&EPA
United States
Environmental Protection
Agency
Research and Development
Municipal Environmental Research EPA-600/2-78-090
Laboratory May 1 978
Cincinnati OH 45268
Conventional and
Advanced Sewer
Design Concepts for
Dual Purpose
Flood and
Pollution Control
A Preliminary
Case Study,
Elizabeth, NJ
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-78-090
May 1978
CONVENTIONAL AND ADVANCED SEWER DESIGN CONCEPTS
FOR DUAL PURPOSE FLOOD AND POLLUTION CONTROL
A Preliminary Case Study, Elizabeth, New Jersey
by
Herbert L. Kaufman
Fu-Hsiung Lai
Clinton Bogert Associates
Fort Lee, New Jersey 07024
Grant No. S-802971
Project Officer
Anthony N. Tafuri
Storm and Combined Sewer Section
Wastewater Research Division
Municipal Environmental Research Laboratory (Cincinnati)
Edison, New Jersey 08817
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
11
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health
and welfare of the American people. , Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environment.
The complexity of that environment and the interplay between its components
require a concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem
solution. It involves defining the problem, measuring its impact, and
searching for solutions. The Municipal Environmental Research Laboratory
develops new and improved technology and systems for the prevention, treat-
ment, and management of wastewater and solid and hazardous waste pollutant
discharges from municipal and community sources. This work is to facilitate
the preservation and treatment of public drinking water supplies and to
minimize the adverse economic, social, health, and aesthetic effects of
pollution. This publication is one of the products of that research; part of
a most vital communications link between the researcher and the user com-
munity.
Alleviation of the deleterious effects of combined sewer overflows
and storm water discharges on the nation's waterways depends upon characteri-
zation of the pattern, quantity and quality of these flows and evaluation of
alternative methods to achieve cost-effective control. This report compares,
using the urban area of the City of Elizabeth, New Jersey, the cost-effec-
tiveness of alternatives for a community-wide sewer system. Advanced sewer
designs, providing controlled flow routing, are compared with conventional
combined and separate sewer systems. The report also describes a practical
methodology for planning cost-effective facilities to abate pollution from
wet-weather flows.
Francis T. Mayo
Director
Municipal Environmental
Research Laboratory
iii
-------
PREFACE
The study was made specifically for Elizabeth, N.J. Because of funding
limitations, some approximate analytical methods were used. These approxi-
mate methods were evaluated to insure the validity of the conclusion reached
as to the main purpose of the report; i.e., the development of relative
cost/benefit relationships for various types of sewer systems under identical
inputs. In Step 1 (Section 201) Facility Planning, with adequate funding,
more exact methods x^ould be used to determine combined sewage overflow
pollutant characteristics and to route flows through the system.
Determination of the preferred method of handling rainfall data was
outside the scope of this work. In addition, the discrete rainfall/runoff
data required for "fine" temporal and distributional analysis and correlation
was not available, nor were the funds for their development. However, the
importance of runoff volume, as opposed to runoff rate, to achieve combined
sewage overflow pollution abatement was recognized. Hence, the rainfall
handling was based on synthetic hyetographs which were designed to approxi-
mate the runoff volume for the stated return interval at all time segments,
rather than peak runoff rates.
The alternative systems developed using the synthetic hyetographs
were tested for pollution abatement benefits using STORM and available
hourly real rainfall records for a continuous 12 year period.
Although the rainfall input data may not be based on the preferred deter-
mination methodology, it still offered a consistent/convenient means for com-
paratively evaluating the most cost-effective alternative drainage system for
combined sewer overflow pollution control.
The concepts developed have been used and the results accepted for
Steps 1, 2 and 3 grants for Trenton, New Je'rsey. Step 1 planning is also
proceeding in Elizabeth, New Jersey, based on the findings of this study.
-------
ABSTRACT
Alternatives for pollution abatement from combined sewer overflows
and stormwater discharges were evaluated. The following types of systems
were compared: separate storm and sanitary system, conventional combined
system, and advanced combined systems with varying amounts of in-pipe and/or
satellite storage and controlled flow routing. The cost-effectiveness
comparison was based on achieving a desired effluent quality assuming a new
sewer system. In addition, the effects on pollution abatement and cost of
changing various elements (collection system, interceptors, storage and
treatment works) of the system were investigated. Manning's equation was
also compared qualitatively with the complete momentum equation for the
hydraulic design of sewers.
Two mathematical models, SWMM and STORM, were employed. SWMM was
used to design sewers and to analyze the characteristics of the quantity and
quality of combined sewage and stormwater runoff for selected synthetic
storms. STORM was used to analyze a continuous 12-year, real rainfall record
to determine the frequency of overflow events, the mass of pollutants dis-
charged in the overflows for the alternatives considered, and the duration
and quantity of the overflows. The runoff and overflow characteristics were
developed for one .drainage district (about 265 hectares or 655 acres) in
Elizabeth, New Jersey. A method was developed to use these characteristics
to determine runoff and overflow characteristics from the remaining 24
drainage districts in the City so that a city-wide sewer system could be
planned.
The study (a) evaluated the effects of differing rainfall patterns
and intensities on pollutant concentrations, the effects of varying inter-
ceptor and treatment capacity on pollutants discharged, and the economics of
peak flow equalizing for different levels of treatment; (b) quantified
the long-term pollutional loads from combined sewage overflows; and (c)
evaluated the effect- on overflow control of varying the amounts of upstream
and downstream storage. The report concluded that for Elizabeth, N.J., the
most economical systems were those which included substantial amounts of
upstream storage to minimize the size of trunk and interceptor sewers,
pumping stations and treatment facilities.
While the study was made for the City of Elizabeth, New Jersey, the
methodology should be generally applicable to other urban areas.
v
-------
This report was submitted in fulfillment of Grant No. S-802971 by
the City of Elizabeth, New Jersey under the partial sponsorship of the
U.S. Environmental Protection Agency. Work was completed as of August
1976.
vi
-------
CONTENTS
Foreword ±±±
Preface iv
Abstract v
List of Figures ix
List of Tables ; xii
Acknowledgments xv
Sections
I. Summary 1
II. Specific Conclusions-City of Elizabeth 5
Pollution Characteristics - Combined Sewage 5
Pollution Control Potential - Conventional
and Existing Combined System 6
Pollution Control - Advanced Combined Systems 8
The "Normalized Hydrograph" - a Planning Aid 9
Costs 10
Additional Hydraulic Studies 13
III. Recommendations '. ..14
TV. Introduction ................16
V. Objectives and Scope of the Study 19
VI. Description of Study Area ,.,.21.
VII. Suggested Methodology. 25
VIII. Design and Cost Comparison of Sewer Systems in District A....30
Design Philosophy 31
Conventional Combined Sewer System 35
Design 35
Outflow Hydrographs and Pollutographs 41
Separate Storm Sewer System 57
, Separate Sanitary Sewer System 59
' Combined System with In-system Storage ....62
Design with In-pipe Storage 62
Outflow Hydrographs and Pollutographs .....67
Sewers with Satellite Storage. 67
Design * 70
Outfow Hydrographs and Pollutographs 76
Pollution Control Potential of Conventional
Combined System 83
vii
-------
Cost Comparison of Alternative Collection Systems 83
Cost of Conventional Sewer Systems .......... 85
Cost of Advanced Sewer Systems .....,, 87
IX. Design and Cost Comparison of Alternative City-Wide Sewer
Systems • . 89
Synthetic Storm Runoff 89
Design of Intercepting Sewers..... 96
Amount of Storage Required vs. Treatment Rate... 97
Cost Comparison of Alternative Sewer Systems.. 100
X. Analysis of Overflow Quantity and Quality, 105
Effect of Storage and Interceptor Capacity. 106
Effect of Street Sweeping Practice. Ill
XI. Interceptors for Pollution Control 115
The "First Flush" Phenomena. ,.115
XII. Alternative Pollution Control Programs 118
Conventional Combined System... .118
Advanced Combined System. 120
Selection of Design Storm for Separate and Combined
Systems 124
XIII. Appendices 126
A. Alternative Model Evaluation 127
B. Modification of SWMM and STORM Programs 147
C. Quantity and Quality Considerations, SWMM vs
STORM 150
D. Model Input Data ..154
E. Calibration of STORM Runoff Coefficients 197
XIV. References 202
XV. Glossary, Abbreviations and Symbols 206
viii
-------
FIGURES
Number
Study area, Elizabeth, New Jersey.
Page
..22
10
11
12
13
14
15
Hyetograph and outflow hydrograph, District A (5-year
intermediate pattern storm) 42
Mass rate SS pollutographs, 4 dry days (5-year intermediate
pattern storm) ,45
Mass rate. BOD pollutographs, 4 dry days (5-year interme-
diate pattern storm) 45
Mass rate SS pollutographs, 30 dry days (5-year interme-
diate pattern storm) 47
Mass rate BOD pollutographs, 30 dry days (5-year interme-
diate pattern storm) ....47
SS, BOD concentration pollutographs, 5^-year intermediate
pattern storm 49
Outflow hydrograph and pollutographs, 1-year intermediate
pattern storm 50
Outflow hydrograph and pollutographs, 1.3-month interme-
diate pattern storm .51
Outflow hydrograph and pollutographs, 5-year advanced
pattern storm 52
Combined sewage pollutographs for various storms 55
Stormwater runoff pollutographs for various storms 56
Sanitary sewer layout, subareas I and II 60
Sanitary sewer layout, subareas III, IV and V 61
Incremental cost of pipe storage. 63
ix
-------
FIGURES (continued)
Number Page
16 Weir-orifice regulator 64
17 Hydrographs modified by in-pipe storage.. 66
18 Outflow hydrographs with in-pipe storage (5-year interme-
diate pattern storm) « 68
19 Wet-weather BOD variation, Trenton, New Jersey 69
20 Outflow hydrographs with satellite storage at various
locations (5-year intermediate pattern storm)..... 71
21 Schematic diagram of satellite storage locations,
District A 73
22 Outflow hydrographs with satellite storage (5-year
intermediate pattern storm) • 75
23 Outflow pollutographs with satellite storage (5-year
intermediate pattern storm) 77
24 Outflow hydrographs and pollutographs with satellite
storage (1-year intermediate pattern storm) 79
25 Outflow hydrographs and pollutographs with satellite
storage (1 • 3-raonth intermediate pattern storm) 80
26 Outflow hydrographs and pollutographs with satellite
storage (1-year advanced pattern storm) - 81
27 Normalized hydrographs 93
28 Outflow hydrographs, Districts A, E and H 95
29 Total cost comparison-city sewer system, in-pipe vs.
satellite storage • 103
30 Correlation of annual SS overflow with annual precipi-
tation 107
31 Annual number of overflow events for various storage and
interceptor capacities « 109
32 Annual SS overflow for various storage and interceptor
capacities • HO
33 Effect of street cleaning interval on SS street washout 113
x
-------
FIGURES ( Continued)
Number Page
34 Pollution control costs - conventional system 121
35 Pollution control costs - advanced system , 123
A-l Definition sketch for equations of motion ....128
A-2 Uniform flow rating curve and loop rating curve 132
A-3 Surcharged simple pipe system. 135
C-l Street SS accumulation, SWMM vs. STORM., 153
D-l Frequency distribution of precipitation amounts 156
D-2 Frequency distribution of precipitation duration 157
D-3 Frequency distribution of antecedent dry hours...... 158
D-4 Rainfall intensity-duration-frequency curves 160
D-5 5-year design storm hyetographs and infiltration curve 161
D-6 Frequency of hourly rainfalls 163
D-7 Subareas of drainage District A 166
D-8 Combined sewer system layout, subarea I 167
D-9 Combined sewer system layout, subarea II 168
D-10 Combined sewer system layout, subarea III 169
D-ll Combined sewer system layout, subarea V 169
D-12 Combined sewer system layout, subarea IV ..170
D-13 Separate storm sewer system layout. 171
D-14 Definition of drainage districts and interceptor system in
Elizabeth T. .. .186
D-15 Unit cost of sewers 193
D-16 Cost of treatment vs. % of BOD removal 196
E-l STORM runoff coefficient C, vs.. rainfall ,.200
xi
-------
TABLES
Number
1
2
3
4
5
6
10
11
12
13
14
15
Page
Lateral Sewer Design Diameters (inches) • • 37
Trunk Sewer Design Diameters (feet) 40
Comparison of Design Diameters and Peak Hydrographs 43
Solids Deposit (Ibs) in Sewers with Four Antecedent
Dry Days - 4"
Summary of Runoff Quantity and Quality Characteristics
for Various Storms • 54
Comparison of Pipe Diameter, Slope, and Excavation
Depth, Separate Storm versus Combined Systems 58
Peak Flow Comparison for Separate Storm and Combined
Sewer Design • ^9
Comparison of Trunk Sewer Sizes (feet) for Storage
Basin at Various Locations 72
Summary of Runoff Quantity and Quality for Combined
System with Satellite Storage .. .82
Storage Volume in Sewer Pipes and Storage Basins,
District A 84
Cost Comparison of Conventional Combined and Separate
Sewer system ($10 ) of District A 86
Cost Summary of Combined System with In-system
Storage • 88
Summary of Normalized Hydrograph Development from
District A Data • • • 91
Adjustment of Peaking Time with Respect to that of
District A • 92
Development of Outflow Hydrographs 94
xix
-------
TABLES (continued)
Number
16
17
18
19
20
21
22 .
23
A-l
A-2
C-l
D-l
D-2
D-3
D-4
D-5
D-6
D-7
D-8
D-9
Intercepting Sewer Pipe Element Data.,
Page
.,.96
Summary of Intercepting Sewer Diameters (feet) and
Cost ($10 ).'
.98
Off-line Storage Required (10 gallons) and Cost for
Equal Weight of SS Discharged to Receiving Waters 99
Total Cost of City-Wide Sewer Systems 101
Effect of Storage and Flow Interception on District A
Sewer Sys tern Discharges 108
Pollutant Overflow as Affected by Street Sweeping Inter-
val and Efficiency . 112
Interceptor Capacity Required to Contain First Flush from
District A 116
Comparison of Alternative Combined Sewer Systems 119
Range of Typical Values for Each Term in Momentum Equa-
tion : 130
Peak Flow Comparison between Rational Method and SWMM......144
Pollutant Default Values for SWMM and STORM ...151
Annual Precipitation and Runoff Events 155
Hourly Rainfall versus Return Interval 164
Storm Characteristics. 164
Area of Land Use Distribution in Each Subarea (acres) 165
Total Sewer Length Comparison 172
Sub catchment Data for SWMM 173
Pipe Element Data for Combined Sewer System 177
Subcatchment Water Quality Input Data for SWMM.. 179
Subarea Curb Length (feet) versus Land Use 182
xiii
-------
TABLES
Number
D-10
D-ll
D-12
D-13
D-14
D-15
D-16
D-17
E-l
E-2
Page
Summary of Drainage District A Land Use Data. 182
Monthly Evaporation Rate. - -•••,.•• 184
Summary of Land Use in The City of Elizabeth.. 187
Population 188
Average Daily Pollutant Concentration of Domestic Waste-
water 189
Diurnal Variation of Domestic Wastewater Quantity :and
Quality for District A 190
SWMM Input Data for Domestic Wastewater Computation,
District A 191
Unit Cos t of Sewers 194
Sensitivity of Hyetographs and Integration Interval on
Runoff Volume 198
Calibration of Runoff Coefficient C 199
xiv
-------
ACKNOWLEDGEMENT S
The support of the project by the City of Elizabeth and the Office
of Research and Development, U.S. Environmental Protection Agency is acknow-
ledged and appreciated.
The assistance and cooperation of Messrs. Richard Field and Anthony
N. Tafuri, Chief and Grant Project Officer, respectively, Storm and Combined
Sewer Section, Municipal Environmental Research Laboratory, U.S.
Environmental Protection Agency, has been invaluable.
This project was directed by Mr. Herbert L. Kaufman, Partner-In-Charge.
Messrs. John H. Scarino, Principal Associate arid Director of Engineering
Management, and William Wheeler, Associate, provided valuable criticism and
review. Dr. Fu-hsiung Lai served as Project Engineer and Mr. Gerald G.
Gardner as Engineer. Dr. Lai developed original and innovative analyses,
which resulted in significant findings as to practical methods for reducing
pollution from combined sewage overflows.
Dr. Brendan M. Harley, Dr. Guillermo J. Vicens and Mr. Richard L.
Laramie of Resource Analysis, Inc., Cambridge, Massachusetts, provided
computer program changes to SWMM and STORM and prepared drafts for parts of
Appendix A in late 1974. Dr. Harley also participated in the preparation of
the "Design Philosophy" described in Section VIII.
Dr. Wayne C. Huber of the University of Florida and Mr.. Harry C.
Torno of the U.S. Environmental Protection Agency reviewed the draft of
this report. Their comments and constructive criticism are appreciated.
Dr. Huber also freely contributed his expertise in many discussions during
the preparation of the report for which we express our sincere thanks.
Finally, the authors are grateful to Mr. M.B. McPherson, Director
of ASCE Urban Water Resources Research Program, and Mr. Chi-Yuan Fan, Storm
and Combined Sewer Section, Municipal Environmental Research Laboratory, U.S.
Environmental Protection Agency, who also contributed constructive comment.
xv
-------
-------
SECTION I
SUMMARY
Pollution resulting from the discharge of untreated wastes from combined
sewers is a national problem which requires correction if the objective
of clean waters is to be achieved. Such discharges are now experienced
from systems serving a population of 54 million located in some 1,300 munici-
palities. They occur, on average, about .every 5 to 10 days, depending on
location, a'nd can reduce the efficiency of pollutant removal from about 90
percent to 15 to 25 percent.
Using computer technology, tools to investigate and analyze the various
elements required for planning cost-effective solutions for combined sewage
overflow pollution abatement have been developed. This study modifies
and improves existing tools and develops new tools. It organizes these tools
into a systematic methodology that can have broad application.
This study developed appropriate criteria for sewerage improvements.
The criteria are to be used to:
identify alternative corrective programs for abating pollution from
combined sewage overflows, and
- select the most cost-effective solution.
STUDY PROCEDURE
Initially, alternatives for combined and separate sewage collection
systems to serve a 1.6-square-kilometer (1.0-square-mile) area in Elizabeth
were analyzed. , The systems were designed to convey, without flooding,
flows resulting from a two-minute interval, synthetic storm hyetograph which
matched the intensity-duration curve applicable to the area at each time
interval for a 5-year return frequency rainfall. The characteristics of
the hydrographs and pollutographs for the alternative collection systems
were determined and applied to the remaining 11.0-square-kilometer (6.9-
square-mile) urban area of the City. The alternatives, including interceptor
and treatment/storage systems, which would provide essentially complete
capture and treatment of pollutants for each of the collection system alter-
natives, were then screened by comparing costs of the total systems. The
reduced cost resulting from reducing the amount of pollutant captured for
treatment were then estimated and an evaluation made of the pollutional
impact of the resulting discharges.
-------
The 5-year return interval, synthetic storm hyetograph developed does
not represent the 5-year return frequency runoff. However, since this
study emphasized pollution abatement, and, since the calculated amounts of
mass pollutants in the runoff from a similarly synthesized 1-year return
interval storm hyetograph are not significantly different (TABLE 9), the
amount of pollutants determined may be expected to represent the quantity to
be expected from a 5-year return frequency runoff volume computed by other
methods. The function served by the selected 5-year synthetic storm hyeto-
graph was to permit design of alternative facilities that were tested for
effectiveness in pollution abatement with a 12-year continuous real rainfall
record.
For the City of Elizabeth, the quantity of pollutants in urban runoff
and combined sewage, which can be intercepted for treatment with alternative
facilities were relatively quantified. The alternatives considered included:
1. Separate sewer systems;
2. Combined sewer systems designed to convey urban runoff from a
synthetic, 5-year return frequency rainfall (Figure D-5);
3. Combined sewer systems with flow routing and storage in the trunk
sewer;
4. Combined sewer systems with enlarged lateral sewers to permit
storing various amounts of runoff from an intermediate pattern
storm (peak rainfall in the middle half of the storm duration);
5. Combined sewer systems, similar to 4, except that storage basins
were provided along the trunk sewer; and •
6. Interceptor sewers of varying capacity together with storage and
treatment facilities.
Evaluation of 32 alternatives for collection system modifications
and of 27 alternatives for total sewer system modifications was made,
assuming all new construction, to determine relative cost-effectiveness. The
interdependent relations between flood alleviation and pollution control in a
combined sewer system were described.
METHODOLOGY
A suggested flexible methodology has been developed which can be applied
for analysis and screening of many alternatives to determine the cost-effec-
tive solution for abatement of combined sewer overflows. It modifies to
some extent the procedures used in this study. The methodology did not
develop a "design storm" for pollution control. It used a long-term, con-
tinuous rainfall record and calculated runoff and quality characteristics of
real rainfalls to determine the relative effectiveness of alternative
facilities.
-------
The EPA Storm Water Management Model (SWMM) was used to:
1. design the many alternative systems for a common synthetic rainfall
event (Figure D-5), and
2. quantify relatively the quality and quantity of combined sewage for
a broad range of specific, lesser, synthetic rainfall events with
varying patterns.
The Corps of Engineers' Storage, Treatment, Overflow and Runoff Model
(STORM) was used to compare the performance of the various systems q-uantita-
tively with respect to pollution abatement for a 12-year continuous period
of real rainfall events. It determined:
1. the number of annual combined sewer overflow events;
2. the mass loading (weight) of pollutants discharged in the overflows
for the different facilities considered; and
3. the duration and quantity of the overflows.
The study evaluated the effects of:
1. random rainfall patterns and intensities;
2. the long-term and shock pollutional loads from combined sewage
overflows;
3. varying amounts, modes of operation, and locations of storage;
4. varying interceptor and treatment capacity;
5. peak flow equalization; and
6. rainfall patterns on pollutant concentration.
The SWMM RUNOFF Block was modified to include design capability in
sizing sewers. STORM was modified to extend its use to combined sewage.
SIGNIFICANT FINDINGS
1. Capture of the low-volume, high-concentration first flush from
combined systems is essential for pollution abatement.
2. Annual discharges of moderate rainfalls contain more pollutant with
higher concentrations than do discharges from severe storms. The
combined sewage discharges of moderate rainfalls should be con-
trolled and treated for protection of the environment.
3. Storage of combined sewage should reduce pollutant concentration as
a result of mixing the highly polluted first flush with later,
-------
less polluted flows. Storage is effective in abating pollution.
4. If storage cannot be provided, diversion of combined sewage flows
of 17 or more times the peak dry-weather flow for treatment should
result in significant pollution abatement benefits.
5. A combined system can be designed to discharge less pollutant than
a conventional separate system in which the storm sewers discharge
all urban runoff directly to water courses.
6. Providing additional storage in a combined sewer system should
result in the least-cost system for pollution control. This
storage can be provided by enlarging the size of the upstream sewer
reaches if site restrictions result in costly satellite storage
basins.
The ability of SWMM and STORM to simulate the pollutant components
reliably in urban runoff and combined sewage is questionable. This also
applies to all known presently available models. However, the "first flush"
phenomenon has been observed, as well as the substantial reduction in pollu-
tant concentrations after the "first flush". Hence, the relative quantifica-
tion of pollutants, through SWMM and STORM, offers a practical, available
measurement by which alternatives for combined sewage overflow abatement
facilities can be developed and evaluated.
-------
SECTION II
SPECIFIC CONCLUSIONS - CITY OF ELIZABETH
POLLUTION CHARACTERISTICS - COMBINED SEWAGE
1. The magnitude of pollution resulting from combined sewage over-
flows in Elizabeth is indicated by:
(a) the probability of about 66 overflows of raw, combined sewage
annually, or about one every 5 or 6 days;
(b) an average of about 60 percent of the raw sewage entering
the sewer system during the overflow periods being discharged
untreated; and
(c) the concentration of pollutants in the initial discharges
possibly being an order of magnitude greater than in normal
dry-weather flow.
2. Combined sewage generated by an "intermediate" (typical) pattern
storm (peak rainfall in the middle half of the storm duration)
exhibits two peak flushes of mass pollutants. The first,peak
flush includes the washout of solids deposited in combined sewers
from sanitary wastes during dry days. The second peak flush
includes the washout of street pollutants. The first peak is of
low-flow volume but high-pollutant concentration. The second
peak is of high-flow volume but low-pollutant concentration.
Combined sewage generated by an "advanced" pattern storm (peak
rainfall in the first quarter of the storm duration) exhibits
only a single peak flush. When compared to the peaks of an inter-
mediate "pattern storm, this peak has less . flow volume but a much
higher pollutant concentration than the second peak and a lower
concentration than the first peak.
With either storm pattern, however, the highest concentration of
pollutants occurs early in the storm.
3. The mass washout of combined sewage solids from an individual
major storm (5-year rainfall return frequency) is greater than
from individual frequent rainfalls (1.3 and 0.35-month rainfall
return frequencies). However, the total pollution resulting
-------
from frequent rainfalls is greater. Rainfalls with an individual
cumulative depth of less than 25.4 mm (1.0 inch) produce about
93 percent of the total annual precipitation and about: 73 percent
of the total pollutant washout for the year. A storm of 0.35-month
(about 10 days) return frequency, which has a peak runoff rate
about ten times the average dry-weather flow, should cleanse the
sewer of most solids deposits. Hence, capturing runoff from
frequent storms is essential for pollution control.
4. A 5-year return frequency rainfall may be expected to remove
about 96 percent of the total suspended solids accumulated on
streets; a 1-year rainfall, 87 percent; and a 1.3-month rain-
fall, 35 percent. However, except when preceded by a protracted
dry period, street washout should contain substantially lower
pollutant concentrations than the sewer solids washout since it
usually occurs at a time when high flow rates provide dilution.
5. With the synthetic 5-year storm hyetograph used, treatment of
essentially all pollutants in the combined sewage and street
washout would require capture for treatment of the first 90
minutes of flow from the storm. Such facilities may be justi-
fied only where the receiving water requires an extremely high
quality discharge to preserve the environment.
6. In a combined sewer system, the lower the flow velocity in dry
weather, the greater would be the amount of solids deposit.
Maintaining or creating flushing velocities in combined systems
during dry weather should provide a significant reduction in the
pollutant concentration in the initial storm flush.
7. Both SWMM and STORM provide an option with internally specified
default values (see TABLE C-l) for computation of suspended
solids (SS), biochemical oxygen demand (BOD), and coliforms from
the mass of dust and dirt in the street washouts. Use of the SWMM
default values results in SS and BOD mass discharges about ten
and five times, respectively, that computed using STORM default
values.
POLLUTION CONTROL POTENTIAL - CONVENTIONAL AND EXISTING COMBINED
SYSTEMS
1. Flow routing devices can be used to reduce the number of annual
combined sewage overflow events. When installed in a conven-
tional combined sewer system, designed using SWMM for a synthetic
intermediate pattern storm, and having interceptor capacity equal
to the peak dry-weather flow, reductions for Elizabeth proved to be
as follows:
-------
2.
Design Storm
Return
Frequency
10 years
5 years
2 years
Reduction of
Overflow
Events
80%
65%
small
Mass Reduction
of Overflow
Pollutants
75%
55%
small
The ineffectiveness of the system designed with a 2-year return
interval storm in abating pollution results from the reduction in
volume of about 40 percent for storage of flows during a 1.3-
month storm. Analysis of real rainfall events, based on a con-
tinuous 12-year record, indicates the available storage volume
would be exceeded frequently, substantially increasing the number
of overflow events and the mass of pollutants discharged.
The cost of sewer facilities to prevent flooding from a 10-year
return frequency rainfall is about 25 percent greater than from a
5-year return frequency rainfall. Since less costly means are
available in the design of new systems to reduce pollution from
combined sewage overflows and urban runoff, the use of a 10-year
return interval storm for pollution control can not be justified.
Increasing interceptor sewer size can reduce the number of over-
flows and pollutants discharged in combined sewage overflows.
Based on analysis using actual rainfall events over a 12-year
continuous period, the calculated relative amounts of SS and BOD
discharged for various interceptor capacities were determined.
The analysis assumed no storage upstream of the interceptor. The
results for a drainage area of about 1.6 square kilometers (1.0
square mile) are summarized as follows:
Interceptor
Capacity
(Times Peak
Dry-Weather
Flow)
1.0*
3.8
6.2
8.2
Average Annual Overflows
Number of
Events
Longer than
1 Hour
65.8
40.3
28.5
18.5
Volume
(%)
100.
50.40
32.50
24.20
Relative
Amounts of Pollutants
SS BOD
(%) (%)
100. 100.
41.70 29.30
26.30 15.00
18.10 10.00
(continued)
Conventional design capacity based on Harmon's Ratio.
-------
Interceptor
Capacity
(Times Peak
Dry-Weather
Flow)
16.5
22.3
23.1
26.8
59.2
151.7
Number of
Events
Longer than
1 Hour
6.6
3.8
3.5
2.3
0.2
0.0
Vo lume
(%)
8.90
4.30
4.00
3.30
0.04
0.00
SS
(%)
5.20
2.30
1.90
1.20
0.04
0.00
BOD
(%)
2.50
1.10
0.90
0.60
0.04
0.00
The mass of pollutants discharged reduced at a faster rate than
the volume due to greater capture of the initial flush.
3. Where the capacity of an existing combined system is such as to
result in frequent combined sewage flooding, but where the system
is adequate to convey dry-weather flows, and adequate storage
facilities cannot be provided economically upstream, it is cost-ef-
fective by a large margin to divert urban runoff in excess of the
combined system capacity to. a separate stormwater system for the
same level of pollution abatement.
4. Based on data developed for Elizabeth, use of the Rational Method
to determine peak runoff rates results in values 30 to 60 percent
less than those obtained using SWMC1 for a synthetic 5-year return
frequency storm (Figure D-5). In a separate study by the U.S. Army
Corps of Engineers at Louisville, Kentucky in 1949 (33), peak
runoff rates obtained with the Rational Method were 20 to 43
percent less than those obtained with the Unit Hydrograph Method.
Sewers sized in accordance with such values would not permit
effective use of flow control devices for pollution control
because of the limited storage volume available.
POLLUTION CONTROL - ADVANCED COMBINED SYSTEMS
1. Storage of wet-weather flows in the collection system, either in
the pipe (in-pipe) or in storage basins (satellite) , is an ef-
fective pollution' control method. By mixing the highly pol-
luted initial flushes with the later, less polluted runoff, the
concentration of pollutants discharged in the untreated wastes is
reduced. By reducing the peak flow rate, more of the waste
can be diverted to the interceptor and subsequent treatment. The
following presents the SS diverted to treatment for the various
synthetic storms selected with varying amounts' of storage (also
see TABLE 9):
-------
Rainfall
Return
Frequency
5 years
1 year
1.3 month
0*
0
0
0
SS Diverted in Percent
9.4* 17.0*
34
37
55
37
46
56
41.3*
51
78
91
* Storage in percent of 5-year synthetic storm runoff volume. The
pollution control benefits increase with the amount of storage
provided and with a shorter return interval storm.
