Environmental Protection Technology Series
  STORM  WATER  MANAGEMENT MODEL:
DISSEMINATION AND USER ASSISTANCE
                       National Environmental Research Center
                         Office of Research and Development
                        U.S. Environmental Protection Agency
                                 Cincinnati, Ohio 45268

-------
                                         EPA-670/2-75-041
                                         May 1975
            STORM WATER MANAGEMENT MODEL:
          DISSEMINATION AND USER ASSISTANCE
                         By

                  James A. Hagarman
                         and
                  F. R. S. Dressier

            University City Science Center
           Philadelphia, Pennsylvania  19104
                Project No. R-802716
             Program Element No. 1BB034
                   Project Officer

                    Chi-Yuan Fan
Storm and Combined Sewer Section (Edison, New Jersey)
    Advanced Waste Treatment Research Laboratory
       National Environmental Research Center
               Cincinnati, Ohio  45268
       NATIONAL ENVIRONMENTAL RESEARCH CENTER
         OFFICE OF RESEARCH AND DEVELOPMENT
        U.S. ENVIRONMENTAL PROTECTION AGENCY
               CINCINNATI, OHIO  45268

-------
                            REVIEW NOTICE
The National Environmental Research Center - Cincinnati has reviewed
this report and approved its publication.  Approval does not signify
that the contents necessarily reflect the views and policies of the
U.S. Environmental Protection Agency, nor does mention of trade names
or commercial products constitute endorsement or recommendation for
use.
                                  11

-------
                            FORWORD
     Man and his environment must be protected from the adverse effects
of pesticides, radiation, noise and other forms of pollution, and the
unwise management of solid waste.  Efforts to protect the environment
require a focus that recognizes the interplay between the components of
our physical environment—air, water, and land.  The National Environ-
mental Research Centers provide this multidisciplinary focus through
programs engaged in

          . studies on the effects of environmental contaminants
            on man and the biosphere, and

          . a search for ways to prevent contamination and to
            recycle valuable resources.
     This study documents a program of dissemination and user assis-
tance for the EPA Storm Water Management Model.  The program offers
municipalities a simple, inexpensive method of utilizing SWMM for
stormwater management.
                                            A. W. Breidenbach, Ph.D.
                                            Director
                                            National Environmental
                                            Research Center, Cincinnati
                                111

-------
                             ABSTRACT
     A program of dissemination and user-assistance for the EPA Storm
Water Management Model (SWMM) has been developed and implemented at
the University City Science Center (UCSC).

     Services available to SWMM users under this grant include dis-
tribution of the SWMM program itself  and technical .assistance in
problem delineation, data preparation, execution debug, and output
interpretation.  Costs of this service extend only to actual compu-
ting costs, with all technical assistance covered by the EPA grant.

     Several case studies of SWMM applications completed with UCSC
assistance in the past year are included in this report.  These
studies include a combined sewer overflow problem in Binghamton,
New-York; a land use plan in the Stony Brook basin in Princeton,
New Jersey; and RUNOFF/TRANSPORT calculations on the Wingohocking
basin in Philadelphia, Pennsylvania.

     The UCSC SWMM dissemination program is now self-sustaining
and continues to assist the user community.

     This report was submitted in fulfillment of the Office of
Research and Development, U.S. Environmental Protection Agency (EPA)
research grant EPA No. R-802716 by the UCSC under the sponsorship
of the EPA.
                                 IV

-------
                            CONTENTS
               Abstract




               List of Figures




               List of Tables




               Acknowledgements
                                        Page




                                          iv




                                          vi




                                         vii




                                        viii
Sections




  I




  II




  III




  IV




  V




  VI




  VII




  VIII




  IX
Conclusions




Recommendations




Introduction




Use of SWMM at the Science Center




Costs for Model Application




Case Studies




Discussion




References




Appendix
 1




 2




 3




 6




14




16




44




45




46
                              v

-------
                         FIGURES

No.                                                               Page

 1       Modes of SWMM Use                                          7

 2       SWMM Data Sets                                             8

 3       SWMM Program Block Structure                               9

 4       Binghamton, New York - SWMM Discretization                19

 5       Binghamton, New York - Storm 1-Hyetograph                 21

 6       Binghamton, New York - Storm 2-Hyetograph                 21

 7       Binghamton, New York - Storm 1-Subcatchment 36
           Hydrograph                                              24

 8       Binghamton, New York - Storm 2-Subcatchment 36
           Hydrograph                                              24

 9       Binghamton, New Yrok - Storm 1-Outfall Hydrograph         26

10       Binghamton, New York - Storm 2-Outfall Hydrograph         27

11       Stony Brook-Baldwin Creek - Subbasins                     31

12       Stony Brook-Baldwin Creek - Storm 1-Hyetograph            32

13       Stony Brook-Baldwin Creek - Storm 2-Hyetograph            32

14       Stony Brook-Baldwin Creek - Storm 1-Outfall
           Hydrograph                                              34

15       Stony Brook-Baldwin Creek - Storm 2-Outfall
           Hydrograph                                              34

16       Wingohocking - Runoff for Storm 3                         40

17       Wingohocking - Runoff for Storm 4                         40

18       Wingohocking - Runoff for Storm 5                         41

19       Wingohocking - Runoff for Storm 6                         41

20       Wingohocking - Runoff for Storm 7                         42

21       Wingohocking - Runoff for Storm 8                         42

22       Wingohocking - Science Center Hyetograph for Storm 6      43

23       Wingohocking - SWMM Hyetograph from Reference 1
           for Storm 6                                             43
                                  vi

-------
                          TABLES


No.                                                             Page


 1        UNI-COLL Computing Rates                                6

 2        RUNOFF BLOCK Data Requirements                         10

 3        TRANSPORT BLOCK Data Requirements                      11

 4        STORAGE/TREATMENT BLOCK Data Requirements              12

 5        RECEIVE BLOCK Data Requirements                        12

 6        SWMM Execution Costs                                   15

 7        Binghamton, New York - Subcatchment Data               22

 8        Binghamton, New York - Infiltration and Dry
            Weather Flow                                         25

 9        Binghamton, New York - Storm 2-Surcharge Quantity
            and BOD                                              28

10        Stony Brook-Baldwin Creek - Subcatchment Data -
            Storm 1                                              35

11        Stony Brook-Baldwin Creek - Subcatchment Data -
            Storm 2                                              35

12        Wingohocking - Summary of SWMM Testing                 39

13        Conversion Table                                       46
                                vii

-------
                         ACKNOWLEDGMENTS
Support from EPA project personnel, especially Mr.  Richard Field,
Mr. Chi-Yuan Fan and Mr. Harry Torno,  was invaluable in this project.
The support of the staff of UNI-COLL Corporation, Philadelphia, Penn-
sylvania was appreciated for general computing support.
Assistance in modeling in Binghamton was provided by the Staff of Quirk,
Lawler and Metusky,  Inc., Tappan,  New York,  especially Dr.  Mohamed M.
Elsahragty.
Assistance in modeling the Stony Brook basin was provided by the Depart-
ment of Regional Planning, University of Pennsylvania,  Philadelphia,
Pennsylvania, especially Mr.  Lanny Maxwell.
                                 viii

-------
                            SECTION I

                           CONCLUSIONS


1.  The SWMM dissemination method utilized in this study has introduced
    the SWMM technology to a wide range of potential-users.

2.  The SWMM user-assistance program utilized in this study provides
    the computer program in several easy-to-use modes and covers techni-
    cal assistance in problem delineation, data reduction, debugging,
    and output interpretation.

3.  The dissemination effort reached a wide geographic assortment of
    consulting engineers, but touched only a rather narrow geographic
    sampling of municipal planning groups.

4.  SWMM modeling assistance was supplied to an engineering consulting
    firm for a study of combined sewer overflows in Binghamton, New
    York.  The first phase of this study was completed and served to
    locate and quantify overflows in the Binghamton combined sewer
    system.

5.  The adaptability of SWMM was demonstrated in the Binghamton study
    by modification of the TRANSPORT block of SWMM to allow direct rou-
    ting of sewer surcharges to the receiving water rather than holding
    in the sewers for later release.

6.  SWMM can be applied to stormwater runoff problems in developing
    watersheds, as demonstrated in a study of the Stony Brook basin near
    Princeton, New Jersey.  Proper accounting of soil infiltration
    losses in pervious areas is essential for good results.  Small
    streams can be idealized as "gutters" within the RUNOFF block.

7.  Quality data for verification of SWMM quality results was found to
    be very scarce.

8.  SWMM modeling serves purposes apart from its primary output of quan-
    tity and quality of storm or combined sewer flows.   Setting up for
    SWMM modeling naturally directs thought and data collection along
    sensible lines.  The overall organization of SWMM forces users to
    realize the importance of dynamic modeling of both quantity and
    quality of storm flows and thus directs data collection towards
    continuous measurement of pollutant levels and flow quantities.

9.  Assistance in defining the problem, relating it to SWMM, and, out-
    lining the capabilities and requirements of SWMM is necessary.  It
    appears, so far, that simply mailing a SWMM tape to a potential-user
    is insufficient support to generate use.  Although heavy support is
    required in the early stages of modeling, effective transfer must
    also include education to promote user independence.

-------
                              SECTION II

                            RECOMMENDATIONS
1.   Use of the Storm Water Management Model (SWMM)  and appropriate data
    collection as an integral part of municipal planning for control of
    storm or combined sewer pollution is highly recommended.

2.   Use of SWMM for storm water pollution control in developing water-
    sheds should be further explored.

3.   The quality portions of SWMM must be better verified.   Programs of
    data collection for such verification should be undertaken both in
    urbanized areas and in developing watersheds.

4.   The SWMM dissemination effort should be expanded to cover a wider
    geographic range of municipal planners.

5.   The SWMM user-assistance program should be extended to support con-
    tinuing modeling efforts as well as data-collection programs presently
    underway.

6.   A simple, eye-catching document should be prepared to  introduce SWMM
    to the user-community.  A follow-up document outlining capabilities
    and data requirements of SWMM is also required  for further dissemin-
    ation of this technology.

 7-  Use of SWMM is strongly recommended for all EPA 208 and 303-type
    grants.  Appropriate data collection for SWMM verification
    should be closely tied to modeling efforts.

