United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-83-086 Nov. 1983
AER& Project Summary
EPA Macroscopic Planning
Model (EPAMAC)for Stormwater
and Combined Sewer Overflow
Control: Application Guide
and User's Manual
William G. Smith and Marianne E. Strickfaden
A simplified stormwater management
model known as EPAMAC (EPA Macro-
scopic Planning Model) has been
developed to provide an inexpensive.
flexible tool for planning and preliminary
sizing of stormwater facilities. The
purposes of this project were (1) to
describe the application of the model to
San Francisco (SFMAC), which lead to
the generalized EPAMAC and (2) to
write a user's guide. Documentation of
the model and instructions for data
preparation and application are included
in the user's guide portion of the report.
The model was effectively used to
compare a large number of alternative
storage and treatment facilities and to
optimize their sizing. The SFMAC was
then generalized for use at any location
as EPAMAC.
EPAMAC is a special purpose, con-
tinuous simulation model for surface
runoff. The model has minimal input
data requirements, broad areal and
temporal coverage, and both flexibility
and ease of application.
EPAMAC was developed as part of a
methodology for managing stormwater
that uses both simple computer programs
and hand computations. The model
consists of three uncomplicated but
interrelated programs that can be used
singularly or together: EPAMAC, HISTO.
and VDAY. In addition to quantity and
quality analyses for stormwater flows,
model capabilities include dry-weather
flow, hourly simulation of flows,
overflow event analysis, determination
of pollutant removals as a result of
sedimentation in storage, and addition
of dry-weather flow and lateral inflows
from adjacent areas. .Postprocessor
programs can also provide histogram
plots of flow and quality (HISTO) and
analyses of violation days for coliform
limits in receiving water for several
coliform standards (VDAY).
The model proved to be useful to
planners and engineers as well as public
enforcement agencies. Information
developed from application of the
model in San Francisco (SFMAC) was
used by the regional water quality
enforcement agency in an evaluation of
appropriate stormwater overflow fre-
quency levels to be included in NPDES
permits. Pollutant loadings developed
from SFMAC application and coliform
violation day data from VDAY were
used in cost/benefit analyses to deter-
mine an allowable overflow frequency
consistent with beneficial uses of the
shoreline, ocean, and bay waters.
This Project Summary was developed
by EPA's Municipal Environmental
Research Laboratory. Cincinnati. OH.
to announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Complex stormwater models provide
valuable data for the design and final
sizing of stormwater facilities. But the
-------
detailed models generally require compre-
hensive, detailed input data and large
blocks of time on relatively large computers
for execution. In an attempt to provide an
intermediate step between the large,
complex models and the traditional
desktop analysis used in facilities planning
projects, the U.S. Environmental Protec-
tion Agency (EPA) has participated in the
development of several simplified comput-
er models for planning and preliminary
sizing of facilities. These simplified
models require very moderate expenditures
for data preparation and execution, and
they provide a flexible tool for analyzing
a variety of system configurations.
The application guide and user's
manual presented in the full report
summarize the mathematical modeling
work performed in support of the storm-
water and combined sewage treatment
master plan for the City and County of
San Francisco and provide a compact
reference base for applying the model to
future locations. The model consists of
three uncomplicated but interrelated
programs that can be used singularly or
together (EPAMAC, HISTO, and VDAY).
This design allows the user to build on his
individual data strengths and to focus on
individual study objectives.
Background
San Francisco has a unique physical
setting that must be considered in the
development of wastewater facilities.
The City is located on a peninsula, which
allows for the option of effluent discharge
to either the bay or ocean. The planning
area contains 23 watersheds comprising
11,554 ha (28,550 acres) with numerous
hills and peaks. Much of the land slopes
steeply to the water, but a flat coastal
strip exists along the east side of the City.
Rainfall occurs mainly from October
through May, and average annual rainfall
is about 51 cm (20 in.).
The San Francisco combined sewer
system conveys both dry- and wet-
weather flows. To advance the planning
concepts and to meet water quality
requirements, a citywide computer
modeling approach was necessary.
