowa were particularly high in spring.
Since nitrogen was assumed to be applied in nitrate form it was immedi-
ately available for crop uptake or transport with water. The temporal
372
-------
variations in nitrogen losses listed in Table D-4 are a result of seasonal
water movement, crop uptake, and application timing. If other nitrate trans-
formations, particularly denitrification, were included in the model the
distributions of losses might be different. Smid and Beauchamp (1976) have
TABLE D-4. AVERAGE SEASONAL NITROGEN LOSSES IN KG/HA WHEN NITRATE IS APPLIED
AT THE BASE RATE AS PREDICTED BY THE CNS MODEL FOR A 2-YEAR PERIOD
Application
0
Season
Winter
Spring
Summer
Fall
b
NY
11
9
8
8
May
Iowa
4
16
9
0
Ga
11
11
6
0
NY
8
38
5
3
March
Iowa
3
42
4
1
Time
Ga
11
32
1
1
NY
25
15
1
30
Sept
Iowa
15
34
3
20
Ga
25
7
0
38
Winter, Dec-Feb; Spring, Mar-May; Summer, June-Aug; Fall, Sept-Nov.
Application date June 1.
shown that at typical application rates all nitrate can denitrify in only one
or two days if conditions are favorable for denitrification. Conditions
favorable for denitrification are fairly common on some cropland. If this were
the case (e.g., denitrification was important) total nitrogen losses would be
greater but losses to runoff and leachate would be smaller.
Summary
The CNS model was used to simulate 25-years of nitrogen losses at three
locations for selected nitrate application rates and times. Results indica-
ted that for all locations, losses increased substantially as the time period
prior to planting was increased. Average expected nitrogen losses approach
or exceed the quantity applied for fall application. When the nitrate appli-
cation rate was double the quantity taken up by the crop, losses typically
increased by about 70 percent over those at the base rate.
REFERENCES APPENDIX D
Chichester, F.W. 1977. Effects of Increased .Fertilizer Rates on Nitrogen
Content of Runoff and Percolate from ,Mon61ith Lysimeters. Journal of
Environmental Quality 6(2): 211-217.
McDowell, L.L. and E.H. Grissinger. 1976. Erosion and Water Quality.
Proceedings of the 23rd National Watershed Congress. Biloxi, Mississippi.
pp. 40-56.
Whitaker, F.D., H.G. Heineroann and R.E. Russell. 1978. Fertilizing Corn
Adequately With Less Nitrogen. Soil Conservation Society American Journal
pp. 28-32.
373
-------
Smid, A.E. and E.G. Beauchamp. 1976. Effects of Temperature and Organic
Matter on Denitrification in Soil. Canadian Journal of Soil Science
56:385-391.
374
-------
APPENDIX E
CALCULATION OF BUDGETS, SOIL EROSION
AND SEDIMENT DELIVERY FOR THE LINEAR PROGRAMMING MODEL
Budgets and Constraints
A separate Linear Programming (LP) model was used for each of the ex-
ample farms (New York, Iowa, and Texas). Each model is similar to the others.
Therefore, to avoid repetition, the New York example farm will be discussed,
followed by a discussion of the similarities and differences of the Iowa
and Texas farms.
The New York example farm is a dairy farm, where the primary source of
income is milk sales. The maximum number of cows is 80, with an average
replacement rate of 25%. To feed the dairy herd and heifers, three least
cost feed rations using 1978 feed prices were used (Knoblauch et al. 19781.
They are: (1) hay crop base, (2) one-half corn silage base and (3) corn
silage base. Both forages (hay and silage) must be grown on the farm. Corn
grain may be grown or bought as needed in the ration, and soybean meal can
only be bought. These three alternatives allow the least cost ration to be
selected from an almost total row crop operation (corn silage base) down
to a total sod crop operation (hay crop base) if necessary. The Iowa example
farm is a cash crop/hog feeding operation with a nutrient requirement ration.
The Texas farm is cash crop and hay.
The amount of land in New York is limited to 70 hectares which must
grow all the corn silage and hay used by the dairy herd. Of this, one-half
is moderately well drained, highly productive soil and has an 8% slope
(Honeoye-Lima). The remaining one-half is somewhat-poorly drained soil with
a 4% slope and has lower yields (Lima-Kendaia). The Iowa farm is 100 hectares
of Muscatine-Tama-Garwin Association which consists of the following soils:
40% Tama, 30% Muscatine, 15% Dinsdale, and 15% Garwin. The Texas farm is
100 hectares of Houston Black-Heiden soil Association, consisting of three
soils: 45% Houston Black Clay, 35% Heiden Clay, and 20% Trinity Clay.
Crop budgets for each soil type used in Iowa were provided by Jim
McGrann (Personal Communication, 1978) and budgets for the Texas Blackland
Region were provided by Cecil Parker (Personal Communication, 1978).
In New York, dairy farmers grow three basic feeds: corn grain, corn
silage, and hay. These crops can be grown using conservation tillage or
no-till, straight-row or contouring, and continuously or in rotation. Two
rotations are considered, four years corn and four years hay or one year
375
-------
TABLE E-l,
CROPPING COMBINATIONS
New York Cropping Combinations
Corn silage chisel
Corn silage no-till
Corn silage chisel 4x4 rotation**
Corn silage chisel 1x4 rotation**
Corn silage no-till 4x4 rotation
Corn silage no-till 1x4 rotation
Corn grain chisel
Corn grain no—till
Corn grain chisel 4x4 rotation
Corn grain chisel 1x4 rotation
Corn grain no-till 4x4 rotation
Corn grain no-till 1x4 rotation
Iowa Cropping Combinations
Corn grain
Corn grain, soybeans
Corn oats, meadow, meadow, meadow
Corn, corn, oats, meadow, meadow
Corn, corn, corn, oats, meadow
On contour
Diversion system # 1***
Diversion system # 2
Terrace system # 1***
Terrace system # 2
Terrace system ff 3
Terrace system # 4
On contour
Terrace system # 1
Terrace system # 2
Terrace system # 3
Terrace system # 4
Texas Cropping Combinations
Cotton
Wheat
Sorghum
Hay
Cotton, cotton, sorghum
Cotton, sorghum, sorghum
Cotton, wheat, wheat
Sorghum, wheat, wheat
Cotton, sorghum
Cotton, sorghum, wheat
On contour
Terrace system # 1
Terrace system # 2
Terrace system # 3
Terrace system # 4
* All possible combinations of the two columns were used.
** 4x4 Rotation = four year corn and four year hay
1x4 Rotation = one year corn and four year hay
***
Terrace and diversion systems are explained later in this Appendix.
376
-------
TABLE E-2.
NEW YORK BASE BUDGET-CONSERVATION TILLAGE
Machinery Cost
Tillage and planting
Cultivation
Harvest
Corn
Grain
$ 10.19
1.88
8.15
Corn
Silage
$ 10.19
1.88
16.68
Hay1age
$ 3.27
47.44
Fertilizer
Nitrogen
Phosphorus and potassium
37.80
21.10
37.80
21.10
23.50
Pesticide and Herbicide
31.85
31.85
18.50
Miscellaneous
38.74
34.72
37.07
Sub Total
$149.71
$154.22
$129.78
Labor (hr.)
Spring
Summer
Fall
(2.3) 8.60 (2.3) 8.60
(0.7) 2.60 (0.7) 2.60
(4.7) 17.60 (10.5) 39.40
( 0.9) 3.40
(I/.2) 64.50
( 0.2) 7.50
Total
Yield MT/ha^
$178.51
5.19
$204.82
36
$205.18
17.8
Cost per hectare includes only variable cost
Harvest costs vary according to yield
Fertilizer use varies according to soil response
Yields vary according to soil type and response to rotation
377
-------
corn and four years hay. The cropping practices shown in the first column
of Table E-l can be used in conjunction with the terrace or diversion
systems shown in the second column to make 84 possible combinations of SWCPs
for each soil type in New York. Likewise there are 25 for each soil type in
Iowa, and 50 for each soil type in Texas (Table E-l ). Some of these
practices, however, may not be feasible, for example, working on the contour
is only effective for shorter slope lengths.
For New York, budgets for planting, cultivating and harvesting operations
(using conservation tillage) were constructed (Table E-2). Machine time
efficiency, repair cost and fuel consumption information came from "Selecting
Field Machinery" Campbell (1973). The machinery complement which goes with
a nine foot chisel plow and four row planter were used. In New York, neither
conservation tillage nor no-till have a significant effect on yields for
the soils on the example farms. (Fred Swader, Personal Communication, 1978).
In Iowa this is true for conservation tillage, however, yield decreases of
around 5% are expected with no-till (Minora Ameniya, Personal Communication,
1978). In Texas, yield decreases are expected with both reduced tillage
practices, thus neither is widely used on the soils of the example farm.
Some no-till budgets were also used in New York because the yield
response is the same as for conservation tillage (Fred Swader, Personal
Communication, 1978). The no-till budgets show a decrease in machinery and
labor expense but an increase in herbicide and pesticide expense for a net
increase of $4.00 per hectare (Table E-3 ). In Iowa a reduction in yield
is expected with no-till (William Shrader, Personal Communication, 1978),
making it a poorly accepted practice. Thus it was not budgeted for Iowa.
TABLE E-3,
PARTIAL BUDGET - NEW YORK NO-TILL OPERATION
VS. CONSERVATION TILLAGE
Cost of Conservation
Tillage Cost of No-till
Spring Machine $10.19 $ 5.99
Spring Labor 8.60 4.55
Summer Machine 1.80 1.14
Summer Labor 2.60 1.09
Herbicide and Pesticides 31.85 46.35
$55.12 $59.12
Terracing, diversion ditches, and contouring require more turning time
378
-------
and an extra trip around the field to work the end rows and corners. Hence,
machine times were increased by 10% and labor increased by 12%.
Installation cost for terraces with tile drainage is $6.50 per meter
and diversions cost $2.50 per meter in New York (Lauren Johnson, Personal
Communication, 1978). Maintenance costs for terracing and diversions ranged
from 1.5% to 3.5% of installation cost with the higher percentage going to
structures with wider intervals. For both terraces and diversions, con-
struction cost are amortized over an expected life of 40 years at an 8%
interest rate. The annual cost per hectare for each terrace system was cal-
culated and added to the cost of growing and harvesting the crop using
conservation tillage. Because diversions take a percentage of land out of
production, that percentage of the budgeted cost are subtracted from the base
budget before adding the annual cost for diversions.
In Iowa the installation cost of tile drain terraces varied according
to slope of land being terraced. On Muscatine and Garwin soils, the cost
was $7.35 per meter, Tama soil cost was $7.85, and Dinsdale soil cost was
$8.40 (Jim McGrann, Personal Communication, 1978). In Texas the cost of
terraces was $0.46 per meter plus $25 per hectare for construction of the gras-
sed waterway used for drainage (Cliff Williams, Personal Communication, 1978).
New York crop yields remained the same on each soil type for all pro-
duction practices except where diversion were used. In this instance, the
amount of land used by the diversion was taken out of production and a poor
quality low yield hay was used instead. On all soil types in Iowa, corn grain
yield increased 5% when preceded by one, two or three years of meadow production.
In Texas grain sorghum production increased by 23% when following cotton in
the rotation, and wheat production decreased 5% after cotton or grain sorghum
production.
In all three areas, each soil type results in different yields for
each crop, giving some soil types a comparative advantage over other soil types
for producing a particular crop. However, a different crop may gain the com-
parative advantage on a given soil type due to changes in production practices
needed in order to meet soil loss standards. There are two reasons for this
changing advantage: The first is that production costs for a different crop
may be less then production cost of the original crop when meeting soil loss
standards. The second reason is that the original crop may have to be grown
in rotation with another crop in order to meet the soil loss standards.
For each SWCP used with each crop, soil and sediment losses per hectare
were calculated (shown later )• Using varying levels of soil erosion as the
constraint, the LP was allowed to pick the most profitable combination of
crops and production practices.
Soil Erosion Calculations
Soil erosion is calculated from the Universal Soil Loss Equation (USLE)
(Wischmeier, et al. 1965). This equation is given by:
A=2.25RKLSCP where
379
-------
A = Average Annual gross soil erosion in MT/ha.
2.25 = Metric conversion factor from tons/acre to metric tons
per hectare.
R = Rainfall erosivity factor
K = Soil credibility factor
L = Slope length factor
S = Slope gradient factor
C = Cropping practice factor
P = Conservation practice factor
R, K and S Factors
The values for R, K, and S are independent of SWCPs and are given in
Table E-4.
TABLE E-4.
R, K, and S FACTORS FOR THE EXAMPLE FARMS
State and Soil Type
New York
Honeoye-Lima
Lima-Kendaia
Iowa
Muscatine
Tama
Dinsdale
Garwin
Texas
Houston- Black
Heidon
Trinity
R
100*
100*
175***
175***
175***
175***
320****
320****
320****
K
.32*
.32*
.28***
.28***
.32***
.28***
.32****
22****
22****
S**
.840
.350
.117
.304
.701
.117
.182
.260
.089
* Swader (1974)
** S = (0.43 G2 + 0..3Q G + 0.43) /6.613
where G = slope gradient in percent (Wischmeier, et al. 1965).
*** Earnest Hintz, Personal Communication, 1978.
**** Cliff Williams, Personal Communication, 1978.
380
-------
L Factor
The L factor is a_jFunction of the slope length. On simple slopes
(uniform slope) L = • 2H?73 where X is the slope length in meters, or the
interval between structures where these have been implemented (Wischmeier,
et^ al^. 1965). On complex slopes, L is calculated separately for each
segment of the slope having a different gradient. Thus for the example
farms, when no terraces or diversions were installed, the slope length was
the distance from the top of the farm to the stream, and for each soil type
L was given by the formula:
where Xj = Slope length in meters from top of slope to bottom of soil
type j, and A. j= slope length from top of the slope to top of soil type
j. When terraces or diversions are implemented, the slope length is cal-
culated using the intervals, between structures for A. These intervals are
given in Table E-5. Note that Table E-5 lists all structures which were
budgeted and entered into the LP program. Many of.the structures did not
actually appear in any of the farm plans.
TABLE E-5.
INTERVALS BETWEEN STRUCTURES FOR THE EXAMPLE FARMS
State and Soil D * V2* T/ T2* T^* T * Tg*
New York
Honeoye-Lima 242 75.3 242** 119** 56.5**$ 122 58^
Lima-Kendaia 121 56.5 242** 119** 56.5**$ 122 58Q
Iowa
Muscatine —- -— 67^ 142 142 150
Tama --- --- 32$ 92 192 192
Dinsdale --- — 29.5$ 67 67 75
Garwin — — 69$ 67 75 75
Texas
Houston-Black
Heiden
Trinity — —
31}
24$
50$
49
39
50
104
90
50
104
90
44
* D represents a diversion ditch system; T represents a terrace system. The
number subscripts are keyed to specific intervals evaluated for the example
farms.
** Terrace next to stream
$ Meets SCS specifications (USDA-SCS, 1969).
381
-------
C Factor
For the New York and Iowa example farms, C factors were derived from
recently revised tables furnished by Walter Wischmeier (Personal Communi-
cation, 1978). For the Texas example farm, C factors are from a recent
study in that area on the Lavon Reservoir Watershed (Taylor et^ al. 1978).
The effectiveness of cropping practices for reducing erosion depends, among
other things on crop yields. Thus C factors may vary from soil to soil on
a farm depending on yield levels. This effect was taken into account for
both the New York and Iowa farms, but not in Texas due to insufficient data.
Table E-6 lists the C factors used for each rotation on the example farms.
P Factor
The P factor is 0.5 for all contouring, terraces, and diversions except
where slope gradients are less than 2%. In this case 0.6 is used. The
strip-cropping P factor is one half the contouring factor. Where neither
contouring nor structures are used, the P factor is 1.0.
Sediment Delivery
Sediment delivery is calculated as the product of the level of soil
erosion and the sediment delivery ratio. The method used for estimating the
sediment delivery ratio is fairly arbitrary and based on a loose interpre-
tation of data collected and analyzed by Renfro (1974). Thus it is not ex-
pected that the procedure used would necessarily give accurate estimates of
the SDR in actual field situations. However, the results of the analysis
are assumed to depict the relationship between and relative importance of
different areas with differing SDR's when control of sediment loss is desired.
Calculation of the SDR is based on the formula SDR = 2.5d~' where
d = the distance from the center of a field to a waterway. Waterways include
both natural bodies and man-made structures such as diversions and terrace
channels. For fields directly bordering a waterway, the lower 125 meters of
slope are assumed to have an SDR of 1.0. The distance formula is applied to
calculate the SDR for the area above the lower 125 m and the SDR for the
field is an average weighted according to the proportion of soil erosion
occurring in each area. For areas above tile-outlet terraces, the calculated
SDR is multiplied by 0.2 to account for redeposition in the terrace channel
(Walter Wischmeier, Personal Communication, 1978). A factor of 0.4 is used
for terraces with grassed-waterway outlets (Don Redell, Personal Communication,
1978).
The SDR values used on the example farms are given in Table E-7.
382
-------
TABLE E-6.
C FACTORS USED ON EXAMPLE FARMS
—— — — '- - — - .' • . • . , -.Ml—Ml.—. __ .-._._, _..
State and Soil Tillage CROP ROTATIONS
Type Method SSSS- GGGG- S- G-
S G HHHH HHHH HHHH HHHH
New York (All Spring Conservation Tillage]
Honeoye-Lima
Conservation .402 .127 .154 .069 .040 .040
No-till .381 .073 .142 .041 .038 .029
Lima-Kendaia
Conservation .422 .165 .166 .092 .047 .047
No-till .410 .095 .157 .050 .045 .032
Iowa (All Conservation tillage)
GO- GGO- GGG-
G(Sp) G(F) G-Sb(Sp) C-Sb(F) HHH HH OH
Muscatine .117 .124 .211 .219 .028 .047 .070
Tama .122 .129 .211 .219 .028 .048 .071
Dinsdale .126 .134 .228 .239 .029 .049 .074
Garwin .122 .129 .211 .219 .028 .048 .071
Texas (All Spring Conventional Tillage)
C Sg W H CSCg
All soils .60 .40 .20 .05 .53
CSgSg CWW SgWW C Sg C Sg W
All Soils .45 .30 .23 .49 .35
Sp = Spring tillage, F = Fall tillage
S Corn Silage
G
H
Sb
0
C
Sg
W
Corn grain
Hay
Soybeans
Oats
Cotton
Sorghum
Wheat
383
-------
TABLE E-7
VALUES FOR THE SEDIMENT DELIVERY RATIO
ON THE EXAMPLE FARMS
State and Soil
Type
New York
Honeoye-Lima
Lima-Kendaia
Iowa
Muscatine
Tama
Dins dale
Garwin
Texas
Houston-Black
He i den
Trinity
No D * D2*
Structures
.30 .78 1.0
.78 1.0 1.0
.29
.35
.72
1.0
.38
.75
i.o
V
.16
.16
.20
.20
.20
.20
.20
.20
1.0
V
.20
.20
.19
.20
.20
.20
.20
.20
1.0
V
.20
.20
.19
.17
1.0
1.0
.20
.81
1.0
v
.26
.60
.07
.16
1.0
1.0
.20
.15
.20
V
.23
.40
—
* D represents a diversion ditch system,
T represents a terrace system. The number
subscripts are keyed to specific intervals
evaluated for the example farms.
REFERENCES APPENDIX E
Campbell, J.K. 1973. Selecting Field Machinery. Agricultural Engineering
Extension Bulletin No. 395. Cornell University, Ithaca, New York.
Knoblauch, W.A., R.A. Milligan and M.L. Woodell. 1978. An Economic Analysis
of New York Dairy Farm Enterprises. A.E. Res. 78-1. Cornell University.
Ithaca, New York.
Renfro, G.W. 1975. Use of the Erosion Equations and Sediment Delivery Ratios
for Predicting Sediment Yield. Present and Prospective Technology for Pre-
dicting Sediment Yields and Sources. ARS-S-40. U.S. Department of Agricul-
ture, Agricultural Research Service, Washington, D.C.
Taylor, C.R., D.R. Reneau and B.L. Harris. 1978. An Economic Analysis of
Erosion and Sedimentation in Lavon Reservoir Watershed. Bulletin TR-88.
Texas Water Resources Institute. College Station, Texas.
Wischmeier, W.H. and D.D. Smith. 1965. Predicting Rainfall Erosion Losses
From Cropland East of the Rocky Mountains. Handbook 282. Agricultural
Research Service, U.S. Department of Agriculture. Washington, D.C.
384
-------
APPENDIX F
EFFECTS OF SOIL AND WATER CONSERVATION PRACTICES ON RUNOFF AND POLLUTANT
LOSS FROM SMALL AGRICULTURAL WATERSHEDS: A SIMULATION APPROACH
Prepared by Douglas C. Beyerlein and Anthony S. Donigian, Jr.
Hydrocomp Inc., 1502 Page Mill Road, Palo Alto, California 94304
The effect of soil and water conservation practices (SWCPs) on the trans-
port of pollutants is complex and not always obvious. In an attempt to
better understand how SWCPs affect runoff, sediment loss, and pesticide and
nutrient washoff from agricultural lands, we have conducted a study of SWCPs
using the Agricultural Runoff Management (ARM) Model (Donigian and Crawford
1976; Donigian et al. 1977). The ARM Model is a collection of the physical
processes that describes the hydrologic cycle, sediment fines generation and
transport, pesticide interaction in the soil and washoff, and nutrient trans-
formations and washoff. Thus, the ARM Model can provide the necessary scope
and abundance of information to evaluate the effectiveness of soil and
water conservation practices in controlling runoff, sediment loss, and the
movement of pesticides and nutrients from agricultural watersheds.
Soil and water conservation practices have an impact on the hydrologic
and physical characteristics of a watershed, in addition to the timing and
type of agronomic practices. Specific ARM Model parameters are related to
certain hydrologic and physical characteristics of an agricultural watershed,
while other parameters describe agronomic practices. As these characteristics
are changed for a specific SWCP, the model parameter values can be adjusted
to reflect the use of the SWCP. Output from the model for different SWCPs
is compared to the corresponding output for the conventional practice or
base condition. The total volume (or mass) and frequency of runoff, sedi-
ment loss, and pesticide and nutrient washoff are compared for the different
practices. The change in frequency or volume for a particular constituent
by a SWCP is used to evaluate the effectiveness of that SWCP to control that
constituent. This analysis was done for three SWCPs (no tillage, contouring,
and terracing with contours) on small agricultural watersheds in Watkinsville,
Georgia, and East Lansing, Michigan, using 10 years of meteorologic data as
input to the model. This work was performed for Cornell University under
an EPA research grant to determine the effectiveness of SWCPs for pollution
control.
