United States
Environmental Protection
Agency
Office of Water
Washington D.C.
EPA841-R-92-002
June 1992
EPA COMPENDIUM OF
WATERSHED-SCALE MODELS
FOR TMDL DEVELOPMENT
-------
COMPENDIUM OF WATERSHED-SCALE MODELS
FOR
TMDL DEVELOPMENT
U S Environmental Protection Agency
Office of Wetlands, Oceans and Watersheds
Office of Science and Technology
401 M St, SW
Washington, DC 20460
June 1992
-------
Disclaimer
The information contained in this compendium is based on publications and
literature provided by model developers No verification or testing of model
accuracy or function is implied by this review The Environmental Protection
Agency does not support any model unless support is explicitly mentioned
Mention of trade names or commercial products does not constitute
endorsement or recommendation for use
-------
Foreword
The process for determining specific pollution reductions needed to attain State
water quality standards under the Clean Water Act is set out in Section 303(d)
of that Act and involves the determination of Total Maximum Daily Loads
(TMDLs) As interpreted in EPA regulations and guidance, the TMDL process
can be used to address large geographic areas such as river basins and provides
water quality managers with an analytical method to address more complex
water pollution problems, such as nonpomt sources, and to adopt a more
integrated approach to water quality management Through recent program
initiatives such as the Watershed Protection Approach, EPA has been
encouraging Federal, State, and local agencies concerned with water quality
management to analyze all water quality problems and stressors, and recommend
management measures on a basmwide rather than an individual source basis
This compendium, developed in response to recommendations made at the
Workshop on the Water Quality-based Approach for Point Source and Nonpomt
Source Controls held in Chicago on June 26-28, 1991, has identified and
summarized the most widely used (as well as some of the more obscure)
watershed-scale models that can facilitate the TMDL process It is intended to
help water quality managers and other potential users decide which model best
suits their needs and available resources
Bruce Newton, Chief
Watershed Branch
Office of Wetlands, Oceans
and Watersheds
Russ Kinerson, Chief
Exposure Assessment Branch
Office of Science and Technology
Si
-------
Acknowledgments
This compendium was developed under the direction of Don Brady m EPA's
Watershed Branch It was prepared by Leslie L Shoemaker, Ph D, Mohammed
Lahlou, Ph D, Sharon Thorns, Ph D , Richard Xue, Julie Wright and Marti Martin
of Tetra Tech, Inc , Fairfax, Virginia, under EPA Contract Number 68-C9-0013,
for the Watershed Management Section, Watershed Branch of the Assessment
and Watershed Protection Division of the U S Environmental Protection Agency
The authors would like to thank Donald Brady of the Watershed Management
Section and Bruce Newton of the Watershed Branch for their contributions to the
preparation of this document
IV
-------
Table of Contents
Page
1 INTRODUCTION AND PURPOSE 1
1 1 Background 1
1 2 Classification of Watershed-Scale Models 2
2 REVIEW OF SELECTED WATERSHED-SCALE MODELS 7
2 1 Review Methodology 7
2 2 Overview of Watershed-Scale Models 7
2 3 Model Descriptions 9
231 Simple Methods 13
232 Mid-Range Models 16
233 Detailed Models 19
3 OTHER MODELS 23
4 MODEL SELECTION 27
4 1 Model Characteristics 27
4 2 Model Calibration and Verification 37
5 REFERENCES 40
APPENDIX Watershed-Scale Model - Fact Sheets A 1
-------
List of Tables
Table 1 Evaluation of Model Capabilities - Simple Methods
Table 2 Evaluation of Model Capabilities - Mid-Range Models
Table 3 Evaluation of Model Capabilities - Detailed Models
Table 4 A Descriptive List of Model Components - Simple
Methods
Table 5 A Descriptive List of Model Components - Mid-Range
Models
Table 6 A Descriptive List of Model Components - Detailed
Models
Table 7 Input and Output Data - Simple Methods
Table 8 Input and Output Data - Mid-Range Models
Table 9 Input and Output Data - Detailed Models
Table 10 Input Data Needs for Watershed Models
Table 11 Range of Application of Watershed Models - Simple Methods
Table 12 Range of Application of Watershed Models - Mid-Range Models
Table 13 Range of Application of Watershed Models - Detailed Models
Page
10
11
12
28
29
30
32
33
34
35
37
38
38
VII
-------
-------
1. INTRODUCTION AND PURPOSE
Simulation models are used extensively in
water quality planning and pollution control
Many types of models are available to
simulate different environmental
phenomena and components of pollution
problems For example, there are models
to predict the chemical speciation of
contaminant metals, the fate and transport
of organic compounds, the bioaccumulation
of polar compounds, the mixing of
wastewater plumes, the eutrophication
process, flow and velocity characteristics in
channels and the effects on fishery habitat,
and so on Historically, the Environmental
Protection Agency's (EPA) water programs
and its counterparts in State pollution
control agencies have focused on pollution
problems resulting from wastewater
discharges Attention is now expanding to
encompass nonpomt source problems and
storm-related problems It is also generally
recognized that solving these problems
requires a "watershed approach" to
analyzing the problems, engaging the
parties that have a stake in the process,
and developing solutions to the pollution
problems Examining problems on a
watershed scale is a new challenge for
which there are not well-developed
techniques and models The purpose of
this document is to provide the reader with
a quick compendium of models that are
useful for watershed-scale analysis
7.7 Background
Section 303{d) of the Clean Water Act
(CWA) requires States to develop Total
Maximum Daily Loads (TMDLs) for water
quality-limited waters where existing or
proposed controls do not or are not
expected to result in attainment and/or
maintenance of the applicable water quality
standards {WQSs} Implementation of
section 303{d) of the CWA has traditionally
emphasized point source wasteload
allocations, which were easily enforced by
incorporating them into National Pollution
Discharge Elimination System (NPDES)
permits as discharge limits Nonpomt
sources were generally not included as a
separate component of a TMDL because of
the difficulty in measuring water quality
impacts and the effectiveness of controls
Experience has shown, however, that
controlling point source discharges does not
necessarily ensure attainment of WQSs,
especially when nonpomt sources are a
significant contributor to water quality
problems
It is time to consider nonpomt sources as
distinct from natural background pollutant
loadings and to formally address them as a
distinct component of the TMDL equation
in water quality management schemes
Recognizing this fact, EPA is now stressing
more integrated water quality management
that considers all of the sources and types
of pollution within a watershed and brings
together all shareholders to develop
cooperative solutions
This new focus was clearly stated in EPA's
programmatic guidance outlining the TMDL
process (US EPA, 1991b) By addressing
the technical issues of developing and
implementing TMDLs within a broader
-------
Compendium of Watershed-Scale Models for TMDL Development
water quality-based management strategy,
this document sets the stage for State and
Federal agencies to establish both point and
nonpoint source pollution controls on a
watershed basis
The water quality-based approach consists
of five steps, the first three of which
constitute the TMDL process1
1) identification of water quality-limited
waters that require TMDLs and the
pollutants causing impairment,
2) priority ranking and targeting of
identified waters,
3) TMDL development,
4) implementation of pollution control
actions, and
5) monitoring and assessment of control
effectiveness
In complex situations or where nonpoint
source reductions are part of the TMDL, a
"phased approach" may be used Under
this approach, available knowledge on
water quality conditions and BMP
effectiveness is used in conjunction with
scheduled monitoring and evaluation to
develop TMDLs that consider point and
nonpoint sources of pollution In fact, the
final step of the TMDL process provides for
continuous evaluation and improvement of
the TMDL and the pollution control actions
Models and data analysis techniques may
be used in the implementation of each
phase of the TMDL process Models can
assist in the initial evaluation, ranking and
targeting, TMDL development, evaluation of
controls, and program tracking
As part of steps 1 and 2, an initial
evaluation of water quality may be used to
focus or target resources for TMDL
development The use of such preliminary
screening applications can provide water
quality managers with an initial assessment
of water quality, pollutant loading, and
source determination for key watersheds
Screening applications are typically
performed with minimal input data and
calibration/verification Because of the
preliminary nature of the application and
data limitations, the accuracy of output is
often low The results of screening
applications are nevertheless adequate for
relative comparisons and preliminary
decision making As key watersheds are
identified and additional monitoring data
are collected, more detailed and accurate
model analyses can be initiated
1.2 Classification of Watershed-Scale
Models
One challenge faced by water quality
managers is the lack of highly developed,
scientifically sound approaches to identify
problems in watersheds and to predict the
results of potential control actions on water
quality While a wide variety of models are
available, each comes with limitations on
its use, applicability, and predictive
capabilities The type of study will dictate
the complexity of the development of a
model to be selected For example, a high-
prionty water with a number of pollutant
sources may require a detailed model
analysis In determining where to set
priorities and gathering information prior to
initiating development of a TMDL, it may be
necessary to use several models or portions
of models A wide variety of technical
tools and techniques are available to
-------
Compendium of Watershed-Scale Models for TMDL Development
evaluate point and nonpomt loads from
watersheds containing multiple sources and
land uses. Although these tools were not
specifically designed to address the TMDL
process, many of their capabilities can be
directly applied to comprehensive water
quality-based assessments.
Existing watershed-scale models can be
grouped into three categories - simple
methods, mid-range models, and detailed
models - based on the number of processes
they incorporate and the level of detail they
provide In addition to the different classes
of models, the level of application of a
given model may vary depending on the
objectives of the analysis
Simple Methods The major advantage of
simple methods is that they can provide a
rapid means of identifying critical areas
with minimal effort and data requirements
Simple methods are compilations of expert
j'udgment and empirical relationships
between physiographic characteristics of
the watershed and pollutant export that can
often be applied by using a spreadsheet
program or hand-held calculator Simple
methods are often used when data
limitations and budget and time constraints
preclude the use of complex models They
are used to diagnose nonpomt source
pollution problems based on relatively
limited available information They may be
used to support an assessment of the
relative significance of different sources,
guide decisions for management plans, and
focus continuing monitoring efforts
Simple methods, in general, rely on large-
scale aggregation and neglect important
features of small patches of land They
rely on generalized sources of information
and therefore have low to medium
requirements for site-specific data Default
values provided for these methods are
derived from empirical relationships that are
evaluated based on regional or site-specific
data The estimations are usually
expressed as mean annual values.
Simple methods provide only rough
estimates of sediment and pollutant
loadings and have very limited predictive
capability The empiricism contained in the
models limits their transferabihty to other
regions Because they often neglect
seasonal variability, simple methods may
not be adequate to model water quality
problems for which shorter duration
loadings are important They may be
sufficient for problems such as nutrient
loadings to and eutrophication of long-
residence-time water bodies
Mid-Range Models The advantage of mid-
range watershed-scale models is that they
evaluate pollution sources and impacts over
broad geographic scales and therefore can
assist in defining target areas for pollution
mitigation programs on a watershed basis
Several mid-range models are designed to
interface with geographic information
systems (GIS), which greatly facilitate
parameter estimation Greater reliance on
site-specific data gives mid-range models a
relatively broad range of regional
applicability However, the use of
simplifying assumptions limits the accuracy
of their predictions to within about an order
of magnitude (Dillaha, 1992) and restricts
their analysis to relative comparisons
This class of model attempts a compromise
between the empiricism of the simple
methods and the complexity of detailed
mechanistic models Mid-range models use
-------
Compendium of Watershed-Scale Models for TMDL Development
a management-level approach to assessing
pollutant sources and transport in
watersheds by incorporating simplified
relationships for the generation and
transport of pollutants while still retaining
responsiveness to management objectives
and actions appropriate to watershed
management planning (Clark et al , 1979)
They are relatively simple and are intended
to identify problem areas within large
drainage basins or to make preliminary,
qualitative evaluations of BMP alternatives
(Dillaha, 1992)
Unlike the simple methods, which are
restricted to predictions of annual or storm
loads, mid-range tools can be used to
assess the seasonal or inter-annual
variability of nonpomt source pollutant
loadings and to assess long-term water
quality trends Also, they can be used to
address land use patterns and landscape
configurations in actual watersheds They
are based primarily on empirical
relationships and generalized sources of
information. In addition, they typically
require site-specific data and some
calibration. Mid-range models are designed
to estimate the importance of pollutant
contributions from multiple land uses and
many individual source areas in a
watershed. Thus, they can be used to
target important areas of pollution
generation and identify areas best suited for
controls on a watershed basis Moreover,
the continuous simulation furnished by
some of these models provides an analysis
of the relative importance of sources for a
range of storm events or conditions In an
effort to reduce complexity and data
requirements, these models often are
written for specific applications For
instance, mid-range models can be
designed for application to agricultural,
urban, or mixed watersheds Some mid-
range models simplify the description of
transport processes while emphasizing
possible reductions available with controls,
others simplify the description of control
options and emphasize the changes in
pollutant concentrations as they move
through the watershed
Detailed Models. Detailed models best
represent the current understanding of
watershed processes affecting pollution
generation Detailed models are best able
to address problem causes rather than
simply describe overall conditions If
properly applied and calibrated, detailed
models can provide relatively accurate
predictions of variable flows and water
quality at any point in a watershed. The
additional precision they provide, however,
comes at the expense of considerable time
and resource expenditure
Detailed models use storm event or
continuous simulation to predict flow and
pollutant concentrations for a range of flow
conditions The models are large and were
not designed with emphasis on their
potential use by the typical State or local
planner Many of these models were
developed for research into the
fundamental land surface and instream
processes influencing runoff and pollutant
generation rather than to communicate
information to decision-makers faced with
planning watershed management (Biswas,
1975)
Detailed models incorporate the manner in
which watershed processes change over
time in a continuous fashion rather than
relying on simplified terms for rates of
change (Addiscott and Wagenet, 1985)
They tend to require rate parameters for
-------
Compendium of Watershed-Scale Models for TMDL Development
flow velocities and pollutant accumulation,
settling, and decay instead of capacity
terms The length of time steps is variable
and depends on the stability of numerical
solutions as well as the response time for
the system (Nix, 1991), Algorithms in the
detailed model more closely simulate the
physical processes of infiltration, runoff,
pollutant accumulation, mstream effects,
and groundwater/surface water inter-
actions. The input and output of detailed
models have greater spatial and temporal
resolution Moreover, the manner in which
physical characteristics and processes differ
over space is incorporated within the
governing equations (Nix, 1991) Linkage
to biological modeling is possible due to the
comprehensive nature of continuous
simulation models In addition, detailed
hydrologic simulations can be used to
design controls
-------
Compendium of Watershed-Scale Models for TMDL Development
-------
2. REVIEW OF SELECTED WATERSHED-SCALE MODELS
2.1 Review Methodology
A three-step process was used to review
existing watershed-scale models
• Model identification In addition to
information gained through user
experience, the model identification
process was based on a review of
abstracts obtained from computer
searches, available case studies in which
a modeling approach was used to
address nonpomt source pollution, and
available published literature Receiving
water quality models and models
developed for field-size studies were not
considered in this review
• Model acquisition The acquisition
process was initiated by contacting
model distributors, potential model users,
other consultants, and model developers
A list of models acquired and their
corresponding distributor references is
presented in the Appendix The
collection of models with potential use in
the TMDL process is ongoing
• Model documentation Each model was
thoroughly reviewed with respect to its
theoretical basis, range of applicability,
and input requirements Additional
references concerning model application
were also collected and used in this step
The results of this review were compiled
in the form of fact sheets and are
presented in the Appendix This review
forms the primary basis for the following
discussions
2.2 Overview of Watershed-Scale Models
During the last 20 years, numerous
watershed and nonpomt source models
have been developed Some of these
models have been (and still are) used
successfully by watershed and water
quality managers Other models have had
limited use or have been rapidly
incorporated into larger systems prior to
real world applications or testing This
latter group of models is not addressed in
this review In addition, watershed models
have been continuously updated and
improved in light of new developments in
computing facilities Older versions of
currently used models are also not included
in the present compendium Many of the
models reviewed were developed or
sponsored by Federal or State agencies,
however, a few simple and mid-range
models were developed by universities or
private companies
The U S EPA currently supports and
distributes two complex watershed models
the Hydrologic Simulation Program -
FORTRAN (HSPF), for watersheds with
multiple land use categories, and the Storm
Water Management Model (SWMM),
developed primarily for urban land areas
The versatility of HSPF and SWMM for
simulating a wide range of land uses and
their continual upgrading make these
-------
Compendium of Watershed-Scale Models for TMDL Development
models two of the most detailed and widely
applied to watershed studies EPA also
supports a simple screening procedure for
assessing nonpomt loads This procedure
(McElroy et al., 1976, Amy et al , 1974,
Mills et al., 1985), a compilation of
empirical relations and loading functions,
can be applied using a hand-held calculator
The procedure provides average annual
estimates of loadings of toxic compounds
in addition to those of conventional
pollutants.
The U S. Department of Agriculture - Soil
Conservation Service (USDA-SCS) has
developed several relatively simple, well-
documented models that may potentially be
considered for TMDL development in rural
and agricultural watersheds (e g ,
Agricultural Non-Point Source Pollution
Model (AGNPS); Simulator for Water
Resources in Rural Basin-Water Quality
model (SWRRB or SWRRBQ), Erosion-
Productivity Impact Calculator (EPIC),
Groundwater Loading Effects of Agricultural
Management Systems (GLEAMS),
Chemicals, Runoff, Erosion from
Agricultural Management Systems
(CREAMS); and Nitrate Leaching and
Economic Analysis Package (NLEAP))
These models were designed for various
purposes ranging from the simple
estimation of runoff and sediment loads
from individual storm events to runoff
routing, point and nonpomt source
contributions, and continuous runoff and
subsurface flow modeling on complex
watersheds. Field-scale models, such as
GLEAMS, CREAMS, NLEAP, and EPIC are
designed for specific agricultural
applications and may be used to target
specific pollution sources, land uses, or
land activities and to test the efficiency of
control practices. The SWRRB model can
be applied on a watershed basis for
continuous simulation, but additional
development and testing of model
components for nutrient and pesticide
transport is under way The AGNPS model
is watershed-scale but is currently limited in
application to design storms The USDA is
upgrading AGNPS to include continuous
simulation
The U S Army Corps of Engineers
Hydraulic Engineering Center (HEC)
developed a continuous urban simulation,
including dry-weather sewer flows, in their
Storage, Treatment, Overflow, Runoff
Model (STORM) With support of HEC,
STORM has been used extensively for
planning purposes and for evaluating
control strategies for combined sewer
overflows The U S Geological Survey
(USGS) has developed and applied the
Distributed Routing Rainfall Runoff Model
(DR3M-QUAL), calculating runoff and
pollutant loads in several urban watersheds
The USGS also developed a simple
statistical method based on monitoring data
from several gaging stations The U S
Federal Highway Administration developed
and used a simple pollutant loading model
(FHWA) to assess water quality due to
stormwater runoff from highways and to
develop preliminary pollution control
options
A number of models were developed at
universities or other research institutions
The Areal* Nonpomt Source Watershed
Environment Response Simulation model
(ANSWERS) was developed at the
University of Georgia to predict the
movement of sediment through relatively
large agricultural watersheds The Source
Loading and Management Model (SLAMM)
was developed at the University of
8
-------
Compendium of Watershed-Scale Models for TMDL Development
Alabama for the purpose of evaluating
urban management practices for sediment,
nutrients, and other urban pollutants
including toxics and oxygen-demanding
substances The Generalized Watershed
Loading Function (GWLF) model, a mid-
range model, was developed at Cornell
University The GWLF model considers
runoff from rural and urban land uses and
integrates pollution from both point sources
and nonpomt sources Watershed is a
simple nonpomt source model developed at
the University of Wisconsin to assess the
cost-effectiveness of stormwater control
practices
Several State and local agencies have
participated in the development of nonpomt
source models Water Screen is a loadmg-
function-based method developed by the
Office of Planning and Zoning, Anne
Arundel County, Annapolis, Maryland The
Illinois State Water Survey developed a
mid-range model (Auto-Q-llludas or Auto-
Ql) for continuous simulation of pollutant
loading from built-up areas The
Washington Metropolitan Council of
Governments developed a simplified
approach, "the Simple Method," for
estimating pollutant loads from urban land
uses (Schueler, 1987) This approach
relies on data from the National Urban
Runoff Program (NURP) for default values
Watershed Management Model (WMM) is
under development for the Florida
Department of Environmental Regulation to
evaluate nonpomt source pollution loads
and control strategies The Urban
Catchment Model (P8-UCM), was
developed for the Narragansett Bay Project
The P8-UCM model predicts pollutant
loading and transport of storm water runoff
from urban watersheds
Other models developed by private
companies include the Simplified Particle
Transport Model (SIMPTM) and the
Nonpomt Source Model for Analysis and
Planning (NPSMAP) SIMPTM was
developed by OTAK, Incorporated, to
estimate pollutant loads from impervious
areas and to evaluate the effectiveness of
a number of urban stormwater practices
The NPSMAP is a spreadsheet-based model
designed to simulate stream segment load
capacities (LCs), point source wasteload
allocations (WLAs), and nonpomt source
load allocations (LAs)
2.3 Model Descriptions
Several watershed-scale models have been
selected as potential candidates for use in
the TMDL development process and are
included in this compendium A qualitative
evaluation of each model in terms of its
simulation capabilities, modeling
performance, and ease of use was
developed from available documentation
and published applications to real-world
case studies The results of this qualitative
evaluation for the three categories of
models are summarized in Tables 1, 2, and
3 These models differ in the type and the
manner in which they describe the
hydrologic and pollutant loss processes, the
type of pollutant and pollution sources they
address, the level of accuracy they provide,
and the input data, personnel training, and
resources and computing capabilities they
require
Simple methods generally use simplistic and
empirical relations to describe a limited
number of hydrologic and pollution
processes As can be seen from Table 1,
these methods use large simulation time
steps to provide long-term averages or
-------
Table 1 Evaluation of Model Capabilities - Simple Models
Criteria
Land Uses
Time Scale
Hydrology
Pollutant
Loading
Pollutant
Routing
Model Output
Input Data
BMPs
Documentation
Urban
Rural
Point Sources
Annual
Single Event
Continuous
Runoff
Baseflow
Sediment
Nutrients
Others
Transport
Transformation
Statistics
Graphics
Format Options
Requirements
Calibration
Default Data
User Interface
Evaluation
Design Criteria
EPA1
Screening
O
G
_
•
o
-
_4
-
O
e
G
_
-
-
-
-
O
-
•
-
o
-
•
Simple1
Method
G
-
-
•
O
-
G
-
€
G
€
-
-
-
-
-
O
-
•
-
O
-
•
Regression
Method
G
O
-
•
0
-
-
-
G
G
€
-
-
-
-
-
O
-
G
-
-
-
•
SLOSS-,
PHOSPH
-
G
-
•
-
-
-
-
O
0
-
-
-
-
-
-
O
0
G
-
o
-
«
Water
Screen
O
o
-
-
-
-
-
0
G
-
-
-
-
-
-
o
0
o
o
o
-
G
Watershed
C
o
o
•
-
-
-
-
G
€
0
-
-
€
€
O
O
€
O
«
4
-
•
FHWA
O3
0
-
•
0
-
o
-
-
0
€
-
-
O
-
-
o
-
€
O
€
-
•
WMM
•
•
O
•
-
-
o
o
_
€
€
-
O
0
0
o
o
€
€
O
O
-
O
Not a computer program 4 Extended versions recommend use of • High C Medium O Low - Not Available
Coupled with GIS SCS-curve number method
3 Highway drainage basins for runoff estimation
I
I
s?
