United States
Environmental Protection
Office of Water
Washington DC 20460
                                        July 1987
Guide to
Nonppint Source
Pollution Control

Nonpoint Source
       a guide
 U. S. Environmentai Protection Agency
   Criteria and Standards Division
       Washington, DC

                Regie - .", • "c'•
                T ~7 I 1 ( . .

Prepared under contract 68-01-6986 for the U.S. Environmental Protection
Agency. Research and technical information by Battelle Columbus Divi-
sion; reviewed by Criteria and Standards Division, U.S. Environmental
Protection Agency. Manuscript and design by JT&A, Inc. Approval for
publication does not signify that the contents necessarily reflect the views
and policies of the Environmental Protection Agency, nor does  mention
of trade names or commercial products constitute endorsement or recom-
mendation for use.
PHOTOGRAPHS: The Federal Highway Administration, the National Ag-
ricultural Library, and the Soil Conservation Service and Forest Service
of the U.S. Department of Agriculture.

COVER ART: David Stolar.

Foreword	  v

Introduction  	  1

Evaluation of Modeling and Other Assessment Techniques  . .  5

Nonpoint Source Pollution Models  	  9

Best Management Practices 	33

Sources	111

Glossary  	117

Index: Models and Best Management Practices	121

  This document is designed as a user's guide to the techniques
now available for controlling nonpoint source pollution. The reader
will find first a general evaluation of nonpoint source modeling and
other techniques, followed  by a chapter assessing models now
in use, and a third chapter summarizing best management prac-
  Sources of information used are listed at the conclusion of the
text. In the  third chapter,  end notes  are used to substantiate
studies referred to in the discussion of best management prac-
  This is not, however, a literature search or a summary of re-
search; rather, it has been  developed by intensive research and
distilled into a practical guide for the  decisionmaker who must
choose from among many  techniques for  approaching nonpoint
source pollution control.

Under the Clean Water Act of 1972 this Nation
has steadily progressed toward attaining its na-
tional water quality goals—but now that we have
regulated  the  disposal  of municipal  and in-
dustrial waste, we find that our waters still suffer
from pollutants. It has been easy to understand
how waste flushed into a city's sewers—or dis-
charged directly  by a manufacturer—pollutes
the water it enters. So, using the provisions of the
Clean Water Act, we have cleaned up the waste
that continuously pours from pipes.
   Unfortunately, only half of our pollutants come
from pipes—the rest come from nonpoint sour-
ces: our land, our fields, our streets. One might
explain the difference between point and non-
point as "pipe" and "non-pipe," but the  real dif-
ference lies in the word  "continuous." Nonpoint
source pollution is extremely variable—because
nonpoint source pollution  occurs only  when it
    Rain washes  pollutants from  our land into
our water. Moving  water  is the driving force;
however, the pollutants that enter the water
result directly from man's activities on the land,
and  therefore,  vary greatly both by time  and
space. For example, manure deposited  and
frozen on an  open field during the winter proba-
bly won't release its nutrients into a nearby lake
until spring, when the thaw is followed by rain.
   Understanding   this  relationship  between
hydrology  and the  variability of the specific cir-
cumstances is the  key to understanding non-
poinf source  pollution. This relationship dispels
the mystery of nonpoint source pollution. While
it may not  lend itself to traditional collection and
discharge control methods, nonpoint source pol-
lution can be analyzed statistically. And  that
analysis can point the way to the  solutions—
which can be as diverse as the problems.
  Diverse, nonpoint point sources may be, but
complex they are not. They must be approached
on a logical basis, and they must be solved by
us all. Obviously, we  cannot maneuver all the
nitrogen and phosphorus running off a 40-acre
(or even 1-acre) field  into a sewage treatment
plant that removes these nutrients before they
enter the water. Absurd as that task may be, so is
placing the sole responsibility for water pollution
control on county and  State employees hired to
do the job.
  The water professional's first task is to edu-
cate his public. The terms "runoff" and "nonpoint
source pollution" must become as familiar to the
citizen as "sewers" and "pipes." Citizens must un-
derstand that runoff originates in their yards, on
their farms, and in their streets, and that when
they change some of  the ways they do things,
they prevent this pollution. The widespread insis-
tence  on nonphosphate detergents  is an ex-
ample of public acceptance of such a challenge.
  Backed by a knowledgeable public, State and
local  governments can develop management
strategies to control nonpoint source pollution.
Although nonpoint sources vary  by area, agricul-
tural  pollutants  are the most  pervasive, with
urban sources  next in importance.  In addition,
runoff from highways and waste disposal sites,
failed septic systems, mining and logging activi-
ties, and construction  sites all contribute to the
   Pollutants carried by these nonpoint sources
can be grouped by source and effects:

  • Sediments resulting from erosion (of both
    cropland  and streambanks), livestock ac-
    tivities, and construction  site runoff com-
    prise  the  greatest volume  by  weight of
    materials transported. Sediments—and the

               pollutants  they  carry—eventually affect
               recreational,  industrial,  and   municipal
               water uses  as  well  as aquatic  habitats.
               Sediments can also fill reservoirs, harbors,
               and navigable waterways, often necessitat-
               ing dredging.
               Fertilizers, phosphorous and nitrogen, are
               found in both point and nonpoint dischar-
               ges, and are largely responsible for ac-
               celerating the aging and decay (eutrophica-
               tion) of lakes and streams. Often,  large
               quantities  of  organic  matter  such as
               manure carry  these  nutrients  into  the
               water. The nutrients in themselves are not
               pollutants:  their effects  depend  on  the
               chemical form  of  the  material, the  con-
               centrations, and the physical properties of
               the receiving water.
               Pathogenic (disease-bearing) microorgan-
               isms are introduced from agricultural sour-
               ces such as livestock  feedlots and  from
               leaky septic tank  systems and leach bed
               systems. Although urban stormwater runoff
               also  contributes   measurable   bacteria,
               most bacteria originate from animals rather
               than humans. These microorganisms may
               encourage  the  spread of  infectious dis-
     eases, eventually creating a major public
     health problem.
  •  Pesticides, herbicides, metals,  and other
     toxics, particularly from agriculture, silvicul-
     ture, mining, and lawn and landscape care
     not only  threaten surface water\ but  are
     also being found with increasing frequency
     in ground water. Where the surface is per-
     meable, the water percolates  downward,
     carrying materials in solution with it. Harm-
     ful pollutants carried down into the soil can
     become fixed to clay or soil particles, per-
     haps  eventually entering a  groundwater
     aquifer. Mining  may  expose toxic metals
     that can then be mobilized, usually through
     the high  acidities associated  with mine
     drainage. In northern climates, trace con-
     taminants in road salts also  contribute to
     the release of some metals. Since ground
     water is used more and more for irrigation,
     livestock, and human consumption, the ef-
     fects of these toxic substances must be

   In the 15 years that have passed since the
Clean  Water Act took the first concrete step
toward coping  with the pollution of our  Nation's
                                      POLLUTANT  PATHWAYS
                                                   RUNOFF (SURFACE WATER)
Figure 1. —


waters,  we have  largely  controlled point sour-
ces, and learned  a great deal  about  nonpoint
  Nonpoint  source pollution  affects far more
than the water clarity: studies demonstrate that
controlling nonpoint source pollution produces
economic  benefits  even  beyond the obvious
relationship   between  apparent  lake   water
quality and its use  by swimmers and boaters. For
example, farmers can reduce  cultivation costs
by using conservation tillage;  communities can
cut  dredging  costs  and  improve recreation by
controlling runoff to decrease siltation.
   Now, we find we must also be aware of the ef-
fects  of  nonpoint  source pollution  on human
health—not swimmer's itch that one can avoid by
staying out of the lake, but disease that comes
almost invisibly through our drinking water that
is drawn from ground water. Ultimately, a team
must  be formed to solve this problem: an in-
formed  public backing alert  State and local
professionals who  are  using the most effective
technical solutions to  reduce  nonpoint source
pollution of our surface and ground waters.

      Evaluation of Modeling
      and Other Assessment
  Numerous mathematical models and other as-
sessment techniques have been developed to
help make reliable and cost-effective decisions
about nonpoint source control methods and
their costs,  and to relate land use to pollutant
transport and effects on water bodies. This sec-
tion identifies criteria for selecting and evaluat-
ing these decisionmaking  tools, and describes
the most useful categories for managing non-
point source pollution. Particularly emphasized
are those techniques that  exhibit three  key at-

  • Account for the role hydrology plays in in-
    fluencing pollutant behavior.
  • Address spatial and temporal variability in
    pollutant   generation,  transport,  and
  • Relate contaminant concentrations to best
    mangement practices (BMPs).
  The nonpoint source assessment techniques
either employ statistics or simulate the transport

Nonpoint source models are divided into two
categories—physical and decision-oriented (see
Figure 2). Physical models address the causes
and effects in terms of physical variables of a
process. They use hydrologic characteristics to
estimate pollutant delivery to receiving waters
from land use in agricultural, silvicultural, con-
struction, and urban areas.
  The information developed from these techni-
ques can be used to identify the environmental
effects of nonpoint source pollution. For ex-
ample, by applying a probabilistic analytical tech-
nique to determine the effect of  urban water-
shed runoff on instream water quality, the mean
recurrence intervals for a specific pollutant's con-
centrations during storm events can be com-
puted from representative values for the stream
and runoff conditions.  This information helps
identify how often pollutant concentrations occur
that could be a problem, for example, to the
aquatic organisms in receiving waters. Decision-
makers may use such information to evaluate the
effects of alternative urban runoff management
practices on water quality.
process, and generally estimate either water
quality characteristics or mass transport in non-
point source pollution. Physical models  are
based on deterministic or stochastic simulation of

relevant  physical,  chemical,  and   biological
processes. The methods can range from simple
techniques that estimate average annual pol-
lutant loadings to ones that predict detailed tem-
poral and spatial distribution of pollutants.
  Physical  models  predict  runoff and  mass
transport.  They may  calculate  annual  sum-
maries or time-varying pollutant  loadings, and
may estimate pollutant concentrations in various
environmental media. Physical models  do not
necessarily address every parameter that af-
fects water quality. Users may modify and adapt
them on a case-by-case basis. Many physical
models also have been extended to link a non-
point source pollutant loading rate to different
receiving water bodies.
 Decision-oriented models go beyond the physical
 models to calculate cause and effect relation-
 ships in terms of decision variables. They may
 evaluate the effects of a control practice on water
 quality  and  mass  transport, and also  may
 analyze its environmental impacts based on dif-
 ferent criteria for decisionmaking. A control prac-
 tice  such as  urban runoff treatment  can be
 evaluated based on an analysis of its effects (for
 example, protection of aquatic life).
   In listing the attributes of both types of models,
 Figure 2 shows that decision-oriented techniques
 may go beyond physical models to also
   • stress understanding  of  relationships be-
     tween  management activities and  water
   • focus on ecosystems   and watersheds
     rather  than  on  individual  environmental
   • permit the evaluation of options and trade-
     offs of combinations of BMPs from both ef-
     fects and cost standpoints.
   Assessment tools such  as  the fish  habitat
 models (for example, COWFISH used  by the
 Forest Service) are  decision-oriented. They are
 designed to evaluate the effect of  changes in
 the  riparian  environment  (such  as  might be
 caused by sediment or grazing) on  the aquatic
   Another example of a decision-oriented tech-
 nique is an agricultural nonpoint source  model
 that incorporates BMPs  such as  sedimentation
 and erosion control  systems, plant nutrient loss
 control, and  agricultural waste  management.
 Some of these models can  predict a BMP's ef-
 fects on the ecology and water quality of receiv-
 ing water bodies. Other models may include an
 economic  analysis  component that permits  a
 decisionmaker to compare  capital  investment
 and operation and management costs for various
 BMPs or combinations of them.
 The user's specific requirements determine how
 any model is to be applied. It is important to un-
 derstand that the two categories may be used
 together. That is, one may employ a decision-
 oriented  technique to screen  a number of dif-
 ferent management  practices based on a cost-
 benefit  analysis or  an  environmental impact
 analysis. Subsequently, physical models can be
 used to comprehensively analyze the selected
   Figure 3 illustrates how to evaluate and sel-
 ect a technique. In this approach, one must first
 identify  those  environmental  processes  and
 management techniques that may  prevent or
 mitigate the nonpoint source pollution problem.
 The available models are then characterized as
 shown in Figure 2. Next, it is important to estab-
 lish whether the model is operational and has
 been used successfully.  Nonvalidated  models
 should be used only with extreme caution, if at
 all, because of potential inaccuracies and unac-
 ceptability of results.
   Validated models should then be evaluated by
 comparing  the  information  desired  with  the
 costs of  using the model.  Using  a pollutant
 runoff model is worthwhile if the value of the in-
 formation obtained (ultimately expressed as im-
 proved decisionmaking) exceeds the cost of its
 use.  Examples  of such  values  include  the
 model's ability to simulate the parameters  (pol-
 lutant loadings, runoff, etc.) and the processes
 (snowmelt,  chemical  adsorption,  biodegrada-
 tion, etc.) and to assess the  effect of BMPs on
 aquatic environment. An additional value would
 be an improved data base for future use.
   The costs include acquiring the model, collect-
 ing data, modifying and calibrating the model,
 and using it for various scenarios. These values
 and costs should be carefully studied to deter-
 mine whether a particular model  can achieve
 the expected goals within resources available.

• Predict mass transport and loading

• Calculate average annual or time varying
  pollutant loads

• Estimate pollutant concentrations in various
  environmental compartments
• May simulate chemical/physical/biological
• May predict water quality changes in receiving
• Include BMPs

• Link capital, operation and management cost
  of BMPs to water quality benefits
• Permit risk/benefit analysis of different BMPs
  on beneficial receiving water uses
• Address impacts on the environments
These attributes also may be relevant to Decision-Oriented Techniques.

 Figure 2. —


              Are the Study Objective;;,
              Information Development
                  and Scope of the
               Technique Compatible?
                 Has the Technique
                  Been Validated?
         Model Parameters of Interest
         Model Processes of Interest
           Results/Output Options
        Documentation and Availability
             of User Assistance
      Incorporates BMPs, Policy Choices*
Capital Investment
    O&M Cost
'For Decision Models
                                         Validate After
                                        Studying Values
                                           and Cost
                                       Model Acquisition
                                    Data Gathering/Generation
                                   Model Calibration/Trial Runs
                                       Model Execution for
                                       Different Scenarios
                                                           Cost to Validate the Model
                Figure 3. — Selection of NPS pollution assessment techniques.

                  Nonpoint Source
                  Pollution   Models
  The models discussed  in this section esti-
mate one or more of the following parameters:
(1) runoff, (2) sediment concentrations, and (3)
nonpoint   source  pollutant loads.  Although
Tables 1 and 2 list both physical and decision-
oriented models,  only the  latter are described
here  because they focus  on implementation.
These models are  operational, having  been
used successfully at least once.
   A third category, receiving water models, is
listed in Table 3. Although they are not dev-
eloped  specifically for  nonpoint source pollu-
tion, these models can simulate effects on the
receiving water—an ability some of the physical
and decision-oriented models lack.  Decision-
makers can  use  receiving  water models to
select appropriate implementation models.
  The decision-oriented techniques  establish
relationships between BMPs and water quality,
approach ecosystems and watersheds as an in-
tegrated whole, and, in a few instances, permit
both  a cost and  benefit evaluation of BMPs.
Many  of them are modifications of  physical
models, which are process-oriented, simulating
hydrologic, transport, and other physical, chemi-
cal, and biological processes.
  This assessment of decision-oriented techni-
ques is based on a number of different criteria.
Of course, the model should be operational and
validated by successful use. Techniques under
development  or that need further refinement
should be assessed after  the models  are
  The capability of the model to simulate the fol-
lowing parameters also needs to be considered:
    • Meteorology  (rainfall,   temperature,
      snowfall), hydrology (subsurface flow,
      surface runoff, stream flow), and water
      body (lakes, estuaries, oceans).
    • Spatial  (single   catchment,  multiple
      catchments)  and  temporal  (annual,
      event-based, continuous) simulations.
    • Land use (agricultural, silvicultural, con-
      struction, urban).
    • Policy choices, BMPs,  and  associated
      costs of implementation.
    • Environmental effects on beneficial use
      of receiving waters.
    • Ability to simulate on  a field or land
      management unit basis.
    • Extent   of  input  data requirement
      (detailed, moderate, minimal), type of
      data, and relative availability.
    • Need to modify or calibrate  model for
      specific applications.
    • User-friendliness of the model.
    • Availability of user's manuals, reports,
      and support to facilitate implementation
      of the program.
    • Hardware and software required.
    • Costs associated with purchasing neces-
      sary items, services, and implementa-

          Table 1.-NPS pollution assessment techniques/models: physical models.



         EPA Screening Procedures








To simulate runoff and transport from agricultural lands

To simulate urban watershed runoff, sediment yield and water quality

To estimate nonpoint source loads

To estimate urban runoff using an event-based analysis

To estimate the accumulation of pollutants on urban streets

To evaluate the effects of precipitation, climate and land use or
general basin hydrology

To simulate runoff; and other hydrologic quantities

To estimate the runoff, sediment and pollutant delivery of
urban watersheds

To estimate the concentrations of pollutants and their fate and transport
in the environment

To estimate runoff, sediment and pollutant concentrations in runoff in large
agricultural watersheds
         Table 2.-NFS pollution assessment techniques/models: decision models.




         CREAMS/CREAMS 2


         ESRFPP (Feedlot model)









To simulate sediment, nutrient and pollutant transport in an
agricultural watershed

To simulate runoff and other contributions in streams

To predict hydrologic and erosion response of agricultural watersheds

To simulate hydrologic quantities, erosion and chemical transport

To assess the effect of current and past livestock grazing on associated
aquatic resources

To evaluate and rate the pollution potential of feedlot operations

To assess the effect of sediment yields on stream habitat and fish
populations for planning purposes

To simulate pesticides and nutrients leaching from agricultural watersheds

To continuously simulate hydrologic processes

To evaluate the effect of urban location, management practices, etc.
on urban runoff and receiving water bodies

To evaluate the effects of different land use and management practices
on a small watershed

To estimate runoff and water quality in an urban watershed

To simulate runoff, sediment and nutrient transport in an urban watershed

To evaluate the alternative management decisions used in silviculture

Table 3.-Receiving water models.



















To simulate time varying distributions of sediments and chemicals
in receiving waters

To account for dissolved and steady-state concentrations of pollutants
in the water column and bed sediment

To simulate the unsteady tidal flow and dispersion characteristics
of an estuary

Rapid screening and evaluation of the behavior of synthetic organic
chemicals in freshwater ecosystems

To simulate the transport of sediments and contaminants in rivers
and estuaries

To evaluate the consequences of remedial measures for lakes

To estimate the quantities of metals likely to be in solution

To simulate the advective transport of dissolved and adsorbed pollutants

To evaluate nonlinear, nutrient-algae cycles, multi species
and phytoplankton

To simulate the dispersion and flow characteristics of stream systems
and rivers

To evaluate receiving water by representing physical properties

To predict distributions of sediments and toxic contaminants in rivers

To analyze chemicals in simplified lake and stream settings

To simulate sediment transport, dissolved contaminant transport,
and sorbed contaminant transport

To simulate the behavior of pesticides in a reservoir and bioconcentration
of pesticides in aquatic life

To simulate the transport and transformation of organic toxic chemicals
in the water column and the sediment of stratified lakes, reservoirs, rivers,
estuaries and coastal waters

To allow the specification of time-variable exchange coefficients,
advective flows, wasteloads, and water quality boundary conditions

To simulate the transport and transformation of organic chemicals in the
water column and the sediment of streams and estuaries

     AGNPS is a single-event based model intended to simulate sediment and nutrient
     transport from agricultural watersheds  in Minnesota. It is also being used and
     tested in neighboring States (principally Nebraska and Iowa),  with the intention
     of adding pesticide simulation. The model works on a cell basis, and the water-
     shed (ranging from 2.5 to 23,000 acres) can be divided into  1-acre elements.
     The model predicts runoff volume and peak rate, eroded and delivered sediment,
     nutrient (nitrogen and phosphorous) concentration, and chemical oxygen demand
     in the runoff, and the sediment for single storm events for all the cells in the water-

     • Compares the effects of various BMPs.
     • Analyzes pollutant loads from feedlots.
     • Estimates  water quality parameters at intermediate points throughout  the
       watershed network.
     • Estimates erosion for five different particle sizes (clay, silt, small aggregates,
       large aggregates, and sand).
     • Subdivides transport portion into soluble pollutants and sediment-attached pol-

     • Used only for single events.
     • Has undergone only limited testing for pollutant transport.
     • Not adequately tested for particle size distribution during transport.
     • Does not simulate receiving waters.