SUMMARY OF CONCLUSIONS
The ARM Model has been used to analyze the relative effectiveness of
385
-------
three SWCPs (no tillage, contours, and terraces and contours) for controlling
runoff, sediment loss, and pesticide and nutrient washoff from two small
agricultural watersheds. The results show that, in general, the three SWCPs
reduce surface runoff, sediment loss, and associated constituents. Sub-
surface flow is increased, as is soluble nitrogen (NOs). Adsorbed pollutants
(atrazine, methyl parathion, and soluble NH^ and PO^) do not exhibit this
increase although their major form of transport is in solution. Sediment
and associated constituents decrease for all SWCPs studied.
Terraces with contours is the most effective SWCP of the three studied
for controlling most constituents. No tillage is the most effective in
reducing sediment-associated pollutants. Contours is more effective in de-
creasing soluble constituents than no tillage; however, it is less effective
than terraces plus contours.
The two watersheds studied are in different climatic regions of the
United States and therefore exhibit different responses to the introduction
of the three SWCPs. In general, the difference in effectiveness of the
SWCPs is not large. Where major differences do occur they are a result of
differing climatic conditions. The influence of snow in winter months on the
P6 watershed (Michigan) is significant. Snow cover delays and dampens the
hydrologic response of the watershed and often reduces what otherwise would
be peak rainfall and runoff events.
The parameter value changes in the ARM Model to represent the alternative
SWCPs are estimates. As such, they and their associated algorithms are
subjected to potential refinement once we have a better understanding of the
actual mechanisms at work. However, the informalionvgained from this study
is useful in providing a better understanding of how the different hydro-
logic, sediment, pesticide, and nutrient processes interrelate and react
to changes imposed by man.
AGRICULTURAL RUNOFF MANAGEMENT (ARM) MODEL
Model Structure
The ARM Model is a continuous model that simulates runoff (including
snow accumulation and melt), sediment, pesticides, and nutrient contributions
to stream channels from both surface and subsurface sources. Figure F-l
demonstrates the general structure and operation of the ARM Model. The
major components of the model individually simulate the hydrologic response
(LANDS) of the watershed, sediment production (SEDT), pesticide adsorption/
desorption (ADSRB), pesticide degradation (DEGRAD), and nutrient trans-
formations (NUTRNT). The executive routine, MAIN, controls the overall
execution of the program: calling subroutines at proper intervals, trans-
ferring information between routines, and performing the necessary input
and output functions.
In order to simulate vertical movement and transformations of pesti-
cides and nutrients in the soil profile, specific soil zones (and depths)
are established so that the total soil mass in each zone can be computed.
Total soil mass is a necessary ingredient in the pesticide adsorption/
desorption reactions and nutrient transformations. The vertical soil
386
-------
INPUT
OUTPUT--
U)
oo
MAIN
EXECUTIVE
PROGRAM
NUTRNT
NUTRIENT TRANSFORMATION
AND REMOVAL
-*-CHECKR CHECK INPUT SEQUENCE
-»-NUTRIO READ NUTRIENT INPUT
-^OUTMON, OUTYR OUTPUT SUMMARIES
LANDS
HYDROLOGY
AND SNOW
SEDT
SEDIMENT
PRODUCTION
PEST
YES
yes
NO
ADSRB
PESTICIDE ADSORPTION
AND REMOVAL
DEGRAD
PESTICIDE
DEGRADATION
FIGURE F-l. ABM MODEL STRUCTURE AND OPERATION
-------
zones simulated in the ARM Model include the surface, upper, lower, and
groundwater zones. The depths of the surface and upper soil zones are
specified by the model input parameters and are generally 2-8 ram and
75-150 mm, respectively. The upper zone depth corresponds to the depth
of the incorporation of soil-incorporated chemicals. It also indicates
the depth used to calculate the mass of soil in the upper zone whether
agricultural chemicals are soil-incorporated or surface-applied.
The transport and vertical movement of pesticides and nutrients, as
conceived in the ARM Model, is indicated in Figure F-2. Pollutant contri-
butions to the stream can occur from the surface zone, the upper zone, and
the groundwater zone. Surface runoff is the major transport mechanism
carrying dissolved chemicals, pesticide particles, sediment, and adsorbed
chemicals. The interflow component of runoff can transport dissolved pesti-
cides or nutrients occurring in the upper zone. Vertical chemical move-
ment between the soil zones is the result of infiltrating and percolating
water. From the surface, upper, and lower zones, uptake and transformation
of nutrients and degradation of pesticides is allowed. The groundwater
zone is presently considered a sink for deep percolating chemicals.
Model Components
The algorithms or equations used to describe the processes simulated
by the ARM Model are fully discussed in the original model reports (Crawford
and Donigian, 1973; Donigian and Crawford, 1976; Donigian et al. 1977).
A brief presentation of the general methodology is included here.
Hydrology—
Hydrologic simulation by the LANDS subprogram is derived from modi-
fications of the Stanford Watershed Model (Crawford and Linsley, 1966) and
the Hydrocomp Simulation Program (Hydrocomp, Inc., 1976). Through a set
of mathematical functions, LANDS simulates continuously the major components
of the hydrologic cycle, including interception, surface runoff, interflow,
infiltration, and percolation to groundwater. In addition, energy balance
calculations are performed to simulate the processes of snow accumulation
and melt.
Sediment--
The algorithms for simulating soil loss, or erosion, were initially
derived from research by Negev (1967) at Stanford University and have been
subsequently influenced by the work of Meyer and Wischmeier (1969) and
Onstad and Foster (1975).
Although Negev simulated the entire spectrum of the erosion process,
only sheet and rill erosion are included in the ARM Model. The two com-
ponent processes of sheet and rill erosion pertain to (1) detachment of
soil fines (generally the silt and clay fraction) by raindrop impact, and
(2) pick-up and transport of soil fines by overland flow. These processes
are represented as follows:
388
-------
00
VO
TOTAL U
AND DEI
t
\ APPLICATION /
PTAKE
iRADATION ,
,_ APPLICATION
P MODE
SOIL INCORPORATED p/N nN crn,MFNT
SURFACE APPLIED K/N UH itulMtNI
llPTtKf AND -*- SURFACE P/N « » SURFACE P/N PESTICIDE PARTICLES
DEGRADATION STORAGE ~" " INTERACTIONS
I \ P/N IN OVERLAND HOW
INHLIRATION
*
UPTAKE AND * IIPPFR 7HHE P/N - » MPPFR 70NF P/N P/N IN INTERFLOW
nFBB»pftTioN -« STORAGE ' ~~ "" INTERACTIONS
PERCOLATION
UPTAKE AND ^ LOWER ZONE P/N ^ fc LOWER ZONE P/N
DEGRADATION STORAGE ^ *" INTERACTIONS
LOSSES TO GROUNDWATER
GROUNDWATER .-__. G
P/N STORAGE *
rn <;TBFdM ^
KEY
( INPUT )
FUNCTION
j STORAGE
P-PESTICIDE
N- NUTRIENT
*._
»,
ROUNDWATER P/N _fc.
INTERACTIONS
\
FIGURE F-2. PESTICIDE AND NUTRIENT MOVEMENT IN THE ARM MODEL
-------
Soil fines detachment:
TRFR
RER(t) = (l-COVER(T))*SMPF*KRER*PR(t) (F.I)
Soil fines transport:
SER(t) =
fKSER*OVQ(t)JSER, for SER(t) SRER(t) (F.3)
ERSN(t) = SER(t)*F (F.4)
where RER(t) = soil fines detached during time interval t, metric
tons/ha
COVER(T) = fraction of vegetal cover as a function of time, T
within the growing season
KRER = detachment coefficient for soil properties
SMPF = supporting management practice factor (P-factor in
Universal Soil Loss Equation, Wischmeier and Smith
1965)
PR(t) = precipitation during the time interval, mm
JRER = exponent for soil detachment
SER(t) = transport of fines by overland flow, metric tons/ha
JSER = exponent for fines transport by overland flow
KSER = coefficient of transport
SRER = reservoir of soil fines at the beginning of time
interval, t, metric tons/ha
OVQ(t) = overland flow occurring during the time interval, t,mm
F = fraction of overland flow reaching the stream during
the time interval, t
ERSN(t) = sediment loss to the stream during the time interval,
t, metric tons/ha
In the operation of the algorithms, the soil fines detachment (RER) dur-
ing each time (5 or 15 min) interval is calculated by Equation F.I and
added to the total fines storage or reservoir (SRER). Next, the total
transport capacity of the overland flow (SER) is determined by Equation F.2.
Sediment is assumed to be transported at capacity if sufficient fines are
available, otherwise the amount of fines in transport is limited by the
fines storage, SRER (Equation F.3). The sediment loss to the waterway in
the time interval is calculated in Equation F.4 by the fraction of total
overland flow that reaches the stream. An overland flow routing technique
determines the flow contribution to the stream in each interval. After
the fines storage (SRER) is reduced by the actual sediment loss to the stream
(ERSN), the algorithms are ready for simulation of the next time interval.
Thus, the sediment that does not reach the stream is returned to the fines
storage and is available for transport in the next time interval. The
methodology attempts to represent the major processes of importance in
soil erosion so that the impact of land management practices (for example,
tillage, terracing, mulching, etc.) can be specified by their effects on the
sediment parameters.
390
-------
Since land cover by growing crops and crop residues has a major impact on
sediment loss, the variability in the land surface cover is explicitly
represented in the ARM Model. The land cover variable in Equation F.I,
COVER(T), represents the fraction of the land surface effectively protected
from the kinetic energy and detachment capability of rainfall. Monthly
cover values as of the first day of the month are specified by the user.
The model interpolates linearly between the monthly values to evaluate land
cover on each day. Figure F-3 demonstrates the land cover function in the
model. The kinetic energy of rainfall is effectively dissipated by the
land cover with values of 90 to 95 percent of the area. Thus, judicious
use of land cover function allows simulation of various land surface condi-
tions for different practices.
The timing and severity of tillage operations have a controlling effect
on the sediment loss from an agricultural watershed. With regard to sedi-
ment production, the effect of tillage operations is to increase the mass
of soil fines available for transport and produce a reasonably uniform
distribution of fines across the watershed. Consequently, the ARM Model
allows the user to specify the dates of tillage, planting, or other land- ,
surface disturbing operations. For each of these dates the user must specify
a new detached soil fines storage resulting from the operation. At the
beginning of each tillage day the ARM Model resets the fines storage to the
new value, resulting in a uniform fines distribution across the watershed.
The amount of fines storage produced by different tillage operations is
related to the depth and extent of the operation and edaphic characteristics.
Pesticide—
The process of pesticide adsorption/desorption onto sediment particles
is a major determinant of the amount of pesticide loss that will occur.
This process establishes the division of available pesticide between the
water and sediment phases, and thus specifies the amounts of pesticide
transported in solution and on sediment. To simulate this process in the
ARM Model, the following equation is used:
X/M = KC1/N + F/M (F.5)
where X/M = pesticide adsorbed per unit soil, yg/g
F/M = pesticide adsorbed in permanent fixed state per unit soil.
F/M is less than or equal to FP/M, where FP/M is the permanent
fixed capacity of the soil in yg/g for each pesticide.
C = equilibrium pesticide concentration in solution, mg/1
N = exponent
K = coefficient
This algorithm is comprised of an empirical term, F/M, plus the standard
Freundlich single-valued (SV) adsorption/desorption isotherm. The empirical
term F/M, accounts for pesticides (for example, paraquat) that are permanent-
ly adsorbed to soil particles and will not desorb under repeated washing.
The ARM Model includes an option to use a non-single-valued (NSV) ad-
sorption/desorption function because research has indicated that the assump-
391
-------
Without Crop Residue
With Crop Residue
• Input Values
P2 Watershed
i i i i
P6 Watershed
JFMAMJJASOND
FIGURE F-3. LAND COVER WITH AND WITHOUT CROP RESIDUE
392
-------
tion of single-valued adsorption/desorption is not valid for many pesticides
(Davidson et a^. 1973). In these cases, the adsorption and desorption
processes result in different pesticide concentrations. The form of the
desorption equation is identical to Equation F.5 except that K and N values
are replaced by K1 and N1 respectively, with the prime denoting the desorp-
tion process. The user specifies the N1 value as an input parameter and
the ARM Model calculates K1 as a function of the adsorption/desorption
parameters (K, N, N') and the pesticide solution concentration (Davidson
et^ ai^. 1973). The NSV function simulates higher pesticide concentrations
on sediment than the SV function in order to represent the irreversibility
of the adsorption process.
Attenuation of applied pesticides, through volatilization and degrada-
tion processes, is critical to the accurate simulation of pesticide runoff
because these mechanisms control the amount of chemical available for trans-
port. These processes are not well understood and are topics of continuing
research. To approximate the pesticide attenuation following application,
the ARM Model includes a step-wise first-order attenuation function that
allows the use of different degradation rates for separate time periods after
application. The function calculates the combined degradation of pesticides
by volatilization, microbial degradation, and other attenuation mechanisms.
This approach was chosen after evaluating both simpler and more sophisticated
degradation models (Donigian et^ al^. 1977).
The simulation of methyl parathion washoff requires the addition of
algorithms to represent pesticide storage, decay, and washoff from the crop
cover. All previous pesticides studied with the ARM Model are applied to
the soil surface at planting time. Methyl parathion, an insecticide used
on cotton, is applied by spraying the crop every one to three weeks during
the crop's growing season once the cotton squares begin to form. The
amount of methyl parathion applied on the crop compared to the amount miss-
ing the crop and reaching the soil surface is assumed to be proportional to
the fraction of crop cover. For example, if the crop cover equals 0.9,
then nine-tenths of the pesticide applied by spraying is stored on the
crop and one-tenth is stored on the surface layer of the soil.
Pesticide is removed from the crop by rainfall according to the fol-
lowing relationship:
PRFC = CSTR*(1.0 - EXP(PCK*PR*TIMFAC/60.)) (F.6)
where PRFC - pesticide removed from crop storage, grams
CSTR = pesticide stored on crop, grams
PCK = pesticide crop constant
PR = rainfall, mm/interval (min)
TIMFAC = interval time, min
This equation assumes that the amount of pesticide washed off the
crop is exponentially related to the rainfall intensity. Data to confirm
this equation and determine PCK are scarce. Information on parathion wash-
off collected by Gunther (1977) was the most pertinent of the literature
reviewed.
393
-------
Pesticide washed off the crop cover by rainfall is added to that in the
soil surface layer. In the following time intervals this pesticide now in
the soil surface layer can either stay in solution, adsorb onto sediment par-
ticles, or infiltrate into the upper and lower soil zones.
The remaining pesticide on the crop is decreased by decay or degradation.
Degradation on the crop is a first-order decay process, same in concept as
degradation in the surface, upper, and lower soil zones. However, the daily
decay rate for the pesticide stored in the crop cover can be different from
the rate used for pesticide in the soil. This is because of the different
mechanisms involved in breaking down the pesticide on the crop and in the
soil. Degradation of methyl parathion in the soil is mainly by biological
organisms; methyl parathion's half-life in the soil is 45 days (Menzie, 1972).
On the crop surface methyl parathion is degraded by photochemical processes
and resulting half-life is only 8 days (Brown, EPA, personal communication,
1978) . The ARM Model uses each degradation rate for methyl parathion, one
for the pesticide on the crop and the other for the pesticide in the soil.
Finally, any methyl parathion remaining on the crop at harvest time is
removed in proportion to the percent of crop cover removed by the harvest.
This allows for complete accounting of the pesticide by the model at all
stages of its existence in the watershed.
Nutrients--
Nutrient simulation in the ARM Model attempts to represent the reactions
and transformations of nitrogen and phosphorus compounds in the soil profile
as a basis for predicting the nutrient content of agricultural runoff. The
nutrient model assumes first-order reaction rates and is derived from work
by Mehran and Tanji (1974), and Hagin and Amberger (1974). The processes
simulated include immobilization, mineralization, nitrification/denitrifica-
tion, plant uptake, and adsorption/desorption. In the model, fertilizer,
plant residue, or animal waste is applied in their chemical form
Cl, and organic N and P) to the land surface or into the soil.
The ARM Model simulates nutrient movement in the watershed by water or
sediment. Transformations of nutrients determine the nutrient forms in
each soil zone and their resulting susceptibility to movement. The ARM
Model imitates nutrient transport processes only to the extent that is needed
to predict runoff quality and quantity. The model does not consider the
generally secondary movement of soil chemicals by concentration and thermal
gradients. However, it does model the lateral and downward transport of
chemicals by water from the soil zones. Lateral transport of nutrients
towards the stream can occur from the surface and upper zone storages.
Groundwater transport is not currently modeled.
Figure F-4 diagrams the transformation pathways and storages and gives
the names of the reaction rate input parameters. Reaction rates are input
on a per day basis for each soil zone. Nitrite (N02) transforms so quickly
in most agricultural areas that it is not considered separately. The
adsorbed phase represents the nutrients in a complex form along with those
adsorbed on the soil. The plant uptake rates (KPL) are modified monthly by
394
-------
N2
PLNT-N
KD
N03
KPL
(+NO2)
K1
NH4-A
KAS
NH4-S
KAM
KIM
ORG-N
KKIM
A. Nitrogen transformations in ARM model
PLNT-P
PO4-A
KAS
KSA
KPL
PO4-S
KIM
KM
Key
N- Nitrogen
P - Phosphorus
A -Adsorbed
S- Solution
K- Reaction Rate
Parameters
ORG-P
B. Phosphorus transformations in ARM model
FIGURE F-4. NUTRIENT TRANSFORMATIONS
395
-------
an input parameter which depends on the stage of crop growth. These monthly
input parameters are adjusted to represent the crop uptake of N and P from
the soil storages and to distribute it throughout the growing season.
Soil temperature is presently the only environmental factor that is
modeled as affecting the reaction rates with the exception that the trans-
formations are stopped entirely at very low moisture levels. Soil tempera-
tures for the surface and upper zone are determined by regression equations
based on air temperature, while the daily soil temperature for the lower
zone and groundwater is interpolated from average monthly input values.
Other factors deemed as having a constant influence during the simulation
period, such as soil pH, are represented by adjustments to the input reaction
rates.
DESCRIPTION OF SOIL AND WATER CONSERVATION PRACTICES
In a joint study by the EPA and ARS (Stewart e_t al. 19751, 18 soil and
water conservation practices (SWCPsJ were determined to be important in
controlling runoff and sediment loss from agricultural areas. Of these 18
practices we selected three for detailed study. These three practices are
the use of contours, terraces and contours, and no tillage. These practices
were selected based on their relative importance as a SWCP and the ability
of the model to represent the particular changes involved in each practice.
To represent a SWCP by changing model parameter values, it is important
to define how the SWCP differs from the base conditions or standard practice.
In addition, definition of the base conditions is necessary, as the base
conditions change for each region of the country. The ARM Model has been
tested on the ARS-EPA P2 watershed near Watkinsville, Georgia, and the
MSU-EPA P6 watershed in East Lansing, Michigan, (Donigian et^ al. 1977) and
was found capable of representing the hydrology, sediment movement, and
pesticide and nutrient washoff measured on the two watersheds. This previous
work provided initial parameter values for the model. It also provided
an initial practice on which to base these parameter values. A comparison
of the management practices used on the watershed with three SWCPs to be
studied helped to define the necessary parameter value changes. But before
the parameter changes can be discussed it is important to clearly define
each soil and water conservation practice relative to the base conditions.
Base Conditions
For this study, the base conditions represent conventional practices in
common use in the region of the test watershed. For our two watersheds this
means planting row crops in straight rows parallel to (that is, up and down)
the land slope. Tillage of the soil is in the form of discing and is done
in the spring to prepare the soil for planting. After harvest the crop
residue is removed from the field and the soil is fallow through the winter
until a new cro is planted in the spring.
No Tillage
The no tillage SWCP differs from the base conditions in two ways. There
396
-------
is no tillage prior to spring planting and the crop residue is left on the
field after harvest. Planting is done with the crop residue remaining from
the previous fall. This protects the soil from erosive forces until the com-
ing spring when a new crop is planted.
Contours
The contour SWCP differs from the base conditions only in that the crops
are planted in rows perpendicular to the slope of the land, which by defini-
tion are contoured rows. For the purposes of this study, the contour SWCP
is assumed to use discing in the spring prior to planting and removal of
crop residue after harvest.
Contours and Terraces
The terrace SWCP requires physical alteration of the land to establish
and maintain terraces. Both the P2 and P6 watersheds are assumed to be div-
ided into two terraces if this SWCP is used. Contoured rows are used with
the terraces. However, it is assumed that discing and crop residue removal
practices are retained from the base condition practice.
These three SWCPs plus the base condition are modeled by a selection of
parameter values which represent each watershed condition. How the parameters
and their values were selected for each SWCP is discussed in the following
pages.
ARM MODEL PARAMETER VALUES AND SWCPs
The selection of the ARM Model parameters and their values for different
SWCPs were based on the expected physical and agronomic changes associated
with a particular SWCP. A total of eight input parameters for which adjust-
ments could be reasonably estimated were identified as being related to
SWCPs. These parameters and values for each watershed condition are listed
in Table F-l.
Determining how to change parameter values to represent a particular
SWCP on a watershed is often difficult. For some characteristics we have
related parameter value changes to established techniques (for example,
Universal Soil Loss Equation, SCS Curve Number Runoff Method) derived from
field data. However, for certain parameters, our own experience, judgment,
and knowledge of the physical processes and their representation in the
model was our only guide. The general procedure was to estimate initial
parameter changes from the literature (which was non-existent for some
parameters) and then adjust the values to conform with our experience on
other watersheds. These estimating procedures allowed us to establish
parameter values for each of the three SWCPs included in this study.