-------
                           SECTION III

                          INTRODUCTION
GENERAL

    The problems arising from stormwater runoff in urbanized areas
have been proven to be extensive in a number of studies.   Stormwater
runoff naturally collects the refuse of human and animal activity and
efficiently transports this refuse to receiving waters.  In this manner,
storm flows become major pollutant sources during rainy periods.  Com-
bined sanitary and stormwater collection systems worsen the problem
by including sanitary waste with the already polluted storm runoff.
Extensive impervious cover in urbanized areas aggravates the storm
runoff problem by decreasing infiltration and increasing total storm
flows.  Since impervious cover is generally smoother than natural, per-
vious cover, the storm flows are not only increased in volume but also
transported across land surface more quickly, increasing peak flows
downstream and shortening peak arrival times at downstream points.

    In the past, treatment of combined storm and sanitary flows has been
minimal.  The excess quantity of the storm flow necessitated diver-
sion of combined sewage from the normal treatment methods to the receiv-
ing waters.  Most attempts to relieve pollution during wet weather have
centered on holding as much storm flow as possible in either in-line
sewers or off-line storage facilities until the storm period has passed.
Stored stormwater is then routed to treatment during dry periods.

    In order to plan sensibly for storage and treatment of wet-weather
flows, a considerable amount of information concerning the duration,
quantity, and quality of wet-weather flows must be available.

    This necessary information depends upon the nature of the storm
event in question along with a large number of physical characteristics
of the watershed.  The connection between a storm event, the physical
characteristics of the watershed, and the timing, quantity and quality
of storm flows is exceedingly complex.  Simple rainfall/runoff models '
do not consider the time-variance of wet-weather flows.  For this reason
simple models are inadequate for detailed stormwater planning.
STORM WATER MANAGEMENT MODEL

    To assist anyone planning for the control of pollution from storm-
water or combined sewer overflows, the U.S. Environmental Protection
Agency has sponsored development of the Storm Water Management Model
(SWMM) by a consortium of contractors including the University of Florida,
Water Resources Engineers, and Metcalf & Eddy, Inc.

-------
     SWMM is a dynamic event-simulation model which can predict a time-
history of quantity and quality of stormwater runoff at all points in a
watershed.  This model has been programed for computer-based use.

     Starting with a time-history of a storm event (hyetograph), SWMM
models a watershed (using data on land use, topography, population
density, and natural and man-made drainage systems) to route storm-
water overland, through open channels and pipes, and into a collection
system.  If the collection system is a storm or combined sewer, computer
modeling of the sewer allows SWMM to further route the stormwater
through the existing network of sewer elements.  SWMM presents options
to hold stormwater in storage systems for later removal, and, allows
modeling of various types of storage devices.  Stormwater treatment
options can be examined (using a treatment package in SWMM) prior to
routing into receiving waters.

     Effects of quantity and quality of stormwater upon the receiving
waters are also predicted based upon natural flow or tides in the
receiving-water system.  Time-histories of quantity and quality of flow
can be obtained at all points in the stormwater conveyance system.

     SWMM is designed to allow analysis of large areas using a "building
block" approach, starting from intensive analyses of smaller constituent
drainage basins.
OBJECTIVES

     The University City Science Center (UCSC) has, for the past -sixteen
months, been executing an EPA grant specifically designed to disseminate
the existence and capabilities of SWMM to the potential-user community,
and, provide technical assistance in SWMM use where appropriate.  The
Science Center was specifically charged with responsibility for instal-
ling SWMM on in-house computers in a complete and correct format as test-
ed by data supplied by the EPA.  All revisions and corrections to SWMM
as well as major new versions were to be implemented as they became
available.  SWMM was to be made available to users in several formats,
depending on user need.  Technical assistance for users was to be
supplied in describing required data, reduction of engineering data to
formats suitable for SWMM, debugging of SWMM execution runs and inter-
pretation of SWMM output.
DISSEMINATION OF SWMM

     The first step in the dissemination part of this project was the
development of an informational brochure about SWMM.  This document was
designed to capture the interest of a prospective user through a simple
and attractive format.  The content of the brochure is technically simple
and answers the basic question —What is SWMM?  Examples of past and

-------
potential SWMM applications were outlined.  Details of the available
technical assistance were also given.  The brochure was reviewed and
approved by EPA personnel and 2,000 copies were printed.  Mailing lists
of potential users were then compiled and brochures sent.  Approximately
750 brochures have been mailed to Engineering and Design Firms (550),
government agencies in the states of Pennsylvania and New Jersey (5),
planning boards (Pennsylvania and New Jersey - 100), educational insti-
tutions in the Delaware Valley (25), and conservationists and environ-
mentalists (50).  To date, approximately thirty responses have been
received requesting more information on SWMM.  Each response was con-
tacted by mail or telephone and sent a SWMM OVERVIEW excerpted from
Reference 1 giving further detail on SWMM capabilities and data require-
ments.  Further discussions with interested parties led to the develop-
ment of a talk and slide show on SWMM which was given five times over
the course of the project to a wide variety of audiences.  One talk
was given at the 25th National Plant Engineering and Maintenance Confer-
ence in Cleveland, Ohio, March 20, 1974.  Requests were also received
for SWMM program tapes from four potential users.  Although modeling is
planned in each of these four cases, none has been actually undertaken
to date.

-------
                              SECTION IV
                         USE OF  SWMM AT UCSC

COMPUTER SYSTEM DESCRIPTION

      SWMM has been set up on  a  computing system at UNI-COLL Corporation.
UNI-COLL Corporation is a neighbor to the University  City Science Center
and  functions as an extremely large,  regional computing  center servicing
universities, hospitals, government agencies and'private corporations
needing  high-level computing.   UNI-COLL houses an IBM 370/168 with 3
million  bytes of fast core operating  under the OS/VS2 operating system.
OS/VS2 is a virtual system extending  core space beyond real memory to
virtual  space available on high-speed disk.  This feature not only
allows large programs to reside in core without use of OVERLAY, but also
facilitates job scheduling, allowing  more efficient use  of fast core.
UNI-COLL has extensive service  facilities for users and  interactive
capabilities supporting all conventional dial-up terminals.   A schedule
of UNI-COLL computing rates is  given  in Table 1.

                  Table 1. UNI-COLL COMPUTING RATES
          USAGE TYPE
                                                      USAGE RATE3
          CENTRAL PROCESSING UNIT (CPU)
              Batch Processing

              Terminal Processing (TSO)
$0.50/sec.

$0.75/sec.
$1800/hr.

$2700/hr.
          STORAGE
              Real CPU                 $.00056 per kilobyte per sec. $2.00/hr.
              Virtual CPU              $.00028 per kilobyte per sec. $1.00/hr.
              On-Line-Storage (OLS-3330 disk)                  $0.20/trk/day
          TERMINAL CONNECT TIME

          INPUT/OUTPUT
              Cards Read
              Cards Punched
              Lines Printed
                   $3.60/hr.
                   $0.60/1000
                   $2.50/1000
                   $0.60/1000
a.  DISCOUNT  -  A discount is applied  to  the basic charge  for  CPU, Storage,
Channel and Input/Output depending  on the shift on which  the  job is sub-
mitted or specified to be run.  CPU,  Storage and Channel  charges are dis-
counted to 85 percent of base charges for second shift  (1800-2400 hrs.)
and 65 percent  of base charges for  third shift (2400-0830  hrs.).  A
weekend shift discount to 55 percent  of  base charges  (Sunday  1800-2400)
is also available.   Print and punch charges are discounted by approximate-
ly eight percent and sixteen percent  for second and third  shift respectively

-------
SWMM SETUP

     The  most recent version of SWMM has been made operational on an IBM
370/168 at UNI-COLL Corporation.  The program has been stored  for dis-
semination in several formats depending upon user need.  A LOAD MODULE
is kept in on-line disk  storage and can be  used for batch processing and
remote processing.  This LOAD MODULE can be copied to tape and sent to
IBM users operating under IBM OS (operating system) with 328 K of core
storage.   Use of a LOAD  MODULE is the simplest method of accessing SWMM
and avoids costly compilation, overlay, and linkage editing.   Both
FORTRAN SOURCE and OBJECT files may also be copied to tape and sent to
users unable to utilize  a LOAD MODULE.  SAMPLE DATA sets and JOB CONTROL
LANGUAGE  are also available to verify correctness of SWMM program tapes
and are setup for simple batch or remote terminal processing.  SWMM pro-
gram correctness was established using data from the University of
Florida study on the Stevens Avenue Drainage Basin in Lancaster,
Pennsylvania.4

     Two  general modes of SWMM use are available through this  grant as
shown in  Figure 1.  SWMM may be used in-house at UNI-COLL
                                        Process
                                                       Services
                                                       •
                                                       Problem
                                                       Definition

                                                       Data
                                                       Requirements

                                                       Data Setup

                                                       Access to
                                                       SWMM Program
                                                       Job Control
                                                       Language
                                                       Execut ion.
                                                       Debugging

                                                       Output interpretation
                                                       model rebuild
                                                       SVMM program
                                                       secup-out-of-house
                                                       Sample and
                                                       case-study data
                                                       Sample and
                                                       case-study outputs
                                                       SWMM tailoring
                                                       Co specific needs
                     Figure 1. MODES OF SWMM USE

-------
 through batch  processing  or  terminal (ISO).   SWMM output from this use-
 mode  can  be  examined  under ISO or can be mailed to users in hand-copy
 form  from the  Science Center.
      Both in-house and out-of-house users  can  obtain  all  appropriate
 grant services as shown in Figure 1.
 SWMM EXECUTION

      In order to  execute  SWMM at  UNI-COLL,  three data sets are required
 as  shown in  Figure  2.
     JOB CONTROL
      LANGUAGE
        CJCL)
                      SWMM Program
                       LOAD MODULE
 SWMM
OUTPUT
     INPUT
     DATA
                        Figure  2 .  SWMM DATA SETS
 Job Control  Language

     A Job Control Language (JCL) file is required to direct computer
processing.  JCL files can be prepared for specific users.  Modifications
to JCL files  can be made by users through dial-up terminals using the
simple IBM interactive language (TSO)-  This language is conversational
and may be learned quickly.
 SWMM Program

      The  SWMM computer  program is  written in FORTRAN IV and is construct-
 ed  in a  block format  according to  function as shown in Figure 3.