During dry weather, major concerns are
transportation and treatment of the
flows, and the resulting quality of the
effluent. Wet-weather concerns also
include minimizing overflow volume and
frequency through storage and pumping
balance, and maximizing quality of
overflow and effluent. Since flows from
different sections of the city influence
one another, the reaction of the whole
system to one or a series of storm events
must be examined. Runoff volumes from
all 23 watersheds need identification and
integration to obtain guidance on facilities
necessary to process the flows cost
effectively. Rainfall data for 70 years of
record are available for analysis. Such an
analysis needs to be accurate enough to
be reliable but not so involved that it
would hinder the planning process.
Computer simulation using an appropriate
model would achieve analysis of a large
number of alternatives reasonably short
time.
The types of alternatives to be analyzed
involve major regrouping of watersheds
and routing of flows to multiple locations.
Four alternative concepts are presented
in the rnaster plan (Figure 1). Some of the
NSPS
SWOO
Dual Crossing.
: Outfall
NPWPCP
NSPS
swoo
Modified Ocean Discharge.
Legend
« Wet - Weather Flow Routes
Pump Station
Treatment Plant
Outfall
alternative routes pass through politically
sensitive areas. The ultimate planning
goal is to determine for each alternative
the ability to meet effluent discharge
requirements, overflow frequency limita-
tions, and the minimum cost for an
optimized system.
Existing computer models were either
too detailed to quickly survey multiple
alternatives or too broad to examine
storage and treatment interactions. Thus
SFMAC (later generalized as EPAMAC)
was developed to meet the needs of the
San Francisco planning process. The
relationship of the computer modeling to
the overall planning process is shown in
Figure 2.
Low Level Bay side.
.Outfall
NPWPCP
NSPS
swoo\SWWPCP
Bay and Ocean Discharge.
NSPS North Shore Pumping Station
IPS Islais Pumping Station
WSPS Westside Pumping Station
SWWPCP Southwest Water Pollution Control Plant
SWOO Southwest Ocean Outfall
Figure 1. Alternative concepts for wet-weather master plan.
-------
Initial Preparation
Data Gathering
Input Preparation
EPA MA C and Post Processor
Computer Runs
Result Analysis
Selection of New Values or
Selection of New Alternatives
Narrowed Set of Possible
Alternatives
Evaluation and Ultimate
Selection of Alternatives
Using Social, Environmental
and Cost Factors
Rainfall input data were developed
from two sources. The first source was a
network of 30 gages that the City of San
Francisco established in December 1971.
The second source was the National
Weather Service (NWS), which has
maintained a recording gage in San
Francisco since January 1907. The NWS
gage record for the 69-year period (1907-
76) was compared with the 4-year (1972-
76) data set. The storm duration, intensity,
and magnitude characteristics of both
sets of data showed very close agreement.
Water quality also was determined in
the preparatory phase. Dry-weather data
were obtained from influent records for
the three existing sewage treatment
plants. Projected per capita dry-weather
loadings were used as input to the model.
Treatment plant influent data for storm
days were analyzed to obtain a wet-
weather quality characterization. Since
the plant records were based on daily
composite samples, a sampling program
was developed to obtain additional wet-
weather characterization data.
Watershed Data Analysis
After input preparation was completed,
EPAMAC was run for each subarea. The
hourly runoff volume and quality were
established for each subarea and water-
shed in the initial EPAMAC runs. Pollutant
loads from each watershed were obtained
from the model. The next phase of the
application was to route the subarea and
watershed flows to storage/treatment
facilities. A flow-routing diagram was
developed for each system configuration.
EPAMAC was then run again to model
the system flow routes for the alterna-
tives.
Balance of Storage and
Treatment Requirements
In addition to routing the flows, the
second series of EPAMAC runs was used
to determine the optimal sizes of storage
and treatment facilities. During this
stage, storage and pumping requirements
along with treatment plant capacity were
balanced to obtain the desired overflow
frequency during wet weather.
The relationship between pumping or
treatment rates and overflows is illustrated
in Figure 3. Since several storage and
treatment size combinations can produce
a given overflow frequency, a separate
hand cost computation optimized the
selection for each system configuration.