Upper Zone Nominal Moisture Storage
UZSN (upper zone nominal moisture storage) can be related to the initial
abstraction parameter, la, as defined by the SCS procedures for estimating
storm runoff volume (Chow 1964). The relationship is assumed to be propor-
397
-------
TABLE F-l. ARM MODEL PARAMETER VALUE CHANGES FOR SWCPs
Soil and Water Conservation Practices
a
Parameters
P2 Watershed
UZSN
NN
L
SS
COVPMO
SMPF
KSER
SRERTL
P6 Watershed
UZSN
NN
L
SS
COVPMO
SMPF
KSER
SRERTL0
Base
Condition
0.42
0.20
No
Tillage
0.42
0.32
100 100
0.025
(see Figure F-3)
1.0
0.6
1.5
0.17
0.20
60
0.06
(see Figure F-3)
1.0
0.6
1.0
0.025
1.0
0.55
0.15
0.17
0.32
60
0.06
1.0
0.55
0.10
Contours
0.50
0.25
100
0.025
0.5
0.5
1.5
0.20
0.25
60
0.06
0.5
0.5
1.0
Terraces
Contours
0.65
0.25
250
0.015
0.35
0.5
1.5
0.26
0.25
200
0.018
0.35
0.5
1.0
Parameters are defined as follows:
UZSN = nominal upper zone moisture storage, in.
NN = Manning's n (roughness) for overland flow
L = length of overland flow path, ft.
SS = slope of overland flow path
COVPMO = fraction of land covered by vegetation, monthly value
SMPF = supporting management practice factor (equal to the P factor
in the USLE)
KSER = sediment washoff coefficient
SRERTL = sediment fines produced by tillage operations, tons/acre
Tillage occurs once each year on day 115 (April 25) on P2 for the 10-yr.
simulation.
°Tillage occurs once each year on day 139 (May 19) on P6 for the 10-yr.
simulation.
398
-------
tional: UZSN should be changed by the same percent as la when comparing
a SWCP with a base condition. This was done for each of three antecedent
moisture conditions included with SCS method. The average percent change
for a particular SWCP was used to determine the UZSN value for that SWCP.
The calibrated UZSN values for the P2 and P6 watersheds with contours
were the baselines from which the other values were calculated.
Manning's Roughness Coefficient
NN (Manning's roughness coefficient for overland flow) is proportional
to the amount and type of ground cover and soil type. An inexact, but
relative, value of NN can be found from the literature (Chow 1964) for
comparison of SWCPs with the base condition value. However, the NN values
in Table F-l are based largely on our own experience and judgment.
Length of Overland Plow
L (length of the overland flow path) changes only for the terrace SWCP
when compared to the base condition. The modification of the watershed with
terraces changes the overland flow path such that instead of draining to
a central location to form a stream channel the flow must instead cross
the width of the terrace and then traverse the length of the terrace to
its outflow point. This increase in length will vary with the terrace con-
figuration. For our use on the P2 watershed the addition of two terraces
increased L from 30.5 m (100 ft) to 76.2 (250 ft). On the P6 watershed,
L changed from 18.3 m (60 ft) to 61.0 m (200 ft) with the addition of two
terraces.
Slope of Overland Flow Path
SS (slope of the overland flow path) is affected only by the terrace
SWCP. Installation of terraces both increases the length of the flow path
and decreases the slope. For this specific situation the slope of the P2
watershed is decreased from 0.025 to 0.015. For the P6 watershed, the slope
is changed from 0.06 to 0.018.
Crop Cover
COVPMO (cover per month) is the fraction of the land surface covered
by both live and dead vegetation and effectively protects the soil surface
from the erosive force of raindrops. When crop residue is left on the water-
shed after harvesting, as is done with the no tillage SWCP, the fraction
of land covered each month is equal to values measured by a joint EPA-ARS
study (Smith et aJ_. 1977). Monthly cover values for the period between
harvest and planting are close to zero for conditions where the crop residue
is removed and the land left without cover. On the P2 watershed, a minimum
cover value of 0.1 was used as a reasonable value for this situation (Figure
F-3).
Supporting Management Practice Factor
SMPF (supporting management practice factor) has been added to the ARM
399
-------
Model to provide a means of incorporating the effect of management practices
on the generation of sediment fines (Equation F-l). SMPF is equal to the
P-factor of the USLE. As such, data have been collected from which SMPF
can be calculated for the contour and terrace SWCPs (Stewart et^ al^. 1975).
Sediment Washoff Coefficient
KSER (sediment washoff coefficient) is affected by changes to soil sur-
face conditions brought about by the introduction of SWCPs. Fleming and
Leytham (1976) show a relationship between KSER, Manning's roughness coef-
ficient, overland flow length and slope, and mean surface particle size, al-
though the relationship's reliability is unknown. The first three of these
factors are related to the SWCPs we are investigating. Thus, although we
cannot establish any direct correspondence between KSER and a particular
SWCP, relative changes can be estimated from past experience with the model.
KSER is expected to decrease with practices that retard overland flow and
include crop resiudes. Adjustments to KSER similar to the changes in Table
F-l were included in a previous study (Hydrocomp 1978) and were found to
produce reasonable results.
Soil Fines Produced by Tillage
SRERTL (soil fines produced by tillage) is affected by the no tillage
SWCP in much the same way as COVPMO. Model testing on watershed with similar
practices provided SRERTL values for spring discing of the P2 and P6 water-
sheds. Discing is a conventional practice for a number of agricultural
regions of the United States. Thus, for our study it represents the base
condition. With no tillage, sediment fines are only produced by planting
in the crop residue. SRERTL for the no tillage SWCP is reduced to one-tenth
of the base condition value. A comparison of sediment loss data from similar
watersheds of which one was tilled and one was not supports this reduction
of SRERTL (Smith et aJ_. 1977).
Nutrient Applications
Nitrogen and phosphorus transport is directly affected by the runoff
and sediment parameter values adjusted for SWCPs. Removal of adsorbed
phosphate, ammonium, organic nitrogen, and phosphate phosphorus is in-
fluenced directly by the amount of soil eroded, while removal of ammonium
and phosphate in solution is controlled by the runoff.
In addition, SWCPs will have subtle changes on the soil nutrient trans-
formation processes and storages. Residue incorporation will affect the
mineralization and immobilization rates as well as the soil organic content.
Changes in soil organic matter content will influence the adsorptive capacity.
Varying practices will affect crop growth resulting in different rates of
plant nutrient uptake. There are obviously many more impacts which affect
transformations which feedback on the runoff and erosion processes. No
present erosion and hydrologic model considers all of these interactions,
and for our purposes they are assumed to be minor.
In our evaluation of SWCPs, only the change to the soil organic nitrogen
400
-------
and phosphorus storages under differing residue management is considered.
The need for this change is obvious. Residue is left on the watersheds under
no tillage practice, so more organic nitrogen and phosphorus is available
to be removed on sediment. The other practices assumed that only a minimum
of debris remains. Numerical estimates of stubble and root organic nitrogen
and phosphorus additions to the surface and upper zones at time of harvest
are given in Table F-2. These values were used in the ARM Model simulations
on P2 and P6 for the different SWCPs.
WATERSHED SIMULATION AND METHODOLOGY
P2 Watershed, Watkinsville, Georgia
The P2 watershed is a small (1-3 hectare), non-terraced agricultural
watershed near Watkinsville, Georgia (Figure F-5). From 1973 through 1975
EPA and ARS intensively monitored this watershed (Smith et al. 1977). The
data collected by these agencies made possible a thorough calibration of the
ARM Model's hydrology and sediment algorithms and testing of the pesticide
and nutrient routines in the model. This has been done (Donigian et al.
1977) and the calibrated model was found to reasonably represent the hydro-
logy, sediment loss, and pesticide and nutrient washoff of P2.
The management of the P2 watershed during the EPA-ARS study was a mix-
ture of SWCPs. Corn was grown on the watershed in contoured rows. The
watershed was disced each spring prior to planting, but in the autumn
after harvest the residue was left on the field until the coming spring.
The ARM Model was calibrated to represent this combination of the contour
and some no tillage practices; the calibrated parameter values reflect these
conditions. Thus, the calibrated parameter values are not identical to
the base .condition values (which assume no SWCP) or any particular SWCP.
However, with knowledge of how the actual watershed management practices
differ from the base conditions and the SWCPs, the appropriate parameter
adjustments for each condition were evaluated.
To study the effects of SWCPs on runoff, sediment loss, and pesticide
and nutrient washoff the P2 watershed was simulated for the 10-yr period
of 1966 through 1975. Simulation of hydrology, sediment, and pesticides
by the ARM Model requires 5 or 15-min rainfall and daily evaporation.
Nutrient simulation further requires daily max-min air temperature. Hourly
rainfall data collected by the National Weather Service (NWS) at the nearby
Athens Airport were disaggregated to 15-min amounts for use in the model.
Disaggregation of hourly to 15-min rainfall uses a technique developed by
Kraeger (1971) where recorded 15-min data are divided into categories by
the season and size of storm. A maximum of ten different recorded storms
are placed in each category. Recorded hourly rainfall to be disaggregated
is matched by season and size of storm to the 15-min data. Within each
category one of the ten storms is randomly selected to provide the distribu-
tion that divides the hourly data into the four 15-min values. This is
done for the entire 10-yr period of hourly rainfall using the 29 months
of 15-min rainfall to develop the categories and distributions. Daily
evaporation and max-min air temperature data used in the simulation were
also obtained from the NEW and were collected at stations within a 100-kilo-
401
-------
meter radius of the watershed.
Three different pesticides (atrazine, paraquat, and methyl parathion)
were selected for study on the watershed. Atrazine and paraquat were ap-
plied and monitored during the EPA-ARS study and reasonable simulation re-
sults were obtained with the ARM Model. In the 10-yr simulation period of
this study, atrazine and paraquat were applied once in the spring of each
year at planting time. The application amounts (2.91 kg/ha for atrazine,
2.52 kg/ha for paraquat) and application date were selected based on average
values and dates from the 3-yr EPA-ARS study. Methyl parathion was simulated
with six applications of 1.12 kg/ha each during the crop growing season of
each year. Fertilizer (N and P) was applied twice each year (Table F-3).
The ARM Model simulated hydrology, sediment, and pesticides on a 15-min
time step. Nutrient washoff was also computed on a 15-min time step al-
though the nutrient transformations in the soil were computed once every
6 hr. When runoff greater than a minimum value (0.003 nr/s) occurred a
complete list of runoff constituents was written to a computer file for
later analysis. Separate simulation runs were made for atrazine, paraquat,
methyl parathion, and nutrients for each of the three SWCPs and the base
conditions.
TABLE F-2. NUTRIENT APPLICATION CHANGES FOR SWCPs
Base
Condition
No
Tillage
Contours
Contours
and
Terraces
P2 Watershed
Day 297
Surface Zone
Organic N (kg/ha)
Organic P (kg/ha)
Upper Zone
Organic N (kg/ha)
Organic P (kg/ha)
P6 Watershed
Day 273
Surface Zone
Organic N (kg/ha)
Organic P (kg/ha)
Upper Zone
Organic N (kg/ha)
Organic P (kg/ha)
2.52
0.34
22.4
4.48
2.52
0.28
17.9
3.36
8.96
1.68
22.4
4.48
6.72
1.34
17.9
3.36
2.52
0.34
22.4
4.48
2.52
0.28
17.9
3.36
2.52
0.34
22.4
4.48
2.52
0.28
17.9
3.36
402
-------
231.5
231.0
230.5
230.
0
0 10 20 METERS
232.0
232.5
DRAINAGE PATTERN
CONTOUR LINES
I METERS ABOVE M.S.I.
SAMPLING STATION
FIGURE F-5. P2 WATERSHED, WATKINSVILLE, GEORGIA (1.3 ha)
403
-------
TABLE F-3. NITROGEN AND PHOSPHORUS FERTILIZER APPLICATIONS
Watershed
P2
Day of
Application
Each Year
119
162
Form
Applied
sulphate of
ammonia
superphosphate
50% urea
50% ammonia
N
(kg/ha)
38.0
100.7
P
(kg/ha)
33.0
0.0
P6
140
189
ammonium
nitrate
monocalcium
phosphate
ammonium
nitrate
68.3
129.9
92.9
0.0
P6 Watershed, East Lansing, Michigan
The P6 watershed (0.8 hectares) is a non-terraced agricultural water-
shed located on the soil science farm at Michigan State University in East
Lansing, Michigan (Figure F-6). Between 1973 and 1975, MSU monitored this
watershed for an EPA-sponsored study (Ellis et_ a\_. 1978). The data collect-
ed during this study were later used to calibrate the hydrology and sediment
algorithms of the ARM Model and test the pesticide and nutrient routines in
the same manner as was done on the P2 watershed (Donigian et^ al. 1977).
Like the P2 watershed, the management of P6 consisted of a mixture of
SWCPs. During the 1974 and 1975 crop seasons (the period on which the ARM
Model was calibrated), contf was grown in contoured rows. The soil was tilled
each spring to a depth of 7.6 cm before planting and fertilizer application
(Hubbard 1975). After harvest in the autumn the watershed was left bare of
residual until the new crop emerged in the spring. The ARM Model was cali-
brated to represent these practices, and thus the parameter values differ
from the base condition values. Values were then changed to represent base
conditions in the same manner as was done on the P2 watershed.
The P6 watershed was simulated for the 10-yr period of 1966 through
1975. Simulation of hydrology on this watershed included the simulation
of snowfall and snow accumulation and melt during the winter months. Snow
simulation requires daily wind, radiation, and dewpoint data in addition
to 15-min precipitation, daily evaporation, and daily max-min air tempera-
ture required for all months. Hourly precipitation and daily meteorological
data at Lansing and East Lansing were obtained form the NWS. Hourly
404
-------
I
270.0
0 10 20 METERS
270.5
DRAINAGE PATTERN
CONTOUR LINES
[METERS ABOVE M.S.L.
SAMPLING STATION
P6 WATERSHED, ESST UNSING, MICHIGAN (0.8 ha)
405
-------
precipitation data were disaggregated into 15-min amounts in the same manner
as was done for the P2 watershed.
Atrazine and paraquat were simulated on the P6 watershed. These pesti-
cides were applied and monitored during the MSU study and were adequately
simulated with the ARM Model. Each pesticide was applied in the spring of
each year at planting time. The application amounts for atrazine (2.8
kg/ha) and paraquat (1.46 kg/ha) and the application date were selected based
on average values and dates from the MSU study. In a like fashion fertilizer
was applied on the watershed (Table F-3).
The simulation interval, criteria for writing runoff constituents to a
computer file, and the number of simulation computer runs for P6 was the
same as was done for P2, except that methyl parathion was not simulated on
the P6 watershed.
SIMULATION RESULTS AND DISCUSSION
The continuous information produced by the 10-yr ARM Model simulations
was analyzed in two ways: (1) mean annual runoff and pollutant losses, and
(2) frequency of occurrence of runoff rates, pollutant concentrations, and
pollutant flux (that is, mass removal in mass per minute). The SWCPs are
represented only by the parameter changes discussed above. Other effects of
SWCPs, such as changes in infiltration characteristics, crop growth, soil
fertility, etc. are not considered due to lack of sufficient evidence to in-
dicate appropriate parameter adjustments. Also, the pesticide and nutrient
parameters (for example, application rates and time, adsorption coefficients,
reaction rates) were the same for all conditions to provide a basis for
comparing the effects of specific SWCPs. Thus, increasing pesticide appli-
cations to compensate for no tillage practices was not included in this
analysis, but such a practice can be evaluated in a similar manner.
Mean Annual Values
P2 Watershed (Georgia) Results-
Table F-4 summarizes the mean annual runoff and pollutant losses (in
solution and on sediment) for each watershed condition on the P2 watershed
and provides the percent change for each SWCP from the base conditions. The
results show that the total annual runoff decreases with each SWCP, with
minor reductions (<10%) for no tillage, 10 to 12 percent for contours, and
30 to 40 percent reductions for contours and terraces. Since overland flow
comprises more than 90 percent of the total runoff for this small watershed,
the reductions for overland flow are similar. Interflow, which travels
subsurface for nart of its flow nath. increases 10 to 12 nercent for no
tillage, less than 5 percent for contours, and approximately 30 percent for
contours and terraces. Thus, the practices that retard and reduce overland
flow also increase the subsurface flow. Both interflow and groundwater
contributions increase although groundwater is not present in the runoff
from this small watershed.
Sediment loss for each SWCP partially reflects the reductions in surface
406
-------
TABLE F-4. P2 MEAN ANNUAL RUNOFF AND POLLUTANT LOSSES
o
-j
Total Runoff (mm)
Overland Flow (mm)
Interflow (mm)
Sediment Loss
(metric tons/ha)
Total Atrazine Loss
(g/ha)
in solution (g/ha)
on sediment (g/ha)
Total Methyl Parathion
Loss (g/ha)
in solution (g/ha)
on sediment (g/ha)
Paraquat on sediment
(g/ha)
Nitrogen (kg/ha)
Organic
NH^ - solution
NH^ - adsorbed
NO 3 + N02
Phosphorus (kg/ha)
Organic
PO^ - solution
P0i+ - adsorbed
Mean
Base
Conditions
198.6
185.6
13.0
6.33
53.5
53.1
0.41
310.7
283.1
27.6
340.6
7.72
2.08
0.75
2.90
0.88
0.48
2.46
Annual Value
No
Tillage
185.8
171.3
14.5
3.30
47.2
46.9
0.34
263.7
245.6
18.1
187.5
7.01
2.29
0.44
3.34
0.87
0.56
1.64
Contours
177.4
163.9
13.5
3.99
40.9
40.6
0.28
256.5
241.2
15.3
225.0
5.89
1.98
0.51
3.11
0.66
0.47
1.85
Terraces
and
Contours
127.7
110.8
16.9
2.84
19.0
18.8
0.16
154.7
145.2
9.5
162.5
4.60
1.95
0.37
3.91
0.50
0.46
1.42
Percent
No
Change
Tillage Contours
- 6
- 8
+12
-48
-12
-12
-17
-15
-13
-34
-45
- 9
+10
-41
+15
- 1
+ 17
-33
-11
-12
+ 4
-37
-24
-24
-32
-17
-14
-45
-34
-24
- 5
-32
+ 7
-25
- 2
-25
from Base
Terraces
and
Contours
-36
-40
+30
-55
-64
-65
-61
-50
-49
-66
-52
-40
- 6
-51
+35
-43
- 4
-42
-------
runoff. Contours and terraces reduce sediment loss from 6.3 metric tons/
hectare for base conditions to 2.8 metric tons/hectare, a 55 percent reduc-
tion. The sediment reduction for no tillage is 48 percent; for contours it
is 37 percent. No tillage reduces sediment loss by minimizing the disrup-
tion of the soil surface and protecting the surface with crop residues re-
maining on the watershed. In contrast, contours reduce sediment loss by
reducing overland flow.
Pesticide and nutrient contents of runoff for each SWCP are the result
of the changes in runoff and sediment loss discussed above. Although atra-
zine is 'detected both in solution and on sediment, the solution for the
SWCPs are in the same order as for total runoff; the negative values increase
with no tillage, contours, and contours plus terraces. Atrazine loss on
sediment is relatively minor and, as expected, the reductions are similar
to those for sediment loss. Total atrazine loss is reduced by 64 percent
for contours and terraces, 24 percent for contours alone, and 12 percent
for no tillage.
Paraquat is a highly ionic compound that is strongly and irreversibly
adsorbed to sediment particles. Thus, the values for paraquat loss are
shown only for sediment in Table F-4 because paraquat is not transported in
solution. For each SWCP, the percent reductions for paraquat and sediment
are basically the same.
Methyl parathion behaves somewhat like both atrazine and paraquat. It
easily and quickly adsorbs on sediment, yet the large fraction of its wash-
off is present in solution form. The percent reductions for the three SWCPs
are close to those for total runoff. Total methyl parathion loss is reduced
by 50 percent for contours and terraces and 15 to 20 percent for no tillage
and contours.
The nutrient values in Table F-4 are presented by the separate nitrogen
and phosphorus forms simulated with the ARM Model. The sediment -associated
nutrients (organic N and P, adsorbed NH^, and adsorbed POit ) show the same
relative percent reductions for contours and contours plus terraces as
sediment loss. The reductions are in the range of 25 to 35 percent for
contours and 40 to 50 percent for contours and terraces. The variations
reflect the specific reaction and adsorption behavior of the individual com-
ponents. The no tillage SWCP reduces adsorbed NHi, and POi* by 30 to 40
percent. Organic N and P are reduced by less than 10 percent, however. This
is a result of adding organic N and P to the surface zone in the form of
crop residue, and this offsets the larger expected decrease due to the
SWCP.
Both soluble NH^ and soluble POi^ show relatively minor changes for any
of the SWCPs. Both constituents increase slightly under no tillage and de-
crease slightly for contours with and without terraces. Except to show the
expected direction of a change, percentage changes less than 10.0 are not
likely to be significant within the accuracy of the model. Thus, these SWCPs
do not appear to have a major impact on the mean annual losses of these
constituents.
408
-------
On the other hand, N03 (and N02) loss is increased by 35 percent when
terraces are used, with only minor increases under no tillage and contours.
This change is a result of the increase in interflow when terraces are in-
stalled since interflow is the major transporting mechanism for N03. The
other solution nutrients are not as drastically affected by interflow because
they also adsorb onto sediment particles, and thus do not move as readily
through the subsurface environment.
P6 Watershed (Michigan) Results—
The mean annual results simulated on the P6 watershed are presented in
Table F-5.
The P6 results show that annual total runoff decreases 4 percent for no
tillage, 6 percent for contours, and 25 percent for contours and terraces.
Interflow is increased for both no tillage (4 percent) and terraces (24
percent). Unlike the other SWCPs, the introduction of the contour SWCP does
not change the volume of interflow from that of the base conditions. The
percent changes from base conditions for the three SWCPs for runoff are of
similar magnitude as the P2 (Georgia) results. The major difference be-
tween the runoff percent changes on the P2 and P6 watersheds is the slightly
less variability of the P6 percent changes in comparison to the P2 results.
This is probably because the P6 watershed (Michigan) receives most of its
winter precipitation as snowfall and P2 (Georgia) does not. Runoff from
snowmelt is generally more gradual than that produced by rainfall. Thus,
the SWCPs appear to be less effective in controlling slow snowmelt runoff
than the more transitory runoff produced by rainfall.