-------
           BLOCKS                      FUNCTION
                     m  models overland and gutter flow of quantity
                        and quality of stormwater
                       •routes storm or combined flow through sewer
                        system. Allows for dry weather flow,  infiltra-
                        tion and in-line storage

                      - models out-of-line storage and/or treatment
                        facilities

                      - models effects of storm or combined flows
                        on receiving waters
                     •  collates  outputs  from above  for  cumulative
                        modeling

                    -   produces  hydrographs/pollutographs
GRAPH
                 Figure 3 .  SWMM PROGRAM BLOCK STRUCTURE
The block  structure of  SWMM naturally  divides  the  required  data  inputs
as described  in  the section below.  The LOAD MODULE  (Ref. p.  7)  for
SWMM uses  approximately 325K bytes  of  core storage with OVERLAY  and
approximately 715K of core without  OVERLAY.  Data  files for SWMM use
can be prepared  as a joint effort between user and UCSC.  Collaboration
between users and UCSC  personnel is initially  required to define the
technical  problem at hand, and the  function and capabilities  of  SWMM
with respect  to  the problem.  Once  the function of SWMM is ascertained •,
the data requirements may then be outlined.  Potential sources for
required data may be provided by the Science Center.  After required
data is collected, users must reduce this data to  formats acceptable to
the SWMM program (either cards or coding forms), Users' Manual refer-
ence 5.  A summary of the types of  data for each block is given  below.
Runoff Block—
     The RUNOFF block of SWMM generally requires rainfall data and phy-
sical characteristics of the drainage basin to model overland flow of
stormwater and routing of overland flow through gutters and small pipes.

Rainfall input — A continuous record of rainfall over the period of
several storm events is a necessity for SWMM modeling, especially veri-
fication runs.  The recording rain gage or rain gages used for these  mea-
surements should adequately cover the watershed in question and provide
adequate account of the movement of the storm.

-------
Surface flow — The major data preparation task in SWMM use involves dis-
cretization of the watershed into subcatchments according to natural
drainage patterns and existing collection devices (gutters and storm
sewers).  The discretization level can vary for a single watershed from
coarse to fine grade depending on the required accuracy of results.  In
the finest sense, a subcatchment is an area that drains to a common point
with uniform slope, roughness, surface detention, and surface infiltration.
It is possible to use coarser discretization schemes providing verifi-
cation data are available.

Gutter and pipe flow — Outputs from surface (overland) flows may be routed
through gutters and small pipes within the RUNOFF section.  Pipes and
gutters are modeled by specifying length, size, slope and roughness.

Surface quality — Subcatchments are further divided into subareas having
a single land use.  SWMM recognizes five distinct land use patterns:
single-family residential, multi-family residential, commercial, indus-
trial, and undeveloped or park lands.  Quality of surface flows is com-
puted by SWMM from street cleaning data, catchbasin data, and the total
length of gutters in each subarea.

     A summary of RUNOFF block data requirements is given in Table 2.

                Table 2. RUNOFF BLOCK DATA REQUIREMENTS
     Data type
             Description
Rainfall input

Subcatchments
(each)
Gutters/pipes
(each)
Surface quality
(each subcatchment)
time-history of rainfall  (in./hr.) for each
time increment  (i.e., every 5 minutes)
area
width (perpendicular to flow direction)
imperviousness
ground slope
Manning's  roughness factor
retention storage
infiltration rate constants (Hortons  Eq.)
  maximum initial infiltration rate
  minimum final infiltration rate
  decay rate constant
width/diameter
length
slope
roughness factor

street cleaning frequency
antecedent moisture conditions
number of catchbasins, capacity, catchbasin
  solids and 8005,total length of gutters
                                    10

-------
     A connectivity  matrix between subc*tchments  and subareas,  and gut-
ters and pipes must  also be constructed so that flow from subcatchments
can be directed  through gutters  and pipes in proper sequence  to inlet
manholes for the TRANSPORT block.
Transport Block  —

     The SWMM TRANSPORT section  takes as input  the hydrographs  and pol-
lutographs generated by the RUNOFF model.  Manholes specified  in RUNOFF
as input manholes  are thus the input points in  TRANSPORT.  The  quantity
section of transport then routes these flows  through the sewer  system.
All  sewer elements,  manholes and pipes must be  specified for SWMM ac-
cording to type, length, size, slope, and roughness.  A connectivity
matrix must be constructed by the user to sequence flows correctly.
Pipes  are connected  by manholes, some of which  may be stormwater inputs.
Infiltration into  the sewers may be specified provided data on  infiltra-
tion levels are  available; otherwise, best estimates are used  to account
for  dry-weather, groundwater, and wet-weather infiltration.

Combined sewers  -  dry-weather flows — In the case of combined  sewer
systems, the quantity and quality of dry-weather flows (DWF) as well
as the daily and hourly variation must be specified.  Daily and hourly
correction factors for DWF can be obtained from treatment plant records
or measurements.   DWF quantity and quality for  each subarea may be either
directly measured, or, calculated by SWMM from  water use measurements
or population data on each subarea.  Contributing industrial process flows
must also be specified.

     Table 3 lists the data requirements for  the TRANSPORT  block of
SWMM.

               Table  3. TRANSPORT BLOCK DATA REQUIREMENTS

           Input (from RUNOFF) - RUNOFF hydrographs and pollutographs
               •ewer  elements - type (circular, rectangular, etc.)
                           length
                           •lope
                           Mannings  roughness
                           connectivity matrix
           Infiltration      - dry weather Infiltration (base estlnate)
                           ground water Infiltration
                           wet weather infiltration
           Dry weather flow   - measurements of DWF for each subcatchoont
           (DVT)                         or
                           •etered water use far each subcacchnent
                                        or
                           population  data for each subcatchment
                            I.e., area
                                population density
                                auober of duelling units
                                people In average unit
                                •arket value of average unit
                           Industrial  process flows
                           d»ilv and hourly flow variation for DWF
                                       11

-------
Storage/Treatment Block—

     STORAGE  and TREATMENT options  within SWMM can  be used to provide
for a number  of different types  of  storage and treatment at any point
in the TRANSPORT system.  STORAGE requires information about the
physical shape and dimensions of the storage unit as  well as hydraulic
information about flow within the storage unit and  the outlet method
from the storage unit.  TREATMENT requires specification of the treat-
ment alternatives to be included from among those modeled within  SWMM.
Costs of STORAGE and TREATMENT options can also be  calculated within
SWMM.

              Table 4. STORAGE/TREATMENT DATA REQUIREMENTS
      Storage
     Treatment
Physical dimensions of storage  unit
Type of outlet  device
Routing method  in storage unit
Unit costs

Specify treatment string
Unit Costs
Receive  Block —

     The major effort in SWMM modeling of receiving waters involves  ideal-
ization  into a system of channels connected by nodes (junctions).  Op-
tions  exist for inclusion of  tides,  wind and rain  effects, as well as
down-stream weir (dam) conditions.   Elevation, area, and Manning's co-
efficients must be specified  for junctions.  Width,  length, depth, and
Manning's coefficients are  required  for channels.   Storm water may be
input  at any mode from cards  as  well as from the primary transport out-
put  from disk file.  RECEIVE  data requirements are summarized in Table 5.
                Table 5- RECEIVE BLOCK DATA REQUIREMENTS
                   Junctions    -  water surface elevation
                                surface area
                                depth
                                Manning's coefficient
                   Channels     -  length
                                width
                                depth
                                Manning'« coefficient
                   Tides          cldal stage history
                   B»ln        -  precipitation ratrs
                   Heir        -  weir factor, elevation, power law
                   Storm Inputs  -  hydrographs-cards or disk file
                   Quality
                   constituents  -  pollutographs or loading rates for each constituent
                                      12

-------
     The preceeding data descriptions are intended only to indicate the
types of data required for SWMM.  For actual input data preparation,
the user is referred to the SWMM Users Manual.

     For verification and testing of SWMM models, measured runoff rates
and quality loadings taken continously over storm periods are required.
Measurements should be made at points of primary interest to the modeler
(i.e., outfalls, overflow points).  This verification data is especially
critical in modeling of quality constituents of storm or combined sewer
flows.
                                   13

-------
                             SECTION V
                    COSTS FOR MODEL APPLICATION

     Cost for modification, application and verification of SWMM models
may be divided into two parts, costs for' personnel and computer costs.
PERSONNEL COSTS

     Personnel costs include collection of physical data required for
model set-up, data reduction to actually build the computer model,
program execution and debug, and output interpretation.  Although
personnel costs vary widely based upon the availability of data and
the fineness of the discretization effort, a general rule of thumb has
been developed at UCSC to estimate personnel costs.  Three man-weeks
are allocated to physical data collection, discretization, data re-
duction, execution, debug and output interpretation for each basin
being studied.  Thus for an urban area with five distinct drainage
basins, fifteen man-weeks would be required for model set-up, verifi-
cation and application.

     As SWMM is utilized at UCSC these manpower tasks may be divided
between a primary user and UCSC.  For example, data collection and
discretization can be done in-house by the primary user, while actual
input setup, verification, debug, etc., can be done by UCSC.   Man-
power costs are thus divided between primary user and UCSC.
COMPUTER COSTS

     Computer costs include data file setup costs and verification and
application execution runs.  Typically,  setup of data files for the
initial run can be finalized for approximately $50.00.  Costs of execu-
tion runs vary heavily with the number of time steps, the number of
subcatchments, the number of sewer elements, and the portions of SWMM
to be run.  Typical costs for SWMM execution runs done at UNI-COLL
Corporation are given in Table 6.
                                  14

-------
Table 6, SWMM EXECUTION COSTS

Application
Binghamton,
N.Y.
Stony Brook,
N.J.
Wingoho eking
Phila. , Pa.
tt
SWMM
blocks
run
RUNOFF &
TRANSPORT
RUNOFF
RUNOFF &
TRANSPORT
RECEIVE
No.
of
time-
steps
68
64
60
2 days
No.
of
subcatch-
ments
36
20
57
NA
No.
of
sewer
elements
115
NA
120
NA

Cost($)
35.00
6.00
44.00
11.00
 NA:  Not Applicable
            15

-------
                          SECTION VI
                         CASE STUDIES

     Over the past year, six respondees to the dissemination program
described in Section III indicated a definite interest in using SWMM
on existing stormwater pollution problems.  Early discussions with
these potential users confirmed that much of the data required for
modeling had already been collected or was easily available.

     Extensive discussions were held with each of these potential users
and the function of SWMM modeling in the context of their particular
problem was defined.  Data requirements for the needed modeling were
generated.  In four of the six cases, continuous discharge and quality
data required for verification of SWMM was found lacking, and collection
programs to obtain all data were begun by the potential users.  Each of
these modeling sites is briefly discussed below.