!
1
I
i
-------
Table 2 Evaluation of Model Capabilities - Mid-Range Models
Criteria
Land Uses
Time Scale
Hydrology
Pollutant
Loading
Pollutant
Roubng
Model Output
Input Data
BMPs
Documentation
Urban
Rural
Point Sources
Annual
Single Event
Continuous
Runoff
Baseflow
Sediment
Nutrients
Others
Transport
Transformation
Statistics
Graphics
Format Options
Requirements
Calibration
Default Data
User Interface
Evaluation
Design Criteria
NPSMAP
•
•
€
-
0
•
9
o
-
•
-
o
-
€
G
•
€
O
•
9
O
-
•
GWLF
•
•
G
-
-
•
•
•
•
•
-
O
-
O
G
•
G
O
•
•
O
-
O
P8-UCM
•
-
•
-
•
-
•
_
-
•
•
0
-
-
•
•
G
O
G
•
•
•
*
SIMPTM
•
-
_
-
_
-
•
O
•
•
•
o
-
o
-
•
G
o
o
4)
G
O
O
Aiito-QI
•
-
-
-
-
•
•
O
•
•
•
G
-
-
-
O
O
G
O
G
G
C
G
AGNPS
-
•
•
-
•
-
•
-
•
•
-
•
-
-
•
•
c
0
o
€
€
O
•
SLAMM
•
-
•
*•
-
•
•
0
•
•
•
€
-
O
O
•
o
o
€
•
O
o
€
I
I
or
I
I
I
I
i
• High C Medium O Low - Not Available
-------
Table 3 Evaluation of Model Capabilities - Detailed Models
Criteria
Land Uses
Time Scale
Hydrology
Pollutant
Loading
Pollutant
Routing
Model Output
Input Data
BMPs
Documentation
Urban
Rural
Point Sources
Annual
Single Event
Continuous
Runoff
Baseflow
Sediment
Nutrients
Others
Transport
Transformation
Statistics
Graphics
Format Options
Requirements
Calibration
Default Data
User Interface
Evaluaton
Design Criteria
STORM
•
-
•
-
O
•
•
o
•
•
•
_
-
o
-
•
€
O
€
_
€
€
•
ANSWERS
-
•
-
-
•
-
•
-
•
•
-
€
-
-
•
•
•
O
O
_
€
€
O
DR3M
•
-
•
-
O
•
•
o
•
•
-
•
-
•
C
•
•
Q
•
O
•
€
O
SWRRBWQ
O
•
•
-
0
•
•
•
•
•
•
•
-
•
c
•
o
o
•
o
€
-
•
SWMM
•
0
•
-
•
•
0
•
•
•
•
o
o
•
0
•
•
•
0
-
•
•
•
HSPF
•
€
•
-
•
O
•
•
•
€
-
•
•
•
f
9-
i
I
f
High Q Medium
O Low
- Not Available
-------
Compendium of Watershed-Scale Models for TMDL Development
annual estimates However, although they
can easily be adopted to estimate seasonal
or storm event loadings, their accuracy
decreases since they cannot capture the
large fluctuations of pollutant loading or
concentration usually observed at smaller
time steps In general, these methods rely
on available default values and do not
implicitly relate pollutant loads to
hydrological processes Pollutant loads are
determined from export coefficients (e g ,
Watershed) or as a function of the
sediment yield (e g , EPA screening
procedures, SLOSS-PHOSPH, Water
Screen) The Simple Method, regression
method, and FHWA methods are
statistically based approaches developed
from past monitoring information Their
application is limited to the areas for which
they were developed and to watersheds
with similar land uses or activities
Mid-range models attempt to use smaller
time steps in order to represent seasonal
variability, therefore they require additional
meteorologic data (e g , daily weather data
for the GWLF, hourly rainfall for NPSMAP)
They also attempt to relate pollutant
loadings to hydrologic (e g , runoff) and
erosion (e g , sediment yield) processes
These models usually include adequate
input-outputfeatures (e g ,AGNPS, GWLF),
making applications easier to process
Several of these models (NPSMAP, Auto-
Ql) were developed in existing computing
environments (e g , Lotus 1-2-3®) to make
use of their built-in graphical and statistical
capabilities It should be noted from Tables
1 and 2 that neither the simple nor the mid-
range models consider degradation and
transformation processes, and few
incorporate adequate representation of
pollutant transport within and from the
watershed Although their applications
may be limited to relative comparisons,
however, they may provide water quality
managers with necessary information for
watershed-plannmg-level decisions
More detailed models, on the other hand,
attempt to simulate the physical processes
governing hydrologic and pollutant
transport and transformation mechanisms
Table 3 shows that these models use small
time steps to allow for continuous and
storm event simulations However, input
data file preparation and calibration require
professional training and adequate
resources Some of these models (e g ,
STORM, SWMM, ANSWERS) were
developed not only to support planning-
level evaluations but also to provide design
criteria for pollution control practices If
appropriately applied, state-of-the-art
models such as HSPF and SWMM can
provide accurate estimations of pollutant
loads and expected impacts on water
quality However, their added accuracy
may not always justify the amount of effort
and resources they require Application of
such detailed models is more cost-effective
when used to address complex situations or
objectives
A qualitative description of each model is
presented in the following section to
supplement the information reported in
Tables 1, 2, and 3 For a more technical
description, the reader is referred to the
Appendix to this compendium
2.3.1 Simple Methods
EPA Screening Procedures The EPA
Screening Procedures were developed by
the EPA Environmental Research Laboratory
in Athens, Georgia, (McElroy et al , 1976,
Mills, 1985) to calculate pollutant loads
13
-------
Compendium of Watershed-Scale Models for TMDL Development
from point and nonpomt sources, including
atmospheric deposition, for preliminary
assessment of water quality The
procedures consist of loading functions and
simple empirical expressions relating
nonpomt pollutant loads to other readily
available parameters Data required
generally include information on land
use/land cover, management practices,
soils, and topography Although these
procedures are not coded into a computer
program, several computer-based models
have adapted the loading function concept
to predict pollutant loadings An advantage
of this approach is the possibility of using
readily available data as default values
when site-specific information is lacking
Application of these procedures requires
minimum personnel training and practically
no calibration. However, application to
large, complex watersheds should be
limited to pre-planning activities
The Simple Method. The Simple Method is
an empirical approach developed for
estimating pollutant export from urban
development sites in the Washington, DC
area (Schueler, 1987) It is used at the
site-planning level to predict pollutant
loadings under a variety of development
scenarios. Its application is limited to small
drainage areas of less than a square mile
Pollutant concentrations of phosphorus,
nitrogen, chemical oxygen demand,
biological oxygen demand (BOD), and
metals are calculated from flow-weighted
concentration values for new suburban
areas, older urban areas, central business
districts, hardwood forests, and urban
highways. The method relies on the NURP
data for default values (US EPA, 1983) A
graphical relationship is used to determine
the event mean sediment concentration
based on readily available information This
method is not coded into a computer
program but can be easily implemented
with a hand-held calculator
USGS Regression Approach The
regression approach developed by USGS
researchers is based on a statistical
description of historic records of storm
runoff responses on a watershed level
(Tasker and Driver, 1988) This method
may be used for rough preliminary
calculations of annual pollutant loads when
data and time are limiting Simple
regression equations were developed using
available monitoring data of pollutant
discharges at over 70 gaging stations in 20
States Separate equations are given for
ten pollutants, including dissolved and total
nutrients, chemical oxygen demand, and
metals Input data include drainage area,
percent imperviousness, mean annual
rainfall, general land use pattern, and mean
minimum monthly temperature Application
of this method provides storm-mean
pollutant loads and corresponding
confidence intervals The use of this
method as a planning tool at a regional or
watershed level may require preliminary
calibration and verification with additional,
more recent monitoring data
Simplified Pollutant Yield Approach
(SLOSS-PHOSPH) This method uses two
simplified loading algorithms to evaluate
soil erosion, sedimentation, and phosphorus
transport from distributed watershed areas
The SLOSS algorithm provides estimates of
sediment yield, while the PHOSPH
algorithm uses a loading function to
evaluate the amount of sediment-bound
phosphorus Application to watershed and
subwatershed levels was developed by Tim
et al (1991) based on an integrated
approach coupling these algorithms with
14
-------
Compendium of Watershed-Scale Models for TMDL Development
the Virginia Geographical Information
System (VirGIS) The approach was
applied to the Nommi Creek Watershed,
Westmoreland County, Virginia, to target
critical areas of nonpomt source pollution at
the subwatershed level In this application,
analysis was limited to phosphorus loading,
however, other pollutants for which input
data or default values are available can be
modeled in a similar fashion The approach
requires full-scale GIS capability and trained
personnel
Water Screen Water Screen was
developed by the Office of Planning and
Zoning, Anne Arundel County, Annapolis,
Maryland, for the Apple II microcomputer
The program can be used at the planning
level to estimate loadings of sediment,
nitrogen, phosphorus, BOD, lead, and zinc
from various land uses in a watershed
This model uses a loading function
approach similar to that developed in the
EPA screening procedures (McElroy et al ,
1976) and does not require meteorologic
data since pollutant loads are computed
from estimates of sediment yield
Estimation of nitrogen loads considers both
losses from surface soils and input from
precipitation. This model was tested on the
Church Creek watershed, south of
Annapolis, Maryland (Bird and Conaway,
1985)
Watershed. Watershed is a spreadsheet
model developed at the University of
Wisconsin to calculate phosphorus loading
from point sources, combined sewer
overflows (CSOs), septic tanks, rural
croplands, and other urban and rural
sources The Watershed program can be
used to evaluate the tradeoffs between
control of point and nonpomt sources
(Walker, Pickard, and Sonzogni, 1989) It
uses an annual time step to calculate total
pollution loads and to evaluate the cost-
effectiveness of pollution control practices
in term of cost per unit load reduction The
program uses a series of worksheets to
summarize watershed characteristics and to
estimate pollutant loadings for uncontrolled
and controlled conditions Because of the
simple formulation describing the various
pollutant loading processes, the model can
be applied using available default values
with minimum calibration effort
Watershed was applied to study the
tradeoffs between controlling point and
nonpomt sources in the Delavan Lake
watershed in Wisconsin
The Federal Highway Administration
(FHWA) Model The Office of Engineering
and Highway Operations has developed a
simple statistical spreadsheet procedure to
estimate pollutant loading and impacts to
streams and lakes that receive highway
stormwater runoff {Federal Highway
Administration, 1990) The procedure uses
several worksheets to tabulate site
characteristics and other input parameters,
as well as to calculate runoff volumes,
pollutant loads, and the magnitude and
frequency of occurrence of instream
pollutant concentrations The FHWA model
uses a set of default values for pollutant
event-mean concentrations that depend on
traffic volume and the rural or urban setting
of the highway's pathway This method is
used by the Federal Highway
Administration to identify and quantify the
constituents of highway runoff and their
potential effects on receiving waters and to
identify areas that may require controls
Watershed Management Model (WMM)
The Watershed Management Model was
developed for the Florida Department of
15
-------
Compendium of Watershed-Scale Models for TMDL Development
Environmental Regulation for watershed
management planning and estimation of
watershed pollutant loads (Camp, Dresser,
and McKee, 1992). Pollutants simulated
include nitrogen, phosphorus, lead, and zinc
from point and nonpomt sources The
model is implemented in the Lotus 1-2-3®
spreadsheet environment and will thus
calculate standard statistics and produce
plots and bar charts of results Although it
was developed to predict annual loadings,
this model can be adapted to predict
seasonal loads provided that seasonal event
mean concentration data are available In
the absence of site-specific information, the
event concentration derived from the NURP
surveys may be used as default values The
model includes computational components
for stream and lake water quality analysis
using simple transport and transformation
formulations based on travel time The
WMM has been applied to several
watersheds including the development of a
master plan for Jacksonville, Florida, and
the Part II estimation of watershed loadings
for the NPDES permitting process It has
also been applied in Norfolk County,
Virginia; to a Watershed Management Plan
for North Carolina, and to a wasteload
allocation study for Lake Tohopekaliga, near
Orlando, Florida.
2.3.2 Mid-Range Models
Nonpoint Pollution Source Model for
Analysis and Planning (NPSMAP).