Sediment loss is reduced by 71 percent with the use of no tillage. The
reductions for contours and terraces are 21 and 32 percent, respectively.
Snowfall, unlike rainfall, does not produce soil fines available for trans-
port from the watershed. No tillage further reduces soil fines generation
by increasing soil surface cover and, therefore, is much more effective for
control of sediment and associated pollutants on the P6 watershed. The two
effects combine to severely limit fines generation and washoff. Contours
and terraces only limit soil washoff through decrease of runoff volume and
velocity.
Pesticide washoff is largely a function of runoff and sediment loss,
as is noted in the discussion of the P2 results. Atrazine is mainly removed
in solution. Total atrazine loss is decreased 3 percent for no tillage, 20
percent for contours, and 62 percent for contours and terraces. These re-
sults are comparable to those for the P2 watershed.
The paraquat results are directly related to sediment loss, since
paraquat is irreversibly adsorbed to sediment particles. The paraquat
washoff results for the three SWCPs are nearly identical to the sediment
loss results as shown in Table F-5.
Methyl parathion was not simulated on the P6 watershed. Methyl para-
thion is most commonly used on cotton to control insect damage. Cotton is
409
-------
TABLE F-5. P6 MEAN ANNUAL RUNOFF AND POLLUTANT LOSSES
Mean Annual Value
Percent Change from Base
Total Runoff (ram)
Overland Flow (mm)
Interflow (mm)
Sediment Loss
(metric tons/ha)
Total Atrazine Loss
(g/ha)
in solution (g/ha)
on sediment (g/ha)
Paraquat on sediment
(g/ha)
Nitrogen (kg/ha)
Organic
- solution
- adsorbed
NO3 + N02
Phosphorus (kg/ha)
Organic
- solution
- adsorbed
Base
Conditions
190.9
165.1
25.8
No
Tillage
182.6
155.8
26.8
Contours
179.2
153.5
26.7
Terraces
and
Contours
144.0
111.9
32.1
No
Tillage
- 4
- 6
+ 4
Contours
- 6
- 7
+ 0
Terraces
and
Contours
-25
-32
+24
3.51
79.2
77.4
1.84
114.8
1.58
6.55
1.56
1.01
76.8
75.7
1.11
36.3
0.74
6.74
0.58
2.78
63.0
61.7
1.27
93.4
2.40
29.8
28.9
0.86
81.2
1.35
6.01
1.31
1.18
5.96
1.17
-71
- 3
- 2
-40
-68
-53
+ 3
-63
-21
-20
-20
-31
-19
4.22
2.14
0.10
14.71
2.19
2.52
0.03
15.61
3.60
1.93
0.08
13.80
3.16
1.87
0.06
14.48
-48
+18
-70
+ 6
-15
-10
-20
- 6
-15
- 8
-16
-32
-62
-63
-53
-29
-25
-13
-40
- 2
-25
- 9
-25
-------
not a major commercial crop in upper Midwest, and therefore was not considered
to be of enough interest to be simulated with applications of methyl parathion.
The nutrient results on the P6 watershed in general show the same type
of results as were found for P2. The sediment associated nutrients (organic
N and P, adsorbed NH4, and adsorbed POiJ decrease by 50 to 70 percent for no
tillage, 15 to 20 percent for contours, and 25 to 40 percent for contours
and terraces. These ranges are very close to the reductions measured for
sediment loss for the three SWCPs.
In general, the soluble nutrients (soluble NH^, soluble PO^, and
nitrate and nitrite) are only slightly increased for no tillage and are
decreased by less than 10 percent for contours and contours and terraces.
The exceptions are the increase of soluble NH^ by 18 percent for no tillage
and the decrease of the same constituent by 13 percent for contours plus
terraces. The magnitude of these changes are due to the effect of crop
residue for no tillage and the decrease in total runoff for contours plus
terraces.
There is one noticeable difference in the P2 and P6 nutrient results.
In the P6 watershed the PO^ in solution is larger than adsorbed POi^ sim-
ulated in the runoff. The reverse was simulated on P2. The removal of
POi,. in solution is a function of interflow, the kinetic reaction rate between
soluble and adsorbed PO^, and other factors. On the P6 watershed the pro-
portion of runoff that is interflow is twice that from P2. Also the simula-
tion of soluble PO^ on the P6 watershed by Donigian, &t_ al. (1977) resulted
in more PO^ washoff in solution than on sediment. This does not neces-
sarily mean that the simulation of PO^ on the P6 watershed is accurate, but
it is consistent for different conditions and, therefore, is useful in
measuring the size of change in POi* washoff for different SWCPs. This is
our primary goal in nutrient simulation in this study.
Frequency Analysis
Frequency analysis was used to determine the frequency or probability
of occurrence of runoff and pollutant levels on the two watersheds. Anal-
ysis was conducted on the interval runoff data produced during the 10-yr
simulation periods under a range of meteorologic and environmental condi-
tions. The results of the frequency analysis are compared to determirte
how these frequencies change with each different SWCP.
P2 Watershed (Georgia) Results--
The frequency curves for the P2 watershed are shown in Figures F-7
through F-ll. The curves show the percent of time the particular consti-
tuent (for example, runoff in m3/s) is greater than the ordinate value. For
example, Figure F-7 shows that sediment concentrations for the terrace SWCP
are greater than 9.0 g/1 for 2 percent of the time (with time defined as
time during which runoff is occurring and is printed to a computer file).
In the same figure it can be seen that the base condition produces sediment
concentrations greater than 12.7 mg/1 for 2 percent of the time. Each of
the SWCP curves for each constituent can be analyzed in this manner.
411
-------
BASE CONDITIONS
NO TILLAGE
CONTOURING
CONTOURING &
0.00
4 6 8 10 12 14 16
% OF TIME GREATER THAN NOTED VALUE
FIGURE F-7. P2 RUNOFF AND SEDIMENT FREQUENCY ANALYSIS
412
-------
5.60
•H
-H 4.80
bo
K" 4.00
t-H
^ 3.20
CD
C
•H
N
rt
N 2.40
^ 1.60
rH
Oj
p 0.80
5.60
00 4.80
•s
« 4.00
Cti
53 3.20
•H
2.40
u,
C 1-60
•H
2 0.80
bo
CD
28.00
24.00
20.00
•S 16.00
g 12.00
u
g 8.00
•H
N
£ 4.00
^
0.00
BASE CONDITIONS
NO TILLAGE
CONTOURING
CONTOURING &
TERRACING
4 6 8 10 12 14 16 18
% OF TIME GREATER THAN NOTED VALUE
FIGURE F-8. P2 ATRAZINE FREQUENCY ANALYSIS
413
-------
X
3
r-*
U,
28.00
1 1 1 1
1 1 1 1 — 1 1
BASE CONDITIONS
NO TILLAGE
CONTOURING
CONTOURING &
6 8 10 12 14 16 18
% OF TIME GREATER THAN NOTED VALUE
FIGURE F-9. P2 METHYL PARATHION FREQUENCY ANALYSIS
414
-------
'560.00
480.00
•H
•| 400.00
*\
3 320.00
r—I
fc,
Z 240.00
I 1 —
1 1 1 1 1 1 — 1 — 1 —
cti
•M
O
H
•H
160.00
80.00
70.00
60.00
50.00
O 40.00
•rH
30.00
20.00
10.00
CO
.5
"i 1 1 1 1 1 r
i i I i i r
• BASE CONDITIONS
NO TILLAGE
CONTOURING
- CONTOURING*
TERRACING
6 8 10 12 14 16 18
% OF TIME GREATER THAN NOTED VALUE
FIGURE F-10. P2 NITROGEN FREQUENCY ANALYSIS
415
-------
BASE CONDITIONS
NO TILLAGE
CONTOURING
CONTOURING ft
TERRACING
4 6 8 10 12 14 16 18 20 22
% OF TIME GREATER THAN NOTED VALUE
FIGURE F-ll. P2 PHOSPHORUS FREQUENCY ANALYSIS
416
-------
The runoff frequency graph (top graph in Figure F-7) shows relatively
small difference among the curves for base conditions, no tillage, and con-
tours, with the combined contours and terraces curve being somewhat lower.
The mean annual runoff values in Table F-4 confirm this. The sediment
concentration frequency (middle graph in Figure F-7) indicates the base
condition Concentrations are substantially higher than the three SWCPs which
produce similar values. However, the sediment flux frequency (bottom graph
in Figure F-4) more closely demonstrates the relative ranking of the SWCPs as
shown by the percent reductions in annual values in Table F-4. In general,
flux or mass removal is more representative of the effects of a particular
SWCP because it reflects changes in both concentration and flow.
Figure F-8 includes the frequency curves for total atrazine flux, atra-
zine flux in solution, and atrazine concentration in solution. For 1.0 per-
cent of the time the atrazine flux is greater than 0.3 g/min with contours
and terraces, and greater than 0.6 g/min for the other watershed conditions.
Since very little atrazine is transported on sediment, there is no signifi-
cant difference between the flux curves for total atrazine and solution
atrazine. The concentration curve in Figure F-8 shows that for 1.0
percent of the time, the atrazine concentration in water is greater 1.0 mg/1
with contours and terraces, and greater than about 1.5 mg/1 for the other
conditions. Unfortunately the curve definition in these figures is not
sufficient to clarify the differences between the curves. However, with
expanded scales the differences would be more readily apparent.
The frequency curves for methyl parathion (Figure F-9) are similar
to the atrazine curves. The curves of both pesticides show contours and
terraces to produce the largest reduction in the pesticides' flux values.
These results are expected. Although the two pesticides have different
characteristics, both are primarily removed from the watershed in solution
form. Thus, the same runoff results for a particular SWCP will produce
washoff of atrazine and methyl parathion in approximately the same propor-
tions, only the ordinate values differ.
The frequency curves can be analyzed to determine how often specific
runoff volumes, flow rates, pollutant concentrations, or flux rates will
occur. To evaluate potential ecologic impact, the frequency curves and
toxicity data can be used to estimate how often acute or chronic pesticide
levels toxic to specific organisms will exist.
Paraquat flux and concentration frequency curves are not shown here
because paraquat exhibits relatively constant concentrations and, except
for scale, the flux curves are identical to the sediment flux curves in
Figure F-7.
Total-flux frequency curves for nitrogen and phosphorus and their
solution and sediment components are shown in Figures F-10 and F-ll, respec-
tively. Solution nitrogen includes NOs (plus N02 ) and soluble NH^ while
sediment nitrogen is composed of organic nitrogen and adsorbed NH^. Con-
tours and terraces produce reductions in total and sediment nitrogen flux
curves However, for solution nitrogen the SWCP of combined contours and
terraces reduces the high flux rates but increase the rates that occur 96
417
-------
percent of the time. Also, contours alone increase the solution nitrogen
export above the base conditions for values that occur 2 percent of the time
or less. Decisions on implementing contours and /or terraces should con-
sider the relative impacts of solution and sediment nitrogen.
The phosphorus flux curves in Figure F-ll show similar results to the
nitrogen flux curves. Sediment phosphorus includes organic phosphorus and
adsorbed PO^ while solution phosphorus is comprised only of dissolved PO^.
The effects of contours and terraces on solution phosphorus are similar to
those noted above for solution nitrogen. Contours produce lower flux rates
while terraces and contours produce a flatter frequency curve with the
highest flux rates for a portion of the time.
To evaluate the net or overall impact of the alternative practices
(and to quantify the differences between the curves), the area beneath the
curve for each practice can be calculated and compared. From elementary
decision theory this area represents the expected value of the ordinate
variable under all conditions; that is, the value of the variable times
its probability of occurrence, summed over all possible occurrences. For
example, the area beneath the base sediment curve in Figure F-7 is the
measured units of the y-axis, mg/1; each block (1 x-axis unit*l y-axis unit)
is 0.08 mg/1 (4 mg/1*.02).
The differences in area beneath each curve, or the area between the
curves, can be used to evaluate the impact of a particular practice. Table
F-6 lists the area beneath each frequency curve and the percent change for
each practice from the base conditions for the P2 watershed. The percent
changes are similar to and generally slightly lower than the corresponding
values in Table F-4. The difference in the expected value results and the
mean annual value results is not usually significant and is due to the
decision to only analyze frequency data when the runoff from the watershed
is above a minimum value (set to 0.003 m3/s). This eliminates some small
portion of the flow and pollutant load from the frequency analysis that is
included in the mean annual values. Frequency analysis provided values
for expected pollutant concentrations which are important for analyzing the
impact of toxic pollutants, and our analysis procedure eliminates spurious
concentrations that may occur at low flows. However, both mean annual values
and expected values can mask the effects of extreme conditions. For that
reason, visual interpretation of the frequency curves is important to a
full understanding of how conditions vary when alternatives are compared.
P6 Watershed (Michigan) Results--
The P6 frequency curves are found in Figures F-12 through F-15. The
frequency results are presented in the same format as those for the P2
watershed. The frequency results presented and discussed for the P6 water-
shed include runoff, sediment, atrazine, nitrogen and phosphorus.
Analysis of the total runoff curves (Figure F-12) shows that for 2
percent of the time that peak runoff is greater for the contour and terrace
SWCP than for any other SWCP or the base condition. This is not expected,
but can be understood in terms of the flatter slopes of the terrace SWCP not
418
-------
TABLE F-6. P2 FREQUENCY ANALYSIS OF ALTERNATIVE SWCPs
Expected Value
Percent Change from Base
Total Runoff (cm • 10" 2)
Overland Flow (cm • 10~2)
Interflow (cm • 1Q-2)
Sediment Loss
concentration (g/1)
flux (kg/min)
Total Atrazine Flux (g/min)
Atrazine Loss in solution
concentration (mg/1)
flux (g/min)
Atrazine Loss on sediment
concentration (mg/1)
flux (g/min)
Paraquat Loss on sediment
concentration (mg/1)
flux (g/min)
Total Methyl Parathion
Flux (g/min)
Methyl Parathion Loss in
solution
concentration (mg/1)
flux (g/min)
Methyl Parathion Loss on
sediment
concentration (mg/1)
flux (g/min)
Total X Flux (g/min)
in solution
on sediment
Total P Flux (g/m'in)
in solution
on sediment
Base
Condition
1.94
1.87
0.082
No
Tillage
1.90
1.82
0.093
Contours
1.86
1.80
0.085
2.07
4.83
0.0772
0.199
0.0764
0.608
0.0008
50.74
0.322
0.2736
0.1281
0.2418
3.3439
0.0316
8.02
1.95
6.25
2.62
0.186
2.48
1.39
2.42
0.0657
0.169
0.0649
0.553
0.0007
55.27
0.168
0.2258
0.1167
0.2069
3.0886
0.0192
7.66
2.17
5.56
2.02
0.218
1.82
1.56
3.30
0.0642
0.172
0.0637
0.536
0.0006
53.85
0.232
0.2353
0.1280
0.2178
3.3933
0.0183
7.15
2.01
5.19
2.14
0.182
1.97
Terraces
and Contours
1.43
1.32
0.113
1.75
2.48
0.0293
0.108
0.0290
0.456
0.0004
56.04
0.174
0.1596
0.1268
0.1476
3.6092
0.0115
6.60
2.23
4.32
1.81
0.165
1.63
- 2
- 3
14
-33
-50
-15
-15
-15.
- 9
-12
+ 9
-48
-18
- 9
-14
- 8
-39
- 4
+ 11
-11
-23
+ 17
-27
Contours
- 4
- 4
+ 3
-25
-32
-17
-14
-17
-12
-25
+ 6
-28
-14
+ 0
-10
+ 2
-42
-11
+ 3
-17
-18
- 2
-21
Terraces
and Contours
-26
-30
+38
-16
-49
-62
-46
-62
-25
-50
+ 10
-46
-42
- 1
-39
+ 8
-64
-18
+ 14
-31
-31
-11
-34
-------
O
OS
1—I
rt
•P
O
I
ti
CJ
0)
1)
0.20
0.15
0.10
0.05
J i
.70
.90
.10
.30
,50
.70
,90
.10
,30
.50
X
X
X
\
60
0 2 4 6 8 10 12 14 16 18 20 22
% OF TIME GREATER THAN NOTED VALUE
FIGURE F-12. P6 RUNOFF AND SEDIMENT FREntlENCY ANALYSIS
420
-------
NO TILLAGE
CONTOURING
CONTOURING &
10 12 14 16
% OF TIME GREATER THAN NOTED VALUE
22
FIGURE F-13. P6 ATRAZINE FREQUENCY ANALYSIS
421
-------
140
120
BASE CONDITIONS
NO TILLAGE
CONTOURING
6 8 10 12 14 16
% OF TIME GREATER THAN NOTED VALUE
FIGURE F-14. P6 NITROGEN FREQUENCY ANALYSIS
422
-------
BASE CONDITIONS
NO TILLAGE
CONTOURING
CONTOURING &
TERRACING
6 8 10 12 14 16 18
% OF TIME GREATER THAN NOTED VALUE
FIGURE F-15. P6 PHOSPHORUS FREQUENCY ANALYSIS
423
-------
draining as quickly after the spring snowmelt as is the case for the other
SWCPs and the base conditions. Thus, when a large storm event occurs early
in the spring season, higher peak flows are produced by the terraces than
under terraces than under other conditions because the terrace soil moisture
levels are quite high compared to the other practices. Sediment flux is af-
fected in the same way. No tillage gives the largest reduction in sediment
flux and concentration. This reduction in sediment loss is a result of the
decrease in soil fines generated due to residue ground cover.
The frequency curves for total atrazine flux, atrazine flux in solution,
and atrazine concentration in solution are shown in Figure F-13. Both
contours and contours and terraces decrease atrazine washoff as would be
expected. However, no tillage increases atrazine loss. The increase is
not evident, nor may it be significant as the increase in atrazine flux is
less than 10 percent. The increase is most likely the result of one par-
ticular storm event which produced a large volume of runoff soon after the
application of atrazine in the spring. Whether one wants to consider this
increase significant or not, it is an example of how the timing of a
meteorological event together with a SWCP can produce the opposite response
of what is expected and desired.
The nitrogen results in Figure F-14 show a combination of effects. The
terrace SWCP produces the largest total nitrogen flux of the three practices
because of the large volume of interflow which transports nitrogen in
solution from the watershed. The nitrogen flux in solution is also high for
no tillage. This can be related to the large amount of nitrogen left on the
watershed after harvest in the form of crop residue. The amount of organic
N left as residue is approximately 2.5 times greater than the amount when
the residue is removed after harvest (see Table F-2). Much of this organic
N is leached from the residue and becomes soluble in the spring snowmelt
and washoff.
The phosphorus flux curves (Figure F-15) are similar to the nitrogen
flux curves previously discussed. The effect of no tillage on phosphorus
flux is less than that observed for nitrogen flux because PO^ adsorbs onto
soil particles, and thus is not as easily removed by subsurface flow, as
N03.
The area under each curve for the different SWCPs and constituents and
the percent change from base conditions are summarized in Table F-7.
CONCLUSIONS
The purpose of this study is to use the ARM Model to analyze the rela-
tive effectiveness of alternative SWCPs for controlling runoff, sediment loss,
and pesticide and nutrient washoff from small agricultural watersheds. The
results of this study provide interesting insight into the influence various
SWCPs have in altering the hydrologic response of agricultural areas arid how
these practices can be represented in a modeling approach.
It has been shown that, in general, the SWCPs studied in this report
reduce surface runoff, sediment loss, and associated constituent loads.
424
-------
TABLE F-7. WATERSHED P6 FREQUENCY ANALYSIS OF ALTERNATIVE SWCPs
to
en
Expected Value
Base
Condition
Total Runoff
(cms • 10-2)
Overland Flow
(cms • 10-2)
Interflow (cms • 10~2)
Sediment Loss
concentration (g/1)
flux (kg/min)
Total Atrazine Flux
(g/min)
Atrazine Loss in solution
concentration (g/1)
flux (g/min)
Atrazine Loss on sediment
concentration (g/1)
flux (g/min)
Paraquat Loss on sediment
concentration (g/1)
flux (g/min)
Total N Flux (g/min)
in solution
on sediment
Total P Flux (g/min)
in solution
on sediment
1.11
1.05
0.055
1.56
1.74
0.140
0.220
0.136
1.621
0.0048
49.43
0.0995
4.77
2.81
2.09
2.46
0.897
1.57
No
Tillage
1.11
1.04
0.065
0.50
0.51
0.147
0.257
0.145
1.942
0.0032
54.43
0.0300
4.20
3.14
1.13
1.65
0.994
0.67
Contours
1.10
1.05
0.059
1.34
1.49
0.123
0.222
0.119
1.666
0.0036
51.00
0.0877
4.48
2.66
1.89
2.25
0.827
1.43
Contours $
Terraces
1.00
0.91
0.074
1.74
1.56
0.052
0.104
0.050
1.102
0.0023
52.35
0.0930
4.89
2.91
1.94
2.22
0.707
1.50
Percent
No
Tillage
+ 0
- 1
+10
-68
-71
+ 5
+17
+ 7
+20
-33
+10
-70
-12
+12
-46
-33
+11
-57
Change from
Contours
+ 0
+ 0
+ 0
-14
-14
-12
+ 1
-12
+ 3
-25
+ 3
-12
- 6
- 5
-10
- 8
- 8
- 9
Base
Contours §
Terraces
-10
-13
+24
-12
-10
-63
-53
-63
-32
-52
+ 6
- 6
+ 2
+ 4
- 7
-10
-21
- 4
-------
However, notable exceptions do exist. Practices that reduce and retard
overland flow tend to increase subsurface flow components. On both water-
sheds interflow was increased for each SWCP with the largest increase of 38
percent computed on the P2 watershed for terraces plus contours. This can re-
sult in an increase in soluble pollutant washoff. The soluble nitrogen re-
sults (mostly N03) from the P2 watershed demonstrate this increase. Soluble
NH^ and PO^ do not exhibit such a strong relationship, but increase for no
tillage and decrease for contours with and without terraces. Atrazine and
methyl parathion loss occur mostly in solution form and decrease with the
different SWCPs in relation to their decrease in total runoff. The adsorp-
tion characteristics appear to minimize any increase in the removal of
soluble pollutants. This is not true for non-adsorbing constituents such
as N03, which can substantially increase with subsurface flow.
Sediment and associated constituents decrease for all SWCPs studied.