     In the two remaining basins, discharge data was available but no
continuous quality data had been collected.  In both of these cases
SWMM modeling was done and is described in detail below.  Testing was
also done on a series of ten storms which fell on the Wingohocking
basin of Philadelphia.  Storm data was supplied by the EPA.
DATA COLLECTION PROGRAMS

     In each of the sites discussed below, all data required for modeling
with exception of continuous quality and quantity testing (for verifi-
cation of SWMM results) have been collected or are easily available.  In
each case the problem is similar.  Development of  suburban or near-rural
areas is occurring rapidly with little concern for water planning.  The
local effects of increases in runoff quantity and pollution as well as
the regional effects of development are being largely ignored, especially
in the case of stormwater runoff.  Local conservation groups and plan-
ning associations are attempting to sensibly control development and
need assistance in quantifying both local and regional impacts.
Upper Raritan

     The Upper Raritan Watershed Association (URWA) is concerned about
the effects upon the Upper Raritan River of a proposed AT&T administra-
tive office building.  The building is planned for a 50 acre site in Far
Hills, New Jersey,  and includes stormwater drains for buildings and
parking lots, and,  a stormwater retention basin.  The URWA is concerned
that the retention basin has been improperly sized and that  moderate
storms could have disastrous effects on the stream and community below
the site.  All necessary data collection, except site discharge, has
                                    16

-------
been collected and reduced to SWMM input.  Modeling is planned for the
site under natural cover initially, and then with the AT&T offices and
parking lots imposed upon the natural cover.  A study of varying intensity
storms could determine the maximum storm which the retention basin can
contain, as well as quality constituents of stormwater runoff as it is
drained to the Upper Raritan.
Sandy Run

     Sandy Run in Abington, Pennsylvania, is a tributary of the Wissa-
hickon Creek and has been chosen by the Wissahickon Watershed Associa-
tion (WWA) as a model subbasin for the Wissahickon Watershed.  Six sep-
arate storm sewers drain most of this subbasin.  Modeling of this basin
as it presently exists, followed by extrapolation to future development,
will form the basis for a land-use plan for similar watersheds in the
Wissahickon basin.  To date, data on one of the areas drained by one
of the storm sewers has been collected and examined.  Discretization of
the basin has been done and data on sewers collected.  Measurements of
rainfall and discharge for this area are in planning stages in the WWA.
Schoeneke Creek

     The Schoeneke Creek drains most of the city of Nazareth, Pennsyl-
vania, and empties into the Bushkill River.  Lafayette College in Easton,
Pennsylvania has done several studies on the Bushkill and its upper
tributaries and has routinely collected all necessary data for SWMM
modeling except continuous quantity and quality of discharge.  A heavy
suspended sediment loading is introduced to the Bushkill through the
Schoeneke and represents the first pollution of the Bushkill.  This
basin is especially interesting as it represents the edge of develop-
ment in the area.  The resources of both the Chemistry and Civil Engi-
neering Departments at Lafayette College are also available for testing.
Collection of required data is in the planning stage in this basin.
Rocky Run

     The Rocky Run in Lima, Pennsylvania, is a small stream that has
been heavily affected by development at the Penn State Extension at
Lima.  A large parking lot was built at the headwaters of this stream.
Over the past year, the increased runoff from the parking lot has sub-
stantially altered both the size and channel of the Rocky Run.  Severe
erosion and undercutting have destroyed trees and has resulted in heavy
buildup of rubble in downstream areas.  These effects have interested
both local residents and several faculty members at the Penn State
Campus.  A combined program of data collection between the Rocky Run
Watershed Association (RRWA) and faculty at the Penn State Extension is
planned to provide data required for SWMM modeling of the entire Rocky
Run Basin.  The primary purpose of this modeling is to avoid future
damage and form the basis of a stormwater plan for the area.  All re-
quired data for SWMM modeling has been collected or is easily accessible
except for continuous monitoring of quality and quantity of discharge.
                                    17

-------
     In each of the cases above, SWMM modeling is desired as a beginning
step in sensible stormwater planning.  All users will require moderate
assistance in finalizing data collection programs and in the early stages
of modeling.
MODELING ASSISTANCE

     In the Stony Brook basin in New Jersey and in Binghamton, New York,
discussions with parties interested in SWMM modeling uncovered required
SWMM data including continuous discharge monitoring.  No quality data
were available.

     Each application was carried through to initial verification studies,
These modeling efforts have resulted in, at least, a partial solution
to the existing stormwater pollution problem.  In the sections that
follow, a complete description of the application site, the application
of SWMM, results of the modeling, and future use of SWMM are given.
Binghamton, New York

     The city of Binghamton, New York, supports a population of 120,000.
The storm sewer system is partially combined with sanitary sewers, has
heavy infiltration in older sewers near the Susquehanna River, and allows
direct overflows of combined sewage at surcharge points along the river.
A new treatment plant with 30MGD capacity handles a daily dry-weather
flow (DWF) of 15MGD regularly and routes combined sewage overflows di-
rectly to the river even during small storms (Figure 4).

     The consulting engineering and design firm of Quirk, Lawler and
Matusky (QLM) has been contracted by the U.S. Army Corps of Engineers
and the City of Binghamton to produce a plan that would alleviate combined
sewage overflows.  A limited budget precludes rebuilding and separating
major sections of the combined sewer.  Possible alternatives include
increases in treatment plant capacity to alleviate plant surcharging and
storage of combined sewage during storm periods to eliminate interceptor
surchages.   Rebuilding of sewers is a possibility in areas known to be
subject to heavy  infiltration.
Study Objectives—

     Discussions with engineers at QLM indicated a need for establish-
ment of baseline data on surcharging.  Of key interest was the quantity
and quality (BOD) as well as location of sewer overflows into the Sus-
quehanna River.  Treatment plant records included continuous quantity
monitoring at the plant, but no measure of surcharging was made at over-
flow points.  In addition, neither the location nor the relative size
of overflow points was well characterized.  It was decided that SWMM
would be used to locate and determine the quantity and quality of over-
flows as the first step toward water resource planning for the City of
Binghamton.

                                    18

-------
                                                   SUBCATCHMENTS
                                                   t~*Trn\r v •*•*<••-.
        — wa_rx>*»A X^AAl'lJ
— —-- CITY LINE
	 SEWER LINES
     O DRAINAGE MANHOLES
     n OVERFLOW POINTS
Figure 4.  BINGHAMTON, NEW YORK -  SWMM DISCRETIZATION

-------
Study Development—

     A plan for quantifying the surcharge problem in Binghamton using
SWMM was developed.  The RUNOFF portion of SWMM was used to generate
inputs to the major sewer lines.  Only areas drained by combined sewers
were considered.  Accumulated dust and dirt  as well as sanitary flows
from the entire city were also modeled.  The TRANSPORT section of SWMM
was used to route combined sewage to the treatment plant.  At sewer
surcharge points, SWMM was modified to route surcharges directly to the
Susquehanna River rather than storing for later drainage.  With these
surcharges removed' from the sewer system, correlation of treatment plant
flows could be attempted.  Primary outputs were:  (1) quantity and
quality (BOD) of flow at treatment plant, and (2) location, quantity,
and quality (BOD) of sewer overflows.
Storm Inputs—

     Two storms were chosen for verification studies.  The first storm
totaled 1.25 in. of rainfall which fell on October 30, 1973, over a period
of 24 hrs.    The second totaled .25 in. and occurred October 2, 1973,
over 2 hrs.    In both storms, hourly rainfall data from the U.S. Depart-
ment of Commerce were interpolated to 15 minute intervals.  All calcula-
tions were made using 15-minute time steps.  Input hydrographs are given
in Figures 5 and 6.
Runoff Block—

     The city of Binghamton was discretized into 36 subcatchments as
shown in Figure 4 using topographic maps and aerial photographs.  Areas
drained by separate storm sewers were delineated, resulting in combined
connected subcatchments varying from 17.4 acres to 377.8 acres in size.
Gutters and small pipes were ignored.  Only overland flow was used to
route stormwater to the outlet manhole for each subcatchment.  Imper-
viousness of connected subcatchments was established from aerial
photographs resulting in variations of 12 percent to 80 percent imper-
viousness.  Slopes were calculated from topographic maps.  Default
values were taken for Manning's coefficients, detention depths, and
infiltration constants.  Subcatchment data are given in Table 7-  Land-
use types were determined by examination of aerial photographs.  For
each subarea, curb lengths were estimated using the method of Graham
et al.10

     It was determined by these researchers that:

          C=423.7-420.8  (.8797)p

          Where:  C=specific curb length (ft/acre)

                  P=population per acre                           (1)
                                   20

-------
     i.a >	1	1	1
        •.«   10.1    U.J    l«.l
                                  ll-l I.. HOLXS
         Figure 5.  BINGHAMTON,  NEW YORK -  STORM 1-HYETOGRAPH
UI Will.

 IN

IN / m

        i.o    i.»    «.i    I.T    4.1    T.J     >.«

                                  TIKf IN MOWK
                                                      10.2   U.I    12.0
          Figure 6.   BINGHAMTON, NEW  YORK -  STORM  2-HYETOGRAPH
                                   21