2. In a combined system, upstream storage reduces the size of trunk,
interceptor and treatment facilities. For a 1.6-square-kilometer
(1.0-square-mile) area in Elizabeth, about 99 percent of the
pollutants that would overflow from a conventional combined system
could be captured for treatment with storage equal to 9.4 percent
of the 5-year synthetic storm runoff and an interceptor capacity
of 17 times peak dry-weather flow. If the storage is increased to
41.3 percent, the interceptor capacity could be reduced to about
four times peak dry-weather flow for the same amount of capture.
With 41.3 percent storage and interceptor capacity equal to the
peak dry-weather flow, (the conventional capacity), about 95
percent of the pollutants should be captured for treatment.
Storage for pollution control provides flood protection as a side
benefit.
3. The manner of flow routing through storage influences both costs
and the degree of pollution control achieved. If upstream storage
is bypassed until the downstream trunk sewer capacity is exceeded,
interceptor and downstream storage and treatment capacity must be
much larger for a desired pollution abatement goal as compared to
the facilities required if all flow is routed through upstream
storage. For a given facility, bypassing upstream storage basins
until the trunk sewer capacity is reached would result in lowest
operating costs, but routing all flow through upstream storage
basins would result in greatest pollution control. The preferred
method of operation might consider bypassing upstream storage until
interceptor capacity is exceeded.
THE "NORMALIZED HYDROGRAPH" - A PLANNING AID
1. For a given rainfall pattern, the primary factors defining the
outflow hydrograph of a combined drainage system serving, an urban
environment, are area and percent imperviousness. The area
influences peak flow reduction and time of occurrence. Percent
imperviousness affects flow volume. Within reasonable ranges of
area and percent imperviousness, the shape of the hydrographs is
sufficiently similar to permit development of a family of
-------
"normalized hydrographs" whose ordinates are defined by a percen-
tage of the peak runoff rate and whose area is measured in seconds.
The volume of runoff of a storm event may be determined by a
simple computation (as in STORM). The runoff at each ordinate
then is determined so that this volume is obtained. These ordi-
nates provide an adequate representation of the hydrograph from the
given area for flow routing through an interceptor. This concept is
an aid in reducing the amount of segmentation required and the
costs for planning.
COSTS
1.
The amount of in-pipe storage which may be provided is limited
by cost since:
(a) The incremental unit cost of in-pipe storage volume increases
with sewer size (Figure 15),
(b) storage resulting from a large increase in sewer diameter
should be more costly per unit of volume than for a rela-
tively small increase,
(c) pipe storage in the smaller lateral sewers provides not only
lower costs for in-pipe storage but also permits reduction
in the size of the more costly trunk sewers, and
(d) satellite (off-line) storage, which is located so as to
reduce trunk sewer sizes, would probably be a cost-effective
alternative since its cost per unit of volume may be as
little as 30 percent of in-pipe storage. .
2. Storage to reduce costs should result in:
(a) reduction of the peak flow of the outflow hydrograph to the
extent practical and economical,
(b) the multiple peaks of the hydrograph being relatively equal,
and
(c) reduction of the concentration of pollutants in the initial
peak by dilution.
3. Comparative costs of the alternative collection systems con-
sidered are as follows (TABLE 11, TABLE 12):
(a) a conventional combined system should cost about the same as
separate systems when surcharge of the separate drainage
system is permitted,
(b) the combined system permitting significant overland drainage
flow is the least cost conventional system,
10
-------
(c) separate systems in which surcharge of the storm drainage
system is not permitted are most costly,
(d) surcharge of a separate storm drainage system reduces
costs by about 20 percent,
its
(e) in the case of the City of Elizabeth storing about eight
percent of the 5-year synthetic storm runoff volume in
the upstream lateral system results in the least cost for
in-pipe storage systems since the increased cost of the
lateral sewers is more than compensated by the reduced cost of
trunk sewers. However, the increased cost resulting from
doubling or tripling the amount of storage in the laterals is
relatively small,
(f) if satellite storage basins can be built at low cost, about
$0.08 per liter ($0.3 per gallon), the least-cost collection
system results from storing about 17 percent of the 5-year
synthetic storm runoff volume, and
(g) if satellite storage is costly, about $0.26 per liter ($1.0
per gallon), the cost of the collection system increases with
the increase in storage volume. '
4. Comparative costs of the alternative systems considered for
control of urban runoff and combined sewage overflows from the
entire City, assuming essentially complete capture (92% using the 5
year synthetic hyetograph) of pollutants in both sewage and urban
runoff for treatment, are:(TABLE 19 and Figure 29)
(a) Where the cost of off-line storage is high, $0.26 per liter
($1.0 per gallon), and in-pipe storage cannot be provided,
separate sewers would be more economical than combined sewers.
However, where the storage cost is low, $0.08 per liter ($0.3
per gallon), the combined system is more economical.
(b) The amount of in-pipe storage that may be optimally provided
is about 30 percent of the runoff developed using the 5-year
synthetic storm hyetograph with the assumed range in cost of
off-line (downstream) storage. ,
(c) The conventional separate system would be the most costly
choice if storage (off-line or satellite) can be built at a
cost of $0.08 or less per liter ($0-30 per gallon).
(d) At a satellite storage cost of $0.26 per liter ($1.0 per
gallon), the conventional separate system would be more
economical than any alternative advanced combined system.
(e) In all instances, the advanced combined system offers advan-
tage over the conventional system.
11
-------
5. The least-cost system for essentially complete control of pollu-
tants from urban runoff and combined sewage is expensive, being
estimated at more than $114,000 per hectare ($29.5 million per
square mile) (ENR Construction Cost Index equals 1800).
6. The criteria of removing essentially all pollutants from urban
runoff and combined sewage overflows is economically imprac-
tical. Capturing about 97.5 percent of the wet-weather BOD for
treatment, rather than substantially 100 percent, would reduce
costs of the overall system by about 25 percent. Hence, alter-
native criteria were explored which would result in acceptable
pollution control at a more reasonable cost. These_ are tabulated
below:
Description
A
A
B
C
A
Interception
Rate
% of Peak
DWF
380
2310
1380 .
220
100
Pollutant*
Capture
58
98
98+
98
0
Cost
$/Hectare
79,000
96,000
90,000
80,000
75,500
A. Conventional combined system.
B. Conventional combined system with flow routing to utilize storage
capacity available in sewers during normal rainfalls.
C. Advanced combined system with about 40% in-system storage.
* With respect to SS overflow from a conventional combined system
with intercepting rate equal to peak dry-weather flow.
A conventional combined system, without provision for pollu-
tion control, would cost about $75,500 per hectare ($19.6 million
per square mile). Upgrading this system by increasing interceptor
capacity from 1.0 to 3.8 times peak dry-weather flow would increase
costs by about five percent but provide about 58 percent pollutant
capture. Increasing interceptor capacity, only, to obtain 98
percent pollutant capture would increase base costs by about ^27
percent. Using a conventional system with flow routing and in-
creased interceptor capacity to obtain a pollutant capture of 98
percent would increase the base cost by about 20 percent. If
in-system storage equal to about 40 percent of the 5-year synthetic
storm runoff can be provided, 98 percent of the annual pollutants
in the combined wastes should be captured with an increase of base
cost of about seven percent.
The cost of providing pollution control by in-system storage
12
-------
and flow control devices for urban runoff and combined sewage is
quite small when considering the overall investment in sewerage
systems.
7. The optimum cost relation between off-line storage and treatment
capacity depends on the degree of treatment required. If simple
processes such as micros training will suffice, the cost of addi-
tional storage would about balance the cost of reduced treatment
facilities so that the total cost is about the same within any
reasonable range (TABLE 19). Should more sophisticated treatment
be required, larger storage and smaller treatment facilities would
afford greatest economy.
ADDITIONAL HYDRAULIC STUDIES
1. Overland flow can be adequately modeled by retaining only the
bottom slope and friction slope terms in the complete momentum
equation (Kinematic Wave Model).
2. Simulation of pipes in urban system may require that the gravita-
tional and convective acceleration terms be retained, as done in
the SWMM TRANSPORT Block over individual pipe segments. These
terms will introduce dynamic effects which exist in urban sewer
routing.
3. In cases where backwater effect (say, near a control structure)
or surcharging of pipes is present, it is inappropriate to use
the assumption that the pipe slope and the friction slope are
equal. Manning's equation can only be used in these cases if the
appropriate friction slope is computed. But in these cases, the
friction slope depends on flow conditions and is not known in
advance. In the absence of such effects, use of Manning's equation
and assuming friction slope equals pipe slope, appears to be a good
approximation in sewer design.
4. No detailed comparison of cost reduction versus number of terms
modeled for the momentum equation has been published. Most
evaluations and comparisons have focused on overall differences
between models and not on a comparison of the details of the
solutions to the basic equations. However, models which include
more than the bottom and friction slope terms would increase -the
cost of computer use. These "full equation" models need be used
only when extreme accuracy can be justified.
However, other factors such as inaccurate rainfall data, inaccu-
rate modeling of control structures, and the modifications used
to model a prototype system with a limited set of model elements,
may produce errors more significant than those introduced by
using a simple model such as the Kinematic Wave Model which
neglects the gravitational and convective acceleration terms
in momentum equation.
13
-------
SECTION III
RECOMMENDATIONS
The Rational Method should be used with extreme caution. It should
not be used for systems in which controlled routing or storage for
pollution control is desired.
In urban areas with relatively high imperviousness, combined sewers
should be considered for new development.
New combined sewers should be designed to maintain flushing velo-
city for dry—weather flows.
Urban runoff and combined sewage overflow pollution control should
be predicated on containing pollution from the less intense, frequent
storms rather than from intense rare storms.
The storage contained in trunk sewers of a combined sewer system
designed with SWMM for a 5-year or greater return frequency storm
hyetograph should be exploited by installing suitable regulators for
pollution control of runoff from less intense, frequent storms. De-
velopment of simple inexpensive regulators for utilization of in-pipe
storage is needed.
Upstream storage in the collection system (either in—pipe or satellite)
should be required in new combined systems and, where practical, for the
improvement of existing systems.
More reliable quantification of runoff in urban areas would be desirable
to determine absolute rather than relative quantities of pollutants
discharged in urban runoff and combined sewage overflows. Further
investigations might include street washout mechanisms, pollutant
constituents in dust and dirt, and the contribution of solids deposits
in combined sewers to BOD loading.
Based on computed data, control of first flush pollutants appears
to be a better criterion for Elizabeth than control of the total mass of
pollutants.
To achieve pollution control benefits from controlled routing in
Elizabeth, design based on a 5-year return interval synthetic storm
14
-------
hyetograph using SWMM can be justified. Use of a 2-year return interval
storm largely forfeits the benefits of flow routing, and the additional
storage obtained from using a 10-year return interval storm may be more
economically achieved by satellite basins.
15
-------
SECTION IV
INTRODUCTION
Concern with storm runoff primarily has considered street and base-
ment flooding. Sewers were installed to correct such problems. With
recognition of the pollutants contained in storm runoff, reduction of the
amounts of pollutants reaching natural water bodies has become a signifi-
cant factor. The pollution load placed on a stream or other receiving
waters by combined sewer overflows is substantial. It has been estimated
in a study conducted at Northhampton, England, that the total pounds of BOD
contributed yearly to receiving waters by storm generated overflows may be
equal to about 15 percent of the BOD in the dry-weather flows (1). A U.S.
Environmental Protection Agency (EPA) funded study in Durham, North Carolina,
resulted in similar findings (2).
The earliest sewers were built to convey stormwater runoff. They
were converted to combined sewers in later years as waterborne wastes were
discharged to them. Subsequently, the use of combined sewers became wide-
spread. A nationwide survey by the American Public Works Association (APWA)
indicated that drainage areas totalling over 1,214,100 hectares (3,000,000
acres) in more than 1,300 municipalities and a population of 54 million are
served by some 875,000 kilometers (550,000 miles) of combined sewers. Of the
641 jurisdictions surveyed, 493 reported about 14,200 combined sewer overflow
points. Three hundred and forty reported infiltration problems during
wet weather and 96 indicated combined sewer overflows during dry weather
(3). The cost of sewer separation in the U.S. has been estimated at about $96
billion in 1974 (4). The merit of sewer separation has been argued with
respect to effectiveness in storm runoff pollution abatement (4).
Elimination of all pollution from wet-weather flow could be prohibi-
tively expensive. There is, however, an optimal level of expenditure for
pollution control that should be justified by environmental benefits. To
enhance the urban environment and to conserve water and other resources,
the design of either combined or separate sewer systems should:
1. provide sufficient relief of street and basement flooding from
urban runoff or combined sewage,
2. limit pollution from municipal wastes and urban runoff that may
enter natural waterways to an acceptable level that will not
adversely affect the environment, and
16
-------
3. be economically feasible in context with society's needs.
Alternatives for the alleviation of street flooding include:
1. gutter inlets and collection sewers of adequate capacity to collect
and convey the defined peak urban runoff,
2. storage basins for the containment of urban runoff to reduce peak
flow entering or flowing in sewers,
3. a combination of collection sewer capacity and in-pipe storage,
and
4.' source control, including porous pavement to reduce runoff volumes,
and roof detention or ponding over designated areas to reduce peak
flows.
Alternatives for pollution control in the natural waterways from
wet-weather flow include:
1. increased interceptor and treatment capacity to convey and treat
wet-weather flows before discharge,
2. off-line storage for equalization of the treatment rate and reduc-
tion of capital investment in interceptors and treatment plants,
3. in-pipe storage for containment of pollutants in runoff and for
modulating the runoff hydrograph of flow to the interceptor system,
and
4. source control, including sewer flushing to reduce solids deposit
in sewers=, street sweeping to reduce the pollutant washout and/or
porous pavement to reduce the runoff quantity.
The alternatives for alleviation of urban flooding and for urban runoff
and combined sewage pollution abatement are interdependent. The upstream
sewer facilities planned for flood control determine the character of the
flow hydrographs and pollutographs reaching the downstream interceptor,
storage and treatment facilities and impacting on the receiving water.
Upstream storage, source control or other quantity control measures have
potential benefits in both pollution and flood control. A plan for a com-
munity-wide sewer collection system should include optimum exploitation of
those benefits.
The complex nature of the urban rainfall/runoff and the pollutant
accumulation/washout/transport processes, and evaluation of the many manage-
ment alternatives have made use of computer models advantageous. Tradi-
tionally, storm sewers are designed using the Rational Method. It
provides an estimated peak flow but not the runoff hydrograph and
17
-------
pollutograph needed for the evaluation of runoff pollutants and the design of
facilities for their control and treatment. Several more recent design
nethods for drainage systems have been proposed (5,6,7). These methods employ
nonentun equations for computing unsteady storm runoff in overland catchments
and in sewers and utilize the micro (fine discretization) approach in the
application of hydrologic and hydraulic theory.
This study was undertaken to analyze the effects of alternatives
available for the design of sewer systems considering both pollution and
flood control. The alternatives were analyzed for the urban area of
Elizabeth, New Jersey, but the principles and general conclusions should be
applicable to other urban areas.
18
-------
SECTION V
OBJECTIVES AND SCOPE OF THE STUDY
This study compares:
1. sewer system design flows using conventional and advance hydro-
logic and hydraulic methodology, and
2. the cost-effectiveness of
(a) a conventional separate storm and sanitary system,
(b) -a conventional combined system, and
(c) advanced combined systems with varying amounts of in-pipe
and/or satellite storage and controlled flow routing.
The cost-effectiveness comparison was initially based on achieving a
very high standard of pollution abatement. As the study progressed, it
became apparent that to provide an. effective basis for evaluation of alterna-
tives for pollution abatement, the scope required extension to evaluate
facilities for less complete, but acceptable, pollutant capture for treat-
ment.
The components of a sewer system include collection sewers, storage
(both in-pipe and satellite) facilities in the collection systems, inter-
ceptors, and downstream off-line storage and treatment facilities. New
systems were assumed to provide a uniform basis of comparison between
the alternatives and permit application of the findings to other cities.
However, the study also includes the case (applicable to Elizabeth and
other cities) in which the existing system can be used as a sanitary system
and a new storm system provided for pollution abatement and alleviation of
combined sewage flooding.
The scope of the study includes:
1. comparison of sewer design equations,
2. review and selection of mathematical models for evaluation of
alternatives,
3. development of synthetic storm hyetographs, hydrographs and pollu-
tographs,
19
-------
4. alternative sewer designs at a master planning level for Drainage
District A of the City of Elizabeth, with an area of 265 hectares
(655 acres),
5. economic evaluation of the alternative designs for Drainage
District A,
6. development of quantity and quality parameters for application to
the entire City,
7. design of interceptors and downstream facilities for pollution
abatement,
8. cost-effective comparison of the alternative systems for the
entire City,
9. determination of the facilities required to provide an acceptable
level of pollution abatement, and
10. the development of a methodology for planning facilities for
pollution abatement.
20
-------
SECTION VI
DESCRIPTION OF STUDY AREA
The City of Elizabeth, New Jersey (Figure 1) encompasses a total
urban area of about 1,780 hectares (4,400 acres) and has been divided
into 25 drainage districts for planning sewerage improvements. The existing
sewers are largely combined. Approximately 90 percent, by length, of the
sewers are from 50 to 100 years old, and the expected deterioration in their
condition has occurred. This is evidenced by occasional sewer collapses,
joint deterioration, and physical inspection when sewers are exposed in
excavations. In 1973, the base sewage flow in the City was estimated
as follows:
Flow Category
Residential
Commercial
Industrial
Municipal
Total
Amount (mgd)
6.5
1.1
3.4
0.6
11.6
There are 37 points in the system at which combined sewage overflows
to natural water courses during wet weather. In 1973, the principal over-
flows on the Westerly Interceptor discharged combined sewage during 61 storm
events while the principal overflows on the Easterly Interceptor discharged
sewage during 44 storm events. Discharges from the Westerly Interceptor
entered the Elizabeth River while those from the Easterly Interceptor entered
the Peripheral Ditch around Newark Airport, the Great Ditch, Newark Bay, the
Arthur Kill and the Elizabeth River. Overflows during dry weather were
experienced at Westfield Avenue, the upstream terminus of the Westerly
Interceptor. These overflows may extend for about 18 hours per day and may
be the result of a blockage in the Westerly Interceptor or of very high
infiltration in its tributary area.
Average infiltration in 1973 was estimated as 0.28 m3/sec (6.4 mgd),
or about 8.7 million m (2.3 billion gallons) per year. Infiltration
equals about 55 percent of the average dry-weather flow and indicates the
deterioration of the system' with age. Excluding dry-weather discharge of
wastes at Westfield Avenue, the total amount of infiltration jirom Elizabeth
treated at the Joint Meeting Plant was about 7.6 million
gallons) in 1973.
m
(2,0 billion
21
-------
EASTERLY
INTERCEPTOR
WESTERLY ,
INTERCEPTOR
JOINT MEETING i
PLANT
Figure 1, Study area, Elizabeth, New Jersey
22
-------
Because the increased infiltration, as noted in other studies, is
coincident with peak inflow rates, the amount of inflow treated in 1973
was only about 1.9 million m (500 million gallons) per year. Assuming
an average number of about 60 events per year during which combined sewage
overflow occurs, t|ie average quantity of inflow accepted to the Plant would
be about 30,300 m (8 million gallons) per event. The quantity of wastes
discharged' as untreated combined sewage during 1973, including the dry-
weather Westfield Avenue discharge, is estimated at about 7.6 million m3
(2.0 billion gallons). Approximately 80 percent of this discharge entered
the Elizabeth River, with the remainder discharging to the Arthur Kill, Great
Ditch, Peripheral Ditch and Newark Bay. Since such discharges can contain
significant quantities of pollutants, including oxygen demanding substances,
suspended solids and coliforms, they exert an adverse environmental effect on
the receiving waterways. The inadequacy of the existing system also resulted
in frequent flooding of streets and cellars with combined sewage. Hence, an
environmentally effective solution for correction of the Elizabeth sewer
system requires:
(1) a reduction of infiltration, and
(2) alleviation of the overflows and street and cellar flooding by
combined sewage,
to the extent justified as cost-effective.
Detailed studies were made in Drainage District A (Figure 1) to develop
criteria for water quality management for the entire City. The drainage area
contains 265 hectares (655 acres); with 47 hectares (115 acres) located in
Rosalie Park; 2 hectares (5 acres) in Union and the remainder in Elizabeth.
The area is largely developed for residential use (about 90 percent), with
some neighborhood commercial (about 5 percent) and small industrial areas
(about 3 percent). It includes most of the northwest region of the City and
is served almost entirely by combined sewers. Dry-weather flow from most of
the District is conveyed to the Westerly Interceptor which generally follows
the course of the Elizabeth River. Part of the District, generally west of
Elmora Avenue and south of Park Avenue, is served by the Joint Meeting Trunk
Sewers, which traverse the District. Dry-weather flows and limited combined
sewage flows are treated at the'Joint Meeting Treatment Plant, located
east of Bayway Avenue and north of South First Street in the City. Excess
combined sewage flows discharge to the Elizabeth River. The existing sewer
system in the District is badly undersized and six areas flood with combined
sewage at almost every significant rainfall. A preliminary design has been
prepared for a separate drainage system to relieve the overloaded combined
system. This system provides for the drainage of the six frequently flooding
areas plus other low spots, and for the separation of urban drainage and
municipal wastes. A storm drain is not provided on every street, but re-
liance is placed, wherever possible without detrimental effects, on overland
flow for the collection of urban drainage.
The population of the City is close to saturation and is expected
to have only a moderate future growth. Inadequacies in the existing sewer
23
-------
system result in numerous complaints of street and basement flooding and
of pollution in the Elizabeth River. Secondary treatment facilities are
now under construction at the Joint Meeting Plant. These facilities are
designed to provide secondary treatment for a nominal average flow of 3.3
m /sec (75 mgd) of which 0.9 m /sec (20 mgd) is assigned to
Elizabeth. By installation of a low head effluent pumping station, the
plant would, during wet weather, hydraulically provide secondary treat-
ment for flows of up to 7.9 ja. /sec (180 mgd) and primary treatment for
flows between 7.9 and 9.6 m /sec (180 and 220 mgd) at all tide eleva-
tions. The City has allocated to it, a peak wet-weather flow capacity
of 1.8 m /sec (40 mgd). The Corps of Engineers has also planned a diked
storage area, with a total capacity of about 80,000 m (21 million gallons)
along the Elizabeth River near the Joint Meeting Plant.
The Elizabeth River is tidal from its mouth at Arthur Kill for about
4,900 m (16,000 feet) to the Penn-Central Railroad. The river is small,
draining only a total of 60 square kilometers (23 square miles) at its
mouth. The analysis of the flow records at Westfield Avenue, about 5,500
m (18,000 feet) upstream from the ..mouth indicated the 7-day, 10-year return
interval low flow to be 0.21 m /sec (4.8 mgd). Samples 'taken during
slack periods at both low and high tides showed the river was polluted below
the location of combined sewage overflow outlets.
24
-------
SECTION VII
SUGGESTED METHODOLOGY
REQUIREMENTS TO BE MET
The need to evaluate the effects of:
1. random rainfall patterns and intensity,
2. the long-term and shock pollutional loads from combined sewage
overflows,
3. varying amounts, modes of operation, and locations of storage,
4. varying interceptor and treatment capacity,
5. peak flow equalization, and
6. rainfall patterns on pollutant concentration
to develop cost-effective facilities for combined sewage overflow pollu-
tion abatement is apparent. This need arises from the following under-
standing of combined sewage pollution characteristics as developed for
Elizabeth:
1. Approximately three out of four storms have peak rainfall rates
during the last three quarters of the rainfall period. Such storms
exhibit two peaks of mass pollutant washout. The first peak is
.of high-concentration, but low-flow volume, and the second of low-
concentration, but high-flow volume peak.
2. The remaining one in four storms have the unusual pattern with peak
rainfall rates occurring in the first quarter of the rainfall
period. Such storms have only one peak with the maximum SS concen-
tration about one order of magnitude less than the first peak of
the usual storm pattern, but a substantially greater volume.
3. Most of the pollutant washout in the first peak of all the usual
pattern storms, and in the frequent return interval, unusual
pattern storms, results from solids deposition in the sewers
during dry weather and is dependent upon the dry-weather sewage
characteristics.
25
-------
4. Significant amounts of street pollutant washout occur with more
intense storms than those causing the severe pollutant discharge
from sewer solids deposits. Urban runoff, therefore, is generally
lower in pollutant concentration than the first flush of combined
sewer flow.
5. Every one or two weeks, on an average, a rainfall can be expected
to cleanse essentially all dry-weather solids deposits in sewers.
Hence, capturing for treatment combined sewage generated by ordinary
rainfalls is the primary requirement for effective, combined sewage pollution
abatement.
The variables that affect the pollutants contained in combined sewage
overflow resulting from average rainfalls are many and complex. The most
significant are
1. Time of day when rainfall occurs. Combined sewage generated by a
rainfall between 3 and 6 a.m. will contain relatively small amounts
of low-strength municipal sewage, (excepting pollution from dry
x^eather flow solids deposit) and hence will be weaker than combined
sewage generated by the same amount of rainfall between noon and 2
p.m., when the greatest amounts and higher strengths of municipal
sewage are usually experienced.
2. Day of the week. Municipal sewage is usually stronger and of
greater volume on weekdays than on Saturdays, Sundays and holidays.
3. Sewer slopes. The flatter the sewer slope, the lower will be
the velocity of dry—weather flows and the more solids will be
settled and later resuspended and discharged in the combined sewage
overflows.
4. Rainfall pattern and intensity. The overflows from storms of
frequent return interval contain a higher concentration of pollu-
tants than do the greater storms.
5. Period since last rainfall. The amount of pollutants deposited in
sewers increases with the duration of the antecedent dry period as
does the street pollutant accumulation. There are large masses of
data existing in the literature indicating pollutant concentration
for various combined systems and rainfall amounts. Much of the
available data has not been correlated, because the nature of the
phenomenon may not have been fully understood, and all the neces-
sary information may not have been obtained. (10,15)
The selection of a "design storm" or a series of "design storms"
would not assess adequately many of the significant pollutional aspects
of combined sewage overflows. Hence, development of a "design storm" for
26
-------
pollution control is not a viable approach. A long-term (12-year) continuous
rainfall record and calculated runoff and quality characteristics, based on
STORM default values, were used in this study to determine the relative
effectiveness of alternative facilities for pollution abatement.
METHODOLOGY OUTLINE
A practical methodology has been developed in this study for the anal-
ysis required to determine cost-effective sewer systems in urban areas for
combined sewage overflow pollution abatement. The following outlines the
tasks of the suggested methodology:
1. Delineate the drainage districts of the study area and select a
typical drainage district for detailed analysis.
2. Prepare data for the typical district for input to computer models.
This would include:
(a) The existing long-term (10-year) precipitation and air
temperature data applicable to the area to account for the
random nature of rainfall events.
(b) Development of synthetic storms of 1-month, 6-month, 1-year
and 5-year return intervals to determine overflow quantity
and quality for the various alternatives for pollution
control.
(c) Subdivision of the typical drainage district and development
of data such as land use, street length, population, sewer
network and sizes, and street sweeping practice for input to
SWMM.
(d) Preparation of district-wide data (land use, population, area
and street length, etc.) of all drainage districts to develop
the total combined sewage flows for the City.
(e) Development of dry-weather flow quantity and quality includ-
ing infiltration.
(f) Preparation of cost data, including costs of sewer pipes,
storage, pumping stations, treatment, flushing, chemicals and
regulators.
3. Evaluate alternatives in the typical drainage district using
computer technology.
This would include:
(a) Field sampling and calibration of computer models for both
quantity and quality of combined sewage.
27
-------
(b) Analyzing the capacity of existing sewer systems.
(c) Establishing the characteristics of runoff quantity and
quality from the existing system for various synthetic or
real storm events.
(d) Using STORM and SWMM to evaluate the combined sewage over-
flows pollution reductions to be achieved by various alter-
natives, including:
1) upstream storage either in detention basins or sewers,
2) controlled routing in sewers,
3) sewer separation,
4) sewer rehabilitation,
5) overflow containment,
6) overflow treatment with and without containment, and
7) interceptor capacity upgrading.
(e) Comparing the cost-effectiveness of alternatives.
(f) Developing parameters of runoff quantity and quality (i.e.,
normalized hydrographs and pollutant loading for various land
uses) for application to the remaining districts of the
planning area.
4. Cost-effectiveness evaluation of alternatives for the entire
planning area. The "Effectiveness" and "Efficiency" Indices may
be used for scanning the various alternatives.
Index" is obtained as follows:
The "Effectiveness
x BOD(mg/l)
x Annual Overflow Volume (MG)
Average Overflow Return Interval (Days)
Physically, it represents the relative quantities of pollutant mass
per day in the combined sewage overflows during overflow periods.
The "Efficiency Index" is the product of the Effectiveness Index
and the cost of the sewer system in dollar units. These parameters
also permit development of the optimum level of treatment beyond
which effectiveness diminishes markedly with increased cost incre-
ments.
RESULTS ATTAINED
The methodology develops, for the various alternatives, the average
28
-------
number of annual overflows, the annual volume of combined sewage overflows,
the duration of the overflows, and the annual mass of BOD and SS discharged.
Combined sewage discharges are, by their nature, a severe shock load
on the receiving waters.
The effectiveness of the various alternatives with respect to mitigating
shock discharge loads are compared using the "Effectiveness Index". The
logic for selecting the parameters which comprise this index follows:
1. The longer the return interval between overflows, the less should
be the impact of such flows on the river system.
2. The impact on the river system would be proportional to the concen-
tration of pollutants discharged thereto. Hence, the use of the
geometric mean weights the effects of both SS and BOD.
3. The total volume of overflow would also affect the river quality.
The lower the value of the "Effectiveness Index", the less should
be the impact of combined sewage overflows on the receiving water.
The "Efficiency Index" determines the most efficient alternative. The
higher the "Efficiency Index", the less desirable is the alternative.
The "Efficiency Index" is a measure of cost-effectiveness. The relationship
of the "Efficiency" and "Effectiveness" Indices should also be compared for
the various alternatives. If Alternative A has a lower "Effectiveness Index"
but a higher "Efficiency Index" than Alternative B, it means the incremental
work proposed to achieve Alternative A may be too expensive and other alter-
natives should be explored.