The percent changes for sediment-associated pollutants are generally far
larger than for water-related pollutants. The difference can be explained
by the fact that the original intent of the three SWCPs studied (no tillage,
contours, and contours with terraces) has been to reduce erosion and soil
loss from croplands. All three SWCPs appear to be well designed for this
purpose.
In this study, we found that terraces with contours is the most effec-
tive SWCP for most constituents on both watersheds studied. The contour
SWCP and the no tillage SWCP can be considered equally effective on the two
watersheds. No tillage is generally better than contours in reducing sedi-
ment-associated pollutants. Contours was found to be more effective in
decreasing soluble constituents. However, the individual practices defined
by the SWCPs vary across the United States, and only two regions are re-
presented in this study. In actual practice various combinations of SWCPs
are often used and the specific names are applied to widely differing prac-
tices across the country.
No detailed attempt has been made to compare the SWCP results for the
two watersheds simulated in this study. However, with one watershed (P2)
representing the Piedmont region of the southeastern United States and the
other (P6) located in the Great Lakes Basin some obvious differences can be
found. The major hydrologic and climatic difference between the two water-
sheds is that P6 is snow covered much of the winter. P2 rarely receives
snow. This difference is important in evaluating the effectiveness of
the SWCPs. Snowfall is a much less intense form of precipitation than
rainfall, especially thunderstorm rainfall commonly produced in the south-
east. Thus, for the months of the year when snow provides a protective
manner, regions of the country which rarely have intensive rainfall, such
as the Pacific Northwest, will .not benefit as much from SWCPs as areas
subject to heavy thunderstorm activity.
Parameter changes in the ARM Model to represent the alternative SWCPs
were based on information in the literature and our own judgment and ex-
perience in applying the ARM Model on other watersheds. When little or no
information was available, we tended to be conservative in the parameter
adjustments so as not to overstate the impacts of the different practices.
426
-------
Also, limitations within the model and lack of data prevented consideration
of other effects such as changes in infiltration characteristics, crop
growth, nutrient transformation, etc. Research is needed to define relation-
ships between model parameters and measurable watershed characteristics,
and additional data collection is needed to characterize the impacts of
various SWCPs as a basis for parameter adjustments.
Although the ARM Model is by no means exact in its representation of all
soil processes, this work has shown that the model can be used to analyze
the relative effects of some alternative SWCPs. Further research is needed
to better represent erosion processes, the effects of tillage operations,
the transport of soluble substances, pesticide adsorption and degradation
mechanisms, and nutrient transformations. However, for most processes the
model includes our best current understanding of the important mechanisms.
With an awareness of the changes affected by a SWCP and the representation
of these changes by the model the important effects of a SWCP can be
investigated.
ACKNOWLEDGEMENTS
This study and report were funded by Cornell University in conjunction
with a research grant from the U.S. Environmental Protection Agency. Dr.
Raymond C. Loehr and Dr. Douglas A. Haith and their staff at the Department
of Agricultural Engineering at Cornell University provided ideas, suggestions,
and comments on our work which were invaluable to the successful completion
of this study. In addition, Mr. Lee .Mulkey and Mr. Charles Smith of the
Environmental Research Laboratory in Athens, Georgia were very helpful in
reviewing some of the ideas and assumptions used in the modeling of methyl
parathion on the P2 watershed and our concepts of how to represent base
conditions for that watershed.
At Hydrocomp, Mr. Douglas Beyerlein was project manager and was responsi-
ble for the use of the ARM Model for simulation of soil and water conserva-
tion practices and the content of this report. Mr. Anthony S. Donigian and
Dr. Norman Crawford provided technical guidance and together with Mr.
Beyerlein represented Hydrocomp, Inc. at Cornell's Panel of Experts dis-
cussions. Mr. Harley Davis, Jr. provided the input for and helped review
the nutrient simulation. Mr. Kevin Gartner assembled the meteorologic data
required for the 10-yr simulations and conducted the mean annual value and
frequency analysis. Mr. Tamas Eger and Mr. Gartner are responsible for the
computer programs which produced these analyses from the ARM Model output.
The production of this report would not be possible without the able editing
of Ms. Donna Mitchell, the swift typing of Ms. Kathy Francies, and the
clear graphics of Mr. Guy Funabiki.
427
-------
REFERENCES FOR APPENDIX F
Chow, V.T. 1964. Handbook of applied hydrology. McGraw-Hill Book Co. New
York, New York.
Crawford, H.H. and R.K. Linsley. 1966. Digital simulation in hydrology:
Stanford Watershed Model IV. Department of Civil Enginnering, Stanford
University. Stanford, California. Technical Report No. 39. 210 p.
Crawford, N.H. and A.S. Donigian, Jr. 1973. Pesticide transport and runoff
model for agricultural lands. EPA 660/2-74-013. Southeast Environment-
al Research Laboratory. U.S. Environmental Protection Agency. Athens,
Georgia. 211 p.
Davidson, J.M., R.S. Mansell, and D.R. Baker. 1973. Herbicide distributions
within a soil profile and their dependence upon adsorption-desorption
Soil Crop Sci. Soc. Florida Proc. 26 p.
Donigian, A.S., Jr. and N.H. Crawford. 1976. Modeling pesticides and nutri-
ents on agricultural lands. Environmental Research Laboratory. Envir-
onmental Protection Agency. Athens, Georgia. EPA 600/3-76-043.
Donigian, A.S., Jr., D.C. Beyerlein, H.H. Davis, and N.H. Crawford. 1977.
Agricultural runoff management (ARM) model version II: refinement and
testing. EPA 600/3-77-098. Environmental Research Laboratory. U.S.
Environmental Protection Agency. Athens, Georgia. 293 p.
Ellis, B.C., and A.E. Erickson, and A.R. Wolcott. 1978. Nitrate and phos-
phorus runoff losses from small watersheds in Great Lakes Basin.
EPA-600/3-78-028. Environmental Research Laboratory. U.S. Environ-
mental Protection Agency. Athens, Georgia. 84 p.
Fleming, G. and K.M. Leytham. 1976. The hydrologic and sediment processes
in natural watershed areas. Proceedings of the Third Federal Inter-
Agency Sedimentation Conference. Denver, Colorado. Marth 22-25, 1976.
Gunther, F.A. 1977. The citrus reentry problem: Research on its causes
and effects, and approaches to its minimization. Residue Reviews.
Volume 67. Springer-Verlag. New York, New York.
Hagin, J. and A. Amberger. 1974. Contribution of fertilizers and manures
to the N and P load of waters. A computer simulation. Report submit-
ted to the Deutsche Forschungs Gemeinschaft. 1974. 123 p.
Hubbard, R.K. 1975. The vertical and horizontal redistribution of nitrogen,
chloride, and phosphorus by precipitation and surface runoff on two
similar watersheds. Master's thesis. Department of Crop and Soil
Sciences. Michigan State University. East Lansing, Michigan. 122 p.
Hydrocomp, Inc. 1976. Hydrocomp simulation program operations manual. 4th
Edition. Palo Alto, California. 115 p.
428
-------
Hydrocomp, Inc. 1978. Sediment loading and agricultural practices in the
Kishwaukee and Thorn Creek watersheds. Prepared for the Northeastern
Illinois Planning Commission. Hydrocomp, Inc. Palo Alto, California.
45 p.
Kraeger, B.A. 1971. Stochastic monthly streamflow by multi-station daily
rainfall generation. Technical Report 152. Department of Civil
Engineering, Stanford University. Stanford, California.
Mehran, M., and K.K. Tanji. 1974. Computer modeling of nitrogen transform-
ations in soils. J. Environ. Qual. 3(4):291-395.
Menzie, C.M. 1972. Fate of pesticides in the environment. Annual Review of
Entomology. 17:199-222.
Meyer, L.D. and W.H. Wischmeier. 1969. Mathematical simulation of the
process of soil erosion by water. Trans. Am. Soc. Agric. Eng.
12(6):754-758, 762.
Negev, M.A. 1967. A sediment model on a digital computer. Department of
Civil Engineering, Stanford University. Stanford, California. Tech-
nical Report No. 76. 109 p.
Onstad, C.A. and G.R. Poster. 1975. Erosion modeling on a watershed. Trans.
Am. Soc. Agric. Eng. 18(2):288-292.
Smith, C.N., R.A. Leonard, G.W, Langdale, and G.W. Bailey. 1977. Transport
of agricultural chemicals from small upland piedmont watersheds. U.S.
Environmental Protection Agency, Athens, Georgia and U.S. Department of
Agriculture, Watkinsville, Georgia. Final report on Agreement No.
D6-0381. (In preparation).
Stewart, B.A., D.A. Woolhiser, W.H. Wischmeier, J.H. Caro, and M.H. Frere.
1975. Control of water pollution from cropland. Volume I: A manual
for guideline development. Agricultural Research Service. U.S. Depart-
ment of Agriculture. Washington, D.C. ARS-H-5-1. EPA-600/2-75-
026a. Ill p.
Wischmeier, W.H. and D.D. Smith. 1965. Predicting rainfall erosion losses
from cropland east of the Rocky Mountains. Department of Agriculture.
Agricultural Handbook No. 282. 47 p.
P2 and P6 PARAMETER VALUES
The P2 and P6 parameter values used in the ARM Model are presented in
the form of input sequences to the model. With each major pollutant
simulated (atrazine, paraquat, methyl parathion, and nutrients) are the
input sequences for the four practices simulated (base condition, minimum
tillage, contours, and terraces and contours).
429
-------
P2 Parameter Values
Atrazine--
P-2 WATKENSVILLE, GA. : BASE CONDITIONS (NO SWCP)
ATRAZINE APPLIED: 2.6 LBS/AC EVERY DAY 129 FOR THE 10-YR PERIOD
HYCAD=CALB
INPUT=ENGL
OUTPUT=ENGL
PRINT=INTR
PEST=YES
NUTR=NO
ICHECKOFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 3.20
BGNDAY= 2 BGNMOSN 1 BGNYR= 1966
ENDDAY=31 ENDMON=12 ENDYR= 1975
UZSN= 0.42 UZS= 0.420 LZSN= 18.00 LZS= 18.00
L= 100. SS= 0.0250 NN= 0.2000 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.100 0.100 0.100 0.100 0.000 0.150 0.600 0.850 0.750 0.100 0.100 0.100
TIMTIL= 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=1.0
YRTID= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.600 SRERI= 2.00 SCMPAC 0.02
PESTICIDE
APMCDE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= O.OC PSL2= 0.00 PSGZ= 0.0
TIMAP= 129 129 129 129 129 129 129 129 129 129 000 000 NPA= 10 NDRPA= 2
YEARAP= 66 67 63 69 70 71 72 73 74 75 99 99
SSTR= 2.600 2.600 2.600 2.600 2.600 2.600 2.600 2.600 2.600 2.600 0.000 0.000
CMAX= 3.5E-5DD= 0.0000 K= 2.000 N= 1.0000 NP= 2.300
DDG= 129 138 129 138 129 138 129 138 129 138 129 138
YDG= 66 66 67 67 68 68 69 69 70 70 71 71
KDG= 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040
SZDPTH=0. 1250 UZDPTH= 6.125 BD3Z= 99.90 BDUZ= 99.90 BDLZ= 99.9 UZF=3.0 LZF=1.5
430
-------
P-2 WATKINSVILLE, GA.: SWCP — NO TILLAGE
ATRAZINE APPLIED: 2.6 IBS/AC EVERY DAY 129 FOR TOE 10-YR PERIOD
HYCAL=CALB
INPtfT=ENGL
OUTPUT=ENGL
PRINT=INTR
SNOW=NO
PEST=YES
NOTR=NO
DISK=*K>
INTRVL=15 HVMIN= 0.100 AREA= 3.20
BGSQAY= 2 BGNMON= 1 BGNYR= 1966
ENDQAY=31 ENDMON=12 ENDYR= 1975
UZSN= 0.42 UZS= 0.420 LZSN= 18.00 LZS= 18.00
L= 100. SS= 0.0250 NN= 0.3200 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.600 0.600 0.600 0.600 0.400 0.490 0.760 0.910 0.850 0.600 0.600 0.600
TIMTIL= 115 115 115 115 115 115 115 1.15 115 115 0 0 SMPF=1.0
YKTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SREKTL= 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.550 SRERI= 0.20 SCMPAC 0.02
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSIZ= 0.00 PSGZ= 0.0
TIMAP= 129 129 129 129 129 129 129 129 129 129 000 000 NPA= 10 NDRPA= 2
YEARAP= 66 67 68 69 70 71 72 73 74 75 99 99
SSTR= 2.600 2.600 2.600 2.600 2.600 2.600 2.600 2.600 2.600 2.600 0.000 0.000
CMAX= 3.5E-5 DD= 0.0000 K= 2.000 N= 1.0000 NP= 2.300
DDG= 129 138 129 138 129 138 129 138 129 138 129 138
YDG= 66 66 67 67 68 68 69 69 70 70 71 71
KDG= 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ= 99.9 UZF=3.0 LZF=1.5
431
-------
P-2 WATKINSVILLE, GA.: SWCP — CONTOURING
ATRAZINE APPLIED: 2.6 LBS/AC EVERY DAY 129 FOR THE 10-YR PERIOD
HYCAL=CALB
INPOT=ENGL
OUTPUT=ENGL
PRINT=INTR
SNOW=NO
PEST=YES
NOTR=NO
ICHECKOFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 3.20
BQODAY= 2 BO!MCN= 1 BGNYR= 1966
ENDDAY=31 ENDMON=12 ENDYR= 1975
UZSN= 0.50 UZS= 0.500 LZSN= 18.00 LZS= 18.00
L= 100. SS= 0.0250 NN= 0.25QO A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= l.COO KK24= 0.600 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.100 0.100 0.100 0.100 0.000 0.150 0.600 0.850 0.750 0.100 0.100 0.100
TIMTIL= 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=0.5
YRTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTIr= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.500 SRERI= 2.00 SCMPAC 0.02
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSI2= 0.00 PSGZ= 0.0
TIMAP= 129 129 129 129 129 129 129 129 129 129 000 000 NPA= 10 NDRPA= 2
YEARAP= 66 67 68 69 70 71 72 73 74 75 99 99
SSTR= 2.500 2.600 2.600 2.600 2.600 2.600 2.600 2.600 2.600 2.600 0.000 0.000
CMAX= 3.5E-5 DD= 0.0000 K= 2.000N= 1.0000NP= 2.300
DDG= 129 138 129 138 129 138 129 138 129 138 129 138
YDG= 66 66 67 67 68 68 69 69 70 70 71 71
KDG= 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ= 99.9 UZF=3.0 LZF=1.5
432
-------
P-2 WATKINSVILLE, GA.: SWCP - TERRACES (2) WITH CONTOURING
ATRAZINE APPLIED: 2.6 LBS/AC EVERY DAY 129 FOR THE 10-YR PERIOD
HYCAL=CAL8
INPOT=ENGL
OUTPUT=ENGL
PRINT=INTR
SNOW=NO
PEST=YES
NOTR=NO
ICHECK=OFF
DISK=NO
IOTRVL=15 HYMIN= 0.100 AREA= 3.20
BGNEAY= 2 B<3QMON= 1 BGNYR= 1966
ENDDAY=31 ENEMCN=12 ENDYR= 1975
UZSN= 0.65 UZS= 0.650 LZSN= 18.00 L2S= 18.00
L= 250. SS= 0.0150 NN= 0.2500 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SOJ= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.100 0.100 0.100 0.100 0.000 0.150 0.600 0.850 0.750 0.100 0.100 0.100
TIMTII/= 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=0.35
YKTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTD= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.500 SRERI= 2.00 SCMPAC 0.02
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSIZ= 0.00 PSGZ= 0.0
TIMAP= 129 129 129 129 129 129 129 129 129 129 000 000 NPA= 10 NDRPA= 2
YEARAP= 66 67 68 69 70 71 72 73 74 75 99 99
SSTR= 2.600 2.600 2.600 2.600 2.600 2.600 2.600 2.600 2.600 2.600 0.000 0.000
CMAX= 3.5E-5 DD= 0.0000 K= 2.000N= 1.0000 NP= 2.300
DDG= 129 138 129 138 129 138 129 138 129 138 129 138
YDG= 66 66 67 67 68 68 69 69 70 70 71 71
KDG= 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040 0.100 0.040
SZDPTH=0.1250 02DPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ= 99.9 UZF=3.0 LZF=1.5
433
-------
Paraquat--
P-2 WATKINSVILLE, GA.: BASE CONDITIONS (NO SWCP)
PARAQUAT APPLIED: 2.25 LBS/AC EVERY DAY 129 FOR THE 10-YR PERIOD
HYCAL=CALB
INPOT=ENGL
OOTPOT=ENGL
PRINT=INTR
SNOW=NO
PEST=YES
NOTR=NO
ICHECKOFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 3.20
BGNDAY= 2 BGNMON= 1 BGNYR= 1966
ENDDAY=31 ENEMON=12 ENDYR= 1975
UZSN= 0.42 UZS= 0.420 LZSN= 18.00 LZS= 18.00
I> 100. SS= 0.0250 NN= 0.2000 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SGW= O.OOOGWS= 0.000 KV= 0.000 ICS= O.OOOOFS= 0.000 IFS= 0.000
COVPMO= 0.100 0.100 0.100 0.100 0.000 0.150 0.600 0.850 0.750 0.100 0.100 0.100
TIMTID= 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=1.0
YKTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.600 SRERI= 2.00 SCMPAC 0.02
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSDZ= 0.00 PSI£= 0.00 PSGZ= 0.0
TIMAP= 129 129 129 129 129 129 129 129 129 129 000 000 NPA= 10 NDRPA= 1
YEARAP= 66 67 68 69 70 71 72 73 74 75 99 99
SSTR= 2.250 2.250 2.250 2.250 2.250 2.250 2.250 2.250 2.250 2.250 0.000 0.000
CMAX= l.OE-5 DD= 0.0003 K= 120.0 N= 2.0000 NP= 4.600
DDG= 129
YDG= 66
KDG= 0.002
SZDPTH=0.1250 UZDPTH= 6.125 BD3Z= 99.90 BDUZ= 99.90 BDLZ= 99.9 UZF=3.0 LZF=1.5
434
-------
P-2 WATKINSVILLE, GA.: SWCP — NO TILLAGE
PARAQUAT APPLIED: 2.25 LBS/AC EVERY DAY 129 FOR THE 10-YR PERIOD
HYCAIXMB
INPUT=eNGL
OOTPUT=ENGL
PRINT=INTR
SNOW=NO
PEST=YES
NOTR=NO
ICHECKOFF
DISK=tTO
INTRVL=15 HYMIN= 0.100 AREA= 3.20
BGNDAY= 2 BGNMCN= 1 BQWR= 1966
ENDDAY=31 ENCMON=12 ENDYR= 1975
UZSN= 0.42 UZS= 0.420 LZSN= 18.00 LZS= 18.00
L= 100. SS= 0.0250 NN= 0.3200 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SQW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COWMO= 0.600 0.600 0.600 0.600 0.400 0.490 0.760 0.910 0.850 0.600 0.600 0.600
TIMTIL= 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=1.0
YRTIL= 66 67 63 69 70 71 72 73 74 75 99 99
SRERTL= 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER=. 1.700 KSER= 0.550 SRERI= 0.20 SCMPAC 0.02
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSLZ= 0.00 PSGZ= 0.0
TIMAP= 129 129 129 129 129 129 129 129 129 129 000 000 NPA= 10 NDRPA= 1
YEARAP= 66 67 68 69 70 71 72 73 74 75 99 99
SSTR= 2.250 2.250 2.250 2.250 2.250 2.250 2.250 2.250 2.250 2.250 0.000 0.000
CMAX= l.OE-5 DD= 0.0003 K= 120.0 N= 2.0000 NP= 4.600
DDG= 129
YDG= 66
KDG= 0.002
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ= 99.9 UZP=3.0 LZF=1.5
435
-------
P-2 WATKINSVILLE, GA.: SWCP — CONTOURING
PARAQUAT APPLIED: 2.25 LBS/AC EVERY DAY 129 FOR THE 10-YR PERIOD
HYCAL=CALB
INPOT=ENGL
OUTPUT=ENGL
PRINT=INTR
SNCW=NO
PEST=YES
NUTR=NO
ICHECKOFF
DISK=flO
INTKVL=15 HYMIN= 0.100 AREA= 3.20
BGNDAY= 2 BGNMCN= 1 BGNYR= 1966
ENDDaY=31 ENIMCN=12 ENDYR= 1975
UZSN= 0.50 UZS= 0.500LZSN= 18.00 LZS= 18.00
L= 100. SS= 0.0250 NN= 0.2500 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.100 0.100 0.100 0.100 0.000 0.150 0.600 0.850 0.750 0.100 0.100 0.100
TIMTIL= 115 115 115 115 115 115 115 115 115 115 0 OSMPF=0.5
YRTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.500 SRERI= 2.00 SCMPAC 0.02
PESTICIDE
APMCDE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSLZ= 0.00 PSGZ= 0.0
TIMAP= 129 129 129 129 129 129 129 129 129 129 000 000 NPA= 10 NDRPA= 1
YEARAP= 66 67 68 69 70 71 72 73 74 75 99 99
SSTR= 2.250 2.250 2.250 2.250 2.250 2.250 2.250 2.250 2.250 2.250 0.000 0.000
CMAX= l.OE-5 DD= 0.0003 K= 120.0 N= 2.0000 NP= 4.600
DDG= 129
YDG= 66
KDG= 0.002
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ= 99.9 UZF=3.0 LZF=1.5
436
-------
P-2 WATKINSVILIE, GA. : SWCP - TERRACES (2) WITH CONTOURING
PARAQUAT APPLIED: 2.25 LBS/AC EVERY IAY 129 FOR THE 10-YR PERIOD
HYCAL=CALB
INPUT=ENGL
OUTPOT^NGL
PRINT=INTR
SNOW=NO
PEST=YES
ICHECKOFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 3.20
BGNDAY= 2 BGOT.1ON= 1 BGNYR= 1966
ENDEAY=31 ENCMON=12 ENDYR= 1975
UZS»= 0.65 UZS= 0.650 LZSN= 18.00 LZS= 18.00
L= 250. SS= 0.0150 NN= 0.2500 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV^= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.100 0.100 0.100 0.100 0.000 0.150 0.600 0.850 0.750 0.100 0.100 0.100
TIMTID* 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=0.35
YRTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.500 SRERI= 2.00 SCMPAC 0.02
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSI2= 0.00 PSGZ= 0.0
0 NPA= 10 NDRPA= 1
9
0.000
TIMAP=
129 129 129
YEARAP=
SSTR=
CMAX=
DDG=
YDG=
KDG=
2
1.