Input Data
     Classified into two categories: watershed data that includes information that  ap-
     plies to the entire watershed and to the storm event to be simulated, and cell
     data that includes physical information and parameters based on the land conser-
     vationpractice. Data required may be obtained through Minnesota Land Manage-
     ment Information Service (LL45 Metro Square, 7th and Robert, St. Paul, Min-
     nesota 55101), visual analysis, maps, topographic and soils data, technical publi-
     cations, or the AGNPS manual.

Output Description
     The basic output  from  AGNPS includes  hydrology;  runoff; and  sediment,
     nutrient, and chemical oxygen demand. The output can be examined for a single
     cell  or for the entire watershed. Detailed  sediment  and  nutrient  analyses
     (weighted and mean) are also available.

     AGNPS is a fairly  new model. The manual written by Robert A. Young and
     others in 1986 provides a list of references for additional data and a number of in-
     dividuals to contact for further information. The Guide to Model  Users is in press.
     For further information, contact Robert A. Young, Agricultural Research Service,
     USDA, North  Central Soil  Conservation Research Lab, Morris,  Minn. 56267;
     phone 612/589-3411.

Resource Requirements
     AGNPS is written in FORTRAN IV computer and developed on a Hewlett-Packard
     1000 computer. A IBM-PC compatible version  requiring 256K memory is available
     for monochrome screen (version 1.0)  or graphics (version 1.1).


    ANSWERS is an event-based surface hydrology model that estimates hydrologic
    and erosion response of agricultural watersheds. The area to be studied must be
    subdivided into a finite number of square grids, with parameter values specified
    for each grid. The model simulates interception, infiltration, surface storage, sur-
    face and subsurface flow, and sediment detachment, transport, and deposition.
    Has been extensively validated in the Midwest.

    •  Can evaluate different erosion control management practices for agricultural
       lands and construction sites.
    •  BMPs can be evaluated by modifying soil infiltration values and surface condi-
    •  Modular program structure permits modification of existing program code and
       the addition of user-supplied algorithms.
    •  Grid analysis allows for consideration of spatial variation of hydrologic and sedi-
       ment processes.

    •  Simulates only single events.
    •  Pesticide fate/transport and snowmelt processes cannot be simulated.
    •  Watershed site is small.
    •  Does not process on a land management unit basis.

Input Data
    Input information for the ANSWERS model contains simulation requirements,
    rainfall information, soils data, land use and surface information, channel descrip-
    tions, and individual element information that includes BMPs.

Output Description
    The output listing consists of input data, watershed characteristics, flow and sedi-
    ment information at the watershed outlet, effectiveness of structural BMPs, net
    transported sediment yield  or deposition for each element, and channel deposi-

    Documentation and user support, including a User's Manual,  are available free
    from U.S. EPA Region V. The ANSWERS program was developed by D. B. Beas-
    ley and L. F. Muggins at Purdue University under EPA sponsorship. For further in-
    formation, contact Professor Beasley, Department of Agricultural Engineering,
    Purdue University, West Lafayette, Ind. 47907; phone 317/494-1198.

Resource Requirements
    Written in Fortran, the program will run either on a PC or a mainframe. Requires a
    large memory, especially to simulate large watersheds. The data files have been
    designed to use Soil Conservation Service soil surveys, U.S. Geological Survey
    topographic maps, and crop and management surveys.

    The ARM model simulates the hydrologic, sediment production, pesticide, and
    nutrient processes on the land surface and in the soil profile that determine the
    quantity and quality of runoff  in small agricultural watersheds. The major com-
    ponents of the model individually simulate the hydrologic response of the water-
    shed, sediment production, pesticide adsorption/desorption, pesticide degrada-
    tion, and nutrient transformations.

    • Can simulate surface runoff, subsurface flow, and snowmelt.
    • Both event-based and continuous simulations are available options.
    • Includes different management practices.

    • Application is limited to agricultural watersheds less than 5 km2 (1.9 sq. miles).
    • No channel routing procedures are  included.
    • Model does not link the  cost associated with different BMPs to pollutant load-

Input Data
    Detailed input data are required for simulating hydrology, snowmelt, sediment,
    pesticides, and  nutrients. Some data must be generated from  physical water-
    shed and pollutant characteristics, land surface conditions, agricultural cropping,
    and management practices. Calibration and verification data are also required.

Output Description
    The printout provides summaries of runoff, sediment, pesticides, and  nutrient
    loss in addition to nutrients  remaining in the various soil zones. Generally, the out-
    put summaries are printed daily or monthly, but can be obtained hourly.

    The model was developed  by  Hydrocomp, Inc., Palo Alto, Calif., and is available
    through Tom Barnwell at  the Water Quality Modeling Center,  Environmental
    Research Laboratory, U.S. EPA, College Station  Rd., Athens, Ga. 30613; phone
    404/546-3175. The model is adequately documented.

Resource Requirements
    The model has  been tested on the IBM 370/168 using the FORTRAN  H com-
    piler. The program requires approximately 360K bytes of storage for compilation
    of the largest subroutine, whereas program execution requires up to 230K bytes
    of storage, depending on the model options selected. However, Version  II of the
    ARM model has been adapted to run on a Hewlett-Packard 3000 Series II com-
    puter, which is substantially smaller than the IBM machines on which the model
    was developed and tested.


 Runoff, and
Erosion from

    CREAMS and CREAMS 2 are field-scale models that simulate surface and sub-
    surface runoff, evapotranspiration, erosion, sediment yield, and plant nutrient
    and pesticide delivery. One purpose of these models is to evaluate BMPs. User-
    defined management activities simulated by CREAMS 2 include aerial spraying or
    soil incorporation of pesticides, animal waste management, and alternative
    agricultural practices such as minimum tillage  and terracing. USDA is modifying
    CREAMS 2 to make the model more user friendly. The model does not need to be
    specifically calibrated for a given watershed. Most of the required parameter
    values are physically measurable.

    •  Represents soil processes with reasonable accuracy.
    •  Simulates continuously; considers event loads.
    •  Can simulate up to 20 pesticides at one time.
    •  Includes BMPs.

    •  Subsurface drainage is not simulated.
    •  Data management/handling capabilities are limited.
    •  Maximum size of simulation area is limited to field plots.
    •  Receiving waters are not simulated.

Input Data
    CREAMS and CREAMS 2  require extensive  data on meteorology, hydrology,
    erosion, and chemistry of the pollutants.

Output Description
    Output  can be very detailed. Erosion data are available for each element con-

    Program manuals, tapes, and  floppy disks can be obtained from the USDA-ARS
    Southeast Watershed Research Laboratory, P.O.  Box 946, Tifton, Ga. 31793;
    phone 912/386-3462.

Resource  Requirements
    The model can be run on mainframe computers (IBM) or on personal computers
    (IBM-PC, AT&T).

     The COWFISH model is designed to assist resource specialists analyze the con-
     dition of the riparian environment in relation to past and current livestock grazing
     management and to estimate the compatibility of grazing with associated aquatic
     resources. It is not intended to replace presently used stream  surveys or fish
     population analyses. Rather, it uses existing information to derive an initial indica-
     tion of how livestock grazing may be affecting trout populations.

     The model considers six variables in determining a stream's suitability to support
     trout: (1) the extent of the streambank which is undercut, (2) the extent of the
     stream edge with vegetational overhang, (3) the extent of the streambank show-
     ing bare soil or trampling, (4) stream embeddedness, (5) stream width, and (6)
     stream depth. Two additional variables, stream gradient and  the drainage soil
     type, are used to calculate fish production and recreational and economic value.

     The field value obtained for each variable  is converted to a parameter suitability
     index (PSI) based on principles similar to those developed by the U.S.  Fish and
     Wildlife Service in its habitat suitability index models. The PSI values  are then
     averaged to compare the stream's existing habitat conditions with its  potential
     habitat suitability index.

     • Although originally developed for the mountainous regions of central Montana,
       after some adjustments this model can be used throughout the western United
     • Can be used to determine  stream habitat  productivity any time during the
       season prior to snow cover.
     • Can accurately assess current habitat conditions, provided the sampling area is
       at least 100 feet long.
     • Can be used to evaluate larger sections of uniform streams. Data from five sites
       per stream  mile would be needed to provide a 10 percent sampling  of the
       study area.
     • Can analyze a wide variety of riparian and stream types (the variation being in
       dimensions, flow conditions, streambank conditions, and surrounding environ-

     • Accuracy diminishes when the estimated analysis of grazing effects on fish
       production does not immediately follow the modeled livestock use.
     • Less accurate for use along streams with rocky streambanks that do not follow
       the natural development of undercut banks.
     • When sample  areas smaller than  100  feet are used, the results will  reflect
       population numbers only for the immediate area.

Input Data
     Requires field data that include descriptive information about  the  stream, allot-
     ment, and  sample size being evaluated.  Specifically, information  is needed on
     sample size, vegetative type, side valley slope gradient, percentage of undercut
     banks and banks supporting vegetative overhang, embeddedness, streambank
     alteration, width/depth ratio, and stream gradient.


Cows and

Output Description
    The printout is in tabular form and contains information on the optimum number of
    catcnable trout per 300 m of stream for (1) optional conditions for this stream and
    (2) existing conditions. The losses from optional conditions are also displayed,
    that is, the number of trout per 300 m of stream per year, recreation loss in
    wildlife and fish user days, and economic loss in dollars per 300 m of stream per

    The model was developed by the U.S. Forest Service's Northern Region Wildlife
    and Fish Habitat Relationships Program. Copies are available from USDA Forest
    Services, Federal Bldg., P.O. Box 7669, Missoula, Mont. 59807; phone 406/329-

Resource Requirements
    The user may obtain  the results either manually by following the Guide to the
    Field Form or through a computer by recording the variables in the field and
    using the Data General software program developed for this model. The manual
    procedure provides the flexibility of obtaining the results while still on the site.
    No programing knowledge is required.

     The animal lot evaluation system, developed to evaluate and rate the pollution
     potential of feedlot operations, consists of two parts: (1) a simple screening proce-
     dure that evaluates the potential pollution hazard associated with the feedlot,
     and (2) a more detailed analysis that is better able to identify feedlots that are not
     potential pollution hazards. The Soil Conservation Service's curve  number
     method is used to estimate the  runoff. Chemical oxygen demand and phos-
     phorous are the two parameters used as the pollutant indicators. The ESRFPP
     model has been tested in several States. Currently, the Minnesota Pollution Con-
     trol Agency requires that all animal lots in the State be rated by employing this

     • Estimates pollutant discharge using simple techniques.
     • Can be used as a screening procedure.
     • Considers both surface and groundwater pollution potential.
     • Evaluates the effects of different land management practices.

     • Runoff calculations may  not  be valid for large tributary areas (more than 100
     • Model does not deal with receiving water bodies.
     • The discharge point defined in the model may be difficult to apply in the field.
     • Potential pollution threats to ground water are treated lightly.

Input Data
     Data requirement  is minimal. Most of the required data are presented in the
     ESRFPP manual.

Output Description
     The model estimates the concentration of pollutant indicators at the discharge

     The method is well documented in the ESRFPP manual. Further assistance is
     available from Robert A. Young, North Central Soil Conservation Research Lab,
     Agricultural Research Service, USDA, Morris, Minn. 56267; phone 612/589-3411.

Resource Requirements
     All the calculations can be performed using a small desktop calculator. Programs
     have been developed to use with  Hewlett-Packard 67/97/41C, Monroe 325, and
     Compucorp 327 calculators.
To Rate


    Effects on

    GLEAMS is an extension of USDA's CREAMS models. GLEAMS simulates
    leaching of pesticides and nutrients from agricultural watersheds. The leaching
    behavior of pesticides in root zones has been tested and validated. The model is
    being modified to incorporate subsurface nutrient transport and the effects of
    variability in soil porosity at different depths. The GLEAMS model is in develop-
    ment/testing stage. For more information, contact Walter G. Knisel, Jr., USDA-
    ARS, Southeast Watershed  Research Laboratory, P. O. Box 946, Tifton, Ga.
    31793; phone 912/386-3462.

     Sometimes  referred to as GAWS,  this guide  is not a computer program. It
     provides a standard method for predicting the effect of sediment on stream habitat
     and fish populations for planning purposes. GAWS estimates sediment yields
     resulting from past activities such as fire, road construction, and logging. On-site
     erosion  is  modified according to general characteristics and delivered to a
     stream channel where it is routed to a critical stream reach—a segment of the
     stream that biologists select to predict changes in fish habitat, fish embryo sur-
     vival, summer rearing capacity, and  winter carrying capacity. Model outputs are
     reasonable  estimates that are intended to be used with sound biological judg-
     ment. The model will help land managers quantify existing and potential impacts
     and evaluate trade-offs to fish resources from forest management.

     • Can determine sediment yields.
     • Can predict habitat changes resulting from sediment yields.
     • Can predict fish population changes caused by habitat changes.

     • Average sediment deposition in high gradient streams may not realistically rep-
       resent deposition in fish habitat.
     • The efficiency of high gradient channels for  sediment transport may lead to
       sediment concentration in downstream channels with lower gradient.
     • Increased sediment production from land types drained by high gradient chan-
       nels may have a  greater  impact downstream  than  analysis of  only  high
       gradient channel drainage would indicate.
     • The laboratory-determined response of salmonid fish populations to increased
       sediment levels that the model uses may not approximate the response in the
     • The model was developed and tested only for salmonid species associated
       with the Idaho Batholith. Application in other systems would require testing.

Input Data
     The following types of data are needed to operate this model: (1) estimates of
     sediment yield, (2) substrate core samples to determine existing conditions and
     natural conditions, (3)  measurements  of substrate embeddedness  in critical
     reaches, (4) stratification of stream by channel  type, and  (5) sufficient informa-
     tion on the fish populations.
     The following additional types of data would assist in interpreting the results: (1)
     substrate coring data from several surrounding streams over time, (2) substrate
     embeddedness for most of the fish production areas, (3) redd count or adult es-
     capement data, (4) fish density/standing crop data, (5)  classification of  the
     stream by geomorphic and channel  type, (6) stock-recruitment data over time,
     (7) empirical relationships between sediment yields and fish  habitat, and (8) a
     calibrated watershed model.

Output Description
     No special  output as such. The step-by-step procedure described in the manual
     can be used to calculate changes in the fish populations.
Guide for
to Sediment
Yields in


   Guide for
to Sediment
    Yields in

    This guide was developed by and may be obtained from the U.S. Forest Service's
    Northern Region and Intermountain Region, Federal Bldg., P.O. Box 7669, Mis-
    soula, Mont. 59807; phone 406/329-3101.
Resource Requirements
    Since a computer is not required for using this guide, minimal resources are

     The HSPF is a series of fully integrated computer codes that simulate watershed
     hydrology and the behavior of conventional and organic  pollutants in surface
     runoff and receiving waters. The processes that affect the fate and transport of
     pesticides and nutrients from agricultural land are derived from the Agricultural
     Runoff Management (ARM) model, whereas the sediment delivery algorithms
     are from the Nonpoint Source (NPS) model. The HSPF contains three application
     modules: the PERLND (pervious land) and IMPLND (impervious land) modules
     perform the land and soil simulation for those land surfaces;  the RCHRES
     (reach/reservoir) module simulates the processes that occur in a single reach and
     the bed sediments of a receiving water body (a stream or well-mixed reservoir).
     Extensive and flexible data management and statistical routines are available for
     analyzing simulated or observed time series data.

     • Consists  of systematic modular framework that allows a variety of operating
       modes, including continuous hydrologic simulation.
     • Nonpoint source loading (including alternative control practices) and receiving
       water quality simulation are integrated into a single package.
     • HSPF can analyze  relative contributions and  impacts of both point and non-
       point source loadings; offers options to use either simplified or detailed repre-
       sentations of nonpoint source runoff processes.
     • Models risk assessment of the exposure of aquatic organisms to toxic chemi-
       cals delivered to receiving waters.
     • By adjusting parameters, can include different agricultural management prac-

     • Calibration is usually needed for site-specific applications.
     • A long time period (2-3 months) may be required to learn the operational details
       of applying HSPF if the user has no prior experience.
     • Model does not link the cost  associated with different BMPs to pollutant
     • Depending on the extent of model use, computer costs for model operation
       and data  storage can be a significant fraction (10-15 percent) of total applica-
       tion costs.

Input Data
     If all modules are selected for model  implementation, the HSPF requires an ex-
     tensive amount  of data. The meteorological and some hydrologic data are time
     series inputs.

Output Description
     The HSPF output includes system state variables, temporal variation of pollutant
     concentrations at a given  spatial distribution, and annual summaries describing
     pollutant duration and flux. A summary of time-varying contaminant  concentra-
     tions  is provided  along with the link between simulated receiving water pollutant
     concentration and risk assessment.

     HSPF is in the public domain and can be obtained from the  Center for  Water
     Quality Modeling, Environmental Research Laboratory, U.S. EPA, College Station
     Road, Athens, Ga.  30613; phone 404/546-3175.  The EPA Water Quality Model-
     ing Center provides user assistance on an ongoing basis, periodically scheduling
     free training  sessions.
Program —


Resource Requirements
    HSPF requires a FORTRAN compiler that supports direct access I/O. Twelve ex-
    ternal files are required. The system requires 128K bytes of instruction and data
    storage on virtual memory machines or about 250K bytes with extensive overlay-
    ing on overlay-type machines. It has been installed on several systems, including
    IBM, DEC, VAX, System 10/20, Data General, MV4000, CDC Cyber, HP3000,
    HP1000, and Burroughs and Harris.

     The NPS model continuously simulates hydrologic processes, including snow ac-
     cumulation and melt, and pollutant accumulation, generation, and washoff from
     the land surface. Sediment and other suspended material are used as basic in-
     dicators of nonpoint pollutants. Simulates erosion on both  pervious  and imper-
     vious areas.

     • Can simulate urban, agricultural, and silvicultural nonpoint source pollution.
     • Both event-based and continuous simulations are available options.
     • Can simulate nonpoint pollution from a maximum of five different land use prac-
       tices in a single simulation run.
     • Different agricultural and construction management practices can be simulated.

     • Does not consider subsurface flow, groundwater pollution, or channel proces-
     • Simulates only sediment and nutrient transport processes;  does not estimate
       pesticide delivery.
     • Does not simulate the relationship between cost of BMPs  and runoff as pol-
       lutant loadings.

Input Data
     Requires  extensive input data,  including those related to  model  operation,
     parameter evaluation, and calibration.

Output Description
     The output from the NPS model includes (1) output heading (summary of the
     watershed characteristics, simulation run characteristics, and input data),  (2)
     time interval output (can be for 15-minute intervals) and storm summaries,  (3)
     monthly and yearly summaries, and (4) output to interface with other models (op-

     This model, developed by A. S. Donigian and N. H. Crawford of Hydrocomp, Inc.,
     Mountain View, Calif., under U.S. EPA sponsorship, is available  through EPA's
     Water  Quality  Modeling Center,  Environmental  Research  Laboratory, College
     Station Rd., Athens, Ga. 30613; phone 404/546-3175. The model is well docu-
     mented and  has been tested for the simulation  of  nutrient loadings  in surface

Resource Requirements
     The NPS model was developed  on IBM 360/67 and 370/168 computers. Later,
     it was adopted to UNIVAC 1108, CDC 6000, and Honeywell Series 32. The com-
     puter core requirements for compilation  and execution can be as high as 194K
     and 144K bytes, respectively. With a reasonable level of technical support, it is
     expected that 2 to  3 man-months will be required  to use and apply the NPS


Technique Description
     The NURP is not a computer model: rather, it may be considered to be a statisti-
     cal-based technique that addresses the effects of urban locations, management
     practices, etc., on urban runoff and receiving waters.

     EPA used the NURP to build on prior work on urban stormwater pollution and to
     provide practical information and insights to guide policy and planning. The NURP
     program  included 28 projects, all of which were involved in one or more of the fol-
     lowing activities:
     • Characterizing pollutant types and their effects on receiving water quality.
     • Determining the need for controlling stormwater pollution.
     • Evaluating alternatives for controlling stormwater pollution.

   •  Includes methodologies and information on urban runoff  characteristics and
      controls, stormwater  management, data analysis,  results interpretation,  and
      receiving water quality effects of urban runoff.

   •  Facilitates  decisionmaking by using  qualitative statements, quantitative es-
      timates, and graphic illustrations.
     A large number of reports (about 77) resulted from this program.  The program
     report, Results  of  the  Nationwide  Urban  Runoff  Program,  Complete  Set,
     published by EPA in 1984, consists of the executive summary, final report, appen-
     dices,  and data appendix.  Other reports  include case studies.  The reports are
     available from the National Technical Information Service,  U. S. Department of
     Commerce,  5285 Port Royal Road, Springfield,  Virginia 22161;  phone 703/487-
     4650. For additional information, write to the EPA Headquarters Nonpoint Sour-
     ces Branch, WH-585, 401 M St. SW, Washington, DC 20460. A floppy disk con-
     taining the program also is available from that office.