NPSMAP is spreadsheet program developed
to simulate nonpomt source runoff and
nutrient loadings, in addition to point
source discharges (Omicron Associates,
1990). Nonpoint source runoff simulations
incorporate irrigation, evapotranspiration,
and drainage to groundwater Point source
discharge simulations include infiltration,
overflows and bypasses, and changes in
treatment plant performance NPSMAP can
also evaluate surface water storage in
reservoirs and wetlands, evaluate water
uses, and perform preliminary water quality
analyses The model can be used to
evaluate user-specified alternative control
strategies, and it simulates stream segment
load capacities (LCs) in an attempt to
develop point source wasteload allocations
(WLAs) and nonpomt source load
allocations (LAs) Probability distributions
for runoff and nutrient loadings can be
calculated by the model based on either
smgle-eventor continuous simulations The
model can be applied in urban, agriculture,
or complex watershed simulations The
spreadsheet-based NPSMAP operates
within the Lotus 1-2-3® programming
environment and is capable of producing
graphic output Although this model
requires a minimum calibration effort, it
requires moderate effort to prepare input
data files The current version of the
program considers only nutrient loading,
sediment and other pollutants are not yet
incorporated into the program The model is
easily interfaced with CIS (ARC/INFO) to
facilitate preparation of land use files The
developers also plan to interface the model
with remote sensing facilities for updated
land use information NPSMAP has been
applied to the Tualatin River basin for the
Oregon Department of Environmental
Quality
Generalized Watershed Loading Functions
(GWLF) Model The GWLF model was
developed at Cornell University to assess
the point and nonpomt loadings of nitrogen
and phosphorus from a relatively large,
agricultural and urban watershed and to
16
-------
Compendium of Watershed-Scale Models for TMDL Development
evaluate the effectiveness of certain land
use management practices (Harm and
Shoemaker, 1987) One advantage of this
model is that it was written with the
express purpose of requiring no calibration,
making extensive use of default
parameters. The GWLF model includes
rainfall/runoff and erosion and sediment
generation components, as well as total
and dissolved nitrogen and phosphorus
loadings The current version of this model
does not account for loadings of toxics and
metals, but with minimal effort improve-
ments can be made to add this feature
This model uses daily time steps and allows
analysis of annual and seasonal time series
The model also uses simple transport
routing, based on the delivery ratio
concept In addition, simulation results can
be used to identify and rank pollution
sources and evaluate basmwide
management programs and land use
changes The model also includes several
reporting and graphical representations of
simulation output to aid in interpretation of
the results This model was successfully
tested on a medium-size watershed in New
York (Haith and Shoemaker, 1987) It is
currently being updated to include an
enhanced user interface with potential
Jinking to available national databases for
rapid and effective assessment of
watershed point and nonpomt source
pollution problems
Urban Catchment Model (P8-UCM) The
P8-UCM program was developed for the
Narragansett Bay Project to simulate the
generation and transport of stormwater
runoff pollutants in small urban catchments
and to assess impacts of development on
water quality, with minimum site-specific
data It includes several routines for
evaluating the expected removal efficiency
for particular site plans, selecting or siting
best management practices (BMPs)
necessary to achieve a specified pollutant
removal, and comparing the relative
changes in pollutant loads as a watershed
develops (Palmstrom and Walker, 1990)
Default input parameters can be derived
from NURP data and are available as a
function of land use, land cover, and soil
properties However, without calibration,
the use of model results should be limited
to relative comparisons Spreadsheet-like
menus and on-line help documentation
make extensive user interface possible
On-screen graphical representations of
output are developed for a better
interpretation of simulation results The
model also includes components for
performing monthly or cumulative
frequency distributions for flows and
pollutant loadings
Simplified Particle Transport Model
(SIMPTM) The Simplified Particle
Transport Model was developed by OTAK,
Inc to simulate runoff, sediment, and
pollutant yield from urban watersheds and
to determine the effectiveness of controls
such as street sweeping (Sutherland,
Green, and Jelen, 1990) Runoff volumes,
durations, pollutant loadings, and event
mean concentrations are reported for each
rainfall event on a watershed and a
subcatchment basis Included in the model
results are standard statistics for average
monthly and annual loadings SIMPTM is
an event- or multiple-event-based model for
simulation of urban nonpomt source
loadings for six pollutants, including
sediment, metals, nutrients, and chemical
oxygen demand The model simulates the
accumulation, washoff, and mechanical
removal of pollutants contained in
particulate matter that builds up on streets
77
-------
Compendium of Watershed-Scale Models for TMDL Development
SIMPTM was calibrated using the NURP
data from Lake Hills basin in Bellevue,
Washington NURP data from nearby
Surrey Downs basin were used for testing
SIMPTM The observed variability of runoff
was greater than was predicted by the
model, but the estimates were good
considering the simplistic approach used for
generating runoff SIMPTM was used for
planning stormwater management for the
area surrounding Reno, Nevada
Automated Q-ILLUDAS (AUTO-QI) AUTO-
QI is a watershed model developed by the
Illinois State Water Survey to perform
continuous simulations of stormwater
runoff from pervious and impervious urban
lands (Terstnep et al , 1990) It also allows
the examination of storm events or storm
sequence impacts on receiving water
Critical events are also identified by the
model However, hourly weather input
data are required Several pollutants,
including nutrients, chemical oxygen
demand, metals, and bacteria can be
analyzed simultaneously This model also
includes a component to evaluate the
relative effectiveness of best management
practices An updated version of Q-
ILLUDAS, with an improved user interface
and linkage to the Geographic Information
System (ARC/INFO on PRIME computer),
has been completed by the Illinois State
Water Survey This interface is provided to
generate the necessary input files relating
to land use, soils, and control measures
AUTO-QI was recently verified on the
Boneyard Creek in Champaign, Illinois The
model was also applied to the Greater Lake
Calumet area to determine annual pollutant
loadings to the Calumet and Little Rivers,
south of Chicago
Agricultural Nonpomt Source Pollution
Model (AGNPS) Developed by the USDA
Agricultural Research Service, AGNPS
addresses concerns related to the potential
impacts of point and nonpomt source
pollution on surface and groundwater
quality (Young et al , 1989) It was
designed to quantitatively estimate
pollution loads from agricultural watersheds
and to assess the relative effects of
alternative management programs The
model simulates surface water runoff along
with nutrient and sediment constituents
associated with agricultural nonpomt
sources, and point sources such as
feedlots, wastewater treatment plants, and
stream bank or gully erosion The available
version of AGNPS is storm-event based,
however, the USDA will soon be releasing
a continuous simulation version,
ANNAGNPS The structure of the model
consists of a square grid cell system to
represent the spatial distribution of
watershed properties This grid system
allows the model to be connected to other
software such as Geographical Information
System (GIS) and Digital Elevation Models
(DEM) This connectivity can facilitate the
development of a number of the model's
input parameters Two new terrain-
enhanced versions of the model, AGNPS-C,
a contour-based version, and AGNPS-G, a
grid-based version, have been developed to
automatically generate the grid network
and the required topographic parameters
(Panuska et al , 1991) The model also
includes enhanced graphical represen-
tations of input and output information
AGNPS has been extensively used by
resource agencies in Minnesota and
surrounding States It was also applied to
develop cost-effective nonpomt pollution
control alternatives in Idaho and Illinois and
St Albans Bay, Vermont GIS was linked
18
-------
Compendium of Watershed-Scale Models for TMDL Development
with AGNPS to model sediment loading to
surface waters of the Wet Beaver Creek
watershed of North-Central Arizona
Source Loading and Management Model
(SLAMM) SLAMM was developed at the
University of Alabama to assist in
evaluating the effects of alternative control
practices and development characteristics
on urban runoff quality and quantity (Pitt,
1986) It evaluates only runoff charac-
teristics at the source areas in the
watershed and the discharge outfall, it does
not directly evaluate receiving water
responses However, output data from the
model have been used in conjunction with
other receiving water models to examine
the ultimate effects of urban runoff The
model performs continuous storm mass
balances for both particulate and dissolved
pollutant loadings, and computes runoff
flow volume for different development and
ram characteristics It is intended to be
used as a plannmg-level tool and to provide
rapid assessment of the effects of a
number of urban stormwater control
practices including detention basins,
infiltration devices, porous pavement, street
cleaning, grass swales, and roof runoff
disconnections Proposed features to be
added in an upcoming version of the model
include cost estimates of stormwater
control practices, graphical summaries, and
baseflow and snowmelt predictions
SLAMM was used in Wisconsin to evaluate
the trade-offs between increased street
sweeping, diversion of roof drains to
pervious areas, and detention basins, as
well as combinations of these practices
The costs per unit of suspended solids
removal was compared in order to rank the
control options The model has also been
used to evaluate pathogens from CSOs, for
Madison, Wisconsin's stormwater permit
program, and in Birmingham, Alabama and
in Ottawa and Toronto, Canada
Sewer Overflow Model (SOM) SOM was
developed by Limnotech as a hybridized
version of the urban STORM and SWMM
models It generates pollutant loadings
from urban land uses and evaluates
transport through interconnected sewer line
networks SOM has been applied in six
locations to study CSOs and stormwater
pollution A 15-year continuous simulation
was performed for Portland, Oregon, and
options for controlling dissolved oxygen
(DO) problems associated with large storm
events in Richmond, Virginia were
analyzed It also has been applied for
evaluation of CSOs in Wayne County,
Michigan, and South Bend, Indiana SOM
can perform simulations of an event or a
continuous series of events It is used to
evaluate existing conditions as well as to
perform plannmg-level analysis of controls
2.3.3 Detailed Models
Storage, Treatment, Overflow Runoff Model
(STORM). STORM is a U S Army Corps of
Engineers (COE) model developed for
continuous simulation of runoff quantity
and quality, including sediments and several
conservative pollutants It also simulates
combined sewer systems (Hydrologic
Engineering Center, 1977) STORM has
been widely used for planning and
evaluation of the trade-off between
treatment and storage control options for
CSOs Long-term simulations of runoff
quantity and quality lend themselves to the
construction of duration-frequency
diagrams These diagrams are useful in
developing urban planning alternatives and
designing structural control practices
STORM was primarily designed for
19
-------
Compendium of Watershed-Scale Models for TMDL Development
modeling stormwater runoff from urban
areas. It requires relatively moderate to
high calibration and input data STORM
was initially developed for mainframe
computer usage; however, several versions
have been adapted by various individual
consultants for use on microcomputers
Areal Nonpoint Source Watershed
Environment Response Simulation Model
(ANSWERS). ANSWERS is a
comprehensive model developed at the
University of Georgia to evaluate the
effects of land use, management schemes,
and conservation practices or structures on
the quantity and quality of water from both
agricultural and nonagncultural watersheds
(Beasley, 1986) The distributed structure
of this model allows for a better analysis of
the spatial as well as temporal variability of
pollution sources and loads It was initially
developed on a storm event basis to
enhance the physical description of erosion
and sediment transport processes Data
file preparation for the ANSWERS program
is rather complex and requires mainframe
capabilities, especially when dealing with
large watersheds The output routines are
quite flexible; results may be obtained in
several tabular and graphical forms The
program has been used to evaluate
management practices for agricultural
watersheds and construction sites in
Indiana. It has been combined with
extensive monitoring programs to evaluate
the relative importance of point and
nonpomt source contributions to Saginaw
Bay. This application involved the
computation of unit area loadings under
different land use scenarios for evaluation
of the trade-offs between LAs and WLAs
Recent model revisions include
improvments to the nutrient transport and
transformation subroutines (Dillaha et al ,
1988) Future improvements may include
linkage of the grid-based system with CIS
technology
Distributed Routing Rainfall Runoff Model
(DR3M or DR3M-QUAL) The DR3M
model, including a quality simulation
component, is supported by the U S
Geologic Survey (USGS) (Alley, 1986)
This model was developed for the study of
conventional pollutants m predominantly
urban watersheds and includes components
for sewer flow routing It can be used at
the planning assessment level and as a
design tool It is both a continuous and
storm event-based simulator of nutrient and
sediment loadings The model requires
moderate to high calibration combined with
a high level of personnel training It also
requires mainframe computing capabilities
Adequate input/output features are
incorporated into the model to facilitate use
and result interpretations Time series
hydrographs and pollutographs can be
presented in tabular or graphical form The
USGS has used DR3M-QUAL for calculating
urban runoff pollutographs in Southern
Florida, Anchorage, and Fresno (Donigian
and Huber, 1991)
Simulation for Water Resources m Rural
Basins - Water Quality (SWRRB or
SWRRBQ) The SWRRBQ model was
adapted from the field-scale CREAMS
model by USDA to simulate hydrologic,
sedimentation, nutrient, and pesticide
movement in large, complex rural
watersheds (Arnold et al , 1989)
SWRRBQ uses a daily time step to evaluate
the effect of management decisions on
water, sediment yields, and pollutant
loadings The processes simulated within
this model include surface runoff,
percolation, irrigation return flow,
20
-------
Compendium of Watershed-Scale Models for TMDL Development
evapotranspiration, transmission losses,
pond and reservoir storage, sedimentation,
and crop growth The model is useful for
estimation of the order of magnitude of
pollutant loadings from relatively small
watersheds or watersheds with fairly
uniform properties Input requirements are
relatively high, and experienced personnel
are required for successful simulations
The National Oceanic and Atmospheric
Administration (NOAA) is using SWRRBQ to
evaluate pollutant loadings to coastal
estuaries and embayments as part of its
national Coastal Pollution Discharge
Inventory The model has been run for all
major estuaries on the east coast, west
coast, and gulf coast for a wide range of
pollutants (Donigian and Huber, 1991)
Storm Water Management Model (SWMM)
SWMM is a comprehensive watershed-scale
model developed by EPA (Huber and
Dickinson, 1988) It was initially developed
to address urban stormwater and assist in
storm-event analysis and derivation of
design criteria for structural control of
urban stormwater pollution Recently,
SWMM was upgraded to allow continuous
simulation and application to complex
watersheds and land uses SWMM can be
used to model several types of pollutants
provided that input data are available No
user-enhanced interface is currently
provided to assist in input data entry
However a windows-based user interface is
under development Recent versions of the
model can be used for either continuous or
storm event simulation with user-specified
variable time steps The model is relatively
data-intensive and requires special effort for
validation and calibration Its application in
detailed studies of complex watersheds
may require a team effort and highly trained
personnel. SWMM has been applied to
address various urban water quantity and
quality problems in many locations in the
United States and other countries (Huber,
1989, Donigian and Huber, 1991) In
addition to developing comprehensive
watershed-scale planning, typical uses of
SWMM include predicting CSOs, assessing
the effectiveness of BMPs, providing input
to short-time- increment dynamic receiving
water quality models, and interpreting
receiving water quality monitoring data
(Donigian and Huber, 1991)
The Hydrological Simulation Program -
FORTRAN (HSPF) HSPF is a
comprehensive package developed by the
U S EPA for simulating water quantity and
quality for a wide range of organic and
inorganic pollutants from agricultural
watersheds (Barnwell and Johanson,
1981) The model uses continuous
simulations of water balance and pollutant
generation, transformation, and transport
Time series of the runoff flow rate,
sediment yield, and user-specified pollutant
concentrations can be generated at any
point m the watershed The model also
includes in-stream quality components for
nutrient fate and transport, BOD, DO, pH,
phytoplankton, zooplankton, and benthic
algae Statistical features are incorporated
in the model to allow for frequency-duration
analysis of specific output parameters
Data requirements for HSPF are extensive,
and calibration and verification are
recommended The program is maintained
on IBM microcomputers and DEC/VAX
systems Because of its comprehensive
nature, the HSPF model requires highly
trained personnel It is recommended that
its application to real case studies be
carried out as a team effort The HSPF
model has been extensively used for both
screening-level and detailed analyses,
21
-------
Compendium of Watershed-Scale Models for TMDL Development
including application to pesticide runoff 1986), and analysis of best management
testing (Lorberand Mulkey, 1982), aquatic practices (Donigian et al , 1983)
fate and transport testing (Mulkey et al ,
22
-------
3. OTHER MODELS
Many models have been developed to
assess pollutant movement, transformation,
and leaching in field-scale studies Like the
watershed-scale models, these models
range in detail from empirical relationships
to physical and deterministic simulation
models Many of these models provide for
detailed analysis of the efficiencies of
management practices and evaluation of
alternative control options Although these
models are not applicable to watershed or
subwatershed analysis, they may be used
in the TMDL process to assess more
localized problems and to help develop
appropriate nonpomt source pollution
control strategies
The Occoquan Method The Occoquan
Method was developed by the Northern
Virginia Planning District Commission,
Annandaie, Virginia, to determine the
design criteria required to achieve a given
level of pollutant removal from development
areas using structural practices The
approach develops a site runoff coefficient
as a function of imperviousness after
development Overall pollutant removal
efficiency for a given site is derived from
the type of control practices used and the
fraction of the total area controlled This
method can be used for reviewing the
adequacy of proposed control practices It
was used in the development of BMPs for
the Occoquan watershed (Northern Virginia
Planning District Commission, 1987)
Water Resources Evaluation of Nonpoint
Silvicultura! Sources (WRENS) WRENS
was originally developed by EPA to
evaluate the effects of forest and
silvicultural activities on water quality
(USEPA, 1980) The method is compiled in
a handbook that provides various
quantitative techniques to estimate
potential changes in stream flow, surface
erosion and sediment yield, and water
temperature The handbook also provides
directions on how to compare alternative
silvicultural management practices The
basic formulations of this approach are
similar to those developed in the EPA
screening procedures
Chemicals, Runoff, and Erosion from
Agricultural Management Systems
(CREAMS) CREAMS is the most detailed
field-scale model available to evaluate
agricultural runoff It was developed by the
U S Department of Agriculture (USDA) to
provide detailed information necessary for
designing agricultural management systems
(Knisel, 1980) CREAMS is a physically
based, daily simulation model that
estimates runoff, erosion/sediment
transport, plant nutrients, and pesticide
yield from field-size areas CREAMS can
compare relative effects of different
agricultural BMPs The model is data-
intensive and requires highly trained
personnel However, it may be applied
with minimum calibration efforts A
microcomputer version is currently
available The hydrology, erosion and
sediment, and pesticides components of
CREAMS were used to develop
Groundwater Loading Effects of Agricultural
Management Systems (GLEAMS) (Leonard
et al , 1987) to simulate the vertical
23
-------
Compendium of Watershed-Scale Models for TMDL Development
movement of pesticides and evaluate
effects on groundwater quality
Pesticides Root Zone Model (PRZM) PRZM
was developed by the EPA Environmental
Research Laboratory, Athens, Georgia, to
simulate chemical movement in the
unsaturated zone within and below the
plant root zone (Carsel et al , 1984) It
considers pesticide application and
accounts for plant uptake, decay, and
transformation, and the dissolved,
adsorbed, and vapor phase concentrations
This model is linked to other models to
simulate the transport and fate of
pesticides and to assess the risk to drinking
water wells. PRZM is a relatively
comprehensive model requiring highly
trained personnel However, it requires a
moderate level of input data
The Water Erosion Prediction Project
{WEPP). WEPP is a simulation model using
daily time steps It is the result of a multi-
agency effort to enhance researchers'
ability to model erosion and sediment
transport processes at field-scale and
watershed levels (Laflan, Lane, and Foster,
1991). The model is currently under
development and is intended to be used in
soil erosion assessment and to provide
necessary information for siting agricultural
management practices on specific fields
Eutromod. Eutromod is a spreadsheet-
based modeling procedure for
eutrophication management developed at
Duke University and distributed by the
North American Lake Management Society
(Reckhow, 1990). It is a watershed and
lake model designed to estimate nutrient
loadings, various trophic state parameters,
and tnhalomethane concentrations in lake
water. The computation algorithms used in
Eutromod were developed based on
statistical relationships and a continuously
stirred tank reactor model At present, the
model is specific to watersheds in the
southeastern United States with multiple
land uses including rural and urban areas,
feedlots, septic tanks, and discharges from
wastewater treatment plants and
construction sites Model results include
the most likely predicted phosphorus and
nitrogen loading for the watershed and for
each land use category The model also
determines the lake response to various
pollution loading rates The spreadsheet
capabilities of the model allow graphical
representations of the results and data
export to other spreadsheet systems for
statistical analyses
Precipitation-Runoff Modeling System
(PRMS) PRMS is a deterministic physical-
process model developed by the USGS to
evaluate the impacts of various
combinations of precipitation, climate, and
land use distribution on surface water
runoff, sediment yield, and general
watershed hydrology (Leavesley et al ,
1983) It simulates watershed response to
normal and extreme rainfall and snowmelt
and determines changes in water balance
relationships, flood peaks, and groundwater
recharge It includes several capabilities for
parameter optimization and sensitivity
analysis The model divides the watershed
into homogeneous hydrologic-response
units (HRUs) and can be applied to both
agricultural and urban land uses Input
variables include descriptive data on the
physiography, vegetation, soil, and
hydrologic and climatic characteristics It
is, however, designed to run with data
retrieved directly from the USGS's National
Water Data Storage and Retrieval
(WATSTORE) system The modular system
24
-------
Compendium of Watershed-Scale Models for TMDL Development
approach provides an adaptable modeling
system for both management and research
applications PRMS was initially developed
for use on mainframe computers New
model components proposed by the USGS
as future additions to PRMS include water-
quality routines and expanded saturated-
unsaturated flow and groundwater flow
components
The Unified Transport Model for Toxic
Materials (UTM-TOX). UTM-TOX was
developed by the Oak Ridge National
Laboratory for the analysis of hydrologic,
atmospheric, and sediment transport of
pesticides and toxic substances (Patterson
et al , 1983) This model consists of a
multi-media simulation approach
Considering a chemical release to the
atmosphere from a given source (e g ,
stack, area, or line source), the model uses
mass balance formulations to compute
chemical fluxes from the source through
the atmosphere, deposition on watersheds,
transport in surface runoff and percolation
through the soil profile, and transport in
sediment and streamflow The model
generates summary tables and plots of the
average monthly and annual chemical
concentrations in the various media It also
considers biotic processes and computes
chemical accumulation m stems, leaves,
and fruits of impacted vegetation Limited
applications of this model were reported in
the literature due primarily to the complex
nature of the model and the lack of user
support (Donigian and Huber, 1991)
Exposure Analysis Modeling System
(EXAMS) EXAMS was developed for the
USEPA to provide rapid evaluations of the
behavior of synthetic organic chemicals in
aquatic systems (Burns and Cline, 1985)
The initial version of EXAMS computes the
long term results of continual, steady
discharges of single chemicals into typical
aquatic systems The new version includes
updated routines which consider the
seasonal variations of transport and
transformation kinetics and computes the
fate and transport of products resulting
from transforation reactions The updated
version also includes capabilities for flexible
timing and durations of chemical loadings.
The model requires laboratory information
on the reactivity and transformation of
chemicals as well as state variables
describing the transport mechanisms and
physical/chemical properties of the
receiving water This model is widely used
for exposure analysis and for deriving basic
information necessary to perform health
and ecological risk assessments
The Water quality Analysis Simulation
Program (WASP4) WASP4 is a detailed
receiving water quality model supported by
the US EPA (USEPA, 1988) It allows
users to interpret and predict water quality
responses to natural phenomena and man-
made stresses for various pollution
management decisions WASP4 is a
dynamic compartment modeling program
for aquatic systems, including both the
water column and the underlying benthos
The model includes the time-varying
processes of advection, dispersion, point
and nonpomt mass loading, and boundary
exchanges WASP4 may be applied in two
modes (1) EUTROWASP for nutrient and
eutrophication analyses and (2) TOXIWASP
for analysis of toxic pollutants and metals
The flexibility of WASP4 is unique in that it
permits the modeler to structure one, two,
or three dimensional model applications to
rivers, lakes, estuaries, or open coastal
areas The model's computer code is
structured to permit easy development of
25
-------
Compendium of Watershed-Scale Models for TMDL Development
new kinetic or reactive subroutines without
having to rewrite large sections of
computer code The mam advantage to
using WASP4 is the ability to more
realistically portray the relative spatial
influence of each major pollutant source on
the receiving water as well as a more
accurate representation of the transport/
water quality kinetics phenomena (i e
mixing and diurnal process) It may also be
useful to assist in designing long term
monitoring programs for receiving waters
The model has been successfully applied to
a wide range of estuary water bodies such
as the Peconic Bay, Long Island, and the
Bird River estuary, Maryland
Enhanced Stream Water Quality Mode!