SUPPORTING APPENDICES
This report contains five appendices which provide background data for
the methods used in this study. Appendix A presents the justification for
using SWMM and STORM. Appendix B discusses modifications made to SWMM and
STORM. Appendix C compares the differences in the two models with respect to
determinations of quantity and quality. Appendix D describes, in detail, the
required data . for computer models, including meteorological data,- the syn-
thetic hyetograph development, sewer network configuration, hydraulic proper-
ties and land use of District A for both SWMM and STORM, discrete data, the
area and land use characters for the remaining urban area of Elizabeth,
dry—weather flow quantity, quality and characteristics, and basic cost data.
Appendix E describes the method of calibrating the STORM runoff coefficient.
29
-------
SECTION VIII
DESIGN AND COST COMPARISON OF SEWER SYSTEMS IN DISTRICT A
Designs of the three types of alternative sewer systems (conventional
combined, conventional separate, and advanced combined with varying amounts
of storage) were made for District A and respective costs estimated, assuming
new sewer systems. The effectiveness of the alternative sewer systems in
District A in controlling wet-weather flow problems (flooding and pollution)
were also evaluated. New sewer systems were assumed in the evaluation to
permit a uniform comparison and the transfer of the study findings to other
districts within the City and to other urban areas. —
The conventional separate sewer system consisted of two systems, one to
carry wastes from domestic, commercial and industrial discharges together
with such infiltration/inflow as must be anticipated, and the second to carry
storm runoff. There are no storage facilities or flow regulators in either
system. In the sanitary sewer system, almost every street contains a
sewer, while in a storm sewer system, street gutter and overland flow to the
maximum extent practical is allowed. Consequently, the total length of the
storm sewer system is shorter than that of the sanitary sewer system. In
newly developing areas, where ponding for urban runoff can be provided, storm
sewer sizes can also be significantly reduced.
In a conventional combined system, sewers receive both municipal waste-
water and stormwater. In this study, by definition, flow control devices
and/or storage facilities were not provided in the collection system for
control of combined sewage flow rates and quality. Collection sewers (in-
cluding laterals and trunks) were used for conveyance of both dry-weather and
wet-weather flows. Combined sewage flow rates in excess of the capacity of
the intercepting sewers (which carry flow to storage/treatment facilities)
overflow to receiving waters. A combined sewer system and a separate sani-
tary sewer system have the same total length. As street runoff normally
enters inlets located at the downstream end of a street, the sewer in the
upstream reach of a sewer network branch would usually carry sanitary waste-
water only. Consequently, the length of sewers carrying combined sewage is
less than the total length of a separate sanitary system. In the case of
District A, about 40 percent of the length of sewers in the combined system
would accept only sanitary sewage. Modification of the combined system
which permitted the same amount of overland flow as the separate system was
also evaluated.
The advanced combined sewer systems were designed to provide various
amounts of storage for wet-weather flow control. Two types of storage were
30
-------
considered: in-pipe storage and satellite storage. In in-pipe storage
systems, sewers are increased in size to provide volumetric storage in
addition to flow conveyance. Regulators are installed ,to permit use of
in-pipe storage to reduce peak flows, thus increasing the amount of polluted
runoff intercepted and later treated for equal interceptor capacities.
In-pipe storage has been successfully implemented in the existing combined
sewer systems of Seattle, Detroit, Cleveland and Minneapolis-St. Paul. The
storage may be incorporated in upstream lateral sewers or downstream trunk
sewers or both. In this study, in-pipe storage was found to be more econom-
ically located in upstream lateral sewers, as will be described more fully
hereinafter. There is only about half the volume in the lateral sewers
as in the larger size trunk sewers in a conventional combined system, and the
additional storage for both pollution abatement and flood control may be
obtained more economically by enlarging smaller size rather than larger
size sewers.
In a satellite storage system, collection sewers are designed for flow
conveyance, but detention basins are located in the collection sewer system
to reduce peak flows. There would be little volume available in the sewers
for storage because of the small volume in the laterals and reduced size and
volume in the trunks. The detention basins could be open, covered or tun-
neled structures. In the literature, "satellite storage" is frequently
referred to as "off-line storage". In this study, however, the term "off-
line storage" is reserved for flow equalization basins located near the
sewage treatment plant. In both the conventional and advanced combined
systems, off-line storage may be provided. Satellite storage can be con-
structed at any location within a drainage district where there is sufficient
flow to use it effectively. With respect to the city-wide sewer network,
satellite storage should be viewed as upstream as opposed to off-line or
downstream storage. There could be several satellite storage structures at
strategic locations in each of the 25 drainage districts in Elizabeth.
The number of off-line storage basins would be much fewer. As both in-pipe
and satellite storages are located in the collection sewer system, both will
be generally termed as "in-system" storage.
Since the basic difference between an advanced and a conventional sewer
system is the in-system storage provided, the sewer layout and total sewer
length of the two systems are identical. However, the pipe diameters differ
significantly.
DESIGN PHILOSOPHY
The selection of the rainfall event or events on which to base the
design of a combined or separate system for urban runoff is of major impor-
tance, and has, too often, been only casually addressed. The traditional
approach has been to:
1. develop the intensity-duration-frequency curves for the location
under study. These are normally available from U.S. National
Weather Service Technical Paper No-s. 25 and 40 (37,38), or other
such sources.
31
-------
2. select a return interval for which it is desired to provide flood
protection. Typically, a return interval of two, five or ten years
has been used.
3. develop a design storm hyetograph from the intensity-duration curve
for the selected return interval. This storin is a synthetic
rainfall event which may be based on meteorological probabili-
ties or may be an envelope storm which contains within a single
event the depth—duration characteristics of all naturally occurring
events likely within the chosen return interval (9)•
4. Use the intensity-duration curve directly in situations where a
synthetic storm hyetograph is not developed, e.g., when the analy-
sis of the system is to be performed using the Rational Method.
This use has similar characteristics as an envelope event.
During the past five years, as use of relatively powerful computer-
based models such as SWMM has grown, users continued to develop the required
"design storm" input in the manner described above. However, the .increased
capabilities of the model compared to hand-computation techniques has led to
numerous questions concerning the "shape" of the design storm and the appro-
priateness of any single storm for the design of the system. The usual
questions are related to the following items:
1. Where should the peak of the "design storm" be located? The
intensity-duration—frequency curves do not include any information
on the "shape" characteristics of naturally occurring rainfall
events.
2. What antecedent conditions should be used?
At the same time as these issues were being raised, the developemnt of
the various Water Quality Control Acts has forced the designer to consider
quality as well as quantity in the design of a storm sewer system. Only
recently has it been appreciated that the return interval appropriate for
quantity control is not appropriate for the elimination of the typical
pollution-causing discharge. The runoff quality from any event is largely
a function of the pollutant which has accumulated in the sewer and on the
overland surface since the last rainfall event. The following aspects
of storm selection now have been recognized:
1. Rainfall rates of about 25-4 mra/hr (1.0. in./hr) are sufficient to
washoff about 90% of the pollutant from the surface (6) . Slightly
lower rates will be sufficient for paved surfaces, while higher
rates may be required to cause significant erosion from non-urban
areas. Substantially lower rates should flush out most sewer
solids deposits.
2. Rainfall depths of between 5.1 and 12.7 trim (0.2 and 0.5 inches) are
sufficient to washoff most of the accumulated pollutants from a
32
-------
typical urban area.
3. Since the available pollutant is largely a function of the inter-
vals between rainfall events, the distribution of storms within the
typical year is significant.
The majority of the polluting overflows result from small events
(as compared to the traditional design event). Hence, the system must
be able to handle the hydraulic design storm without excessive flooding,
while, at the same time, it must be capable of regulating the frequent
rainfall runoff which, if allowed to discharge freely, would result in
major pollution. Several models such as STORM were developed to allow analy-
sis of the performance of the proposed/present system under long-tern rain-
fall behavior. These models employ simplified hydraulics, and little or no
sewer network data, and can not adequately estimate the response of the
system to the large storm event. Rather, they serve as a mechanism for
evaluating the type of control strategies which should be employed.
The availability of both SWMM and STORM provides the mechanisms to
evaluate both the "day-to-day" and "rare-design-storm" characteristics
of this system. However, the inputs to the models must be both compatible
and representative of the rainfall characteristics of the area. For the
present study, the following inputs were adopted:
1. The continuous hourly precipitation data at the Newark
International Airport gauge was used as input to the STORM program.
This location is quite close to the study area, and its rainfall
characteristics are quite similar. Rainfall data for the 12-year
period from 1963 to 1974 was used, and was assumed to be uniform
within the study area.
2. A 5-year return interval intermediate patterns envelope type
storm hyetograph was used as the design criteria for flood protec-
tion for the City of Elizabeth. The intensity-duration curves for
Sandy Hook, New Jersey, were used as a basis for the synthetic
storm generation. An analysis of the 12-year Newark Airport data
used, indicated close agreement between the intensity-duration
curves for Newark and Sandy Hook. However, the latter was used for
the development of the hyetograph since they represent an analysis
of a much longer period of record than the 12-year Newark data.
It is felt that this combination of rainfall inputs can confidently be
used to design a system which is capable of controlling the pollutional
aspects of typical daily or weekly flows, while at the same time providing
runoff control capabilities sufficient to handle the hydraulic flows of the
5-year event.
The synthetic hyetograph, as developed in Appendix D, was used to
design alternative facilities for evaluation. This type of hyetograph was
conceived in 1959 by the senior author after investigation of failures
resulting with drainage structures designed using meterological
33
-------
hyetograph (9) or other methods. The total rainfall precipitation repre-
sented in this hyetograph at any elapsed time should about equal the histori-
cal total for that time interval, and the total volume of runoff may be
expected to equal, within reasonable limits, the total for the interval at
the stated return period interval. However, the actual rainfall intensities
for shorter increments within the selected time interval nay be expected to
vary from, and be less than, those assumed in the synthetic ,_hyetograph.
Hence, the peak flows experienced will probably be less than flows computed
with the hyetograph.
The magnitude of the difference between the flow rates obtained using
the synthetic rainfall hyetograph and the probable 5-year peak runoff
flows is influenced by the short time of travel for peak runoff flows from
the upstream reaches to the overflow point. Previous studies in
Elizabeth (8) determined that, for 93 drainage areas containing less than
49 hectares (120 acres) each, the time of travel was less than 12 minutes.
For District A, the time of travel is about 20 minutes. Such short times of
travel are frequently found in cities with populations of about 100,000 or
less. Based on analyses by others (9), the short increment rainfall rates of
5—year storms of various durations would be as follows, when expressed as a
multiple of the increments of a 360-minute duration rainfall.
Time
Increment
(minutes)
5
10
15
20
<
30
60
180
Total Rainfall Duration (Minutes)
5 30
1.37 1.22
1.20
1.18
1.18
1.16
-
_ _
60
1.16
1.15
1.14
1.13
1.11
1.10
_
180
1.06
1.05
1.04
1.05
1.05
1.04
1.03
360
1.0
1.0
1.0
1.0
1.0
1.0
1.0
Hence, at the point where a system would be stressed by a 5-ninute
peak rainfall rate, computations using the synthetic hyetograph, which
satisfies the intensity-duration curve at each time interval, would be about
30 percent greater than that obtained from the peak 5-minute flow from a
180-minute duration storni. By extrapolation, the following relations with
respect to the 180^-minute storm were developed for reaches which would be
stressed by other time intervals.
34
-------
Time
Interval (min.)
% Greater Flow from
Synthetic Hydrograph
10
15
20
25
30
27
24
21
18
15
It is improbable that a 5-year return frequency storm of duration
equal to the most severe flow generating time for each sewer reach would be
experienced every five years. Hence, the synthetic hyetograph chosen may
generate peak flow rates that are, perhaps, 15 to 30 percent greater than
those experienced, although the total runoff volume required for design of
pollution abatement facilities should be about right. As demonstrated
hereinafter, the use of the 180-minute duration hyetograph should permit
determination of the runoff volume required to develop pollution abatement
facilities.
The design philosophy and methodology developed in this study is based
on a synthetic hyetograph, which reasonably establishes runoff volumes
required for effective pollution abatement for design of alternatives. The
methodology permits determination of the effects of a wide variation in
storage volumes on the design of facilities. The alternatives developed are
tested using a long-term continuous record of real rainfall events to deter-
mine their effectiveness for abatement of pollution from combined sewage
overflows. No attempt is made to develop a design storm or < storms with
respect to pollution abatement.
.CONVENTIONAL COMBINED SEWER SYSTEM
Design
The new conventional combined sewers in District A were designed using
SWMM with capacity to convey runoff from a 5-year return interval, inter-
mediate pattern synthetic storm. The combined sewer system layout is shown
in Figures D-8 to D-12. The total length of the system is 32,308 m (106,000
feet), of which 19,397 m (63,640 feet) carry combined sewage and 12,911 m
35
-------
(42,360 feet) carry sanitary sewage. The sewers carrying sanitary sewage
are all 0.2 m (8 inches) diameter. >
The total length of lateral sewers conveying combined sewage (those pipe
elements numbered less than 300) is 15,651 m (51,350 feet) and of trunk
sewers 3,746 m (12,290 feet). There are 139 designated lateral sewers
conveying combined sewage and 31 trunk sewers in District A, with pertinent
pipe elements shown in TABLE D-7. These sewers receive runoff from 279
subcatchments with subcatchment characteristics tabulated in TABLE D-6.
SWMM allows routing rainfall runoff in sequence from overland through
lateral sewers and trunk sewers to the outfall. Lateral sewers were sized
with the SWMM RUNOFF Block and trunk sewers with the SWMM TRANSPORT Block.
The design diameters of the lateral sewers for the combined system are shown
in TABLE 1 and of the trunk sewers in TABLE 2, under 0 percent in-pipe
storage. Design diameters for the advanced combined system are also shown
for comparison and later discussion.
Two sets of design diameters are shown in TABLE 1 (laterals) and TABLE 2
(trunks), for the combined system. The lateral sewers were designed with two
methods of flow routing. In both cases, the hydrographs of the 5-year
synthetic storm from tributary subcatchments to lateral sewers (gutter and
pipes) are identical and are computed with the SWMM RUNOFF Block (up to
Subroutine WSHED). In computing the set of design diameters for lateral
sewers not in parentheses, the design flow was the peak of the inflow hydro-
graph developed by adding the inflow hydrographs of upstream pipes and local
tributary subcatchments using a subroutine derived for this study. The
design flow, without reduction due to pipe storage, was assumed to enter the
upstream end of the pipe. Manning's equation was used to compute the re<-
quired pipe diameter for full flow condition and the next largest commercial
size was used as the design diameter. This method was developed to permit
determination of peak flow with in-pipe storage as later described. The
lateral sewer sizes in. parentheses were computed with the Subroutine GUTTER
of the SWMM RUNOFF Block which uses Manning's equation to determine maximum
pipe flow together with storage routing using the continuity equation. Where
only numbers not in parentheses are shown, there is no difference in pipe
diameter under the two methods.
Sewers sized assuming no storage routing or time lag in the hydrograph
are generally at least as large as those sized considering storage routing or
time lag of hydrograph. Of the 139 lateral sewers considered, 111 have the
same sewer diameter and 19 designed with the first method are greater and 9
are smaller in diameter than those designed with the second method. However,
some significant reductions are found in element numbers 210, 225, 239, 237
and 219, which are the larger diameter and more costly pipe reaches.
With respect to in-pipe storage pipe sizes, discussed later, pipe
element number 105 decreases in diameter with increased total volumes
of in-pipe storage. This results because of the moderating effect of outflow
from pipe element number 49 and no tributary area directly drains to it.
Pipe element number 77 is the minimum size used in a combined sewer and
36
-------
TABLE 1. LATERAL SEWER DESIGN DIAMETERS (inches)
Pipe
Element
No.
82
101
99
98
95
94
92
91
189
187
85
104
103
108
112
119
78
,80
83
121
123
125
126
84
128
148
149
147
146
143
141
139
138
135
87
89
133
1
3
4
7
8
0%*
27
36
30
27
36 (42)
36
42 (48)
36 (30)
48
36 (40)
36
21
27
36
36 (30)
30
36
36
30 (36)
27
30
36
42
42
36
27
18
21
36
48
48
60
21 (18)
48
48
54
54
27
30 (36)
27
18
24
In-pipe
8.3%
60
48
48
48
36
42
48
42
48
65
48
42 '
36
48
48
48
36
42
24
48
42
42
30
60
54
36
24
42
60
60
60
66
30
60
42
60
48
48
48
54
36
36
Storage
16.5%
84
60
60
60
42
54
60
48
60
90
60
60
48
60
60
60
42
54
21
60
54
54
27
78
66
42
36
54
72
72
78
78
42
72
48
66
48
60
66
66
48
48
Volume**
25.0%
102 .
72
66
72
42
60
66
60
66
108
66
72
54
72
66
72
48
60
18
72
60
60
24
96
78
48
42
66
90
84
90
84
48
84
54
78
54
72
78
78
54
54
41.3%
126
90
84
90
54
72
84
72
84
132
84
90
72
90
84
90
60
72
18
90
78
72
18
120
96
60
48
84
114
108
114
108
60
102
60
96
60
90
96
102
66
66
NOTE:
Conventional Combined Sewer System. Numbers in the
parenthesis are pipe sizes obtained from SWMM and
should replace the numbers not in parenthesis as
the combined sewer sizes.
Advanced Combined Sewer System. Storage volumes as
percent of 119,000 m (31.5 million gallons) (runoff
volume of the 5-year synthetic storm for District A).
37
-------
TABLE 1. (continued)
Pipe
Element
No.
10
13
14
15
20
19
24
25
29
30
33
31
35
42
40
38
36
47
49
56
52
54
55
114
117
59
72
74
75
67
105
109
129
77
76
71
69
64
62
60
58
201
202
203
204
205
206
207
208
212
213
214
215
0%*
27 (30)
27 (24)
30
21
27
30
36 (30)
21
24
36
27
36
30
21
27
30
48
24
27 (24)
27
36
36
42
42
27
36
27
27
36 (30)
27 (24)
27 (24)
36
21
18
21 (18)
36
42
54
54
54
48
36
54
54
54
54
54
42
30
42
42
48
27
In-pipe
8.3%
48
48
36
42
42
48
48
36
42
54
48
48
24
42
42
42
48
42
36
42
54
42
42
36
42
42
42
54
36
42
21
48
24
18
36
36
60
54
54
50
42
84
54
96
48
60
54
42
48
60
42
60
54
Storage
16.5%
60
60
42
54
54
60
60
48
60
72
66
66
21
54
60
54
54
54
42
54
66
48
54
27
54
48
60
72
42
60
18
54
27
18
42
42
72
60
54
72
48
114
60
132
48
72
60
48
60
84
48
78
66
Volume**
25.0%
72
72
54
60
60
66
72
60
66
84 •
78
78
18
66
66
66
60
60
48
66
78
54
60
24
66
54
66
84
54
66
18
66
30
18
48
42
84
66
60
78
48
138
72
156
54
84
66
48
72
96
54
90
78
41.3%
90
90
60
78
78
84
84
78
84
108
102
96
18
78
84
84
72
78
60
84
96
56
72
21
84
66
84
108
60
84
18
84
42
18
60
54
108
84
72
102
60
174
84
192
60
108
84
60
84
120
66
114
102
38
-------
TABLE 1. (continued)
Pipe
Element
No.
216
209
210
211
220
221
227
226
225
228 ,
229
230
233
222
232
235
234
231
224
223
236
238
246
249
251
253
252
250
248
247
245
244
243
242
241
240
239
237
260
261
262
263
264
219
0%*
. 36
48 (54)
60 (54)
48
30
36
27 (24)
30
42 (36.)
27 (24)
30
36
30
36 (42)
42
24
27
42 (36)
48
66
24
27 (30)
24
27
27
27
36
42
48
54
60
24
66
54
54
54
60 (54)
66 (60)
36
36
54 (48)
21
36
102 (96)
In-pipe
8.3%
42
48
54
42
48
60
30
48
54
' 42
54
36
60
66
48
36
30
54
36
60
36
60
36
42
42
42
48
42
36
54
60
36
60
60
60
60
48
60
54
54
54
30
60
84
Storage
16.5%
48
48
54
48
66
84
36
54
66
54
72
42
78
84
60
48
36
66
30
60
48
78
48
54
60
54
60 .
48
36
60
66
48
66
66
72
72
54
72
72
66
66
42
78
78
Volume**
25.0%
54
54
60
54
78
96
42
66
78
60
84
48
90
102
66
60
42
78
27
66
54
90
60
66
66
66
72
54
42
66
72
60
78
78
84
84
60
78
84
78
78
48
90
78
41.3%
66
66
72
60
96
126
54
84
96
78
108
54
114
126
•78
72
48
102
24
72
66
120
72
84
84
84
90
60
42
84
84
72
90
96
102
102
66
96
108
96
96
60
120
84
NOTE:
Conventional Combined Sewer System. Numbers in the
parenthesis are pipe sizes obtained from SWMM and
should replace the numbers not in parenthesis as
the combined sewer sizes.
Advanced Combined Sewer System. Storage volumes as
percent of 119,000 m (31.5 million gallons) (runoff
volume of the 5-year synthetic storm for District A).
39
-------
IT1OOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
co r^inoinoinoininooinooooinininooooooininoooo
•if
0)
« ^000000000000000000000000000000
O I
4J 1
CO
. . pptnplOCDpOOCDOO.OOOOOOOOOOOOOOOOOQO
CO
-a-
a)
^ fr? motnoooooooooooooooooooooooooooo
* o
OJ <
(0
u
o
CO^S oinOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
_D|v£3 MNcoc^co| mo ooomoooo o m _!^ _^_I_1_!I
£2u3 •• •••••••• • • CO-
o o
M 4-1
cu
a.<-~
CO
iw C
O O
et)'
00
a -2
•H rH
co •
0) rH
g en
rH
CXO
> 0
a o
ooo
ni o
rJ «
o a\
4J rH
CO rH
40
-------
is not reduced in size as a result of moderated outflows from pipe element
number 129. Pipe element number 219 is reduced in size with in-pipe storage
because of the moderating effects of outflow from upstream pipes.
A similar comparison of trunk sewer sizes for the conventional combined
system is given in TABLE 2. The difference in sewer sizes, with four excep-
tions, is no more than one commercial size and is due to differences in
hydrographs entering the trunk sewer inlet points since the same TRANSPORT
Block was used to size the diameters. Of the 31 trunk sewer conduit elements
considered, 13 have the same sizes. Twelve associated with the first method
are greater and six are smaller than those associated with the second
method. However, the increases are in the larger size pipe although the
largest and second largest size pipes, (element numbers 349 and 351) are
reduced in size. The difference in diameters sized with two design methods
reflects the small pipe storage effect for a 5-year synthetic storm. Those
diameters obtained with subroutine GUTTER of SWMM (as modified) were used as
the combined sewer sizes.
The initial diameters specified for each conduit element when using SWMM
for conduit sizing affect the final design diameters as well as the outflow
hydrograph and pollutograph. This results since SWMM does not recompute the
previously computed portion of the hydrograph when pipes are resized due to
the occurrence of surcharge at a later time in the design storm. Generally,
the effect on the design diameter is small but could be significant on the
peak outflow hydrograph.
The trunk sewer decreases as the amount of in-system storage increases.
This results from the reduced peaks of the outflow hydrographs from either
satellite storage or in-pipe storage. This characteristic is a significant
factor in determining the least-cost system for both pollution and flood
control.
Outflow Hydrographs and Pollutographs
The 5-year return interval synthetic storm and outflow hydrograph from
District A for the conventional combined system is shown in Figure 2. The
peak outflow rate is 39.8 m /sec (1,407 cfs). A 2-minute integration
interval was used. The outflow hydrograph maintains a relatively low value
until about one hour after the rainfall starts. The time lag between the
peak of the hyetograph and the peak of the outflow hydrograph is about 12
minutes, indicating a short response time of the drainage system to the
synthetic storm. The time of travel in the trunk sewers, estimated assuming
all pipes flowing full, is about twice as long as the time required for the
peak flow to be generated at the downstream outlet.
As discussed previously, the initial diameters specified as input to
SWMM affect the design diameters and the flow routing. Hence, the outflow
hydrograph and pollutographs are influenced. For example, Case A of TABLE 3
specified initial pipe diameters to be about one-half of the final design
diameter and results in a peak hydrograph of value 44.4 m /sec (1,569
cfs). In Case B, initial diameters were assumed to be one or two commercial
41
-------
(jnoq/ui) A1ISN31NI
in
o>
c
'i
cr
o
i-
co
I-
cc
- CO
UJ
O
z
CO
O
ni
tu
4J
O
•H
M
V
w
•H
Q
00
O
>,
=)
0
0
J-l
cu
T3 ft
n3 0)
4-1
,£ n)
ft -H
a T)
i-i QJ
oo d
O M
4-1 0)
a) 4-1
CX|
0)
60
•rf
(sp) 31Vd MOHd
42
-------
TABLE 3. COMPARISON OF DESIGN DIAMETERS AND PEAK
HYDROGRAPHS
Trunk
Sewer
No.
301
303
305
307
309
311
313
315
317
319
321
323
325
327
329
331
333
335
337
339
341
343
345
347
349
351
353
357
359
361
363
Case A
(1) (2)
1.5
1.5
1.5
1.5
2.0
2.0
3.0
3.0
3.0
2.5
4.0
4.0
4.0
3.0
3.0
4.0
3.0
3.0
3.0
4.0
4.0
4.0
5.0
4.0
4.5
4.5
6.0
5.0
5.0
5.0
5.0
3.0
4.0
4.0
4.5
5.0
6.0
6.5
7.0
7.5
6.0
8.5
8.0
• 8.5
6.5
7.5
9.0
6.5
7.5
7.5
9.0
10.0
10.0
10.0
12.0
13.0
14.0
12.5
11.0
11.0
11.0
11.5
Case B
(1) (2)
2.5
3.5
* 3.5
4.0
4.5
5.5
6.0
6.5
7.0
5.5
8.0
7.5
8.0
6.0
7.0
8.5
6.0
7.0
7.0
8.5
9.5
9.5
9.5
12.0
13.0
13.0
11.0
10.5
10.5
10.5
10.5
3.0
3.5
4.0
4.5
5.0
6.0
6.0
6.5
7.0
6.0
8.0
8.0
9.0
6.5
8.0
9.0
6.5
7.5
7.5
9.5
10.0
10.0
10.0
13.5
13.0
14.0
12.5
11.0
11.0
11.5
11.5
Case C
(1) (2)
3.0
3.5
4.0
4.5
5.0
6.0
6.0
6.5
7.0
6.0
8.0
8.0
9.0
6.5
8.0
9.0
6.5
7.5
7.5
9.5
10.0
10.0
10.0
13.5
13.0
14.0
12.5
11.0
11.0
11.5
11.5
3.0
3.5
4.0
4.5
5.0
6.0
6.0
6.5
7.0
6.0
8.0
8.0
9.0
6.5
8.0
9.0
6.5
7.5
7.5
9.5
10.0
10.0
10.0
13.5
13.0
14.0
12.5
11.0
11.0
11.5
11.5
Peak hydro-
graph flow
Note:
1569 cfs
1509 cfs
(1) Input diameters in feet
(2) Design diameter in feet
1 foot = 0.3048 meters
1 cfs = 0.0283 m /sec
1407 cfs
43
-------
sizes less (using Case A as a guide) than the final design diameters.
The final design diameters were generally the same as or one commercial
size different from those of Case A and the peak outflow hydrograph value
from District A was reduced from 44.4 to 42.7 m /sec (1,569 to 1,509
cfs). In Case C, the initial pipe diameters were specified to be equal to the
final design diameters of Case B. Sewer routings were consistently made
throughout the duration of the synthetic storm. Because of proper sewer
routing, of design storm flow, the peak flow rate was reduced from 42.7 to
39.8 .BI /sec (1,509 to 1,407 cfs). The hydrographs corresponding to Case C
of TABLE 3 were used for analysis.
+
Figures 3 and 4 preserit the outflow pollutographs in pounds per minute
from the combined sewer system for SS and BOD, respectively. They assume
four dry days prior , to the occurrence of the storm. The street sweeping
interval assumed was seven days (as now practiced in Elizabeth) and the
sweeping efficiency 75 percent. STORM default values were used in computing
pollutant constituents (SS, BOD) from the amount of dust and dirt in the
street washout.
Figure 3 illustrates the two flushes in a combined sewer system.
The first flush ends about 50 minutes from the start of the storm and the
second flush, about 40 minutes later. The first flush reflects the deposit
of solids in combined sewers from sanitary wastes during dry days and their
resuspension in the initial phase of the storm. TABLE 4 shows the amount
of solids deposited in each trunk sewer segment in District A during four
antecedent dry days as computed with the SWMM TRANSPORT Block. The average
daily solids deposit is about 127 kg (281 pounds). Most of these .deposits
occur in pipe element numbers 347, 349 and 351, which have the relatively
flat slope of 0.0004. The solids deposits in these three elements account
for about 84 percent of the total. This indicates the advantage of
maintaining flushing velocities for dry-weather flows. A rainfall with an
intensity of about 1/16 of the 5-year synthetic storm and with a return
frequency of less than one month (about ten days)- should cleanse these
deposits in the sewers. The peak runoff rate from this small, frequent
rainfall is 0.76 m /sec (27 cfs) as compared to 39.8 m /sec (1,407 cfs)
for the 5-year design storm. This flow rate is about ten times the average
dry—weather flow.
Figure 4 presents mass rate BOD pollutographs associated with the
5_year synthetic storm for both the separate storm and combined systems.
The double peak, observed in the SS pollutograph, is not as apparent since,
in the SWMM used, solids deposited in sewers are assumed to contribute only
to SS and not to BOD. Such contribution should be considered based on the
anticipated organic content associated with all sewer solids.
Had the number of antecedent dry days been more than four, the first
peak of the mass rate SS pollutograph would be increased since solids deposit
in combined sewers would become a dominant factor in total runoff polutants.
The pollutograph, shown in Figure 5, .assumes 30 antecedent dry days. The
BOD for either the combined or separate system would mainly result from
street washout as shown in Figure 6 by the similarity of pollutographs
for both the combined and separate systems.