     U.S. EPA's SWMM package has several versions whose use depends on the
     level of effort available and the amount of information required to estimate runoff
     and water quality in an urban watershed. These versions include:
     •  SWMM—Level I
     •  Simplified SWMM
     •  SWMM

     SWMM—Level I is designed as a screening tool to provide a rough estimate of
     quantity and quality during a precipitation event  that lasts a few hours in an
     urban watershed. The calculations can be performed with  a hand calculator.
     Runoff and water quality parameters are determined on the basis of land use
     characteristics, precipitation, population density, sewerage system, and street
     sweeping operations. A graphic procedure permits the analyst to examine a wide
     variety of control options operating either parallel or in series with one another.

     •  No computer expenses.
     •  Includes economic analysis of sewerage management practices.

     •  Relatively crude model.
     •  Does not consider the water quality changes in storage.

Input Data
     Does not need a great deal of input data.

Output Description
     Minimal output: the calculations are performed with a hand calculator.

     The technical manual for this preliminary screening procedure is available from
     EPA.  Write to the Nonpoint  Source Branch,  WH-585,  401   M  St.  SW,
     Washington, DC 20460.

Resource Requirements
     Requires few resources because the data need is minimal and computers are
     not used.

Level I

    The simplified SWMM simulates runoff and nutrient transport in an urban water-
    shed. The five tasks performed in this model include data preparation,  rainfall
    characterization, storage-treatment balance, overflow-quality assessment, and
    receiving water response.  Each task generally combines  small computer
    programs with hand calculations.

    •  Can be linked to simplified, single-event receiving water model.
    •  Available options include continuous and event-based simulations.
    •  Models effect of pollutant delivery on receiving waters.

    •  Does not consider water quality change or treatment during storage.
    •  Does not simulate snowmelt and sediment transport.
    •  Overflow quantities and qualities must be measured to calibrate model.

Input Data
  Input data include hourly precipitation, runoff coefficient, treatment rate, storage
volume, and receiving water characteristics.

Output  Description
    Output contains time-varying overflows and runoff and summation of these data.
    Pollutant loadings and receiving water response also are included.

    The various SWMM models are available from EPA's Nonpoint Source Branch,
    401 M St. SW, Washington, DC 20460. However, incomplete documentation and
    inadequate user support are problems with using the model.

Resource Requirements
    The computer programs for simplified SWMM have been developed on an IBM
    360/67 digital computer. The storage-treatment program has been used success-
    fully on Xerox  560 and IBM 1130 computers.

     SWMM is a comprehensive, mathematical model capable of representing urban
     stormwater runoff and combined sewer overflow phenomena. Level surface runoff
     generated by precipitation is routed through channels and  pipe networks. The
     analysis uses finite difference approaches. The water quality parameters simu-
     lated in SWMM include sediment and nutrients.

     SWMM contains its own receiving water model, RECEIV. It also can be linked to
     STORM, QUAL-II, and other simplified receiving water models.

     • Both combined and separate sewerage systems may be evaluated.
     • Considers treatment in five different storage systems.
     • Includes several physical and chemical treatment options to evaluate water
       quality changes.
     • Includes capital and operating and management costs for treatment options.
     • Can simulate the effects of pollutant delivery on the quality of receiving waters.
     • Different storage and treatment techniques may be considered as options for
       controlling nonpoint source pollution.

     • Is a large model with complex and detailed input requirements.
     • Statistical summaries are limited.
     • Uses a monthly flow routing method that can be costly because it requires short
       time steps.
     • It is comprehensive, but data management capabilities are not advanced.

Input Data
     Detailed input data  sets are required. Some of the  input information typically re-
     quired includes precipitation, air temperature, wind speed, channel, pipe net-
     works, land use patterns, and storage and treatment facilities.

Output Description
     Output consists of summaries of treatment options and costs, variation of water
     quality with time, and hydrographs and pollutographs with daily and hourly varia-

     The model is well documented and available from EPA in four versions: Final
     Report, Verification  and Testing, User's Manual, and Program Listing. The exist-
     ence of established user groups that meet semiannually is considered an impor-
     tant factor that promotes its use. For a copy, write to EPA's Nonpoint Source
     Branch, 401 M St. SW, Washington, DC 20460.

Resource Requirements
     The computer hardware system should be the equivalent of the IBM 360/65 with
     peripheral storage devices and a usable core capacity of no less than 360K bytes.
     Data  requirements  are common to engineering design  and analysis, and are
     mainly descriptive of the real system.


    SWAM is a continuous simulation model that estimates change in hydrologic,
    sediment,  and chemical characteristics of a small agricultural watershed in
    response to different land use and management practices. The overland flow
    and pollutant transport are estimated by CREAMS 2. The movement and interac-
    tions of sediments, pesticides, and nutrients in the surface drainage (channel net-
    work), and surface retention basins (reservoirs)  are simulated. The model also
    can simulate surface/groundwater interactions.

    •  Is an integrated watershed model that uses a  dynamic version of CREAMS 2
       together with channel, reservoir, and groundwater routing of water, sediment,
       nutrients, and pesticides.
    •  Incorporates backwater effects in channel routing.
    •  Provides a detailed representation of soil and watershed processes.

    •  A complex model to use.
    •  Watershed area is less than 10 km2.
    •  Not practical for long-term (20 years or more) simulations.

Input Data
    Input data include rainfall, soil characteristics, topography, land use, and  manage-
    ment practices.

Output Description
    The model developers are still working on output structure.

    Model is still being tested and has not been released yet.  Intended  as a  re-
    search model that can later be upgraded to basin scale. The model will  be avail-
    able from  Dr.  Donald  DeCoursey, USDA-ARS,  P.O. Box E, Fort Collins, Colo.
    80522;phone 303/221- 0578.

Resource Requirements
    Initially written in FORTRAN for mainframe computers.

    WRENS is a procedural handbook for evaluating the effects of forest-related ac-
    tivities on water quality and making management decisions.

    WRENS  describes  procedures that can be used for quantitatively estimating
    changes in the stream flow, surface erosion, soil movement, sediment discharge,
    and temperature.  In  addition, WRENS  includes  qualitative  assessment of
    forestry activities on pesticide concentration, nutrients, organic matter, and dis-
    solved oxygen. WRENS also  lists criteria for  selecting appropriate control
    measures to minimize adverse effects of forestry activities on water resources.
    Runoff volumes necessary for  water quality analyses can  be  estimated from
    precipitation and daily minimum-maximum air temperature data. One procedure
    (WATBAL) is used to calculate runoff calculation in areas where snowmelt is the
    major source of energy. The other procedure (PROSPER) is used for nonhuman
    areas. WRENS uses the modified soil loss equation (MSLE) to calculate sediment
    erosion in forests.

    •  No computer required.
    •  WRENS handbook is self-contained and includes examples.
    •  Handbook identifies BMPs.

    •  Designed only for small site-specific areas.
    •  Nutrients, pesticides, dissolved oxygen, and organic mater are evaluated only

Input Data
    Input information includes watershed hydrology and daily meteorological data.

Output Description
    No special output as such. Procedures described in the handbook can be used
    to estimate some physical and chemical characteristics.

    Handbook available from the  National  Technical  Information  Service, U.S.
    Department of Commerce, 5825 Port Royal Rd., Springfield, Va. 22161; phone
    703/487-4650. Specify No. EPA 600/8-80-012. Cost: $58.95.

Resource  Requirements
    Resource requirements for this analysis are minimal.
of Nonpoint


Suggestions to Improve Models
  Models can be important tools for assessing
nonpoint source pollution as well as projecting
the effects of BMPs. However, most of the exist-
ing decision-oriented models have distinct limita-
tions and require further refinement. The follow-
ing improvements could be most useful:

    • Develop the ability to simulate proces-
      ses in grids or cells so that one com-
      puter run could evaluate different BMPs
      for  different spatial  distributions of a
      given watershed.
    • Improve the capability for analyzing the
      cost/benefit relationship.
    • Improve the ability to simulate environ-
      mental impact  and risk analysis: for ex-
      ample, pesticide  levels toxic to various
      organisms in receiving waters, eutrophi-
      cation of receiving  water  bodies by
      nutrient  loadings, and effects of bio-
      chemical oxygen demand on dissolved
  oxygen level in streams and consequent
  strain on aquatic organisms.

• Develop the ability to evaluate total im-
  pact resulting from redispersion  of the
  bottom sediments and sorbed or settled
  pesticides and nutrients.

• Improve the models' capabilities for han-
  dling  data and  results  for statistical
  analyses and interpretations.

• Incorporate graphic illustrations in  the

• Simulate nonpoint source pollution  of
  ground water from subsurface runoff.
  This problem, that is, separation of inter-
  flow from surface hydrograph, is not ade-
  quately addressed in most of the exist-
  ing  models. More  fundamental work
  should be done on hydrograph separa-

              Best   Management
  Since  nonpoint source pollution can be at-
tributed to various  land disturbances,  specific
methods  have been developed to minimize both
these disturbances and the runoff they generate.
These methods are known  as BMPs,  for .best
management practices.
  Synonymous with prevention, BMPs use the
land in fhe wisest possible ways—whether it be
for growing crops or grazing cattle, building high-
ways or cutting trees.  BMPs are exactly what
the phrase implies:  coordinated, judicious timing
of activities and use of vegetation and materials
(including some structures),  as components
within a total land management system.
  Rarely viewed as remedies to past problems,
BMPs  function as  ingredients in a  so-called
"prescription"  to control a  specific  nonpoint
source effect.  Perhaps the best example of this
approach can be  seen  in  farming,  where  a
management scenario might combine half  a
dozen  practices and structures—from contour
terraces and no-till  planting to buffer strips and
dikes.  Each specific practice would be  selected
and applied according  to the overall use of the
  Applications for best management practices
overlap: fertilizer and pest management techni-
ques, for example, can be applied to both agricul-
ture and forestry. BMPs also vary with specific
application, with standard practices taking on dif-
ferent forms when used, for example, in a mining
situation as opposed to highway construction.
  In an attempt to describe the most commonly-
used BMPs, this Guide classifies them into
broad categories defined by land use:
  • agriculture
  • construction/urban runoff
  • silviculture
  • mining
  A last classification—multicategory—includes
BMPs that are used in all the others.

  Agriculture: Agricultural BMPs are particular-
ly well-known (and numerous) because agricul-
ture is regarded as the primary source of pollution
to rivers and lakes, contributing at least half of the
sediment deposited in streams and lakes.  The
American Water  Works Association has named
agricultural runoff as the largest contributor of
nitrogen and phosphorus to water.
  More detailed information  about  agricultural
BMPs is  available  from  the National  Water
Quality Evaluation Project (NWQEP), a USDA-
EPA cooperative venture to determine the most
effective  ways for controlling  agricultural non-
point source pollution  at the  watershed level.
NWQEP's 1985  Annual  Report describes 20
projects that studied watershed controls inten-
sively.   The report  is available  from North
Carolina State University at Raleigh.

  Mining: Commercially-mined and processed
minerals range from fuels such as coal and oil to
metals to nonmetallic  minerals like sand  and
gravel, stone, phosphate, and clays.  Mining
operations include deep  mining,  strip mining,
auger mining, open pit mining, placer mining and
dredging, well extraction,  solution mining,  and
other types.
  In addition to the extraction of minerals, mini-
ng involves transport,  exploration, processing,
storage, and waste disposal.  Each phase has
its own activities—and its own potential for pollut-
ing water, either during operation  or following
  Specific  preventive  measures  must   be
designed for each operation on the basis of an
examination of the site and the consequences

expected  from its  use for mining. EPA  has
developed a list of 17 general control principles
that   should   be  applied   to  selecting   and
designed BMPs for mining:
   1. Choose least hazardous methods.
   2. Manage water.
   3. Control erosion and trap sediment.
   4. Segregate water from toxics.
   5. Collect and treat runoff when other  ap-
proaches fail.
   6.  Quickly stabilize disturbed area.
   7.  Properly store minerals and  dispose of
mineral wastes.
   8. Correct pollution-causing hydrologic distur-
   9. Prevent and control pollution from roads.
  10. Avoid disturbing  streambeds, stream-
banks, and natural drainageways.
   11. Use stringent controls in high risk areas.
   12. Apply sound engineering.
   13. Properly locate and seal shafts and bore-
   14. Control fugitive dust.
   15. Maintain control measures.
   16. Use temporary stabilization  and control
when needed.
   17. Prevent and control pollution after close-
   The mining section describes several specific
BMPs  used  in mining that are based  on the
cited control principles.
   The matrix in Table 4 relates the  individual
BMPs  in all categories to one another, and to
the total management plan.
   By combining Table 4 with the text, which in-
cludes cost information for each BMP, the reader
will have the tools to weigh alternative solutions
to specific nonpoint source pollution problems.
 Table 4. — Best management practice activity matrix.
Conservation tillage
Contour strip cropping
Cover crops
Integrated pest management
Range and pasture management
Sod-based rotations
Waste management practices
Structural control practices
Nonvegetativesoil stabilization
Porous pavements
Runoff detention/retention
Street cleaning
Surface roughening
Limiting disturbed areas
Log removal techniques
Ground cover
Removal of debris
Proper handling of haul roads
Water diversion
Block-cut or haul-back
Buffer Strips
Grassed waterway
Devices to encourage infiltration
Material ground cover
Sediment traps
Vegetative stabilization/mulching
V &







' &f



' &









' /









                                   Conservation tillage refers to any planting sys-
                                   tem that reduces soil disturbance and water loss
                                   by retaining crop residues on the land (covering at
                                   least 30 percent) and  leaving the  surface rough,
                                   porous, cloddy,  or  ridged.  Identified under sev-
                                   eral different names—minimum tillage, reduced
                                   tillage, stubble mulching—conservation tillage in-
                                   cludes  no-till,  ridge-till,  strip-till,  mulch-till, and
                                   reduced-till.  Conservation tillage reduces  runoff
                                   and directly benefits  farmers, but  may require
                                   special equipment and additional costs.
     Experience and Application
The conservation tillage system that best fits a
farm operation depends  on the  climate,  soil
characteristics, and the  crops  grown.  While
lower yields have been experienced on some
soils,  research  has  generally  found no   sig-
nificant yield differences  between  conventional
and conservation tillage practices.1 Conservation
tillage also reduces labor,  time, fuel usage,  and
machinery wear. The preferred conservation til-
lage method depends on the soil.  Although ap-
plicable to a wide range of soils,  reduced til-
lage is considered more  suitable to cold  and
wet soils  than  no-till,  with soil drainage largely
determining its economic success. Reduced til-
lage  maintains productivity because it reduces
erosion  and soil compaction and  conserves
moisture, improving soil  structure and optimiz-
ing land resources.  Reduced  tillage is more
widely  adaptable than no-till planting but is
somewhat less effective in controlling water pol-
lution. No-tillage is most applicable  on highly
erodible, well-drained, coarse to medium-tex-
tured soils planted in dormant  grass  or small
grain crops. Farm use of conservation tillage is
increasing rapidly, from about 27 million acres in
1972 to 74 million acres  in 1979, and 99.6 mil-



lion acres in 1985.   The Conservation Tillage
Information Center in 1986 estimated a 2.8 mil-
lion-acre  increase in  conservation tillage  prac-
tices from 1984 to 1985.  The Center also com-
pared 1984 and 1985 according to conservation
tillage type: no-till increased by 700,000 acres;
ridge-till by  45 percent;  strip-till declined by
50,000 acres; and mulch-till, accounting for 64
percent of all conservation tillage, grew by 6.7
million acres'3
                          Capital and  Operating Costs
                     An important reason for the increasing use of
                     conservation tillage  is  the decreased overall
                     capital expense and increased net return.  One
                     study found the average diesel fuel consump-
                     tion for reduced-till was 1.4 gallons/acre and 1.0
                     gallon/acre  for no-till,   compared to  4.2 gal-
                     lons/acre for conventional tilling.4 An analysis of
                     crop production costs indicated that reduced til-
                     lage decreased equipment and labor  costs by

                     •  Effectiveness
                      All conservation tillage practices reduce erosion
                     potential below that of conventional  tillage, in
                     some estimates by 30 percent, with runoff declin-
                     ing by about 61 percent.  Leaving crop residue
                     in  the field protects the soil surface from  the
                     erosive forces of  rain  water  and snowmelt by
                     preventing surface sealing, which increases in-
                     filtration and decreases the volume and velocity
                     of runoff.  Conservation tillage is particularly ef-
                     fective in controlling sediment loss  and phos-
                     phorus  and   pesticide  transport,   although
                     declines in transport vary with the type of soil
                     and reductions in  sediment. On-site effective-
                     ness of preventing these pollutants from enter-
                     ing surface waters ranges from 40 to 90 percent
                     reduction for conservation tillage to 50 to 95 per-
                     cent reduction for no-tillage.7  In South Dakota,
                     researchers found that  as much as 7.5 cm rain-
                     fall produced no surface runoff with conservation
                     tillage, while only 2.0 cm rainfall produced run-
                                              about 8 percent and no-tillage by about 10 per-
                                              cent5  Another researcher concluded that in-
                                              creased material  costs for conservation tillage
                                              practices were more than offset  by reduced
                                              costs for machinery and  items.6 To encourage
                                              conservation tillage,  Virginia  and  North  arid
                                              South Carolina offer farmers a state income tax
                                              credit of 25 percent (to a maximum $2,500 per
                                              year) for purchasing no-till farm equipment.
                                              off with conventional tillage.   Another study
                                              found that surface  runoff averaged 1 mm per
                                              season for no-till plots and 11mm per season on
                                              conventionally tilled plots.9 Conservation tillage
                                              may  increase groundwater contamination if it
                                              relies more on  more pesticide  usage than con-
                                              ventional  tillage.     Research   has  shown,
                                              however,  that  conservation  tillage  does not
                                              necessarily require  more  pesticide usage.10
                                              The Integrated Pest Management BMP discus-
                                              ses this issue in more detail.


                                Plowing follows the contours of the field (perpen-
                                dicular to the slope of the land).  Crops are then
                                planted along these tilled contours.
    Experience and Application
Especially applicable on cropland with 2 to 8 per-
cent slope, contouring provides more protection
from erosion than tilling parallel to the  slope.
Contouring is limited by soil, climate, and topog-
raphy, and may not be usable with large farming
equipment under some topographic conditions.
    Capital and Operating Costs
Contouring costs are slight because it does not
require any specialized farming equipment, nor
does it affect fertilizer and pesticide application
rates.  However, a proper plowing design must
be established.


•  Effectiveness
Contour plowing can reduce average soil loss
(and therefore phosphorus and pesticide runoff)
by 50  percent on  moderate slopes,  less on
steep slopes.  Contouring loses its effectiveness
if the rows break down, so for long slopes ter-
races may be needed. The contours provide ex-
cellent  erosion control during  moderate  rains-
torms,  collecting  water and thereby reducing
runoff velocity and  increasing the infiltration
time.  However, contouring loses its effective-
ness during  extreme storms when rainfall ex-
ceeds the surface storage capacity. When prac-
ticed alone on gentle slopes, or in combination
with  strip cropping  or terracing on moderate
slopes, contouring reduces erosion.

                                                                                     I Strip-

                                  Stripcropping  alternates  plowed strips  of  row
                                  crops  and close- grown crops such as pasture,
                                  hay, or grasses to reduce erosion on tilled soils.
                                  Strips of close-growing crops are planted between
                                  tilled  row crops  to serve as  sediment  filters or
                                  buffer strips in controlling erosion. The system of
                                  cropping where the strips are  laid out nearly per-
                                  pendicular to the direction of the slope is referred
                                  to as contour Stripcropping.
     Experience and Application
Stripcropping  is  particularly  applicable  on
cropland with 8 to 15 percent slope, its primary
advantage  being that it permits row crops on
slopes. Contour Stripcropping is nearly twice as
effective in  controlling erosion as seeding grain
in the fall to replace pasture. However, for con-
tour Stripcropping, the farming area must be
suited for across-slope farming  and for using
pasture as part of a crop rotation.  Alternating
corn  with  spring grain  is  not  effective for


     Capital and Operating  Costs
Costs are slight because contour stripcropping
does not require the purchase of specialized
farming equipment, nor does it affect fertilizer
and pesticide  application  rates.  However, a
farmer must have a use for both crops, and may

•  Effectiveness
The practice reduces the velocity of  the water
as it leaves the tilled areas, because  the buffer
strip  absorbs runoff  and retains soil particles,
thus minimizing nutrient and pesticide entry into
suffer some production loss because of the al-
ternating  system  of  planting.    On  some
topographies, contouring may not be compati-
ble with the use of large farming equipment.
                                                          surface water bodies.  Planting row crop and
                                                          hay in alternate 50- to 100-foot strips reduces
                                                          soil loss by about 50 percent compared to the
                                                          same rotation contoured only.