(QUAL2E). QUAL2E is an US EPA
supported model. It is a one dimensional
(longitudinal water quality model) that
assumes steady state flow but allows
simulation of diurnal variations in
temperature or algal photosynthesis and
respiration. QUAL2E simulates a series of
nonuniform segments that make up a river
reach and incorporates the effects of
withdrawals, branches, and tributaries
Water quality parameters simulated include
conservative substances, temperature,
bacteria, BOD, DO, ammonia, nitrate,
nitrate and organic nitrogen, phosphate and
organic phosphorus, and algae QUAL2E is
widely used for stream waste load
allocations and discharge permit
determinations in the United States and
other countries (USEPA, 1992a)
Simplified Method Program - Toxics
(SMPTOX) SMPTOX is an USEPA
supported model It was developed to
provide a user-friendly microcomputer
program for performing screening level
modeling of toxic chemical concentrations
resulting from discharges from POTWs into
streams and rivers It provides a simplified
technique for performing load allocation for
dissolved oxygen/ammonia/CBOD Because
of its simplistic approach, several versions
of SMPTOX are currently available (USEPA,
1992b) The program contains a full
screen editor to facilitate the entry and
modification of input data as well as high
resolution graphics to present model
results
26
-------
4. MODEL SELECTION
Watershed models are becoming more and
more integrated at various levels of water
quality analysis At the same time,
selecting the model that best matches the
project or objective in hand is becoming
difficult as more models are being
developed The watershed models
presented in the previous section cover a
wide range of functions and can be applied,
either directly or with minimum
modification, to the majority of planning
problems associated with point and
nonpomt source pollution Selection of a
watershed or water quality model may be
considered an important decision process,
not only because of the time and resources
a modeling effort involves, but also because
of the wide variety and amount of
information a water quality project may
require
4.1 Model Characteristics
In addition to information presented in the
previous sections of this compendium,
relevant characteristics generally associated
with model selection for specific
applications are presented in this section
Tables 4, 5, and 6 present the basic
simulation functions used in each model to
generate pollutant loadings Most water-
shed models include three components a
hydrology component, which estimates the
quantity of runoff and streamflow
generated from the watershed or
subwatersheds, an erosion and sediment
component, which derives the amount of
sediment delivered to a receiving water
body, and a quality component, which
computes the pollutant loadings Tables 4-
6 also present the type of pollutant handled
by each model and the corresponding
computation time steps
As shown in Tables 4-6, most models are
based on similar mathematical formulations
The curve number equation (CNE)
developed by the USDA-SCS is widely used
for simulating runoff and stream flows
(e g , NPSMAP, GWLF, P8-UCM, AGNPS,
STORM, SWRRBQ), and the Universal Soil
Loss Equation (USLE) is commonly used for
determining erosion and sediment yield
from rural areas or watersheds (e g , EPA
screening procedures. Water Screen,
Watershed, SLOSS-PHOSPH, GWLF,
AGNPS, STORM, SWRRBQ) Pollutant
loadings from rural areas are often
calculated based on loading functions or
potency factors (e g , EPA screening
procedures, Water Screen, AGNPS,
SWRRBQ, HSPF) For urban areas, unit
area loading rates (e g , GWLF) or build-up
and washoff functions (e g , STORM,
SWMM) are widely used The advantage
of the CNE- and USLE-based models is that
detailed default parameters are available for
a wide variety of soil conditions and
agricultural management techniques The
differences among models using similar
simulation functions reside in the degree of
spatial discretization they use, the number
27
-------
Table 4. A Descriptive List of Model Components - Simple Methods
Model
EPA Screening
Procedures
The Simple
Method
Water Screen
Watershed
FHWA
WMM
SLOSS/PHOSPH
Regression
Mam Land
Use
Mixed
watershed
Urban
Mixed
watershed
Mixed
watershed
Highways
Mixed
watershed
Rural
Urban
Hydrology
N/A
Runoff coefficient
N/A
N/A
Runoff coefficient,
observed data
Runoff coefficient
N/A
N/A
Erosion/
Sediment
USLE-
MUSLE
N/A
USLE
USLE
N/A
N/A
USLE
N/A
Pollutant
Load
Loading functions,
potency factors
Mean
concentration
Loading functions,
potency factors
Unit area loadings
Median
concentration
Event mean
concentration
Loading functions
Regression
equations
Pollutants
Wide range1
NURP data
TSS, P, metals,
O&G
N, P, organics
Wide range1
TSS, N, P
Organics, metals
N, P, lead,
zinc
P
TSS, N, P,
COD, metals
Time
Scale
Mean annual
Variable
(annual,
monthly,
event)
Mean annual
Annual
Storm event
Annual
Annual
Storm event
I
I
0?
I
I
«•
i
I
t
i
s
I
i
1 Depends on available pollutant parameters and default data
N Nitrogen
0 + G Oil and gas
P Phosphorus
TSS Total suspended solids
COD Chemical oxygen demand
-------
Table 5 A Descriptive List of Model Components - Mid-Range Models
Model
NPSMAP
GWLF
P8-UCM
SIMPTM
Auto-QI
AGNPS
SLAMM
Main Land
Use
Mixed
watershed
Mixed
watershed
Urban
Urban
Urban
Agriculture
Urban
watershed
Hydrology
SCS curve number
SCS curve number
SCS curve number
-modified
TR20
Trapezoidal
hyetograph
Water balance
SCS curve number
Small storm-based
coefficient
Erosion/
Sediment
N/A
Modified USLE
N/A
Modified Yalm
equation
N/A
Modified USLE
N/A
Pollutant
Load
Runoff
concentration
Unit loading
rates
Nonlinear
accumulation
Nonlinear
accumulation
Accumulation
and washoff
Potency
factors
Nonlinear
accumulation
and washoff
Pollutants
N, P
N, P
TSS, N, P,
metals
Wide Range1
Wide Range1
N, P
N, P, COD
bacteria,
metals
Time
Scale
Continuous
Continuous
Storm
sequence
Storm
sequence
Storm event,
Continuous
Storm event
Continuous
o
I
I
I
t
1 Depends on available pollutant parameters and default values
N Nitrogen
0 + G Oil and gas
P Phosphorus
TSS Total suspended solids
COD Chemical oxygen demand
-------
Table 6. A Descriptive List of Model Components - Detailed Models
Model
ANSWERS
SWMM
HSPF
STORM
SWRRB
DR3M
Main Land
Use
Agriculture
Urban
Mixed
watershed
Urban
Agriculture
Urban
Hydrology
Distributed storage
model
Nonlinear reservoir
Water balance of land
surface and soil
processes
Runoff coefficient -
SCS curve numbers -
Unit hydrograph
SCS curve number
Surface storage
balance
kinematic wave
method
Erosion/
Sediment
Detachment
transport
equations
Modified USLE
Detachment/
washoff
equations
USLE
Modified USLE
Related to runoff
volume and peak
Pollutant
Load
Potency factors
(correlation with
sediment)
Buildup/washoff
functions
Loadmg/washoff
functions and
subsurface
concentrations
Buildup/washoff
functions
Loading functions
Buildup/washoff
functions
Pollutants
N/A
Wide range1
Wide range1
P, N, OD,
metals
N, P, OD,
metals,
bacteria
TSS, N, P,
organics,
metals
Time
Scale
Storm event
Storm event,
continuous
Storm event,
Continuous
Continuous
Continuous
Continuous
s
o
-s
I
I
$
i
1 Depends on available pollutant parameters and default values
N Nitrogen
0 + G Oil and gas
P Phosphorus
TSS Total suspended solids
COD Chemical oxygen demand
-------
Compendium of Watershed-Scale Models for TMDL Development
of processes for which they account, and
the computational time steps they use
Many of the simple methods do not take
hydrologic processes into account when
simulating pollutant loads When dealing
with urbanized areas, simple methods
usually generate runoff based on empirical
or statistical relationships between runoff
coefficients and the degree of
imperviousness (e g , the Simple Method,
FHWA, and WMM) It is, however, difficult
to extrapolate such relationships to rural
and agricultural areas
Detailed models use more complex
formulations for simulating runoff and
sediment yield The hydrology component
generally involves a set of deterministic
equations to represent the elements of the
water balance equation (e g , infiltration,
evapotranspiration, groundwater recharge
and/or seepage, depression storage)
These models also use a physical
description of the erosion and sediment
yield mechanisms (e g , soil detachment,
transport, and deposition) Predictions of
pollutant washoff are usually made based
on exponential decay functions (e g ,
SWMM) with hourly time steps Default
values for parameters are pollutant- and
site-specific and therefore may not be
readily available, making calibration difficult
and time-consuming In most cases,
additional laboratory testing and field
measurement may be required
The type and amount of input data required
for operation, calibration, and verification of
the model and the output results should be
considered in the model selection process
Depending on the type of formulations the
model uses, input data may range from
simple watershed characteristics to hourly
meteorological parameters, pollutant
transformation kinetic coefficients, and field
monitoring data Tables 7, 8, and 9
present a brief summary of input and
output information for each of the models
reviewed Novotny and Chesters (1981)
have developed three sets of input
parameters that may be required for a
typical modeling application (Table 10)
Interpretation of the type and amount of
data required, along with information
contained in the preceding tables, may be
used to evaluate the time and resources
required to apply a given model for a given
situation or project For a detailed listing of
input requirements, please refer directly to
the model documentation
Watershed models are usually developed to
target a specific setting, characterized
primarily by land use or land activity Few
models are developed to evaluate
watersheds with mixed land uses Among
the detailed models, HSPF appears to be
the most versatile for watersheds with
complex land use/land cover SWMM,
STORM, and DR3M-QUAL are designed
primarily for urban areas, while ANSWERS
and SWRRB are primarily agricultural
models Among the mid-range models,
NPSMAP and GWLF are the two models
that account for both rural and agricultural
watersheds The GWLF model offers the
possibility of generating long-term time
series of pollutant loadings at various time
steps, allowing analysis of seasonal and
inter-annual variabilities GWLF also allows
evaluation of watershed response to
changes in land use patterns and point and
nonpomt source loadings Urban models
such as P8-UCM, SIMPTM, and SLAMM
were mainly designed for evaluating
management practices to control urban
stormwater runoff Simple methods use
31
-------
Jo
Nj
Table 7. Input and Output Data - Simple Methods
Models
EPA Screening Procedures
The Simple Method
Regression
SLOSS/PHOSPH
Water Screen
Watershed
FHWA
WMM
Main Input Data
Watershed and land use data
Loading factors (default values)
Annual rainfall data
Land use and imperviousness data
Pollutant mean concentration
BMPs removal efficiencies
Mean annual rainfall
Mean minimum January temperature
Drainage areas and land use
Percent imperviousness
Rainfall erosivity factor
Soil, crop, topography, and land use data
Rainfall erosivity factor
Watershed and land use data
Loading factors (default values)
Rainfall erosivity factor
Land use and soil parameters
Unit loading rates
BMP cost information
Site and receiving water data
Flow and storm event concentrations
Land use and soil data
Annual precipitation and evaporation
Inputs from baseflow and precipitation
Event mean concentrations in runoff
Reservoir, lake or stream hydraulic
characteristics
Removal efficiencies of proposed BMPs
Output Information
Mean annual sediment and pollutant loads
Runoff volume and pollutant
concentration/load, storm or annual
Mean annual storm event load and
confidence interval
Mean annual loads of sediment and
phosphorus
Mean annual sediment and pollutant loads
Mean annual pollutant loads
BMP cost-effectiveness
Statistics on storm runoff and
concentrations
Impacts on receiving water
Annual urban and rural pollutant loads
from point and nonpoint sources,
including septic tanks
Load reductions from combined effects of
multiple BMPs
In-lake nutrient concentrations as related
to trophic state, also concentrations
of metals "^r
I
i
o
i
I
I
-------
Table 8. Input and Output Data - Mid-Range Models
f
o
I
I
I
I
I
i
f
Models
NPSMAP
GWLF
P8-UCM
SIMPTM
Auto-QI
AGNPS
SLAMM
Mam Input Data
Meteorologic and hydrologic data, hourly or daily
maximum one year
Watershed and channel parameters
Point sources and pollutant parameters (e.g , decay)
Meteorologic and hydrologic data, daily
Land use and soil data parameters
Nutrient loading rates
Meteorologic and hydrologic data, hourly storm or
storm sequence
Land use and soil parameters
BMP information
Rainfall event statistics
Watershed parameters
Accumulation and washoff rates (default values)
Hourly/daily rainfall
Watershed and land use data
BMP removal rates
Watershed, land use, management, and soil data
Rainfall data, topography
BMP removal data
Hourly rainfall data
Pollution source characteristics, areas, soil type,
imperviousness, and traffic
Structure characteristics
Output Information
Runoff and nutrient loadings
Pollution load allocations
Monthly and annual time series of runoff,
sediment, and nutrients
Daily runoff and pollutant loads
BMP removal efficiencies
Storm runoff volume and hydrograph
characteristics
TSS/sediment and pollutant washoff
Continuous or storm event simulation of
runoff and selected pollutants
Storm runoff volume and peak flow
Sediment, nutrient, and COD concentrations
Pollutant load by source area
BMP evaluation and cost estimates
-------
Table 9. Input and Output Data - Detailed Models
I
Models
Main Input Data
Output Information
STORM
Hourly rainfall data
Buildup and washoff parameters
Runoff coefficient and soil type
Event-based runoff and pollutant loads
Storage and treatment utilization and
number of overflows
Hourly hydrographs and pollutographs
I
ANSWERS
Hourly rainfall data
Watershed, land use, and soil data
BMP design data
Predicts storm runoff (volume and peak flow),
Sediment detachment and transport
Analysis of relative effectiveness of
agricultural BMPs
DR3M
Meteorologic and hydrologic data
Watershed characteristics related to runoff
Channel dimensions and kinematic wave parameters
Characteristics of storage basins
Buildup and washoff coefficients
Continuous series of runoff and pollutant yield
at any location in the drainage system.
Summaries for storm events
Hydrographs and pollutographs
I
SWRRB
Meteorologic and hydrologic data
Watershed and receiving waterbody parameters
Land use and soil data
Ponds and reservoir data
Continuous water and sediment yield
Peak discharge
Water quality concentrations and loads
SWMM
Meteorologic and hydrologic data
Land use distribution and characteristics
Accumulation and washoff parameters
Decay coefficients
Continuous and event-based runoff and
pollutant loads
Transport through streams and reservoirs
Analysis of control strategies
HSPF
Meteorologic and hydrologic data
Land use distribution and characteristics
Loading factors and washoff parameters
Receiving water characteristics
Decay coefficients
Time series for runoff and pollutant loadings
Analysis of impacts on receiving water
Analysis of controls
-------
Compendium of Watershed-Scale Models for TMDL Development
Table 10. Input Data Needs for Watershed Models
(after Novotny and Chesters, 1981)
System Parameters
Watershed size
Subdivision of the watershed into homogenous subareas
Imperviousness of each subarea
Slopes
Fraction of impervious areas directly connected to a channel
Maximum surface storage (depression plus interception storage)
Soil characteristics including texture, permeability, credibility, and composition
Crop and vegetative cover
Curb density or street gutter length
Sewer system or natural drainage characteristics
State Variables
Ambient temperature
Reaction rate coefficients
Adsorption/desorption coefficients
Growth stage of crops
Daily accumulation rates of litter
Traffic density and speed
Potency factors for pollutants (pollutant strength on sediment)
Solar radiation (for some models)
Input Variables
Precipitation
Atmospheric fallout
Evaporation rates
55
-------
Compendium of Watershed-Scale Models for TMDL Development
generic empirical relationships that can be
used in both rural and urban settings
provided site-specific or default values are
available.
Model applications may be classified as
screening, intermediate, or detailed
depending on the focus and objectives of
the application. Simple methods are most
frequently used for screening applications,
however, mid-range and detailed models
may allow for a wider range of applications
Screening applications are generally
performed at the preplanning level, with
specific objectives such as comparisons of
the relative contribution of point and
nonpoint sources using a relatively limited
set of available information Screening
analyses may consider a broad range of
land use types and sources and may be
performed at various stages of project
development (e g , planning, evaluation of
alternatives, preliminary design) At the
planning level, screening applications may
be directed toward scoping the project
objective and identifying general areas
where controls or additional sampling may
be required.
Intermediate applications provide a more
detailed description of the geographic
variables contributing to nonpoint pollution,
in addition to consideration of multiple point
sources. Intermediate applications may
assist in identifying specific point and
nonpoint source activities and in preliminary
selection of pollution control options
incorporating a higher degree of spatial
variation within land uses
As it becomes necessary to accurately
distinguish differences in pollutant
characteristics from multiple source areas,
pollutant behavior is considered in more
detail and a more mechanistic description
of pollutant generation, transformation, and
removal by various control practices is
required Detailed applications are,
therefore, necessary to provide either
storm-based or continuous simulation of
water and water quality processes and to
assist in developing design criteria for
achieving project objectives
The potential range of applications of
watershed models m planning, evaluation of
management measures, and analysis of
impacts on the quality of receiving waters
is illustrated in Tables 11, 12, and 13 The
tables show that the majority of the models
may be used for screening-level
applications The simple methods in
particular provide only an order-of-
magnitude estimate on an annual basis and
therefore are limited to screening
applications at the planning level Some of
the mid-range models (e g , GWLF,
NPSMAP, and AGNPS) incorporate point
and nonpoint source pollution routines and
are also good candidates for screening
activities SLAMM, P8-UCM, and SIMPTM
are primarily urban runoff models, and their
application to evaluation of urban
stormwater control practices and strategies
may be useful at an intermediate level
SWMM, HSPF, DR3M, STORM, and
SWRRB stand out from the others as
models capable of providing a detailed
indication of the contribution of pollutants
from various point and nonpoint sources
Their simulation capabilities allow for
evaluation of control strategies and
development of design criteria
Application of detailed models such as
HSPF and SWMM for screening purposes,
using estimated default values for a number
of parameters, may reduce time and input
36
-------
Compendium of Watershed-Scale Models for TMDL Development
requirements However, representative
default values for many of the detailed
models are difficult to obtain In addition,
their accuracy as screening tools may be
jeopardized by replacing mechanistic
equations with their simplified forms and
including inappropriate default values
Urban stormwater runoff models, such as
SWMM, HSPF, SLAMM, P8-UCM, and
DR3M-QUAL, are capable of providing
design criteria for a number of structural
practices Models with such capabilities,
however, are data-intensive and require
trained professionals to operate the model,
select appropriate default values, and
interpret the results
4.2 Model Calibration and Verification
The results of watershed simulations are
more meaningful when they are
accompanied by some sort of confirmatory
analysis The capability of any model to
accurately depict water quality conditions is
directly related to the accuracy of input
data and the level of expertise required to
operate the model It is also largely
dependent on the amount of data available
Detailed models lacking the required
calibration and verification data are limited
in accuracy.
Table 11 Range of Application of Watershed Models - Simple Methods
Simple
Methods
EPA Screening
The Simple Method
Regression
SLOSS/PHOSPH
Water Screen
Watershed
FWHA
WMM
Watershed Analysis
Screening
•
•
•
o
•
•
•
•
Intermediate
-
-
-
-
-
-
-
O
Detailed
-
-
-
-
-
-
-
-
Control Analysis
Planning
-
O
-
-
-
0
o
€
Design
-
-
_
-
-
-
-
-
Receiving
Water
Quality
O
-
-
-
-
-
O
€
High
Medium
O Low
- Not Available
37
-------
Compendium of Watershed-Scale Models for TMDL Development
Table 12 Range of Application of Watershed Models - Mid-Range Models
Mid-Range
Methods
NPSMAP
GWLF
P8-UCM
SIMPTM
Auto-QI
AGNPS
SLAMM
Watershed Analysis
Screening
•
•
•
o
•
•
•
Intermediate
O
€
€
€
•
•
G
Detailed
0
o
G
G
0
O
0
Control Analysis
Planning
G
-
O
€
O
•
•
Design
-
-
•
O
o
o
o
Receiving
Water
Quality
0
_
-
-
0
0
0
High
Medium
O Low
- Not Available
Table 13. Range of Application of Watershed Models - Detailed Models
Detailed
Methods
STORM
ANSWERS
SWRRBQ
DR3M-Q
SWMM
HSPF
Watershed Analysis
Screening
•
•
0
G
G
G
Intermediate
•
•
•
•
•
•
Detailed
O
€
•
•
•
•
Control Analysis
Planning
•
•
•
•
•
•
Design
O
O
O
C~
G
0
Receiving
Water
Quality
O
O
G
G
-
•
High
Q Medium
O Low
- Not Available
-------
Compendium of Watershed-Scale Models for TMDL Development
Calibration involves minimization of
deviation between measured field
conditions and model output by adjusting
parameters of the model (Jewell et al ,
1978) Data required for this step are a set
of known input values along with
corresponding field observation results
The results of the sensitivity analysis
provide information as to which parameters
have the greatest effect on output For the
best results, CSO models should be
calibrated during storm events as opposed
to dry flow periods (Water Pollution Control
Federation, 1989)
Verification involves the use of a second
set of independent information to check the
model calibration The data used for
verification should consist of field
measurements of the same type as the data
output from the model Specific features
such as mean values, variability, extreme
values, or all predicted values may be of
interest to the modeler and require testing
(Reckhow and Chapra, 1983) Models are
tested based on the levels of their
predictions, whether descriptive or
predictive More accuracy is required of a
model designed for absolute versus relative
predictions If the model is calibrated
properly, the model predictions will be
acceptably close to the field observations
Observed data for model calibration and
verification may, in many cases, be
insufficient or unavailable Model selection
must be based on an assessment of the
available data Screening-level applications
may be possible with limited input data As
noted by Donigian and Rao (1988), most
models are more accurate when applied in
a relative rather than an absolute manner
Model output data concerning the relative
contribution of a watershed to overall
pollutant loads is more reliable than an
absolute prediction of the impacts of one
control alternative viewed alone When
examining model output from watershed-
pollution sources, it is important to note
three factors that may influence the model
output and produce unreasonable data
First, suspect data may result from
calibration or verification data that are
insufficient or inappropriately applied
Second, any given model, including detailed
models, may not represent enough detail to
adequately describe existing conditions and
generate reliable output Finally, modelers
should remember that all models have
limitations and the selected model may not
be capable of simulating desired conditions
Model results must therefore be interpreted
within the limitations of their testing and
their range of application Inadequate
model calibration and verification can result
in spurious model results, particularly when
used for absolute predictions Data
limitations may require that model results
be used only for relative comparisons
39
-------
5. REFERENCES
Addiscott, T. M., and R J Wagenet
1985. Concepts of solute leaching in soils
A review of modeling approaches Journal
of Soil Science 36 411-424
Alley, W.M 1986 Summary of experience
with the distributed routing rainfall-runoff
model (DR3M). In Urban Drainage
Modeling, ed. C Maksimovic and M
Radojkovic, pp 403-414 Pergamon Press,
New York.