Effect of Sedimentation in
Storage Facilities
One feature of EPAMAC is the ability to
model the pollutant removal effect of
storage (sedimentation) on overflows
from storage. A constant pollutant
removal efficiency factor is applied only to
the overflow volume to determine the
effect of storage on improvements in
overflow water quality. Additional EPAMAC
Figure 2. Relationship of EPAMAC appli-
cation to the planning process.
Model Application
Input Preparation
The first step in applying SFMAC was
the preparation of input data, which
involved identification and gathering of
watershed and collection system charac-
teristic data.
The population, area, and general
character of each watershed were
determined. Information on watershed
characteristics, potential sites for future
storage facilities, and storage available in
the existing system were identified.
Watershed data were reviewed and
regrouped because systemwide alterna-
tives modified individual watershed
configurations.
The a reas a nd ru noff coeff ici ents for six
different land use types were determined
for each subarea. Runoff coefficients
were determined from actual rainfall and
sewer flow records.
o
u.
I
r
WW &DW Max Pumping Rate
(Includes Effects of Storage)
DW Max Pumping (Treatment
Rate)
Time-
Overflow Portion
Treatment Portion
WW Treatment
DW Treatment
Figure 3. Overflow and dry- and wet- weather flow relationships to pumping rates.
3
-------
runs compared the effects of pollutant
removal during storage on receiving
water quality.
Analysis of Receiving
Water Quality
After EPAMAC runs established water-
shed characteristics and system treatment
and storage interactions, the postprocessor
programs were used on EPAMAC output to
glean information about water quality im-
pacts on receiving waters. The violation day
postprocessor program (VDAY)was used to
determine the effect of the overflows for va-
rious storage/treatment capacity options on
the receiving water coliform concentra-
tion limits.
The postprocessor program HISTO, which
produces histogram plots of flow, concen-
tration, and mass loads, was used to evalu-
ate the frequency, volume, and pollutant
load of overflows and treatment plant dis-
charges.
Plant Operation Frequency
Once the storage and treatment capaci-
ties were established, it was necessary to
estimate the length of time the wet-weather
portion of the plant would operate for each
storm and the time available between storm
events to drain and clean sedimentation
tanks and grit chambers.
HISTO was run on EPAMAC output during
a year of typical rainfall. The time between
storms was determined for each event Thus
it was possible to determine a representa-
tive number of times the tanks would be
clean at the start of a storm and also the
number of times the tanks would still con-
tain water or sediment at the start of the
next storm event.
Model Description
The EPAMAC model (Figure 4) computes
the runoff and eventual discharge to the re-
ceiving water. Multiple lateral inflows, dry-
weather flows, pumping rates, and storage
operation modes are included in the model.
HISTO prints summary tables and graphs
from EPAMAC results for statistical analy-
sis. The VDAY program features include
coliform concentration calculation in re-
ceiving water (including die-off overtime),
a statistical analysis of overflow events and
violation days, and the option to select
specific time periods for analysis.
Technical Basis and Assumptions
The fundamental theory behind EPAMAC
is the rational method of computing
runoff. The rational method relates flow
to the contributing area, rainfall, and
runoff coefficient for the area to determine
the runoff volume for each time step.
Maximum
Storage
Capacity
Pollutant
Removals
Receiving Water
Receiving Water
Figure 4. EPAMAC schematic.
By using continuous rainfall records,
the method can be used to approximate a
continuous simulation model.
The rational method is most appropriate-
ly used in urban areas of less than 100
acres (41 ha) or where runoff is spread
over the surface and collected in numer-
ous inlets. The method is somewhat sim-
plistic; but the simplicity is appropriate for
EPAMAC, since it is a broad-scale model
for use in estimating long-term results
rather than simulating individuals storm
events.
Another theory used in EPAMAC
involves mixing in storage. The complete
mix theory assumes that flow and the
associated pollutants are immediately
dispersed throughout storage. The parti-
cles then leave in proportion to the
average concentration and flow volume
removed from storage at any time step.