44
-------
in
XI
CONVENTIONAL
COMBINED SYSTEM
CONVENTIONAL
SEPARATE SYSTEM
ASSUME:
7 DAY STREET SWEEPING INTERVAL
0.75 SWEEPING EFFICIENCY
STORM DEFAULT VALUES FOR
POLLUTANT CONVERSION FROM DD
5-YEAR INTERMEDIATE
PATTERN STORM
6O |20
TIME SINCE START OF STORM (minutes)
180
Figure 3. Mass rate SS pollutographs, 4 dry days (5-year intermediate
pattern storm)
CONVENTIONAL
COMBINED SYSTEM
CONVENTIONAL
SEPARATE SYSTEM
5-YEAR INTERMEDIATE
PATTERN STORM
60 120
TIME SINCE START OF STORM (minutes)
ISO
Figure 4. Mass rate BOD pollutographs, 4 dry days (5-year interme-
diate pattern storm)
45
-------
TABLE 4. SOLIDS DEPOSIT (LBS) IN SEWER WITH FOUR
ANTECEDENT DRY DAYS
Element
Number
301
303
305
307
309
311
313
315
317
319
321
323
325
327
329
331
333
335
337
339
341
343
345
347
349
351
353
357
359
361
363
Solids
fibs)
6.48
4.88
3.52
3.61
4.23
5.65
5.87
6.53
7.18
0.37
15.36
12.41
11'. 66
0.40
0.67
12.78
0.29
0.52
0.51
2.60
13.80
13.76
13.75
311.61
306.40
322.76
28.69
1.73
1.86
1.92
1.92
Note: 1 Ib = 0.453 kg
46
-------
400
300
100
CONVENTIONAL
COMBINED SYSTEM
CONVENTIONAL
SEPARATE SYSTEM
ASSUME:
7 DAY STREET SWEEPING INTERVAL
0.75 SWEEPING EFFICIENCY
STORM DEFAULT VALUES FOR
POLLUTANT CONVERSION FROM DD
5-YEAR INTERMEDIATE
PATTERN STORM
Figure 5.
60 120
TIME SINCE START OF STORM (minutes)
180
Mass rate SS pollutographs, 30 dry days (5-year interme-
diate pattern storm)
20 i-
CONVENT10NAL
COMBINED SYSTEM
CONVENTIONAL
SEPARATE SYSTEM
5-YEAR INTERMEDIATE
PATTERN STORM
60 120
TIME SINCE START OF STORM (minutes)
180
Figure 6.
Mass rate BOD pollutographs, 30 dry days (5-year interme-
diate pattern storm)
47
-------
To control the initial flush of pollution from solids deposits in
the sewers, it would be necessary to store and treat a volume of about 1.6 mm
(0.063 inches) over the drainage basin for a 5-year synthetic storm of
intermediate pattern. The volume requiring storage is small since most of
the solids are deposited in the larger and more flatly sloped downstream
reaches of the sewer system. This condition is generally typical in most
urban areas. If the solids are distributed more uniformly throughout the
system, the volume to be stored could be larger. If the pollutants from
street washout are to be controlled, a runoff volume of 23.0 mm (0.906
inches) should be stored and treated for the 5-year synthetic storm.
In terms of pollutant concentration (mg/1), a different pollutograph
character is observed. Figure 7 shows the concentration pollutographs of the
5-year synthetic storm for BOD and SS for both the conventional combined and
separate storm systems. The pollutograph contains only one peak which occurs
early in the storm when the runoff quantity is small. In the combined
system, the peak SS concentration of 1,972 mg/1 occurs 18 minutes after the
start of rainfall and 38 minutes later (or 56 minutes after the start of the
storm) the concentration drops below 20 mg/1. The peak SS concentration from
the separate storm system occurs 50 minutes after the storm starts. The peak
has a relatively low value of 17 mg/1 using STORM default values. The
BOD concentration pollutograph has a similar general character except that
the peak of pollutograph occurs earlier since there is no consideration of
the BOD fraction in the solids deposit in storm sewers in SWMM.
The pollutographs associated with the separate storm sewer system
were obtained using the combined sewer network and design diameters but
without the DWF inputs to the SWMM TRANSPORT Block. Although there are
differences in sewer network and design diameters between the two systems,
the resulting differences in outflow hydrographs and pollutographs are within
the range of accuracy required in this study for the comparison of the
alternative systems.
Figures 8 and 9 present the outflow hydrographs and pollutographs of
1-year and 1.3-month synthetic storms, respectively. They are similar to
those of the 5-year synthetic storm except in flow and pollutant magnitude.
Runoff characteristics from storms of advanced pattern were also inves-
tigated. Figure 10 shows the District A outflow hydrograph and pollutographs
for a 5-year return interval storm with the advanced pattern hyetograph shown
in Figure D-5. The peak runoff rate from District A occurs about 18 minutes
(as compared to 74 minutes for intermediate pattern storm) after the start of
the storm with a peak rate of 21.7 m /sec (766 cfs), about one-half of
the corresponding intermediate pattern storm. The peak combined sewage SS
concentration (138 mg/1) is one order of magnitude less than the peak concen-
tration (1,972 mg/1) for the intermediate pattern storm. This results since
peak runoff from an advanced storm occurs in the early stages of the storm
event when the results of sewer flushing is combined with street washout.
However, the peak SS concentration of stprmwater runoff alone is 92 mg/1,
about five times higher than the peak (17mg/l) for an intermediate pattern
storm. There is one peak flush of pollutant mass in advanced pattern
storm runoff. The peak flush ends about 45 minutes after the .storm starts.
48
-------
O
CM
(i/.6uj)
(l/Biu.oi *)
39VM3S Q3NiaiA)OD
00
CO
bi 00
CO
LU
o
I
0>
E
±: CO
OD CO
O
CD
u_
V-
CO
§
*
t! - /
5^ 1
5 D> 1
S £ £
s ^ j
o^ /
t- o 1
CO OQ *
V /
N^1'
//
\ /' i
* ' 1
A/I
\'i
\"
I1
V
v/
U/-
iX
K\
//*»
^y i.\.
io
00
?
CO
OJ
0 1
CO J.
2
cc
R
S ~
LL- .
O
H-
£E
0
CO
•
r~~
O a)
— M
an
g
j_i
o
4J
CO
S
M
0)
to
o
CJ
^L
CO
oo
CO
(l/Suioix) MOIJinO SS H31VMlAiy01S
(l/Buj 001 x ) SS 39VM3S Q3NiaiAIOO
(sp 001 x) 31Vd MOld
49
-------
o
u>
1
o
OJ
1
(iniu/sqi) SS 39VM3S Q3NI9IA100
o o o o o
In * ro CJ —
r~
(01 X 860
o
o
CM
!_.,
1
Mas psuiqoioo
0
u>
i
I i i
'1 X J81DMUIJCHS ) (I/BOI) SS
§ § s
1 / 1 / 1 1
o
o
00 -IS
cr
CO
o o
oc
y, ^
o
CO
UJ
o 1
* H
n)
•H
13
0)
4->
d
ed
0)
J2
ex
60
0
o
ex
§
S1
Ig
T) O
>>, 4J
JS CO
O 1-1
rH
o
o
IT)
o
o
<3-
O
o
ro
O
O
CM
O
O
50
-------
(mui/sq|) SS 39VAA3S Q3NiaiAlOO
X J8JDMUIJCHS ) ( |/6UU) 35
01
4-J
nJ
•H
t)
0)
0)
•U
e
•H
a
o
CO
to
•a
cfl
M
00
o
4-1
3
.H
I-I
o
(3
cd
ex.
ctf
M
60
o e
!-l t-t
-d o
>~. JJ
J3 CO
JJ 4J
3 cS
O D.
60
•H
Pn
1
o
1
O
CM
1
0
CO
i Vi 1
o
•tf-
31Va Moid
51
-------
o
OJ
ro
o
00
CVl
o
ID
o
<*
CVJ
o
o
CVJ
O
to
O
CM
O
co
(i/6ui) aog 'ss
o
o
co
o
o
r-
o
o
to
o
o
10
o
o
-------
Characteristics of runoff from 1-year and 1.3-month synthetic storms are
similar to those of the the 5-year synthetic storm except for magnitude of
runoff rate and pollutant amounts.
Although runoff characteristics between storms of intermediate and
advanced patterns are substantially different, differences appear insigni-
ficant in the total pounds of pollutants reaching the outlet of the sewer
system as shown in TABLE 5. The computed SS loadings in District A combined
system are 1,331 and.1,329 kg (2,939 and 2,933 pounds), respectively, for the
synthetic 5-year intermediate and, advanced pattern storms. For a separate
storm system, the computed values are 506 and 500 kg (1,116 and 1,104
pounds), respectively. For BOD, the values are 321 versus 320 kg (709
versus 707 pounds) for a combined sewer system, and 58 versus 57 kg (128
versus 126 pounds) for a storm sewer systems. However, the peak concentra-
tions for the advanced pattern are much lower than for the intermediate
pattern due to the dilution afforded to the first flush.
The following is noted from TABLE 5:
(1) For a given rainfall amount, both the runoff volume and peak runoff
rate from an intermediate pattern storm are greater than those from
an advanced pattern storm with the exception 6f the runoff volumes
for 1.3-month return frequency storms, which are equal, since most
of the runoff from this storm is from impervious areas.
(2) Frequent, smaller storms, regardless of pattern, generally produce
higher pollutant concentrations than infrequent, rare storms. This
is illustrated in Figures 11 and 12, which, respectively, compare
the SS concentration pollutographs versus storm return frequencies
for the combined and separate storm sewer systems in District A.
Conclusions from BOD concentration pollutographs are similar.
(3) Rains with a return interval of about 10 days (0.35 months) (inten-
sity 0.1 inches/hour) should cleanse combined sewers. The total SS
loading reaching the drainage district outlet from this storm is
computed as 827 kg (1,826 pounds), of which 325 kg (718 pounds) is
estimated to be from DWF. The remaining pollutant mass is attri-
butable to the sewer solid deposits, as the street washout from
this small storm would be insignificant. During the four dry days
assumed in the computation, the total amount of solids deposit is
estimated to be 509 kg (1,124 pounds).
(4) The total SS accumulation on the' streets of District A, assuming
four antecedent dry days, a 7-day street sweeping interval, and
75 percent sweeping efficiency, is computed as 525 kg (1,160
pounds). A 5—year synthetic storm would remove about 96 percent of
the total, a 1-year synthetic storm, 87 percent, and a 1.3-month
synthetic storm, 35 percent. Corresponding synthetic advanced
storms would remove slightly less pollutant.
Effective pollution control in an urban area associated with wet-weather
53
-------
CO
Q*
o
H
CO
w
CD
o
*^d
**"^
.
O
PR
CO
o
H
H
CO
*i
PH
P-l
s
iy|
3
tri
><
H
1
O1
CO
m
a
S
*
•H
i-l
cd
3
O
•
4J
•rt
U
d
ra
3
O-
O
M O
oi ca
Sm
B "o
^ o d
CO JJ 3
.0 en oi en
H en
&I
•r!
cd Q
0-0 0
d cu W
r-4. irl 60
CO ,0 CO
4-1 g g
O O 01
H cj en en
en
a
M O
a) pa
x-^ -U
i-l CO,
10 6 IH
^-* o d
•4-1 3
d en erf co
o en
•1-1
co
",
d
CU
0 O
§TJ O
Q) P3
o d o)
•H 00
"cO "§ §
01 O CU
p-i o en en
en
IH
3
O ^
v^ cu »2
iS « ^
CO
IH & r^
d t— i d
3 O *H
p^ {^ ^^
d
o
JJ
CO
4J X-N
•H CO
H O. CU
3 -H j:
O O CJ
J3 0) d
1 U vH
gf
4J CO
en CM
' ^
o
C!
t-t C7
o cu
en £4
00
CM
•-I
VO
1-1
cr»
O
o>
CO
CM
in
CM
Q
^
r-l
O
-a-
r-I
CM
r--
1-1
0
1-1
VO
r--
i-H
CO
-a-
CM
01
CO
•H
•O
Interme
13
01
1
f^
1-1
1-1
1-1
o
CO
CTS
VO
r--
CM
CO
CM
CO
CM
CM
O-v
i-l
O
•a-
CO
CM
VO
i-l
VO
CM
CO
O
,
r-l
»a-
! — |
CU
cd
•H
Interme
cd
cu
i
H
O
VO
VO
O
-a-
,__{
CO
VO
in
CM
CM
VO
CM
O
P-I
CM
Q
O
VO
VD
CM
in
r-l
CM
VO
CM
O
O
VO
O
01
4J
.2
•i)
Interme
r*
4-J
C
|
CO
•
r-H
1
f«^
m
vO
CM
00
I
1
o
-a-
r-l
0
00
CM
CM
m
o
o
in
o
01
4J
co
•H
•o
r*
Interme
y (-1
1 1
d
O
S
in
CO
•
o
VD
CM
i-H
^j-
O
r-l
i-l
p^
O
CO
CO
CTi
CM
^.
«^-
•
CM
0\
00
r>
00
CO
1-1
VO
VO
CM
m
i-H
CO
t-l
CO
fc
0
4J
d
•H
00
d
•H
01
a)
w
4-1
01
0)
4-1
CO
to
•a
d
01
> .
0)
CO d
0
0) • 4-1
>1 a CO
CO O H
T3 -H 0
4-1 Pi
>-. cd co
•034-1
4-1 g ID
d o
0) O >,
•a fi
01 4-1
o d TS
oi co oi oi
4-1 4-1 4J r-l
d 3 to ^
cO iH g cd
t-l -rl iH
H O 4J 'i-l
3 & to CO
0 01 >
IH n co
O P->
d t-< o u
o do
co o) d
TJ 01 3
01 3 0- CO
CO r-l CU 4J
CO CO H CO
pq > fn Q
*
•K -K 1
to
CO
in
-a-
c^
d
If
—
, — |
r-H
O
(I)
CO
ro"
PS
CO
CO
CM
O
II
iw
CJ
t 1
tf.
g
"
-i.
tn
CM
II
0)
4J
jg;
54
-------
4000
3000
2000
1000
800
600
500
400
300
200
^ 100
\ 80
-§ 60
co 50
10 40
30
20
10
8
6
5
4
•I-YEAR
INTERMEDIATE PATTERN STORM
ADVANCED PATTERN STORM
5-YEAR
I
1
I
30 60 90 120
TIME SINCE START OF STORM (minutes)
150
180
Figure 11. Combined sewage pollutographs for various storms
55
-------
CO
3
S-4
o
O
•rl
^
O
D.
ni
(50
O
o
a.
4-1
m
o
a
^
M
CD
O
JJ
w
-------
flows can be largely achieved by the containment and treatment of the first
flush from frequently occurring storms.
SEPARATE STORM SEWER SYSTEM
In 1972, a separate storm sewer system was designed for District
A (11). The existing combined sewers were to be maintained as a separate
sanitary sewer system accepting limited amounts of urban runoff. The storm
sewer layout is shown in Figure D-13. Its function was to relieve combined
sewage flooding by intercepting runoff. The new combined sewer layouts were
extensions of the separate storm system. Storm sewer slopes, diameters and
average excavation depths for the trunk sewer elements in the system are
listed in TABLE 6. Also listed is the corresponding information for the
combined sewer system. The effect of eliminating surcharges in combined
sewer systems is apparent. The storm sewer slopes were determined from
detailed topographic surveys and field inspection of underground utilities to
eliminate interference. The average excavation depth for each pipe element
is computed taking into account the upstream and downstream pipes.
The separate storm sewer system was designed with surcharged sewers to
reduce sizes and excavation cost. The following was used to determine the
flows and hydraulic grade in the system. Hydrographs for both impervious and
pervious areas were developed from the 5-year synthetic storm hyetograph.
These hydrographs were determined for a 0.405 hectare (1.0 acre) area and a
ground slope of one percent, by subtracting surface infiltration, depression
storage and surface detention from the rainfall, for each minute of the
storm. The flow in any particular section of the trunk sewer at any minute
was determined by time routing the incremental volumes from upstream hydro-
graphs. For each minute of the storm, the flow in each section of the trunk
was successively determined. For trial pipe sizes, the hydraulic gradient
was determined for each minute of 'the storm based on flowing full friction
and secondary losses calculated for each section of trunk sewer. The pipe
sizes selected were those which raised the peak hydraulic gradient to about
one foot below the drain inlet grating at the lowest point in the system.
This would still allow design flow from the critical low points to simul-
taneously enter the system. Different trunk sewer sizes necessitated
reevaluating the time of travel in each section for proper time routing.
TABLE 7 compares peak flows at several locations of District A generated
for the storm sewer design and for the design of combined sewer system using
SWMM. The peak flow rate used in the separate storm sewer design is gener-
ally ten percent less than the flow used in the design of combined sewers to
reduce the frequency of combined sewer surcharging. This results since the
critical factor in a surcharged system is the hydraulic gradient. The peak
hydraulic gradient in any sewer reach, except the most downstream reach, is
influenced by backwater effects and occurs after the peak flow has passed
downstream.
57
-------
TABLE 6. COMPARISON OF PIPE DIAMETER, SLOPE, AND EXCAVATION DEPTH,
SEPARATE STORM VS. COMBINED SYSTEMS
Trunk
Sewer
Element
301
303
305
307
309
311
313
315
317
319
321
323
325
327
329
331
333
335
337
339
341
343
345
347
349
351
353
357
359
361
363
Separate Storm
Diameter
(feet)
3.0
3.5
4.0
4.5
5.0
6.0
6.0
6.5
7.0
7.0
7.0
7.0
7.5
7.5
7.5
7.5
7.5
7.5
7.5
8.0
8.0
8.0
8.0
10.0
10.0
10.0
11.0
11.5
11.5
11.5
11.5
Slope
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0013
.0010
.0013
.0017
.0020
.0090
.0090
.0005
.0100
.0094
.0094
.0034
.0034
.0088
.0088
.0004
.0004
.0004
.0013
.0003
.0003
.0003
.0003
System
Excavation
Depth (feet)
9.0
9.6
9.8
10.4
10.9
12.2
14.3
13.5
14.4
13.7
13.4
13.5
14.2
13.6
14.0
14.8
14.1
15.3
16.3
17.0
16.4
17.6
18.7
19.9
22.6
25.9
28.0
28.5
29.5
32.3
31.0
Combined System
Diameter
(feet)
3.0
3.5
4.0 .
4.5
5.0
6.0
6.0
6.5
7.0
6.0
8.0
8.0
9.0
6.5
8.0
9.0
6.5
7.5
7.5
9.5
10.0
10.0
10.0
13.5
13.0
14.0
13.0
11.0
11.0
11.5
11.5
Slope
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0090
.0013
.0017
.0025
.0099
.0048
.0019
.0133
.0089
.0091
.0034
.0019
.0019
.0019
.0004
.0004
.0004
.0013
.0030
.0030
.0030
.0030
Excavation
Depth (feet)
9.2
9.8
10.0
10.6
11.0
12.7
14.3
14.4
14.8
13.3
15.4
15.4
16.2
13.1
14.5
16.0
13.8
16.5
17.4
23.1
22.8
22.4
20.8
22.4
24.7
29.0
30.2
28.5
29.5
32.8
31.5
Note: 1 foot = 0.3048 meters
58
-------
TABLE 7. PEAK FLOW COMPARISON FOR
SEPARATE STORM AND COMBINED SEWER DESIGN
(5 YEAR RETURN FREQUENCY)
Trunk Sewer
Inlet No.
338
342
356
364
Drainage Area
(Acres)
349.6
360.0
600.4
655.1
Peak Flow
Separate Storm
System
766
795
1259
1276
(cfs)
Combined
System
898
909
1321
1407
Note: 1 acre = 0.405 hectare
1 cfs = 0.0283 m /sec
There are noted differences of sewer sizes for the conveyance of storm-
water runoff and combined sewage as shown in TABLE 6. The differences result
from allowing surcharged flow in the separate sewer system and not in the
combined system under design flows. All other conditions being similar,
design diameters for a combined sewer system are generally equal to or larger
than those for a separate storm system.
There is little difference in runoff hydrographs for the separate storm
and combined system since dry—weather flow quantity is a small fraction of
total runoff during wet weather. For instance, the peak 1.3-month synthetic
storm runoff rates are 6.08 and 5.95 m /sec (215 and 210 cfs), respec-
tively, for the combined and separate storm system.
SEPARATE SANITARY SEWER SYSTEM
The layout of the separate sanitary sewer system (Figures 13 and 14)
generally follows the existing sanitary or combined sewers, except that the
existing three outlets have been reduced to two. Each city street is served
by one sanitary sewer. There are 260 sewer pipe elements with a total length
of about 32,308 m (106,000 feet).
The quantity of domestic sewage was determined using the estimated
saturation population and the estimated daily wastewater flow per capita.
For District A, the total saturation population was estimated at 16,500 and
the average daily domestic flow was calculated as 4.34 m per min. (0.65
mgd), assuming a rate of 379 liters (100 gallons) per capita per day in-
cluding infiltration. The ratio of peak flow to average daily flow was
estimated using the formula suggested by Harmon (12)
M
Q
1 +
14
(1)
where:
M = the instantaneous peak flow,
Q = the average daily domestic flow, and
P = the tributary population in thousands.
59
-------
•a
0)
1-1
dJ
•§
CO
a)
§
to
4J
•r*
td
to
bO
60
-------
CO
CO
LU
CO Ld
LU O
o: 2
UJ
-I X
_J (-
< O
ttf
>
1-1
ft
1-1
M
ffl
cd
-------
The formula indicates that relatively higher maximum rates of flow are
expected from small numbers of people than from large numbers. This formula
was found applicable to Elizabeth in previous studies(S). For District A,
the ratio of peak daily flow to average daily flow is computed as 2.74 and
peak daily flow, as 11.82 m per min. (4.50 mgd). For commercial and
industrial areas, peak sewage flow was assumed as 187,000 liters per hectare
(20,000 gallons per acre) and equaled 2.76 m per min. (1. 05 mgd). The
peak dry-weather flow at the outlets was about 14.58 m per min. (5.5
mgd). This peak was rounded up to 15.8 m per min. (6.0 mgd) for compu-
tations in this report.
The minimum size used for sanitary sewers was 0.20 m (8 inches).
The minimum pipe slopes selected assured a minimum flowing full velocity of
0.61 m (2.0 feet) per second. Sewer diameters were calculated based on
Manning's equation with a roughness coefficient of 0.013.
The sanitary sewer diameters are shown in Figures 13 and 14. Most of
the pipes are 0.20 m (8 inches) in diameter with the largest being 0.381 m
(15 inches). .
COMBINED SYSTEM WITH IN-SYSTEM STORAGE
As discussed previously, in-system storage, as considered in this study,
includes (a) storage provided in enlarged upstream lateral sewers due to
the economy achieved, and (b) satellite storage provided in the collection
system along the trunk sewer. Both storage schemes have the effect of
reducing the size of trunk sewers. In some cases, trunk sewers designed to
convey peak flows are the practical maximum that can be physically con-
structed due to space limitations. Pipe storage provided in upstream lat-
erals and satellite storage would permit a reduction in trunk sewer sizes.
Such storage serves to reduce the peak of outflow hydrograph and modify the
outflow pollutograph from a drainage basin thus reducing the capacity of
interceptors required to abate overflow pollution.
I
Design with In-Pipe Storage
In-pipe storage can be provided in both lateral and trunk sewers.
Preliminary cost analysis indicates storage in the laterals, rather than
in the trunks, to be less costly for either pollution or flood control.
Figure 15 shows the incremental cost of pipe storage resulting from
increasing the diameters of the pipes required to convey storm runoff. The
pipe cost is obtained from Figure D-15 or TABLE D-17 for an excavation depth
allowing 1.52 meters (5 feet) of cover over sewers. The following can be
observed:
1. A slight enlargement of sewers to provide storage volume is less
costly than a large sewer enlargement,
2. Pipe storage in the smaller lateral sewers would be less costly
than pipe storage in large trunk sewers, and
62
-------
3. Satellite storage which also'reduces trunk sewer sizes may be the
least costly alternative since in-pipe storage costs average
about $0.26 per liter ($1.00 per gallon).
10 lOi-
LU
cc
O
en
LJ
a.
a_
u.
O
C/5
O
O
UJ
2 4
UJ
(T
O
E 3
NOTES:
NUMBER ARE BASIC PIPE SIZES
REQUIRED FOR CONVEYANCE OF
DESIGN STORM RUNOFF.
GRAPHS INDICATE-COST OF STORAGE
OBTAINED BY INCREASING PIPE
DIAMETER TO SIZE INDICATED
BY THE ABSCISSA.
I
I
67 8 9 10 II
INCREASED PIPE DIAMETER (feet)
12-
13
14
Figure 15. Incremental cost of pipe storage
In-pipe storage reduces peak runoff rates while permitting flow convey-
ance. The pipe volume, in excess of that required to pass peak flow, can
provide regulation by installing either a downstream orifice coupled with an
overflow weir (Figure 16) or a collapsible dam coupled with a regulator.
Other devices such as the "Hydro-Brake" (13) and the VBB Flow Regulator
(14) have been proposed. The backwater created by the regulators results
in storing flows in excess of the desired downstream rate reaching £he
regulator. This reduces the design flow rate in the downstream sewers. It
may also be designed to moderate the initial pollutant concentration by
diluting the concentrated pollutant in the first flush with subsequent, less
concentrated runoff.
The in-pipe storage analysis was accomplished in three steps.
63
-------
LENGTH AND DIAMETER AS REQUIRED
TO PROVIDE DESIRED. FLOW CONTROL
-MANHOLE
OVERFLOW WEIR
ELEVATION AND LENGTH
DETERMINED TO PREVENT
SURCHARGE OF
UPSTREAM SEWER
CONTROL
STRUCTURE
PLAN
OVERFLOW
WEIR
FLOW CONTROL SEWER
-c
SECTION A-A
Figure ,16. Weir-orifice regulator
64
-------
Step I - The overland flow hydrograph to lateral sewers from each
subcatctment was generated for the 5-year synthetic storm using a
computer program adapted from the SV7MM RUNOFF Block. Summing
arithmetically the corresponding time steps of the tributary
subcatchment hydrographs, the overland flow hydrograph to a pipe
element was derived. This hydrograph was integrated to obtain the
total direct runoff. The volume that should be stored to obtain
the desired outflow peak rate was then determined by integration of
the flow in the overland flow hydrograph in excess of that rate.
The peak constant flow rate shown in the modified hydrographs of
Figure 17 assumes a variable orifice outlet control.
; Step II - The desired peak flows of the upstream tributary sewers
were sumried to obtain the peak flow to be passed for the pipe under
consideration. Manning's equation was used to compute the pipe
diameter and slope required to convey the peak design flow. The
pipe diameter required for storage was separately computed and then
was combined with the pipe diameter for flow conveyance to obtain
an aggregate design pipe size having the same cross-sectional area
as the two separate pipes. The next available 'commercial pipe size
was used as the design diameter.
Step III - The design of trunk sewers was made using the SUMM
TRANSPORT Block with the modified hydrographs entering the trunk
sewer from the lateral sewer branches. No surcharge was allowed in
either the lateral or trunk sewers.
Figure 17 illustrates the controlled hydrographs to be routed through
the pipe element 201 for storage of 0,10, 20 and 50 percent of total runoff
volume drained to that pipe element from its tributary subcatchment. Zero
percent storage represents the case without in-pipe storage. Storage reduces
the magnitude of the peak flow but extends its duration.
The design diameters of lateral and trunk sewers for ,in-pipe storage
volumes equal to 8.3, 16.5, 25.0 and 41.3 percent of the 5-year synthetic
storm runoff volume in District A are.listed in TABLE 1 and TABLE 2, respec-
tively. These percentages correspond to a total of 10,200; 20,100; 30,300;
and 50,300 m (2-7, 5.3, 8.0 and 13.3 million gallons), respectively,
stored in the lateral sewer pipes. The respective percentages of tributary
subcatchment runoff volume stored in each lateral sewer element are, however,
10, 20, 30 and 50 since about 14 percent of District A drains directly to
t runk s ewe rs.
As the volume stored in lateral sewers increases, the lateral sewer
sizes are consistently increased except for some at the lower end of lateral
sewer branches. Trunk sewer, sizes are consistently reduced because of the
reduction in peak flox/s discharged by the branches. The reduction in size is
reasonably uniform over the entire length of trunk sewer. This type of
design provides storage for pollution control at least cost, as previously
described. •
65
-------
PIPE ELEMENT 201
TRIBUTARY AREA = 8.9 ACRES
% IMPERVIOUSNESS = 80
NUMBERS ON CURVES ARE
PERCENTAGES OF TOTAL
RUNOFF STORED IN PIPE
FROM AREA DIRECTLY TRIBUTARY
TO THE PIPE
5-YEAR INTERMEDIATE
PATTERN STORM
50
30
60 90 120 150 180
TIME SINCE START OF STORM (minutes)
210
240
Figure 17. Hydrographs modified by in-pipe storage
66
-------
Outflow Hydrographs and Pollutographs
Figure 18 presents outflow hydrographs for various amounts of runoff
temporarily stored^in upstream laterals. The peak runoff rates are 293, 228,
162, 126 and 93 m /sec (1,407, 1,096, 788, 607, and 448 cfs) for amounts of
storage eq,ual to 0, 8.3, 16.5, 25.0, and 41.3 percent -respectively of about
119,000 m (31.5 million gallons). The computer program developed in this
project for in-pipe storage does not currently permit routing pollutants.
Therefore", pollutographs were not generated.
The pollution abatement benefits of in-pipe s.torage depend on the
flow rate selected at which storage of combined waste begins. Where a
pollutant sink does not exist, the higher the flow-through rate, the smaller
would be the benefit achieved and the smaller the amount of storage required.
Less pollution abatement benefit results since less dilution of the high
pollutant concentration in the first flush would occur.
Increasing the amount of in-pipe storage reduces the flow velocity
and may result in solids settlement in the sewers. About 60 percent of
the pipes, when in-pipe storage equals 8-3 percent of the total 5-year
synthetic storm runoff volume, would have a flowing full velocity greater
than 0-30 m (1.0 feet) per second. As the amount of storage increases to 25
percent, less than 20 percent have flowing full velocity exceeding 0.30 m
(1.0 feet) per second with about 40 percent have a velocity less than 0.06 m
(0.2 feet) per second. Solids deposited in sewer pipes during the pipe
full, low velocity period may be resuspended as flow recedes and velocity
increases. However, where in-pipe storage of about 40% of the overland flow
volume is provided, additional maintenance for cleaning may be required. The
Boston sewer flushing study (15) indicates that sewage impounded in upstream
sewer segments by manhole stoppers are self-cleansing upon stopper release
due to sudden increase in velocity with pipe drawdown.