Amy, G.R., R Pitt, W L Singh, Bradford,
and M.B. LaGraff. 1974 Water quality
management for urban runoff. , U S
Environmental Protection Agency,
Washington, D.C , EPA 440/9-75/004
(NTIS PB 241 689/AS)
Arnold, J.G., J R. Williams, A D Nicks, and
N.B. Sammons 1989. SWRRB, a basin
scale simulation model for soil and water
resources management Texas A&M Press
Barnwell, T.O., and R Johanson 1981
HSPF: A comprehensive package for
simulation of watershed hydrology and
water quality. Nonpoint pollution control:
Tools and techniques for the future.
Interstate Commission on the Potomac
River Basin, Rockville, MD.
Beasley, D B. 1986 Distributed parameter
hydrologic and water quality modeling
Agricultural Nonpoint Source Pollution:
Model Selection and Application, ed
Giorgmi and F. Zmgales. pp 345-362
Bird, B L. and KM Conaway 1985
WATER SCREEN - A microcomputer
program for estimating nutrient and
pollutant loadings In Proceedings of the
Stormwater and Water Quality Model Users
Group Meeting, April 12-13, 1984 ed
TO Barnwell, pp 121-174 EPA-600/9-
85-0013
Biswas, A K 1975 Mathematical
modelling and environmental decision-
making Ecological Modelling 1 31-48.
Burns, LA, and DM Clme 1985
Exposure Analysis Modeling System -
Reference manual for EXAMS II US
Environmental Protection Agency,
Environmental Research Laboratory,
Athens, Georgia EPA/600/3-85/038.
Camp, Dresser, and McKee (COM) 1992.
Watershed Management Model user's
manual, version 2 0, Prepared for the
Florida Department of Environmental
Regulation, Tallahassee, FL
Carsel R F , C.N Smith, L A Mulkey, J D
Dean, and P Jowise. 1984 User's
manual for the Pesticide Root Zone Model
(PRZM): release 1 US Environmental
Protection Agency, Environmental Research
Laboratory, Athens, GA EPA-600/3-84-
109
Clark, W C , D D Jones, and C S Hollmg
1979 Lessons for ecological policy design
A case study of ecosystem management
Ecological Modelling 7 1-53
40
-------
Compendium of Watershed-Scale Models for TMDL Development
Dillaha, TA 1992 Nonpomt source
modelling for evaluating the effectiveness
of best management practices NWQEP
Notes 52(March) 5-7
Dillaha, T A , C D Heatwole, M R Bennett,
S Mostaghimi, V O Shanholtz, and B B
Ross 1988 Water quality modeling for
nonpoint source pollution control planning:
Nutrient transport. Virginia Polytechnic
Institute and State University, Dept of
Agricultural Engineering Report No SW-
88-02
Donigian, A S , G C Imhoff, B R Bicknell
1983 Modeling water quality and the
effects of BMPs in Four Mile Creek, Iowa.
U S. Environmental Protection Agency,
Environmental Research Laboratory,
Athens, GA
Donigian, AS., D W Meier, and P P
Jowise. 1986 Stream transport and
agricultural runoff for exposure assessmen t:
A methodology. Environmental Research
Laboratory, U S EPA, Athens, GA
EPA/600/3-86-011
Donigian, A S , and W C Huber 1991
Modeling of nonpoint source water quality
in urban and non-urban areas U S
Environmental Protection Agency
EPA/600/3-91/039
Donigian, A.S , and PSC Rao 1988
Selection, application, and validation of
environmental models. Draft Prepared for
presentation at the International
Symposium on Water Quality Modeling of
Agricultural Non-Point Sources Utah State
University, Logan, UT June 19-23, 1988.
Federal Highway Administration, 1990
Pollutant loadings and impacts from
highway stormwater runoff Research,
Development, and Technology U S
Department of Transportation, McLean, VA
Haith, DA, and LL Shoemaker 1987
Generalized watershed loading functions for
stream flow nutrients Water Resources
Bulletin 107(EEI) 121-137
Huber, WC 1989 User feedback on
public domain software SWMM case
study In Proceedings of the Sixth
Conference on Computing in Civil
Engineering, Atlanta, GA
Huber, WC and RE Dickinson 1988
Storm Water Management Model Version
4, User's manual. U S Environmental
Protection Agency, Athens, GA EPA/600/
3-88/001 a (NTIS PB88-236641/AS),
Hydrologic Engineering Center 1977.
Storage, Treatment, Overflow, Runoff
Model, STORM, User's manual
Generalized Computer Program 723-S8-
L7520 U.S Army Corps of Engineers,
Davis, CA
Jewell, T.K , T J Nunno, and D D Adrian
1978 Methodology for calibrating
stormwater models. Journal of The
Environmental Engineering Division, 104
485
Knisel W G 1980 CREAMS, A field scale
model for Chemicals, Runoff, and Erosion
from Agricultural Management Systems.
U S Department of Agriculture,
Conservation Research Report No 26
Laflan J M., L.J. Lane, and G R Foster
1991 WEPP A new generation of erosion
prediction technology. Journal of Soil and
Water Conservation 46 (1). 34-38
41
-------
Compendium of Watershed-Scale Models for TMDL Development
Leavesley, G H, R W Lichty, B M
Troutman, and LG Saindon 1983
Precipitation-runoff modeling system
User's manual. U S Geological Survey,
Central Region, Water Resources Division,
Denver, CO. Water-Resources Inves-
tigation 83-4238
Leonard, R.A , W G Knisel, and D A Still
1987. GLEAMS. Groundwater Loading
Effects of Agricultural Management
Systems Transactions of the ASAE 31 (3)
776-788.
Lorber, M.N , and L A Mulkey 1982 An
evaluation of three pesticide runoff loading
models. Journal of Environmental Quality
11(3): 519-529.
McElroy, A.D , S W Chiu, J W Nabgen, A
Aleti, R.W Bennett 1976 Loading
functions for assessment of water pollution
for non-point sources U S Environmental
Protection Agency EPA 600/2-76/151
{NTIS PB-253325)
Mills, W.B. 1985 Water quality
assessment: A screening procedure for
toxic and conventional pollutants in surface
and ground water. Part 1 Environmental
Research Laboratory, U S Environmental
Protection Agency, Athens, GA
EPA/600/6-85/002a.
Nix, J.S. 1991. Applying urban runoff
models. Water Environment and
Technology, June 1991
Northern Virginia Planning District
Commission 1987 BMP handbook for
the Occoquan watershed Northern Virginia
Planning District Commission, Annandale,
VA.
Novotny, V and G Chesters 1981
Handbook of nonpoint pollution- Sources
and management. Van Nolstrand Remhold
Company, New York, NY
Omicron Associates 1990 Nonpoint
Pollution Source Model for Analysis and
Planning (NFSMAP) - Users manual
Omicron Associates, Portland, OR
Palmstrom, N , and W W Walker, Jr
1990 P8 urban catchment model- User's
guide. Program documentation, and
evaluation of existing models, design
concepts, and Hunt-Potowomut data
inventory The Narragansett Bay Project
Report No NBP-90-50
Panuska, J C , ID Moore, and L A
Kramer 1991 Terrain analysis
Integration into the agricultural nonpoint
source (AGNPS) pollution model Journal
of Soil and Water Conservation 46(1) 59-
64
Patterson, M
Sjoreen, M G
D M Hetrick,
Randon 1983
TOX, A unified
by Oak Ridge
Ridge, TN, for
Substances
R , T J. Sworski, A L
Brownman, C C Coutant,
B D Murphy, and R J
A user's manual for UTM-
transport model. Prepared
National Laboratory, Oak
U S EPA Office of Toxic
Pitt, R 1986 Runoff controls in
Wisconsin's priority watersheds In Urban
Runoff Quality, ed Urbonas, B , and L A
Roesner Proceedings of the Engineering
Foundation Conference on Urban Runoff
Quality (ASCE), June 22-27, 1986
Henniker, NH
Reckhow, KH 1990 EUTROMOD
Watershed and lake modeling software.
42
-------
Compendium of Watershed-Scale Models for TMDL Development
Software package No. 1. North American
Lake Management Society, Alachua, FL
Reckhow, K H , and S C Chapra 1983
Confirmation of water quality models
Ecological Modelling 20 113-133
Schueler, T 1987 Controlling urban
runoff: A practical manual for planning and
designing urban BMPs Metropolitan
Washington Council of Governments,
Washington, DC
Stewart, B A , DA Woolhiser, W H
Wischmeier, J H Cara, and M H Frere
1975 Control of Water Pollution From
Croplands. Vol I U S Environmental
Protection Agency, Washington, DC
EPA600/2-75/026a
Sutherland, R C , D L Green, and S L
Jelen 1990 Simplified Paniculate
Transport Model. User's manual OTAK,
Inc , Lake Oswego, OR
Tasker, G D , and N E Driver 1988
Nationwide regression models for predicting
urban runoff water quality at unmonitored
sites Water Resources Bulletin
24(5) 1091-1101
Terstnep, M L , M T Lee, E P Mills, A
V Greene, and M R Rahman 1990
Simulation of urban runoff and pollutant
loading from the greater Lake Calumet area,
Prepared by the Illinois State Water Survey
for the U S Environmental Protection
Agency, Region V, Water Division,
Watershed Management Unit, Chicago, IL
Tim, U S , S Mostaghimi, V O Shanholtz,
and N Zhang 1991 Identification of
critical nonpomt pollution source area using
geographic information systems and
simulation modeling In Proceedings of the
American Society of Agricultural Engineers
(ASAE) International Winter Meeting,
Albuquerque, New Mexico, June 23-26,
1991
USEPA 1980 An approach to water
resources evaluation of nonpoint
silvicultural sources (A procedural
handbook). U S Environmental Protection
Agency, Environmental Reserach
Laboratory, Athens, GA EPA 600/8-
80/012
USEPA 1988 WASP4, A hydrodynamic
and water quality model - Model theory,
user's manual, and programmer's guide
U S Environmental Protection Agency,
Environmental Research Laboratory,
Athens, Georgia EPA 600/3 87/039
USEPA 1991 a Workshop on the Water
Quality-based Approach for Point Source
and Nonpoint Source Controls. U S
Environmental Protection Agency EPA
503/9-92-001
USEPA 1991b Guidance for water
quality-based decisions: The TMDL
process U S Environmental Protection
Agency EPA 440/4-91-001
USEPA 1992a Technical guidance
manual for performing waste load
allocations - Book II Streams and rivers
(Draft) U S Evironmental Protection
Agency, Washington, DC
USEPA 1992b Simplified Method
Program - Toxics (SMPTOX), User's
manual U S Environmental Protection
Agency, Monitoring and Data Support
Division, Washington, DC
43
-------
Compendium of Watershed-Scale Models for TMDL Development
Walker, J. F., S. A Pickard, and W C
Sonzogni 1989. Spreadsheet watershed
modeling for nonpomt-source pollution
management in a Wisconsin basin Water
Resources Bulletin 25(1).139-147
Water Pollution Control Federation. 1989
Combined sewer overflow abatement:
Manual of practice no. FD-17.
Young, R A , C A. Onstad, D D Bosch, and
W.P. Anderson 1986 Agricultural
nonpoint source pollution model: A
watershed analysis tool U S Department
of Agriculture, Agricultural Research
Service, Morris, MN
Young, R A , C A Onstad, D D Bosch, and
WP Anderson 1989 AGNPS A
nonpomt-source pollution model for
evaluating agriculture watersheds Journal
of Soil and Water Conservation 44 168-
173
44
-------
APPENDIX
WATERSHED-SCALE MODEL - FACT SHEETS
-------
-------
EPA Screening Procedures
1 Distributor
Center for Exposure Assessment
Modeling
U S Environmental Protection Agency
Environmental Research Laboratory
Athens, Georgia 30613
(404) 546-3123
2 Type of Modeling:
• Not a computer program, consists of
a series of equations/techniques
• Pollutant loadings
• Multiple diffuse source
• Annual time steps, but may be used
for storm events
• Screening application
3. Model Components
• Loading functions for various land
uses/land activities
• Irrigation return flows
• Atmospheric inputs
• Pesticides in runoff and sediment
4 Method/Techniques
Loading functions estimate pollutant
loadings for screening applications and
comparison purposes A generic loading
function is of the following form
= a
X0'Cs}' fir,
where
Le, =
Xe
Cs,
Er, =
pollutant load in sediment
sediment yield
concentration of pollutant i in
soil
enrichment factor for pollutant i
a = dimensional constant
Additional pollutant load from rainfall may be
accounted for using the following
formulation
where
Lr, =
Qr =
QP =
Cn =
b =
pollutant load to streams from
rainfall
overland flow
rainfall amount
concentration of pollutant in rainfall
attenuation factor
Runoff can be estimated from the SCS curve
number equation and sediment yield can be
calculated from the USLE Soil concentra-
tion and enrichment factors can be
determined from field measurement or from
available default values Several variations
of these forms of loading function can be
used to represent other pollutant and
pollution sources
5 Applications-
Loading functions have been incorporated
into several hydrologic models to estimate
pollutant loadings They are also widely
used for screening purposes using desktop
calculators They can be used to evaluate
atmospheric inputs, urban, and agricultural
sources including irrigation return flows,
mining activities, and feedlots
A 1
-------
6. Number of Pollutants:
11 Simulation Output.
Nitrogen, phosphorus, sediment, heavy
metals, pesticides, organics, salinity
7. Limitations:
• Accuracy is limited when default
parameters are substituted for site
specific data
• Neglects seasonal variation
• Can be adapted to predict event or
seasonal loadings
• Does not evaluate control practices
except through assumption of a
constant removal rate or changes in
USLE parameters
• Application can be tedious and time-
consuming for basins with complex
multiple land uses and/or pollution
sources
8. Experience.
EPA Screening Procedures have been
applied (Donigian and Huber, 1991) to
the Sandusky River in Northern Ohio and
the Patuxent, Ware, Chester, and
Occoquan basins in the Chesapeake Bay
region (Davis et al., 1981, Dean et al ,
1981).
9. Updating Version:
N/A
10. Input Data Requirements:
• Pollutant concentrations in soils
• Enrichment ratios
• Nutrient concentrations in
precipitation
• Parameters in the USLE
• Land use data
• Annual pollutant loads (may be adopted
for seasonal or storm events)
• Nitrogen inputs to groundwater
• Salinity in irrigation return flows
12. References Available.
Davis, M J , M K Snyder, and J W Nebgen
1981 River basin validation of the water
quality assessment methodology for
screening nondesignated 208 areas -
Volume I: Nonpoint source load estimation
U S Environmental Protection Agency,
Athens, GA
Dean, J D , B Hudson, and W B Mills
1981 River basin validation of the MRI
nonpoint calculator and Tetra Tech's
nondesignated 208 screening
methodologies. Vol. II. Chesapeake-
Sandusky nondesignated 208 screening
methodology demonstration U S
Environmental Protection Agency, Athens,
Georgia
McElroy, A D , S W Chiu, J W Nabgen, A
Aleti, and R W Bennett 1976 Loading
functions for assessment of water pollution
for non-point sources. U S Environmental
Protection Agency, Washington, DC EPA
600/2-76/151 (NTIS PB-253325)
Mills W B , B B Borcella, M J Ungs, S A
Ghermi, K V Summers, Mok Lmgsung, G L
Rupp, G L Bowie, and D A Haith 1985
Water quality assessment: A screening
procedure for toxic and conventional
pollutants in surface and ground water, Part
1 US Environmental Protection Agency,
Environmental Reaserch Laboratory, Athens,
GA EPA/600/6-85/002a
A2
-------
The Simple Method
1 Distributor:
Metropolitan Washington Council of
Governments (MW-COG)
777 North Capitol St , Suite 300
Washington, DC 20002
(202) 962-3200
2 Type of Modeling:
• Not a computer program
• Pollutant concentration from urban
drainage areas
• Diffuse source
• Storm-based computations
• Screening application
3. Model Components-
• Pollutant export in storm runoff
• Sediment event mean concentration
estimates
• Threshold exceedance frequencies
4 Method/Techniques:
The Simple Method uses the following
expression as its governing equation
where
L, = pollutant loading (Ibs/year)
P = average annual rainfall (inches)
P, = unitless correction factor to account
for storms that produce no runoff
Rv = runoff coefficient (dimensionless)
C = flow-weighted mean pollutant
concentration (mg/L)
A =area of development (acres)
Runoff is estimated using runoff coefficients
for the fraction of rainfall converted to
runoff The portion of storms that do not
produce runoff are accounted for by a
correction factor determined based on
analyis of site specific or regional
precipitation pattern (p = 0 9 for
Washington, DC area). Runoff coefficients
are determined based on the following
equation
Rv = 0.05 + 0.009 • PI
where
PI = percent imperviousness
Pollutant concentrations in runoff depend on
the land use/land activity and can be
obtained from sampling programs such as
the NURP program Sediment event mean
concentrations are calculated as a function
of the surface area of the drainage basin It
is assumed that the channels in urban
watersheds are a major source of sediment
and thus larger watersheds will have higher
event mean concentrations Factors such as
the channel stability, storage, and stream
velocity are taken into account in the event
mean concentration determination
5. Applications.
• Estimate increased pollutant loading from
an uncontrolled development site
• Estimate expected extreme concentration
that occurs over a specified interval of
time
A3
-------
6. Number of Pollutants
Phosphorus, nitrogen, COD, BOD,
metals, including zinc, copper, and lead
7. Limitations:
• Limited to watersheds where data are
available or must assume national
NURP values
• It is intended for recently stabilized
suburban watersheds
• It is limited to small watersheds (less
than 1 square mile)
• Application limited to relative
comparisons
8. Experience:
The Simple Method is used to evaluate
development plans in the metropolitan
Washington, D.C area
9. Updating Version:
N/A
10. Input Data Requirements:
• Characteristics of pollution sources
• Flow and concentrations of point
sources
• Areas served by urban land uses such
as storm sewers, combined sewers,
and unsewered areas along with their
corresponding unit area loads for the
pollutant of concern
• Areas and unit area loads for grass and
woodland areas
• Parameters for the USLE for croplands
• Pollutant delivery ratios and pollutant
reduction efficiency ratio
• Treatment schemes and associated costs
11 Simulation Output-
• Total annual loads and load reductions
achieved by controls for the site or
watershed
• Program costs and cost per unit load
removed
12 References Available
Northern Virginia Planning District
Commission 1981 Comparison of nonpoint
pollution loadings from suburban and
downtown central business districts
Annandale, VA
Northern Virginia Planning District
Commission 1990 Analysis of the
recommended guidance calculation
procedure for the Chesapeake Bay
Preservation Act. Draft report, Northern
Virginia Planning District Commission.
Annandale, VA
Schueler, T R 1987 Controlling urban
runoff- A practical manual for planning and
designing urban BMPs Metropolitan
Washington Council of Governments
Document No 87703, Washington, DC
A.4
-------
USGS Regression Method
1 Name of Distributor.
Gary D Tasker
U S Geological Survey
430 National Center
Reston, VA 22092
2 Type of Modeling:
• Not a computer program
• Pollutant concentration from
urbanized watersheds
• Statistical approach
• Annual, seasonal, or storm event
mean pollutant loads
• Screening applications
3 Model Components
• Regression equations for mean storm
event pollutant load estimation
• Confidence interval around the mean
4. Method/Techniques
Regression equations were developed
from historical records of storm loads for
10 pollutants at 76 gaging stations in 20
States Ten explanatory parameters
were used to reflect possible site
variability associated with pollutant
processes The nonuniformity of the
variance required a generalized least
squares analysis The general form of
the regression model is as follows
W =
clA
eMJT
„ ~
where
W =
DA =
IA =
MAR =
Xa =
BCF =
mean storm event pollutant load
watershed drainage area
impervious area
mean annual rainfall
indicator variable
bias correction factor
a,b,c,d,e,f = regression coefficients
The mean annual pollutant load can then be
calculated by multiplying W by the mean
annual number of storm events
5 Applications.
• Estimation of average mean annual storm
event loads when data are severely
limited
• Comparing different locations
6. Number of Pollutants-
Chemical oxygen demand, suspended solids,
dissolved solids, total nitrogen, total
ammonia-nitrogen (NHa-N), total
phosphorus, dissolved phosphorus, total
copper, total lead, and total zinc
7. Limitations
• Valid only for areas for which regression
coefficients are provided, i e , regional
transferabihty is severely limited
• Valid only within the range of observed
values of pollutant loads and explanatory
variables
• Tends to underestimate the contributions
of snowmelt or extreme events
• Does not address causation
• Applies only to small watersheds
A 5
-------
8. Experience: 11. Simulation Output.
Limited • Average annual storm event load and
confidence interval
9. Updating Version:
12 References Available:
N/A
Tasker, G D , and N E Driver 1988
10. Input Data Requirements: Nationwide regression models for predicting
urban runoff water quality at unmonitored
• Drainage areas sites Water Resources Bulletin 24(5) 1091-
• Percent imperviousness 1101
• Mean annual rainfall
• Land use indicator
• Mean minimum January temperature
• Mean annual number of storm events
A6
-------
Sediment and Phosphorus Prediction (SLOSS, PHOSPH)
1. Name of Distributor.
N/A (see reference below)
2 Type of Modeling
• Not a computer program
• Sediment yield and phosphorus
loading from a watershed
• Annual simulation (may be adopted to
storm events)
• It is used in combination with GIS
capabilities
• Screening application
3. Model Components
• Two simple models for sediments and
phosphorus
4 Method/Techniques
SLOSS uses the Universal Soil Loss
Equation (USLE) to predict erosion and a
sediment delivery ratio is used to
estimate the sediment yreld Phosphorus
loading is calculated as the product of
the average phosphorus content of the
surface soil and a phosphorus enrichment
ratio The unique feature of this
approach is not the manner in which the
pollutant loads are calculated but the
ability to integrate simple algorithms with
the Virginia Geographic Information
System (VirGIS), which greatly facilitates
parameter input.