                                  Cover crops are grown when the ground is  nor-
                                  mally bare, to protect the soil from leaching  and
                                  erosion.   Grasses and  other  crops offer better
                                  protection than row crops such as corn and grain
                                  sorghum.   Crops  that  leave large quantities  of
                                  residue  after  harvest offer more soil  protection
                                  than crops with small quantities of residue.
     Experience and Applications
The cover crop technique is applicable to all
cropland.  In appropriate climates winter cover
crops provide a good base for slot-planting the
next crop.  For example, winter rye can be
seeded immediately after a corn crop is  har-
vested for silage. The growing rye protects the
soil during the fall, winter, and early spring when
the field would otherwise be bare and subject to
erosion.  Many cover crops are left on the soil
as a protective mulch or are plowed under to im-
prove the soil.  Cover crops also may reduce
input costs and increase soil productivity. Heavy
soil cover, such as chopped corn or sorghum
stalks or straw, usually provides as much soil
protection as winter cover crops.
    Capital and Operating Costs
The cost of using cover crops is moderate. No
special equipment is  needed  to  incorporate
winter cover crops into a farming operation;
however, planting the cover crop will involve ad-
ditional man  hours, machine use and main-
tenance, fuel, and seed. Research also has sug-
gested that water  use  by  winter cover may
reduce the following year's cash crop yield.11



Cover crops reduce soil and  water loss,  thus
minimizing loss of  nutrients (and pesticides, if
present).  Winter cover crops reduce erosion
where corn stubble has been removed and low-
residue  crops  harvested.   They may reduce
leaching of nitrate. If the cover  crop is chemical-
ly killed and left in place for no- till planting of a
row crop, it provides excellent control during the
crucial erosion period. One study reported that a
winter cover crop significantly reduced both soil
and water  loss for  no- till corn harvested for
silage,  compared  to growing the corn without
using  winter cover.12  In general,  cover  crops
provide  better protection  from  the  erosive ef-
fects of precipitation than continuous intertilling
of crops. Cover crops reduce pollutant transport
to adjacent surface water bodies by 40 to 60 per-

                                    Fertilizers are used to increase the productivity of
                                    the land. However, the very act of adding them to
                                    the land increases the  potential  amount of pol-
                                    lutants that can be carried away by rainfall.  Judi-
                                    cious use  of fertilizers is therefore advisable to
                                    achieve both increased  productivity and minimal
                                    effect on water.
     Experience and Application
If the quantity and composition of fertilizers ap-
plied are based on crop needs and soil fertility,
plants will use the nutrients, thereby reducing the
amount of nutrients lost in runoff. Fertilizers also
may increase root density, which will make the
soil more permeable.  On the other hand, if too
much fertilizer is used a nutrient imbalance may
result; followed  closely  by heavy rainfall,  the
potential for polluting water bodies  may be
great. Fertilizer transport to water varies with
crop absorption rates, rainfall, slope, soil type,
the closeness to a waterway, and the fertilizer's
propensity for movement by water  or sediment.
When more than an inch of rainfall occurs within
a week of application, the delivery  rate may in-
crease substantially. The consensus reached at
an EPA workshop in June 1986 was that to mini-
mize groundwater pollution, both fertilizers and
pesticides must be better managed. The follow-
ing summary is based  on the  conclusions of
EPA's  Great Lakes National  Program Office
Special Workshop:
   Conservation tillage does not differ radically from
   conventional  tillage,  and  therefore,  fertilizer
   management is similar. The types of chemicals
   may differ, but not the total quantity used: for ex-
   ample, nitrogen fertilizer rates are similar with both.
   However,  nitrogen leaching is a problem for all til-
   lage systems, although  potentially greater for no-
   till. Good fertilizer management can include
   •  optimizing crop planting time
   •  optimizingfertilizerformulation
   •  optimizing time of day for application
   •  optimizing date of application
   •  using lower application rates
     Capital and Operating Cost
Improved fertilizer  management is  extremely
cost effective; in fact, the NWQEP 1985 Annual
Report found this to be the most cost-effective
BMP for reducing nutrient losses in most of the

•  Effectiveness
Fertilizer  (and  pesticide)  management is  em-
phasized  in  all Rural  Clean Water  Program
(RCWP) projects as part of an EPA-USDA effort
to  limit contamination of ground water.  Using
studies described.  Wise  management of fer-
tilizer application can actually reduce the capital
invested in fertilizer, and may reduce the man-
hours and equipment and fuel cost of applying it.
conservation  tillage  without  appropriate  fer-
tilizer/pesticide management is not considered
an acceptable BMP.


                                   Integrated  pest management combines tradition-
                                   al pest  control  methods (such as  crop rotation
                                   and  pesticides)  with   sophisticated  measures
                                   such as trapping  insects and analyzing their life
                                   cycles to determine how best to destroy them,
                                   and monitoring pests to improve the efficiency of
                                   pesticides and other controls. Pesticides are ap-
                                   plied at a  minimal  rate, the method and  timing
                                   carefully selected according to the targeted pest,
                                   and using  pesticides with  the least persistence
                                   and volatility.
     Experience and  Application
To reduce environmental impacts,  agricultural
pesticides  have changed in recent  years, with
EPA also mandating application requirements to
minimize problems. The newer pesticides are
less persistent in the environment and, therefore,
have fewer long- term impacts, but they are also
more likely to be water soluble. This  means that
water (instead of  sediment) may carry them to
the water bodies.  Because their water-soluble
forms may be more biologically available when
freely waterborne than sediment-bound forms,
these toxic chemicals may cause serious short-
term surface  water problems  and  eventually
degrade groundwater resources.
   Pesticide transport  to water bodies varies
with crop adsorption rates,  rainfall,  slope, soil
type, the proximity of the land to a waterway, and
the pesticide's propensity for movement by water
or sediment.  Normally, only about 5 percent of
total  pesticides applied  enter water bodies;
however, when more than an inch of rainfall oc-
curs within a week of pesticide application, the
delivery rate increases substantially and fish may
be killed. With this in mind, it is obvious that pes-
ticides must be managed better for all types of til-
lage, including more care with application rates,
timing, and methods.
  Conservation tillage  may  increase ground-
water contamination. The  consensus reached
at an EPA workshop on this subject in June 1986
was that to minimize groundwater pollution, pes-
ticides and fertilizers must be better managed
for all types of tillage. The following summary is
based on the conclusions of the EPA's Great
Lakes National Program  Office  Special Work-
Conservation tillage does not differ radically from con-
ventional tillage, and,  therefore, pesticide/fertilizer
management is similar.  The types of pesticides may



differ, but not the total  quantity  used: for  ex-
ample, nitrogen  fertilizer rates are similar with
both.  However,  nitrogen leaching is a problem
for all tillage systems, but is potentially greater
for no-till.  Farmers may use more insecticides
and herbicides as they begin to practice conser-
vation tillage, but often  decrease these inputs
as they become  more experienced with the sys-
  Farmers can help control pesticide losses in a
number of ways, including
  • combining  mechanical cultivation with dis-
     ease-resistant crop varieties,
  • trying other pesticides,
  • optimizing   pesticide   placement  with
     respect to loss,
  •  rotating crops,
  •  using resistant crop varieties,
  •  optimizing crop planting time,
  •  optimizing pesticide formulation,
  •  using mechanical control methods,
  •  reducing excessive treatment, and
  •  optimizing time of day for pesticide applica-
  Other practices that have limited applicability
include  optimizing date of pesticide application,
  •  using integrated control programs,
  •  using biological control methods,
  •  using lower pesticide application rates,
  •  managing aerial applications, and
  •  planting between rows in minimum tillage.
                        Capital and Operating Costs
                    The costs  of an integrated pest  management
                    program can vary widely according to practices
                    chosen.   However, pesticide management  is

                    •   Effectiveness
                    An effective integrated pesticide  management
                    program can reduce pollutant loadings by 20 to
                    40 percent, depending on the practices used.
                    A 1983 Report to the  Great  Lakes  Water
                    Quality Board claims a 50 to 90 percent reduc-
                    tion  in pollutant loading, as well as a direct im-
                    pact on instream water quality.  Pesticide and
                                             highly cost effective and maximizes profits and
                                             input costs.14
                                             fertilizer  management is being emphasized  in
                                             all  Rural  Clean  Water  Program  (RCWP)
                                             projects, as part of an EPA/USDA effort to limit
                                             contamination of ground water. Using conserva-
                                             tion tillage  without  appropriate  pesticide/fer-
                                             tilizer management is not considered an accept-
                                             able BMP.

Rangeland and pastureland can be  managed to
reduce erosion.  Lands used for grazing  vary in
climate, topography,  soils, and vegetative type
and condition, a diversity that creates the poten-
tial for varying degrees of erosion. Erosion from
grazing land can be prevented and controlled by
minimizing  the intensity of livestock use, and/or
increasing  the productivity  of  the vegetation.
Grazing management  practices  should  restrict
livestock use to the carrying capacity of the land,
thus minimizing erosion.
•  Experience and Application

Overgrazing changes the soil structure because
the soil compacts, therefore becoming less per-
meable. Overgrazing also can change the den-
sity, vigor,  and species  composition of vegeta-
tion, thus exposing the soil to the erosive forces
of wind and water.
  An  adequate  grazing regime  should use
recommended  stocking   rates,  discourage
animal  congregation in critical areas, break  up
animal  distribution, and  incorporate or remove
manure accumulations.   Following are some
practices recommended  for  rangeland and pas-
tureland management:

  • Rotation grazing permits intensive use of
    fields or portions of fields on an alternating
    basis.  The nonuse period allows vegeta-
    tion to recover before the livestock return.
  • Water supply dispersal distributes the live-
    stock better, thus reducing overuse or over-
    grazing  near water supplies,  and  sub-
    sequent erosion hazards.
  • Ponds in pastures  conserve water while
             providing water for livestock.
             Seasonal  grazing that is compatible with
             the  specific vegetation's most productive
             period permits recovery and reseeding.
             The dispersal and occasional relocation of
             salt, mineral, and feed supplement sites
             avoids  concentrated   overuse of  these



    Capital and Operating Costs
Most of these management practices involve ap-
plying common sense and thoughtful manage-
ment to range and pasture use; they should not
require significant investments by farmers or

• Effectiveness
Restricting animal  access to highly erodible
areas such as bare slopes will reduce erosion
ranchers. However, to use the various practices
most effectively, a farmer or rancher must know
such factors  as stocking rates and vegetation
types and conditions.
                                                         and improve surface water quality.

                                 Sod-based  crop  rotation  involves  planting  a
                                 planned sequence of crops in regular succession
                                 on  the  same land, rather than cultivating one
                                 crop  continuously.   Sod-forming  grasses and
                                 legumes are used, with  hay a part of the cycle.
                                 Meadowless rotation (using crops that do not form
                                 sod) employed by farmers to restore fertility  to
                                 their soils is far less successful than sod-based
                                 rotation in combatting erosion.  However,  mead-
                                 owless  rotations do help  control  disease and
                                 pests and  may  provide  more continuous soil
                                 protection than one-crop  systems.
    Experience and Application
Sod-based  rotations-  may  be used  on all
cropland, particularly those farm operations with
livestock that can eat the hay grown as part of
the cycle. This system has been used for many
years to reduce erosion from the conventional
plow-based systems in regions adapted to rotat-
ing pasture for one or more years.  Soil and
water  loss  from a good  quality grass and
legume meadow is negligible, and plowing the
sod improves infiltration and reduces erosion.
Sod is most frequently used in two- to four-year
rotations. Legumes in the meadow mixture can
help restore the soil's nitrogen balance through
fixation of atmospheric nitrogen.  Sod- based
rotations help control some diseases and pests
and also give the farmer more  fertilizer place-
ment options.  Along with  increased labor re-
quirements,  some climatic restrictions are as-
sociated with sod-based rotation farming.
    Capital and Operating Costs
Sod-based rotations  can  be costly since  the
farm's income is reduced by substituting hay for
as feed.  Other than that, no additional equip-
ment  or  accessories  need  be purchased;



major crops for  one year during the rotation
cycle. This is less of a problem where livestock
are part of the operation, since hay can be used
 •  Effectiveness
 Sod-forming grasses and legume crops used in
 rotation  with row crops are highly effective in
 maintaining the soil structure and tilth.  In addi-
 tion,  this type of  rotation cycle can reduce  soil
 and nutrient losses by 20 to 50 percent.  The
 rotation of crops often provides for planting both
 shallow- and deep-rooted plants; this pattern im-
 proves the  physical condition  and the internal
 drainage of  both the soil and the subsoil.
however, labor  hours  may increase and cash
sales decline.
   Hay fields lose virtually no soil and reduce
 erosion from succeeding crops.  Total soil loss
 is greatly reduced; however, losses are unequal-
 ly distributed over the rotation cycle.  The poten-
 tial for transport of water-soluble phosphorus in-
 creases.  Sod-based  rotations  improve  weed
 and insect control, thus reducing pesticide ap-
 plications and the possibility of groundwater con-

                                   A terrace is a ridge or embankment constructed
                                   across a  slope to  control erosion.  Terraces
                                   reduce the slope and divert  or  store surface
                                   runoff instead of permitting it  to flow uninter-
                                   rupted down  the  slope.  Terraces  break the
                                   length of the slope into shorter segments, reduc-
                                   ing the slope effect on erosion rates by dividing
                                   the field into segments with lesser or even  near-
                                   horizontal slopes. The excess water either  is
                                   conveyed from the terraces to grassed outlets  or
                                   removed by subsurface drains. Some terraces
                                   on permeable soils are designed to stop runoff
                                   and hold the water until it is absorbed.

    Experience and Applications
Terracing is generally applied to fields where
contouring,  strip-  cropping, and tillage  opera-
tions do not protect the soil adequately. Par-
ticularly applicable on land with up to 12 percent
slope, terraces fill a niche in cropland conserva-
tion systems that no other practice can by con-
trolling sheet, rill, and gully erosion on steeper,
longer slopes.  In 1977, terraces were used on


31.3 million acres in the  United States.15  Ter-
race design requires detailed knowledge of prob-
able rainfall totals and intensity, soil characteris-
tics, and cropping systems. Terraces often re-
quire new management  practices to maintain
their desired effects.  They should be planned to
permit farming with large, modern equipment.  If
terraces are improperly designed or used with
poor cultural and management practices, they
may increase rather than reduce soil losses.
Terracing can facilitate more intense cropping
and reduce downstream flood peaks.
                    Capital and Operating Costs
               Terraces involve substantial initial cost, along
               with  some   periodic   maintenance   costs;
               however, in the long term, income usually in-
               creases.  Conventional  gradient terraces have
               been incompatible with use of large equipment,
               but new designs have alleviated this problem.
               Terraces can be constructed with a moldboard
               or  disk  plow,  whirlwind terracer, bulldozer,
               motor grader, scraper,  or similar equipment.
               Many computer  programs  are  available to
               select the best terracing design.

               • Effectiveness
               By  shortening  long  sloping  areas, terracing
               slows runoff and prevents the formation of gul-
               lies, reduces soil loss, and conserves soil mois-
               ture. Terraces have proven to be more effective
               in reducing erosion than in reducing total runoff
               volume  per se.   Farmers report better water
               management,  higher yields,  and  fewer  wet
               spots on terraced fields.  One study reported
               that a terrace with  a vegetative outlet traps 60 to
               80 percent  of the sediment moving into  the ter-
               race channel.16 A terrace with  a closed outlet
               traps 92 to 98 percent of the sediment  moving
                                             into the channel. The Conestoga Rural Clean
                                             Water  Project in Lancaster County,  Pennsyl-
                                             vania, is trying to assess the impact of terraces
                                             on groundwater quality  by monitoring ground
                                             water before and after terraces are built.  The
                                             monitoring has not yet continued long enough tor
                                             any conclusions to be made.

                                  Waste storage structures temporarily store animal
                                  waste until  they can  be safely used.  Although
                                  usually  constructed  of reinforced  concrete or
                                  coated steel, earthen berms may also be used to
                                  build  ponds that can  handle  liquid  and  solid
                                  animal wastes and runoff containing various pol-
                                  lutants.  However, earthen waste storage ponds
                                  may  leak,  thus  increasing  the  potential  for
                                  groundwater pollution.  Diversions and dikes can
                                  be  used to direct  contaminated runoff  into the
                                  storage pond or to route uncontaminated runoff
                                  away from  a  confined animal facility.  Waste
                                  storage ponds  must  be designed to handle ex-
                                  pected precipitation and runoff as well as the es-
                                  timated volume of waste.
     Experience and Application
Animal  waste management practices must be
designed to meet the  site conditions, type of
animal  wastes,  and farm management prac-
tices. In addition to storage facilities for animal
waste,  runoff diversion  structures  are often
needed in  barnyard areas  to reduce waste
transport.  Given the variability of site topog-
raphy and layout of barns, each farm requires
an  individually-designed  waste  management
system. Proper  design  factors also must incor-
porate estimates of land  application frequency
and duration. If convenient to operate, these sys-
tems can directly control  the animal  waste as
well as concentrate the manpower work load.
Piling  or spreading  animal wastes on frozen
ground (commonly done in northern states)  or
during  periods  of high rainfall can increase
nutrient and organic loadings to water bodies. An
interesting  example  of the  effectiveness  of
manure storage has  been demonstrated in an
EPA Clean  Lakes project on the Cobbossee
watershed   near  August,   Maine.  Applying
manure to the fields during the winter months
was identified as the major source of pollution to
the  lakes in the watershed.  A  cost-sharing
program with USDA  along with some pressure
applied by  the  Cobbossee Watershed  District
convinced most of the farmers in the watershed



to construct manure hold facilities: in 1980, some
30  separate  agricultural  waste  management
facilities were built at a total estimated cost of
$662,000. These facilities store manure for ap-
proximaley 80 percent of the animal  units in the
watersheds of the three lakes. Ongoing monitor-
ing  of water quality in the watershed's lakes is
proving  the effectiveness of this  management
system.  Lake  Parker, located  in northeastern
Vermont, suffered from weeds,  algae, and bac-
teria growths over the past decade. The Vermont
Department  of Water  Resources  determined
that the  lake's problems were caused by  exces-
sive phosphorus and bacteria loads from 8 of
the  11 dairy farms located in the watershed. The
Soil  Conservation Service,  of the  town  of
Glover, and other sponsors joined together to im-
plement a Resource Conservation and Develop-
ment Project. All eight farms participated in the
pollution control project, which  included proper
use and disposal  of  wastes  through manure
storage, barnyard runoff control, and milkhouse
waste  management.  The  project  began  in
January 1981 and ended in June 1982. The Lake
Parker Association claims that the  project has
been highly successful; however, like the Cob-
bossee watershed, a longer period of evaluation
will be needed to establish trends in the lake's
                         Capital  and Operating Costs
                    Waste management  practices  are generally
                    costly, requiring farmers to make significant per-
                    sonal investments in agricultural pollution con-
                    trol (see Table 5). In  considering the high cost of
                    building a waste storage structure, farmers must
                    be  assured  that manure  storage  systems not
                    only control  pollution but benefit farm manage-

                    •  Effectiveness
                    Effective  containment  of  animal  waste can
                    reduce phosphorus runoff by as much as 50 to
                    70 percent, thereby minimizing water quality im-
                    pacts and consverving fertilizer for food produc-
                                              ment and farm productivity. Often the motivation
                                              for implementing these practices is not belter
                                              water quality but convenience to the farmer.
                                              Runoff controls are the cheapest waste manage-
                                              ment practice to implement and have been found
                                              to be very cost effective.
                                              tion during the summer months. Waste storage
                                              structures have advantages over less expensive
                                              waste storage ponds such as less potential for
                                              groundwater pollution and fewer odor problems.
                    Table 5. — Typical manure storage facilities and costs.
                                                       MANURE SYSTEM
                             TYPE FARM                 COMPONENTS

                       90 milkers

                       28 milkers
                     Dairy Replacement
                       20 animals
                     Poultry Litter
                     Stacking Site
                       20,000 Broilers
                                  50'x80'x 10'
                                  Concrete storage
                                  with push off ramps
                                  and roof

                                  40' x 40'
                                  Asphalt Pad with
                                  8' Concrete headwall
                                  and earth sides