In a wastewater facilities plan study for Trenton, New Jersey "(16),
several wet-weather flows were sampled to find the values of BOD and SS
in combined sewage. Automatic samplers were used and started at the begin-
ning of the rainfall with samples taken at 15-minute intervals. The results
of two sampling periods during small storms are shown in Figure 19. An
increase in BOD occurred in a short time as the storm runoff increased
flow to the treatment plant. The BOD then receded as the sewage flow exceed-
ed the treatment plant's pumping, capacity and sedimentation occurred in the
interceptor due to reduced velocities resulting from storage in the sewer.
After the end of the rainfall, wastes stored in the interceptor were drawn
upon, the sewer velocity increased, settled material was resuspended and peak
concentrations of BOD reached the plant.
The deposition and resuspension of solids would be reflected in the
outflow pollutographs.
SEWERS WITH SATELLITE STORAGE
Satellite storage along the trunk was investigated using the ST7MM
TRANSPORT Block. The simulation included no more than two such basins. The
67
-------
0
•H
LO
CO
g?
i-l
O
4-1
W
a)
ft
•H
ft
•H
IS
CO 1=i
rC H
ft O
Cd 4-1
M CD
M
O fl
M M
•X) CO
13 ft
rH CD
m 4-j
4-) n)
oo
rH
CO
•H
1
0
O
to
1
O
O
"fr
1
O
O
-------
(I/Bui) 009 30VM3S Q3NI8WOO
! S S 18188*8*
Oft
-------
effect of storage quantity and location on the trunk sewer sizes and outflow
hydrographs and pollutographs was studied. All flows were assumed diverted
to storage basins with outflow occurring only after storage was full. In
addition, at any point in time, the flow contained in a storage basin was
assumed to be completely mixed. After storage basin locations were deter-
mined for various amounts of storage capacity, runoff characteristics for
various combinations of storm intensity and patterns were developed. These
runoff characteristics were used in determining comparative costs.
Design
Design of trunk sewers with satellite storage was made with SWMM.
Cylindrical storage basins were assumed with outlets controlled by overflow
weirs. The width of the overflow weir was determined to minimize the hydrau-
lic head above the crest of the weir and reduce the effect of surcharge
volume on the routing of storm runoff. Since the flow divider option in SWMM
was not used, dry-weather flows above a storage basin are intercepted by the
basin. Consequently, the amount of solids deposit in the sewer reflected the
location of storage basins. In practice, dry-weather flows would be diverted
from the basin.
The locations and volumes of storage basins in a drainage watershed were
selected to conform to the following criteria:
1. The peak of outflow hydrograph should be attenuated as much as
practical.
2. For outflow hydrographs with multiple peaks, the peaks should be
relatively equal to reduce intercepting sewer sizes.
3. The rising limb of the outflow hydrograph should be delayed as much
as possible to provide reduction of the high concentrations of
pollutants in the initial flush by dilution with later runoff.
Figure 20 shows the outflow hydrographs for storage basins_at six
different locations in District A. In. each case, about 50,300 m (13.3
million gallons) of the total 119,000 m (31.5 million gallons) of assumed
storm runoff is captured. TABLE 8 compares the corresponding trunk sewer
sizes. The curve labeled "no storage" applies to a conventional combined
system and is shown for reference only. Figure 21 presents schematically the
various alternatives for storage location.
Case A assumes a single basin located at the downstream end of the
system. Overflow occurs 82 minutes after the start of rainfall assuming the
5-year synthetic storm hyetograph. The peak overflow rate is 25 m /
sec (890 cfs). The trunk sewer diameters are the same as in the conventional
combined system, and the cost is about 75 percent more than Case B, the least
expensive.
In Case B, the two storage basins of equal capacity were located at
points where 26 and 54 percent of the watershed drained to them. The peak
runoff is about the same as Case A but the peak occurs about 20 minutes
70
-------
o
CM
O OT
-co ..
or
^
CO
O LiJ
ID O
Z
CO
LiJ
O
<*•
s
CO IO
o
•H
0
o
'to"
CO
"5
c
'e
^-^
«g
cc
o
1-
4-J
cd
OJ
60
cd
M
O
4-J
CO
0)
4-1
•H
Is
o
4-J
to
a
^
(I)
4-1
4-1
cd
PH
H a)
OJ 4-1
4-) cd
cd -H
^ 1
| ) ,
0 I
c
£ o
O -H
iH 4-1
MH Cd
4-1 O
3 O
O
OJ
60
'•H
fe
o
o
to
o
o
o
o
-------
TABLE 8. COMPARISON OF TRUNK SEWER SIZES (FEET) FOR STORAGE
BASIN AT VARIOUS LOCATIONS
Sewer
Element No
301
303
305
307
309
311
313
315
317
319
321
323
325
327
329
331
333
335
337
339
341
343
345
347
349
351
353
357
359
361
363
A
3.0
3.50
4.00
4.50
5. 00
6.00
6.00
6.50
7.00
6.00
8.00
8.00
9.00
6.50
8.00
9.00
6.50
7.50
7.50
9.50
10.00
10.00
10.00
13.50
13.00
14.00
12.50
11.00
11.00
11.50
11.50
B
3.0
4.00
4.00
4.50
5.00
6.00
6.50
7.00
4.00
3.50
4.50
4.50
6.00
5.00
5.50
7.00
5.00
6.00
6.00
7.50
5.00
5.00
5.00
6.50
7.00
7.00
9.00
8.50
8.50
9.00
9.50
C
2.75
3.50
4.00
4.50
5.00
5.50
6.00
6.50
7.50
6.00
8.00
8.00
5.00
4. 00 •
.4.50
5.00
4.00
4.50
4.50
6.50
7.00
7.00
7.00
9.50
9.00
10.50
9.50
8.50
8.00
8.50
8.50
D
2.75
3.50
4.00
4.50
5.00
5.50
6.00
6.^50
7.50
6.00
8.00
8.00
9.50
7.00
7.50
8.50
6.00
7.00
7.00
8.50
9.00
9.00
9.00
12.00
12.00
12.00
9.50
8.50
8.50
8.00
8.50
E
2.75
3.50
4.00
4.50
5.00
5.50
6.00
6.50
7.50
6.00
8.00
8.00
8.00
6.50
7.50
8.50
6.50
7.00
7.00
9.00
10.00
10.00
10.00
13.00
13.00
13.50
9.50
8.50
8.00
8.00
8.50
F
, 2.75
3.50
4.00
4.50
5.00
5.50
6.00
6.50
7.50
6.00
5.00
4.50
5.00
4.00
5.00
6.00
4.50
4.50
5.50
7.00
8.00
8.00
8.00
10.00
11.00
10.50
9.50
8.00
8.00
8.00
8.00
Peak Runoff
Rate (cfs)
890
897
665
644
630
607
TOTAL STORAGE VOLUME = 13.3 MILLION GALLONS OR 41.3% OF
TOTAL 5-YEAR SYNTHETIC STORM RUNOFF VOLUME
NOTE: Case A - One basin at downstream end of watershed
B - Two basins of equal capacity at 26 and 54% of watershed area
C - Two basins at 35 and 82% of watershed area, capacity ratio 3:2
D - Two basins at 35 and 82% of watershed area, capacity ratio 1:3
E - One basin at 82% of watershed area
F - Two basins of equal capacity at 28 and 82% of watershed area.
3 3
1 foot * 0.3048 meters; 1 cfs = 0.0283 m /sec; 1 million gallons = 3,785 m
72
-------
u.
LU
CO
o
IO
/s\
co
-------
earlier. Hence, more polluted runoff would have to be intercepted and more
off—line storage capacity would be required. On the other hand, trunk sewer
sizes are substantially reduced.
Case C has a lower peak in the outflow hydrograph than Case B with
about the same cost of trunk sewers. Two basins, having a capacity ratio of
3:2, were located where 35 and 82 percent of the watershed drains to them.
Further improvements of the outflow hydrograph is achieved by using the same
basin locations but with a capacity ratio of 1:3 (Case D). However, trunk
sewer sizes are generally greater than those in Case C. Case E, with one
large basin located where 82 percent of the area drains to it, has a compar-
able outflow hydrograph and trunk sewer sizes (except for the reaches immedi-
ately upstream of the basin) as those of Case D.
Case F assumes two basins of equal capacity located where 28 and 82
percent of the watershed drains to them (trunk sewer element numbers 320 and
352). The outflow hydrograph improved somewhat over that of Case C with
trunk sewer sizes generally comparable. Based on the criteria stated pre-
viously for the selection of basin location, Case F offers most advantage and
was used to represent the desirable locations for storage equalling 41.3
percent of the 5-year synthetic storm runoff. In practice, the locations for
storage would be dictated by site conditions.
For satellite storage volumes equal to 9.4 and 17 percent of 5-year
storm runoff, two basins of equal capacity located where 26 and 54 percent of
watershed drains to them, or at trunk sewer element numbers 316 and 340
respectively, appear to result in reasonably good outflow hydrographs
(Figure 22). The two peaks in the hydrographs are of about the same magni-
tude and the reduction in peak flow rate is appreciable. An additional
computer run was made for 17 percent storage with basins located where 28 and
82 percent of the watershed drained to them. The peak of the outflow hydro-
graph was about 15 percent greater and the trunk sewer sizes slightly
greater.
The trunk sewer diameters for three different amounts of satellite
storage are listed in TABLE 2 together with the sewer sizes for a conven-
tional combined system and advanced systems with in-pipe storage. The
effect of storage on sewer sizes is.more pronounced for capacities of about
17 or more percent of the total runoff. A major reduction in sewer sizes
results as the storage capacity increases to 41.3 percent of the 5-year
synthetic storm runoff volume. For a storage capacity equal to 9.4 percent
of the total 5-year synthetic storm runoff volume, the trunk sewer dia-
meters are generally less than that for conventional combined system but not
by a significant amount. Similar findings have been reported by Yen and
Sevuk(17).
The sewer size of element number 317 for 9.4 percent satellite storage
is greater than for a conventional combined system. This is caused by the
release of waste over a gravity weir after the storage basin is full. The
rate of release is affected by the plan area of the basin and the characte-
ristics of the overflow weir. In this particular case, the effect of basin
74
-------
O
o
LU
O
CO
z
CO
m
O
H
— CM ro
- O
en
O <*>
~ "5
c
tr
o
t/>
tr
^
CO
o
CM
1
o
o
CO
1
o
o
t
1
o
o
CM
1
0
o
o
1
0
o
CO
1 -
o
o
CO
1
0
o
*
1
o
0
CM
c
i-J
cfl
CD
CU
60
tfl
!-l
O
4J
CO
i-H
rH
CU
^3 O
•U JJ
•H W
o
?
UJ
0
z
CO
UJ
2>
p
CO
cd
oo
O
(-4
T3
^
•^
o
,H
U
J-l
CU
4-)
ctf
0,
-------
volume on discharge rate is of particular significance since the storage
basin becomes full at about the same time as the peak inflow, and hence, is
not available for storage. The same explanation applies in the case of 17
percent storage to pipe elements numbers 321 and 323 whose diameters are
greater than those of 9.4 percent storage.
Outflow Hydrographs and Pollutographs
The outflow hydrographs from District A, for satellite storage amounts
equal to 0, 9.4, 17, and 41.3 percent of 5-year synthetic storm runoff
(about 119,000 m or 31.5 million gallons), are in Figure 22. 3The peaks of
hydrographs are respectively 39.8, 34.2, 24.9 and 17.2 m /sec (1,407
,1,208, 879 and 607 cfs). There are two peaks in the outflow hydrographs.
The first peak results from runoff below the downstream storage basin. The
first 60 minutes of the hydrograph rising limbs for 9.4 and 17 percent
storage are identical, since the storage basins are identically located in
each case. The shape of hydrograph following the first peak is influenced by
the modulating effect of the unfilled storage volume remaining, if any, and
the reduction in local runoff from areas downstream of the second basin.
Overflow from the storage basin can result in a' second peak. With greater
storage (41.3 percent), the first peak is less than with less storage (17
percent) due to the further downstream location of storage. The first peak
for 9.4 percent storage is greater than for 17 percent storage, due to the
small storage volume and earlier spillage from basins. The time lag between
the two peaks increases as the storage capacity increases.
Figure 23 presents the 5-year, intermediate-pattern, synthetic storm
outflow SS concentration and mass rate pollutographs from District A for
various amounts of satellite storage. The effect of satellite storage on
pollutant loadings discharged is indicated. The peak concentration of SS is
lower with satellite storage. Although the peak SS concentration for 41.3
percent storage is higher than for 9.4 and 17 percent storage, this is not
significant with respect to pollution control since the flow rates are low
when the high concentrations occur. For example, at the time of the peak SS
concentrations for 0, 9.4, 17, and 41.3 percent storage, the flow rates are
respectively 0.13, 0.07, 0.07 and 0.02 m /sec (4.6, 2.6, 2.6 and 0.6 cfs).
At the time of peak outflow, the SS concentrations are respectively 7.9, 8.6,
5.8 and 13.4 mg/1.
Based on the current SWMM program, the low discharge pollutant concen-
trations of intense storms imply that intercepting of urban runoff for
treatment during such times may not be required where the concentration,
rather than mass, of pollutants discharged is governing and there is not a
need for extremely high quality water.
The higher SS concentration with 41.3 percent storage is coincident with
peak storage basin spillage. The same peak concentrations are found with 9.4
percent and 17 percent storage but at an earlier time since spillage takes
place earlier. The mass rate pollutograph for 41.3 percent storage during
the first 70 minutes of storm runoff is substantially less than for 9.4 and
17 percent storage. No storage results in highest pollutant levels. Hence,
76
-------
1,000 -
-NO STORAGE
mg/I
Ibs/min
5-YEAR INTERMEDIATE
PATTERN STORM
41.3%
20-X 40 60 80
TIME SINCE START OF STORM (minutes)
100
120
Figure 23. Outflow pollutographs with satellite storage
(5—year intermediate pattern storm)
77
-------
satellite storage provides benefit in controlling both storm runoff quantity
and quality.
Figures 24 and 25 present the 1-year and 1.3-month intermediate pattern
storm outflow hydrographs and SS concentration pollutographs, respectively,
for satellite storage amounts equal to 0, 9.4, and 41.3 percent of 5-year
synthetic storm runoff. Their characteristics are similar to those of the
5—year synthetic storm. Differences are in the magnitude of flow rate and
pollutant concentration. If satellite storage is located downstream of the
system, storage capacity equal to 17 percent of 5-year synthetic storm runoff
volume may contain all combined sewage from a 1.3-month return frequency
rainfall.
The effect of satellite storage on runoff characteristics from storms
of advanced pattern was also investigated. Figure 26 presents the
District A outflow hydrographs and pollutographs for a 1-year return
interval storm. The shapes of the hydrographs and pollutographs from ad-
vanced pattern synthetic storms of 5-year and 1.3-month return frequencies
are similar. Runoff from the 1.3-month synthetic storm would be fully
contained by a satellite storage basin of capacity equal to 17 percent of the
5-year synthetic storm runoff located at the downstream end of the system.
TABLE 9 provides a summary of peak discharge, peak SS concentration
and the total SS mass loading from a combined system serving District A
with various amounts of satellite storage for 5-year, 1-year and 1.3-month
return frequency storms of intermediate and advanced patterns. The amount of
solids deposited in sewers during dry days, as computed with SWMM is depen-
dent upon the storage basin locations selected. For the cases being com-
pared, the total pounds of solids deposited in sewers during the assumed four
antecedent dry days are 509, 172, 172 and 395 kg (1,124, 380, 380 and 872
pounds) respectively for 0, 9.4, 17.0 and 41.3 percent of satellite storage.
The following may be noted from TABLE 9:
1. Peak runoff rates are effectively reduced by satellite storage.
2. Satellite storage is an effective pollution control device.
3. The difference between the total pounds of SS discharged from 5
-year and 1.3-month intermediate pattern storms is 328 kg (724
pounds) for a conventional combined system. For a system storing 9.4
percent, 17 percent and 41.3 percent of 5-year synthetic storm
runoff, the differences are 429, 396 and 553 kg (948, 875 and 1,221
pounds) respectively.
4. The peak flow, peak pollutant concentration, and total mass loading
discharge is less for an advanced pattern storm than for an interme-
diate pattern storm of the same frequency.
Satellite storage basins along the trunk line of a combined system
result in the reduction of trunk sewer sizes to control both flooding and
78
-------
(I/6LU 001) NOIlVyiN30NOO SS
C\J
OJ
o
CM
CO
a.
i <
o-
-------
(l/Biuooi) NOIlVaiN30NOD SS
< * cj o
CU
4-1
,-H
1-7!
H
0)
4-1 ^
o) S
CO 5-4
•H
is a.
CO CU
ft 4-1
of cd
S-l ft
60
o cu
4-1 4-1
r-H -H
O 0)
ft S
5-1
!§
•H
CO
II
M O
60 g
O I
5-i ^0
>, H
0)
60
l[ 1 C_j
4J O
O CO
CU
s
60
80
-------
MOI1VH1N30NOO SS
81
-------
H
o
05
O
H
CO
w
H
W
^
CO
M
H
g
-g
W
EH
CO
CO
Q
pq
•S
M
^
o
CJ
pt
o
x
M
*3
»-•»
o*
Q
*5
^
EH
M
S
3
D
&
fV|
PTJ
O
a
0^
PCI
o
^
pv*
cd •
cu o\
PH
B*s
O
d
M
8 cu
J-l 4-1
0 4-1
4-1 Cd
co PM
.j.
CJ
d
cu
e 3
n a
o cu
t \ ti
CO pi.
^,
CO CO
—) CM
, — 1
s»x
_^
rH rH
rH in
m co
ci
^^
rH in
rH CO
d.
CM O\
p^ CO
Cy\ 0%
rH CM
s_x
rs.
O
VO
in
r*^
00
, |
o
CM
rH
r-*.
o
<}•
rH
cu
4-1
cd
•H
13
CU
3
cu
4-1
d
M
M
cd
cu
1
m
^_^
rH rH
rH rH
CO VO
N^X
_^
in m
CM CO
co m
i
t—i
^-^
m cr\
P^ CO
OO P~
rH
•si- f~T
VO CM
CO CO
CM CM
s_x
rs.
CO
rH
rH
VO
CO
CO
m
CO
vO
rH
VO
a)
cd
•rH
13
cu
e
M
d)
4-J
d
M
f_]
cd
cu
^-1
I
rH
CM
m o
in CM
CO ^
^^
r~ vo
O P~.
O CT»
j [ ^_y
,— N
CT\ P*«
m oo
O CJ*i
i s-_ ^
o in
vO —1
VD CM
CM CM
>^ *
CTi
s^f
CO
CM
rH
["-^.
CM
rH
m
t-H
CM
cu
4J
cd
•H
13
cu
0
cu
4-1
Cj
^J-j
4-1
d
o
1
CO
•
t-H
1 1
1 1
^^^
CN]
CM ^±
ON co
r-H
^^
oo co
CO ON
tH ON
CM
s^x
1
1
in
00
CM
*
VO
vO
F*.
13
CU
O
d
cd
13
Sf CTt
•* o
CM
•*^*
r1^
CN
•
i-l
o
m
o .
CQ
d
3
0
PM
.
13
cu
too
->-. cd
60 43
0 0
CO
d -rH
•H 13
d co
O CO
•H
4-i m
cd o
4J CQ
R 13
cu d
O 3
d o
O f?\i
a
rH
CO Cd
CO 4-1
o
4*i 4J
cd
cu cu
PU M
cd
cu
i-l CO
cd -H
CO
co a)
•H 43
CQ 4J
CU d
43 CJ
4J H
d cd
CU ft
M
cd d
ft -H
CU 0)
43 CO
4J 0
d EH
•H
4-1 .
O T3
d cu
SO
CO ^
!-i cd
CJ 43
43 0
0 CO
JS5 ^
•%
CM
CM
CU
M
3
60
•rH
pj
•H
13
CD
4J
cd
O
•H
13
•H
CJ
i-l
cd
CO
d
o
•rH
4-i
cd.
O
o
rH
CJ
toO
cd
M
O
4-1
CO
• •
O
m
•
rH
CO
m
o
4J
d
CU
o
i-l
CU
ft
d
•H
CU
60
cd
o
4-1
CO
•X
*
*
60
CO
in
•si"
•
o
II
43
t-H
rH
• A
O
CU
CQ
•s^
CO
0
CO
OO
CM
O
•
0
II
CO
UH
O
rH
• •
cu
4-1
O
82
-------
pollution. The cost-effectiveness of providing satellite storage also
requires consideration of downstream interceptors, storage and treatment
facilities as discussed hereinafter.
POLLUTION CONTROL POTENTIAL OF CONVENTIONAL COMBINED SYSTEMS
Pollution from combined sewage overflows has been abated by installing
flow control systems in large trunks or interceptors of existing, well
designed, conventional systems to contain peak flows for later discharge
to treatment facilities. The Cities of Seattle, Washington, Minneapolis-
St. Paul, Minnesota, and Cleveland, Ohio have installed dynamic flow control
systems and demonstrated the feasibility of remote, computer directed
control. In Seattle, the cost of each flow control device ranged from
about $60,000 to $450,000 with $150,000 being a representative value (18).
The above approach originated the concept of the "advanced system"
which provides either in-pipe or satellite storage in the design process.
TABLE 10 summarizes the total volume in sewer pipes and storage basins
for both conventional and advanced sewer systems. That part of the total
pipe volume available for storage would be dependent upon the number and/or
location of flow control devices, the pipe slopes and sewer network configu-
ration. The trunk sewer of the conventional combined system designed for
District A has a total storage capacity%f 9.4 mm (0.37 inches). This
storage, if fully effective, would contain the total runoff of 6.6 mm
(0.26 inches) which is equivalent to a 1.3-month synthetic storm runoff
volume (See TABLE 5). The 1.3-month synthetic storm represents a high return
frequency storm. By capturing and treating its runoff significant pollution
control can be achieved. The conventional combined system designed using
SWMM for a 5-year or more return frequency contains as much storage (14.2 mm
or 0.56 inches for District A) as the minimum size advanced combined system
investigated and could be equally effective for pollution control.
COST COMPARISON OF ALTERNATIVE COLLECTION SYSTEMS
The collection system includes the lateral and trunk sewers and storage
facilities within each drainage district. House connections are practically
the same for all alternatives and were therefore not considered in making
comparison.
The number of manholes was estimated assuming a uniform spacing interval
of 91 m (300 feet). For excavation depths of less than 8.5 m (28 feet), the
estimated unit cost was $2,500 and of more than 8.5 m (28 feet), $5,000. The
excavation depth of a manhole is measured as the average.excavation depth of
the sewer it serves.
The cost of sewers is computed in accordance with the unit prices
of Figure D-15 or TABLE D-17. The excavation depth was determined as the
average of the depths of the upstream and downstream ends of the sewer at the
junction manholes. The sewer depth is determined by its invert elevation.
At a junction manhole, where more than one upstream sewer connects, the
lowest crown elevation of the upstream connecting sewers was taken as the
83
-------
TABLE 10. VOLUME* IN SEWER PIPES AND STORAGE BASINS, DISTRICT A
Description
Separate Storm
Conventional Combined
8.3% In-Pipe
16.5% In-pipe
25.0% In-pipe
41.3% In-pipe
9.4% Satellite
17.0% Satellite
41.3% Satellite
Lateral
Sewer
(inches)
-
0.19
0.28
0.41
0.55
0.83
0.19
0*19
0.19
Trunk
Sewer
(inches)
-
0.37
0.28
0.23
0.18
0.15
0.43**
0.61**
0.94**
Total
(inches)
0.39
0.56
0.56
0.64
0.73
0.98
0.62**
0.80**
1.13**
* Total Volume expressed as inches of runoff over the entire watershed
of District A.
** Includes satellite storage volume. Note: 1 inch = 25.4 mm
84
-------
crown of the downstream sewer to avoid surcharging upstream sewers above
their crown when the downstream sewer flows full or near full. The minimum
sewer cover was 1.8 m (6 feet). Reduction of the minimum cover to 1.52 and
1.22 m (5 and 4 feet) would reduce the cost of District A sewer system by
about five and eight percent respectively.
A computer program was written to estimate sewer system cost. The
same computer program was used for all sewer system alternatives including
the separate storm system. Although the assumption of matching the down*-
stream sewer crown with the lowest of the upstream sewer crowns would some-
what increase the cost of the separate storm sewer system, the increase
should not alter the conclusion.
Cost of Conventional Sewer Systems
TABLE. 11 summarizes the costs of alternative conventional separate
and combined sewer systems in District A. Compared are the following
alternative systems:
1. A conventional combined system with a minimal amount of overland
flow permitted for urban runoff. In this system, all sewers would
carry combined sewage except those at the extreme upstream end of a
sewer branch. These upstream sewers would convey sanitary flow
only as previously described (Case 1).
2. A combined system with the same amount of overland flow as a
separate storm sewer system. The total length of sewers carrying
combined sewage is the same as the toal length of separate storm
sewers (Case 2).
Separate sanitary and storm systems with storm sewers surcharged. In
this alternative, a significant amount of overland flow is permitted
to reduce the length of storm sewers required (Cases 3 plus 4).
A separate sanitary and storm system with storm sewers designed
without surcharge (Cases 3 plus 5).
The costs summarized in TABLE 11 indicate:
1. A conventional combined system (Case 1) would cost about the same as
a separate system with surcharge (Cases 3 plus 4).
2. The combined system permitting significant overland flow (Case 2) is
the least-cost system with an estimated cost about five percent
lower than the cost of Case 1 and Cases 3 plus 4.
3. The most expensive system is Cases 3 plus 5 with a cost of about 13
percent greater than for Case 1 and Cases 3 plus 4.
4. Surcharging storm drains permits reducing their costs by about 20
percent.
3.
4.
85
-------
n
^-^
vO
O
H
•CO-
CO
2
W
H
CO
CO
erf
^
w
CO
w
H
M
3
CM
W
CO
Q
Q
jgj
M
§
O
t T
«<4
E3
O
M
H
Z
W
gj
O
O
Pn
O
z
0
CO
M
^ -,
M
M
cd
o
CO
CU
&
CU
CO
^
o
tH
>>
cd
4-1
•H
C
cd
CO
60
a
•rl
!-J
cd
o
CO
J_l
cu
cu
CO
cu
CO
cd
o
•X
•X
4J
CO
O
O
i-H
cd
4J
O
E-l
>-
a
o
CU
6t
cd
cu
CO
Tj
cu
0
•rl
d
a
o
CJ
CU
a
£_4
a
CU
CO
^*
co
•
0
^-^
0
o o to
-* o o
• «\ •
OO ^O 1^*
•H O i-l
•co- H
v-x
-X
HO en
C7\ *^" CO
• vO •
m * CM
i — 1 CO i — I
•CO- VO -l CO
60
CU CO
TJ Cd
CM
, 1
•
CO
eg
r— !
•
OO
1
1
1
^
S-f
cd
•H
a
cd
CO
•
CO
,— X
o
o
o
f\
vD
o
**-s
s~^
o
o
o
r\
0
1-1
OO
00
•
CTi
•co-
•~
1
1
1
1
1
TJ
CU
a
60
•H
CO
CU
TJ
a
}_i
0
J-l
CO
CU
cd
cd
ft
cu
CO
•
CN
^^
cu
60
M
cd
X!
O
^1
0
CO
•fi
•H
15
cy\
CO
•
CN
i-l
1
!
1
i
i
Tj
0)
a
M
•rl
CO
CU
TJ
a
o
4J
CO
cu
4J
cd
cd
cu
CO
•
LO
^-^
O
o
o
A
CT"*
CM
•^^
QJ
W)
I-l
cd
^cj
o
M
S
CO
4J
^3
0
•g
•H
13
a
CU
0)
g
CD
CO
0)
O
a
j-
4-1
60
a
CU
CU
j-i
CO
•H
CO
cu
4-J
d
cu
M
cd
ft
id
•rl
CO 4-3
(-1 CU
CU CU
r£3 M— 1
a
d PI
!3 'i-1
•X
TJ
cl
cd
CO
CU
•H
ft
S-l
0
>
CU
CO
M-l
o
4->
CO
O
O
60
d
•H
TJ
d
I~H
o
d
M
•X
•X
86
-------
For urban areas requiring storm and sanitary sewers, a combined sewer
system (Case 2) would be more economical than separate systems with respect
to the cost of collection sewers. For areas where the existing sewers
can be used to convey sanitary wastes, as in District A, sewer separation
is the least-cost alternative (Case 4 as compared to case 2). However,
where an existing system contains substantial capacity to convey part of
the storm flows, it may be desirable to intercept in a separate system
only sufficient runoff to prevent frequent flooding. As previously dis-
cussed, combined systems, designed as described in this report, also offer
advantages for pollution abatement because of the storage that can be made
available in the pipes.
In the subsequent cost comparisons of conventional'systems with advanced
combined systems, Case 1 is used since the configuration of the advanced
combined system was compatible with this case.
Cost of Advanced Sewer Systems
TABLE 12 presents a cost comparison of combined systems with various
amounts of either in-pipe storage provided in upstream lateral sewers to
achieve greatest economy or satellite storage along the trunk sewers.
The costs of collection sewers for the entire City were projected from
District A sewer costs by the ratio of drainage areas. These costs do
not include interceptors, downstream storage and treatment facilities.
Storing about 8.3 percent of the overland runoff from a 5-year synthetic
storm in the lateral sewers would result in the least cost for trunk and
lateral sewers in District A. The increased cost of lateral sewers is
more than offset by the reduced cost of trunk sewers. However, while there
is a small increase in cost (about $1.1 and $1.9 million, respectively) as
the storage volume is doubled and tripled, there is a substantial reduction
of peak outflow which the downstream interceptors and subsequently downstream
storage and treatment facilities may be required to accept. Depending
on specific circumstances, this increase on collection system costs may
be offset by capital investment required for downstream sewer facilities.
For a combined system with satellite storage, the potential cost savings of
providing storage may be greater than for a system with in-pipe storage
depending on the unit cost and location of the storage. In TABLE 12, sewer
system costs are presented with unit costs equal to $0.08 to $0.26 per
liter ($0.30 and $1.00 per gallon) of storage. This cost could vary de-
pending upon the type of storage basin required and the environmental re-
straints. With respect to District A, a satellite storage capacity of 20,800
m (5.5 million gallons) or about 17 percent of 5-year synthetic storm
runoff should result in the least cost in the three cases compared if the
unit storage cost is $0.08 per liter ($0.03 per gallon) of storage. At
a unit cost of $0.26 per liter ($1.00 per gallon) of storage, the cost
of collection sewer system increases with an increase in storage capacity.