66 67 68
.250
OE-5
2.250
DD=
129 129 129
69 70 71
2.250
2.250
0.0003 K=
129 129
72 73
2.250
2.
120.
129 129 000 000 NPA= 10 NDRPA= 1
74 75 99 99
250 2.
0 N=
250 2.250 2.250
2.0000 NP=
2.250 0.
4.600
000
129
0
66
.002
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ= 99.9 UZF=3.0 LZF*1.5
437
-------
Methvl Parathion--
P-2 WATKINSVILLE, GA. : BASE CONDITIONS (NO SWCP)
METHYL PARATHION APPLIED: 1.0 LB/AC 6 TIMES PER YR FOR 10-YR PERIOD
HYCAL=CALB
INPOT=ENGL
OUTPUT=ENGL
PRINT=INTR
SNOW=NO
PEST=YES
NOTR=NO
ICHECK=OPF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 3.20
BOIDAY= 2 BGNMON= 1 BGNYR= 1966
ENDDAY=31 ENDMGN=12 ENDYR= 1975
UZSN= 0.42 UZS= 0.420 LZSN= 18.00 LZS= 18.00
L= 100. SS= 0.0250 NN= 0.2000 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.200 0.200 0.200 0.200 0.000 0.100 0.500 0.750 0.650 0.550 0.200 0.200
TIMTIL= 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=1.0
YRTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.600 SRERI= 2.00 SCMPAC 0.02
PESTICIDE
APMODE=CROP
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSLZ= 0.00 PSGZ= 0.0
TIMAP= 166 180 194 208 222 236 166 180 194 208 222 236 NPA= 60 NDRPA= 1
YEARAP= 66 66 66 66 66 66 67 67 67 67 67 67
SSTR= 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
CMAX= 4.0E-5 DD= 0.0000 K= 25.000 N= 1.0000NP= 1.000
PSCZ= 0.0 PCK= 1.16 KCDG= 0.087
DDG= 166 166 166 166 166 166 166 166 166 166 166 166
YDG= 66 67 68 69 70 71 72 73 74 75 00 00
KDG= .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ= 99.9 UZF»3.0 LZF=1.5
438
-------
P-2 WATKINSVILLE, GA.: SWCP — NO TILLAGE
METHYL PARATHION APPLIED: 1.0 LB/AC 6 TIMES PER YR FOR 10-YR PERIOD
HYCAL=CALB
INPtTT=€NGL
COTPOT=ENGL
PRIOT=IOTP
SNOW=NO
PEST=YES
NOTR=NO
ICHECKOFF
DISK=NO
INTWL=15 HYMDJ= 0.100 AREA= 3.20
BOIDAY= 2 BQWON= 1 BGNYR= 1966
ENDDAY=31 END>1ON=12 ENDYR= 1975
UZSN= 0.42 UZS= 0.420 LZSN= 18.00 LZS= 18.00
L= 100. SS= 0.0250 NN= 0.3200 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SGW= 0.000 OJS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.200 0.200 0.200 0.200 0.200 0.280 0.600 0.800 0.720 0.640 0.200 0.200
TIMTIL= 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=1.0
YKTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.000 0.000
JRER= 1.9 KRER= 0..160 JSER= 1.700 KSER= 0.550 SRERI= 0.20 SCMPAC 0.02
PESTICIDE
APMODE=CPDP
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSLZ= 0.00 PSGZ= 0.0
TIMAP= 166 180 194 208 222 236 166 180 194 208 222 236 NPA= 60 NDRPA= 1
YEARAP= 66 66 66 66 66 66 67 67 67 67 67 67
SSTR= 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
CMAX= 4.0E-5 DD= 0.0000 K= 25.000 N= 1.0000 NP= 1.000
PSCZ= 0.0 PCK= 1.16 KCDG= 0.087
DDG= 166 166 166 166 166 166 166 166 166 166 166 166
YDG= 66 67 68 69 70 71 72 73 74 75 00 00
KDG= .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ= 99.9 UZF=3.0 LZF=1.5
439
-------
P-2 WATKINSVILLE, GA.: SWCP — CONTOURING
METHYL PARATHION APPLIED: 1.0 LB/AC 6 TIMES PER YR FOR 10-YR PERIOD
HYCAL=€ALB
INPOT=ENGL
OUTPOT=ENGL
PRINT=INTR
SNCW=NO
PEST=YES
NOTR=*JO
ICHECK=OFF
DISK=tJO
INTRVL=15 HYMIN= 0.100 AREA= 3.20
BGNDAY= 2 BGNMCN= 1 BGNYR= 1966
ENDDAY=31 ENCMCN=12 ENDYR= 1975
UZSN= 0.50 UZS= 0.500 LZSN= 18.00 LZS= 18.00
L= 100. SS= 0.0250 NN= 0.2500 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.200 0.200 0.200 0.200 0.000 0.100 0.500 0.750 0.650 0.550 0.200 0.200
TIMTIL= 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=0.5
YRTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.500 SRERI= 2.00 SCMPAC 0.02
PESTICIDE
APMODE=CROP
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSLZ= 0.00 PSGZ= 0.0
TIMAP= 166 180 194 208 222 236 166 180 194 208 222 236 NPA= 60 NDRPA= 1
YEABAP= 66 66 66 66 66 66 67 67 67 67 67 67
SSTR= 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
CMAX= 4.0E-5 DD= O.OOOOK= 25.000 N= 1.0000 NP= 1.000
PSCZ= 0.0 PCK= 1.16 KCDG= 0.087
DDG= 166 166 166 166 166 166 166 166 166 166 166 166
YDG= 66 67 68 69 70 71 72 73 74 75 00 00
KD<3= .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ= 99.9 UZF=3.0 LZF=1.5
440
-------
P-2 WATKINSVILLE, GA. : SWCP - TERRACES (2) WITH CONTOURING
METHYL PARATHION APPLIED: 1.0 LB/AC 6 TIMES PER YR FOR 10-YR PERIOD
HYCAL=CALB
INPOT=ENGL
OOTPOT=ENGL
PRINT=INTR
SNOW=NO
PEST=YES
ICHECKOFF
DISK=«O
INTRVL=15 HYMIN= 0.100 AREA= 3.20
BGSIDAY= 2 BGNMON= 1 BOIYR= 1966
ENDDAY=31 ENEMON=12 ENDyR= 1975
UZSN= 0.65 UZS= 0.650 LZSN= 18.00 LZS= 18.00
L= 250. SS= 0.0150 NN= 0.2500 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SQti= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.200 0.200 0.200 0.200 0.000 0.100 0.500 0.750 0.650 0.550 0.200 0.200
TIM-TIL- 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=0.35
YRTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.500 SRERI= 2.00 SCMPAC 0.02
PESTICIDE
APMODE=CROP
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSLZ= 0.00 PSGZ= 0.0
TIMAP= 166 180 194 208 222 236 166 180 194 208 222 236 NPA= 60 NDRPA= 1
YEARAP= 66 66 66 66 66 66 67 67 67 67 67 67
SSTR= 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
CMAX= 4.0E-5 DD= O.OOOOK= 25.000 N= 1.0000 NP= 1.000
PSCZ= 0.0 PCK= 1.16 KCDG= 0.087
DDG= 166 166 166 166 166 166 166 166 166 166 166 166
YDG= 66 67 68 69 70 71 72 73 74 75 00 00
KDG= .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154 .0154
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ* 99.90 BDLZ= 99.9 UZF=3.0 LZF=1.5
441
-------
Nutrients--
P-2 WATKINSVILLE, GA. : BASE CONDITIONS (NO SWCP)
NUTRIENTS APPLIED: N & P ON DAY 119; N ON DAY 162; PLANT DECAY AFTER HARVEST
HYCAL=CALB
INPUT=ENGL
OUTPOT=ENGL
PRINT=INTR
SNOW=NO
PEST=NO
NUTR=YES
ICHECKOFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 3.20
BGNDAY= 2 BGNMON= 1 BGNYR= 1966
ENDDAY=31 ENCMON=12 ENDYR= 1975
UZSN= 0.42 UZS= 0.420 LZSN= 18.00 LZS= 18.00
L= 100. SS= 0.0250 NN= 0.2000 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SQW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.100 0.100 0.100 0.100 0.000 0.150 0.600 0.850 0.750 0.100 0.100 0.100
TIMTIL= 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=1.0
YRTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.600 SRERI= 2.00 SCMPAC 0.02
NUTRIENT
TSTEP= 360 NAPPL= 3 TIMHAR= 297
ULUPTK=0.00 0.05 0.08 0.10 0.15 0.75 0.200 0.07 0.00 0.00 0.00 0.0
LZUPTK=0.00 0.00 0.00 0.05 0.03 0.55 0.450 0.10 0.00 0.00 0.00 0.0
REACTION RATES
NITROGEN
SURFACE
1.00
UPPER
0.20
LOWER
0.10
0.
ZONE
0.
ZONE
0.
0
006
002
0.
0
0
10
.13
.025
0.
0.
0
00
002
.002
0.0
0.00
0.0
0.
0
0.0
0.0
1.0
1.0
1.0
.20
0.25
0.2
GROUNOTATER
0.0
0.
TEMPERATURE
1.05
1.
0
0.
0
0.0
0.0
0.
0
0.0
0.0
COEFFICIENTS
07
1.
07
1.07
1.07
1.
07
1.05
1.05
PHOSPHORUS
SURFACE
0.02 0.0
UPPER ZONE
0.002 0.0
LOWER ZONE
0.002 0.0
GROUNDWATER
0.0 0.0
0.01
0.70
0.80
0.0
1.0
1.0
1.0
0.0
0.015
0.0015
0.005
0.0
TEMPERATURE COEFFICIENTS
1.07 1.07
END
1.07
1.05
1.05
442
-------
INITIAL
NITROGEN
SURFACE
90.0 2.0 20.0 0.20 0.0
UPPER ZONE
375.0 4.0 40.0 3.00 0.0
DOWER ZONE
2500. 6.0 20.0 212. 0.0
GROUNDWATER
0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
12.0 1.0 36.0 0.0
UPPER ZONE
100. 0.35 305.0 0.0
DCWER ZONE
800. 1.0 135.0 0.0
GROUNDWATER
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
3.46
UPPER ZONE
79.54
LOWER ZONE
770.
GROUNDWATER
0.0
END
0.0
0.0
0.0
0.0
APPLICATION 119
NITROGEN
SURFACE
0.0 0.71 0.0 0.0 0.0 0.0
UPPER ZONE
0.0 33.29 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.0 0.2 0.40 0.0
UPPER ZONE
0.0 28.9 0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END i
443
-------
APPLICATION 162
NITROGEN
SURFACE
0.0 25.0 3.0 0.0 0.0 0.0
UPPER ZONE
0.0 61.9 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.0 0.0
UPPER ZONE
0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
0.0 0.0
0.0 0.0
APPLICATION 297
NITROGEN
SURFACE
1.5 0.0 0.0 0.0 0.0 0.0
UPPER ZONE
20.0 0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.3 0.0 0.0 0.0
UPPER ZONE
4.0 0.0 0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
LZTEMP= 48.0 47.0 53.0 57.0 65.0 73.0 77.0 77.0 76.0 71.0 58.0 51.0
ASZT= -21.4 BSZT= 1.345 AUZT= 26.210 BUZT= 0.675
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ=99.90 UZF=5.0 LZF=1.0
444
-------
P-2 WATKINSVILLE, GA.: SWCP — NO TILLAGE
NUTRIENTS APPLIED: N & P ON DAY 119; N ON DAY 162; PLANT DECAY AFTER HARVEST
HYCAL=CALB
INPUT=ENGL
OUTPUT=ENGL
PRINT=INTR
SNOW=NO
PEST=NO
NUTR=YES
ICHECKOFF
DISK=NO
INTKVL=15 HYMIN= 0.100 AREA= 3.20
BGNDAY= 2 BGNMCN= 1 BGNYR= 1966
ENDDAY=31 ENCMGN=12 ENDYR= 1975
UZSN= 0.42 UZS= "0.420 LZSN= 18.00 LZS= 18.00
L= 100. SS= 0.9250 NN= 0.3200 A= 0.0000 EPXM=0.1200 PETOL 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.600 0.600 0.600 0.600 0.400 0.490 0.760 0.910 0.850 0.600 0.600 0.600
TIMTIL= 115 115 115 115 115 115 .115 115 115 115 0 0 SMPF=1.0
YKTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.150 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.550 SRERI= 0.20 SCMPAC 0.02
NUTRIENT
TSTEP= 360 NAPPL= 3 TIMHAR= 297
ULUPTK=0.00 0.05 0.08 0.10 0.15 0.75 0.200 0.07 0.00 0.00 0.00 0.0
LZUPTK=0.00 0.00 0.00 0.05 0.03 0.55 0.450 0.10 0.00 0.00 0.00 0.0
REACTION RATES
NITROGEN
SURFACE
1.00 0.0
UPPER ZONE
0.20 0.006
LOWER ZONE
0.10 0.002
GROUNDWATER
0.0 0.0
0.10
0.13
0.025
.
0.0
0.00
0.002
0.002
0.0
0.0
0.00
0.0
0.0
0.0
0.0
• 0.0
0.0
1.0
1.0
1.0
0.0
.20
0.25
0.2
0.0
TEMPERATURE COEFFICIENTS
1.05 1.07
1.07
1.07
1.07
1.07
1.05
1.05
PHOSPHORUS
SURFACE
0.02 0.0
UPPER ZONE
0.002 0.0
LOWER ZONE
0.002 0.0
GROUNCWATER
0.0 0.0
0.01 1.0
0.70 1.0
0.80 1.0
0.0 0.0
0.015
0.0015
0.005
0.0
TEMPERATURE COEFFICIENTS
1.07 1.07
END
1.07 1.05
1.05
445
-------
INITIAL
NITROGEN
SURFACE
90.0 2.0 20.0 0.20 0.0
UPPER ZONE
375.0 4.0 40.0 3.00 0.0
LOWER ZONE
2500. 6.0 20.0 212. 0.0
GROUNDNATER
0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
12.0 1.0 36.0 0.0
UPPER ZONE
100. 0.35 305.0 0.0
LOWER ZONE
800. 1.0 135.0 0.0
GROUNDWATER
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
3.46
UPPER ZONE
79.54
LOWER ZONE
770.
GFOUNCWATER
0.0
END
0.0
0.0
0.0
0.0
APPLICATION 119
NITROGEN
SURFACE
0,0 0.71 0.0 0.0 0.0 0.0
UPPER ZONE
0.0 33.29 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.0 0.2 0.40 0.0
UPPER ZONE
0.0 28.9 0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
446
-------
APPLICATION 162
NITROGEN
SURFACE
0.0 25.0 3.0 0.0 0.0 0.0
UPPER ZONE
0.0 61.9 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.0 0.0 0.0 0.0
UPPER ZONE
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
APPLICATION 297
NITROGEN
SURFACE
8.0 0.0 0.0 0.0 0.0 0.0
UPPER ZONE
20.0 0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
1.5 0.0 0.0 0,0
UPPER ZONE
4.0 0.0 0.0 0.0
CHLORIEE
SURFACE
0.0
UPPER ZONE
0.0
END
LZTEMP= 48.0 47.0 53.0 57.0 65.0 73.0 77.0 77.0 76.0 71.0 58.0 51.0
ASZT= -21.4 BSZT= 1.345 AUZT= 26.210 BUZT= 0.675
SZDPTH=0.1250 U2DPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ=99.90 UZF=5.0 LZF-1.0
447
-------
P-2 WATKINSVILIE, GA. : SWCP — CONTOURING
NUTRIENTS APPLIED: N & P ON DAY 119; N ON DAY 162; PIANT DECAY AFTER HARVEST
HYCAL=CALB
INPOT=ENGL
OUTPUT=ENGL
PRINT=INTR
SNCW=NO
PEST=NO
NUTR=YES
ICHECKOFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 3.20
BGNDAY= 2 BGNMON= 1 BGNYR= 1966
ENDDAY=31 ENDMON=12 ENDYR= 1975
UZSN= 0.50 UZS= 0.500 LZSN= 18.00 LZS= 18.00
L= 100. SS= 0.0250 NN= 0.2500 A= 0.0000 EPXM=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO= 0.100 0.100 0.100 0.100 0.000 0.150 0.600 0.850 0.750 0.100 0.100 0.100
TIMTIL= 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=0.5
YRTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.500 SRERI= 2.00 SCMPAC 0.02
NUTRIENT
TSTEP= 360 NAPPL= 3 TIMHAR= 297
ULUPTK=0.00 0.05 0.08 0.10 0.15 0.75 0.200 0.07 0.00 0.00 0.00 0.0
LZUPrK=0.00 0.00 0.00 0.05 0.03 0.55 0.450 0.10 0.00 0.00 0.00 0.0
REACTION RATES
NITROGEN
SURFACE
1.00 0.0
UPPER ZONE
0.20 0.006
DOWER ZONE
0.10 0.002
GROUNDWATER
0.0 0.0
0.10
0.13
0.025
0.0
0.00
0.002
0.002
0.0
0.0
0.00
0.0
0.0
0.0
0.0
0.0
0.0
1.0
1.0
1.0
0.0
.20
0.25
0.2
0.0
TEMPERATURE COEFFICIENTS
1.05 1.07
1.07
1.07
1.07
1.07
1.05
1.05
PHOSPHORUS
SURFACE
0.02 0.0
UPPER ZONE
0.002 0.0
LOWER ZONE
0.002 0.0
GROUNOTATER
0.0 0.0
0.01
0.70
0.80
0.0
1.0
1.0
1.0
0.0
0.015
0.0015
0.005
0.0
TEMPERATURE COEFFICIENTS
1.07 1.07
END
1.07
1.05
1.05
448
-------
INITIAL
NITROGEN
SURFACE
90.0 2.0 20.0 0.20 0.0
UPPER ZONE
375.0 4.0 ,u.O 3.00 0.0
LOWER ZONE
2500. 6.0 20.0 212. 0.0
GROUNDWATER
0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
12.0 1.0 36.0 0.0
UPPER ZONE
100. 0.35 305.0 0.0
IOWER ZONE
800. 1.0 135.0 0.0
GROUNEWATER
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
3.46
UPPER ZONE
79.54
LOWER ZONE
770.
GROUNCWATER
0.0
END
0.0
0.0
0.0
0.0
APPLICATION 119
NITROGEN
SURFACE
0.0 0.71 0.0 0.0
UPPER ZONE
0.0 33.29 0.0 0.0
PHOSPHORUS
SURFACE
0.0 0.2 0.40 0.0
UPPER ZONE
0.0 28.9 0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
0.0 0.0
0.0 0.0
449
-------
APPLICATION 162
NITROGEN
SURFACE
0.0 25.0 3.0 0.0 0.0 0.0
UPPER ZONE
0.0 61.9 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.0 0.0 0.0 0.0
UPPER ZONE
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
APPLICATION 297
NITROGEN
SURFACE
1.5 0.0 0.0 0.0 0.0 0.0
UPPER ZONE
20.0 0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.3 0.0 0.0 0.0
UPPER ZONE
4.0 0.0 0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
LZTEMP= 48.0 47.0 53.0 57.0 65.0 73.0 77.0 77.0 76.0 71.0 58.0 51.0
ASZT= -21.4 BSZT= 1.345 AUZT= 26.210 BUZT= 0.675
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ=99.90 UZF=5.0 LZF=1.0
450
-------
P-2 WATKINSVILLE, GA.: SWCP - TERRACES (2) WITH CONTOURING
NUTRIENTS APPLIED: N & P ON DAY 119; N ON DAY 162; PLANT DECAY AFTER HARVEST
HYCAL=CALB
INPOT=ENGL
OUTPUT=ENGL
PRINT=INTR
SNOW=NO
PEST=NO
NOTR=YES
ICHECKOPF
DISK=*JO
INTRVL=15 HYMIN= 0.100 AREA= 3.20
BGNDAY= 2 BGNMON= 1 BGNYR= 1966
ENDDAY=31 ENDMGN=12 ENDYR= 1975
OZSN= 0.65 UZS= 0.650 LZSN= 18.00 LZS= 18.00
L= 250. SS= 0.0150 NN= 0.2500 A= 0.0000 EPXH=0.1200 PETML 1.000
K3= 0.300 0.300 0.300 0.400 0.400 0.500 0.700 0.800 0.600 0.500 0.400 0.300
INFIL 0.100 INTER 0.590 IRC= 0.000 K24L= 1.000 KK24= 0.600 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
COVPMO 0.100 0.100 0.100 0.100 0.000 0.150 0.600 0.850 0.750 0.100 0.100 0.100
TIMTIL= 115 115 115 115 115 115 115 115 115 115 0 0 SMPF=0.35
YRTIL= 66 67 68 69 70 71 72 73 74 75 99 99
SRERTL= 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 1.500 0.000 0.000
JRER= 1.9 KRER= 0.160 JSER= 1.700 KSER= 0.500 SRERI= 2.00 SCMPAC 0.02
NOTRIENT
TSTEP= 360 NAPPL= 3 TIMHAR= 297
ULUPTK=0.00 0.05 0.08 0.10 0.15 0.75 0.200 0.07 0.00 0.00 0.00 0.0
LZUPTK=0.00 0.00 0.00 0.05 0.03 0.55 0.450 0.10 0.00 0.00 0.00 0.0
REACTION RATES
NITROGEN
SURFACE
1.00 0.0
UPPER ZONE
0.20 0.006
LOWER ZONE
0.10 0.002
GROUNDWATER
0.0 0.0
0.10
0.13
0.025
0.0
0.00
0.002
0.002
0.0
0.0
0.00
0.0
0.0
0.0
0.0
0.0
0.0
1.0
1.0
1.0
0.0
.20
0.25
0.2
0.0
TEMPERATURE COEFFICIENTS
1.05 1.07
1.07
1.07
1.07
1.07
1.05
1.05
PHOSPHORUS
SURFACE
0.02 0.0
UPPER ZONE
0.002 0.0
LOWER ZONE
0.002 0.0
GROUNDWATER
0.0 0.0
0.01 1.0
0.70 1.0
0.80 1.0
0.0 0.0
0.015
0.0015
0.005
0.0
TEMPERATURE COEFFICIENTS
1.07 1.07
END
1.07 1.05
1.05
451
-------
INITIAL
NITROGEN
SURFACE
90.0 2.0 20.0 0.20 0.0
UPPER ZONE
375.0 4.0 40.0 3.00 0.0
LOWER ZONE
2500. 6.0 20.0 212. 0.0
GROUNDWATER
0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
12.0 1.0 36.0 0.0
UPPER ZONE
100. 0.35 305.0 0.0
LOWER ZONE
800. 1.0 135.0 0.0
GROUNCWATER
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
3.46
UPPER ZONE
79.54
LOWER ZONE
770.