5 Applications:
• Identify critical areas of pollutant
production in watersheds
• Predict annual soil loss and phosphorus
yields
6. Number of Pollutants1
• SLOSS predicts erosion and sediment
yield
• PHOSPH predicts phosphorus loading
7 Limitations:
• Does not address seasonal variation
• Considers sediment and phosphorus only
• Requires access to GIS data
8 Experience:
Applied to Nommi Creek watershed in
Westmoreland County, Virginia
9. Updating Version:
N/A
10. Input Data Requirements
• Parameters for the USLE (soil erodibihty,
cropping and management factors,
topography, and rainfall erosivity factor)
and channel parameters
• Phosphorus concentration in soil,
phosphorus enrichment ratio
A7
-------
11. Simulation Output:
• Mean annual loads of sediment and
phosphorus
12. References Available*
Tim, U.S , S. Mostaghimi, V O
Shanholtz, and N Zhang 1991
Identification of critical nonpomt pollution
source area using geographic information
systems and simulation modeling In
Proceedings of The American Society of
Agricultural Engineers (ASAE)
International Summer Meeting,
Albuquerque, New Mexico, June 23-26,
1991.
A.8
-------
Water Screen
1 Distributor:
Office of Planning and Zoning
Anne Arundel County
Annapolis, MD
2 Type of Modeling-
• Pollutant loading functions
• Screening application
3 Model Components
• Apple II computer program
• Modified USLE erosion and sediment
yield
• Loading functions for nutrient and
organic loadings
4 Method/Techniques
Loading factors are calculated for
watersheds with sufficient data and
applied to similar watersheds Sediment
yield is estimated using the modified
USLE and a delivery ratio The specific
forms of the loading functions differ for
nitrogen and phosphorus Phosphorus
inputs are based on concentrations in
eroded material and an enrichment ratio
Nitrogen inputs include those associated
with precipitation The loading function
formulations are similar to those used in
the EPA Screening Procedures (McElroy
et al , 1976, and Zison et al , 1977)
5 Applications.
• Estimation of pollutant loadings
• Preplanning and screening applications
6. Number of Pollutants:
Phosphorus, nitrogen, metals, and organics
7 Limitations
• Cannot assess seasonal variability
• Limited to watersheds where data are
available to calculate loading functions
• Limited documentation and applications
8 Experience
Water Screen has been applied on the
Church Creek watershed south of Annapolis,
Maryland, where it showed discrepancy
between predictions of the MUSLE and the
loading functions Either the loading
functions for nitrogen and phosphorus or the
factors used to estimate nitrogen and
phosphorus concentrations in erosion
predicted by the MUSLE were inappropriate
for this watershed
9 Updating Version.
N/A
10 Input Data Requirements
• Parameters for the MUSLE including the
rainfall factor (R) for all subwatersheds
and land uses
• Sediment delivery ratios
A 9
-------
• Nitrogen, phosphorus, and organic
matter contents of eroded soils and
enrichment ratios
• Atmospheric depositions
• Loading factors as a function of land
use
11. Simulation Output
• Sediment, total nitrogen, total
phosphorus, and organic (BOD and
COD) yield for each land use
12. References Available:
Bird, B.L , and K M Conaway 1985
WATER SCREEN - A microcomputer
program for estimating nutrient and
pollutant loadings In Proceedings of the
Stormwater and Water Quality Model Users
Group Meeting, April 12-13, 1984 ed by
TOBarnwell pp 121-174 EPA-600/9-85-
0013
McElroy, A D , D S Chiu, J W Nebgen, A
Aleti, and F W Bennett 1976 Loading
functions for assessment of water pollution
from nonpoint sources. U S Environmental
Protection Agency, Washington, DC EPA-
600/2-76-151
Zison, S W , K F Haven, and W.B Mills
1977 Water quality assessment - A
screening method for nondesignated 208
areas U S Environmental Protection
Agency, Athens, GA EPA600/9-77-023
A.10
-------
WATERSHED
1. Distributor.
John F Walker
U S Geological Survey
6417 Normandy Lane
Madison, Wl 53719-1133
(608) 274-3535
2. Type of Modeling:
• Various multiple point sources plus
continuous and diffuse source/release
• Screening application
3 Mode! Components-
• Program is divided into seven
worksheets. The first summarizes
basic watershed characteristics
The next three worksheets estimate
pollutant loads from point sources and
cropland and non-cropland agricultural
land uses for controlled and
uncontrolled conditions. Sources are
totaled for controlled and uncontrolled
conditions by worksheet 5
• Program costs and cost-effectiveness
per unit load reduction are also
calculated
4 Method/Techniques
Separate methods are used to calculate
urban, rural non-cropland, and rural
cropland loads Urban loads are
calculated from point estimates of flow
and concentration, rural non-cropland
loads are estimated on a unit area basis,
and rural cropland loads are based on the
Universal Soil Loss Equation (USLE) The
rainfall factor (R) in the USLE is
unspecified for use as a calibration
parameter Delivery ratios and trapping
efficiencies for tributary wetlands are used
to convert eroded sediment to sediment
delivered These values are also calibrated
The model used the sorting features of the
EXCEL® spreadsheet program for the
Macintosh computer to rank the most cost-
effective alternatives
5 Applications:
• Phosphorus loading from point sources,
CSOs, septic tanks, rural cropland, and
non-cropland rural sources was estimated
for Delavan Lake watershed in
Wisconsin
• Evaluation of the trade-offs between
control of point and nonpomt sources
6 Number of Pollutants:
Used for only one at a time, e g phosphorus
7 Limitations:
• Cannot assess seasonal variability
• Can assess only a limited number of land
management control practices
• Requires calibration to determine the
rainfall factor and the sediment delivery
ratio
• Can assess only contaminants associated
with soils and sediments
8. Experience:
Watershed was applied to the study of point
and nonpomt sources in the Delavan Lake
watershed in Wisconsin It was determined
that runoff controls would be insufficient to
A 11
-------
meet water quality standards Instead of
focusing controls for phosphorus on
nonpomt sources, the study
recommended several in-lake controls
9. Updating Version
N/A
10. Input Data Requirements:
• Sources of pollution along with their
respective position and point of entry
to the basin
• Flows and concentrations of point
sources
• Areas served by urban land uses such
as storm sewers, combined sewers,
and unsewered areas along with their
corresponding unit area loads for the
pollutant of concern
• Areas and unit area loads for grass
and woodland areas
• Parameters for the USLE for croplands
• Pollutant delivery ratios and pollutant
reduction efficiency ratio
• Treatment schemes and associated costs
11. Simulation Output.
• Total annual loads and load reductions
achieved by controls for the site or
watershed
• Program costs and cost per unit load
removed
12 References Available-
Walker, J F , S A Pickard, and W C
Sonzogni 1989 Spreadsheet watershed
modeling for nonpomt-source pollution
management in a Wisconsin basin Water
Resources Bulletin 25(1) 139-147
A 12
-------
FHWA: The Federal Highway Administration Model
1. Distributor.
Office of Engineering and Highway
Operations R&D
Federal Highway Administration
6300 Georgetown Pike
McLean, VA 22101
2 Type of Modeling
- Statistical
• Screening application
3 Model Components
• Computation of water quality impact
from site data for either lakes or
streams
• Simple evaluation of controls
4 Method/Techniques.
Pollutant loadings and the variability of
loadings are estimated from runoff
volume distributions and event mean
concentrations for the median runoff
event at a site. Rainfall is converted to
runoff using a runoff coefficient
calculated from the percent
imperviousness Runoff velocity is
estimated from runoff intensity. Mean
runoff concentrations are calculated from
site median pollutant concentrations,
coefficient of variation for event mean
concentrations (EMCs), and the mean
EMC as
MCR = TCR • Vd+CVC/?2)
where
MCR= mean EMC for site (mg/L)
TCR= site median pollutant concentration
(mg/L)
CVCR = coefficient of variation of EMCs
Mean event mass loading is computed as
M(Mass) = MCR-MVR- (62 45 • 10'6}
where
M(Mass) = mean pollutant mass loading
(pounds per event)
MCR= mean runoff concentration (mg/L)
MVR = mean storm event runoff volume
(cf)
Annual loads are calculated by multiplying
by the number of storms per year Pollutant
build-up is based on traffic volumes and
surrounding area characteristics
5 Applications
• Evaluation of lake and stream impacts of
highway stormwater discharges
• Uncertainty analysis of runoff and
pollutant concentrations, or loads
• Highway stormwater runoff management
6 Number of Pollutants.
Heavy metals (copper, lead, and zinc),
nitrogen, and phosphorus
7 Limitations:
• Assesses seasonal variability in a limited
manner as expressed in the probability
distributions of the output
• Limited in its evaluation of controls
A 13
-------
• Does not consider the soluble fraction
of pollutants or the precipitation and
settling of phosphorus in lakes
8. Experience:
The FHWA model was used by the
Federal Highway Administration to
evaluate the impacts of stormwater
runoff from highways and their
surrounding drainage areas
9. Updating Version:
N/A
10. Input Data Requirements-
• Hourly rainfall data to be transformed
into mean and coefficient of variation
• Drainage and paved areas, average
rainfall volumes, intensities, and
durations
• Coefficients of variation are required
for all average rainfall characteristics
• Traffic volumes for the surrounding
area are required
• Runoff concentrations (average and
coefficient of variations) for each
pollutant
11 Simulation Output
• Mean and variance of m-stream or lake
concentrations
• Mean and variance of pollutant loadings
and concentrations in runoff
12 References Available
Dnscoll, E D , P E Shelley, and E W
Strecker 1990 Pollutant loadings and
impacts from highway stormwater runoff.
Volume I. Design procedure Prepared for
the Office of Engineering and Highway
Operations R&D, Federal Highway
Administration
Dnscoll, E D , P E Shelley and E W
Strecker 1990 Pollutant loadings and
impacts from highway stormwater runoff.
Volume II. Users guide for interactive
computer implementation of design
procedure Prepared for the Office of
Engineering and Highway Operations R&D,
Federal Highway Administration
A.14
-------
I/I/MM/ Watershed Management Model
1 Distributor:
Prepared by Camp Dresser & McKee Inc
for Stormwater Management Division,
Florida Department of Environmental
Regulation
Twin Towers Office Building
2600 Blair Stone Road
Tallahassee, Florida 32301-8241
(904) 488-6221
2. Type of Modeling:
• Watershed Stormwater pollutant loads
• Multiple diffuse source release
• Annual time steps
• Screening application
3. Model Components
• Computation of annual nutrient and
metal loads to reservoirs
• Computation of m-lake or m-stream
water quality from pollutant loads
• Load reduction estimates for site or
regional BMP implementation
• Uptake and removal in stream courses
• Estimates of annual pollutant loads
from baseflow
• Comparison with point sources
• Failing septic tank loads
• Chlorophyll-a and nutrient
concentrations in downstream lakes
and reservoirs
4. Method/Techniques.
Runoff coefficients are used for rural
areas, for urban areas runoff is based on
a linear function of the percent
imperviousness Loading of nutrients and
metals is based on event mean
concentrations measured locally or from
NURP data Baseflow is estimated from
flow records and concentrations There is a
choice of three lake water quality routines
that output mean annual concentrations of
chlorophyll-a. (The model can be adapted to
predict seasonal loads or chlorophyll-a
concentrations provided that seasonal event
mean concentration data are available )
Simple calculations are included for m-
stream transport and transformation based
on travel time The program can assess the
relative contributions of point and nonpomt
sources Resultant water quality is
predicted with a version of the Vollenweider
eutrophication model, adapted to lakes in
the southeastern United States Removal of
metals associated with sediments in
reservoirs is estimated from the sediment-
trapping efficiency of the reservoir
5 Applications:
Estimates the annual nonpomt source loads,
including baseflow and precipitation inputs,
for management planning
6. Number of Pollutants
Total phosphorus, total nitrogen, lead, and
zinc
7. Limitations
• Accuracy is limited when default
parameters are substituted for site-
specific data
• Neglects seasonal variation
• Does not predict sediment yields
A 15
-------
• Does not evaluate control practices
except through assumption of a
constant removal fraction
• Does not consider loadings associated
with snowmelt events
• Can assess only relative impacts of
land use categories or controls
8. Experience:
The model has been applied to between
10 and 15 watersheds It has been used
as part of a wasteload allocation study
for Lake Tohopekaliga and for
Jacksonville, Florida, watershed's Master
Plan. It has been applied in Norfolk
County, Virginia, and to a Watershed
Management Plan for North Carolina
9. Updating Version
Under development
10. Input Data Requirements
• Land use and soil types
• Average annual precipitation,
evaporation, and evapotranspiration
• Nutrient concentrations in
precipitation
• Annual baseflow and baseflow pollutant
concentrations
• Event mean concentrations in runoff
• Reservoir, lake, or stream hydraulic
characteristics
• Removal efficiencies of proposed BMPs
11 Simulation Output
• Annual pollutant loads from point and
nonpomt sources, including both
agricultural and urban land use
• Relative magnitude of inputs from point
sources and septic tanks
• Load reductions from combined effects
of multiple BMPs
• In-lake nutrient concentrations as related
to trophic state, also, concentrations of
metals are evaluated for the reservoir
• Standard statistics and bar graphs of
results
12. References Available
Camp, Dresser and McKee (COM) 1992
Watershed Management Model user's
manual. Version 2 0 Prepared for the Florida
Department of Environmental Regulation,
Tallahassee, FL
A16
-------
NPSMAP: Nonpoint Pollution Source Model for
Analysis and Planning
1 Distributor
Dr Jack Douglas Smith
Project Manager
Water Resources/Quality
Fetrow Engineering, Inc
12300 S E Mallard Way, Suite 205
Milwaukee, OR 97222
(503) 652-1526
2 Type of Modeling
• Multiple land use watersheds
• Water quality analysis
• Continuous simulation (hourly data,
one year at a time)
• Point and nonpomt sources
• Wasteload and load allocations
• Screening application
3. Model Components:
• Runoff and pollutant loading
assessment
• Water quality and resources
management analysis
• Simulation of wet detention or
wetland system controls for each
subbasm or stream segment
• Irrigation and drainage
• Snowfall and snowmelt
• Uses Lotus 1-2-3® spreadsheet
4. Method/Techniques
NPSMAP is a spreadsheet-based program
that operates within the Lotus 1-2-3®
programming environment NPSMAP is a
dynamic simulation program that includes
three primary computation modules
(SIMULATE, ALLOCATE, and
PROBABILITY). SIMULATE computes daily
runoff, pollutant loadings, streamflow, and
water quality ALLOCATE simulates stream
segment load capacities (LCs), point source
wasteload allocations (WLAs), and nonpomt
source load allocations (LAs) PROBABILITY
computes the probability distributions of
runoff and loadings
5 Applications
• Nonpoint source runoff and pollutant
(nutrient) loadings, including surface
water storage in reservoirs and wetlands
• Point source discharges and streamflow
• Groundwater levels, irrigation and other
water uses, and water quality
6 Number of Pollutants*
Nitrogen and phosphorus
7 Limitations.
• No sediment is evaluated
• Limit on simulation period (one year at a
time)
8 Experience
Applied to the Tualatin River basin for
Oregon Department of Environmental Quality
9 Updating Version
Version 1 0 (1990)
A 17
-------
10. Input Data Requirements.
• Hydrology data stream segment
volumes and streamflow recession
coefficients, parameters for upland
groundwater, parameters and
exchange matrix for valley fill
groundwater model, and rainfall
intensity
• Land use data land use distributions
and parameter values in each stream
segment, soil parameters, and SCS
curve number and retention
coefficients
• Weather records including rainfall,
snowfall, temperature, and
evapotranspiration
11. Simulation Output
• Daily runoff and streamflow
• Nonpomt source runoff and loads
• Treatment plant discharges, loadings,
overflows/bypasses
• Groundwater, streamflow, and water
quality (loads and concentrations)
• Bar graphs, two- and three-dimensional
graphs are available for simulation results
• Statistical summary
• Probability distributions of runoff and
loadings
• Stream segment load capacities (LCs),
point source wasteload allocations
(WLAs), and nonpomt source load
allocations(LAs)
12 References Available
Omicron Associates 1990 Nonpomt
Pollution Source Model for Analysis and
Planning (NPSMAP) - Users manual
Omicron Associates, 11265 NW Rammont
Road, Portland, OR
A.18
-------
GWLF: Generalized Watershed Loading Functions
1 Distributor
Dr Douglas A Haith
Department of Agricultural and
Biological Engineering
Cornell University
Ithaca, NY 14853
(607) 255-2802
2 Type of Modeling
• Pollutant loads from urban and
agricultural watersheds
• Continuous simulation using daily time
step
• Point and nonpomt sources
• Screening to intermediate application
• Evaluation of effects of land use
changes
3 Model Components:
• Rainfall/runoff assessment
• Surface water/groundwater quality
analysis
4. Method/Techniques
This model is based on simple runoff,
sediment, and groundwater relationships
combined with empirical chemical
parameters. It evaluates streamflow,
nutrients, soil erosion, and sediment yield
values from complex watersheds Runoff
is calculated by means of the SCS curve
number equation The Universal Soil
Loss Equation (USLE) is applied to
simulate erosion Urban nutrient loads
are computed by exponential wash-off
functions. Groundwater runoff and
discharge are obtained from a lumped-
parameter watershed water balance for both
shallow saturated and unsaturated zones
Calibration is not required for water quality
data
5. Applications:
Relatively large watersheds with multiple
land uses and point sources
6. Number of Pollutants:
Total and dissolved nutrients (nitrogen and
phosphorus) and sediment
7. Limitations:
• Simulation of peak nutrient fluxes is
weak
• Stormwater storage and treatment are
not considered
8 Experience:
GWLF was validated for an 85,000-hectare
watershed from the West Branch Delaware
River Basin in New York using a 3-year
period of record
9. Updating Version
Under development
10. Input Data Requirements.
• Daily precipitation and temperature data
and runoff source areas
• Transport parameters' runoff curve
numbers, soil loss factor,
evapotranspiration cover coefficient,
erosion product, groundwater recession
A 19
-------
and seepage coefficients, and
sediment delivery ratio
• Chemical parameters urban nutrient
accumulation rates, dissolved nutrient
concentrations in runoff, and solid-
phase nutrient concentrations in
sediment
• Point sources
11. Simulation Output-
• Annual and seasonal runoff,
streamflow, watershed erosion, and
sediment yield
• Annual and seasonal total and
dissolved nitrogen and phophorus
loads in streamflow and groundwater
discharge to streamflow
• Annual erosion and total/dissolved
nitrogen and phosphorus loads from
each land use
• Annual and seasonal pollutant
loadings by land use type and
pollution sources
12. References Available
Brown, M P., M.R Rafferty, and P
Longabucco. 1985 Nonpoint source
control of phosphorus - A watershed
evaluation. Report to the U S
Environmental Protection Agency, Ada,
OK.
Delwiche, L L D and D A Haith. 1983
Loading functions for predicting nutrient
losses from complex watersheds Water
Resources Bulletin 19(6) 951-959
Haith, D A 1985 An event-based procedure
for estimating monthly sediment yields
Transactions of the American Society of
Agricultural Engineers 28(6) 1916-1920
Haith, D A and L L Shoemaker 1987
Generalized watershed loading functions for
stream flow nutrients Water Resources
Bulletin 23(3) 471-478
Haith, D A andLJ Tubbs 1981
Watershed loading functions for nonpomt
sources Proceedings of the American
Society of Civil Engineers Journal of the
Environmental Engineering Division
107(EE1) 121-137
Wu, S R , D A Haith, and J H Martin, Jr
1989 GWLF - Generalized Watershed
Loading Functions - User's manual.