-------
                   Table 7.   BINGHAMTON,  NEW YORK - SUBCATCHMENT DATA
SJBAREA
NUHBER
1
2
3
4
5
&
1
13
8
9
10
11
12
20
21
22
23
24
25
26
27
28
29
30
31
32
33
43
41
42
43
44
45
46
47
48
GUTTER
rfUTH
OR MANHOLE (FT)
51
52
53
54
55
56
57
63
59
59
60
61
62
70
71
72
73
74
75
76
77
70
79
B3
81
82
83
90
91
92
93
94
95
96
97
93
2355.
3335.
2)33.
3335.
30J3.
4030.
4 )33.
2033.
4 6 •> •) .
6 6 i 7 .
2!>i3.
2330.
26^8.
23)3.
23JO.
3335.
53J6.
3315.
2668.
6)33.
43)3.
4030.
26'jR.
46i9.
6330.
16i3.
3609.
6333.
333a.
2333.
4330.
26a8.
4336.
3033.
7137.
2603.
AKEA
(AC)
83.5
47.6
36.7
101.6
19.1
203.2
95.2
17.4
142.6
164. 3
2i1.4
25. 1
82.7
46.2
27.4
72. 1
166.3
51.3
211.9
120.4
20d. 8
187.0
66.9
!>*.'*
256.9
19.0
86.1
377.8
28J.4
134.5
33.2
93.1
124.6
111.5
119.4
73.1
PF^CF.NT
I'lPERV.
23.0
15.0
30.0
33.0
12.0
25.0
35.0
73. 0
30.0
20.0
13.0
13.0
20.0
40.0
20.0
53.0
40.0
50.0
50.0
90.0
30.0
30.0
20.0
1" . 3
40.0
40.0
60.0
20.0
60.0
30.0
40.0
50.0
73.0
80.0
40.0
35.0
SLOPE
(FT/FTI
0. 1330
0.0710
0.0120
0.0000
0.0310
0.2250
0. J190
0.0', 10
0.3560
.0. )430
'0.2500
0. 1390
0.3630
0.34BO
0.01 40
0.31 30
0. J?. 30
0.0270
0.0520
0.0260
0. 0560
0.3330
0.0630
0.3HO
0.0600
O.J430
0.03^0
o.otoo
0.0030
0. 3030
0.00.10
0.0010
0. 3150
0.0160
0.3140
0. J140
RFSTSTANC
\ y p p p y ,
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
P. 013
1.013
0.013
0.013
O.C13
0.013
0.013
0.013
0.013
0.013
0 . (1 1 3
0.013
0.013
P. 013
0.013
0.013
0.013
0.013
0.013
0.013
0.013
E PACT3S
PLP.V.
0.250
0.250
0.250
C.250
0.250
0.250
O.?50
0.250
0.25')
0.250
0.250
0.250
0.250
0.253
0.250
0.250
3.250
0.250
0.?53
0.250
0.250
0.250
0.2'J3
0.250
0.250
0.250
0.250
0.250
0.250
0.250
0.250
0.250
0.250
0.250
0.250
0.2 50
SIJOFATF ST
IMPCPV.
0.062
O.P6?
0.062
0.062
0.067
0.06?
0.062
P.P62
0.052
0 ."6?
P. 06?.
n.oft?
0.062
0.06?
O.P62
O.Pfc?
0.36?
P. 062
P. 06?
0.062
O.Of.2
0.062
P. 06?
O.C62
0.062
P. 06?
P. 06?
0.06'?
0.062
0.06?
0.062
0.06?
P.. 162
0.06?
P. 362
0.36?
r)RACr( INI
Pf.RV.
0. 184
0.1B4
0.134
0.104
0.1 H4
0.1P4
o. i 114
0. 134
0.1 P.4
o.i r>4
J.lf!4
0.1R4
0.184
0. 1B4
0 . 1 B 4
0.1 C4
0.1(34
0. 104
0 . 1 .•' 4
0.1P4
0.184
O.lfi4
O.lfi'i
O.lfl4
o. ir.4
0. 1 14
0.1 84
0.1 J4
0.114
0. lf)4
O.I 84
O.I 84
0. 184
0. 134
0. 184
O.lf)4
\I;F |U
MA XI '-Mil'
3.00
3.00
3. CO
3.00
3.00
3. CO
3. no
3.00
3. no
3. OP
3.00
3. no
3. no
3.00
3.00
3. no
3 .no
3. OP
l.PO
?..PO
3. no
3. no
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3. CO
3.00
3.00
3. CP
?.CO
3.00
3.PO

M ! N!I MUM
0.5?
0.5?
0.52
0.5?
0.52
0.52
0.52
0.52
0.52
0. 5?
0.52
0.52
0.52
0.52
0.5?
0.52
0.52
0.52
o.-;2
0.52
0.52
0.52
0.5?
0.52
0.52
0. 52
0.5?
P. 52
0.'J2
0.52
0.52
0.52
0.52
0.52
0.5?
0.52
AT(= ( IN/MB)
DECAY RATE
0.00115
3. 00115
0.00115
0.00115
0.00115
3.001 15
0.001 15
O.PP115
o.nni 15
o.noiis
0.00115
0.0.31 15
0.00115
O.P01 15
0.001 15
O.C01 15
0.001 15
0.00115
0.00115
0.00115
o.oni 15
0.00115
O.P0115
0.001 15
i). 'onus
0.001 15
0.3T115
0.00115
O.P0115
0.001 15
O.OPU5
0.00115
O.P01 15
O.nnj 15
I'. 00115
0.00115
GA.,
MJ
1
1
I
1
1
1
1
1
1
i
1
1
I
I
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
I
1
1
1
•1
1
1
TOTAL NUHBER OF SUDCAtCHIENTS.  36




TOTAL TRIBUTARY AREA (ACrtESI, 41B4.40

-------
     Population densities of each subarfta were obtained from census data
and specific curb lengths were calculated using formula (1).  The
RUNOFF portion of SWMM was run for both storms to produce input for
TRANSPORT.  Sample subcatchment hydrographs are given in Figures 7 and 8.
Transport Block—

     The major combined sewers in Binghamton were modeled using the
TRANSPORT section of SWMM.  A total of 115 sewer elements, including 58
manholes and 57 sewer pipes were included.  Shapes, lengths and slopes
were obtained from sewer maps for each pipe.  Default values were assumed
for Manning's coefficients.

     The TRANSPORT section of SWMM was altered slightly to route surcharges
from the sewer system to the Susquehanna River.  In subroutine TRANSPORT,
surcharges in each sewer element were set to zero in each time step after
printing surcharge quantity.  SWMM was relinked with this modification
and a LOAD MODULE for use with this project was produced.  Engineers in
the city of Binghamton estimated total infiltration of 1,500 gpm into
the combined sewer.  This level was used for all calculations.  Using
subareas and population densities from RUNOFF, and adding 9 cfs DWF from
nearby communities, a total DWF of 25.03 cfs (16.6 MGD) was calculated.
This figure correlates well with the measured average DWF of 15MGD.
Hourly DWF correction factors were obtained from treatment plant records.
Apportioned infiltration, sanitary flows, and total DWF are given in Table
8.
Results—

     The TRANSPORT section of SWMM was run using output from RUNOFF as
input for both storms.  Output hydrographs at the treatment plant are
compared to plant records in Figures 9 and 10.  Correlation with measure-
ments for the 1.25 in. storm was excellent.  However, the smaller storm
produced poorer correlation between calculated and observed plant flow.
A possible explanation for this poor correlation is that overflow occurs
continually at problem points rather than simply during pipe surcharge.

     For the large storm, quantity and BOD loads for overflow points are
given in Table 9.  Locations of overflows are given in Figure 4.  For
this storm, 29.9 MG of a total runoff of 50.9 MG was discharged directly
to the Susquehanna.  Of a total BOD load of 25,000 Ibs. , 16,000 Ibs.
were discharged directly to the river.  Specific quantities of BOD
discharge were determined by calculating total BOD load at the manhole
upstream of a surcharge point and multiplying this load by:

          (total discharge at point/total flow at upstream manhole) =
          discharge BOD at discharge point.
                                    23

-------
      loe.oco
          ••••••••••••—'
          J.O    1.*
                          5.1
                                ».»
                                      T.J
                                      IN MM»

1.4    9.1    10.I   11.1    12.0
Figure  7.   BINGHAMTON,  NEW YORK - STORM  1-SUBCATCHMENT 36 HYDROGRAPH
               10.1    12
Figure  8.   BINGHAMTON, NEW  YORK - STORM 2-SUBCATCHMENT  36 HYDROGRAPH
                                    24

-------
                         Table 8.  BINGHAMTON,  NEW YORK - INFILTRATION AND DRY WEATHER FLOW
                                      .. WA«m
                                                       of  a » f
                                                                  EACH SJBMEA _________
M?-M) • IIT'I."
A1S
_ ._ A1CJI
AUK
KMIt INPJf IHI- »
CF*
S • 1420.1
1 • 2.00EI
F « «.;
	 l*:.ll__a_
CFS
>OL8$PI:*nA»
)0 IDSPFRO/
>ll *V1/UK\
!8 CFS
JWiv-tf 	 W
crs
'/CFS
kY/CFS
r rf H CAP! T»

. ANJ D
7
9
19
II
12
11
2J
21
22
24
25
26
27
2B
T9
30
31
33
43
42
*5
47
46
SI.
52
S3
5*
55
5/
59
to
61
63
7.)
71
7J
74
75
76
7f
11
7)
*1
fl3
9J
11
92
93
95
97
9d
0*1 i •
O.O'i
3.2»
3.52
0. 14
o. ri
3.51
J. I 4
J.J?
5./U
0.,)V
3.12
0. 11
0.65
0. 10
U.13
0. Jl
).-)!
0.0
J.7J
0. J9
0.2J
3.5J
•).'•>
0.3
j.JI
J.40
J.33
0.'35
0. Jl
(1.33
0.17
0. J5
0.3f
0.26
•J.lft
3. J6
0.01
2. 17
0. Jl
0.04
0. )2
•3.16
0.23
0.04
0.35
0. 11
O.JO
1.0
0.20
0. JJ
0. 39
0.21
.1.20
0. )
.3.16
.-•. 3
J. )7
3. 15
0.11
J.17
0.05
0.30
0.71
0.19
0.20
d.-)7
4. 11
0. 60
0.21
0.13
7.01
0.05
0.17
0.16
j'.u2
O.riU
0. 14
1.33
0.42
0.02
J.12
0.0
0.97
0.13
0. 35
3.79
0. 74
0.0
0.62
0. •)
0. 11
U.26
0. 55
0.41
0.11
0.13
0.20
0.13
0. 19
0...4
3. 17
o.-o
n.n
0. 12

0.1-1
0.14
0. 1 3
O.'il
0.09
0.-I9
0.01
O.'lll
0. ,3
O.Jfl
0.23
O.VJ
4 0. J
I O..I
k 0.)
2 0. )f
2 ') . 1 7
1 0.36
1 0.27
0.12
3.04
0.22
O.SI
0.1V
0.20
o. ro
3.46
0.43
J.13
O.02
5.63
0.34
0. 12
O.O-i
0. 1'.
0.45
0 . 64
0.10
0.9(>
0.33
0.01
0.011
0.3
0.71
0.09
0.2>
0.57
0.54
0.0
0.45
0.0
0.03
0.11
3.43
0.30
                           KIIALS
                               H.43
                                       6.63
                                              2!>.03
                                                          24r/."22 ITJ 272.2n.TT5   r?*

-------
FLCW




 IN




 MGD
                                                                   i   O  PLANT RECORDS
         20
         10
               4 AM   5
9   10    11    12   1 PM   2    3    4




  TIME IN HOURS
                 Figure 9.  BINGHAMTON,  NEW YORK - STORM 1-OUTFALL HYDROGRAPH

-------
NJ
           FLOW
           MGD
50

AO

30

20

10

 0
                            *  SWMM
                            O  PLANT RECORDS
                           5AM   7     9    11    IPM   3    5    7     9    11    1AM    3    5     7    9
                                                       TIME IN HOURS
                                Figure 10.  BINGHAMTON, NEW YORK - STORM 2-OUTFALL HYDROGRAPH

-------
         Table 9.  BINGHAMTON, N. Y. - STORM 2 OVERFLOW




                        QUANTITY AND BOD
Sewer Element
No.
124
133
135
119
122
123
152
114
153
172
144
125
151
117
Overflow Volume
(mil. gal.)
3.1
3.0
2.75
2.7
2.7
2.4
1.85
1.8
1.7
1.6
1.5
1.3
1.15
1.0
BODc
UbsJ
1550
1140
1030
2100
3200
1500
790
1100
730
790
710
150
440
850
Total
29.9
16,000
                               28

-------
     Although this computational method is not precisely correct, it
produced the best possible upper estimate of discharge BOD levels.
Discussion—

     The overall result of this study has been to quantify, for the 1.25
in. storm, the location, amount and BOD load of sewer surcharges.  This
represents the first attempt to determine the severity of the pollution
load introduced to the Susquehanna by these combined overflows.  The
calculated severity of overflow pollution represents a major driving force
in establishing programs to alleviate this problem in Binghamton.