87
-------
W
O
Si
o
CO
§
H
CO
X
CO
1
2
1-1
IT*
H
H
3:
CO
t£$
H
CO
X
CO
Jgj
O
I—I
w
r3
•J
O
O
Q
Z;
M
O
p4
O
^-4
p^
.3
*gj
s
CO
H
CO
0
CJ
w
CN
H
w
%
H
S •
O 4J
i-H CO
UH -rH
4J o
3 " ^
o S co
O MH
^i rl O
cd m •-'
0)
*
vo""*
o
•CO-
^
4J
CO
O
o
•K I—I
>-. cd
4-1 4-1
•rl O
0 H
*
^
• rH
• 1 fjt
4-J Cu
CO 4-1
•rl 0
0 H
a)
00
cd d
1-1 -rl
o co
4-1 Cd
co pq
^
•3 3
M CU
H CO
rH
H M
•3 EC
cd o
H) CO
co
d
o
a) I-H
B rH
3 cd
rH d O
O M
> d
t) 0
rH 0) -H
4J 0 rH
O 4J -H
H co S
d
o
•rl
4J
0 P.
d) *H
4J lj
co o
>S CO
co cu
a
a)
CO •
cd o
0 25
P~ vD OO
CD CT\ OO
-* O P-
rH rH
CT\ O 00
CM cy\ co
CN p*^ ^1*
CM rH CN
rH rH rH
O •* ON
oo P-. oo
rH rH rH
O O O
O CTl CO
CN CM CO
O\ P- VO
O m rH
co in in
, OO CTV rH
rH
P-. CO
vD CO
o CN m
CU 0)
rH pi CX
Cd iH -H
d p, p,
O 1 1
•H -o d d
4J CU M M
d d
CU 'rl B-S B-S
> _Q co in
d S •
O O OO vD
O CJ rH
• • •
rH CM CO
p^
O
vD
O
vO
0
rH
m
VD
<7t
rH
O
p~-
0
in
oo
vO
CO
rH
O
00
cu
•rl
Pi
1
d
w
gsg
o
•
in
CN
•
^
CO
•*
OO
o
CO
oo
g
CO
CU
4-1
•rl
rH '
i-H
CU
4-1
cd
CO
B*S
O
•co-
M-l
o
4-1
CO
O
CJ
CU
rH
O
rj
cd
£3
CO
CU
T)
rH
O
d
M
*
vJ3
*""
0
•rl
4-1
cd
cd
cu
j^
cd
t^>
-a
rH
id
4J
0
4.J
•"^
4-1
CJ
•rl
J,^
4J
CO
•H
"
&
0
M-l
•b
cu
•H
cd
,a
o
•K
*
allon) of storage
00
S-l
CU
Pi
o
CO
•
o
s^x
^_1
N^X
^
CU
4-1
•rl
rH
^4
CU
a
vD
CM
c
O
•oo-
d
o
0)
CO
cd
4J
CO '
o
0
•%
^4
^
•^
^
. SJX
CO
}_4
CU
4J
CO -H
& rH
m in
oo oo
P- P-
co •
CO
II
II
CO
d1
o
rH
, J
cd
00
d '
o d
•rl 0
rH rH
rH rH
•rl Cd
0 00
rH rH
CU
4-1
0
88
-------
SECTION IX
DESIGN AND COST COMPARISON OF ALTERNATIVE CITY-WIDE
SEWER SYSTEMS
An objective of this study was the evaluation of the cost—effectiveness
of alternative sewer system designs for a given level of pollution control.
For the various alternative collection systems developed (a total of 9),
the level of pollution control would determine the requirements for (a)
interception and treatment of runoff and combined sewage before discharge
and (b) capital investment required for interceptors, downstream storage
and treatment facilities. Since the hydrographs and pollutographs generated
for the alternative collection sewers are different, the downstream system
facilities to achieve the same level of pollution control can,also be
different. That alternative which achieves the desired level of pollution
control with the least total cost for collector, interceptor, storage and
treatment systems, is the cost-effective system.
Examination of the outflow characteristics of the various synthetic
storms from the alternative collection systems provided insight to the
interception and treatment requirements. The effectiveness in pollution
abatement of varying amounts of interceptor capacity and upstream storage
was evaluated using real rainfall events over a period long enough to provide
statistically significant information.
Near total elimination of runoff pollutants from District A associated
with runoff from the 5-year synthetic storm with intermediate pattern
would require the first 90 minutes of runoff from a conventional combined
system to be captured and conveyed by interceptors to storage and treatment
facilities (Figures 3 and 4). The costs of systems meeting this severe cri-
terion were estimated.
SYNTHETIC STORM RUNOFF
City-wide planning requires development of runoff hydrographs and
pollutographs for each of the City's drainage districts for the storms of
interest to develop the required interceptor and treatment storage facili-
ties. The necessary hydrographs and pollutographs for each drainage district
were derived from information developed in the detailed study of District A.
The two primary factors governing the shape and peak flow rate of
the hydrograph for a given drainage basin are area and percent of impervious-
ness. The area affects the peak flow rate, the flow duration and the peaking
time of the hydrograph. The percent ©f imperviousness, which is dependent on
89
-------
land use, affects the runoff volume and the peak flow rate. Study of 5-year
synthetic storm hydrographs computed with SWMM for various areas within
District A indicated that hydrographs are relatively similar in character
within a reasonable range of -variation in drainage area and percent of
imperviousnesSj and that dividing the volume by the peak runoff rate yields a
relatively constant value in time units. Such hydrographs may be "norma-
lized" by equating the defining ordinates to values based on Q/^peak where
Q is the flow rate at given time and %eak, the peak flow rate. The runoff
volume for the synthetic storm was estimated as in STORM. With the hydro-
graph constant determined from the data developed for District A, the peak
flow was obtained by dividing the volume by the hydrograph constant and the
values of the 5-year synthetic storm hydrograph obtained by multiplying the
peak flow by the ordinates of the appropriate "normalized" .hydrograph. The
maximum ordinate is equal to 1 (peak/ peak) and the hydrograph con-
stant equals the area of the normalized hydrograph. Five normalized hydro-
graphs were developed with data as summarized in TABLE 13. The normalized
hydrograph No. 1 is based on the hydrographs for sewer element^ numbers 205,
209, 219 and 223 with drainage areas of 13.2, 21.8, 58.3 and 21.0 hectares
(32.5, 53.9, 144.0 and 52.0 acres) respectively. The corresponding percents
of imperviousness of the areas drained to these elements are respectively
60.0, 55.3, 53.5, and 43.0. The estimated peak flow of 3.14 m / sec (111.0
cfs) for ,sewer element number 205 was obtained by dividing the runoff volume
of 634 m (22,400 cubic feet) by the hydrograph constant of the normalized
hydrograph No. 1 in seconds.
The estimated peak flow for element numbers 209, 219 and 223 were
computed in the same manner. The agreement between the peak flow computed
with SWMM and that computed using the normalized hydrograph, indicates the
percent of imperviousness affects the runoff volume more than the hydrograph
shape within the range of variation considered. For other normalized hydro-
graphs, the agreement between the estimated peak flow and that obtained from
SWMM is also good. The adjustment of the time of peak flow occurrence with
respect to District A is listed in TABLE 14 for four ranges of drainage
areas. The peak outflow from District A occurs 74 minutes after the start of
the 5-year synthetic storm. Figure 27 presents two of the normalized
hydrographs, one for a drainage area greater than 215 hectares (530 acres)
and one for area less than 60 hectares (150 acres). The hydrograph for the
larger drainage area has a greater spread and later peaking time. The
peaking time of the other normalized hydrographs fall between the two peaks
shown in Figure 27.
The 5-year synthetic storm runoff volume in drainage districts other
than District A were computed using the calibrated runoff coefficients of
0.55 for the pervious area and 1.0 for the impervious area. TABLE 15 sum-
marizes the relevant data involved in the computation of outflow hydrographs
for 24 drainage districts. Runoff from District Y drains toward the City of
Newark, and is not served by the Easterly or Westerly Interceptor.
Figure 28 shows the outflow hydrographs for Districts A, E and H
respectively. To develop outflow hydrographs for drainage districts other
90
-------
?i
o
^
EH
O
Pi
H
CO
w
a
Q
pS
EH
1
0
tJ
W
W
o
P3
PJ
n5
rh
Q
Q
£
o
w
ts]
M
»-J
si
3
o.
h
0
pd
S
§
r^
C/3
*
3'
w
•3
EH
(
-K
14H
D
rj
Pi
P*^
J_j
cd
4J
0
,Q
•H
M
H
j_i
0)
S
CU
* c/3
0)
rH
fi t| [
cd o
o
•H CU
rH 60
ft C
•^ Pi
•K
^
2
60
O
M
13
£>i
W
•x)
CU
N
•H
rH
cd
g
j_j
o
£5
X-N
CO T3
~~^
.
ft
M
X~s
CO
cd cu
CU M
V4 0
< <_
4J.
d JH
>
SB
O
rH
i — |
i — I
-a-
00
rH
i-H
^j.
CM
CM
O
•
O
VD
in
•
CM
OO
in
o
CM
CO
CO
CU
o
o
,«{
CM
r-4
CM
O
CM
•
rH
O CT>
OO CM
r-s, oo
I — 1 "si"
CM 00
00 VO
r*^ r*s
1 — 1 *xl-
CM VD
vo r^
co o1*
co m
• ' •
m co
m m
ON O
• •
CO <±
m -*
rH
CJN CT\
O rH
CM CM
CM
CM
m
rH
OO
rH
m
rH
00
o
co
o
•
CO
"*
o
•
CM
in
CO
CM
CM
0
in
rH
vO
O
in
rH
VD
rH
-
CM OO
00 00
r-~ CTI
r-~ oo
rH *st"
cr\ co
OO CM
rH CM
0 0
OO CO
-*.**
O'O
o o
01
O
0)
e
3
i-l
0
MH
14H
O
Pi
Pi
•K
4-J ^4
C cd
cd 01
en o1
rj
0 II
c_>
4J
£1 ti
ft cd
cd 4-i
M CO
60 C
o o
i-l O
' id
>>> rj
ffl ft
cd
II i-<
60
*c1J2
ft>,
0- M
• — ^ ^^
0) 0)
B e
3 3
rH rH
O O
0 0
0 Pi
f§ Pi
•3'
*5<
91
-------
TABLE 14. ADJUSTMENT OF PEAKING TIME WITH RESPECT
TO THAT OF DISTRICT A
of Drainage
Area A
(Acres)
Peaking Time Ahead of
District A
(minutes)
A <70
70 ^A <250
250 £A <500
500
-------
LLI "-I
gg
< <
o o
o in
in —
XV. "V
< r^
LU UJ
UJ UJ
O O
11
< <
ct cc
Q Q
t>o
o
cu
N
•H
bO
•H
MV3d 0/0
-------
TABLE 15. DEVELOPMENT OF OUTFLOW HYDROGRAPHS
Drainage
District
A
B
C
D
E
F
G
H
I
J
K
L
M
N
0
P
Q
R
S
T
U
V
W
X
Area
(acres)
655
112
111
117
229
88
34
122
70
420
65
66
67
79
103
37
83
303
207
62
331
440
415
138
Normalized
Hydrograph
Used
5
1
1
1
2
1
1
1
1
4
1
1
1
1
1
1
1
3
2
1
3
4
4
1
5-Year Storm
Runoff Volume
(106 ft3)
4.18 :
0.71
0.76
. 0.81
1.57
0.61
0.27
0.92
0.48
2.82
0.46
0.48
0.44
0.52
0.67
0.28
0.60
2.36
1.37
0.44
2.37
3.16
2.77
0.85
Peak
Flow
(cfs)
; 1407
351
376
401
671
302
134
455
237
1015
228
237
218
257
331
- 139
297
967
555
218
971
1137
997
421
Downstream
Inlet
204
200
202
202
202
206
206
208
210
212
210
110
212
: 212
300
300
300
300
108
106
104
102
102
100
Note: 1 acre =.405- hectares; 1 ft3 =0.283 m3
94
-------
en co co
co w to
LU UJ UJ
z z z
CO CO CO
=> => 3
ooo
LU UJ LJ
a_ CL a.
55 5? S5
is r^ ro
sr in N-
co~ to" co
UJ LU= LU
cc a: o:
ooo
o
a:
1-
O)
a
LU
1-
o
ro c\J
oaov/sjo) Moidino
95
-------
than District A for systems with in-pipe or satellite storage, the respective
ratios of outflow hydrographs of the various advanced combined systems to
the conventional combined system in District A were used.
DESIGN OF INTERCEPTING SEWERS
A computer program was developed to obtain the 5-year synthetic storm
outflow hydrographs for the 24 drainage districts using the appropriate
normalized hydrograph, runoff volume, and peaking time. The program output
was stored and used as input to the SWMM TRANSPORT Block for sizing inter-
cepting sewers. The layout of the intercepting sewer system is shown in
Figure D-14. It follows the existing Westerly and Easterly Interceptors.
Runoff from a drainage district is assumed to enter the interceptor at one
inlet only as indicated in TABLE 15. TABLE 16 .summarizes the pipe slope,
length and ground elevation at the upstream non-conduit element. The ground
elevation of each pipe element is used to compute the excavation depth for
the cost estimate of intercepting sewers. The Easterly Interceptor is about
6.4 km (4 miles) in length and the Westerly Interceptor, 4.8 km (3 miles).
The capacity of the existing interceptor is a small fraction of that required
to convey the peak 5-year synthetic storm runoff.
TABLE 16. INTERCEPTING SEWER PIPE ELEMENT DATA
Pipe
Element
No.
101
103
105
107
109
111
201
203
205
207
209
211
213
Slope
.0005
.0005
.0007
.0005
.0005
.0010
.0010
.0015
.0007
.0002
.0050
.0010
.0018
Length
(ft)
2517
5769
3982
3125
3470
2950
2500
2500
1000
2000
2000
3000
2970
Ground*
Elevation
(ft. MSL)
5.
10.
15.
17.
13.
8.
30.
27.
24.
16.
12.
10.
11.
* at upstream non-conduit element
In the analysis, new interceptors were assumed to be required and
only one storage basin and one treatment facility located near element
number 300. As mentioned before, the U.S. Army Corps of Engineers has
planned several diked storage basins with an estimated total storage capacity
of 80,000 m (21 million gallons) near this location. The Join^ Meeting
Sewage Treatment Plant, now being upgraded, will provide 1.76 m /sec (40
mgd) of its wet-weather flow capacity for use by Elizabeth. Hence, the
96
-------
assumption of centralizing storage and treatment facilities is appropriate to
the actual condition.
TABLE 17 presents a summary of intercepting sewer sizes for the alterna-
tives investigated defined by the type of collection system and the amount
and types of in-system storage. Except for sewers carrying sanitary wastes,
all other sewers were sized with the SWMM TRANSPORT Block. Intercepting
sewers for the conventional separate system were assumed to have dual pipes
in each pipe element, one carrying sanitary waste and the other storm flow.
This dual pipe sewer system would be more economical than one pipe carrying
both sanitary waste and storm flow because the off-line storage required
would be substantially less (TABLE 18). The sewers intercepting sanitary
wastes were designed using Equation 1 to determine peak domestic flow
plus peak commercial and industrial flows. The peak domestic flow of a
drainage district was computed using the district population shown in TABLE
D-13. The peak commercial and industrial flows were calculated from land
use distribution shown in TABLE D-12 and the assumed rate of 187,000 liters
per hectare (20,000 gallons per acre). The intercepting sewer sizes for
carrying storm flow only and for combined sewage are the same since the
difference of flow rate between the two systems is the sanitary wastewater,
which is small in comparison with the 5-year synthetic storm runoff rate.
The sewer sizes required for the advanced combined systems decrease
as the amounts of in—system storage increase, since the peaks of outflow
hydrograph from a drainage district decrease.
AMOUNT OF STORAGE REQUIRED VERSUS TREATMENT RATE
Ninety-two percent of the SS would be treated if the initial 90 minutes.
of runoff of the 5-year synthetic storm were captured assuming a
conventional combined system is provided in District A. The remaining SS
would be discharged untreated to the receiving waters. Since in-system
storage modifies the outflow pollutograph characteristics, the duration of
runoff to be captured for equal amounts of pollutant overflow varies with the
amount of storage provided. In District A, satellite storage capacities
equal to 9.4, 17.0 and 41.3 percent of the 5—year synthetic storm runoff
would require the capture of the initial 92, 100 and 124 minutes of runoff
respectively to capture for the treatment the same amount (92 percent of
the total in the 5-year synthetic storm) of SS. The greater duration of
runoff to be captured results since the period of significant pollutant
spillage from storage basins is extended as the storage volume increases.
For advanced combined systems with various amounts of in—pipe storage,
the duration of runoff to be captured was assumed to be the initial 90
minutes for the 5-year synthetic storm. This is the same as for a conven-
tional combined system. In—pipe storage, using a fixed orifice and overflow
weir (Figure 16) for control, can be effective during the peak portion of
runoff hydrograph starting about 60 minutes after the beginning of the
storm. During this period, dry-weather sewer deposits should be fully clean-
sed and most of the pollutants washed off the streets would be contained
in the next 30 minutes of runoff. In-pipe storage would delay somewhat the
97
-------
VD
O
CO
O
Q
w
&
B-S
M rH
M
O
CO B-S
, o
£ H
rH
CU
4J B-S
Cd -3"
o\
B-S
CO
rH
M: O
t_i m
O CM
4J
CO
CU frS
a. in
•H •
p, VD
B-S
CO
co
o o
VO CO
o o
r^ in
i-H
in in
OO VD
rH
o m
VD CM
rH
in o
VD CO
rH
m o
^ a
o o
O> VD
i-H
r-
omoooinomomo-s-
coco-sfcMincocor^ocococM
I-HrHrHrH rHrHi-Ht-HrHm
CM
inoinmomooomoco
^- vo vo - t-H in in in
rH rH r— 1 rH • rH rH rH i-H rH rH l-~
rH
oommommmoomo
r^r^covot — rH r*- CM co oo vo CM
1-HrHrHrH rH rH CM rH rH rH OO
O
omoinooommmoco
CMCMCOt-HinO1CM^a\CMCMVD
rH rH rH i-H i-HrH i-HrHO-
oooomomoinminrH
co^incoino^covoo-^covD
rHrHrHrH rHrHHrHrHm
VO
inmininooomommco
• •••••••••••
<]• in VD &
a e ra
o o >,
O O CO
•H cu
4J 4->
> « 4J
a d. to
O 0) >,
U CO CO
o
rH
o
o
s
m
oo
co
rH
oo
rH
o
l-%
o
CM
CM
t-H
O
cn
a.
co
CM
CO
CM
co
rH
m
CO
CD
CT>
O
oo
o
co
in
CT>
cd
03
m
rH
m
CM
CM
m
CO
o
^
o
^r
o
^r
in
CM
rH
m
CM
CM
O
CO
O
-a-
m
CM
m
CO
m
CO
CO
o
61
f
in
cd
cu
a,
MH
o
o
%
4J
cu
oo
•*
O
cu
4-J
o
4J
a
cu
PH
O
ts
rH
O
CO
o
m P^
o o
rH rH
a\
3
rH
O
CM
CO
O
CM
m
o
CM
o
CM
a\
o
CM
i-H
i-H
CM
VO
O
CO
O
98
-------
TABLE 18. OFFLINE STORAGE REQUIRED (106 GALLONS) AND COST FOR EQUAL
WEIGHT OF SS DISCHARGED TO RECEIVING WATERS
Case
No.
1
2
3
4
5
6
7
8
9
Sewer
System
Conventional*
Separate
Conventional**
Combined
8.3% In-pipe**
16.5% In-pipe**
25.0% In-pipe**
41.3% In-pipe**,
9.4% Satellite ***
17.0% Satellite +
41.3% Satellite"1"1"
Treatment Rate
40 mgd
26.67
(7.88)
84.87
(25.46)
78.80
(23.64)
59.91
(17.97).
49.20
(14.76)
37.66
(11.30)
74.21
(22.26)
f
74.81 .
(22.44)
56.98
(17.09)
200 mgd
22.44
(6.73)
78.32
(23.50)
72.28
(21.68)
53.40 ,
(16.02)
42.63
(12.79)
31.14
(9.34)
67.68
(20.30)
67.44
(20.33)
48.45
(14.54)
400 mgd
18 .'43
(5.53)
71.51
(21.45)
65.51
(19.65)
46 . 6.3
(13.99)
35.85
(10.76)
24.37
(7.31)
61.70
(18.51)
60.33
(18.10)
39.62
(11.89)
Number in the parenthesis are storage costs ($10^) for a unit
cost of $0.30 per gallon of storage.
*Capture of runoff starts 30 minutes and ends 70 minutes after
the start of storm.
**Capture the initial 90 minutes of runoff.
***Capture the initial 92 minutes of runoff.
+Capture the initial 100 minutes of runoff.
++Capture the initial 124 minutes of runoff.
Note: 1 million gallons = 3785 m3
1 mgd = 0.0438 m3/sec
99
-------
discharge of pollutant mass, but this delay would not be as long as with
satellite storage since, under the assumed operating mode, storage would not
be utilized until flows in excess of interceptor pipe capacity are experi-
enced.
Runoff from the separate storm system for the 5-year intermediate
pattern synthetic storm, as computed by SWMM, for the first 30 minutes is
practically pollution-free (Figure 3), and capture of this small portion of
runoff (Figure 2), should not be required. The dry-weather flow would be
fully captured in the separate system and only stormwater would overflow.
Since stormwater is less polluted than the combined sewage, less storage
should be required or a greater amount of overflow may be allowed. The
treatment of 92 percent of the SS would require storing runoff for a duration
of 40 minutes, from 30 to 70 minutes after the start of the synthetic storm.
The amount of storage required would also depend upon the rates of
stormwater runoff and combined sewage treatment. Runoff rate, up to the
treatment capacity, would bypass the storage to treatment. TABLE 18 presents
the appropriate off-line storage capacity for treatment rates of 1.75,8.77
and 17.54 m /sec (40,200 and 400 mgd). This range of treatment rates
indicates the effect of treatment capacity on;downstream storage costs of a
city-wide sewer system.
It was assumed that the required off-line storage volume would be
provided near element number 300.
The off-line storage volume required generally decreases with the amount
of in-system storage provided. The off-line storage volumes required for
9.4 percent and 17.0 percent of satellite storage are about the same since
the outflow hydrographs and pollutographs from these two systems are similar.
(Figures 22 and 23)
The preceding analysis was calculated using SWMM with STORM default
values. The values have not been verified by testing of field samples.
However, the relative relationships developed should be adequate for this
stage of planning.
COST COMPARISON OF ALTERNATIVE SEWER SYSTEMS
The cost of intercepting sewers was estimated as described for the
collection sewers. Sewer sizes greater than 4.3 m (14 feet) in diameter
were replaced by multiple parallel sewer pipes with sizes less than 4.3
m (14 feet) while maintaining the same total cross-sectional area. For
example, a 5.8 m (19 feet) diameter pipe as required for the pipe element
number 205 in a conventional combined system was replaced by two 4.1 m
(13.5 feet) diameter' pipes; and one 7.2,m (23.5 feet) diameter pipe for
element number 207 was replaced by three 4.1 m (13.5 feet) diameter pipes.
This enabled the use of the unit costs shown in Figure D-15 and TABLE D-17.
The cost of the intercepting sewers for the alternative systems are
shown in TABLE 19. The cost of intercepting sewers for the advanced combined
100
-------
H
C/3
><
CO
pel
w
CO
Q
M
IS
•>l
H
M
O
H
CO
O
o
H
O
H
W
S-l
0
4-1
ft
CU
CJ
^1
Ol
4-J
C
M
g
•H
CJ
CU
rH
O
U
CU
60
cd
M
CU
vO
o
4-1 60
CO 0
O"
U O
o
§•*
UJ
4-1
CO
M tj
CU 60
5
CO O
o
I~H CvJ
cd
o
*
rr-i
\J
60
e
o
-a-
/— V
V4 vO
Ol 4-1 O
fe CO rH
CU O -CO-
CO CJ v-'
e /-N
§ vo
4J 4-1 O
CO CO rH
>,o -co-
co C-3 N-^
6
M 3
CU 4-1
ji CO
CU >•>
CO CO
CU
co •
cd o
0 &
•x *
* *
p-^ P^
O 00
P- CM
in r-
CM CM
m in
00 Pv.
00 rH
in PV
CM CM
rH CM
co m
VO CM
in PV.
CM CM
vo m
m co
m cd
C ft
O 01
O CO
rH
00 -*
O 00
P- rH
•* O
CM CO
CM OO
o o
OO 00
•if CTv
CM CM
co m
OO VD
VO rH
•if O
CM CO
CTv O
CO 00
vo m
-* o
CM CO
•
!~v.
m
•
r^
ov
CT>
CM
CM
CM
rH
rH
cd
C
O
•H T3
4-1 CU
fi el
CU -H
> &
o o
O C_?
CM
•if 00
CO 00
0 0
CO 00
CM CM
CTV m
CM rH
rH f-»
CO P^
CM CM
OO OO
O VO
0 0
CO 00
CM CM
-* 0
VO 00
CTv -if
CM CO
CM CM
CO
O
•
r^.
00
o
cr>
t^I
rH
rH
CU
•H
ft
1
cd
H
6*S
CO
•
OO
CO
oo o
•if 00
CM CTv
CM m
CM CM
-a- oo
*^" ^3
CO vo
CM m
CM CM
CO rH
CM VD
CM Ov
CM m
CM CM
CO CM
P- P~
rH CO
CM VO
CM CM
VO
CO
•
f-
[-v^
00
oo
^
CM
rH
01
ft
•H
ft
1
ti
M
6-S
m
' •
vo
rH
•
rH
•
VO
m
o
VO
o
co
*~i
01
ft
•H
ft
1
c
H
6-S
O
r •
m
CM
m
r7in"
P~ •*
O> rH
rH -if
CM CM
VO CM
vO pv-
o r-
CM CO
CM CM
in in
CM
OO O
CM a\
CM CM
o o
in -it-
CD
CM CM
rH
o
•
CM
OO
•K
* *
vo" H"
rH rH
CO P~.
CM CO
rH rH
01
4J
•H
H
rH
01
4-1
cd
CO
6-S
•if
•
CT»
r-.
o'o'
in co
vo cy>
rH 00
CM CM
O VO
CO O
p~. in
rH CO
CM CM
o-\ co
rH CTv
VO 00
H 00
CM CM
O O
O CTi
VO CO
rH CTi
CM CM
CM
CO
•
U">
r^.
r**> CD
rH P**
P~. CM
rH cy,
rH CM
(-»»
-^"
•
CM
m
rH'cM
OO O
VO CTV
CM OO
H rH
CU
4J
•H
H
rH
CU
4-J
CO
B-S
CO
•
rH
~*
o\
rH
CU
•H
4-)
O
pi-
CO
0)
CO
cd
o
13
.200
•j rH rH
'j rH rH
rH Cd Cti
•rj M 60
3
ft ft
H O rH
H , rl W
i? o o
| CO CO
CJ
0)
CO
CO
&
oo
CO
£-J
.
o
II
4H
O
CO
4-J
§§
gg
01 01
CO CO
cd cd
pq pq
* *
*
0)
4J
O
101
-------
systems is less than for the conventional separate or combined systems.
The cost of off-line storage was estimated using unit costs of $0.08 and
$0.26 per liter ($0.30 and $1,00 per gallon) of storage provided. TABLE 18
uses only a storage cost for $0.08 per liter ($0.30 per gallon) of storage.
Capital cost of treatment for facilities was estimated using the SWMM STORAGE
/TREATMENT Block. The treatment processes were assumed to include bar racks
and microstrainers. Sedimentation is assumed to occur in the storage basins.
This essentially conforms to the contemplated operation of the Joint Meeting
Plant since peak wet-weather flows are proposed to bypass the biological
treatment units. The unit cost of the treatment processes and land, and the
amortization period and interest rate used, were the default values provided
in SWMM. The computed costs were converted to the ENR construction cost
index of 1800. The computed capital costs of treatment facilities were
$1.07, $3.47 and $6.71 million respectively for treatment capacities of 1.75,
8.77 and 17.54 m /sec (40, 200 and 400 mgd). If more sophisticated treat-
ment is required, these costs could increase by about 40 times and would
favor larger basins and lower treatment capacities.
TABLE 19 presents the total cost of the nine alternative sewer systems
including collection sewers, in-system storage, intercepting sewers, off-line
storage and treatment facilities. Two cost totals are given, one assumes a
storage cost of $0.08 per liter ($0.30 per gallon) and the other $0.26 per
liter ($1.00 per gallon). For each alternative, the total cost of the
sewerage system is about the same for the minimum treatment assumed regard-
less of treatment rate. However, if more sophisticated treatment is required
(such as activated sludge at the Joint Meeting Plant),, the lower treatment
rate would provide substantial cost advantage. The total costs are presented
in Figure 29 for the various amounts of in-pipe and satellite storage.
Points A and B are a conventional separate system computed with a unit cost
of off-line storage equal to $0.08 and $0.26 per liter ($0.30 and $1.00 per
gallon), respectively. The following conclusions can be drawn:
1. In areas where the cost of off-line storage is relatively high,
sewer separation would be more economical than a conventional
combined system. On the other hand, when storage can be econo-
mically provided, a conventional combined system would be more
economical than a conventional separate system.
2. The maximum amount of in-pipe storage that may be optimally pro-
vided is about 30 percent of the 5-year synthetic storm runoff,
within the range of costs assumed for off-line storage. •
3. The conventional separate system would be the most expensive system
if storage (off-line and satellite) can be provided at a cost of
$0.08 or less per liter ($0.30 or less per gallon). ,
4. At a storage cost of $0.26 per liter ($1.00 per gallon), the
conventional separate system would be more economical than an
advanced combined system with any amount of satellite storage
provided.
102
-------
.UJ
; o u-
UJ
o
1 *
o
CO
UJ
h-
_l
UJ
£
c^
Ij <
j cc ;
< P <
WOT c
cc c
LU LU L
CL Z C
^ LL C
O LL
=te-0 =t
J< <->(/,
t g |_ CO
^
CM
O
CO
CM
O
1^-
CM
O
CO
CM
O
in
CM
O
CM
/
/
O
CO
CM
O
r-
CM
o
CM
O
in
CM
O
-------
5. In all instances, an advanced combined system offers economic
advantage over the conventional combined system. If the storage
cost is $0.26 per liter ($1.00 per gallon), the total system cost
would be reduced by about four percent when an optimal storage
volume of about 20 percent of the runoff is provided. The cost
would be reduced by 20 percent if the storage cost is $0.08 per
liter ($0.30 per gallon) and the storage volume about 40 percent of
the 5-year synthetic storm runoff.