                                  37' x 37' x 4'
                                  Concrete storage
                                  Asphalted barnyard
                                  Runoff controls:
                                   holding basin
                                   450' diversion
                                  40' x 40 '
                                  Concrete Pad
                                  with earth berms



                       $ 9,240
                     * Costs have been updated to 1985 dollars  Source US EPA, 1980c

  Urban and

                                 Structural  controls are  used  when  vegetation
                                 alone  will  not provide  the desired  degree  of
                                 protection,  or  when  flow  concentrates  in   a
                                 specific  area, as  it does in drainage  courses.
                                 Structural measures include drop spillways, box
                                 inlet spillways, chute  spillways, pipe drop inlets,
                                 sod flumes, debris basins, and other grade con-
                                 trols. These structures supplement sound conser-
                                 vation measures, trap sediment, and reduce the
                                 grade  in water courses,  the velocity of flowing
                                 water, and  peak water flows. Common structural
                                 controls for construction sites include filters (prin-
                                 cipally the gravel inlet and the  filter  berm), traps,
                                 basins, and diversion  structures.  Diversion struc-
                                 tures, basins, and  similar  practices also are used
                                 in other urban environments.
     Experience and Application
Standard designs may be available  for sedi-
ment control structures such as inlet filters and
other small  structures. Structures should  be
built to provide maximum site protection. On
urban  construction sites  and major  highway
projects where storm drains are used, preventing
sediment damage to the  drainage system be-
comes  especially important.  Failure  to  trap
much of the sediment before it reaches an inlet
may  result  in costly damage to the storm
drainage system. It is also important to routinely
remove sediment from settling ponds and sedi-
ment basins, and to dispose of it in a manner
that will preclude its return to downstream areas
during  storms.  Usually  a sediment pond  is
cleaned out when it has  reached 50 percent of
its sediment storage capacity.
  In drainageways, bank protection and grade
stabilization  structures  help  control channel
erosion. Bank protection structures involve either

                                                                                     urban and


   urban and
protective riprap placed directly on the bank or
in-channel structures that deflect or dissipate
the velocity of the flow  impinging on the bank.
Grade stabilization structures consist of a series
of check dams (energy dissipators) that both dis-
sipate the energy of the flowing water and physi-
cally restrict downcutting of the channel.
     Capital and Operating Costs
                 Costs vary widely according to the complexity of
                 the structure and its maintenance requirements.
                 Some  sediment  control  structures  such  as
                 sandbag and straw bale sediment barriers may
                 require daily inspection because they are sub-
                 ject to vandalism. Diversion dikes, filter berms,
                 flexible   downdrains,   interceptordikes,  level

                 •  Effectiveness
                 Based  on the present state of the art, it  is  dif-
                 ficult to assign an accurate universal  effective-
                 ness value to a sediment control practice or com-
                 bination of practices. Many complex factors in-
                 fluence effectiveness, including  soil erodibility,
                 climate, types of control practices being  used,
                 physical characteristics of the sediment, and flow
                 characteristics. Also, adequate standard design
                                              spreaders, sediment retention basins or ponds,
                                              and other structures  require inspection  after
                                              each rainstorm. Actually, the best time to inspect
                                              most structural controls is during a major storm.
                                              Corrective decisions made on-site at that time
                                              can reduce sediment  damages and operating
                                              costs in the long run.
                                              and construction  criteria  are  not available for
                                              many  sediment  control  practices for different
                                              slope  conditions.  Techniques  are  available,
                                              however, for determining trapping efficiency in
                                              large reservoirs.  Sediment basins are generally
                                              designed to have a minimum of 70 percent effec-

                                 Nonvegetative soil stabilization can be either tem-
                                 porary  or  permanent.  Temporary  stabilization
                                 uses covers and binders to shield the soil  sur-
                                 face from rainfall and runoff or to bind the soil par-
                                 ticles into a more resistant mass. Permanent non-
                                 vegetative soil stabilization usually consists  of a
                                 protective  blanket  of  coarse  crushed  stone,
                                 gravel, or other durable materials. On very steep
                                 slopes, more rigid materials are required,  such
                                 as  concrete,  wooden  or metal  retaining  struc-
                                 tures, or concrete or asphalt pavements.
     Experience and Application
Temporary stabilization protects the land during
grading  and construction,  while  permanent
vegetation is developing or until another BMP is
completed. Nonvegetative stabilization can be
used anywhere erosive gradients exist, par-
ticularly at gully headlands. Various  chemical
emulsions recommended for erosion control in-
clude polyvinyl acetate, vinyl acrylic copolymer,
and copolymer emulsions of methacrylates and
acrylates. Hydrated lime and  cement  can be
used for stabilizing clayey soils. Other tem-
porary stabilizers include mulches, nettings, and
textile blankets and mats.
                                             -SSf^"^""'A? <&*"* *

      urban and
• Capital and Operating Costs
The cost of nonvegetative soil stabilization can
vary greatly because there are many available
• Effectiveness
Straw mulch (small grain) has proven effective
on 12 percent slopes at application rates of 5
ton/ha and on a 15 percent slopes at an applica-
tion rate of 10 ton/ha. On-site nonvegetative soil
                                                           stabilization reduces  erosion by 75 to 90per-
                                                           cent.17 Straw mulching loses its effectiveness
                                                           on steep slopes because of rill formation and its
                                                           tendency to be washed away by overland flow.
                                                                             *:.-**'f''*\ %>^W<;?f
                                                                             s•' ^'^.m^^iK;';*

                                 Most porous pavements are made from  asphalt
                                 in which the fine filling particles are missing. The
                                 porous asphalt mixture is installed  on a gravel
                                 base. If the  pavements are designed properly,
                                 most of the runoff can be stored and allowed to in-
                                 filtrate into the ground. Where permeability of un-
                                 derlying natural soil is not adequate or porous
                                 pavement is installed over an impervious  base, a
                                 drainage system  can  be  installed. If the drains
                                 are not installed, the subgrade may soften.
     Experience and Application
First  used in Philadelphia, porous pavements
have been used in a number of States as well
as Denmark and Germany. Porous pavements
can be installed over existing impervious pave-
ments, an advantage in cities with combined
sewers  (because it reduces the frequency of
overflow) or in areas with  inadequate  storm
drainage. The primary benefit of porous pave-
ments is that they significantly reduce runoff from
otherwise impervious areas.  Other benefits in-
clude dissipation of runoff energy and associated
suspended  sediment loss, infiltration of soluble
pollutants and some fine materials, and elimina-
tion of most hydraulic collection systems. Care
should be taken to prevent porous pavements
from polluting ground water.

                                                                                    urban and


   urban and
    Capital and Operating Costs
Porous pavements may sometimes cost slightly
more  than conventional  surfaces for parking
lots, roads, and other urban surfaces; however,
their   benefits  and  savings  on sewer  and
drainage capacities, as well as on treatment, will
offset  any the additional cost of the  pavement.

• Effectiveness
Pollution  loading  by  surface  runoff  from  a
porous pavement should be zero if all water in-
filtrates. However, this may not happen, espe-
cially if the porous pavement is installed on an
impervious surface. In this case, the porous
pavement and the base  act as a filter.  Even
Dust and dirt that accumulate on top of the pave-
ment must be periodically removed or clogging
may occur. Therefore,  porous pavements rely
on proper maintenance to  achieve maximum
                                                           when the ground is impervious, the porous base
                                                           and pavement are beneficial. The gravel base
                                                           serves as a storage area, and if the storm water
                                                           requires treatment, it  may be  stored in  the
                                                           porous  media until treatment capacity becomes

                                                                                            urban and
                                    Runoff  storage  facilities can prevent or  reduce
                                    storm water runoff, keeping associated pollutants
                                    from  entering  combined  sewers  and  surface
                                    water bodies, and,  if properly designed,  ground
                                    water. Detention facilities  treat  or  filter out  pol-
                                    lutants, or  hold  the water for treatment prior to
                                    release. A  retention facility controls  surge flows
                                    such  as runoff from storm events, but provides
                                    no treatment. The  water  in a  retention  facility
                                    either evaporates or infiltrates into the ground.
     Experience and  Application
Many variations of runoff storage facilities are
used on construction sites and in different types
of urban settings. A storage basin or a sewer's
storage can be used to contain storm discharge
or combined sewer overflow before  releasing it
gradually—either after treatment into receiving
waters, or after a storm into a centralized treat-
ment facility. Storage facilities used for this pur-
pose can be either in-line or off- line and include
ponds or surface basins, underground tunnels,
excess sewer storage, and underwater flexible
(collapsible) holding tanks.  A relatively new ap-
proach used in northern Europe and now being
considered by several U.S. cities, is the  use of
floating detention basins in lakes or harbors to
hold runoff for treatment.
   Retention facilities can be important sources
of groundwater  recharge  in highly  impervious
urban areas; however, groundwater  pollution
from these facilities must be considered.
   Detention facilities  can  reduce  the  peak
runoff flow volumes from storms to trap sedi-
ment. Detention basins can be either dry (con-
ventional storm  water management basins) or
wet (designs that maintain a permanent water
pool). Dry basins  are  designed to attenuate
peak runoff  rates and, therefore,  only  briefly
detain portions of flow from  larger storms. Over-
all, detention basins can very effectively remove
some pollutants in  urban  runoff; however, the
design  and the size of the basin in relation to
the urban area served have a critical influence
on this capability.
   Retention basins are increasingly being incor-
porated in  highway designs to contain runoff.
One variation is to construct and contour  high-
way interchanges so that open grassed areas
can be used as runoff storage. However, as with
all simple retention systems, this design can ad-
versely affect ground water. Water quality can be
protected by designing runoff control structures
to prevent  infiltration of pollutants  into ground
water,  and  to provide for settling,  filtration, or
more substantial treatment prior to discharge to
surface water or ground water.
   Runoff storage devices  include small check
dams and  earthen berms that increase the
storage capacity  in  existing streams and wet-
land areas; natural  or excavated depressions
that  store  stormwater flows and  slowly  dis-
charge into  the drainage system; grassed water-
ways (swales) that  moderately improve urban
runoff  quality;  roof tops  and parking   lots
designed to  serve  as  short-term  detention
facilities  in  highly  developed  areas;   and
depressed local recreational areas used for tem-
porary storage of runoff from adjacent areas.
   Designs  and general specifications for reten-
tion basins can be found in a manual developed
by  the Nationwide  Urban  Runoff   Program
(NURP):  Retention/Detention  and  Infiltration
Devices for the Control of Urban Runoff. For a
copy, write  to EPA's Nonpoint Source Branch,
401 M St. SW, Washington, DC 20460.


  urban and
                      Capital and Operating Costs
                 Capital  and operating  costs  largely depend
                 upon the type and size of the runoff storage
                 facility. Check dams and earthen berms may be
                 built inexpensively, but more expensive storage
                 systems (concrete vault storage structure) may
                 require significant excavation and construction.
                   According to NURP, on-site wet ponds serving
                 relatively small urban areas range from $500 to
                 $1,500 per acre of urban area served;  approxi-

                 •  Effectiveness
                 Storage and gradual release of storm water les-
                 sens the downstream  impacts caused by flood-
                 ing, stream bank erosion, resuspension of bot-
                 tom deposits, and disruption of aquatic habitats.
                 Detention basins designed to control a  peak
                 flow rate from a two-year storm can reduce peak
                 flows over 90  percent on the average flow of
                 that storm.  Depending on the storage volume
                 and settling characteristics of suspended  pol-
                 lutants,  detention basins can  reduce pollutant
                 loadings to downstream water bodies. NURP
                 found wet basins provide  more water quality
                 benefits than dry basins.  According to NURP,
                 wet  basins  have  the  greatest  performance
                 capabilities, with pollutant reductions varying
mately $100 to $250 per acre for relatively large
urban areas. The costs include both capital and
initial operating costs; the range reflects differen-
ces in size needed to remove 50 to 90 percent
of particulates. Annual operating costs per acre
of urban area served are estimated  at $60 to
$175 for on-site applications, and $10  to $25 for
off-site applications.18
from excellent  to very poor in the basins that
were monitored. This variability in  effectiveness
demonstrates that the design and the size of the
basin  in relation to the  urban area served are
critically important. As indicated by NURP data,
dry basins are  essentially ineffective for reduc-
ing pollutant loads.
   Wetlands and marshes have proven to act as
giant  sponges  in  removing phosphorous and
suspended  solids from  runoff.   It  has been
demonstrated that  a 7.5-acre wetland retained
77 percent ofall phosphorus and 94 percent of
total suspended solids draining from a 65-acre
sub-watershed of Lake Minnetonka, Minn.19

                                   Street cleaning practices include sweeping streets
                                   and parking  lots  using  mechanical  vehicles or
                                   flushing from  tanker trucks.  Sweeping  handles
                                   coarser dust and litter particles, whereas flushing
                                   carries away the finer fractions.
     Experience and Application
Sweeping is more common in the United States,
whereas street flushing is practiced mainly  in
Europe. Street cleaning or sweeping actually
removes solids from the street, and therefore
reduces the volume, weight, and concentrations
of pollutants that can  reach receiving waters.
Flushing does not remove the particles from the
street; it only moves the street refuse toward the
drainage system. For this reason, flushing has a
negligible effect on reducing pollution. However,
street flushing may be advantageous in areas
with combined sewer systems.
     Capital and Operating Costs
Two types of  sweepers are presently used to
remove solids from impervious urban surfaces.
The  most common design is a mechanical
street cleaner that combines  a rotating gutter
broom with a large cylindrical broom to carry the
material onto a conveyor belt and into the hop-
per. A vacuum-assisted street cleaner uses gut-

•  Effectiveness
Studies indicate that street cleaning is most ef-
fective   in  controlling  heavy  metals,   and
moderately effective in controlling oil and grease,
floating matter, and salts. Street sweeping has
not proven to be effective in controlling sediment,
nutrients,   and oxygen-demanding  materials.
The performance  of a street cleaning program
depends on the condition of the street surface,
the particle size  distribution of pollutants,  the
amount of pollutants present initially, the  num-
ber of passes per  treatment, and the interval be-
tween treatments. Cars parked on congested
urban streets  can reduce the  effectiveness of
street sweeping to zero. Street flushing is not af-
fected by parked  cars. In spite of relatively low
sweeper efficiencies, one study reported water
quality  improved   significantly in areas   with
ter and main pickup brooms to loosen and move
street refuse into the path of a vacuum  intake.
At present, a new mechanical street cleaner can
cost approximately $62,000, whereas a vacuum-
assisted street cleaner is slightly more ($68,000 -
$72,000). The average operator receives $9/hr.
regular sweeping practices. Water flushed from
unswept streets contained on the average 2.3
times more suspended solids and heavy metals
than that from swept and cleaned streets.20 Cur-
rent street sweeping is mostly for aesthetic pur-
poses, with sweepers  removing little of the finer
dust and dirt fractions.
  Based on  five projects that evaluated street
sweeping as a  management  practice to control
pollutants in urban runoff, the Nationwide Urban
Runoff Program (NURP) found  street sweeping
generally ineffective. Four of these projects con-
cluded that street sweeping was not effective for
this purpose  and the fifth, which had a defined
wet and dry season, believed  that sweeping just
prior to the rainy season could reduce some pol-
lution in urban runoff.

                                                                                         urban and

                                 Surface roughening is designed to decrease the
                                 rate  of water  runoff by slowing  the  downhill
                                 movement of water. It is routinely used at  con-
                                 struction sites.  One type of surface roughening,
                                 scarification, roughens the soil along the contour
                                 of a graded slope. Grooves retain runoff and in-
                                 crease  the  rate of water infiltration. Horizontal
                                 grooves also  retain soil additives,  seeds,  and
                                 mulch that may be washed down the slope.  Sof-
                                 tening  the  soil  also  permits  plant  roots  to
                                 develop more  readily.  Other variations of soil
                                 roughening  include tracking and serrating slopes.
      Experience and Application
This practice may be used where the physical
site requires  some means  of runoff control.
Scarification  and serrating  slopes  are  used
more commonly on  gradual slopes,  whereas
tracking  is more adaptable on steep slopes.
Compacted  surfaces from  tracking may  be
more beneficial on long, steep slopes exposed
to high rainfall. The construction sites should be
roughly graded as soon as possible after  ex-
cavation to avoid  the formation of soil  mounds
that easily erode.

                                                                                  urban and


    urban and
     Capital and Operating Costs
Surface roughening is inexpensive to perform but
requires timing and coordination of construction
activities.  Since these practices are carried out
with typical construction equipment (i.e., bulldoz-
ers), no  additional  equipment need be  pur-

•  Effectiveness
Construction typically disturbs areas by removing
vegetation and moving earth from one place to
another. Unprotected soils will be subject to
erosion unless pollution reduction controls are
chased.   Areas disturbed  during construction
can be  roughened  while  the construction  is
taking place, thus minimizing the cost of addition-
al labor to perform the erosion control practice.
                                                             practiced.  Surface roughening is  a very effec-
                                                             tive, cost-efficient method  for preventing soil


                                Limited  disturbance  restricts the  areas where
                                work takes place to the most  effective  use of
                                space,  personnel, and equipment  at any given
                                time. An operational area may be defined by the
                                maximum number of active cut blocks, maximum
                                number of  acres without seeding,  or maximum
                                miles of roads  without permanent  erosion  con-
    Experience and  Application
In large logging areas, it is especially useful to
limit the amount of space in which work is being
done. Exposing large areas of soil to direct rain-
fall may alter the soil's structure, reducing infiltra-
tion rates so that more rain becomes runoff and
overland flow. The use of large equipment also
compacts the soil, resulting in the same effects.
    Capital and Operating  Costs
This practice will generally not increase costs if it
is fully integrated  into operations to make the
most effective  use of  equipment, labor,  and
management. Operating costs may decline be-
cause  management  is  concentrated  on  a
smaller area of operation.



•  Effectiveness
By limiting the logging operation to a clearly
defined area, greater control can be exercised
over the potential causes of nonpoint source pol-
lution, significantly reducing negative effects. The
clear area should be cleaned up as much as pos-
sible as the logging operation moves to the next
defined area. Restoration can begin as soon as
leftover material is disposed of, with reforesta-
tion proceeding even as the logging continues in
another area.

                                  Log transport (yarding) methods that move logs
                                  from the felling location to a landing or transfer
                                  point  can vary drastically in their effect on the
                                  environment.  Yarding methods and the access
                                  roads associated with them are primary causes
                                  of erosion and sedimentation. Various log trans-
                                  port methods include tractor,  high  lead, skyline
                                  cable, balloon, and helicopter.
     Experience and  Application
Research has shown  that silvicultural opera-
tions can  temporarily  change  water quality
characteristics in streams draining forest land, in-
creasing sediment concentrations if erosion ac-
celerates, and stream temperatures if  shade is
removed.  Accumulations  of  slash  (logging
debris)  in a stream can deplete its dissolved
oxygen.  Harvesting or application of pesticides
and fertilizers can increase organic and inorganic
chemical concentrations.21
  Tractor skidding is the most common transport
method employed in the Northeast and South,
and on lands with less than 30 percent slope in
the Intermountain, Northwestern, and California
regions.  To minimize scarification, a skid pan
and a high-wheeled arch yarder are commonly
used, sometimes with a winch to snake logs to
the tractor before skidding them to the yarding
  A  high  lead log  transport system uses  a
metal  tower about 23  meters  (75 feet) high
mounted on a mobile frame. Guy lines hold the
tower in  place, from which a winch and set of
cables drag the logs along the ground to a yard-
ing area, where they are loaded into a truck.