     SWMM will assist in planning to abate combined sewer overflows in
Binghamton.  Three pollution abatement schemes have been suggested to
solve the Binghamton problem.  The first scheme involves construction of
external storage devices at each overflow point to store surcharges
during wet periods.  Existing interceptors will be used to route storm
flow to the treatment plant after the storm periods subside.  SWMM will
be used to size and determine cost of this storage scheme.  The second
abatement scheme would utilize separate, additional treatment facilities
to handle each of the combined overflows, and would be located as near
as possible to the overflow points.  Treatment methods suggested for
scheme II include microstrainers or dissolved air flotation with chemical
addition.  SWMM will be used to determine removal efficiencies and costs
of this scheme.  The third abatement scheme considers use of a centralized
treatment plant specifically for overflows and an appropriate interceptor
system to route overflows to the treatment facility.  Again SWMM will be
used to predict pollutant removal efficiencies and construction costs.

     A model of the receiving waters will also be constructed to deter-
mine effects of untreated overflows and of storage or treatment efflu-
ents upon the susquehanna River.  Measurements of quantity and quality
of overflow in Binghamton are presently being made in order to verify
the SWMM models which will be built for this planning stage.
Stony Brook

Introduction—

     The planning commissions of Princeton, New Jersey and Hopewell
Township, New Jersey, have contracted a group of graduate students at
the Department of Regional Planning at the University of Pennsylvania
to produce a land use plan for the Stony Brook basin in New Jersey.
Part of this work was a water resource planning study including quanti-
tative modeling of stormwater flow in the basin.  The use of SWMM was
investigated to accomplish this modeling.
                                    29

-------
     The Stony Brook Basin is approximately 90 square miles of largely
rural land, with the only sizable population center in Princeton, New
Jersey.  Two gaging stations have been established on the Stony Brook,
one at the mouth of Baldwin Creek, a tributary to the Stony Brook,
the other on the Stony Brook itself near Princeton (Figure 11).
Study Development—

     The general plan of this study was to break the Stony Brook Basin
into subbasins, one of which would be Baldwin Creek.  The RUNOFF portion
of SWMM could then be used to model overland flow in the Baldwin Creek
Basin.  It was anticipated that some parametization of RUNOFF would be
required to produce correlation with gaging results.  Once a parameti-
zation of RUNOFF for Baldwin Creek was obtained, this modeling method
could be extended to the other subbasins of the Stony Brook.  Output
hydrographs from each subbasin would then serve as input to a RECEIVING
WATER model of the Stony Brook itself.  The overall model could then be
verified by comparisons with the Princeton gaging results.  At first
it was thought that suspended sediment measurements were available for
both Baldwin Creek and Stony Brook.  These measurements proved to be
only daily recordings.  Extrapolation of SWMM quantity results to
future development levels could then be accomplished in two ways.
First, a general development level in the entire basin or in specific
areas of the basin could be modeled by appropriate increases in im-
pervious cover.  Effects of this level of development on stormwater
quantity and quality could then be 'assessed.  Second, specific develop-
ment sites could be modeled to determine effects or proposed buildings,
parking lots, etc.
Storm Inputs—

     The first step in this application of SWMM was modeling the Baldwin
Creek subbasin.  To minimize effects of temperature and general ground
water conditions, two storms were chosen from roughly the same time
of the year - the summer of 1973.  The first storm totaled 4.42 in. of
rainfall.  The ten days prior to this storm were very dry.  The second
storm (1.83 in. total rainfall)  was preceded by a moderately wet ten-day
period.   It was anticipated that different infiltration-rate parametiza-
tions would be required to model these storms.  Hourly rainfall data
from the National Weather Service were interpolated to 15 minute inter-
vals for each storm.  Hyetographs are given in Figures 12 and 13.
Runoff Block—

     The RUNOFF portion of SWMM was used to model both overland flow
and small-stream flow.   In some sections of the subbasin, subcatchments
were delineated with two greatly different ground slopes.  No stream
rose to drain the high slope area so that overland flow occurred from
the high slope area to the low slope area before collection by streams
                                   30

-------
                                                              Subbasin Boundary
                                                 — — — — — —  Subcatchment Boundary
                                                 •• in !•*!•• i  Streams
                                                         Q3)  Subcatchraent
                                                         129   Stream
Figure 11.  STONY BROOK-BALDWIN  CREEK

-------
     -1.400
  IN
     0.100
      0.4JO
      0.0
        10.0    12.3    14.S    16.1     19.0   ' 21.3    23.5    2S.«    21.0    30.3

                                 '       fj«~ IN HOURS
    Figure 12.   STONY  BROOK-BALDWIN CREEK -  STORM  1-HYETOGRAPH
•A1NFALL
                    *   ••
                   *     *
                   *     •
                                  »  •       *   *
      "-"
                                     ,1	1.
          7.0     7.3
                              13. «    16.0
     Figure  13.   STONY BROOK-BALDWIN CREEK - STORM 2-HYETOGRAPH
                                    32

-------
began.  In order to model this physical situation, the subcatchment was
broken into two subcatchments.  Runoff from the high-slope area was
drained to a "gutter" having the geometry and slope of the low-slope
area.   Output from this gutter was then combined with runoff from the
low slope area.  This method effectively accounts for runoff over two
widely differing slopes with no collection device; it effectively retards
the hydrograph from the high slope area.  Small streams in the Baldwin
Creek Basin were treated as idealized channels within RUNOFF block.
Overland flows from all subcatchments were combined and routed through
channels to the gaging station.  The Baldwin Creek Basin was discretized
into 20 subcatchments using topographic maps; idealized gutters were
utilized to route storm flows.  First efforts (1.83 in. total rainfall)
with relatively low Manning's coefficients for both subcatchments and
streams produced hydrographs which were high in peak discharge and early
in peak runoff.  Adjustment of Manning's coefficients upwards produced
reasonable correlation with gaging results (Figure 14).  Default values
were kept for infiltration coefficients.  Tables 10 and 11 contain sub-
catchment data for this subbasin.  These values for Manning's coefficients
were then used to study the other storm.  Initial results showed ex-
aggeration of runoff peak.  Infiltration constants were varied until
good correlation was obtained (Figure 15).  The project staff recognized
that the parameter sets used to obtain correlation do not necessarily
represent the only possibilities.  Other combinations of Manning's co-
efficients and infiltration-rate coefficients could produce good results.
The two variables are heavily interconnected since, for example, an in-
crease in Manning's coefficient retains stormwater on land surfaces
longer, allowing more time for infiltration.  Lacking measurements on
infiltration rates for the basin at differing antecedent moisture con-
ditions, these coefficients could only be taken as variables.  The
results obtained and the  method used to obtain them is simply a logical
and effective procedure - it works.

     Correlations of storm discharge at other times of the year (winter,
spring) were very unsuccessful, because of drastic variations in infil-
tration rates due to groundwater or temperature variations.  In addition
to a verified model of this subbasin, a healthy regard for the dependence
and interdependence of RUNOFF results, roughness coefficients, and
infiltration coefficients was obtained.
Discussion—

     The best parameterizations for the Baldwin Creek subbasin were ex-
tended to the rest of the Stony Brook Basin.  Thirty-four separate sub-
basins were identified and discretized.  Outflow hydrographs were
generated with RUNOFF block.  Outfall hydrographs for all sub-basins
in the Stony Brook are available from the authors for further examination.
                                    33

-------

                                                         3z;r
Figure 14.   STONY BOOK-BALDWIN CREEK - STORM 1
                 OUTFALL HYDROGRAPK
160.000




no. ADO

CUWWF
ii
CFS
§5.000



J
41.090




0.0
« • X GAGING STATION
« Q
• •
„' '. * SWMM
• V*\
• / * \
/ * \
1 • \
r ' \ ••
/ . v . .
• \ • •
• \ • •
• \ • •
l • \ • •
/ \
/ « \ « «^w^
* / * ^v * /^* \
* /^ * * ^^--tf'^ * \
*/ * • « \
> . . . V
/ • . . \
/ '. .' : \
/ : : : \
/ " : \ '
I : \
/ : ^
i * 'V
X « • \
/ « * T<- 	
/ "
/ . ...
.«..> */•..!.• •» 	 | 	 | 	 | 	 | 	 | 	 | 	 | 	 .............
 1.3    II. •>    U.
                    l*.0    lit. )    M.S
                       r»S"ls
                                             ii."    J7. 3
Figure 15.   STONY BROOK-BALDWIN  CREEK - STORM  2
                 OUTFALL HYDROGRAPH
                      34

-------
                      Table 10.  STONY BROOK-BALDWIN CREEK - SUBCATCHMENT DATA-STORM  1
OJ
Ul
                      Table  11.   STONY BROOK-BALDWIN CREEK - SUBCATCHMENT DATA-STORM 2
S HASP t
'; j " - f »
l
2
3
4
'i
I
7
0
1 1
11
n
i •>
is
1 1
1 ^
n
2J
MITTf
OR •' A "Jl 1 It 6
1 1 1
112
113
115
lib
1 1 7
in
i n
i •»
i n
i ?i
1 'S
1 '(.
1 ?7

