&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

-------