  Skyline  cable  and  balloons  are  recom-
mended  for  clearout logging operations.  A
skyline cable system employs a cable to carry
the   full  weight  of  the   logs  as  they are
transported.   Aerial cables attached to the
towers at opposite  ends  of the logging site
mechanically lift the logs  off the ground and
move them along the cable to the landing area,
which is usually near one of the towers. Skyline
cable systems are typically used for a large vol-
ume of timber and the system can operate at dis-
tances of more than 900  m (3,000 ft.). Cables
should be installed at a height that will ensure
that logs are lifted off of the ground during most
of the transport opertion.
  The balloon method uses a large  balloon
usually filled with  helium and capable of static
lifts of 5 to 10 tons (4.5-9 metric). A cable system
similar to high lead is used to control the horizon-
tal movement of the balloon over the logging
site. A snubbing  line may be required to winch
the unloaded  balloon close to the ground. Bal-
loon  logging is adapted to steep slopes (415 to
90 percent) with unstable soils. A minimum log-
ging  yield of 70 m3 per hectare (12,000 board-
feet per acre) is necessary to justify this type of
log transport.
   Helicopter log transport is  recommended for
transporting  valuable  timber in   inaccessible
areas where  aesthetic values command  high
priority. With this system, the helicopter lifts the
logs from the ground at the point  of felling and
transports them to the loading area.
                        Capital and Operating  Costs
                    Balloons and helicopters transport timber without
                    causing  extensive  soil   erosion  and  sedi-
                    mentation,  but they  are more expensive than
                    the other techniques. Actual cost  is difficult to
                    determine because balloons or blimp-like lifting

                    •  Effectiveness
                    Tractor skidding  is the worst technique in terms
                    of erosion control, and even on level to rolling
                    land tractors will expose  more bare soil than
                    other methods. The  pollution  potential for high
                    lead log transport systems  is generally less than
                    that for tractor skidding, although when logs are
                    repeatedly  yarded over  a high  spot  on the
                    ground, deep profile cuts into the soil may occur.
                      A skyline cable system  has less potential for
                    pollution than the other two systems because it
                    requires  only one-tenth as  much road construc-
                                              vehicles are only in the experimental stage in the
                                              United States. Heavy lift helicopters are very ex-
                                              pensive to purchase and maintain, and their use
                                              has generally been restricted to  more inacces-
                                              sible terrain.
                                              tion as more conventional methods. Lifting logs
                                              off the  ground,  it thereby  avoids  cutting  the
                                              forest floor and confines soil disturbance to yard-
                                              ing and loading areas.
                                                 Balloon and helicopter logging causes mini-
                                              mal soil  disturbance and erosion. Helicopter log-
                                              ging  requires fewer access roads and is a very
                                              versatile system for moving logs from felling sites
                                              to loading areas.

Log Removal


                                Maintaining ground cover in a disturbed area will
                                help prevent erosion while the vegetation even-
                                tually reestablishes. Grass, shrubs, small trees,
                                and  sod provide good ground cover  for forest
                                areas disturbed by logging.
    Experience and Application
In most situations, ground cover can be intro-
duced successfully to areas disturbed by log-
ging operations, usually after logging debris has
been removed from the  ground. Grass cover
can be difficult to  establish on arid or sterile
soils or on all slopes over 1:1. Jute mats or ex-
celsior pads may be required to hold seeds in
these critical areas.
    Capital and Operating Costs
The costs incurred  for this pollution manage-
ment practice include the purchase  of plant
materials and fertilizer, and the labor for plant-
ing and applying the fertilizer.



•  Effectiveness
Vegetation retains moisture, thus preventing dry
soil masses and saturation of the  subsoils.
Vegetative ground cover also protects the soil
from heavy rainfall that  may carry soil particles
downhill. This practice enhances infiltration and
reduces overland water flow, as it encourages
growth of vegetation required to return the site
to prelogging conditions.

                                 Debris  such as tree  tops and  slash must be
                                 properly  disposed  of away  from  waterways.
                                 Streams near logging operations must be properly
                                 and regularly checked for buildup of such debris,
                                 to prevent  it from damming  the waterway  and
                                 scouring sediment from stream banks.
    Experience and Application
In any work near waterways it is very important
to prevent build up of woody debris. Logging
operations may increase the amount of woody
debris in waterways, and it must be cleaned up
during and after timber harvesting. Although all
slash  should  be  removed,  the  old  debris
anchored in the streambed and around which the
channel has formed should be left in place.
    Capital and Operating Costs
Ultimately, this management practice should cost
nothing. The only requirement is enough super-
vision  over logging operations to  ensure the
proper removal and disposal of debris.
  When cleanup has been  necessary,  hand

• Effectiveness
Maintaining stream channels  by preventing the
buildup of debris from logging activities will sig-
nificantly  reduce the impact  on  the water
cleaning costs were estimated at $160 to $800
(updated to 1985  dollars)  per  100  foot of
stream, averaging about $500 per station when
about 5 tons of material were removed.22
course.  Debris  that  deflects  or  constricts
waterflow  accelerates   bank   and  channel
                                                                                  of Debris


                                  Construction,  use,  and  maintenance of access
                                  roads and skid trails expose the ground surface to
                                  rainfall and can alter drainage patterns to intensify
                                  erosive forces. Appropriate  location and  design
                                  of haul roads will help prevent erosion in disturbed
     Experience and Application
Logging roads must be constructed properly to
prevent nonpoint source pollution. Roadways
should be built away from water courses and ac-
cording  to  recommended  guidelines   for
gradient, drainage,  soil  stabilization, and  filter
strips in the area. Where possible, roads should
be routed  across,  rather  than up  and  down
slopes, with artificial drainage where normal pat-
terns are disrupted  seriously. Use  should be
restricted during wet weather. Skid trails should
be limited in number and avoided when logging
on very steep grades.
    Capital and  Operating Costs
Although road construction may cost the logging
company a great deal before operations begin,
it is very important to provide a roadway that
functions as a route for tree removal and also
causes the least environmental  impact.  Road
surfacing will affect operating cost. In one study,
grass (plus fertilizer) cost $5.37/30 m roadbed;
crushed rock (15 cm), $179.01/30 m roadbed;
and 20 cm rock, $265.67/30 m roadbed.23 Gras-
ses may be cheaper but may require more main-
tenance and periodic replanting.
                                                                                     and Trails


and Trails

•  Effectiveness
Properly designed and constructed road and skid
trails help pre- vent erosion in forest areas.
Roadways  must  be  designed  to   consider
erosion potentials from various cross sections of
the road, choosing a crown-with-ditch-and-cul-
vert design that  has the  least  impact on the
area. Roads must be graded and aligned to mini-
mize mileage on steep slopes.
   Road coverings such as grass or gravel also
reduce sediment  loss. A study discovered that
losses of sediment from roadbeds  without sur-
face covering were eight times the losses from
roadbeds covered by 15-20 cm  gravel.24 In  a
study of types of coverings,  the mean soil loss
rates on five  mountainous roadways were es-
timated as follows:
  •  bare soils, 1.2 ton/ha roadbed/cm rain
  •  grass, 0.65 ton/ha roadbed/cm rain
  •  crushed  rock  (15 cm),  0.1  ton/ha road-
     bed/cm rain.25
  Adequate  gravel cover will last for  several
years whereas grasses must be replenished
routinely and may  not be effective  on highly
traveled roads.
  Good practices include closing temporary tim-
ber access roads when not in use,  closing roads
during wet periods of the  year to  prevent sedi-
ment from being produced, and—when  logging
operations  cease—restoring   the  main  haul
roads to conditions that will prevent erosion.


                                 Water  diversion  involves  collecting the  water
                                 before  it  enters  the  mine, then  conveying  it
                                 around the  mine  site. Ditches, flumes,  pipes,
                                 trench drains, and dikes are commonly used for
                                 water diversion. Riprap and  dumped  rock are
                                 sometimes used to reduce  water velocity  in the
                                 conveyance  system.   Ground  water  can be
                                 diverted by using a well to  intercept it  and then
                                 pump it away from its  normal flow path before it
                                 enters the mine, or by using drainways to carry
                                 the ground water out of the mine before it con-
                                 tacts pollution-forming materials.
    Experience and Application
A water  diversion  system should  be  properly
designed to accommodate expected  volumes
and water velocities. If the capacity of a ditch is
exceeded,  water  can  erode  the sides  and
render the  diversion  structure  useless for any
amount of  rainfall.  Ditches are  usually ex-
cavated upslope of the surface mine. Flumes
and pipes are used to carry water down steep
slopes or across regraded areas. Surface water
diversion can be applied to  many waste piles,
diverting water around or conveying it via closed
channels through the waste material.
    Capital  and Operating Costs
Costs  vary according  to the site,  the type of
mine  and  operation,   and  the  design  and
materials  used.  However,  water  diversion
reduces water treatment costs by reducing the

• Effectiveness
Surface water diversion is an effective techni-
que for preventing water pollution: it can be ap-
plied to almost any surface mine orwaste pile.
volume of water that needs to be treated. Where
ground water is concerned, it may be cheaper to
drill holes and pump ground water away than to
treat the water after it passes through the mine.
This procedure decreases erosion and reduces


                                 Underdrains of tile, rock, or  perforated pipe can
                                 be placed  below  pollution-forming  materials  to
                                 quickly   discharge   infiltrating  water.   These
                                 devices shorten the flow path and residence time
                                 of water in the waste materials.
    Experience and Application
Underdrains are designed to provide zones of
high permeability to collect and transport water
from the bottom of the piles. A common construc-
tion method is to use trenches filled with rock.
Underdrains can be used with large tailings ac-

• Capital  and Operating Costs
Construction should be relatively inexpensive,
higher if the underdrains must be installed in ex-
isting waste piles.

• Effectiveness
Underdrains are recommended  for use with
head-of-hollow mining technique, but should not
be used where the pile has been inundated, be-
cause the  underdrains  will drain the  pile and
cumulations, but should be installed before the
pile is created. Care must be taken during design
to preclude the possibility of fines clogging the
underdrain. Filter fabric is used for this purpose.
cause adverse effects. Underdrains should be
used only in piles where the water table fluc-
tuates and flow is in direct response to rainfall.


                                  The block-cut method is a simple innovation of the
                                  conventional contour  strip  mining  method  for
                                  steep terrain. Instead  of  casting the overburden
                                  from above  the coal seam down the hillside, it is
                                  hauled back and placed in the pit of the previous
                                  cut. No spoil is deposited on the downslope below
                                  the coal  seam,  topsoil is saved, overburden  is
                                  removed in  blocks and deposited in prior cuts,
                                  the outcrop barrier is left intact, and reclamation is
                                  integrated with mining.
     Experience and Application
Once the original cut has been made, mining can
be continuous,working in  either one  or both
directions around the hill. The cuts are mined as
units,  thereby making it easier to retain the
original slope and shape of the mountain after
mining. In all cuts, an unmined outcrop barrier
is left to serve as a notch to support the toe of the
backfilled overburden.
    Capital and Operating Costs
The block-cut method is no more expensive and
may  be  less than  conventional dragline pull-
back mining—an estimated 33 cents  per ton

•  Effectiveness
Block-cut mining makes it  possible to mine on
steeper slopes without the clanger of slides and
with minimal disturbance: approximately 60 per-
cent  less total  acreage is disturbed than  by
other mining techniques.  There is significant
visual evidence that the block-cut method is less
damaging than the old practice of shoving over-
burden down the side  of the  mountain. The
less  in Pennsylvania.  Reclamation costs are
lower, as the overburden is handled only once in-
stead of two or three times.
treeline below the mined area  and above the
highwall is preserved, with results of the opera-
tion generally hidden  from  view. Landslides
caused by spoil on the downslopes have been
totally eliminated. Erosion is significantly reduced
and more easily controlled because of concur-
rent reclamation with mining.



                                  Buffer zones are strips of grass or other erosion-
                                  resistant vegetation planted between a waterway
                                  and intensively-used land.  Buffer strips can be lo-
                                  cated on the edges  of  fields or along  stream-
                                  banks. Grass buffer strips reduce the velocity of
                                  runoff and cause suspended sediments to settle
                                  out of storm flows.  Buffer strips reduce nonpoint
                                  pollution from agricultural, construction,  and sil-
                                  vicultural activities.
     Experience and  Application
Vegetative strips trap sediment and pollutants in
surface runoff because they retard water flow,
thereby increasing  infiltration and  detention of
paniculate matter. This technique is applicable
to  all  areas where streams, lakes, and  open
channels exist.
  Grass filter strips provide excellent trapping ef-
ficiency for suspended solids, especially for con-

• Capital and Operating  Costs
Buffer strips require moderate expenditures to
implement and  should  be incorporated  as
needed along natural waterways.
struction sites and farms. Grass strips can also
be used to control barnyard and feedlot runoff
and to reduce soil and pollutant loss from agricul-
tural  fields.  This  BMP  does not address the
source of pollutants, and may remove agricul-
tural land from production and valuable land from
•  Effectiveness
A 2.5 m grass strip can remove a minimum of 85
percent sediment during shallow overland flow.
Although effective in controlling erosion,  most
grassed areas do not reduce pollution. Grass
buffer strips have been used only  recently to
control  agricultural pollution, and, like grassed
waterways, more research is needed to quantify
their effectiveness. Cropped buffer strips on a 4
percent slope have reduced feedlot runoff by 67
percent and runoff of total solids by 79 percent.
In the same study, movement of nitrogen and
phosphorus in  the  runoff decreased by  an
average of 84 and 83 percent, respectively.


                                 Grassed waterways are  natural or constructed
                                 drainage  channels  used to  conduct  surface
                                 runoff. The waterways are usually broad and shal-
                                 low,  covered  with  erosion-resistant  grasses.
                                 Agricultural grassed waterways are used for safe
                                 disposal of runoff from fields, diversions,  and ter-
                                 races, and for other conservation measures. Also,
                                 grassed waterways serve as nonpoint pollution
                                 control in residential and construction areas.
    Experience and Application
Grassed waterways are a basic conservation
practice commonly used by farmers. The ideal
location for  a waterway is a  well-vegetated
natural draw; however, this technique can be
used only on up to 80 percent of all cropland. A
pasture or meadow strip lying in  a drainageway
may be used instead of a constructed or natural
waterway. The waterway design should ensure
that the strip is wide enough to carry the volume
of flow, and that the type and density of vegeta-
tion are adequate to withstand expected flow
velocities. Vegetative cover such as reed canary
grass and Kentucky 31 tall fescue (unmowed)
provide excellent cover.



     Capital and Operating Costs
Although the cost to implement grassed water-
ways is moderate, this technique is probably the
most effective and economic means of convey-
ing water. If waterways are designed properly, in-
channel  erosion should  be minimal, and the
grass lining may even serve as a sediment trap.

•   Effectiveness
Agricultural grassed outlets facilitate drainage of
graded rows and terrace  channels with minimal
erosion.  By encouraging  drainage, this practice
increases runoff volume and decreases  soil in-
filtration. Grassed waterways also allow for dis-
Agricultural grassed outlets involve costs to es-
tablish and maintain and may interfere with the
use of large equipment. Some shaping or enlarg-
ing may be required to handle the increased
flow;  if this is the case, the design and construc-
tion should provide a stable channel.
                                                             posing of excess surface water from construction
                                                             sites and urban areas without causing erosion.
                                                             Grassed waterways can prevent 60 to 80 percent
                                                             of suspended particles from moving to adjacent
                                                             surface waters.

                                 The  most popular method  of increasing infiltra-
                                 tion—the gradual downward movement of water
                                 from the  surface into the  subsoil—is through
                                 trenches or ponds. Trenches usually are shallow
                                 excavations with a  permeable  bottom material
                                 such as gravel or sand laid over a permeable sub-
                                 strate, thereby allowing runoff water to percolate
                                 freely into the ground water. Other  ways to in-
                                 duce infiltration  include  dry wells, wet and dry
                                 ponds, evaporation ponds, and special impound-
                                 ments.  Infiltration  systems  are  used to control
                                 runoff along highways and manage storm water in
                                 urban developments. (See  Runoff Detention/Re-
     Experience and  Application
Infiltration devices can be an acceptable  ap-
proach to managing stormwater runoff, but they
must be limited to good quality storm water so
as to prevent contamination of the ground water.
In addition, adequate clearance  must be main-
tained between the bottom of the structure  and
the groundwater table. Some type of sediment
trap should be provided to remove sediment  and
other debris from the runoff before it enters the
infiltration area.
  Infiltration systems can  replenish  ground
water and increase stream flows during low flow
periods. Their use depends on soil permeability,
aquifer conditions,  and the topography as well
as the depth of the groundwater table. A pervious
subsoil is necessary to dispose of water at an
adequate rate.



    Capital and Operating Costs
These systems may require a high degree of main-
tenance because of the tendency of the coarse bot-
tom material to clog with fine particles. This can be

• Effectiveness
Increased  infiltration  improves recharge  into
groundwater aquifers.  In addition, infiltration may
corrected by routing the water over grass, vegeta-
tive filters, or sediment traps before it enters the in-
filtration area.
                                                              totally remove fine soil particles and other particu-
                                                              lates as well as dissolved solids.

Diversion structures are conveyances designed to
change the direction of water flow. They represent
any modification of the surface that intercepts and
diverts runoff while  increasing  the  distance of
flow to a larger channel system. Basically, diver-
sion  structures are designed to intercept runoff
before it has a chance to come in contact with an
erodible soil surface. Diversion structures include
soil or stone dikes, ditches (or swales), terraces,
and  benches.  These structures can be  tem-
porary or  permanent and should not cause in-
channel erosion.  They often are constructed to
divert up-slope runoff from a source of nonpoint
pollution (disturbed area), and  can be used on
construction sites, urban  areas, or agricultural



     Experience and Application
Diversion structures may be used to protect bot-
tom land from hillside runoff, divert water from
areal sources of pollution, or protect structures
from runoff.  This technique is particularly ap-
plicable  on slopes up to 12 percent and above
feed lots on any slope. The diversion structure
should allow  a shallow, random flow.  It can also
be used at the top of graded slopes to divert of-
fsite runoff away from an erodible surface. For
example, this technique  has been  used to a
great  extent  to  divert  runoff along  highways.
Diversion  structures can  be  used in urban
developments,   but  require   an  adequate
analysis of the hydrology, geology, and soils in
the area. On agricultural lands, they are general-
ly built above cropland fields, gully headcuts, or
other critical erosion areas. Other BMPs should
be  used in  areas subject  to  slumping and
  Diversion structures should have adequate
outlets that will convey runoff without causing
erosion,  such  as   natural  or  constructed
vegetated outlets capable of safely carrying the
discharge, properly designed and constructed
grade stabilization structures,  or storm sewers.
If the diversion structure is used to collect runoff
from a disturbed area, its outlet should open
into a sediment trap or basin.  Disadvantages of
using this technique: it may interfere with cultiva-
tion, maintenance may be required,  and it is
relatively costly.
                       Capital  and Operating Costs
                   Diversions require some  engineering  design
                   and structure, and could be  more expensive
                   than source controls. The cost of a diversion

                   • Effectiveness
                   This technique can be a very effective way of
                   preventing excessive erosion, but only  if  it is
                   designed, installed,  and maintained correctly.
                   Diversion structures can reduce pollutants (par-
                   ticularly  total phosphorus, suspended  solids,
                                             structure could vary greatly from an inexpensive
                                             earth channel to a large concrete diversion.
                                             pesticides and total nitrogen) entering adjacent
                                             water bodies from 30 to 60 percent. The struc-
                                             ture's  effectiveness depends on the physical
                                             characteristics of the development site as well
                                             as climatic factors.

                                  Loose rock, aggregate, mulches, or fabric can be
                                  layered  over an  credible  soil surface,  providing
                                  an excellent erosion control for all  nonpoint sour-
     Experience and Application
Riprap, the common term for loose rock or ag-
gregate, is used where soil conditions, water tur-
bulence and  velocity, expected  vegetative
cover, and groundwater conditions are such that
soil may erode under  certain flow  conditions.
Riprap may be used at such  places as storm
drain outlets, channel banks and bottoms, road-
side ditches, lake shores, and drop structures.
The area to be riprapped can be shaped with
heavy equipment or the rock can be placed along
the bank without any bank modification. Vinyl or
geo-fabric materials, often of fine mesh construc-
tion, should be placed beneath riprap to prevent
flowing  water  from pulling sediments through
voids in the riprap.
   Deflectors constructed of large rocks, logs, or
wire mesh (gabions) also can be put in place to
control water flow.
    Capital and Operating Costs
The cost of using one of these materials  to
prevent  erosion  could  vary  tremendously,
depending  on  the type and  availability of the
material, whether or  not the site  must be
modified by heavy equipment, and how large an
area needs  to be covered. The initial cost  of



material varies from State to State and from com-
munity to community. A good example  of the
variability  in  the  cost  of  riprapping was
demonstrated  in 1982 on three South Dakota
Lakes (Lake Kampeska, Oakwood, and  Swan)
under EPA's Clean Lakes Program. In summary,
Lake Kampeska  cost $39 per foot of shoreline,

•  Effectiveness
The materials described can be very effective in
reducing erosion on unstable banks. A  good ex-
ample of  the effectiveness of riprap was dem-
onstrated on the South Dakota Lakes where, in
Oakwood Lake cost $38.50 per foot of shore-
line,  and Swan Lake cost $5.22 per foot of
shoreline. The low cost of riprapping Swan Lake
can be attributed to the extensive local donations
and the fact that little or no reshaping was done
to the shoreline prior to riprapping.27
                                                              all three cases,  visual  inspections have con-
                                                              cluded that the riprap has protected severely
                                                              eroded  shorelines  and  is  expected  to  dis-
                                                              courage further erosion.