1 14
»l )'l H
I'll
I I. '2.
Uift.
7 >•>..
I'o2. 	
lit -I!
1 J-l.
! / 1 '. .
IS','..
f\l'\
n:>6.
2/.0 	
la .O
2^ .0
2j .u
Ul . J
Ul .0
5J.O
1 .J
2< .0
2j.J
72 .0
• VI , 0
I/ .5
21.0
20J . J
t)j .3
12u.O
0.0
0.0
0.0
0.0
o.o
b.o
O.-l
0. U
0,0
0.0
0.0
0.0
0. .)
0.0
0.0
:',!. '.-IT
" ISr/i-TJ"
'• . 1 '. "0
• >. ?0 10
<). 
0.012
0.013
0 . C I 3
•J . 0 I :•
O.C13
• o.ni ?
0.013
0 . 0 I i
0.013
0.013
I'.Ol '
o.r-i:,
0.013
V:C;F _rArir:
\/ . l>r i' V
0.7.^0
0. 7 (So
0.7 in
O.f.00
O.OOQ
0.7PO
0.400
o.n oo
O.'jIV)
. o.70(^
O.'iOO
0. JOO
o.r>co
0 . i ft )
C . 7 00
O.IT-?
'0.0>,2
O.OA?
0.0 /x'
o.or.?
O . 0 '> ?
~ 0.0/S2 ~
o.o •',?.
O.()ft2
0.o/,:>
0.0.-.2
P. "f..1
o.r.c,?
n.o-,?
o . n fr t
O.Ofi?
0 . 0 /i 2
0.;i,S2
0.0 r>2
.W ( IN) 	 .!
0. l"/i 	 1.
0.1 ?'• 1.
0.104 1 .
0.1P'. I.
0.1,14 1.
0.104" ' " 1.
0 . 1 .14 1 .
0.1 114 1.
o.l'V, I.
D.I '14 I.
0 . 1 H 4 1 .
0.1 114 1.
0 . 1 f)4 I .
0.104 I.
0.194 1.
0.104 1.
'•,0
50
r.o
in
50
50
50
50
SO
50
50
50
50
50
50
0.25
0.25"
0.25
0.25
0.?5
0.?S 	
n.?5
0.?5
o.?s
0.?5
0.25
0.25
0.25
0.25
0.25
0.25
rc IV/HU i r. iof
r AV ' •- ATP :j"o
0.0l!ll5___l 	
o.«oiif, i
o . n .) 1 1 T> i
0. 00.11 S l
r».n;,iis i
O.'"llr. 1
0.00115 I-
o.ooiin 1
0.00115 1
0.00110 I
0.00115 1
0.00115 1
r. onii5 i
0.00115 I
0.00115 \
0.00115 1
0.00115 I
0.00115 1

-------
     A model of the Stony Brook main stem was constructed with junctions
located at subbasin outflows and channels connecting junctions.  At-
tempts to run the quantity sections of RECEIVE block using hydrograph
inputs have been thwarted, however, by instabilities resulting from
short channel lengths and relatively long time steps.  The standard
fixup in SWMM for this instability is to increase channel length.  This
modification is impossible in this case because of the fixed junction
location.  Presently, changes in time-step size and SWMM print-out
techniques are being investigated to alleviate or "work around" this
problem.

     Once the Stony Brook model is completed and verified with Princeton
gaging results, the planning commissions in the Stony Brook Basin will
be given a description of the model, its results, and its future uses.
They will then decide whether to use the model in future water resource
planning.

     The work group at the University of Pennsylvania will strongly rec-
ommend continued model use.   Technical assistance will be required both
to extend the model to future development levels and to refine the model,
especially in the Princeton area.
Wingohoeking

     EPA provided the Science Center with rainfall and runoff data for
a series of sixteen storms which fell on the Wingohocking basin of
Philadelphia, Pennsylvania between 1964 and 1968.   Data from ten of
these storms was used to test SWMM and to provide experience in SWMM
modeling to Science Center personnel.
Storm Selection—

     Of the sixteen storms for which rainfall and runoff data were pro-
vided, two had rainfall input for only one of the recording gages in the
watershed, six storms had some combination of two or three of the four
accessible gages, and eight had data for all four gages.  In order to
use the storms with two or three gages, reapportionment of the Thiessen
polygons  used to assign rainfall to subcatchments would be required.
Therefore, the ten storms with either data from one or all four gages
were included in this study.  Rainfall data was then reduced to the
SWMM input format in preparation for execution of RUNOFF block (Refer to
P. 9).
                                    36

-------
Runoff Block—

     In order to apply  the RUNOFF block  to  the  rainfall  data,  the Wingo-
hocking basin was discretized by the  same method used  in Reference  6.
A brief investigation was made  into a discretization scheme developed
in a test of the British Road Research Laboratory  (RRL)  method8 but
this proved to be too fine in detail, resulting in  too many sub-areas
and sewer elements for  present  SWMM matrix  dimensions.   Since  the only
runoff data available was taken at the sewer outfall,  no correlation of
intermediate runoff results was possible other  than the  final  results
of TRANSPORT.
Transport Block—

     In a manner analogous to  that used in RUNOFF block calculations,
the sewer system for the Wingohocking basin was modeled exactly as in
Reference 6.  Input from RUNOFF block was routed through this sewer
system, resulting in the outfall graphs in Figures 16-21.  On the same
graphs is given the time-history of actual runoff as obtained from the
Philadelphia Water Department^.  These runoff results are corrected
for an upstream dry-weather flow (DWF) interceptor and an estimated
DWF of 100 cfs.*  Table 12 gives actual and calculated total runoff
volume for these studies.

     Input data for the first  two storms is of poor quality.  A single
rain gage for a variable storm on a basin as large as Wingohocking is
insufficient.  The last two storms have runoff data which is suspect.
Runoff/rainfall ratios for the last two storms are much higher than
the rest of the storms.  Discussions with Philadelphia Water Department
engineers" has substantiated these suspicions.  These engineers also
feel this data is suspect and  suggest ignoring these two storms.

     Therefore, although results for two of these four storms are good,
SWMM cannot be reasonably expected to produce good correlations with data
which may be of low quality.
Results—

     SWMM gives uniformly good results for five of the other six storms
with calculated runoff tending to be slightly early and slightly larger
than reported values (Figures 16-21).  The results for Storm #5 stand
then as the only poor result which is not easily explicable.  It is
interesting to note that this storm produced the lowest runoff ratio
(0.35) of those studies.  Considering quality of data, size of the
basin (5400 acres), and the coarseness of the discretization (57 sub-
catchments, 129 sewer elements), these results are considered to be
quite good.
^Estimated DWF by Philadelphia Water Department Engineers, SWMM calcu-
lated DWF of approximately 15 cfs.

                                    37

-------
     Another point of interest is the disparity between the rainfall
hyetograph used in Reference 6 for the July 3, 1967 storm and that made
available to the Science Center for the same storm.  A comparison of
these two hyetographs (Figures 22, 23) shows they are similar but not
identical.  Tracing this data back to the original recording graphs  »
it was found that the data made available to the Science Center were
correct.
Storage Block—

     TRANSPORT output from Storm #4 was used as input to the STORAGE
block of SWMM.  Storm //4 was similar in magnitude to the storm of
August 3, 1967, which was studied in Reference 6.

     A storage and treatment package similar to that used in SWMM was
setup and the pumping rates (700cfs) and start/stop (14ft/9ft)
levels in the storage system adjusted to hold the entire storm.  Using
this package, 51 percent of the BOD, 47 percent of the suspended
solids and 19 percent of the coliforms were removed.  A minor error
in the STORAGE section of SWMM was detected and has been discussed
with SWMM programmers at the University of Florida.
                                  38

-------
Table 12.  WINGOHOCKING - SUMMARY OF




                 SWMM TESTING
VOLUME x 106 CU. FT.
STORM
1
2
3
4
5
6
7
8
9
10
DATE
4/27/64
4/29/64
11/25/64
8/04/65
10/07/65
7/02/67
7/29/67
8/09/67
7/24/68
8/01/68
SWMM
13.5
10.8
19.8
14.4
14.5
18.8
16.7
17.9
22.8
15.1
ACTUAL
11.6
14.1
12.7
10.3
8.2
21.7
15.2
12.5
32.0
23.1
%
DIFFERENCE
16
23
56
40
77
13
10
43
29
35
                 39

-------
 ieoo-.oo>i
                                                 CALCULATED RUNOFF
                                                              ll.r.    11.2
           Figure 16.   WINGOHOCKING - RUNOFF FOR STORM 3
 H.JM
 IN
eft ~
   •co.ooo
   430.000
        9.)     J.3
                                                                17.U     19.1
           Figure 17-   WINGOHOCKING - RUNOFF FOR STORM 4
                                   40

-------
   1400.000 -
   1200.000
 Km

 M

en
    •OJ.OOj
    400.003
                                   ..  CALCULATED RUNOFF
                                   •  f
     0.0  1"
         0.)
                                                          MEASURED RUNOFF
                 1.6      3.0      *.«.     5. •      7.2      »-6    Iw.U

                          lint IN HOCK]
           Figure  18.   WINGOHOCKING -  RUNOFF  FOR  STORM  5
    ICOO.OOJ
     Itoo.ioa
 Cfi
     iceo.aoj
     soo.ooj
    _ 0.0
           O.1     ».»     «.»
                                                                u.i.     ig.7
                            Tint IN
            Figure  19.   WINGOHOCKING -  RUNOFF FOR STORM  6
                                       41

-------
   3000. CO* -
 now
 IK
CM
   tioo.ooo
   loou.doo
.CALCULATED RUNOFF
                         MEASURED RUNOFF
     0.0.
         0.2      2.J      4.4     t.i     «.>     10.7     U.b     U.V     17.0
                          TI« IN HOUkS

           Figure  20.   WINGOHOCKING  - RUNOFF FOR  STORM 1
           i
    3000.000 -
 IK
 CM
    2000.000
    1000.000
  	0.0.
                                 •\_v/^    NV..  ••«"".
                                        CALCULATED RUNOFF
          O.I      2.1
                                         7.4      9.7     11.»     11.1     14.»
                           Tint IK
           Figure  21.   WINGOHOCKING  - RUNOFF FOR  STORM 8
                                       42

-------
i.ooe •





l.»CJ


•lf*FAU.
IN
l.V / Ml


1.0C3







0,-SCS

•
•
IX
XX «MKG*G{. ICGEND
XX
XX
XX
XX
xx
XX
X1
X*
»•
X X
X X
' XX X
XX X
XX X
XX X
X' X
X« X
X » X'
X » X'
» • » X1
» X' • XX1



.'
•
•
•
t
,'
1
1 1
• »
• •

t
I « • 2 - *



I
*
»
*
*
»
t •
1 •
t •
v *
« 1
• •»












X X
A X

v- « t . i • •' >. • • X X* • • • • • •


21.1    2*.2    25.0    25.4    It.7    27.4    2D.3    29.2    30.0    30.8     31.7
                              T1H£ IK HOUhS
       Figure 22.   WINGOHOCKING - UCSC  HYETOGRAPH
                               FOR STORM  6
J.JW







1.5 JO

»AI SftlL
I'l

IN / HI

1. J30






. 0.^00

O.G
24

;
t
XX
XI'
xx«
IX'
X'X
x'X
X'X
X'X
X'X
X'X
X'X •
X'X*1
X X • X •
XX X'X'
XX X'X'
X« X"X
X X" «
< X'X
X -l» «. '«
X "\ • ' • «


. •) 24.4 2i.'<



1
t
»
•
t
t
1
1
•
• * *
• # •
• * •
,
*
•
•
"
• 4
•> * *
* •


t^.t

-
t
I . • 2»* 1 • •

•*
•I
-

*
"*
•
-
•
•
,-
t
*
•t
X
) * t
» A '. •
*,;; r*
- VS Kl*
•4;^..
^•.j 26.1 :O.M 2V.8 30.4
IK-: it i3u;t
      Figure 23.  WINGOHOCKING  -  SWMM HYTEOGRAPH
                        FOR STORM 6 FROM REF.  1
                              43

-------
                            SECTION VII
                            DISCUSSION

     Successful technology transfer should produce use commensurate
with the capacities of the particular technology being transferred.
Transfer of a high-quality, comprehensive technology must result in
wide-spread application by a user community.