The above comparison demonstrates the effect of storage costs on the
cost-effectiveness of the alternative systems. Assuming a cost of $0.08
per liter ($0.30 per gallon) of storage, the least-cost system for Elizabeth
for treating 92 percent of the combined sewage pollutants from a 5-year
synthetic storm would be about $200 million. This cost is excessive and
alternatives are explored in Section XII.
104
-------
SECTION X
ANALYSIS OF OVERFLOW QUANTITY AND QUALITY
In previous analyses, collection sewers, storage basins and interceptors
have been developed to essentially provide treatment for about 92 percent
of the computed pollutants in runoff and municipal wastes resulting from
a long return interval, synthetic, intermediate pattern storm hyetograph.
Pollutants in combined sewage are subject to random hydrometeorological
fluctuations. Use of the same "design storm" for pollution and flood control
implies the desired degree of water quality protection can be equated to a
level of flood protection in an urban area. This is not valid. The severity
and frequency of critical storms for the planning of facilities for control
of urban flooding and of pollution is different. While the total pounds of
pollutants contained in the runoff from a 5-year synthetic storm is greater
than, in the runoff from individual frequent storms, the total pollution
generated by the frequent storms is greater when weighted by the number
of expected occurrences per year. The concentration of pollutants in a
frequent storm may also be greater. In the Elizabeth, area, about 93 percent
of the precipitation events have a gross rainfall amount of less than 25 mm
(1.0 inch) with the accompanying street pollutant washout aggregating more
than 73 percent of the total.
To develop reasonable criteria for pollution control, the charac-
teristics of runoff quantity and quality from frequent storms (1.3 and 0.35
month return frequencies) were evaluated quantitatively using SWIM. Overflow
control effectiveness was also studied, using STORM, to evaluate (a) the
dynamic response of the catchment to a series of historical rainfall events,
(b) the consequences (overflow quantity and quality) of such events, and (c)
the results in terms of an average performance characteristic. The benefits
of a combination of storage and treatment (or interceptor) capacity were
also comparatively quantified in terms of overall performance under the
variety of hydrometeorological conditions experienced in nature. The bene-
fits were based on the characteristics of overflow quantity and quality,
using long-term precipitation records. Statistical analysis developed the
probable number of annual overflow events, their total volume and their
discharge quality (total pounds of pollutants) for 32 alternatives.
Treatment rate, as normally referred to in STORM, was used as inter-
ceptor capacity. Overflow from each drainage district was controlled by the
interceptor capacity selected. Because storage basins are possible near the
treatment plant for flow equalization, treatment rate and interceptor capa-
city could be different.
105
-------
Since STORM considers pollutant accumulation on streets, it was used to
evaluate the effectiveness of street sweeping practices for source control.
Data, developed for District A, was used to determine the effectiveness of
overflow control measures by extension to other drainage districts in
the City. Simulation of urban runoff from a separate system was made by
specifying zero storage and bypassing the dry—weather flow (which would be
entirely diverted to treatment) routines. Figure 30 presents the annual
amount of SS in street washout and combined sewage overflows versus the
amount of annual rainfall. More pollutants should be removed from streets
and greater pollutant loads discharged in combined sewage overflows during a
wet year than a dry year.
EFFECT OF STORAGE AND INTERCEPTOR CAPACITY
On the basis of a continuous 12-year precipitation record at the Newark
International Airport and an assumed capture rate for treatment equal to
the peak dry-weather flow, 65.8 combined sewage overflow events should
be expected annually from a conventional combined sewer system. Yearly
overflows should have lasted a total of more than 3.2 percent of the time.
The yearly total overflow volume should have equaled-343 mm (13.5 inches)
over the entire watershed of District A or 908,000 m (240.0 million gal-
lons). Based on STORM default values, the pollutant overflow yearly should
equal 23,300 kg (51,300 pounds) of SS and 11,000 kg (24,300 pounds) of
BOD. All sanitary wastes were assumed to be treated during dry periods.
TABLE 20 summarizes the results of computer simulations over a 12-year
period for various combinations of storage capacity and rates of interception
for treatment. The number of overflow events, their volume and the mass
discharges of SS and BOD are listed. Computer runs were made for storage
capacities ranging from 0 to 50,300 m (0 to 13.3 million gallons) and
interceptor capacities from 0.10 to 40 m /sec (2.4 to 910 mgd). Case No.l
in TABLE 20 represents discharge from a conventional separate storm system.
i
Figures 31 and 32 present the annual average number of overflow events
and mass SS discharged for various combinations of storage volume and inter-
ceptor capacity. The number of overflow events and the amount of pollutants
discharged could be controlled by increasing storage or interception rate or
a combination of both. For a given storage volume, ah increasingly greater
interceptor capacity would be required to attain each successive increment of
pollutant reduction. This diminishing return also applies, for a given
interceptor capacity, to storage volume.
The amount of SS and BOD contained in street washout in District A
is estimated as 15,300 and 2,600 kg (33,800 and 5,800 pounds) respectively
(Case No. 1 of TABLE 20) based on the STORM default values for computing
pollutants from dust and dirt washed off the street. This much SS and BOD
would reach the receiving waters through a separate storm system with no
treatment. With an assumed interception rate of 0.26 m /sec (6.0 mgd), the
peak DWF, a conventional combined system would annually discharge about
23,300 kg (51,300 pounds) of SS and 11,000 kg (24,300 pounds) of BOD (Case
No. 3). These amounts are substantially greater than those estimated for
106
-------
-Q
t
o
-------
TABLE 20. EFFECT OF STORAGE AND FLOW INTERCEPTION OF DISTRICT A
SEWER SYSTEM DISCHARGES
Case
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16*
17
18
19
20
21
22
23
24
25
26*
27
28
29
30
31
32
Storage
(MG)
0
0
0
0
0
0
0
0
0
2.7
2.7
2.7
3.0
3.0
3.0
5.3
5.3
5.3
5.3
5.3
8.9
8.0
8.0
8.0
8.0
13.3
13.3
13.3
13.3
13.3
13.3
13.3
Intercepting DWF
Rate
(mad)
0
2.4
6.0
22.6
41.0
98.9
161.0
355.0
910.0
2.4
6.0
41.0
69.8
82.7
140.9
2.4
6.0
69.8
82.7
140.9
2.4
6.0
22.6
41.0
88.6
2.4
6.0
12.9
22.6
31.7
41.0
69.8
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Number
of
Overflow
Events
. 69.5
68.5
65.8
40.3
22.6
6.6
2.3
0.2
0.
28.0
24.4
7.8
3,3
2.7
1.1
16.0
13.1
1.6
1.4
0.6
9.0
7.8
3.3
1.8
0.8
5.0
3.7
1.8
1.3
1.1
0.8
0.5
Quantity
(MG)
298.7
285.5
240.0
121.0
69.4
21.3
7.8
0.9
0.
185.0
138,7
34.5
16.4
13.0
5,3
131.6
90.4
11.0 .
•8.7
3.7
90.2
63,5
26.5
14.4
5.5
62.1
35.0
23.7
16.2
11.6
7.8
4.1
SS
(Ibs.)
33822
68242
51337
21387
11335
2702
611
23
0
37631
23466
3665
1215
859
203
21658
11913
449
313
92
12962
6670
1355
468
122
6373
2345
995
451
268
162
83
BOD
(Ibs.)
5803
35528
24345
7138
3113
620
149
9
0
18573
10199
1002
336
242
73
11315
5542
169
126
46
7588
3592
622
234
75
4109
1496
630
310
192
119
61
NOTE: (a) Assumed Street Sweeping frequency is 7 days and efficiency
is 75 percent.
(b) STORM default values were used in computing pollutants from streets
(c) *Runs made with precipitation data (1965-1971), otherwise (1963-
1974)
Note: 1 mgd
2.628 m3/min; 1 MG = 3785 m3; 1 Ib = 0.453 kg
108
-------
CURVE A-STORAGE
CURVE B-STORAGE
CURVE C-STORAGE
CURVE D-STORAGE
CURVE E-STORAGE
CURVE F-STORAGE
CAPACITY
CAPACITY
CAPACITY
CAPACITY
CAPACITY
CAPACITY
= 0
= 2.7MG (8.3 % OF*)
= 3.0 MG (9.4 % OF *)
= 5.3 MG (16.5 % OF *)
= 8.0 MG (25.0% OF*)
= I3.3M6(4I.3% OF *•)
* 5-YEAR SYNTHETIC STORM RUNOFF VOLUME
• DATA POINT
0 20 40 60 80 100 120 140 160 180 200 220 240 260
INTERCEPTOR CAPACITY (mgd)
Figure 31. Annual number of overflow events for various storage and
interceptor capacities
109
-------
100,000
CURVE A-STORAGE CAPACITY = O
CURVE B-STORAGE CAPACITY = 2,7 MG (8.3% OF*}
CURVE C- STORAGE CAPACITY =3.0 MG (9.4% OF *)
CURVE D-STORAGE CAPACITY = 5.3 MG (16.5 % OF *)
CURVE E -STORAGE CAPACITY = 8.0 MG (25.0% OF #)
CURVE F-STORAGE CAPACITY =13.3 MG (41.3% OF *)
* 5-YEAR SYNTHETIC STORM RUNOFF VOLUME
• DATA POINT
o: 10,000
LU
CC
LJ
DL
I
o:
UJ
CO
CO
U_
o
CO
Q
z
Z)
£ 1,000
100
I
20
40
60
Figure 32.
80 100 120 140 160 180 200 220 240 260 280
INTERCEPTOR CAPACITY (mgd)
Annual SS overflow for various storage and interceptor
capacities
110
-------
a separate system indicating the pollutant contribution of DWF. The sanitary
waste lost through overflows is estimated to be about two percent of the
total. Since overflows may be experienced about 3.2 percent or more of the
time, it appears that as much as 60 percent of the sanitary wastes might be
lost during overflow periods. These percentages were determined from the
estimated total weight of annual SS and BOD contained in the sewerage wastes.
The high percent of sanitary waste lost in the overflows, coupled with the
fact that most of the loss occurs during the frequent rainfalls and is
accompanied by high concentrations of pollutants, makes their control signi-
ficant to upgrading receiving water quality.
To reduce the SS in the overflow from District A by two orders of
magnitude, say to 277 kg (611 pounds) per year (Case No. 7 of TABLE 20), it
would be necessary to upgrade interceptor capacity to 7 m /sec (161 mgd) or
to about 18 percent of the peak 5-year synthetic storm runoff estimated for
District A. The same amount of SS reduction can be accomplished if 50,300
m (13.3 million gallons) of storage is provided and the interceptor
capacity is equal to about 0.9 m /sec (20 mgd). To reduce SS in the
overflow by one order of magnitude, an interceptor capacity of about 2.7
m /sec (62 mgd) would be required. Alternatively, the same reduction
should result from upstream storage of 36,300 m (8.0 million gallons) with
interceptor capacity (Case No. 22 of TABLE 20) equal to the peak dry-weather
flow. Hence, combinations of storage and interceptor capacity could be used
to advantage in meeting pollution control objectives.
Data in TABLE 20 permit a quantitative evaluation of the effectiveness
of various pollution control measures.
EFFECT OF STREET SWEEPING PRACTICE
Dust and dirt accumulated on streets during ,dry days contributes to
water pollution. A source control alternative is more frequent and effective
street sweeping. Computer runs, summarized in TABLE 20, assume a street
sweeping interval of seven days and a sweeping efficiency of 75 percent.
The reported removal efficiency of the dust and dirt averages about 50
percent and improvement to 70 percent and possibly as high as 90 percent may
be feasible under suitable conditions if the parked car problem can be
surmounted (19).
TABLE 21 presents the simulation results, using STORM, for various
street sweeping efficiencies and intervals. The average quantity of SS
in street washout is plotted in Figure 33 for various cleaning intervals
and appears almost linearly proportional to the street sweeping interval. If
the sweeping efficiency is reduced to 0.5, the amount of pollutants expected
would increase by about 35 percent (Case Nos. 3 and 5 of TABLE 21). Case
Nos. 6 and 7 present the effect of street sweeping efficiency on combined
sewage overflow quality. The numerical difference in pollutant loadings
between these two cases is about the same as that between Case Nos. 3 and
5. "
Simulation analysis by others (47) using STORM indicates the amounts of
111
-------
CO
,0
M
O
M
i-H
o
PM
CS
53
M
Pn
CO
H
CO
g
B
w
53
ts
o
CO
4J
0)
(U
PH
CM
pa
•H
o
•H
14-1
03
> to
M
a) ctf
0)
CO •
cd o
U
CO
o
VO
CO
cr\
rH
CO
OO
rH
CTi
CO
m
vo
m
-------
1 OC I
: UJ
-------
pollutant contained in urban runoff increases rapidly as the interval
between successive street sweepings increase until the interval exceeds about
20 days. Beyond this interval, the amount of pollutants in the urban
runoff becomes essentially constant, and the entire cost of street sweeping
may be attributed to aesthetics. At lesser sweeping intervals, part of
the cost may be attributed to reducing pollution.
114
-------
SECTION XI .
INTERCEPTORS FOR POLLUTION CONTROL
For effective .combined sewage overflow pollution abatement, the "first
flush", which usually occurs early in a storm runoff event, should be cap-
tured for treatment. The interceptor capacities required to contain the
"first flush" for the alternative collection sewer systems considered in
District A are defined herein. These interceptor capacities are further
considered in Section XII to develop and compare alternative combined sewage
overflow pollution control programs for the City of Elizabeth.
THE "FIRST FLUSH" PHENOMENA
As previously discussed, the outflow mass rate pollutograph in a com-
bined system exhibits two peaks. The first peak results from the flushing of
dry-weather flow solids deposited in the sewer and the second from street
washout. The first peak is of low volume but with a high concentration and,
perhaps, the larger mass of pollutants, while the second is of high volume
but with much lower concentrations as well as less mass of pollutants. The
outflow concentration pollutograph exhibits only one peak which occurs early
in the storm when the flow rates are small. The peak concentrations may be
an order of magnitude higher than that found in dry-weather flow. In consi-
dering the impact on a free-flowing stream, the peak of the overflow concen-
tration pollutograph may be of primary concern. The characteristics of this
peak are tabulated as follows:
Synthetic
Storm
Return
Frequency
5-year
1-year
1.3-month
Peak
Overflow
Concentrate
SS
(mg/1)
20
25
20
22
92
20
Time to
Peak Overflow
Concentration
(minutes)
56
50
70
60
60
84
Runoff Volume*
To Time of
Peak Overflow
Concentration
(mm) (inches)
Maximum
Flow Rate
to Inter—
ceptor
(m /sec) (cfs)
2.50
1.60
3.16
1.40
0.30
2.62
0.098
0.063
0.144
0.055
0.012
0.103
7.05
5.86
15.55
4.33
0.99
6.09
249
207
449
153
35
215
* Expressed in terms of area of District A (265 hectares or 655.1 acres)
115
-------
A significant finding of the above tabulation is the need to control a
greater volume of combined sewage for a 1.3-month synthetic storm than for a
5-year synthetic storm to achieve equal pollution abatement.
The estimated storage required to contain the critical first flushes
from the entire City is less than 80,000 m (21 million gallons), which
is equivalent to about 4.6 mm (0.18 inches) over the drainage area. This
storage capacity is being developed by the U.S. Army Corps of Engineers
along the Elizabeth River near the Joint Meeting Treatment Plant.
Interceptor capacities required for containment of the first flush
from District A are presented in TABLE 22 for storage basins of varying
size. The location and operating mode of storage basins influence the
amount of interceptor capacity required to limit the pollutants in overflows
TABLE 22. INTERCEPTOR CAPACITY REQUIRED TO CONTAIN
FIRST FLUSH FROM DISTRICT A
„>
Interceptor Capacity *** (cfs)
Storage* Synthetic Storm Return Frequency
Capacity 5-Year 1-Year 1. 3-Month
0
9.4
17.0
41.3
249
218
218
20
549****
358
361
137
215
127
128**
49**
***
****
expressed percent of runoff determined using SWMM and the synthetic,
intermediate pattern, 5-year return period hyetograph. The storage
basins are located as defined in Figure 22.
with storage located at the confluence of the trunk and interceptor,
the entire storm volume would be stored.
assumes intermediate pattern storms and suspended solids concentra-
tion of less than 20 mg/1 in any overflow.
reduces to 153 cfs if the maximum allowable concentration in the
overflow is 22 mg/1.
Note: 1 cfs = 0.0283 m /sec.
116
-------
to a specified value. The operating mode has a greater effect upon storage
located close to or at the confluence of the trunk and interceptor sewers
than if located upstream. If storage is operated so that there is zero
discharge from the basin until it is full, the interceptor capacities re-
quired for upstream and downstream storage locations for storms of various
return intervals are tabulated below. This tabulation assumes intermediate
pattern storms and a maximum permissible concentration of SS in the overflow
of 20 mg/1 using STORM default values.
Interceptor Capacity
Storage
Location
Upstream*
Downstream
Storage*
17
41.3
17
41.3
Synthetic Storm Return Frequency
5-year 1-year 1. 3-month
m /sec cfs
3 3
m /sec cfs m /sec
cfs
6.17 218 10.22 361 3.62 128
0.57 20 3.88 137 1.39 49
39.85 1407 15.52 548 No Discharge
25.20 890 0.28 108 No Discharge
* In percent of runoff computed using synthetic, intermediate pattern,
5-year return interval storm hyetograph.
** Located where defined in Figure 22.
Runoff from the frequent storm (1.3-month return frequency) would be
completely contained within the downstream storage volume. The rare storm
(5-year return frequency) would require interceptor capacity equal to the
peak runoff rate with 17 percent downstream storage volume and equal to about
63 percent of the peak runoff rate with 41.3 percent downstream storage
volume. Less rare, but still severe, storms (1-year return frequency)
would require about 52 percent greater capacity with 17 percent downstream
storage volume, but about 21 percent less capacity with 41.3 percent down-
stream storage volume than those with upstream storage. This reversal in
trend basically results from less dilution of the first flush with smaller
amounts of storage. Where the interceptor capacities indicated above for
downstream storage are in excess of those shown for upstrem storage, they
could be reduced by changing the operating mode to permit discharges equal to
the interceptor capacity during the storage fill period.
117
-------
SECTION XII
ALTERNATIVE POLLUTION CONTROL PROGRAMS
While removing almost all pollutants from a 5-year synthetic storm
might be desirable in areas draining to environmentally sensitive waters
demanding very high quality effluent discharges, it is not normally required
to protect the environment. In the development of a feasible program for
acceptable pollution control in Elizabeth, a maximum treatment rate of
1.75 m /sec (40 mgd) (as is available to the City at the Joint Meeting
Plant), and a storage capacity of about 80,000 m (21 million gallons) near
the treatment plant (as is being developed by the Corps of Engineers) are
assumed to be an integral part of the cost-effective program.
The alternatives compared are shown in TABLE 23. The variables include
(a) ratio of interceptor capacity to the peak 5-year synthetic storm runoff
rate and (b) amount and location of in-system storage. The interceptor
capacities investigated ranged from 1.4 to 100 percent of the peak 5-year
storm runoff rate as computed from the intermediate pattern synthetic hyeto-
graph. The overflow concentrations for the selected standard, intermediate
pattern storms of 5—year, 1-year and 1.3-month return intervals were obtained
using SWMM. The annual overflow frequencies and mass pollutant discharges
were obtained using STORM and are based on analysis of a 12-year, continuous
real rainfall record. The annual overflow characteristics are the primary
influence in determining the "cost-effective" alternative. Hence, the
findings are based on real rainfall events rather than on synthetic hydro-
logic phenomena.
For each alternative, interceptor sewers were selected and the cost
estimated. The ratios of regulated flow to the peak 5-year synthetic
storm runoff obtained from District A were used to modify the unregulated
5-year synthetic storm hydrographs of the remaining drainage districts to
obtain the regulated outflow hydrographs required to size the interceptor
sewer. Downstream storage of 80,000 m (21 million gallons) is more than
adequate for containment of the environmentally significant portion of runoff
except for Case Nos. 6 and 7 where additional storage would be required. The
cost of downstream storage included only the pumping station required to
lift interceptor flow to the storage basins for later treatment. The unit
cost of satellite and off-line storage of $0.08 per liter ($0.30 per gallon)
was used in the comparison.
CONVENTIONAL COMBINED SYSTEM
Case Nos. 1 to 7 of TABLE 23 apply to conventional combined systems
118
-------
M
S3.
CO 01 O irt O i-l CM mvDu-tmiHOONiHC
co co [-* 1-1 o \o o\ CM co co CM CM
rH i-H CM t-t iH ,-|
CO CO vO CO u") CO CM COt
H
3
j o] c-l c-l
O
o
cji o m co o
^ooor-*\oio-a-co^
119
-------
with various rates of flow intercepted for treatment. The effect of increas-
ing interceptor capacity on pollution control and costs are evident. Figure
34 presents for various ratios of interceptor capacity to the peak runoff
rate of the adoped 5-year synthetic storm the estimated (a) combined sewage
SS concentrations discharged to receiving waters, (b) annual number of
overflow events from District A and (c) cost, assuming new sewers. The
difference in cost between the systems results from the alternatives for
interceptor sewer and pumping station capacity. The cost of the collection
system is the same for all alternatives.
The findings in TABLE 23 are summarized as follows:
1
2
3
4
5
6
7
Pollutants*
Captured for
Treatment (%)
58
95
98
98
99
99.9
100
Number of
Overflows per year
(1 hour or more)
40.3
6.6
3.8
3.5
2.3
0.2
0
Overflow in
1-i 3-month Storm
Duration (Min.)
26
18
4
0
0
0
0
* % of annual SS overflow for Case No. A (TABLE 23) captured and '
treated.
In terms of pollution control, there appears to be little reason to
provide facilities better than those included in Case Nos. 3 and 4. Based on
the estimates developed herein, it appears the maximum justifiable cost for
combined sewage pollution control is less than $96,000 per hectare ($25
million per square mile) and is about 21 percent greater than for Case No. 1
which would only capture for treatment about two-thirds of the pollutants.
By utilizing the storage available in the combined system through devices to
control routing as previously discussed, pollution control better than that
offered by Case No. 4 might be achieved for somewhat more than $89,000 per
hectare ($23 million per square mile) (Case No. 11 with a conventional system
plus storage regulators).
ADVANCED COMBINED SYSTEM
An advanced system differs from a conventional system in that storage
capacity and controlled routing is built into the system. The conventional
combined system, designed to convey flows determined using the 5-year return
frequency, synthetic, intermediate-pattern hyetograph, has sufficient volume
to prevent overflow from storms with a return frequency of 1.3-month if flow
routing devices are installed.
A storage facility may be operated to permit either (a) no diversion
to storage until downstream sewer capacity is reached (to minimize operating
cost), or (b) diversion to storage with controlled, low rate outflow until
storage is full (to maximize pollution control benefits). With the first
120
-------
400 i-
-.200
NO STORAGE PROVIDED
NUMBER OF OVERFLOW EVENTS
60
50
40
CO
I-
z
LU
Ld
5?
o
li-
ce
Ld
30
cr
Ld
m
20
<
z
z
<
10
120
O.I 0.2 0.3 0.4
INTERCEPTED FLOW/PEAK 5-YEAR STORM RUNOFF
Figure 34. Pollution Control Costs - Conventional Systems
121
-------
operating mode, the peak flow to be accepted for pollution control is
a~bout 15 percent or less of the peak runoff from the synthetic storm of
5_year return interval used herein. This amount is less than the reduced
peak flow from the 5-year synthetic storm for the amounts of upstream, storage
investigated (up to 41.3 percent of the 5-year synthetic storm runoff vol-
ume). Hence, upstream storage would not reduce the size of intercepting
sewers required for pollution control under this operating mode.
With the second operating mode, the highly polluted initial runoff
is mixed in the basin with the less polluted later runoff. Case Nos. 8
to 14 of TABLE 23 compare alternative advanced combined systems with upstream
in—system storage. The findings are summarized as follows:
Pollutants
Case Captured for
No. Treatment (%)
8 98
9 99+
10 100-
11 99+
12 99+
13 100-
14 .100-
Number of
Overflows per year
(1 hour or more)
1.8
1.1
0.5
2.7
1.4
1.1
0.6
Overflow in
1.3-month Storm
Duration (Min.)
26
0
0
0
0
0
0
There appears little justification in providing facilities better than
those provided in Case No. 8. This alternative costs about $81,000 per
hectare ($21 million per square mile) but is predicated on a large, volume
of upstream storage. Case Nos. 11 and 12, and 13 and 14 compare costs
for lesser amounts of storage with equal interceptor capacity. In both
instances, increased storage reduces costs by about $3,900 per hectare ($1
million per square mile). All the alternatives provided for a high degree of
solids capture at a cost ranging from six percent to 28 percent greater than
a conventinal combined system. The upper permissible limit of expenditures
would.be fixed by that system which exploits the storage available with
the conventional combined system most effectively (similar to Case No.
11). This system cost is estimated at about $89,000 per hectare ($23 million
per square mile) and is about 19 percent greater than the cost of a conven-
tional combined system.
The effect of upstream versus downsteam locations for satellite storage
facilities is shown as Case Nos. 15 and 16. Relatively small amounts of
storage (Case Nos. 11 and 15) are more advantageously located upstream
for pollution control if the capital costs are to be maintained about equal.
Large amounts of storage in the upstream collection system (Case Nos. 10 and
16) result in lower costs than those at the confluence with the interceptor,
since the size of the trunk sewers can be reduced.
The relation between mass pollutant capture and system cost for the
indicated range of interceptor capacity is presented in Figure 35.
122
-------
175
NOTE:
84 i-
STORAGE LOCATIONS ARE
DEFINED IN FIGURE 40.
86 -
88
CO
co
90
NO STORAGE
O
o.
co
CO
92 •
UJ
QL
O.
<
O
UJ
•o
cc
94 -
96 -
98 -
100
170
165
(O
160 —
UJ
I-
co
CO
155
% CAPTURE
COST
ASSUMED STORAGE COST
0.30 PER GALLON
150
a:
ui
S
UJ
CO
u_
o
CO
O
o
145
140
135
0.05 0.10 0.15
INTERCEPTED FLOW/PEAK 5-YEAR STORM RUNOFF
Figure 35. Pollution Control Costs - Advanced System
123
0.20
-------
Interpretation of this figure indicates the following for a capture
rate of 96 percent of the suspended solids:
In-System**
Storage
0
9.4
17.0
41.3
Ratio:
Interceptor Capacity
to 5-year Peak* Storm
Runoff
0.124
0.057
0.031
0.006
Capital
Cost,
($iob)
164.4
151.2
136.0
138.5
* Based on intermediate pattern synthetic hyetograph
adopted for this report.
** In percent of volume determined from intermediate pattern, 5-year
synthetic storm hyetograph.
For lesser capture rates, there is almost no cost reduction if 41.3
percent storage can be provided, but the cost reduction increases with
decreased capture as the amount of storage reduces. The minimum cost still
occurs at about 17 percent storage.
For greater capture rates, the greater the amount of storage within
the range tested, the lower the increase in costs. At about 98 or more
percent capture, it becomes economical to provide 41.3 percent storage.
DESIGN STORM CONSIDERATIONS FOR SEPARATE AND COMBINED SYSTEMS
In design of a separate storm system, the pollution control aspects are
less controlling than the flood damage aspects. Hence, cost-effectiveness
is largely determined by benefits to be gained from relief of flooding.
In design of a combined system, the pollutional control aspects are predomi-
nant and cost-effectiveness is determined by the facilities necessary to
achieve appropriate pollution abatement. These conflicting standards affect
the selection of the design storm.
As the sewered area contributing to a single outlet increases, the
cost of sewers increases, and at a more rapid irate than the drainage area.
This results since the relative lengths of larger size pipe required in-
creases and the unit cost for pipe greater than about 2.1 m (84 inches) in
diameter increases faster than the carrying capacity. Hence, for protection
of a fixed property value from flood damage, the justifiable investment in
storm drains would remain essentially fixed, and as the drainage area^in-
creases, the storm return frequency for which protection can be provided
reduces•
In considering pollution abatement, however, in-pipe storage can be
124
-------
justified as a significant element of a cost-effective program. As pre-
viously discussed, the conventional combined system has about the same
total pipe volume as the advanced combined system with 8.3 percent in-pipe
storage, and can provide storage for about 3.8 mm (0.15 inches) of runoff
with appropriate flow control devices (TABLE 10).
If a 2-year synthetic storm is used, the total volume in the sewer system
would be reduced by about 5.1 mm (0.20 inches) or the potential for in-system
storage would be significantly reduced. However, the cost of the collection
system would be reduced by more than 20 percent, or more than $13,900 per
hectare ($3.7 million per square mile). At a cost of $0.08 pej: liter ($0.30
per gallon), this saving could build more than 45,000 m (12 million
gallons) of storage or more than 13,600 m of storage at $0.26 per liter
(3.7 million gallons at $1.0 per gallon). These volumes are equivalent to
more than 17.5 and 5.3 mm (0.69 and 0.21 inches) of storage respectively, or
37 percent and 12 percent of the runoff volume from the selected 5-year
return interval storm hyetograph respectively. Hence, if sites are available
for construction, satellite storage is more economical than using sewers
sized for a larger return interval storm for pollution control. Designing
sewers to provide flood control protection for less frequent return interval
storms on the basis of pollution control is even less justified. The in-
creased cost per hectare for a sewer system designed based on a 10-year
return period storm is more than $17,400 per hectare ($4.5 million per square
mile). The system would provide a potential storage volume for pollution
control of about 7.1 mm (.28 inches) of runoff or about 18,500 m (4.9
million gallons) of storage. Such design is not cost-effective for pollution
control because of the high cost of such storage. Hence, selection of a
5-year return interval storm design criteria can be justified for pollution
control if there is no other alternative for obtaining storage. If there are
alternatives for providing storage, a 2-year design storm for flood control
could be ample except in high risk areas where the potential for damage
justifies greater protection and investment. With protection based on
a design storm with a 2-year return interval, flooding of perhaps one-half
hour duration would be expected every five years and lesser flooding at more
frequent intervals.