GROUNDWATER
0.0
END
0.0
0.0
0.0
0.0
APPLICATION 119
NITROGEN
SURFACE
0.0 0.71 0.0 0.0 0.0 0.0
UPPER ZONE
0.0 33.29 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.0 0.2 0.40 0.0
UPPER ZONE
0.0 28.9 0.0 0.0
CHLORITE
SURFACE
0.0
UPPER ZONE
0.0
END
452
-------
APPLICATION 162
NITROGEN
SURFACE
0.0 25.0 3.0 0.0 0.0 0.0
UPPER ZONE
0.0 61.9 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.0 0.0 0.0 0.0
UPPER ZONE
0-0 0.0 0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
APPLICATION 297
NITROGEN
SURFACE
1.5 0.0 0.0 0.0 0.0 0.0
UPPER ZONE
20.0 0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.3
UPPER ZONE
4.0
0.0
0.0
0.0
0.0
0.0
0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
LZTEMP= 48.0 47.0 53.0 57.0 65.0 73.0 77.0 77.0 76.0 71.0 58.0 51.0
ASZT= -21.4 BSZT= 1.345 AUZT= 26.210 BUZT= 0.675
SZDPTH=0.1250 UZDPTH= 6.125 BDSZ= 99.90 BDUZ= 99.90 BDLZ=99.90 U2F=5.0 LZF-1.0
453
-------
P6 Parameter Values
Atrazine--
P-6 EAST IANSING, MI.: BASE CONDITIONS (NO SWCP)
ATRAZINE APPLIED: 2.5 LBS/AC EVERY DAY 140 FOR THE 10-YR PERIOD
HYCAL=CALB
INPUT=ENGL
OOTPOT=ENGL
PRINT=INTR
SNOW=YES
PEST=YES
NOTR=t)0
ICHECK=OFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 1.98
BGNDAY= 7 BGNMON= 1 BGNYR= 1966
ENDDAY=31 ENDMON=12 ENDYR= 1975
UZSN= 0.170 UZS= 0.250 LZSN= 9.00 LZS= 7.0
L= 60. SS= 0.0600 NN= 0.2000 A= 0.0000 EPXM=0.1200 PETML 0.700
K3= 0.200 0.200 0.200 0.200 0.300 0.300 0.500 0.450 0.400 0.300 0.200 0.200
INFIL 0.030 INTER 0.800 IRC= 0.000 K24L= 1.000 KK24= 0.000 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
SNOWPRINT=NO
RADCN 1.000 CCFAC 1.000 SCF= 1.400 ELDIF 0.000 IDNS= 0.140 F= 0.000
DGM= 0.000 WC= 0.030 MPACK 1.000 EVAPS 0.400 MELEV 892. TSNOW 32.00
PACK= 1.000 DEPTH= 4.000
PETMIN= 35.00 PETMAX= 40.00 WMUL= 1.000 RMUL= 1.000 KUGI= 0.000
COVPMO= 0.000 0.000 0.000 0.000 0.000 0.050 0.550 0.900 0.900 0.800 0.000 0.000
TIMTIL= 139 139 139 139 139 139 139 139 139 139 0 0 SMPF=1.0
YRTIL= 66 67 68 69 70 71 72 73 74 75 0 0
SRERTL= 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000 0.000
JRER= 2.200 KRER= 0.150 JSER= 1.400 KSER= 0.60 SRERI= 1.0 SCMPAC .001
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= o.oo PSUZ= o.oo PSIZ= o.oo PSGZ= o.o
TIMAP= 140 140 140 140 140 140 140 140 140 140 0 0 NPA= 10 NDRPA= 1
YEARAP= 66 67 68 69 70 71 72 73 74 75 00 00
SSTR= 2.500 2.500 2.500 2.500 2.500 2.500 2.500 2.500 2.500 2.500 0.000 0.000
CMAX= 3.5E-5 DD= 0.0000 K= 4.0000 N= 1.0000 NP= 2.300
DDG= 140 140 140 140 140 140 140 140 140 140 0 0
YDG= 66 67 68 69 70 71 72 73 74 75 00 00
KDG= 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0,050 0.050 0.050 0.0 0.0
SZDPTH=0.1250 UZDPTH=6.1250 BD3Z= 63.70 BDUZ= 72.40 BDLZ= 99.0 UZF=1.0 LZF=1.0
454
-------
P-6 EAST LANSING, MI.: NO TILLAGE SWOP
ATRAZINE APPLIED: 2.5 LBS/AC EVERY DAY 140 FOR THE 10-YR PERIOD
HYCAL
-------
P-6 EAST LANSING, MI.: CONTOUR SWCP
ATRAZINE APPLIED: 2.5 LBS/AC EVERY DAY 140 FOR THE 10-YR PERIOD
HYCAL=CALB
INPUT=ENGL
OOTPUT=ENGL
PRINT=INTR
SNOW=YES
PEST=YES
NUTR=NO
ICHECKOFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 1.98
BGNDAY= 7 BGNMCN= 1 BGSYR= 1966
ENDDAY=31 ENDMOM=12 ENDYR= 1975
UZSN= 0.200 UZS= 0.300 LZSN= 9.00 LZS= 7.0
L= 60. SS= 0.0600 NN= 0.2500 A= 0.0000 EPXM=0.1200 PETML 0.700
K3= 0.200 0.200 0.200 0.200 0.300 0.300 0.500 0.450 0.400 0.300 0.200 0.200
INFIL 0.030 INTER 0.800 IRC= 0.000 K24L= 1.
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS=
SNCWPRINT=*JO
RADCN 1.000 CCFAC 1.000 SCF= 1.400 ELDIF 0.000 IDNS=
DGM= 0.000 WC= 0.030 MPACK 1.000 EVAPS 0.
PACK= 1.000 DEPTH= 4.000
PETMIN= 35.00 PETMAX= 40.00 WMUL= 1.000RMUL= 1.000KUGI= 0.000
COVPMO= 0.000 0.000 0.000 0.000 0.000 0.050 0.550 0.900 0.900 0.800 0.000 0.000
TIMT[L= 139 139 139 139 139 139 139 139 139 139 0 0 SMPF=0.5
YRTIL= 66 67 68 69 70 71 72 73 74 75 0 0
.000 KK24= 0.000 K24EL 0.000
0.000 OFS= 0.000 IFS= 0.000
0.440 F= 0.000
.400 MELEV 892. TSNOH 32.00
SRERTL= 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000 0.000
JRER= 2.200 KRER= 0.150 JSER= 1
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSLZ= 0.00 PSGZ=
TIMAP= 140 140 140 140 140 140 140 140 140 140
YEARAP= 66 67 68 69 70 71 72 73 74 75
400 KSER= 0.50 SRERI= 1.0 SCMPAC .001
0.0
0 0 NPA= 10 NDRPA= 1
00 00
SSTR= 2.500 2.500 2.500 2.500 2.500 2.500 2.500 2.500 2.500 2.500 0.000 0.000
CMAX= 3.5E-5DD= 0.0000 K= 4.0000N= 1.0000 NP= 2.300
DDG= 140 140 140 140 140 140 140 140 140 140 0 0
YDG= 66 67 68 69 70 71 72 73 74 75 00 00
KDG= 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.0 0.0
SZDPTH=0.1250 UZDPTH=6.1250 BDSZ= 63.70 BDUZ= 72.40 BDLZ= 99.0 UZF=1.0 LZF=1.0
456
-------
P-6 EAST LANSING, MI.: TERRACE WITH CONTOURS SWCP
ATOAZINE APPLIED: 2.5 IBS/AC EVERY DAY 140 FOR THE 10-YR PERIOD
HiCAlFCALB
INPUT=ENGL
OUTPUT=ENGL
PRINT=INTR
SNOW=YES
PEST=YES
NOTR=NO
ICHECKOFF
DISK=*JO
INTRVL=15 HYMIN= 0.100 AREA* 1.98
BQ3DAY= 7 BGNMCN= 1 BQJYR= 1966
ENDEftY=31 ENEMCN=12 ENDYR= 1975
UZSN* 0.260 U2S= 0.390USN= 9.00 LZS= 7.0
L= 200. SS= 0.0180 NN= 0.2500 A= 0.0000 EPXM=0.1200 PETML 0.700
K3= 0.200 0.200 0.200 0.200 0.300 0.300 0.500 0.450 0.400 0.300 0.200 0.200
INFIL 0.030 INTER 0.800 IRC= 0.000 K24L= 1.000 KK24= 0.000 K24EL 0.000
SO/= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0,000
SNCWPRINT=NO
RADCN 1.000 CCFAC 1.000 SCF= 1.400 ELDIF 0.000 IDNS= 0.140 F= 0.000
DGM= 0.000 WC= 0.030 MPACK 1.000 EVAPS 0.400 MELEV 892. TSNOW 32.00
PACK= 1.000 DEPTH= 4.000
PETMIN= 35.00 PE1MAX= 40.00 WMUL= 1.000 RMUL= 1.000 KUGI= 0.000
COVPMO= 0.000 0.000 0.000 0.000 0.000 0.050 0.550 0.900 0.900 0.800 0.000 0.000
TIMTIIr* 139 139 139 139 139 139 139 139 139 139 0 0 SMPF=0.35
YRTIL= 66 67 68 69 70 71 72 73 74 75 0 0
SRERTL= 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000 0.000
JRER= 2.200 KRER= 0.150 JSER= 1.400 KSER= 0.50 SRERI= 1.0 SCMPAC .001
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSDZ= 0.00 PSLZ= 0.00 PSGZ= 0.0
TIMAP= 140 140 140 140 140 140 140 140 140 140 0 0 NPA= 10 NDRPA= 1
YEARAP= 66 67 68 69 70 71 72 73 74 75 00 00
SSTR= 2.500 2.500 2.500 2.500 2.500 2.500 2.500 2.500 2.500 2.500 0.000 0.000
CMAX= 3.5E-5 DD= 0.0000 K= 4.0000 N= 1.0000 NP= 2.300
DDG= 140 140 140 140 140 140 140 140 140 140 0 0
YDG= 66 67 68 69 70 71 72 73 74 75 00 00
KDG= 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.0 0.0
SZDPTH=0.1250 UZDPTH=6.1250 BDSZ= 63.70 BDUZ= 72.40 BDLZ= 99.0 UZF=1.0 LZF=1.0
457
-------
Paraquat--
P-6 EAST IANSING, MI.: BASE CONDITIONS (NO SWCP)
PARAQUAT APPLIED: 1.3 LBS/AC EVERY DAY 140 FOR THE 10-YR PERIOD
HYCAL=CALB
INPOT=ENGL
OUTPOT=ENGL
PRINT=INTR
SNOW=YES
PEST=YES
ICHECK=OFF
INTRVL=15 HYMIN= 0.100 AREA= 1.98
BGNDAY= 7 BGNMON= 1 BGNYR= 1966
ENDDAY=31 ENEMON=12 ENDYR== 1975
OZSN= 0.170 UZS= 0.250 LZSN= 9.00 LZS= 7.0
L= 60. SS= 0.0600 NN= 0.2000 A= 0.0000 EPXM=0.1200 PETML 0.700
K3= 0.200 0.200 0.200 0.200 0.300 0.300 0.500 0.450 0.400 0.300 0.200 0.200
INFIL 0.030 INTER 0.800 IRC= 0.000 K24L= 1.000 KK24= 0.000 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
SNOWPRINT=NO
RADCN 1.000 CCFAC 1.000 SCF= 1.400 ELDIF 0.000 IDNS= 0.140 F= 0.000
DGM= 0.000 WC= 0.030 MPACK 1.000 EVAPS 0.400 MELEV 892. TSNOW 32.00
PACK= 1.000 DEPTH= 4.000
PETMIN= 35.00 PEIMAX= 40.00 WMU1> 1.000 RMUL= 1.000 KUGI= 0.000
COVPMO= 0.000 0.000 0.000 0.000 0.000 0.050 0.550 0.900 0.900 0.800 0.000 0.000
TIMTIL- 139 139 139 139 139 139 139 139 139 139 0 0 SMPF=1.0
YRTIL= 66 67 68 69 70 71 72 73 74 75 0 0
SRERTD= 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000 0.000
JRER= 2.200 KRER= 0.150 JSER= 1.400 KSER= 0.6Q SRERI= 1.0 SCMPAC .001
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSLZ= 0.00 PSGZ= 0.0
TIMAP= 140 140 140 140 140 140 140 140 140 140 0 0 NPA= 10 NDRPA= 1
YEARAP= 66 67 68 69 70 71 72 73 74 75 00 00
SSTR= 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 0.000 0.000
CMAX= l.OE-5 DD= 0.0003 K= 120.00 N= 2.0000 NP= 4.600
DDG= 140 140 140 140 140 140 140 140 140 140 0 0
YDG= 66 67 68 69 70 71 72 73 74 75 00 00
KDG= 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.0 0.0
SZDPTH=0.1250 UZDPTH=6.1250 BDSZ= 63.70 BDDZ= 72.40 BDLZ= 99.0 UZF=1.0 LZF=1.0
458
-------
P-6 EAST LANSING, MI.: NO TILLAGE SWCP
PARAOJAT APPLIED: 1.3 LBS/AC EVERY EAY 140 FOR THE 10-YR PERIOD
HYCAL^ALB
INPUT=ENGL
OUTFUT=ENGL
PRINT=INTR
SNOW=YES
PEST=YES
NUTR=NO
ICHECKOFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 1.98
BGNCftY= 7 BGNMON= 1 BGNYR= 1966
ENDEAY=31 ENEMCN=12 ENDYR= 1975
UZSN= 0.170 UZS= . 0.250 LZSN= 9.00 LZS= 7.0
L= 60. SS= 0.0600 NN= 0.3200 A= 0.0000 EPXM=0.1200 PETNL 0.700
K3= 0.200 0.200 0.200 0.200 0.300 0.300 0.500 0.450 0.400 0.300 0.200 0.200
DJFIL 0.030 INTER 0.800 IRC= 0.000 K24L= 1.000 KK24= 0.000 K24EL 0.000
SG*= 0.000 QWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
SNOWPRINT=NO
RADCN 1.000 CCFAC 1.000 SCF= 1.400 ELDIF 0.000 IDNS= 0.140 F= 0.000
DGM= 0.000 WC= 0.030 HPACK 1.000 EVAPS 0.400 MELEV 892. TSNOW 32.00
PACK= 1.000 DEPTH= 4.000
PETMIN= 35.00 PETMAX= 40.00 WMUL= 1.000 RMUL= 1.000 KUGI= 0.000
COVPMO= 0.600 0.600 0.600 0.600 0.400 0.430 0.730 0.940 0.940 0.880 0.600 0.600
TIMTIL= 139 139 139 139 139 139 139 139 139 139 0 0 SMPF=1.0
YRTIL= 66 67 68 69 70 71 72 73 74 75 0 0
SRERTL= 0.100 0.100 0.100 0.100 0.100 0.100 0.100 0.100 0.100 0.100 0.000 0.000
JRER= 2.200 KRER= 0.150 JSER= 1.400 KSER= 0.55 SRERI= 0.1 SCMPAC .001
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSLZ= 0.00 PSGZ= 0.0
TIMAP= 140 140 140 140 140 140 140 140 140 140 0 0 NPA= 10 NDRPA= 1
YEARAP= 66 67 68 69 70 71 72 73 74 75 00 00
SSTR= 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 0.000 0.000
CMAX= l.OE-5 DD= 0.0003K= 120.00 N= 2.0000 NP= 4.600
DDG= 140 140 140 140 140 140 140 140 140 140 0 0
YDG= 66 67 68 69 70 71 72 73 74 75 00 00
KDG= 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.0 0.0
SZDPTH=0.1250 OZDPTH=6.1250 BDSZ= 63.70 BDUZ= 72.40 BDLZ= 99.0 UZF=1.0 LZF=1.0
459
-------
P-6 EAST LANSING, MI.: CONTOUR SWCP
PARAQUAT APPLIED: 1.3 LBS/AC EVERT DAY 140 FOR THE 10-YR PERIOD
HYCAL=€ALB
INPOT=ENGL
OOTPUT=ENGL
PRINT=INTR
SNOW=YES
PEST=YES
NUTR=NO
ICHECKOFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 1.98
BQNDAY= 7 BQIMON= 1 BGNYR= 1966
ENDEAY=31 ENEMCN=12 ENDYR= 1975
OZSN= 0.200 UZS= 0.300 LZSN= 9.00 LZS= 7.0
L= 60. SS= 0.0600 NN= 0.2500 A= 0.0000 EPXM=0.1200 PETML 0.700
K3= 0.200 0.200 0.200 0.200 0.300 0.300 0.500 0.450 0.400 0.300 0.200 0.200
INFIL 0.030 INTER 0.800 IRC= 0.000 K24L= 1.000 KK24= 0.000 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
SNOWPRINT=tro
RADCN 1.000 CCFAC 1.000 SCF= 1.400 ELDIF 0.000 IDNS= 0.140 F= 0.000
DGM= O.OOOWC= 0.030 MPACK 1.000 EVAPS 0.400 MELEV 892. TSNOW 32.00
PACK= 1.000 DEPTH= 4.000
PETMIN= 35.00 PETMAX= 40.00 WMUL= 1.000 RMUJ> 1.000 KUGI= 0.000
COVPMO= 0.000 0.000 0.000 0.000 0.000 0.050 0.550 0.900 0.900 0.800 0.000 0.000
TIMTIL= 139 139 139 139 139 139 139 139 139 139 0 0 SMPF=0.5
YRTIL= 66 67 68 69 70 71 72 73 74 75 0 0
SRERTD= 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000 0.000
JRER= 2.200 KRER= 0.150 JSER= 1.400 KSER= 0.50 SRERI= 1.0 SCMPAC .001
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSI2= 0.00 PSGZ= 0.0
TIMAP= 140 140 140 140 140 140 140 140 140 140 0 0 NPA= 10 NDRPA= 1
YEARAP= 66 67 68 69 70 71 72 73 74 75 00 00
SSTR= 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 0.000 0.000
CMAX= l.OE-5 DD= 0.0003 K= 120.00 N= 2.0000 NP= 4.600
DDG= 140 140 140 140 140 140 140 140 140 140 0 0
YDG= 66 67 68 69 70 71 72 73 74 75 00 00
KDG= 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.0 0.0
SZDPTH=0.1250 UZDPTH=6.1250 BDSZ= 63.70 BDUZ= 72.40 BDLZ= 99.0 UZF=1.0 LZF=1.0
460
-------
P-€ EAST LANSING, MI.: TERRACE WITH CONTOURS SWCP
PARACJJAT APPLIED: 1. 3 LBS/AC EVERY DAY 140 FOR THE 10-YR PERIOD
HYCAL 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000 0.000
ORER= 2.200 KRER= 0.150 JSER= 1.400 KSER= 0.50 SRERI= 1.0 SCMPAC .001
PESTICIDE
APMODE=SURF
DESORP=YES
PSSZ= 0.00 PSUZ= 0.00 PSI3= 0.00 PSGZ= 0.0
TIMAP= 140 140 140 140 140 140 140 140 140 140 0 0 NPA= 10 NDRPA= 1
YEARAP= 66 67 68 69 70 71 72 73 74 75 00 00
SSTR= 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 1.300 0.000 0.000
CMAX= l.OE-5 DD= 0.0003 K= 120.00 N= 2.0000 NP= 4.600
DDG= 140 140 140 140 140 140 140 140 140 140 0 0
YDG= 66 67 68 69 70 71 72 73 74 75 00 00
KDG= 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.0 0.0
SZDPTH=0.1250 UZDPTH=6.1250 BDSZ= 63.70 BDUZ= 72.40 BDLZ= 99.0 UZF=1.0 LZF=1.0
461
-------
Nutrients--
P-6 EAST LANSING, MI.: BASE CONDITIONS (NO SHCP)
NUTRIENTS APPLIED: N & P ON DAY 140; N ON DAY 189; PLANT DECAY AFTER HARVEST
HYCAL=CALB
INPUT=ENGL
OUTPUT=E>3GL
PRINT=INTR
SNOW=YES
PEST=NO
NUTR=YES
ICHECK=OFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 1.98
BGNDAY= 7 BGNMCN= 1 BGNYR= 1966
ENDDAY=31 ENCMON=12 ENDYR= 1975
UZSN= 0.170 UZS= 0.250 LZSN= 9.00 LZS= 7.0
L= 60. SS= 0.0600 NN= 0.2000 A= 0.0000 EPXM=0.1200 PETML 0.700
K3= 0.200 0.200 0.200 0.200 0.300 0.300 0.500 0.450 0.400 0.300 0.200 0.200
INFIL 0.030 INTER 0.800 IRC= 0.000 K24L= 1.000 KK24= 0.000 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
SNOWPRINT=«0
RADCN 1.000 CCFAC 1.000 SCF= 1.400 ELDIF 0.000 IDNS= 0.140 F= 0.000
DGM= 0.000 WC= 0.030 MPACK 1.000 EVAPS 0.400 MELEV 892. TSNCW 32.00
PACK= 1.000 DEPTH= 4.000
PETMIN= 35.00 PETMAX= 40.00 WMUL= 1.000 PMUL= 1.000 KUGI= 0.000
COVPMO= 0.000 0.000 0.000 0.000 0.000 0.050 0.550 0.900 0.900 0.800 0.000 0.000
TIMTIL= 139 139 139 139 139 139 139 139 139 139 0 0 SMPF=1.0
YRTIL= 66 67 68 69 70 71 72 73 74 75 0 0
SRERTL= 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000 0.000
JRER= 2.200 KRER= 0.150 JSER= 1.400 KSER= 0.60 SRERI= 1.0 SCMPAC .001
NUTRIENT
TSTEP= 360 NAPPL= 3 TIMHAR= 273
ULUPTK=0.00 0.00 0.00 0.01 0.05 0.09 1.000 0.060 0.00 0.00 0.00 0.0
LZUPTK=0.00 0.00 0.00 0.00 0.00 0.01 1.000 0.125 0.00 0.00 0.00 0.0
REACTION RATES
NITROGEN
SURFACE
3.00 0.0 0.25 0.015 0.0 0.0 5.0 0.75
UPPER ZONE
1.25 0.05 0.40 0.0015 0.0 0.0 0.75 0.3
LOWER ZONE
0.7 0.0 0.090 0.0015 0.0 0.0 1.0 0.4
GROUNOTATER
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
TEMPERATURE COEFFICIENTS
1.05 1.07 1.07 1.07 1.07 1.07 1.05 1.05
PHOSPHOHJS
SURFACE
0.015 0.0 0.01 1.00 0.01
UPPER ZONE
0.0015 0.0 2.10 0.5 0.006
LOWER ZONE
0.0015 0.0 1.70 0.5 0.005
462
-------
GROUNEWATER
0.0 0.0 0.0 0.0 0.0
TEMPERATURE COEFFICIENTS
1.07 1.07 1.07 1.05 1.05
END
INITIAL
NITROGEN
SURFACE
68.00 0.40 1.00 1.95 0.0
UPPER ZONE
510.0 8.00 20.50 27.4 0.0
LONER ZONE
2087. 9.0 23.0 100.0 0.0
GROUNDWATER
0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
27.3 0.09 6.64 0.0
UPPER ZONE
175. 13.3 101.3 0.0
LOWER ZONE
1000. 26.8 241.5 0.0
GROUNDWATER
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
6.65
UPPER ZONE
108.0
LOWER ZONE
147.6
GROUNEWATER
0.0
END
0.0
0.0
0.0
0.0
APPLICATION 140
NITROGEN
SURFACE
0.0 0.3 1.0 1.3 0.0 0.0
UPPER ZONE
0.0 29.2 0.0 29.2 0.0 0.0
PHOSPHORUS
SURFACE
0.0 0.35 3.10 0.0
UPPER ZONE
0.0 79.5 0.0 0.0
CHLORIDE
SURFACE
463
-------
5.8
UPPER ZONE
134.2
END
APPLICATION 189
NITROGEN
SURFACE
0.0 38.5 1.5 40.0 0.0
UPPER ZONE
0.0 18.0 0.0 18.0 0.0
PHOSPHOHJS
SURFACE
0.0 0.0 0.0 0.0
UPPER ZONE
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
0.0
0.0
APPLICATION 273
NITROGEN
SURFACE
1.5 0.0 0.0 0.0 0.0 0.0
UPPER ZONE
16.0 0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.25 0.0 0.0 0.0
UPPER ZONE
3.0 0.0 0.0 0.0
CHLORIDE
SURFACE
66.0
UPPER ZONE
60.0
END
LZTEMP= 38.2 36.6 37.1 40.1 48.5 56.5 62.4 65.1 64.5 58.7 51.3 44.3
ASZT= 44.96 BSZT= 0.39 ACJZT= 0.0 BUZT= 1.0
SZDPTH=0.1250 UZDPTH= 3.125 BCSZ= 63.70 BDUZ= 72.40 BDLZ= 99.0 UZF=5.0 LZF=1.0
464
-------
P-6 EAST IANSING, MI.: NO TILLAGE SWCP
NUTRIENTS APPLIED: N & P ON DAY 140; N ON DAY 189; P1ANT DECAY AFTER HARVEST
HYCAJ>CALB
DJPUT=ENGI.