Sedimentation theory is involved in
estimating the effect of storage on the
percent removal of solids from storage
overflows. The ideal, theoretical settling
velocity of a material in water is a
function of particle size, shape, and
specific weight. Other factors such as
turbulence and horizontal velocity affect
how a particle will actually settle in a
transport or storage facility. The detailed
procedure for making the estimate is
included as Appendix C.
-------
Several assumptions were chosen to
maintain the EPAMAC model simplicity
and serviceability. The major assumptions
are:
Pollutants are conservative and do
not degrade.
Suspended solids removal efficiencies
that are due to settling overflow
volumes in storage are constant over
varying flowrates.
Three wet-weather flow quality
concentrations can be input as
separate values for the first hour,
second hour, and all subsequent
hours of a storm to simulate first flush
effects.
The HISTO program is simply a data
transformation program: It takes the
input data and prints them in a new
format namely, histogram plots and
range summary tables.
The fundamental theory behind VDAY
is that the time variation of the coliform
concentration in the receiving water after
an overflow event follows Chick's law (i.e.,
the longer the time since the end of the
overflow, the greater the die-off of the
organisms).
The user should keep in mind that the
values for the coliform die-off decay rate
were determined through regression
analysis of data specific to San Francisco.
The user should determine, whenever
possible, the appropriate values for these
coefficients for any other location where
the program is applied. The decay rates
for San Francisco can be used as default
values, if necessary; but the results
should then be used for relative compari-
sons only.
Limitations
As a consequence of the simplifying
assumptions used in the program, the
programs should be executed with at
least 10 to 20 year's worth of rainfall
records to provide results that represent
the actual ranges that can be expected to
occur. Shorter periods of record can be
used, but the results may be less
representative of what can actually be
expected.
Other limitations of EPAMAC include:
BOD and SS are the only pollutants
modeled.
Wet-weather flow quality for each
pollutant is based on three input
values: One for the first hour, one for
the second hour, and one for the third
and any subsequent hours of a storm.
Printout is in units of-Mgal, klb, and
mg/L only for the appropriate time
period selected.
A maximum of four lateral inflows
(flows from previously simulated
adjacent areas) is allowed.
Values from monthly and annual
summaries are considered represen-
tative of the actual periods simulated.
Values from hourly and daily summaries
cannot be considered truly represen-
tative of actual storm events because
of simplifying assumptions used in
the model.
Limitations of the VDAY program include:
The attenuation rate of coliforms in
receiving water can be a default value
based on the overflow volume (included
in program) or a user-supplied value
developed for the specific area being
simulated.
Printout is in units of Mgal and
violation days.
Printout is on an annual basis and is
only generally representative of
conditions being modeled. Values
from shorter time periods (less than
20 years) will produce an incomplete
statistical analysis based on the
ranked values only.
Benefits
The major benefits from application of
EPAMAC were the ability to evaluate a
large number of alternatives, the ease of
understanding the results, the quickness
of obtaining results, and the acceptance
of results in decision-making processes.
The use of EPAMAC met the planning
objectives of examining a large number of
alternatives (58 in the case of San
Francisco) at a technical level sufficient to
determine differences in the water
quality effects and the overflow frequen-
cies. System configurations were identi-
fied and used as a basis for comparing the
cost-effectiveness of the alternatives.
EPAMAC results were also used to
achieve the objective of establishing
reasonable allowable overflow frequency
limits.
For planning and comparing alternatives,
EPAMAC is sufficiently accurate. Verifica-
tion indicated that the real situation is
adequately modeled. As with all models,
the accuracy of the results depends
strongly on the input data.
Applying a model to San Francisco can
become unreasonably complex because
of the size and intricacy of the sewerage
system. But the EPAMAC adequately
handled the task. Results were understand-
able, and the volume of computer output
was under user control. The overflow
event data and mass balance summaries
provided sufficient detail without generat-
ing a volume of data that would be
impossible to review or comprehend. The
model was a tool for comprehending a
complex system without obscuring the
goals. The results could be used to
determine further implications of storage
and treatment balances. The postprocess-
ors also provided necessary information
and understandable results.
The EPAMAC algorithm is simple and
straightforward, unlike those used in
some detailed, single-event simulation
models. An extensive educational program
is not needed to help the user comprehend
the model and its required input.