Department of Agricultural Engineering,
Cornell University, Ithaca, NY
A.20
-------
P8-UCM: Urban Catchment Model
1 Distributor
Narragansett Bay Project
291 Promenade Street
Providence, Rl 02908-5767
(401) 521-4230
2. Type of Modeling
• Urban watersheds
• Storm event/sequence simulation
• Surface water quality analysis
• Evaluation of BMPs and development
of design criteria
• Single, continuous, and diffuse
source/release
• Screening application
3 Model Components:
• Rainfall, stormwater runoff
assessment
• Surface water quality analysis
• Routing through structural controls
4 Method/Techniques
The P8 program predicts the generation
and transport of stormwater runoff
pollutants in small urban catchments It
consists mainly of methods derived from
other tested urban runoff models (i e ,
SWMM, HSPF, D3RM, TR-20). Runoff
from impervious areas is calculated
directly from rainfall once depression
storage is exceeded Particle build-up
and wash-off processes are obtained
using equations derived primarily from
the SWMM program The SCS curve
number equation is used to predict runoff
from pervious areas Water balance
calculates percolation from the pervious
areas Baseflow is simulated by a linear
reservoir Without calibration, use of model
results should be limited to relative
comparisons
5 Applications
• Surface water quantity and quality
routing
• Small urban area assessments
• Watershed-scale land use planning
• Site planning and evaluation for
compliance
• Selecting and sizing BMPs
6. Number of Pollutants*
Ten pollutants, including total suspended
solids (TSS), total phosphorus, total Kjeldahl
nitrogen, lead, copper, zinc, and
hydrocarbons
7 Limitations.
• No snowfall, snowmelt, or erosion is
calculated
• Effects of variations in vegetation type/
cover on evapotranspiration are not
considered
• Watershed lag is not simulated
8 Experience:
N/A
9. Updating Version
Version 1 1 (1990)
A 21
-------
10. Input Data Requirements.
• Device (hydraulic) parameters for
pond, basin, buffer, pipe, splitter, and
aquifer
• Watershed parameters areas,
impervious fraction and depression
storage, street-sweeping frequency,
SCS runoff curve number for pervious
portion
• Particle parameters
accumulation/wash-off parameters,
runoff concentrations, street-sweeper
efficiencies, settling velocities, decay
rates, filtration efficiencies
• Water quality component parameters
pollutant concentrations
• Air temperatures required for stream
baseflow computations
11. Simulation Output:
• Water and mass balances, removal
efficiencies, mean inflow/outflow
concentrations, and statistical summaries
by device and component
• Comparison of flow, loads, and
concentration across devices
• Peak elevation and outflow ranges for
each device
• Sediment accumulation rates by device
• Violation frequencies for event mean
concentrations
12. References Available
Palmstrom, N , and W W Walker, Jr 1990
P8 Urban Catchment Model: User's guide,
program documentation, and evaluation of
existing models, design concepts, and Hunt-
Potowomut data inventory. The
Narragansett Bay Project Report No NBP-
90-50
A.22
-------
SIMPTM: Simplified Particle Transport Model
1. Distributor-
Roger C Sutherland
Software Distribution
OTAK, Inc
17355 Boones Ferry Road
Lake Oswego, OR 97035
(503) 635-3618
Cost: §495
2 Type of Modeling.
• Stormwater from urban watersheds
• Storm event/sequence simulation
• Continuous and diffuse source release
• Screening applications
• Evaluation of BMPs
3 Model Components.
• Rainfall/runoff assessment
• Surface water quality analysis
4 Method/Techniques
SIMPTM simulates the accumulation,
wash-off, and mechanical removal of up
to six different pollutants contained in
total solids or particulate matter
accumulating on paved areas that are
directly connected to storm dram
systems Runoff is calculated using
runoff thresholds and originates only
from paved areas directly connected to
storm dram systems SIMPTM converts
event trapezoidal hyetographs into
trapezoidal hydrographs using the runoff
duration and small storm impervious area
loss equations developed by Pitt (1987)
Street cleaning, catchbasm accumulation,
and on-site retention are expressed as a
function of particle size distribution
Simulation of runoff or pollutant loads from
pervious areas and baseflow is ignored
5 Applications:
• Simulation of runoff and sediment yields
from urban watersheds with small,
storm-based hydrology
• Effectiveness of various, primarily
nonstructural controls, such as street and
catchment cleaning
6. Number of Pollutants
Six pollutants including sediment
components, nitrogen, phosphorus, chemical
oxygen demand
7 Limitations:
• Snow accumulation and snowmelt are
ignored in the hydrology simulation
• Nutrient transformations are not
evaluated
• Runoff and pollutants are assumed to
originate solely from impervious areas
directly connected to the drainage
system
• Model testing in Pacific Northwest only
8 Experience.
The model was calibrated using National
Urban Runoff Program (NURP) data from the
Lake Hills basin in Bellevue, Washington
Subsequently, the model was tested using
NURP data for the Surrey Downs basin, also
in the Bellevue area The simulation of
variability of runoff events was excellent
considering the simplistic approach used to
A 23
-------
convert rainfall into runoff volumes The
predicted variability in pollutant loads
was less variable than was actually
observed.
Version 1 (1980) was applied to over
57,000 acres of urban and urbanizing
land surrounding Reno, Nevada, for the
Washoe County Council of Governments
for the purpose of comprehensive
stormwater management
9. Updating Version:
Version 2 1 November 1990
10. Input Data Requirements:
• Rainfall depths for 1-, 3-, or 6-hour
periods
• Pollutant strengths and accumulation
and wash-off parameters
• Particle size distributions
• Curb lengths and characterization of
impervious surfaces
11. Simulation Output-
• Runoff volumes, durations, pollutant
loadings, and event mean
concentrations for each rainfall event
• Results may be reported for each
subcatchment or for the entire
watershed
• Although graphic output is not available,
results can be imported to spreadsheets,
such as Lotus 1-2-3®
• Average monthly and annual statistics
are calculated for rainfall depths,
durations, and intensities
12 References Available:
Pitt, R E 1987 Small storm urban flow and
particulate wash-off contributions to outfall
discharges Ph D Dissertation, Civil and
Environmental Engineering Department,
University of Wisconsin, Madison, November
1987
Sutherland, R C , D L Green, and S L Jelen
1990 Simplified Particulate Transport
Model, Users manual OTAK, Inc Lake
Oswego, Oregon
Sutherland, R C 1991 Modeling of urban
runoff quality in Bellevue, Washington using
SIMPTM In Proceedings from the technical
sessions of the regional conference on
nonpoint source pollution: The unfinished
agenda for the protection of our water
quality, March 20-21, 1991, Tacoma, WA
Published by the State of Washington Water
Research Center
A.24
-------
Automated Q-ILLUDAS (AUTO-QI)
1 Distributor.
Michael L Terstnep
Illinois State Water Survey
2204 Griffith Drive
Champaign, Illinois 61820-7495
Cost $50
2 Type of Modeling
• Urban stormwater processes
• Storm event simulation
• Continuous and diffuse pollutant
sources
• Screening and intermediate
applications
• Evaluation of BMPs
3 Model Components
• Rainfall/runoff assessment from
pervious and impervious areas
• Water quality analysis (emphasis on
nutrients and sediments)
• Simulation of BMPs, separate or
overlapping
• Linkage to geographic information
system (GIS)
4 Method/Techniques
AUTO-QI is based on continuous
simulation of soil moisture Runoff
volumes are adjusted for soil moisture,
pervious and impervious depression
storage, interception, and infiltration
based on Morton infiltration curves
Exponential pollutant accumulation and
wash-off functions are used to determine
the pollutant loads The impacts of a
series of pollutant reduction practices are
simulated based on user-supplied removal
efficiencies
5 Applications-
• Simulation of runoff volumes, pollutant
loads, and event mean concentrations
• Comparison of pollutant levels with and
without BMPs and with various fertilizer
application rates
6 Number of Pollutants.
Several pollutants including nitrogen,
phosphorus, chemical oxygen demand
(COD), metals, and bacteria (at least three at
once)
7 Limitations-
• Does not calculate pollutant removal
efficiencies, removal efficiencies must be
supplied by the user
• Lacks nutrient transformation and
mstream processes
• Tested in the State of Illinois only
• No simulation of subsurface soil
processes
8. Experience
Simulation of urban pollutant loads for
suspended solids, phosphorus, and lead
from the greater Lake Calumet area after
calibration on Boneyard Creek in Champaign,
Illinois (1990)
A 25
-------
9. Updating Version:
October 1990
10. Input Data Requirements-
• Daily and hourly rainfall data
• Monthly evaporation and
evapotranspiration values
• BMP removal efficiencies
• Soil infiltration parameters
• Land use parameters and soil types
for each subcatchment
• Build-up and wash-off characteristics
of each pollutant
11. Simulation Output:
• A summary for the watershed by
event is created for rainfall, runoff,
and runoff duration
• Event mean concentrations and
loadings
12. References Available.
Terstnep, M. L , M T Lee, E P Mills, A V
Greene, and M R. Rahman 1990
Simulation of urban runoff and pollutant
loading from the Greater Lake Calumet area
Prepared by the Illinois State Water Survey
for the U S Environmental Protection
Agency, Region V, Water Division,
Watershed Management Unit, Chicago, IL
A 26
-------
AGNPS: Agricultural Nonpoint Source Pollution Model
1 Distributor
Basil Meyer
North Central Soil
Conservation Research Laboratory
U S Dept of Agriculture
Agricultural Research Service
Morris, MN 56267
(612) 589-3411
2 Type of Modeling
• Simulation of pollutant loads from
agricultural watersheds
• Storm-event simulation
• Point source/release
• Distributed modeling using a grid
system with square elements
• Screening, intermediate, and detailed
applications
• Evaluation of BMPs
3 Model Components
• Rainfall/runoff assessment
• Water quality analysis (emphasis on
nutrients and sediments)
• Point source inputs available (feedlots,
springs, wastewater treatment plant
discharge, bank and gully erosion)
• Unsaturated/saturated zone routines
• Economic analysis
• Linkage to CIS possible
4 Method/Techniques
This model can identify critical areas of
sediment and nutrient production in a
watershed and can assess the impacts of
best management practices (BMPs). Soil
erosion is simulated by the modified
Universal Soil Loss Equation (USLE) The
unit hydrograph approach is used to predict
water flow, while the SCS curve number
equation is applied to estimate runoff
volume Some versions are linked to GIS
and DEM with automatic generation of
terrain parameters (Panuska et al , 1991)
5 Applications
• Erosion, sediment, and chemical
transport
• Surface water flow routing
6 Number of Pollutants
Four pollutants nitrogen, phosphorus,
chemical oxygen demand (COD), and
sediment
7 Limitations
• Only single event version is currently
available, although a continuous
simulation version (ANNAGNPS) should
be released soon
• Does not handle pesticides
• Lacks nutrient transformation and
mstream processes
• Needs further field testing for pollutant
transport component
• No simulation of subsurface soil
processes
• Rainfall intensity is not considered in the
runoff analysis
8. Experience.
• Economic assessment of soil erosion and
water quality in Idaho (1987)
A 27
-------
• Economic effect of nonpomt pollution
control alternatives (1988)
• Analysis of agricultural nonpomt
pollution control options in St Albans
Bay, Vermont (1987)
• Water quality evaluation of Garvm
Brook watershed (1989)
• Alternative management practices in
Salmonson Creek watershed (1989)
• Applied along with a GIS to the Owl
Run watershed, which is part of the
Chesapeake Bay watershed, to predict
the effectiveness of BMP installation
9. Updating Version
Single event AGNPS Version 3 65
Soon to be released are AGNPS Version
4 0 (written in C) and continuous
simulation AGNPS (ANNAGNPS)
10. Input Data Requirements
• Topography and soil characteristics
• Meteorologic data
. Land use data (cropping history and
nutrient applications)
• Point source data
11. Simulation Output
• Hydrology output storm runoff
volume and peak rate
• Sediment output sediment yield,
concentration, particle size
distribution, upland erosion, amount
of deposition
. Chemical output pollutant
concentration and load
12. References Available
Frevert, K and B M Crowder 1987
Analysis of agriculture nonpoint pollution
control options in the St. Albans Bay
watershed Economic Division, ERS, USDA,
Staff Report No AGES870423
Hewitt, M J 1991 GIS for nonpoint source
watershed modeling applications In EPA
Workshop on the Water Quality-based
Approach for Point Source and Nonpoint
Source Controls, June 1991, p 34
EPA503/9-92-001
Kozloff, K , S J Taff, and W Wang 1992
Microtargetmg the acquistion of cropping
rights to reduce nonpoint source water
po 11 utio n Wa ter Resources Research
28(3) 623-628
Lee, M T 1987 Verification and
applications of a nonpoint source pollution
model In Proceedings of the National
Engineering Hydrology Symposium, ASCE,
New York, NY
Panuska, J C , I D Moore, and L A Kramer
1991 Terrain analysis Integration into the
agricultural nonpoint source (AGNPS)
pollution model Journal of Soil and Water
Conservation 46(1) 59-64
Prato, T , H Shi, R Rhew, and M Brusven
1989 Soil erosion and nonpomt-source
pollution control in an Idaho watershed
Journal of Soil Water Conservation
44(4) 323-328
Setia, P P , R S Magleby, R S , and D G
Carvey 1988 Illinois rural clean water
project - An economic analysis Resources
and Technology Division, ERS, USDA Staff
Report No AGES830617
Young, R A , C A Onstad, D D Bosch, and
W P Anderson 1986 Agricultural Nonpoint
Source Pollution Model A watershed
analysis tool Agriculture Research Service,
U S Department of Agriculture, Morris, MN
A 28
-------
SLAMM: Source Loading and Management Model
1 Distributor
Dr Robert Pitt
Department of Civil Engineering
University Station
Birmingham, AL
(205) 934-8430
2 Type of Modeling
• Continuous and diffuse source/release
• Continuous series of storm events (up
to 150)
• Screening application
• Evaluation of controls
3 Model Components
• Rainfall/runoff assessment
• Water quality analysis
4 Method/Techniques.
This program can evaluate the effects of
a number of different stormwater control
practices on runoff routines SLAMM
performs continuous mass balances for
particulate and dissolved pollutants and
runoff volumes Runoff is calculated by
a method developed by Pitt (1987) for
small storm hydrology Runoff is based
on infiltration minus initial abstraction
and is calculated for both pervious and
impervious areas Triangular
hydrography is used to simulate the
hydrology A statistical approach is used
to parametrize the hydrographs
Exponential build-up and wash-off
functions are used for pollutant loading,
which is assumed to come from
impervious areas Loading from pervious
areas is constant concentration supplied by
user Water and sediment from various
source areas is tracked by source area as it
is routed through various treatment devices
The program considers how particulates
filter or settle out in control devices
Particulate removal is calculated based on
the design characteristics of the basin or
other removal device Storage and overflow
of devices is also considered At the outfall
locations, the characteristics of the source
areas are used to determine pollutant loads
in solid and dissolved phases Loads from
various source areas are summed
5 Applications
• Evaluates multiple control strategies such
as wet detention basins, porous
pavement, infiltration devices, street
cleaning, catchment cleaning, grass
swales, roof runoff disconnections, and
paved parking lot disconnections
• Planning tool for urban runoff quality and
quantity assessments
• Applicable to the study of stormwater
pollutant control from regions frequently
receiving rainfall events of less than 1
inch
6. Number of Pollutants
Particulate and dissolved pollutants
(depending on the calibration information),
such as particulate and filterable forms of
residue, phosphorus, phosphate, total
Kjeldahl nitrogen, chemical oxygen demand
(COD), fecal cohform bacteria, aluminum,
copper, lead, and zinc
A 29
-------
7. Limitations:
• Does not evaluate snowmelt and
baseflow conditions
• Will not provide individual storm
predictions
• Evaluates runoff characteristics at the
source area within the watershed and
at the discharge outfall but does not
consider mstream processes that
remove or transform pollutants
• Does not evaluate receiving water
quality responses
• Requires between 10 and 100 storms
in a study period (maximum 150)
• Does not develop or evaluate specific
hydraulic designs
• Does not model erosion from pervious
areas or construction sites
8. Experience.
SLAMM has been used in conjunction
with receiving water quality models
(HSPF) to examine the ultimate effects
on urban runoff from Toronto for the
Ontario Ministry of the Environment
SLAMM was also used to evaluate
control options for controlling urban
runoff in Madison, Wisconsin, using CIS
information The State of Wisconsin
uses SLAMM as part of its Priority
Watershed Program It was used in
Portland, Oregon, for a study evaluating
CSOs.
9. Updating Version:
Currently being updated
10. Input Data Requirements:
• Rainfall start and end dates (and
times) and rainfall depths
• Areas of each source type, effective
SCS soil type
• Building and traffic density
• Pavement texture, roof pitch, and
presence of alleys
• Land use
• Shape, size, and type of outlet structures
of the wet detention basin
• Soil infiltration rates for infiltration
devices
11 Simulation Output-
• Source area and outfall flow volume
estimates for each rainfall period and land
use
• Source area and outfall particulate
residue mass discharge and
concentration estimates for each rainfall
period and land use
• Relative source area runoff volume and
particulate residue mass
contribution estimates for each rainfall
period
• Mass discharge, concentration, and
relative contribution estimates for each
pollutant selected
• Cost estimates of stormwater control
practices, graphical summaries, baseflow
predictions, and snowmelt predictions are
under development
12 References
Pitt, R 1979 Demonstration of non-point
pollution abatement through improved street
cleaning practices U S Environmental
Protection Agency, Cincinnati, OH PA-
600/2-79-161 (NTIS PB80-108988)
Pitt, R 1986 Runoff controls in Wisconsin's
priority watersheds In Urban runoff quality,
ed B Urbonas, and L A Roesner,
Proceedings of the Engineering Foundation
Conference on Urban Runoff Quality (ASCE),
June 22-27, 1986, in Henniker, NH.
A.30
-------
STORM: Storage, Treatment, Overflow, Runoff Model
1. Distributor:
U S Army Corps of Engineers
The Hydrologic Engineering Center (HEC)
609 Second Street
Davis, CA 95616
Cost. $200 - 9-track magnetic tape
2 Type of Modeling
• Urban runoff processes
• Continuous simulation (hourly time
steps)
• Continuous and diffuse source/release
• Screening application
3 Model Components-
• Rainfall/runoff assessment
• Water quality analysis
• Statistical and sensitivity analysis
4 Method/Techniques.
This is a quasi-dynamic program A
modified rational formula is used for
hydrology simulation Rainfall/runoff
depth and volumes are computed by
means of an area-weighted runoff
coefficient and the SCS curve number
equation, respectively The Universal
Soil Loss Equation (USLE) is applied to
simulate erosion Water quality is
simulated by linear build-up and first-
order exponential wash-off coefficients
Calibration is advisable, but relative
comparisons can be evaluated without
calibration
5. Applications:
• Storm and combined sewer overflows
including dry-weather flow
• Surface water quantity and quality
routing with storage/treatment option
• Urban areas assessments
6 Number of Pollutants
Six prespecified pollutants suspended
solids, settleable solids, BOD, total
cohforms, ortho-phosphate, and total
nitrogen
7. Limitations
• Little flexibility in parameters to calibrate
to observed hydrographs
• Lacks microcomputer version
• Requires a large amount of input data
8 Experience:
STORM was extensively used in the late
1970s and early 1980s. The model was
applied to the San Francisco master drainage
plan for abatement of combined sewer
overflows
9 Updating Version
Version 1 (1977)
10. Input Data Requirements:
• SCS, build-up, and wash-off parameters
• Runoff coefficient and soil type
A 31
-------
11. Simulation Output:
• Storm event summaries (runoff
volume, concentrations, and loads)
• Summaries of storage and treatment,
utilization, total overflow loads and
concentrations
• Hourly hydrographs and pollutographs
(concentration vs time)
• Statistical summaries on annual and
total simulation period basis
(percentage of runoff passing through
storage and the number of overflows)
12. References Available:
Abbott, J 1977 Guidelines for
calibration and application of STORM.
U.S. Army Corps of Engineers,
Hydrologic Engineering Center Davis,
CA Training Document No 8
Abbott, J 1978. Testing of several
runoff models on an urban watershed.