                                Sediment traps are small, temporary  structures
                                used at various  points within or near disturbed
                                areas to detain runoff for a short period and trap
                                coarser sediment particles. They are generally
                                thought of as smaller detention structures used
                                to  trap the coarser sediment. Various types of
                                sediment traps  include  sandbags, straw bales,
                                stone or prefabricated check dams, log and pole
                                structures, excavated ditches, and small pits. A
                                structure called  a  stone  trap  can  be placed
                                across  stream  channels to temporarily  detain
                                flow and trap sediment. This type of trap consists
                                of a dike of randomly placed stone, sized accord-
                                ing to expected flow rates.
    Experience and Application
Sediment traps are considered on-site  prac-
tices. Traps may be placed around storm drain
inlets and in ditches and other small drainage
ways. Sediment should be removed and the
trap restored to its original dimensions when the
accumulated sediment reaches 50  percent of
the trap's depth.



Capital and Operating Costs
                                     spected occasionally, cleaned periodically, and
                                     promptly maintained if they are to function ade-
                                     quately, such costs are minimal.
Sediment traps are usually inexpensive to build
and can be  incorporated  into any construction
project. Although sediment traps  must be  in-

• Effectiveness
When these  structures are positioned at regular    coarser particles can be achieved. They have I it-
intervals along a drainage way or at storm drain    tie effect on retaining fine soil particles and their
inlets,  a high degree of trapping efficiency  for    associated pollutants.

Vegetation can  be established to stabilize  the
soil  either  temporarily  or  permanently.  Tem-
porary stabilization uses fast- growing  annual
and perennial plants that provide interim protec-
tion for less than a year. Permanent stabilization
uses long-lived perennial plants. Selected accord-
ing to specific site conditions, these  plants in-
clude grasses,  legumes, ground cover, vines,
shrubs, and native herbaceous plants and trees.
Areas around vegetative cover can  be mulched
(with  plant residues)  or geo-fabrics can be in-
stalled to  further reduce erosion and protect the
ground. Vegetative  stabilization  and  mulching
are  applicable  to   agricultural,   construction,
urban, mining, and  silvicultural areas.



     Experience and Application
Vegetation is a very desirable material for con-
trolling soil erosion. Vegetative cover performs a
number of important functions,  including shield-
ing  the soil from the impact of the raindrops,
retarding surface flow of water and thereby per-
mitting greater  infiltration,  maintaining a per-
vious  soil surface  capable  of absorbing water,
and removing water from the soil between storm
events by transpiration. The installation of plant
material requires  good  soil preparation and
proper planting techniques. Site factors that may
prevent plants from becoming established on ex-
posed materials typical of construction sites  in-
clude  the  chemical  and physical properties,
steepness of slopes, and the site's biological fea-
  Grasses and legumes are considered superior
to trees,  shrubs, and ground covers for  initial
soil  stabilization because their fibrous root sys-
tems characteristically bind soil particles  en-
courage the formation of water-stable soil ag-
gregates, protect the soil surface from erosion by
water and wind, and grow quickly. Vegetative
cover not only reduces pollution to lakes and
streams but improves the aesthetics of the en-
vironment  and,   where  desirable,   provides
wildlife habitat.
                        Capital and Operating Costs
                    The cost of vegetative stabilization varies greatly,
                    depending on the size of the area to be stabilized
                    and the difficulty of surface preparation as well as
                    the  type of vegetation used.   Ground  cover
                    along  highways can be much  less costly than
                    elaborate  landscape design incorporated  into
                    urban  developments. After the  vegetative cover

                    •  Effectiveness
                    Vegetation in any form is a very effective means
                    of controlling soil erosion. However, the condition
                    of the installed vegetation will determine its effec-
                    tiveness. For  example, a cover of  vegetation
                    that is not properly established  or maintained
                    will  not be fully effective  in controlling erosion.
                                             becomes established, regular maintenance is re-
                                             quired to achieve a long-term cover that ade-
                                             quately controls soil erosion. However,  plant
                                             materials vary in the amount of maintenance re-
                                             quired to sustain them. For example, on inacces-
                                             sible slopes, a low maintenance cover would be
                                             Various grass species reduce soil loss by 95 to
                                             99 percent compared to bare surfaces. Overall,
                                             research must continue into determining the ef-
                                             fectiveness  of  vegetative  stabilization  as an
                                             erosion control practice.


 End  Notes
  1Wendt, R. C. and R. E. Burwell. 1985. Runoff and
  soil losses for conventional, reduced, and no-till corn.
  J. Soil Water Conserv. Sept.-Oct. 450-4.
  2Novotny, V., and G. Chesters. 1981. Handbook of
  Nonpoint  Pollution Sources and Management.  Van
  Nostrand   Reinhold   Environmental   Engineering
  Series, New York.
  ^Conservation Tillage Information Center. 1986.1985
  National Survey Conservation Tillage Practices Ohio
  County Summary. Natl. Ass. Conserv. Distr.
  4Mueller, D., T. Daniel, and R. C. Wendt. 1981. Con-
  servation tillage: Best management  practice for non-
  point runoff. Environ. Manage. 5(1 )33-53.
  5Loehr, R. C. 1984. Pollution Control for Agriculture.
  2nd ed. Academic Press, Inc., New York.
  6Yaksich,  S.  M.  1983. Summary report of the Lake
  Erie  wastewater management study. U.S.  Army
  Corps of Engineers, Buffalo District, New York.
  7Report to the Great Lakes Water Quality Board. 1983.
  Nonpointsource abatement in the Great Lakes basin.
  An overview of post— PLUARG development. Interna-
  tional Joint Commission. Windsor, Ontario.
  8Onstad, C. A. and T. C. Olson. 1970. Water budget
  accounting on two corn-cropped watersheds.  J. Soil
  Water Conserv. 25:150-152.
  9Harrold, L. L, and W. M. Edwards.  1970. Watershed
  studies of agricultural pollution. Ohio Rep. 55(4):85-
  10Myers, Carl F., July 9,  1986. Memorandum To: Wil-
  liam A. Whittington, Director, Office  of Water Regula-
  tions and Standards (WH-551). Subject: Pesticides in
  groundwaterfrom agricultural use.
  11Novotny, V., and G. Chesters. 1981. Handbook of
  Nonpoint  Pollution Sources  and Management. Van
  Nostrand   Reinhold   Environmental   Engineering
  Series, New York.
  12Wendt, R. C., and R.  E. Burwell. 1985. Runoff and
  soil losses for conventional, reduced, and no-till corn.
  J. Soil Water Conserv. Sept.-Oct. 450-4.
  13Report to the Great  Lakes Water Quality  Board.
  1983. Nonpoint source  abatement in the Great Lakes
basin. An overview of post— PLUARG development.
International Joint Commission. Windsor, Ontario.
14Report to the Great Lakes Water  Quality  Board.
1983. Nonpoint source abatement in the Great Lakes
basin. An overview of post— PLUARG development.
International Joint Commission. Windsor, Ontario.
15'16Highfill, R. E. 1983. Modern terracing systems. J.
Soil Water Conserv. 38(4) :336-8.
17Report to the Great Lakes Water  Quality  Board.
1983. Nonpoint source abatement in the Great Lakes
basin. An overview of post— PLUARG development.
International Joint Commission. Windsor, Ontario.
18U.S.  Environmental  Protection  Agency.  1983.
Results of the  nationwide urban runoff program.  Ex-
ecutive summary.  Water Planning Division, U.S.  En-
viron. Prot. Agency, Washington, DC. Natl. Tech. In-
form. Serv., Springfield, VA. NTIS#PB84-185545.
19Hickock, E. A., M. C. Hannaman, and N. C. Wenck.
1977. Urban Runoff Treatment Methods. Vol.  I. Non-
structural wetland treatment. EPA-600/2-77-217. U.S.
Environ. Prot. Agency, Washington, DC.
20Malmquist, Per-Arne. 1978. Atmospheric fallout  and
street cleaning—effect  on  urban storm  water  and
snow. Prog. Water Tech. 10:495-505.
21Brown, G. W. 1985. Controlling nonpoint source  pol-
lution from silvicultural operations: What we know  and
don't know. Pages 332-3 in Perspectives on Nonpoint
Source Pollution,  Proc. Natl. Conf.  EPA 440/5-85-
001. U.S. Environ.  Prot. Agency, Washington, DC.
22Dykstrat, D.P., and  H.A. Forehlich. 1976. Cost
of stream protection  during timber harvest. J.
Forestry October: 684-7.
23,24,25Swjft> L w  Jr  1984  gravel and grass surfac-
ing reduces soil loss from mountain roads. Forest Sci.
26Young, R., T. Huntrods, and W. Anderson. 1980. Ef-
fectiveness of vegetated buffer strips in controlling pol-
lution from feedlot runoff. J. Environ. Qual. 9(3): 483-7.
27Final Report: Swan lake restoration project. 1982.

Models and Assessment Techniques
Alley, W. M., and P. E. Smith.  1982a.  Distributed
   routing rainfall-runoff model - version II. Computer
   program documentation user's manual. Open File
   Rep. 82-344. U. S. Geological Survey, NSTL Sta-
   tion, MS.

	. 1982b. Multi-event urban runoff quality
   model, computer program documentation user's
   manual.  Open File Rep. 82- 764. U. S. Geological
   Survey, NSTL Station, MS.

Ambrose, R., S. Hill, and  L. Mulkey.  1983.  User's
   manual for the chemical transport and fate model
   TOXIWASP. Version I.  Draft document. Office of
   Research and Development, U.S. Environ.  Prot.
   Agency, Athens Research Laboratory, Athens, GA.

Beasley, D. B., E. J. Monke and L. F. Muggins.  1980.
   ANSWERS:  A  model  for watershed planning.
   Trans. Am. Soc. Agric. Eng. 23(4): 938-44.

Beasley, D. B. and L. F. Muggins.  1981. ANSWERS
   users manual. EPA 905/9-82-001.  U.  S. Environ.
   Prot. Agency, Region V. Chicago, IL.

Burns, L. A., D. M. Cline,  and R. R. Lassiter.  1982.
   Exposure analysis modeling  system  (EXAMS)
   user manual and system  documentation.   EPA
   600/3-82-023. U.  S.  Environ. Prot. Agency,
   Athens, GA.

Computer Sciences Corporation.  1980.  Pesticide
   Runoff Simulator User's Manual.  Office of  Pes-
   ticides and Toxic Substances, U.S. Environ.  Prot.
   Agency, Washington, DC.

Connolly, T.  P.   1982.  Preliminary estuary and
   stream version documentation of WASTOX.  EPA
   Cooperative Agreement No. R807 827- 02.  U.S.
   Environ. Prot. Agency, Gulf Breeze, FL. Manhattan
   College, Bronx, NY.
DeCoursey,  D.  G.  1982.  ARS small  watershed
  model.  Paper No. 82- 2094. Am. Soc. Agric. Eng.
  Summer  Meeting,  University  of  Wisconsin,
  Madison, Wl. June 27-30.

DePinto, J. V., W. L. Richardson, K. Rygwelski, et al.
  n.d. Technical guidance manual for performing
  waste load allocations. Draft rep. U.S. Environ.
  Prot. Agency,  Washington, DC.

DiToro, D. M., J. J. Fitzpatrick and R. V. Thomann.
  1983.  Documentation  for water quality analysis
  simulation program (WASP) and model verification
  program (MVP).    Prepared  for the Office of
  Research and Development,  U.S. Environ. Prot.
  Agency, Duluth,   MN,  by  Hydroscience,  Inc.
  (presently HydroQual, Inc.), under Contract No. 68-

Donigian, A.  S., Jr. and N. H. Crawford.   1977.
  Simulation of nutrient loadings in surface runoff
  with the NPS  model. EPA 600/3-77-065. U. S. En-
  viron. Prot. Agency, Athens, GA.

	.  1979.  User's manual for the nonpoint
  source (NPS) model.  U.S. Environ. Prot. Agency,
  Environmental Research Laboratory, Athens, GA.

Donigian, A. S.,  Jr.,  D. C. Beyerlein, H. H. Davis, Jr.
  and N. H. Crawford.   1977.   Agricultural runoff
  management  (ARM) model version II: Refinement
  and testing.  EPA 600/3-77-098. U. S. Environ.
  Prot.    Agency,    Environmental   Research
  Laboratory, Athens, GA.

Donigian, A.  S. Jr., and D.  C. Beyerlein.   1985.
  Review and  analysis  of available NPS and in-
  tegrated watershed models. Woodward-Clyde Con-
  sultants, Walnut Creek, CA.

Donigian, A. S. Jr., J. L. Baker, D. A. Haith, and M. F.
  Walters. 1983.  HSPF parameter adjustments to

  evaluate the effects of agricultural best manage-
  ment practices. U.S. Environ. Prot. Agency.  PB83-
  247171  Natl. Tech. Inf. Serv., Springfield, VA

Evenson, D. E., J. R. Monser, W. R. Norton, and W. R.
  Roesner. 1974.  Computer program documenta-
  tion for the stream quality model QUAL-II. Prepared
  for U.S. Environ. Prot. Agency, Systems Develop-
  ment Branch, Washington, DC, by Water Resour-
  ces  Engineers, Inc., under Contract No.  68-01-

Felmy, A. R., S. M. Brown, Y. Onishi, R. S. Argo and
  S. B. Yabusaki. 1982. MEXAMS - The metals ex-
  posure  analyses   modeling  system.   Battelle,
  Pacific  Northwest  Laboratories,  Richland,  WA.
  Contract No. 68-03-3089.

Frere, M. H., C. A. Onstead and  H. N. Holton. 1975.
  ACTMO—An  agricultural  chemical  transport
  model.   Publ. No.  ARS-H-3.  Hyattsville,  MD.
  Agricultural Research Service. U. S. Dep. Agric.
Genet, L A., D. J. Smith, and M. B. Sonnen.  1974.
   Computer program documentation for the dynamic
   estuary model.  Report by Water Resources En-
   gineers, Inc., Walnut Creek, CA, to U. S. Environ.
   Prot. Agency,  Systems Development  Branch,
   Washington, DC.

Haith, D. A., and L. J. Tubbs. 1981. Watershed load-
   ing functions for nonpoint sources. J. Environ. Eng.

Heaney, J.  R., W. C.  Huber, and S. J.  Nix.  1976.
   Storm water management model, level I. Prelimi-
   nary screening procedures.  EPA 600/2-76-275.
   U. S. Environ. Prot. Agency, Cincinnati, OH.
 Huber,  W. C., J. P. Heaney, M. A. Medina, W. A.
   Peltz, H. Sheikh, and G. F. Smith. 1975.  Storm
   water management model user's manual, version
   II. EPA 670/2-75-017.  U.S. Environ. Prot. Agen-
   cy, Cincinnati, OH.

 HydroQual, Inc. 1981. CTAP documentation - chemi-
   cal transport analysis  program   Prepared  for
   Chemical Manufacturers Association, Washington,

 	.  1982.  Application guide  for CMA -
   HydroQual  chemical fate models.  Prepared  for
   Chemical Manufacturers Association, Washington,
  Johanson, R. C., J. C. Imhoff, H. H. Davis, J. L. Kittle,
    Jr. and A. S. Donigian, Jr. 1984. User's manual tor
    hydrological   simulation  program  -  FORTRAN
    (HSPF):  Release 7.0. U. S. Environ. Prot. Agen-
    cy, Athens, GA.
Knisel, W. G.   1986.   Personal  communication.
   Southeast  Watershed Research  Laboratory,  Ar-
   gicultural Research Service, U. S. Dep. Agric.  Tif-
   ton, GA.

Leavesley, G. H. R. W. Lichty, B. M. Troutman and L.
   G.  Saindaon. 1983. Precipitation-runoff modeling
   system:  user's manual. Water Resources Inves-
   tigation Rep. 83-4238. U.  S. Geological  Survey,
   Denver, CO.

Lynch, J., E.  Corbett and K.  Mussallem. 1985.  J.
   Soil Water Conserv. 40(1 ):164-7.

McElroy, A. D. S. Y. Chiu, J. W. Nebgen, A. Aleti and
   F. W.  Bennett.  1976.  Loading functions for as-
   sessment of water  pollution from  nonpoint sour-
   ces.   EPA 600/2-76-151.   U.  S. Environ. Prot.
   Agency. Washington, DC.

Mills, W. B., J. D. Dean, D. B. Porcella, S. A.  Gherini,
   R. J. M. Hudson, W. E. Frick, G. L. Rupp and G. L.
   Bowie.   1982.  Water  quality  assessment:   A
   screening procedure for toxic and conventional pol-
   lutants. EPA 600/6-82-004a,b.  Vol. I and II. U. S.
   Environ. Prot. Agency, Washington, DC.

Onishi, Y.  1981.  Sediment-contaminant transport
   model.   J. Hydraul.  Div.  Am.  Soc. Civil Eng.
   107(HY9) Proc. Pap. 16505:1089-1107.

Onishi, Y., and S. E. Wise.  1982.  Users manual for
   the  instream  sediment-contaminant transport
   model  SERATRA.   EPA 600/3/82-055, U.S. En-
   viron.  Prot.  Agency,  Environmental  Research
   Laboratory, Athens, GA.

Onishi, Y., G. Whelan, and  R. L. Skaggs.  1982.
   Development  of a multimedia  radionuclide  ex-
   posure assessment  methodology for low- level
   waste  management.  PNL-3370. Battelle Pacific
   Northwest Laboratory, Richland, WA.

Onstad, C. A. and T. C. Olson. 1970.  Water budget
   accounting on two corn-cropped watersheds.  J.
   Soil Water Conserv. 25:150-2.

Orlob, G.  T.  1975.  Ecological simulation for aquatic
   environment. In B. C. Patten, ed. System Analysis
   and Simulation in Ecology. Vol. III.  Academic
   Press,  NY.

 Park, R.  A., C. D. Collins, D. K. Leung, C. W. Boylen,
   J.  Albanses,  P. deCaprariis,  and H.   Forstner.
   1978.      The    aquatic  ecosystem    model
   MS CLEANER.  Center  for Ecologic Modeling,
   Rensselaer Polytechnic Institute, Troy, NY.

 Patterson, M. R., T. J. Sworski, A. L. Sjoreen, M. G.
   Browman, C. C. Coutant, D. M. Hetrick, B. D. Mur-
   phy and R. J. Raridon.  1983. A users manual for
   UTM-TOX, a unified transport  model.   Draft.

   Prepared by Oak Ridge National Laboratory, Oak
   Ridge, TN, for U. S. Environ. Prot. Agency Office
   of Toxic Substances, Washington, DC.

Reckhow, K. H., J. R. Butcher, and C. M. Marin. 1985.
   Pollutant runoff  models:  Selection  and use in
   decision making.  Water Resour. Bull. 21(2):185-

Schnoor, J. L, and D.  C. McAvoy.  1981. A pesticide
   transport and bioconcentration model. J. Environ.
   Eng. Div.

Sutherland, R. C., and R. H. McCuen.  1978. Simula-
   tion of Urban  Nonpoint Source Pollution.  Water
   Resour. Bull. 14(2):409-28

Terstriep, M.  L., and J. B. Stall.  1974. The  Illinois
   Urban Drainage  Area Simulator, ILLUDAS.  Bull.
   58. Illinois State Water Survey. Urbana, IL.

U.S. Department  of Agriculture. 1980.  CREAMS2: A
   field scale model for chemicals, runoff, and erosion
   from agricultural  management systems.  Conserv.
   Res.  Rep.  No. 26. Science  and Education  Ad-
   ministration, U.S. Dep. Agric., Tifton, GA.

U.S. Environmental Protection Agency. 1971.  Storm
   water management model, Vol. 1  -  final report.
   Water Quality  Office, Water Pollut. Control Res.
   Series: 11024, Doc 07/71.

	.  1976a.   Storm  water  management
   model, level 1, preliminary screening procedures.
   EPA 600/2-76-275. Cincinnati, OH.

	. 1976b. Development and application of
   a simplified stormwater management model. EPA
   600/2-76-218. Cincinnati, OH.

	. 1977.  Simulation of nutrient loadings in
   surface runoff with the NPS model.  Environmental
   Research Laboratory, Athens, GA.
  	.  1978.   User's manual for agricultural
   runoff management (ARM) model. EPA 600/3-78-
   080. Environmental  Research Laboratory,  Office
   of Research and Development, Athens, GA.

  	.  1980.  User's manual for hydrological
   simulation  program  - FORTRAN (HSPF).   EPA
   600/9-80-015. Environmental Research Laboratory,
   Athens, GA.

	.  1981.  ANSWERS user's manual. EPA
   905/9-82-001. Chicago, IL.

	.  1983.  Results of the Nationwide Urban
   Runoff Program, Complete Set. Water Planning
   Division. PB84-185537. Natl.  Tech.  Inf.  Sen/.,
   Springfield, VA.