The most significant benefit of EPAMAC
application was not just in analysis of
alternatives in the facilities plan. Informa-
tion developed was also used by the
regional water quality enforcement
agency to evaluate appropriate, allowable
overflow frequency levels to be included
in NPDES regulations. Pollutant loadings
developed from EPAMAC simulation
runs and violation day data from VDAY
were used in cost/benefit analyses to
determine allowable overflow frequency
levels consistent with beneficial uses of
the shoreline, ocean, and bay waters. In
conclusion, the model results were
accepted at several levels of the decision-
making process and were used to
establish policy.
Model Documentation and
User's Guide
The report includes a section that
provides EPAMAC documentation and
serves as a user's guide. This section
includes (1) general description, (2) user
instruction, (3) technical concepts, and
(4) program documentation.
The general description includes the
capabilities, features, restrictions, and
limitations of the EPAMAC and postproc-
essor programs.
User instructions are provided for the
program processing steps, system schemat-
ics, alternative analysis, execution
logistics and input data preparation.
Example input data sets, example outputs,
and interpretation of the results are also
described. The theory, assumptions, and
limitations for EPAMAC, HISTO, and
VDAY are presented in the technical
concepts section of the report.
Program documentation for EPAMAC
and the postprocessor programs includes
the mode of operation, program functions
and internal algorithms, and program
segments and subroutines. A section on
job control statements and input/output
file descriptions is provided for the
EPAMAC program.
-------
A list of each program and a description
of the variables are included in Appendix
A.
Summary and Recommendations
The EPAMAC is a generalized version
of the SFMAC used during the
development of San Francisco's
Southwest Water Pollution Control
Plant Facility Plan and Master Plan.
EPAMAC is a special purpose, contin-
uous simulation model of surface
runoff with minimal input data
requirements, broad areal and tempor-
al coverage, and both flexibility and
ease of application.
In addition to quantity and quality
analyses for stormwater flows, capabil-
ities of the model include dry-
weather flow, hourly simulation,
overflow event analysis, pollutant
removal as a result of sedimentation
in storage, and lateral inflows.
Postprocessor programs can also
provide histogram plots of flow and
quality as well as analyses of violation
days for coliform limits in receiving
water for several coliform standards.
Information developed by the model
was used by decision-makers for
establishing allowable overflow fre-
quency limitations for NPDES permits
for wet-weather discharges.
The overall effectiveness of the model
in this application was high. The
client and consultant were both able
to contribute significantly to the
model application process. Also, the
flow and quality concerns in a
combined sewer system were adequate-
ly addressed. In addition, the planning
objectives were met, and preliminary
sizing of a large number of diversified
alternatives was accomplished.
The EPAMAC and postprocessor
programs should be implemented as
a preliminary design and planning
tool.
The EPAMAC should be used repeated-
ly to analyze various combinations of
storage capacities and treatment
plant rates to determine possible
optimum conditions. The EPAMAC
should be used to examine seasonal
and yearly periods.
The full report was submitted in partial
fulfillment of Contract No. 68-03-2877
by Metcalf & Eddy, Inc., under the
sponsorship of the U.S. Environmental
Protection Agency.
William G. Smith and Marianne E. Strickfaden are with Metcalf & Eddy,
Engineers, Inc., Palo Alto, CA 94303.
Richard Field is the EPA Project Officer (see below).
The complete report, entitled "EPA Macroscopic Planning Model (EPAMAC) for
Stormwater and Combined Sewer Overflow Control: Application Guide and
User's Manual," (Order No. PB 83-259 689; Cost: $26.50, subject to change)
will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Storm and Combined Sewer Program
Municipal Environmental Research LaboratoryCincinnati
U.S. Environmental Protection Agency
Edison, NJ 08837
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
U.S. POSTAGE
PAID
Cincinnati, Ohio
Permit No. G35
Official Business
Penalty for Private Use $300
LOU rt TILLEY
REGiON V EPA
LIBRARIAN
?SO S DEARBORN ST
CHICAGO IL 60604
U.S. GOVERNMENT PRINTING OFFICE: 1983-759-102/0779
------- |