ASCE Urban Water Resources Research
Program. ASCE, New York, NY
Technical Memorandum No 34
Donigian, A S , Jr , and W C Huber 1991
Modeling of nonpoint source water quality m
urban and non-urban areas. U S
Environmental Protection Agency,
Environmental Research Laboratory, Athens,
Georgia EPA/600/3-91/039
Hydrologic Engineering Center. 1977
Storage, Treatment, Overflow, Runoff
Model, STORM, User's manual. Generalized
Computer Program 723-S8-L7520 U S
Army Corps of Engineers, Davis, CA
Najanan, T 0 , T T Griffin, and V K
Gunawardana 1986 Development impacts
on water quality A case study Journal of
Water Resources Planning and Management,
ASCE, 112(1) 20-35
Shubmski, R P , A J Knepp, and C R
Bristol 1977 Computer program
documentation for the continuous storm
runoff model SEM-STORM Report to the
Southeast Michigan Council of
Governments, Detroit, Ml
A 32
-------
ANSWERS: Area/ Nonpoint Source Watershed Environment
Response Simulation
1. Distributor
Dr David Beasley
Department of Agriculture Engineering
North Carolina State University
Raleigh, North Carolina
(919) 515-2694
2 Type of Modeling
• Simulation of agricultural watersheds
with emphasis on erosion and
sediment yield
• Distributed simulation using a grid
system
• Storm event simulation
• Single and diffuse source/release
• Screening and intermediate
applications
• Evaluation of BMPs
3 Model Components:
• Rainfall/runoff assessment
• Overland flow and channel flow
• Loading of nutrients and pesticides
• Erosion, sediment transport, and
deposition
4 Method/Techniques
This model simulates the effects of land
use, management, and conservation
practices on the quality and quantity of
water in a watershed The hydrology
component is based on surface and
subsurface water movement relationships
using a modified form of the Holton
infiltration model Erosion processes are
predicted by an event-based particle
detachment and transport model The
quality component was added to the model
to compute pollutant loadings based on
correlation relationships between
concentration, sediment yield, and runoff
volume Improvements to the pollutant
loading and transformation routines have
been incorporated by Dillaha et al (1988)
5 Applications
• Hydrologic and erosion response of
agriculture land and construction sites
• Movement of water m overland,
subsurface, and channel flow phases
• Identification of critical areas for erosion
and sedimentation control
• Siting and evaluation of BMPs
6 Number of Pollutants
Nutrients (phosphorus and nitrogen) and
sediment (Some versions include
pesticides )
7 Limitations:
• Mainframe computer required for large
watershed simulation
• Complexity of input data file
• Snowmelt processes and pesticide
modeling are not included
• No chemical transformation of nitrogen
and phosphorus
• Small time steps are necessary for finite
difference algorithms and restrict the
simulation to a single event
A 33
-------
• Requires small element grid, assumes
homogeneous condition within each
element
8. Experience:
Applied successfully in Indiana on
agricultural watersheds and construction
sites for best management practice
(BMP) evaluation Evaluated the relative
importance of point and nonpomt source
contributions to Sagmaw Bay
9. Updating Version:
N/A
10. Input Data Requirements
Detailed description of the watershed
topography, drainage network, soils, and
land use (available from USDA-SCS soil
surveys, land use, and cropping surveys)
11 Simulation Output:
• Alternative erosion control
management practices on an element
basis or entire watersheds (flow and
sediment)
• Limited graphical representation of
output results
12. References Available.
Amm-Sichani, S. 1982 Modeling of
phosphorus transport in surface runoff
from agricultural watersheds Ph D
Thesis, Purdue University, W Lafayette,
IN.
Beasley, D.B and L F Huggms 1981
ANSWERS User's Manual. U S
Environmental Protection Agency, Region
V Chicago, IL EPA905/9-82-001
Beasley, D B 1986 Distributed parameter
hydrologic and water quality modeling In
Agricultural Nonpoint Source Pollution:
Model Selection and Application, ed A
Giorgmi and F Zingales, pp 345-362
Diilaha, T A III, D B Beasley, and L F
Huggms 1982 Using the ANSWERS model
to estimate sediment yields on construction
sites Journal of Soil and Water
Conservation 37(2) 117-120
Diilaha, T A , C D Heatwole, M R Bennett,
S Mostaghimi, V O Shanholtz, and B B
Ross 1988 Water quality modeling for
nonpomt source pollution control planning
Nutrient transport. Virginia Polytechnic
Institute and State University, Dept of
Agricultural Engineering Report No SW-88-
02
Donigian, A S , Jr , and W C Huber 1991
Modeling of nonpomt source water quality in
urban and non-urban areas. Environmental
Protection Agency, Environmental Research
Laboratory, Athens, Georgia EPA/600/3-
91/039
Freedmann, PL and D W Dilks 1991
Model capabilities - A user focus In EPA
Workshop on the Water Quality-based
Approach for Point Source and Nonpoint
Source Controls, June 1991 pp 26-28.
EPA 503/9-92-001
A 34
-------
DR3M-QUAL: Distributed Routing Rainfall Runoff Model - Quality
1. Distributor:
Ms Kate Flynn
410 National Center
U S Geological Survey
Reston, VA 22092
(703) 648-5313
2 Type of Modeling
• Urban stormwater pollutant loads
• Continuous simulation
• Continuous, intermittent, and diffuse
source/release
• Intermediate and detailed applications
3 Model Components
• Rainfall/runoff assessment
• Water quality analysis
4. Method/Techniques:
This model simulates rainfall/runoff
processes (hydrographs) and water
quality processes (pollutographs) in urban
and other areas The kinematic wave
method is used over multiple
subcatchments to predict runoff from
rainfall and drainage pathways A built-in
optimization routine and a storage-
indication routine are available for
estimating water quantity parameters
Exponential build-up and wash-off
functions derived from experience with
model calibration are applied to predict
water quality Empirical equations use
relationships between sediment yield and
runoff volume and peak to simulate
erosion The erosion parameters are
selected based on the USLE The
transport process is modeled assuming plug
flow and using a Lagrangian scheme.
Calibration is required for accurate quality
predictions However, default values may
be used for screening level analysis
7. Applications
• Rainfall/runoff assessment
• Surface water quality analysis
6. Number of Pollutants:
• Sediment, nitrogen, and phosphorus,
metals, and organics
7. Limitations.
• No interaction among quality parameters
• Sediment transport simulation is weak
8 Experience.
The program has been extensively reviewed
within the USGS and applied to several
urban modeling studies, including South
Florida (1980), Rochester (1985 and 1988),
Anchorage (1986), Denver (1987), and
Fresno (1988)
9. Updating Version.
N/A
10. Input Data Requirements*
• Subcatchment data area,
imperviousness, length, slope,
roughness, and infiltration parameters
A 35
-------
• Trapezoidal or circular channel
dimensions and kinematic wave
parameters
• Stage-area-discharge relationships for
storage basins
• Water quality parameters, including
build-up and wash-off coefficients
13. Simulation Output:
• Time series of runoff hydrographs and
quality pollutographs (concentration or
load vs. time) at any location in the
drainage system
• Summaries for storm events
• Graphical output of water quality and
quantity analysis
14. References Available
Alley, W.M. 1981 Estimation of
impervious-area wash-off parameters
Water Resources Research 17(4) 1161-
1166.
Alley, W.M. 1986 Summary of
experience with the distributed routing
rainfall-runoff model (DR3M) In Urban
Drainage Modeling ed C Maksimovic
and M. Radojkovic pp 403-415
Pergamon Press, New York
Alley, W.M. and P E. Smith 1981
Estimation of accumulation parameters
for urban runoff quality modeling Water
Resources Research 17(6) 1657-1664
Alley, W.M. and P E Smith 1982a
Distributed Routing Rainfall-Runoff Model
- Version II. U.S. Geological Survey,
Reston, VA. Open file report 82-344
Alley, W.M and P.E. Smith 1982b
Multi-event urban runoff quality model.
U.S. Geological Survey Reston, VA
Open file report 82-764,
Brabets, T P 1987 Quantity and quality of
urban runoff from the Chester Creek basin
Anchorage, Alaska. U S Geological Survey,
Anchorage, AL Water-Resources
Investigations Report 86-4312
Donigian, A S , Jr , and W C Huber 1991
Modeling of nonpoint source water quality in
urban and non-urban areas U S
Environmental Protection Agency,
Environmental Research Laboratory, Athens,
Georgia EPA/600/3-91/039
Quay, J R and P E Smith 1988 Simulation
of quantity and quality of storm runoff for
urban catchments in Fresno, California. U S
Geological Survey, Sacramento, CA USGS
Water-Resources Investigations Report 88-
4125,
Kappel, W M , Yager, R M and P J
Zarnello 1986 Quantity and Quality of
Urban Storm Runoff in the Irondequoit Creek
Basin near Rochester, New York, Part 2
Quality of Storm Runoff and Atmospheric
Deposition, Ramfall-Runoff-Quality Modeling,
and Potential of Wetlands for Sediment and
Nutrient Retention U S Geological Survey,
Ithaca, NY USGS Water-Resources
Investigations Report 85-4113
Lmdner-Lunsford, J B and S R Ellis 1987
Comparison of Conceptually Based and
Regression Rainfall-Runoff Models, Denver
Metropolitan Area, Colorado, and Potential
Applications in Urban Areas. U S Geological
Survey, Denver, CO USGS Water-Resources
Investigations Report 87-4104,
Zarnello, P J 1988 Simulated water-quality
changes in detention basins In Design of
Urban Runoff Quality Controls, ed L A
Roesner, B Urbonas, and M B Sonnen
Proceedings of Engineering Foundation
Conferences, Potosi, Ml. ASCE, New York
pp 268-277
A 36
-------
SWRRB: Simulation for Water Resources in Rural Basins
1. Name of Distributor.
Dr Jimmy R Williams
Agriculture Research Service
U S Department of Agriculture
Grassland, Soil and Water Research Lab
Temple, TX 76502
(817) 770-6502
2 Type of Modeling
• Stormwater from agricultural
watersheds
• Diffuse source/release
• Screening, intermediate, and detailed
applications
3 Model Components
• Rainfall/runoff assessment
• Surface water quality analysis
• Soil/groundwater contamination
4 Method/Techniques
This program evaluates basin-scale
(large, complex, and rural basins) water
quality Surface runoff is described by
the SCS curve number equation The
modified Universal Soil Loss Equation
(USLE) is applied to simulate erosion for
each basin Degradation and deposition
are considered for the channel and
floodplam sediment routing model
Bagnold's stream power concept is used
for degradation processes, while the fall
velocity of sediment particles is used for
deposition Return flow, percolation, and
crop growth are calculated Soluble and
sediment-attached pollutants are
considered for pollutant transport
Nutrients (nitrogen and phosphorus) are
simulated by using relationships between
chemical concentration, sediment yield, and
runoff volume Calibration is not specifically
required but is desirable
5. Applications
• Water and sediment yields from ungaged
rural basins
• Return flow, pond and reservoir storage,
and crop growth
• Sediment movement through ponds,
reservoirs, streams, and valleys
• Flood routing
6. Number of Pollutants.
Sediment components, nitrogen,
phosphorus, and pesticides
7 Limitations:
• Nutrient transformations are not
evaluated
8 Experience:
The model was tested on 11 large
watersheds The testing results showed that
SWRRB can simulate water and sediment
yield under a wide range of soils, climate,
land use, topography, and management
systems
9 Updating Version
N/A
A 37
-------
10. Input Data Requirements-
* Meteorological data (daily precipitation
and solar radiation)
• Soils, land use, and fertilizer and
pesticide application
11. Simulation Output.
* Daily runoff volume and peak rate,
sediment yield, evapotranspiration,
percolation, return flow, and pesticide
concentration in both runoff and
sediment
• Nutrient concentrations/loads
12. References Available:
Arnold, J.G., J.R Williams, A D Nicks,
and N.B. Sammons 1989 SWRRB, a
basin scale simulation model for soil and
water resources management Texas
A&M Press.
Arnold, J G., and J.R Williams 1987
Validation of SWRRB - simulator for
water resources in rural basins Journal of
Water Resources Planning and Management
113(2) 243-256
Computer Science Corporation. 1980
Pesticide runoff simulator user's manual.
U S Environmental Protection Agency,
Office of Pesticides and Toxic Substances,
Washington, DC
Donigian, A S , Jr , and W C Huber 1991
Modeling of nonpomt source water quality in
urban and non-urban areas U S
Environmental Protection Agency,
Environmental Research Laboratory, Athens,
Georgia EPA/600/3-91/039
Williams, J R ,and H D Berndt 1977
Sediment yield prediction based on
watershed hydrology Transactions of the
ASAE 20(6) 1100-1104
Williams, J R , A D Nicks, and J G Arnold
1985 Simulator for water resources in rural
basins Journal of Hydraulic Engineering
ASCE 111(6) 970-986
A.38
-------
SWMM: Storm Water Management Model
1 Name of Distributor
Mr. David Disney
U S Environmental Protection Agency
Environmental Research Laboratory
College Station Road
Athens, GA 30613
(404) 546-3123
2 Type of Modeling
• Urban stormwater processes
• Continuous and storm event
simulation with variable and user-
specified time steps (wet and dry
weather periods)
• Single, continuous, intermittent,
multiple, and diffuse source/release
• Screening, intermediate, and detailed
planing applications
• Evaluation of BMPs and development
of design criteria
3 Model Components
• Rainfall/runoff assessment
• Water quality analysis
• Soil/groundwater contamination
• Point source inputs available
4 Method/Techniques
This model simulates overland water
quantity and quality produced by storms
in urban watersheds. Several modules or
blocks are included to model a wide
range of quality and quantity watershed
processes A distributed parameter sub-
model (RUNOFF) describes runoff based
on the concept of surface storage
balance The rainfall/runoff simulation is
accomplished by the non-linear reservoir
approach The lumped storage scheme is
applied for soil/groundwater modeling For
impervious areas, a linear formulation is
used to compute daily/hourly increases in
particle accumulation For pervious areas, a
modified Universal Soil Loss Equation (USLE)
determines sediment load The concept of
potency factors is applied to simulate
pollutants other than sediment.
5 Applications
• Urban stormwater and combined systems
• Surface water routing
• Urban watershed analysis, including
baseflow contributions
6 Number of Pollutants
Limited to ten pollutants, including sediment
7 Limitations.
• Lacks graphics routines
• Quality and solids transport simulations
are weak
8. Experience:
Applied to urban hydrologic quantity/quality
problems in over 100 locations in the United
States and Canada
9 Updating Version:
Version 404 (1989)
A 39
-------
10. Input Data Requirements
• Rainfall hyetographs, antecedent
conditions, land use, and topography
• Dry-weather flow and soil
characteristics
• Gutter/pipes - hydraulic inputs
• Pollutant accumulation and wash-off
parameters
• Hydraulics and kinetic parameters
11. Simulation Output:
• Time series of flow, stage, and
constituent concentration at any point
in watershed
• Seasonal and annual summaries
12. References Available
Cunningham, B A and W C Huber
1987. Economic and predictive reliability
implications of stormwater design
methodologies, Florida Water Resources
Research Center, University of Florida,
Gainesville, FL Publication No 98
Dever, R J., Jr , L A Roesner, and J A
Aid rich. 1983 Urban highway storm
drainage model vol. 4, Surface runoff
program user' manual and
documentation. Federal Highway
Administration, Washington, DC
FHWA/RD-83/044.
Dever, R J , Jr , L A Roesner, and D-C ,
Woo 1981 Development and application of
a dynamic urban highway drainage model In
Urban Stormwater Hydraulics and
Hydrology, Proceedings of the Second
International Conference on Urban Storm
Drainage, Urbana, IL pp 229-235 Water
Resources Publications, Littleton, CO
Donigian, A S , Jr , and W C Huber 1991
Modeling of nonpomt source water quality in
urban and non-urban areas U S
Environmental Protection Agency,
Environmental Research Laboratory, Athens,
Georgia EPA/600/3-91/039
Huber, W C 1986 Deterministic Modeling
of Urban Runoff Quality In Urban runoff
pollution, Proceedings of the NA TO
Advanced Research Workshop on Urban
Runoff Pollution, Montpellier, France ed
H C Torno, J Marsalek, and M Desbordes
pp 167-242 Series G Ecological
Sciences Vol 10 Springer-Verlag, New
York
Huber, W C 1989 User feedback on public
domain software SWMM case study In
Proceedings of the Sixth Conference on
Computing in Civil Engineering, Atlanta,
Georgia American Society of Civil
Engineers, New York, NY
Huber, W C and R E Dickinson 1988
Storm Water Management Model Version 4,
User's manual U S Environmental
Protection Agency, Athens, GA EPA600/3-
88/001a (NTIS PB88-236641/AS)
A 40
-------
HSPF: Hydrological Simulation Program - FORTRAN
1. Distributor-
Mr David Disney
U S Environmental Protection Agency
Environmental Research Laboratory
College Station Road
Athens, GA 30613
(404) 546-3123
2. Type of Modeling
• Pollutant load and water quality in
complex watersheds
• Continuous and storm event
simulation
• Single, continuous, intermittent,
multiple, and diffuse source/release
• Screening, intermediate, and detailed
applications
• BMP evaluation and design criteria
3 Model Components:
• Watershed hydrology assessment
• Surface water quality analysis
(conventional and toxic organic
pollutants)
• Soil/groundwater contaminant runoff
processes with mstream hydraulic and
sediment-chemical interactions
(saturated and unsaturated zones)
• Pollutant decay and transformation
4. Method/Techniques.
This model calculates surface and
subsurface pollutant transport from
complex watersheds to receiving waters
Hydrolysis, oxidation, photolysis,
biodegradation, volatilization, and
sorption are used to describe the transfer
and reaction processes First-order kinetic
processes are employed to model sorption
Water quality is simulated by a lumped
parameter model Three sediment types
(sand, silt, and clay) and a single organic
chemical as well as transformation products
of that chemical can be simulated The
program can assess the water quality
impacts of alternative best management
practices (BMPs) Calibration is required for
model application Because of the modular
approach, detail of application can be varied
depending on data availability and modeling
needs
5 Applications
• Surface and subsurface pollutant
transport to receiving water with
subsequent simulation of mstream
transport and transformations
• Watershed hydrology and water quality
for both conventional and toxic organic
pollutants
• Evaluation of BMPs and development of
design criteria
6 Number of Pollutants
Seven pollutants three sediment
components (sand, silt, and clay), one
pesticide or other toxic pollutant (user-
specified), BOD, ammonia or nitrate, and
orthophosphate
7 Limitations
• Limited to well-mixed rivers and
reservoirs
• Extensive water quality sampling data
required for calibration or verification
A 41
-------
* Highly trained staff required for model
application
8. Experience:
• Developed based on field data
gathering and testing at an Iowa site
• Recently used to model erosion and
BMP effects on a west Tennessee
watershed with calibration and
verification
• Extensively applied in a wide variety
of hydrologic and water quality
studies
9. Updating Version:
Version 9.01, 1989
10. Input Data Requirements:
• Continuous rainfall records
• Continuous records of
evapotranspiration, temperature, and
solar intensity
• A large number of parameters need to
be specified (some default values are
available)
11. Simulation Output.
• Time series of the runoff flow rate,
sediment load, and nutrient and
pesticide concentrations
• Time series of water quantity and
quality at any point in a watershed
• Frequency and duration analysis
routine
12. References Available.
Barnwell, T O., and R Johanson 1981
HSPF: A comprehensive package for
simulation of watershed hydrology and
water quality. In Nonpoint pollution
control: tools and techniques for the
future. Interstate Commission on the
Potomac River Basin, Rockville, MD
Barnwell, T O ,and J L Kittle 1984
Hydrologic Simulation Program - FORTRAN
Development, maintenance and application
In Proceedings Third International
Conference on Urban Storm Drainage.
Chalmers Institute of Technology, Goteborg,
Sweden
Bicknell, B R , A S Donigian, and T O
Barnwell 1984 Modeling water quality and
the effects of best management practices in
the Iowa River basin Journal of Water
Science Technology 17 1141-1153
Donigian, A S , J C Imhoff, B R Bicknell,
and J L Kittle 1984 Application Guide for
the Hydrologic Simulation Program -
FORTRAN Environmental Research
Laboratory, U S Environmental Protection
Agency, Athens, GA EPA 600/3-84-066
Donigian, A S , D W Meier, and P P
Jowise 1986 Stream transport and
agricultural runoff for exposure assessment.
A methodology Environmental Research
Laboratory, U S EPA, Athens, GA
EPA/600/3-86-011
Johanson, R C , J C Imhoff, J L Kittle,
A S Donigian 1984 Hydrological
Simulation Program - FORTRAN (HSPF).
User's manual for release 8.0. ,
Environmental Research Laboratory, U S
Environmental Protection Agency, Athens,
GA EPA600/3-84-066
Schnoor, J L , C Sato, D McKetchnie, D
Sahoo 1987 Processes, coefficients, and
models for simulating toxic organics and
heavy metals in surface waters.
Environmental Research Laboratory, U.S
Environmental Protection Agency, Athens,
GA EPA/600/3-87/015.
A.42
------- |