     The potential SWMM user community includes everyone involved with
water planning; including engineering and design firms, government
officials at all levels,  planning boards, the academic community, envi-
ronmentalists, and conservationists.   In order to utilize SWMM, each
type user requires a different mode of support.  Sophisticated engineer-
ing and design firms with experienced engineering staffs and well-
developed computing capabilities should require moderate assistance in
only the early phases of SWMM use and should quickly develop in-house
expertise in SWMM use.  A local planning board or conservationist
group would probably need extensive technical assistance.

     The dissemination of SWMM is seen then as a process of integrating
a program announcing the existence and capabilities of the model with
a program of flexible technical assistance.   This assistance should,
as rapidly as is proper,  instruct in use of the model.  The end result
is optimum use-levels, done with minimum effort, producing a set of
users skills commensurate with the abilities of the user group.

     There are several important lessons to be learned about SWMM tech-
nology transfer from this project.  First,  a simple but eye-catching
document for wide-spread mailings is essential.  Several of the target
potential-users were contacted in the local area only.  All mailings
should be expanded geographically to cover a larger region.  Second,
responses to mailings must bring competent assistance in both managing
and technically developing a modeling project.  Assistance in defining
the problem, relating it  to SWMM, and, outlining the capabilities and
requirements of SWMM is essential.  It appears, so far, that simply
mailing a SWMM tape to a potential-user is insufficient support to gen-
erate use.   Although heavy support is required in the early stages of
modeling, effective transfer must also include education to promote
user independence.

     The EPA grant for this project has resulted in a self-sustaining
SWMM dissemination-user assistance program under the UCSC, a non-profit
organization, for the benefit of potential SWMM users.
                                  44

-------
                         SECTION VIII
                          REFERENCES
 1.   Environmental Quality.  "Third Annual Report-Council on Environ-
     mental Quality."  3:16, August 1972.

 2.   McPherson, M.B., "Some Notes on the Rational Method of Storm
     Drain Design."  ASCE Urban Water Resources Program Technical
     Memorandum No. 6-ASCE.  January 1969.

 3.   Design and Construction of Sanitary Storm Sewers, ASCE Manual of
     Reports on Engineering Practice.  37, 1969.

 4.   Heaney, J.P., W.C.  Huber, et al., "Stormwater Management Model:
     Decision-Making."  Office of Research and Development, US EPA,
     EPA-670/2-75-022.

 5.   Huber, W.C., J.P- Heaney, et al., "Storm Water Management Model
     User's Manual - Version II."  Office of Research and Development,
     US EPA, EPA-670/2-75-017.

 6.   Metcalf & Eddy, Inc., University of Florida, and Water Resources
     Engineers, Inc., "Storm Water Management Model   Vol. II-Verifica-
     tion and Testing-" US EPA Water Pollution Control Research Series,
     11024DOC08/71.  97-128.

 7.   Chow, V.T. (ed), Handbook of Applied Hydrology,  McGraw-Hill Book
     Company.  9-24, 9-29, 1964.

 8.   Stall, J.B. and M.L. Terstriep, "Storm Sewer Design - An Evaluation
     of the RRL Method." Office of Research and Development, US EPA,
     EPA-12-72-068.  55-58, October 1972.

 9.   Personal conversations with Philadelphia Water Department engineers
     (presently involved in rainfall/runoff studies of the same storm
     events used in SWMM testing).

10.   Graham, P.H., L.S.  Costello, and, H.J. Mallin, "Estimation of
     Imperviousness and Specific Curb Lengths for Forecasting Storm-
     water Quantity and Quality."  Journal Water Pollution Control
     Federation.  46:717-725, 1974.
                                   45

-------
                    SECTION IX




                     APPENDIX





Table 13.   CONVERSION FACTORS  -  ENGLISH TO METRIC
English unit
•era
•ere- foot
cubic foot

cubic inch
cubic yard
feet per Minute
foot (feet)
gallon(s)

gallons per day



horsepower
inch(es)
inches per hour
•illion gallons
Billion gallons per
acre per day
•illion gallons per day
mile
parts per •illion
pound(s)

pounds per 1,000 cubic feet
pounds per cubic foot
pounds per square foot
pounds per square inch
square foot
square inch
square Bile
square yard
standard cubic feet
per Minute
ton (short)
yard
Abbr.
acre
•cre-ft
cf
cf.
cfs
cu in.
cy
deg f
IP-
fps
ft
gal.

gpd
gpd/sq ft
gp»
gp«/sq ft
gpsf
hp
in.
in./hr
• il gal.
• sad
• gd
• i
ppb
pp.
Ib
lb/«cre/day
Ib/day/acre
lb/1.000 cf
PCf
psf
psi
sq ft
sq in.
sq ni
sq yd
scfa>
too
yd
Multiplier
0.405
1,235.5
29.32
0.0283
28.32
16.39
0.0164
0.765
764.6
O.S55 (*F-32)
o.oosoa
0.305
0.305
3.785
9.353
3. 785
4.381 I 10"S
1.698 I 10"*
0. 213
0.0631
Z.445
0.679
40.743
0.746
2.54
2.54
3.785
3,785.0
0.039
43.808
0.0438
1.609
0.001
1.0
0.454
453.6
0.112
1.121
16.077
0.120
16.02
4.882 I 10"4
0.0703
0.0929
6.452
2.590
0.836
1.699
907.2
0.907
0.914
Abbr.
ha
cu •
1
I/sec
cu cm
1
cu •
1
deg C
•/sec
• /sec
•
1
1/day/ha
I/capita/day
I/sec
cu a/hr/sq •
cu i/Bin/ha
I/sec
cu/a/hr/sq •
1/sec/sq •
1/sq •
lev
en
CB/hr
HI
cu •
1/scc
ka
Bg/1
mg/1
kg
g
kg/day/ha
g/cu •
•g/1
kg/cu •
kg/sq CB
kg/sq CB
sq B
sq CB
Sq ta
sq B
cu B/hr
k!
•etric ton
B
Metric unit
hectare
cubic Beter
liter
P

liter
cubic Beter
liter
degree Celsius

«eter(s)
liter(s)
liters per capita per day
cubic aeters per hour
per square meter
cubic oeters per »inute
per hectare
cubic Meters per hour
square oeter
ki lowatts


cubic meters
per square aeter
kiloaeter
milligrams per liter
k ilogram
grams
neter
hectare


kilograms per square
cen t ime ter
kilograns per square
cent meter
square no ter
square k i lame ter
cubic net e rs per hour
k t log rams
net r ic ton
neter
                          46

-------
                                   TECHNICAL REPORT DATA
                            (Please read faUructions on the reverse before completing)
 . REPORT NO.
     EPA-670/2-75-041
                                                           3. RECIPIENT'S ACCESSION NO.
 4. TITLE AND SUBTITLE

     Storm Water Management Model:
     Dissemination and User Assistance
             5. REPORT DATE
              May 1975; Issuing Date
             6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
             8. PERFORMING ORGANIZATION REPORT NO,
             James A. Hagarman
             F.R.S. Dressier
 9. PERFORMING ORG ^NIZATION NAME AND ADDRESS

              University City  Science Center
              3508 Science Center
              Philadelphia, Pa.  19104
              10. PROGRAM ELEMENT NO.
               1BB034  ROAP: ATA Task 020
              11. CONTRACT/GRANT NO.
                                                                802716
 12. SPONSORING AGENCY NAME AND ADDRESS
     National  Environmental Research  Center
     Office of Research and Development
     U.S. Environmental Protection Agency
     Cincinnati,  Ohio  45268
              13. TYPE OF REPORT AND PERIOD COVERED
                  Final
             14. SPONSORING AGENCY CODE
 15. SUPPLEMENTARY NOTES
 16. ABSTRACT
      A program of dissemination and user-assistance  for the EPA Storm Water  Management
 Model (SWMM) has been developed and implemented  at  the University City Science  Center
 (UCSC).
      Services available  to  SWMM users under this grant include distribution  of the
 SWMM program itself and  technical assistance in problem delineation, data  preparation,
 execution debug, and output  interpretation.  Costs of this service extend  only to
 actual computing costs,  with all technical assistance covered by the EPA grant.
      Several case studies of SWMM applications  completed with UCSC assistance  in the
 past year are included in this report.  These studies include a combined sewer over-
 flow problem in Binghamton,  New York; a land use plan in the Stony Brook basin in
 Princeton,  New Jersey; and  RUNOFF/TRANSPORT calculations on the Wingohocking basin in
 Philadelphia, Pennsylvania.
      The UCSC SWMM dissemination program is now self-sustaining and continues  to assist
 the  user community.
      This report was submitted in fulfillment of the  Office of Research and  Develop-
 ment,  U.S.  Environmental Protection Agency (EPA) research grant EPA No. R-802716
 by the UCSC under the sponsorship of the EPA.
 7.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS
                             COSATI l-'icld/Group
 Water  Quality,* Mathematical Models,*
 Rainfall,  Surface Water Runoff,  Waste
 Treatment,* Erosion Control,
        Water Pollution*
  Water Quality Control,
  Stormwater  Management,
  Urban Stormwater, Water
  Pollution Control,
  Binghamton,  New York,
  U.S. EPA Storm Water
  Management  Model (SWMM)
   13B
 8. DISTRIBUTION STATEMENT

   RELEASE TO PUBLIC
19. SECURITY CLASS {This Report)
   UNCLASSIFIED
21. NO. OF PAGES
   55
20. SECURITY CLASS (This page)
   UNCLASSIFIED
                           22. PRICE
EPA Form 2220-1 (9-73)
                                             47
                                                     U.S. GOVERNMENT PRINTING OFFICE: 1975-657-593/537't Reg ion No. 5-11

-------