U.S. Forest Service.  1980. An  approach to water
   rsources evaluation of nonpoint silvicultural sources
   (a procedural handbook). EPA-600/8-80-012. U.S.
   EPA, Athens, GA.

	.  1983.   Guide for Predicting Salmonid
   Response to Sediment Yields in Idaho Batholith
   Watersheds.  Northern Region, Missoula, MT.

	.   1986.  COWFISH, habitat capability
   model. Northern Region, Missoula, MT.

Ward, A.  J., C.  T. Haan, and B.  J. Barfield.  1977.
   Simulation  of sediment  detection basins.   Res.
   Rep. No.  103.  Water Resources Research  In-
   stitute, University of Kentucky, Lexington.

Woodward-Clyde Consultants.  1985.  Summary of
   available models for receiving water impacts from
   nonpoint sources.  Prepared by HydroQual, Inc.,

Yeh, G. T.  1982.  CHNTRN:  A  channel transport
   model  for simulating sediment and chemical dis-
   tribution in  a stream/river network.   ORNL-5882.
   Oak Ridge National Laboratory, Oak Ridge, TN.

Young, R. A.,  M. A. Otterby, and Amos Roos.  1982.
   An evaluation  system  to  rate feedlot pollution
   potential. Agricultural Reviews and Manuals. ARM-
   NC-17. Agricultural Research Service, U. S. Dep.

Young, R. A.,  C. A. Onstad, D. D.  Bosch, and W. P.
   Anderson. 1986. Agricultural nonpoint source pol-
   lution model:  A watershed analysis tool.  Agricul-
   tural Research Service.  U. S. Dep. Agric., Morris,
Best Management  Practices
 Anonymous.  1986.  Virginia Offers Tax Credit for No-
   Till. J. Soil Water Conserv. 41(1):36.

 Bilby, R. E.  1984.  Removal of woody debris may af-
   fect   stream  channel  stability.    J.  Forestry
Chesters, G. and L. Schierow.  1985. A primer on
   nonpoint pollution. J. Soil Water Conserv. 40(1 ):9-

Conservation Tillage Information Center. 1984. 1984
   National Survey Conservation Tillage Practices.

Croft, R. J.  1985.  A project to manage agriculture
   wastes has improved the quality of Vermont's lake
   parker. Pages 153-6 in Perspectives on Nonpoint
   Source Pollution. Proc. Natl. Conf. EPA 440/5-85-
   001. U.S. Environ. Prot. Agency, Washington, DC.

Gordon, T. 1980. Local commitment to lake restora-
   tion:  The Cobbossee watershed example. In Res-
   toration of Lakes and  Inland Waters. Proc. Natl.
   Conf.  U.S. Environ. Prot. Agency,  Washington,
   DC.    PB82-158478.   Natl.  Tech.  Inf.  Serv.,
   Springfield, VA.

Herman, F. R.  1960.  A test of skyline cable logging
   on steep slopes-a progress report.  USDA Forest
   Service Station Paper No. 53, October.

Impact of Land Treatment on the Restoration of Skin-
   ner Lake, Noble County, Indiana.  1982.

Inside EPA. 1986.7(3):January 17.

Kao, D. T. Y., B. J.  Barfield, and A. E.  Lyons.  1975.
   On site sediment filtration using grass strips. Proc.
   Natl. Symp. Urban Hydrology and Sediment Con-
   trol.  University of Kentucky, Lexington. July 28-

Lake Fenwick Restoration Project. 1981.

Lake Kampeske  Shoreline Protection  Project Final
   Report. 1982.

Lyles, L., J. Tatarko, and  J. D. Dickerson.   1984.
   Windbreak effects on soil water and wheat yield.
   Trans. Am. Soc. Agric. Eng. 27(1 ):69-72.

Mannering, J., and C. Fenoter. 1983. What is conser-
   vation tillage. J. Soil Water Conserv. 38(3):141 -3.

Merritt,  D. S., ed. 1968.  Standard Handbook for Civil
   Engineers. Section 21. McGraw Hill, NY.

Moldenhauer, W., G. Langdale, W. Frye, D. McCool,
   R. Papensick, D. Smika, and D. Fryrear.  1983.
   Conservation tillage for erosion  control.  J.  Soil
   Water Conserv. 38(3) :144-51.
Moore,  I. C., F. W. Madison, and  R.  R. Schneider.
   1979.  Estimating phosphorus  loadings from live-
   stock wastes:  Some  Wisconsin results.  Pages
   175-92 in R. C. Loehr, ed. Best Management Prac-
   tices  in Agriculture  and Silviculture.  Ann Arbor
   Science, Ann Arbor, Ml.

Oakwood Lake Bank  Shoreline Protection  Project
   Final Report. 1982.

Pitt, R.   1979. Demonstration of nonpoint pollution
   abatement through improved street cleaning prac-
   tices. EPA 600/2-79-161. U.S. Environ. Prot. Agen-
   cy, Washington, DC.
Porter, K. S., ed.  1975.  Nitrogen and Phosphorus:
   Food Production,  Waste, and  the  Environment.
   Ann Arbor Science Publishers. Ann Arbor, Mich.

Soil Conservation Society of America.   1985.  Non-
   point water pollution, a special issue. J. Soil Water
   Conserv. 40(1).

Stewart, B. A., D. A. Woolhiser, W. J. Wishmeiser, J.
   H.  Caro, and M. H. Frere. 1975. Control of water
   pollution from cropland.  Vol.  I.  EPA 600/2/75-
   026a. U.S. Environ. Prot. Agency, Washington, DC.

Stoeckeler, J. H.   1962.   Shelterbelt influence on
   great plains field environment crops.  Prod. Res.
   Rep. No. 62.  U.S. Dep. Agric.

Tinus, R. W., ed.   1976.  Shelterbelts on the great
   plains.  Proc. Symp. Great Plains Agric. Council

Unger, D. G.  1986. Comments on decision maker's
   guide on nonpoint pollution control.  U.S. Dep.
   Agric., Watershed and Air  Management, to U.S.
   Enviro.  Prot. Agency, Nonpoint Source Branch,
   Washington, D.C. August 14.

U.S.  Environmental  Protection   Agency.  1973a.
   Methods for identifying and evacuating the nature
   and extent of nonpoint sources of pollution. EPA
   430/9-73-014. Office of Air and Water Programs,
   Washington, DC.

	. 1973b.  Methods and practices for con-
   trolling water pollution from agricultural  nonpoint
   sources.  EPA   430/9-73-015.   Office of  Water
   Program Operations, Water Quality and  Nonpoint
   Source Control Division, Washington, DC.

	. 1973c.  Processes,  procedures,  and
   methods to control pollution resulting from all con-
   struction activity. EPA 430/9-73- 007. Office of Air
   and Water Programs, Washington, DC.

	. 1979. Demonstration of nonpoint pollu-
   tion abatement through improved street cleaning
   practices. EPA 600/2-79-161. Municipal  Environ-
   mental Research Laboratory, Cincinnati, OH.

	.  1980a.     Clean   Lakes   Program
   Guidance Manual. Prepared by JACA Corporation,
   Ft. Washington, PA. EPA 440/5-81-003.

 	. 1980b.  An approach to water resour-
   ces evaluation of nonpoint silvicultural sources (a
   procedural handbook). EPA600/8-80-012. Environ-
   mental  Research Laboratory,  Office of  Research
   and Development, U.S.  Environmental Protection
   Agency, Athens, GA.

	.  1980c. Capsule report: lake restoration
 in Cobbossee watershed.  Technol. Transfer EPA
	.  1980d.  An approach to water resour-
 ces evaluation of nonpoint silvicultural sources (A
 procedural handbook by the Forest Service). EPA
 600/8-80-012. Athens, GA.
           1981. Clean  Lakes
Program  Status
 Office of Water
 1975-1981. EPA 440/5- 82-024
 Regulations and Standards, Washington, DC. EPA

	.  1984. Report to Congress:  Nonpoint
 Source Pollution in the United States.  Office of
 Water  Program  Operations,  Water  Planning
 Division, Washington, DC.
	.  1985a.   Perspectives  on  Nonpoint
   Source Pollution. Proc. Natl. Conf. Coordinated by
   N.  Am. Lake  Manage.  Soc., Kansas City, MO,
   May. EPA 440/5-85-001.

	.  1985b.   Clean Lakes  Program.  A
   review of the first decade.  EPA 440/5-85-033. Of-
   fice  of Water  Regulations   and  Standards,
   Washington, DC.

U.S.  Soil  Conservation Service.   1978.    Water
   management and sediment control  for urbanizing
   areas, Columbus, Ohio. U.S. Dep. Agric.

Woodruff,  N.P., L. Lyles, F. H. Siddoway, and D.  W.
   Fryrear. 1972. How to control wind  erosion. U.S.
   Dep. Agric. Info. Bull. No. 354.

Aeration of lawns - Periodic perforation of lawns to
increase infiltration.  Usually performed by punching
holes  into  the lawn by mechanical means.  Urban
lawns have very low infiltration rates in comparison to
woods and grasslands.
Best management  practices (BMPs) -  Measures,
sometimes structural, that are determined to be the
most effective, practical means of preventing or reduc-
ing pollution  inputs from nonpoint  sources to water
Buffer strip or zone - Strips of grass or other erosion-
resistant vegetation between a waterway and an  area
of more intensive land use.
Calibration (Model) - Determination of some model
parameters that  depend on the  application  (site,
process, etc.).
Chisel plowing  - Seedbed preparation by plowing
(with a chisel cultivator or chisel plow) without com-
plete inversion of the soil, leaving a protective cover
of crop residues on the surface for erosion control.
Combined sewer -  A sewer receiving both inter-
cepted surface runoff and municipal sewage.
Conservation tillage - Any tillage and planting  sys-
tem that maintains at least 30 percent of the soil sur-
face covered by  residue after planting to reduce soil
erosion by water; where soil erosion by wind is the
primary concern,  maintains at least 1,000 pounds of
flat, small  grain  residue equivalent on the surface
during the critical erosion period.
Contour farming - Conducting field operations, such
as plowing, planting, cultivating, and harvesting on
the contour (across the slope).
Contour stripcropping  - Farming  operations  per-
formed on the contour with crops planted in narrow
Conventional tillage - Standard method of preparing
a seedbed by completely inverting the soil and incor-
porating all residues with a  moldboard plow.  The
field is gone over more than once in order to prepare a
smooth, fine surface.
 Cover crop - A close-growing crop grown primarily
 for the purpose of protecting and improving soil be-
 tween periods of regular crop production or between
 trees and vines in orchards and vineyards.
 Crop residue - The portion of a plant or crop left in the
 field after harvest.
 Crop rotation - The growing  of different crops in
 recurring succession on the same land, as opposed
 to continuous culture of one crop.
 Detention - The  slowing of flows-either entering the
 sewer system or draining over the surface-by  tem-
 porarily  holding  the  water on  a surface area  in a
 storage basin or within the sewer.
 Detention dam - A dam constructed for temporary
 storage of stream flow or surface runoff and for releas-
 ing the stored water at controlled rates.
 Diversion-Individually designed conveyances across
 a hillside. They may be used to protect bottom  land
 from hillside runoff, divert water from areal sources of
 pollution  or protect structures from runoff.
 Ecology -  Interrelationship  between organisms and
 the environment.
 Ecosystem (ecological system) - a functional  sys-
 tem that  includes  the organisms of a natural com-
 munity together with their environment.
 Erosion  - The wearing away of the land surface by
 running water, wind, ice, or other geological agents.
 Eutrophication - The process of enrichment of water
 bodies by nutrients. Eutrophication of a lake normally
 contributes to its  slow evolution into a bog or marsh
 and ultimately to dry land. Eutrophication may be ac-
 celerated by human activities.
 Filter berm - A  large stone or gravel dike placed
 across graded areas where runoff concentrates or at
 the disposal points along diversion dikes.
 First flush - The  condition, often occurring in storm
sewer discharges and combined sewer overflows, in
which a disproportionately high pollutional load is car-
ried in the first portion of the discharge or overflow.

Gravel inlet filter - A pile of stone or gravel placed
around or in front of an inlet to a culvert or other type
of drainage device.
Grassed waterway - A natural or constructed water-
way (usually broad  and shallow covered with erosion-
resistant grasses) used to conduct surface runoff.
Ground cover - Any vegetation  producing a protect-
ing mat on or just above the soil surface. In forestry,
low-growing shrubs, vines, and herbaceous plants
under the trees.
Ground water  -  Subsurface water  in  the zone of
Herbicide - A  chemical substance used for killing
plants, especially weeds.
Impoundment  - Generally an artificial  collection or
storage of water, as a reservoir, pit, dug out, sump,
Infiltration -  The gradual downward movement of
water from the surface into the subsoil.
Intensive cropping  - Maximum use  of the land by
means of frequent succession of  harvested crops.
Leaching - The removal of materials in solution from
the soil.
Minimum tillage - See reduced tillage.
Model  -  A set  of algorithms  organized  in a logical
structure that  represents a process.
Mulch - Any material such as straw, sawdust, leaves,
plastic film, etc., that is spread on the surface of the
soil to protect the soil and plant roots from the effects
of raindrops, freezing, evaporation, etc.
Mulch-till - The total soil surface  is disturbed by tillage
prior to planting. Tillage tools such as chisels,  field
cultivators, discs, sweeps, or blades are used.  Weed
control is accomplished with a  combination of  her-
bicides and cultivation.
Nonpoint source (NPS) pollution - Pollution caused
by sediment, nutrients, and organic and toxic substan-
ces originating  from land-use activities  and/or  from
the atmosphere, which  are  carried to  lakes  and
streams by runoff.   Nonpoint source pollution occurs
when the rate at which these materials entering water
bodies exceeds natural levels.
No-tillage - A method of planting crops that involves
no seedbed preparation  other than opening the soil
for the purpose of  placing the  seed  at  an intended
depth.  This  usually involves opening a small slit or
punching a hole into the soil. Planting is completed in
a narrow seedbed approximately  1-3  inches  wide.
Chemical weed control is normally used. Also referred
to as "zero tillage" or "no-till."
Nutrients - Elements or substances, such as nitrogen
or phosphorus, that are  necessary for  plant growth.
Large  amounts of  these substances reaching  water
bodies can become a nuisance by promoting exces-
sive aquatic algae growth.
Pesticide -  Any chemical agent used for control of
specific organisms; such as insecticides, herbicides,
fungicides, etc.
Plow-plant  - Plowing  and  planting  a crop in one
operation, with no additional seedbed preparation.
Point source pollution - This type of water pollution
results from  the discharges into receiving waters from
sewers and other identifiable "points." Common point
sources of pollution are discharges from industries and
municipal sewage treatment plants.
Porous pavement - A surface that will allow water to
penetrate through and percolate into  soil (porous as-
phalt pavement).  Pavement is comprised of irregular
shaped crush rock precoated with  asphalt binder.
Water  seeps through into lower  layers of gravel for
temporary storage, then filters naturally into the soil.
Reduced tillage -  Any tillage system  that involves
less soil disturbance and retains more plant residue on
the surface than conventional tillage.  Any tillage and
planting system that meets the 30 percent residue re-
quirement. Sometimes called "conservation" or "mini-
mum" tillage. Common practices include plow-plant-
ing, double-disking, chisel plowing, and strip tillage.
Retention - The prevention of  runoff from entering
the drainage system by storing it on a surface area or
in a storage basin.
Ridge till - The soil is left undisturbed prior to planting.
Approximately 1/3 of the soil surface is tilled at plant-
ing with  sweeps or row cleaners.  Planting is com-
pleted on ridges usually 4-6 inches higher than the row
middles. Weed control  is accomplished with a com-
bination of herbicides and cultivation. Cultivation is
used to rebuild ridges.
Riprap - A facing layer or protective mound of stones
placed to prevent erosion or sloughing of a stream-
bank or embankment.
Row crop -  A crop planted in rows, normally to allow
cultivation between rows during the growing season.
Runoff - The portion of rainfall, melted snow, or irriga-
tion water that flows across the surface or through un-
derground zones and eventually runs into streams.
The  runoff has three components:  surface runoff, in-
terflow, and groundwater flow. Runoff may pick up pol-
lutants from the air or the land  and carry them to
receiving waters.
Sediment -  Transported  and  deposited particles
derived from rocks, soil, or biological materials.
Sedimentation - Erosion, transport,  and deposition
of detached sediment particles  by flowing water or
Serrating -  Serrations  made on  the  contour on  cut
slopes by conventional bulldozers at varying intervals.
Simulation - The process that mimics some or all of
the behavior of one system with a different, dissimilar
system, particularly with computers or models.

Slash - Debris from logging activities littering a clear-
ing in a forest.
Slumping - To sink down suddenly or collapse.
Skid trails - Ruts caused by dragging logs from stump
area to loading station.
Stochastic - Pertaining to random variables (statisti-
Storm sewer -  A sewer that carries  only surface
runoff, street wash, and  snow  melt  from the  land.
Storm sewers are completely separated from those
that  carry domestic and  industrial and commercial
Stripcropping - The crop growing practice that re-
quires different types of tillage, such as corn and alfal-
fa, in alternate strips.
Strip-till - The soil is left undisturbed prior to planting.
Approximately 1/3 of the soil surface is tilled at plant-
ing time. Tillage in the row may consist of  a rototiller,
in-row chisel, row cleaners, etc.  Weed control is ac-
complished with a combination of herbicides and cul-
Stubble  mulch  - The   stubble of  crops  or  crop
residues left essentially in place on the land as a sur-
face cover during fallow and the growing of a succeed-
ing crop.
Surface roughening - Any practice that provides for
rougher,  more  permeable surfaces  that  will  slow
runoff velocity and reduce erosion potential.  Porous
pavements, grassed, and gravel driveways are ex-
Surface  runoff  - Precipitation  excess that is not
retained on the  vegetation  or surface depressions
and is not lost by infiltration, and thereby is collected
on the surface and runs off.
Terraces- An embankment, or combination of an em-
bankment and channel, constructed across a slope to
control erosion by reducing the slope and by diverting
or storing surface runoff instead of permitting it to flow
uninterrupted down the slope.
Tillage  - The operation of implements through  the
soil to prepare seedbeds and rootbeds.
Tilth - The state or degree of being tilled.
Time Series - A series  of chronologically  ordered
values giving a discrete representation of the varia-
tion in time of a given entity.
Tracking - Moving a cleated  dozer up and  down a
graded slope to provide a roughened, serrated slope.
More adaptable to steep slopes than scarification and
Urban  runoff  - Surface runoff from  an  urban
drainage area that reaches a stream or other body of
water or a sewer.
Validation (Models) - The testing of a model for com-
pliance with existing data or information.
Watershed - The total land area drained by a stream,
lake, or  river system.  Also, the area of land that con-
tributes runoff of a given point in a drainage system.
Wheel-track planting -  Plowing and  planting  in
separate operations with the seed planted in the wheel
Windbreak  - A living barrier of trees or combination
of trees and shrubs located adjacent to farm or ranch
headquarters and designed to protect the area from
erosion, winds  and drifting snow. Windbreaks  also
may be used in urban and recreational settings, and
to protect livestock.
Wind erosion - The detachment  and transportation
of soil by wind.

AGNPS                                                     13
ANSWERS                                                  14
ARM                                                       15
Block Cutting                                                91
Buffer Strips                                                 95
Conservation Tillage                                          37
Contouring                                                  39
Cover Crops                                                 43
COWFISH                                                   17
CREAMS                                                   16
ESRFPP                                                    19
Fertilizer Management                                         45
GAWS                                                      21
GLEAMS                                                   20
Grassed Waterways                                          97
Ground Cover                                               79
HSPF                                                      23
Infiltration Devices                                            99
Integrated Pest Management                                   47
Interception/Diversion Practices                                101
Limiting Disturbed Areas                                       73
Log Removal Techniques                                      75
Material Ground Cover                                        103
Nonvegetative Soil Stabilization                                 61
NPS                                                       25
NURP                                                      26
Porous Pavement                                            63
Proper Roads and Trails                                       83
Range and Pasture Management                                49
Removal of Debris                                           81
Runoff Detention/Retention                                    65
Sediment Traps                                              105
Sod-Based Crop Rotation                                      51
Street Cleaning                                              67
Stripcropping                                                41
Structural Controls                                           59
Surface Roughening                                          69
SWAM                                                     30
SWMM-Level 1                                               27
SWMM-Simplified                                            28
SWMM                                                     29
Terracing                                                   53
Underdrains                                                 89
Vegetative Stabilization                                        107
Waste Management                                          55
Water Diversion                                              87
WRENS                                                    31

               -        -
-hicago.. tL  60134-3590