OOTPOT=ENGL
PRINT=INTR
SNOW=YES
PEST=NO
NUTR=YES
ICHECKOFF
DISK=NO
INTRVL=15 HYMIN= 0.100 AREA= 1.98
BGNDAY= 7 BGNMON= 1 BGNYR= 1966
ENDDAY=31 ENDMCN=12 ENDYR= 1975
OZSN= 0.170 UZS= 0.250 LZSN= 9.00 LZS= 7.0
L= 60. SS= 0:0600 NN= 0.3200 A= 0.0000 EPXM=0.1200 PETOL 0.700
K3= 0.200 0.200 0.200 0.200 0.300 0.300 0.500 0.450 0.400 0.300 0.200 0.200
INFIL 0.030 INTER 0.800 IRC= 0.000 K24L= 1.000 KK24= 0.000 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
SNOWPRINT=NO
RADCN 1.000 CCFAC 1.000 SCF= 1.400 ELDIF 0.000 IDNS= 0.140 F= 0.000
D3M= 0.000 WC= 0.030 MPACK 1.000 EVAPS 0.400 MELEV 892. TSNCW 32.00
PACK= 1.000 DEPTH= 4.000
PETMIN= 35.00 PETMAX= 40.00 WMU1> 1.000 RMUL= 1.000 KUGI= 0.000
COVPMO= 0.600 0.600 0.600 0.600 0.400 0.430 0.730 0.940 0.940 0.880 0.600 0.600
TIMTII> 139 139 139 139 139 139 139 139 139 139 0 0 SMPF=1.0
YRTII> 66 67 68 69 70 71 72 73 74 75 0 0
SRERTL= 0.100 0.100 0.100 0.100 0.100 0.100 0.100 0.100 0.100 0.100 0.000 0.000
JRER= 2.200 KRER= 0.150 JSER= 1.400 KSER= 0.55 SRERI= 0.1 SCMPAC .001
NUTRIENT
TSTEP= 360 NAPPI> 3 TIMHAR= 273
ULUPTK=0.00 0.00 0.00 0.01 0.05 0.09 1.000 0.060 0.00 0.00 0.00 0.0
LZUPTK=0.00 0.00 0.00 0.00 0.00 0.01 1.000 0.125 0.00 0.00 0.00 0.0
REACTION RATES
NITROGEN
SURFACE
3.00 0.0
UPPER ZONE
1.25 0.05
UOWER ZONE
0.7 0.0
GRQUNDWATER
0.0 0.0
0.25
0.40
0.090
0.0
0.015
0.0015
0.0015
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
TEMPERATURE COEFFICIENTS
1.05 1.07
1.07
1.07
1.07
1.07
5.0
0.75
1.0
0.0
1.05
0.75
0.3
0.4
0.0
1.05
PHOSPHORUS
SURFACE
0.015 0.0 0.01 1.00 0.01
UPPER ZONE
0.0015 0.0 2.10 0.5 0.006
DOWER ZONE
0.0015 0.0 1.70 0.5 0.005
465
-------
GROUNEWATER
0.0 0.0 0.0 0.0 0.0
TEMPERATURE COEFFICIENTS
1.07 1.07 1.07 1.05 1.05
END
INITIAL
NITROGEN
SURFACE
68.00 0.40 1.00 1.95 0.0
UPPER ZONE
510.0 8.00 20.50 27.4 0.0
DOWER ZONE
2087. 9.0 23.0 100.0 0.0
GROUNEWATER
0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
27.3 0.09 6.64 0.0
UPPER ZONE
175. 13.3 101.3 0.0
LOWER ZONE
1000. 26.8 241.5 0.0
GROUNEWATER
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
6.65
UPPER ZONE
108.0
LOWER ZONE
147.6
GROUNEWATER
0.0
END
0.0
0.0
0.0
0.0
APPLICATION 140
NITROGEN
SURFACE
0.0 0.3 1.0 1.3 0.0 0.0
UPPER ZONE
0.0 29.2 0.0 29.2 0.0 0.0
PHOSPHORUS
SURFACE
0.0 0.35 3.10 0.0
UPPER ZONE
0.0 79.5 0.0 0.0
CHLORIDE
SURFACE
466
-------
5.8
UPPER ZONE
134.2
END
APPLICATION 189
NITROGEN
SURFACE
0.0 38.5 1.5 40.0 0.0
UPPER ZONE
0.0 18.0 0.0 18.0 0.0
PHOSPHORUS
SURFACE
0.0 0.0 0.0 0.0
UPPER ZONE
0.0 0.0 0.0 0.0
CHLORITE
SURFACE
0.0
UPPER ZONE
0.0
END
0.0
0.0
APPLICATION 273
NITROffiN
SURFACE
6.0 0.0 0.0 0.0 0.0 0.0
UPPER ZONE
16.0 0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
1.20 0.0 0.0 0.0
UPPER ZONE
3.0 0.0 0.0 0.0
CHLORITE
SURFACE
66.0
UPPER ZONE
60.0
END
LZTEMP= 38.2 36.6 37.1 40.1 48.5 56.5 62.4 65.1 64.5 58.7 51.3 44.3
ASZT= 44.96 BSZT= 0.39 AUZT= 0.0 BUZT= 1.0 „„,„,_.,„
SZDPTH-0.1250 UZEPTH= 3.125 BD3Z= 63.70 BDUZ= 72.40 BDLZ= 99.0 UZF=5.0 LZF=1.0
467
-------
P-6 EAST IANSINS, MI.: CONTOUR SWOP
NUTRIENTS APPLIED: N & P ON CAY 140; N ON DAY 189; PLANT DECAY AFTER HARVEST
HTCAL^ALB
INPOT=ENGL
O7rPOT=ENGL
TOINT=INTR
SNOW=YES
PBST=NO
NUER=YES
ICHECKOFF
DISKED
INTRVL=15 HYMIN= 0.100 AREA= 1.98
BGNDAY= 7 BGNMON= 1 BGNYR= 1966
ENDDAY=31 ENCMCN=12 ENDYR= 1975
UZSN= 0.200 UZS= 0.300 LZSN= 9.00 LZS= 7.0
L= 60. SS= 0.0600 NN= 0.2500 A= 0.0000 EPXM=0.1200 PETML 0.700
K3= 0.200 0.200 0.200 0.200 0.300 0.300 0.500 0.450 0.400 0.300 0.200 0.200
INFIL 0.030 INTER 0.800 IRC= 0.000 K24L= 1.000 KK24= 0.000 K24EL 0.000
SGW= 0.000 GWS= 0.000 KV= 0.000 ICS= 0.000 OFS= 0.000 IFS= 0.000
SNOWPRINT=*IO
RADCN 1.000 CCFAC 1.000 SCF= 1.400 ELDIF 0.000 IDNS= 0.140 F= 0.000
DGM= 0.000 WC= 0.030 MPACK 1.000 EVAPS 0.400 MELEV 892. TSNOW 32.00
PACK= 1.000 DEPTH= 4.000
PETMIN= 35.00 PETMAX= 40.00 WMUL= 1.000 RMUL= 1.000 KUGI= 0.000
COVPMO= 0.000 0.000 0.000 0.000 0.000 0.050 0.550 0.900 0.900 0.800 0.000 0.000
TIMTIL= 139 139 139 139 139 139 139 139 139 139 0 0 SMPF=0.5
YRTIL= 66 67 68 69 70 71 72 73 74 75 0 0
SRERTL= 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000 0.000
JRER= 2.200 KRER= 0.150 JSER= 1.400 KSER= 0.50 SRERI= 1.0 SCMPAC .001
NUTRIENT
TSTEP= 360 NAPPL= 3 TIMHAR= 273
ULUPTK=0.00 0.00 0.00 0.01 0.05 0.09 1.000 0.060 0.00 0.00 0.00 0.0
LZUPTK=0.00 0.00 0.00 0.00 0.00 0.01 1.000 0.125 0.00 0.00 0.00 0.0
REACTION RATES
NITROGEN
SURFACE
3.00 0.0 0.25 0.015 0.0 0.0 5.0 0.75
UPPER ZONE
1.25 0.05 0.40 0.0015 0.0 0.0 0.75 0.3
LOWER ZONE
0.7 0.0 0.090 0.0015 0.0 0.0 1.0 0.4
GROUNCWATER
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
TEMPERATURE COEFFICIENTS
1.05 1.07 1.07 1.07 1.07 1.07 1.05 1.05
PHOSPHORUS
SURFACE
0.015 0.0 0.01 1.00 0.01
UPPER ZONE
0.0015 0.0 2.10 0.5 0.006
LOWER ZONE
0.0015 0.0 1.70 0.5 0.005
468
-------
GROUNCWATER
0.0 0.0 0.0 0.0
TEMPERATURE COEFFICIENTS
1.07 1.07 1.07 1.05
END
INITIAL
NITROGEN
SURFACE
68.00 0.40 1.00 1.95
UPPER ZONE
510.0 8.00 20.50 27.4
DOWER ZONE
2087. 9.0 23.0 100.0
GROUNEWATER
0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
27.3 0.09 6.64 0.0
UPPER ZONE
175. 13.3 101.3 0.0
LONER ZONE
1000. 26.8 241.5 0.0
GROUNDWATER
0.0 0.0 0.0 0.0
CHLORITE
SURFACE
6.65
UPPER ZONE
108.0
LOWER ZONE
147.6
GROUNEWATER
0.0
END
0.0
1.05
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
APPLICATION 140
NITROGEN
SURFACE
0.0 0.3 1.0 1.3 0.0 0.0
UPPER ZONE
0.0 29.2 0.0 29.2 0.0 0.0
PHOSPHORUS
SURFACE
0.0 0.35 3.10 0.0
UPPER ZONE
0.0 79.5 0.0 0.0
CHLORITE
SURFACE
469
-------
5.8
UPPER ZONE
134.2
END
APPLICATION 189
NITROGEN
SURFACE
0.0 38.5 1.5 40.0 0.0
UPPER ZONE
0.0 18.0 0.0 18.0 0.0
PHOSPHORUS
SURFACE
0.0 0.0 0.0 0.0
UPPER ZONE
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
0.0
0.0
APPLICATION 273
NITROGEN
SURFACE
1.5 0.0 0.0 0.0 0.0 0.0
UPPER ZONE
16.0 0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.25 0.0 0.0 0.0
UPPER ZONE
3.0 0.0 0.0 0.0
CHLORIDE
SURFACE
66.0
UPPER ZONE
60.0
END
LZTEMP= 38.2 36.6 37.1 40.1 48.5 56.5 62.4 65.1 64.5 58.7 51.3 44.3
ASZT= 44.96 BSZT= 0.39 AUZT= 0.0 BUZT= 1.0
SZDPTH=0.1250 UZDPTH= 3.125 BDSZ= 63.70 BDUZ= 72.40 BDLZ= 99.0 UZF=5.0 LZF=1.0
470
-------
P-6 EAST IANSING, MI.: TERRACE WITH CONTOURS SWCP
NUTRIENTS APPLIED: N & P ON DAY 140; N ON DAY 189; PIANT DECAY AFTER HARVEST
HYCAL=CAIB
INPUT=ENGL
OUTPUT=ENGL
PRINT=INTR
SNOW=YES
PEST=NO
NOTR=YES
ICHECKOFF
DISK=80
INTRVL=15 HYMIN= 0.100 AREA= 1.98
BGNDAY= 7'BGNMON= 1 BGNYR= 1966
ENDDAY=31 ENDMON=12 ENDYR= 1975
UZSN= 0.260 UZS= 0.390 LZSN= 9.00 LZS= 7.0
L= 200. SS= 0.0180 NN= 0.2500 A= 0.0000 EPXM=0.1200 PETML 0.700
K3= 0.200 0.200 0.200 0.200 0.300 0.300 0.500 0.450 0.400 0.300 0.200 0.200
INFIL 0.030 INTER 0.800 IRC= 0.000 K24L= 1
0.000 KV= 0.000 ICS= 0
.000
.000
KK24=
OFS=
000 K24EL 0.000
000 IFS= 0.000
400 ELDIF 0.000 IDNS= 0.140 F= 0.000
000 EVAPS 0.400 MELEV 892. TSNCW 32.00
SGW= 0.000 GWS=
SNOWPRINT=NO
RADCN 1.000 CCFAC 1.000 SCF=
DGM= 0.000 WC= 0.030 MPACK
PACK= 1.000 DEPTH= 4.000
PETMIN= 35.00 PETMAX= 40.00 WMDL= 1.000 RMUL= 1.000 KUGI= 0.000
COVPMO= 0.000 0.000 0.000 0.000 0.000 0.050 0.550 0.900 0.900 0.800 0.000 0.000
TIMTII> 139 139 139 139 139 139 139 139 139 139 0 0 SMPF=0.35
YRTID= 66 67 68 69 70 71 72 73 74 75
SRERTL= 1.000 1.000 1.000 1.000 1.000 1.000 1.000
JRER= 2.200 KRER= 0.150 JSER= 1.400 KSER= 0.50 SRERI=
NUTRIENT
TSTEP= 360 NAPPL= 3 TIMHAR= 273
ULUPTK=0.00 0.00 0.00 0.01 0.05 0.09 1.000 0.060 0.00
LZUPTK=0.00 0.00 0.00 0.00 0.00 0.01
000 1.000 1.000 0.000 0.000
1.0 SCMPAC .001
1.000 0.125 0.00
0.00
0.00
0.00
0.00
0.0
0.0
REACTION RATES
NITROGEN
SURFACE
3.00 0.0
UPPER ZONE
1.25 0.05
DOWER ZONE
0.7 0.0
GROUNDWATER
0.0 0.0
0.25
0.40
0.090
0.0
0.015
0.0015
0.0015
0.0
TEMPERATURE COEFFICIENTS
1.05 1.07
1.07
1.07
PHOSPHORUS
SURFACE
0.015 0.0
UPPER ZONE
0.0015 0.0
DOWER ZONE
0.0015 0.0
0.01
2.10
1.70
1.00
0.5
0.5
0.0
0.0
0.0
0.0
1.07
0.01
0.006
0.005
0.0 5.0 0.75
0.0 0.75 0.3
0.0 1.0 0.4
0.0 0.0 0.0
1.07 1.05 1.05
471
-------
GROUNCWATER
0.0 0.0 0.0 0.0 0.0
TEMPERATURE COEFFICIENTS
1.07 1.07 1.07 1.05 1.05
END
INITIAL
NITROGEN
SURFACE
68.00 0.40 1.00 1.95 0.0
UPPER ZONE
510.0 8.00 20.50 27.4 0.0
DOWER ZONE
2087. 9.0 23.0 100.0 0.0
GROUNDWATER
0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
27.3 0.09 6.64 0.0
UPPER ZONE
175. 13.3 101.3 0.0
LOWER ZONE
1000. 26.8 241.5 0.0
GROUNDWATER
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
6.65
UPPER ZONE
108.0
LOWER ZONE
147.6
GROUNDWATER
0.0
END
APPLICATION 140
NITROGEN
SURFACE
0.0 0.3 1.0 1.3 0.0
UPPER ZONE
0.0 29.2 0.0 29.2 0.0
PHOSPHORUS
SURFACE
0.0 0.35 3.10 0.0
UPPER ZONE
0.0 79.5 0.0 0.0
CHLORIDE
SURFACE
0.0
0.0
0.0
0.0
0.0
0.0
472
-------
5.8
UPPER ZONE
134.2
END
APPLICATION 189
NITROGEN
SURFACE
0.0 38.5 1.5 40.0 0.0
UPPER ZONE
0.0 18.0 0.0 18.0 0.0
PHOSPHOHJS
SURFACE
0.0 0.0 0.0 0.0
UPPER ZONE
0.0 0.0 0.0 0.0
CHLORIDE
SURFACE
0.0
UPPER ZONE
0.0
END
0.0
0.0
APPLICATION 273
NITROGEN
SURFACE
1.5 0.0 0.0 0.0 0.0 0.0
UPPER ZONE
16.0 0.0 0.0 0.0 0.0 0.0
PHOSPHORUS
SURFACE
0.25 0.0 0.0 0.0
UPPER ZONE
3.0 0.0 0.0 0.0
CHLORIDE
SURFACE
66.0
UPPER ZONE
60.0
END
LZTEMP= 38.2 36.6 37.1 40.1 48.5 56.5 62.4 65.1 64.5 58.7 51.3 44.3
ASZT= 44.96 BSZT= 0.39 AUZT= 0.0 BUZT= 1.0
SZDPTH=0.1250 UZDPTH= 3.125 BDSZ= 63.70 BDUZ= 72.40 BDLZ= 99.0 UZF=5.0 LZF=1.0
473
-------
TECHNICAL REPORT DATA .
(Please read Instructions on the reverse before completing)
REPORT \o.
EPA-600/3-79-106
I. RECIPIENT'S ACCESSION>NO.
4. TITLE A\D SUBTITLE
Effectiveness of Soil and Water Conservation
Practices for Pollution Control
5. REPORT DATE
October 1979 issuing date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Douglas A. Haith and Raymond C. Loehr (Eds.)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
College of Agriculture and Life Sciences
Cornell University
Ithaca, New York 14853
10. PROGRAM ELEMENT NO.
1BB770
11. CONTRACT/GRANT NO.
R804925010
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Research Laboratory—Athens, Georgia
Office of Research and Development
U.S. Environmental Protection Agency
Athens, Georgia 30605
13. TYPE OF REPORT AND PERIOD COVERED
Final, 10/76-10/78
14. SPONSORING AGENCY CODE
EPA/600/01
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The potential water quality effects and economic implications of soil and water
conservation practices (SWCPs) are identified. Methods for estimating the effects of
SWCPs on pollutant losses from croplands are presented. Mathematical simulation and
linear programming models were used to estimate the effects of SWCPs on edge-of-field
losses of sediment, nutrients, and pesticides and to evaluate the impacts on farm in-
come of implementing SWCPs for pollution control. The models were applied to hypo-
thetical farms and fields in New York, Iowa, Texas, and Georgia.
The major environmental benefit of SWCPs was determined to be their usefulness
in controlling edge-of-field losses of sediment and total phosphorus from croplands.
Farm plans for erosion control, however, are not necessarily the same as plans for
sediment control. SWCPs may not be effective at reducing total losses of dissolved
nitrogen although runoff losses will generally be reduced. These practices will also
reduce losses of sediment-adsorbed pesticides but are not as effective as other tech-
niques. Conservation tillage is often a cost-effective method of reducing losses of
sediment, nutrients, and pesticides because it can be implemented with little or no
decrease in crop yields.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
Water Quality
Agricultural Economics
Water Pollution
Mathematical Models
Management
12A
48G
68D
91A
3. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
480
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
474
------- |