125
-------
SECTION XIII
APPENDICES
Page
A. ALTERNATIVE MODEL EVALUATION 127
EQUATIONS OF MOTION 127
COMPARISON WITH MANNING'S EQUATION 131
COST OF SOLUTIONS 136
FINDINGS ..137
MODEL OPERATING CHARACTERISTICS 137
MODELS EVALUATED 138
Corps of Engineers' STORM 139
EPA - Storm Water Management Model (SWMM) 141
University of Illinois - Storm Sewer Simulation
Model 142
COMPARATIVE EVALUATION 142
THE RATIONAL METHOD 143
METHODOLOGY .145
B. MODIFICATION OF SWMM AND STORM PROGRAMS 147
SWMM 147
STORM 148
C. QUANTITY AND QUALITY CONSIDERATIONS, SWMM VS STORM.. 150
QUANTITY 150
QUALITY 151
D. MODEL INPUT DATA .154
LONG-TERM METEOROLOGICAL DATA 154
FIVE-YEAR SYNTHETIC STORM HYETOGRAPH 159
HYETOGRAPHS OF LESS INTENSE STORMS 161
DEFINITION OF DISCRETE ELEMENTS 165
HYDRAULIC PROPERTIES AND LAND USE .172
DISTRICT A DATA FOR STORM. .172
AREA AND LAND USE FOR OTHER DISTRICTS 184
DRY-WEATHER FLOW QUANTITY AND QUALITY 185
UNIT COST DATA ..192
Sewer Costs .192
Storage Costs 192
Pumping Station Costs 195
Treatment Costs 195
E. CALIBRATION OF STORM RUNOFF COEFFICIENTS ..197
126
-------
APPENDIX A
ALTERNATIVE MODEL EVALUATION
There are at least 24 models available for the assessment, planning',
design and control of sewer systems (20). These models compute time-vary-
ing runoff as opposed to steady state analysis. However, they differ in
scope and purpose and in the mathematical detail of routing flow through
sewers. In essence, the difference is in the approximations made to model
the complete momentum equation. The simplest models use Manning's equation
or similar formulae. Simplification of the momentum equation affects the
sizing of collection sewers and downstream interceptor, storage and treatment
facilities (17). As part of the study, Resource Analysis, Inc. (RAI),
Cambridge, Massachusetts made an evaluation of sewer design equations. In
addition, RAI evaluated the applicability of available stormwater management
models.
Stormwater simulation models use the governing equations of motion
- continuity and momentum - to numerically predict the flow in stormwater
systems. Fo.llowing is an analysis of the momentum equation, evaluating the
relative importance of the various terms in this equation. A qualitative
analysis of the effect of eliminating some of these terms was investigated,
focusing on the accuracy of the simulation predictions, the cost of unproved
accuracy, and the potential mis-design of the resulting structure.
EQUATIONS OF MOTION
The governing equations of motion for one-dimensional unsteady open
channel flow (21,22) are:
Continuity Equation:
A-
(a) (b) (c) (d)
(A-l)
Momentum Equation:
(e) (f) (g)
dx
-------
Where:
A s the cross-sectional area of the channel,
v s the mean flow velocity,
x = the downstream distance along the channel,
B - the surface width of the channel,
S = the friction slope, ,
S = the bed slope,
o
y = the water depth,
t = the time,
g s the gravitational constant, and
q = the lateral inflow.
Figure A-l presents a definition sketch for the equations of motion.
Figure A-l. Definition sketch for equations of motion
The assumptions made to arrive at these equations are:
1. The flow is one dimensional, i.e. uniform velocity distribution
128
-------
across each cross section, and the free surface is horizontal across
the section,
2. pressure is hydrostatic, i.e., vertical acceleration is neglected,
and
3. boundary friction and/or turbulence may be accounted for through
Manning's equation.
These equations can also be applied to cases of flow in partially full
pipes, with the proper definition of area and friction terms. The terms in
Equations A-l and A-2 can be described as:
(a) prism storage term due to variations in velocity with space,
(b) wedge storage term due to variations in cross-sectional area with
space,
(c) rate-of-rise term that describes the changes in storage due to
water surface elevation variations over time,
(d) lateral inflow term that represents the net inflow from lateral
sources,
(e) bottom slope term,
(f) friction slope term,
(g) water surface slope term (pressure force),
(h) convective acceleration due to spatial variation in velocity,
(i) local acceleration due to temporal variation in velocity, and
(j) term indicating momentum change due to lateral inflow.
Various mathematical models for flow routing have been proposed in
the literature. These models can be classified according to which terms of
the governing equations they include in their solution. Some general classi-
fications are (neglecting lateral inflow):
1. Storage Routing Models - use continuity equation only,
2. Kinematic Wave Models - use continuity equation, and friction (f)
and bottom slope (eX terms of momentum equation,
3. Diffusion Wave Models - also include pressure term (g) of momentum
equation,
4. Dynamic Wave models - use all terms in momentum equation.
129
-------
Only the Dynamic Wave Models can describe exactly all of the effects
present in a flow motion since these are the only models that retain all
of the terms in the momentum equation. For many practical situations,
some of these terms may be ignored resulting in a significant reduction in
the cost of simulation runs. An order of magnitude analysis of the terms
in the momentum equation can give some indication as to the importance
of each of these terms. A compilation of the range of typical values for
these terms is presented in TABLE A-l. These values were obtained from
Schaake (23), Henderson (21), Eagleson (24) and experience with these
models.
TABLE A-l. RANGE OF TYPICAL VALUES FOR EACH TERM IN THE MOMENTUM EQUATION
Term
Overland Flow
(ft/sec2)
Pipe & Gutter Flow
(ft/sec2)
Streamflow
(ft/sec2)
(e)
0 - 3.20
0 - 1.60
0 - 0.30
gSf (f) 0 - 3.20
3x
(g;
vl* (h)
ox
£ <«
* «)
0 - 0.10
0 - 0.01
0 - 0.01
0 - 0.001
0 - 1.60
0 - 0.60
0 - 0.30
0 - 0.30
0 - 0.05
0 - 0.30
0 - 0.005
0 - 0.002
0 - 0.0005
0 - 0.005
Note: 1 ft/sec2 = 0.305 m/sec2
The local acceleration (i) and lateral inflow (j) terms appear to
be relatively small for all cases. For overland flow and steep streams,
the water surface slope (g) and convective acceleration (h) terms also
appear to be of small magnitude when compared to the bottom (e) and friction
(f) slope terms. Henderson (21) has shown that for most streams and for
Froude numbers less than 1, the acceleration terms due to spatial (h) and
temporal variations (i) in velocity are much smaller than the acceleration
due to the water surface slopes (g). For Froude numbers higher than 1
130
-------
all three terms (g), (h) and (i) are of the same order of magnitude but
these cases only occur in mountain torrents where the bottom slopes (e) are
much larger. Flows with high Froude numbers may also occur in urban storm
drainage systems. But in these systems, the effects of control structures,
e.g., weirs, flow dividers, etc., produce more noticeable flow changes than
the additional terms (g), (h) and (i) in the momentum equation.
For flow in pipes, the water surface slope (g) and convective accelea-
tion Ch) terms may be large enough to be significant relative to the bottom
(e) and friction (f) slope terms. Terms (g) and (h) are the key to*modeling
the dynamic (e.g., backwater, subsidence) effects that are sometimes present
in pipe flows. The major difference between a pipe system and stream system
is that pipe systems are usually surrounded by highly urbanized catchments.
These catchments respond much more quickly and with higher lateral inflow
rates than non-urban catchments. Therefore, terms (g) through (j) are
relatively larger for pipes than for streams.
An order of magnitude analysis can only give an indication as to the
relative importance of the different terms. But it is unlikely that a
very small term in the momentum equation could have any significant effect on
the solution to the equations of motion. With this condition, only the
bottom (e) and friction (f) slopes appear significant for overland flow and
steep sloped streams. For mild sloped streams and for pipe flows, the water
surface slope (g) may be significant. In addition, for pipes the convective
acceleration term (h) may also be significant.
COMPARISON WITH MANNING'S EQUATION
If Manning's equation is used as the basis to establish a relationship
between depth of flow and discharge, the influence of the smaller terms
in the momentum equation is of interest. Manning's equation provides a
relationship among the flow velocity, v, depth of flow and shape of the
channel cross section as expressed through the hydraulic radius, R, and the
friction or energy grade line slope, Sf. The relationship is:
(A-3)
In which "n" is Manning's resistance coefficient. Equation A-3 may be
solved for the friction slope, Sf, to yield:
(l.49)2R
(A-4)
131
-------
which may then be substituted into the complete momentum Equation A-2.
The latter is solved for the flow velocity, v, giving:
1.49 ^2/5,
v = R '
a y v
>x g
'
g
(A-5)
If the flow is steady, (^=0), uniform, (y^=0), and there is no lateral
inflow, (q = 0), all but the first 'of the terms in parentheses disappears and
Equation A-5 is reduced to the familiar form of Manning's equation that
is:
1.49 D2/3C
v=—R S0
(A-6)
The significant difference between Equations A-3 and A-6 must be noted.
In Equation A-6 the term S , is a known geometric property of the channel,
hence a unique, known relation exists between velocity and hydraulic radius
(or, as more commonly expressed, between stage and discharge). But, in
Equation A-3 the slope, Sf, is in general not known, and, from Equation
A-5, the friction slope depends on the rates at which the depth and velocity
are changing.
Equations A-5 and A-6 may both be considered to be forms of Manning's
equation. However, the latter is a special case and thus limited in its
application. The effect of the extra terms in Equation A-5 is readily
demonstrated for the case of a prismatic channel of constant bottom slope and
having no lateral inflow. For this case, the two equations provide relations
between stage and discharge which may be plotted in Figure A-2.
ui
o
<
t-
V)
EQUATION A-5
EQUATION A-6
DISCHARGE, Q
Figure A-2. Uniform flow rating curve and loop rating curve
132
-------
Equation A-6 gives a single-valued relation, while Equation A-5 is multi-
valued according to whether the stage is rising or falling with time.
The latter describes a loop—type rating curve.
The relation between stage and discharge at a given channel section
depends on whether one is on the rising or falling limb of the hydrograph.
Although not shown in Figure A-2, the shape the loop rating curve is not
unique, but is directly dependent on the shape of the inflow hydrograph.
Three points on the loop rating curve, A, B, and C, are of particular
interest. At point A, the instantaneous gradient of depth with distance
is identically zero and Manning's equation in the form of Equation A-6
is applicable. However, neither the peak discharge nor the peak stage
occur at this point. The peak discharge occurs at point C prior to the
occurrence of the peak stage which is at point B. Thus, the peak stage and
discharge develop a phase shift which makes it difficult to determine the
peak stage from the peak discharge.
If by routing an accurate value for the peak discharge, Q , is deter-
mined, the peak stage at point B must still be determined. Substitution
of the peak discharge, Q , into Manning's Equation A-6 would yield a stage
at point D which is too tiigh. In this sense, use of Manning's equation will
in general lead to conservative predictions of peak stage.
This tendency to overestimate the peak stage is reinforced since the
peak discharge and peak stage do tend to subside with distance down the
channel. This effect would not be'predicted without the higher order terms of
the momentum equation. Hence, any procedure such as direct translation of
the inflow hydrograph to downstream locations or the kinematic routing model,
both of which fail to reduce the peak discharge, will overestimate the
discharge and thus add to overestimation of the peak stage.
The problem may also be reversed. Substitution of the peak stage,
YB, into Equation A-6 will yield a discharge, Q , which is clearly
smaller than the true peak, Q . Hence, Manning's equation can tend to
underestimate the peak discharge.
A similar situation occurs when one uses the Manning's equation to
estimate the discharge capacity of a structure. The geometric properties of
the structure together with the given bed slope, S , can be substituted
into Equation A-6 to yield the capacity of the structure. However, the
friction slope, S , observed in the neighborhood of such control is fre-
quently less than the bed slope, S , and the true discharge capacity as
yielded by Equation A-3 will be less than that obtained using the bed slope
directly. In this sense, use (or misuse) of the Manning's equation can lead
to overprediction of the discharge capacity of a structure.
In summary, the use of Manning's equation, without the other terms
in the momentum equation, is deficient in at least the following respects:
1. It fails to produce the loop-type rating relation.
133
-------
2. It fails to produce a reduction of the peak discharge.
The magnitude and importance of these effects depend on the particular
problem under study and may often be negligibly small. In some classes of
problems, the need to include these effects is obvious. For example, in long
natural channels with extensive flood plain storage or in river systems which
include ponding areas and reservoirs, these effects may be very pronounced.
On the other hand, urban drainage systems typically involve short, steep
channels with well defined boundaries; in these cases the indicated effects
may be secondary and are often negligible.
Simple criteria for determining when these effects may reasonably
be neglected are not readily available. Until these criteria are forth-
coming, considerable reliance will continue to be placed on the judgment and
experience of the design engineer. Hopefully, the existence of computer
models which allow retention of all terms in the momentum equation will
permit the development of sounder judgment and the eventual preparation of
precise criteria.
Other factors affect the accuracy of results from simulation models.
For example, inaccuracies in rainfall measurements and/or,the simplifications
used to model a prototype system within a limited set of model elements
(i.e., catchments, pipes, flow dividers, etc.) may introduce significant
errors. In fact, these errors will in general be much larger than those
resulting from neglecting some terms in the momentum equation. It is usually
more important to model the effects of a flow diversion structure accurately
than to model the subsidence of a flood wave in the input pipe exactly.
The preceding conclusions apply only to the case of open-channel flow
in which downstream conditions do not have a significant effect on the
upstream stages. This idealized situation frequently is not applicable. If
a channel is terminated by a control structure such as a weir, constriction,
culvert, etc., with discharge characteristics such that the stage in the
channel immediately upstream of the structure must build up to a level
greater than normal, a backwater profile will form for some distance upstream
where the stage will be above normal. The actual values of stage will exceed
those calculated by a simple application of Manning's equation with Sf =
S . Manning's equation will then yield stage values which are too low,
rather than too high as in the preceding discussion. The same type of
behavior can arise when anlyzing flows in sewer' systems. When the flow rates
and/or downstream boundary conditions are such that the sewer pipes flow full
and surcharinging occurs, the relationships defined by the normal flow
condition, Sf = S , are no longer applicable. The friction slope, Sf,
in these cases is related to differences in water level in the surcharged
sewers which, as in the backwater example, will be higher than the levels
given by Manning's equation with S_=S . The difficulty lies not with
Manning's equation per se. Rather, the difficulty stems from the inappro-
priate use of the assumption that the channel (or pipe) slope and the
friction slope are equal, i.e. Sf = S . Manning's equation is valid
where written in terms of an appropriate friction slope. However, the
134
-------
friction slope depends on the flow conditions. Thus, in general, the true
friction slope is a dependent variable of the problem and is not known in
advance. The assumption of S = S can be made only when the validity of
the assumption is known. If this assumption is indiscriminately built into a
sewer design program (for example), considerable caution should be used in
interpreting the model's results, especially if the results are to include
predictions of stage or piezometric grade line elevation.
Figure A-3 presents a surcharged simple pipe system consisting of
three links and manholes at each of the two nodes.
M.H*2 M.H.*
!>
v
El. Z
3
V
El. Z
2
LINK 3
LINK 2
El. Z.
LINK I
Figure A-3. Surcharged simple pipe system
The inflow to the system is a discharge from upstream sources and is de-
signated as QL. The outflow, QQ, is into a reservoir whose free surface
is at elevation Z . The instantaneous water levels in the manholes are at
elevations Z2 and Z^ as indicated. If the reservoir elevation, Z , is
known as a function of time (e.g., constant, as in the case of a large
reservoir), a mathematical relationship can be derived which expresses the
manhole water surface elevations as a function of the inflow, Q . A
particularly simple form of the relationship is obtained if a linear relation
is assumed between discharge and head and the inertial effects of the water
columns can be neglected. In this simplest representation the relationship
between Q and Z« takes the form:
d2Z3
-
dZ3
(A-7)
135
-------
in which the coefficients, a,, a« and a^, depend on the geometry
and hydraulic resistance of the system. The left-hand side of the equation
is analogous to a simple spring-mass-dashpot system. The right-hand side is
a function of the boundary conditions which serve as a forcing function.
One property of the solution of this equation is a phase shift between
the forcing function and the response,, Z^. Among other things this implies
that in general the peak inflow, (Q1>max.' and the Peak water level in
the manhole, (Z_) , do not occur at the same time. Thus, it is not
possible to relate "th'e' peak flow and stage (i.e., manhole water level) in a
simple manner. This behavior is analogous to that attributed earlier to the
effects of channel storage in the open channel case, and again points out the
problem of any routing procedure which fails to take these storage effects
into account.
COST OF SOLUTIONS
Cost reductions for simulation runs may be attained by neglecting
terms from the momentum equation. These reductions in cost are indirectly a
result of solving a simpler set of equations. Such an action would allow the
models to:
1. use larger At and Ax grids in the finite different formulations
or
2. step-down to a less numerically complex solution of the equations
(e.g., from the method of characteristics to a finite difference
scheme).
Any of these two steps would result in significant cost reductions in simula-
tion models.
From a qualitative point of view, some estimates of the effect of
each term can be made. Schaake (23) has shown that neglecting the temporal
variation in velocity term (i) can reduce the cost of simulation runs by at
least a factor of 2. This is a result of the larger time steps that can be
used for the finite difference schemes when Av/At is not in the equations.
A similar reduction in cost can be obtained when the spatial variation
in velocity term (h) is eliminated. In this case larger Ax.steps can be
used. Neglecting the water surface slope term (g') may not have as large an
effect on the cost of solutions as the previous two terms. But this term can
cause stability problems. In some cases, neglecting it may allow for much
larger At and Ax grids and therefore result in significant cost reductions.
The final term (j), the acceleration imparted on the lateral inflow, would
have no significant effect on the cost of simulations since it only involves
one additional operation.
In summary, if the variation in velocity terms (h) and (i) could be
neglected, significant reductions in simulation costs could be obtained.
No detailed comparison of cost reductions versus number of terms
136
-------
modeled for the small errors introduced by using a simple model such as the
Kinematic Wave Model.
6. In cases where backwater effects or surcharging of pipes is present,
it is inappropriate to use the assumption that the channel (or pipe)
slope and the friction slope are equal, i.e., S = S .
Manning's equation can only be used in these cases if the appro-
priate friction slope is computed. But in these cases the friction
slope is a dependent variable and is not known in advance.
MODEL OPERATING CHARACTERISTICS
The desired operating characteristics of the chosen simulation model
(or models) are:
1. Ability to simulate •
(a) the surface runoff quantity,
137
-------
(b) the operation of the trunk sewers, and
(c) any storage (in-pipe or satellite).
2. Ability to run on continuous records to check the overflow charac-
teristics of the system over a period of record.
3. Ability to simulate quantity based on physical parameters.
4. Capability to be transferable from one area to another within a city
and its environs without extensive calibration exercises.
5. Ability to predict the quality of surface runoff during wet-weather
conditions.
6. Capability of handling the sanitary and dry-weather quality aspects
of the system.
7. Ability to handle the pollutant transport in the trunk system on
a conservative (i.e., no time decay) basis due to the small size of
the drainage systems and hence short in-system time.
8. Ability to simulate directly the quality of overflows.
In the above synopsis "quality" has been used in the generic sense. For
urban purposes it appears that the following constituents are of importance.
a) Biochemical oxygen demand (BOD)
b) Chemical oxygen demand (COD)
c) Suspended solids (SS)
d) Settleable solids
e) Nitrogen/nitrates
f) Phosphates
g) Grease
h) Coliforms
i) Heavy metals
The ability to simulate the pick-up, transport, etc. of the above pollutants
is relatively crude at present. In particular, little or no capability to
simulate heavy metal pollutants exists. However, the models discussed below
are evaluated on their ability to handle pollutants in terms of the "ideal"
set presented here.
MODELS EVALUATED
Following an initial survey, attention was focused on three models for
a detailed review. These models were:
Corps of Engineers Storage, Treatment, Overflow, Runoff Model (STORM)
138
-------
Environmental Protection Agency Storm Water Management Model (SWMM)
University of Illinois Storm Sewer Simulation Model
The thorough analysis of urban models by Brandstetter (25) has been used as
a, general reference in much of the discussion on models which follows.
Recently, the same author prepared a more comprehensive comparison of urban
runoff models (26). The following sections review each of the above models,
and recommend that both STORM and SWMM be used to provide adequate simulation
capability.
Corps of Engineers'" STORM
STORM was developed in early 1974 by Water Resources Engineers, Inc.
(WRE) of Walnut Creek, California under contract to the Hydrologic Engineer-
ing Center (HEC) of the U.S. Army Corps of Engineers. The program is an
outgrowth of work performed by WRE for the Environmental Protection Agency
and the City of San Francisco. The following comments on the program are
based primarily on the User's Manual (27) and discussion with WRE and HEC
personnel. The program is intended primarily for the evaluation of storm-
water storage and treatment capacity required to reduce overflows to the
receiving waters to acceptable (defined externally) levels. The model
recognizes that the intense short-duration storms, so often used as "design"
storms, may well be completely contained through natural and artificial
storage mechanisms so that no untreated overflows occur. Alternatively, a
series of moderately sized storms may load the system to the point where
untreated releases occur. The program is thus designed to account for the
characteristics of the area. Lumped storage and treatment capacities are
also considered. The model considers the interaction of eight variables in
determining the operation of the system:
1. Precipitation
2. Air temperature (for snowmelt computations)
3. Runoff
4. Pollutant accumulation on the land surface
5. Land surface erosion
6. Treatment rate
7. Storage
— tj
8. Overflows from the storage/treatment systems
The model operates on land uses, including single family residential, multi-
ple family residential, commercial, industrial, parks and undeveloped areas.
The program is designed to simulate hourly stormwater runoff and quality for
a single catchment. It is a continuous simulation model with the capability
of simulating a number of years of record but can also be used for selected
single events. The model, as prepared by WRE, does not simulate dry-weather
flow and its associated quality components. Modifications to include dry-
weather flow simulation and up to 20 land uses have been made in the latest
version of STORM (28).
139
-------
The runoff from the catchment is computed using runoff coefficients and
hourly precipitation data for a single rain gauge in the basin. Depression
storage is simulated as the prime mechanism of reducing rainfall to effective
rainfall. This depression storage is allowed to recover at a constant rate
to account for evapotranspiration. The runoff from the area is computed
by applying a simple runoff coefficient to the effective rainfall. This
coefficient is a weighted average of the coefficients for pervious and
impervious areas and does represent losses due to infiltration. A similar
coefficient is used to simulate runoff during periods of no precipitation.
Runoff from snowmelt is computed using a degree-day technique.
Six water quality constituents are computed for different land uses.
These constituents are:
1.
2.
3.
4.
5.
6.
Suspended solids
Settleable solids
Biochemical oxygen demand (BOD)
Nitrogen (N)
Orthophosphates (PO,)
Coliforms
These quality components are computed from non-linear functions considering
the daily rate of dust and dirt accumulation, the percent of each pollutant
contained in the dust and dirt, street sweeping practices and days between
runoff events. Erosion is computed using the universal soil loss equation.
The BOD, N and PO, runoff rates depend on the pickup rates of the suspended
and settleable solids.
The model is very much a "black-box", type model. It does not route
the runoff quantity or quality in a sewer or channel network. The routing
aspects have to be reflected in the runoff coefficients. The computations of
the treatment, storage and overflow processes at the single outflow from
the system are performed by volume and pollutant mass balance only. A very
simple logic is employed as follows: if the hourly runoff volume exceeds the
treatment capacity, the excess runoff is diverted to storage; once the
storage capacity is exceeded, the excess runoff becomes untreated overflow.
During dry-weather conditions, the treatment capacity is used to draw down
the volume in the storage element. The quality of the runoff is not modified
in storage. Simple plug flow routing is used. Similarly, the simulation of
the treatment facility does not include the quality improvement to be ex-
pected.
The model is completely based on the use of empirical coefficients
for both stormwater runoff and quality computations. These coefficients
were provided by the user for the runoff case, and are internal to the
program in the quality aspects. Verification of the various equations
appears to be limited to date, and the dependence of the runoff and quality
coefficients on the catchment characteristics does not appear to have been
fully established.
The model is useful primarily as a planning tool to estimate the approx-
imate number and magnitude of overflows for various combinations of storage
140
-------
and
treatment. It is a useful screening tool in these aspects.
EPA - Storm Water Management Model (SWMM) (6,29)
This is probably the most comprehensive mathematical model for the
simulation of storm and combined sewer systems. It computes runoff quantity
and quality from the watershed by using a number of overland flow elements
to simulate the initial collection process, and a converging branch sewer
network to simulate the trunk sewer system. It can also handle up to three
storage basins and one treatment plant, together with the appropriate flow
diversion structures.
The model, as initially developed and used in this work, was primarily
an event simulator - its complexity and cost effectively limiting its
use to a single storm at a time. However, the latest version has incorpor-
ated continuous simulation capability. Spatial rainfall variations over a
basin can be taken into effect since independent inputs can be applied to
each subcatchment. The runoff from the overland flow areas is simulated
using the kinematic wave equations. Depression storage is accounted for,
and infiltration from pervious areas is computed using Horton's equation.
The stormwater quality is computed using non-linear functions which
are a function of land uses, street cleaning practices, pollutant accumula-
tion and rate of runoff. The pollutants considered are of a wider range than
considered in STORM and include suspended and settleable solids, BOD, COD,
coliform bacteria, nitrogen, orthophosphates, oil and grease. • Daily and
diurnal variations in the dry-weather flow and quality are accounted for and
are estimated from land use characteristics or directly from input data.
The flow in the sewer network, as included initially in the model and
used in this work, was simulated using the kinematic wave equations. Com-
plete continuity and momentum equations are included in the current model
version (30). The elements in the network can include manholes and conduits
from the selection of 12 internally stored geometric shapes, as well as three
additional arbitrary shapes. Other elements in the network are diversion
structures and up to two satellite storage facilities. Pumping stations with
constant pumping rates can be included. Surcharging can be handled by assum-
ing all flow in excess of full pipe flow is stored in the next upstream
manhole. The model also allows the option of letting the program size the
sewer to eliminate such surcharging.
Quality routing through the trunk sewer system is primarily based
on an approach of pure mixing in each pipe element. Decay is allowed for
the BOD, and scour and deposition of suspended solids is simulated. Other
quality reactions and interactions are not considered.
Unlike STORM, SWMM can simulate the operation of up to nine unit treat-
ment processes which can be inserted in series for a single overflow treat-
ment facility. It thus allows more accurate estimation of the pollutant
mass loads on the receiving waters.
141
-------
The model also has the ability to simulate both quantity and quality
behavior in the receiving water system.
The model has been applied to a large number of watersheds. In general,
it has been found that the accuracy of the runoff quantity computations
for basins ranging from 4—2000 hectares (10-5000 acres) was relatively good.
If adequate segmentation rules are applied, the model can be successfully
used for runoff prediction in most urban areas. On the other hand, as in
similar models, significant doubts still exist as to the adequacy of the
runoff quality prediction mechanisms. It is not yet known how generally
applicable are the regression equations relating runoff quality to land use.
Although they can be calibrated for individual basins, they can hot yet be
used with complete confidence for prediction purposes.
University of Illinois — Storm Sewer System Simulation Model (7)
"This model basically computes nonsteady flows in a converging sewerage
network using a solution of the dynamic wave equations. The routing is based
on a first-order explicit finite difference solution to the characteristic
equations of the dynamic wave equation. It can consequently consider up-
stream and downstream flow controls, backwater effects and flow reversals.
It does not, however, include, surcharging and pressure flows. A certain
number of controls, diversions and pumps can be included but only at the
outlet from the system. The model does not include any quality effects.
Similarly, overland runoff from precipitation is not computed. Rather, the
required .flows must be entered as separate hydrographs. Dry-weather flows
must be handled in a similar fashion.
The model includes a comprehensive formulation of open channel flow
routing in circular sewers. It includes a feature for the sizing of circular
pipes to accommodate peak flows. However, its lack of overland runoff and
quality simulation capabilities limits its usefulness for the present study.
In a recent study by Chow and Yen (31) for EPA, these capabilities have
been incorporated into the model.
COMPARATIVE EVALUATION
The work under this project mandated use of a model or models which
would be capable of being used in:
a) Planning of required facility locations and sizes, and
b) Evaluating alternative designs.
In addition, the models must be readily transferable from basin to basin
within the urban area.
In light of the above discussion it appeared:
1. The University of Illinois model was of no advantage.
142
-------
2. STORM included the storage/treatment capabilities required and, in
particular, allowed comparison of a wider range of treatment/storage
alternatives than could be economically considered using SWMM. It
suffers from calibration problems.
3. SWMM was the most suitable model for the design aspects of the
current study. Its RUNOFF and TRANSPORT Blocks have the required
capability to simulate the stormwater quantities involved. Addi-
tionally, it considered a wide range of pollutants. However, the
pollutant pick-up simulation mechanism must still be regarded as
suspect and subject to calibration.
THE RATIONAL METHOD (32)
This method has been widely used for the determination of peak rates
of stormwater runoff. It relates peak runoff rate to rainfall intensity
by the formula,
Q = Cia
(A-8)
where Q, the peak flow, is dependent upon rainfall intensity, i, and a
runoff coefficient C which attempts to weight all the variables affecting
runoff rates, and the tributary area, a. When applying the Rational Method
the intensity is the average rainfall intensity for a storm having a duration
equal to'the longest time of travel in the system being designed (overland
+ pipe) and the runoff coefficient is a composite of published coefficients
as weighted by land use characteristics. In this study, the maximum and
minimum values of the composite runoff coefficient were obtained using the
runoff coefficients published on page 51 of Reference 32.
To compare runoff rates determined by the Rational Method with those
using SWMM, flows, developed using SWMM Model and the synthetic, intermediate
pattern, hyetograph shown in Figure D-5, were used in the Rational Method
formula. A corresponding runoff coefficient was obtained using the rainfall
intensity determined from the appropriate intensity-duration frequency curve
with the estimated time of concentration. The time of concentration includes
the suggested inlet time of five minutes (32) and time of travel in trunk
sewers, which is estimated from SWMM output assuming pipes are flowing full.
TABLE A-2 summarizes the SWMM "C" values calculated at various points in
District A and compares the peak 5-year synthetic storm flows obtained using
the Rational Method with peak flows using SWMM. The Rational Method results
in runoff rates from 30 to 60 percent lower than runoff rates determined
using SWMM.
143
-------
TABLE A-2 PEAK FLOW COMPARISON BETWEEN RATIONAL METHOD AND SWMM
Area
(acres)
353
366
599
655
SWMM
Flow
(cfs)
898
909
1370
1406
Intensity
(in/hr)
3.5
3.4
2.7
2.5
SWMM
npii
.726
.730
.847
.859
Rational
"C" Range
.507-. 311
.519-. 321
.549-. 356
.562-. 372
Time of
Concentration
(Minutes)
15.2
16.0
22. '5
25.2
Rational
Flow Range
626-384
646-399
887-576
920-609
(cfs)
Note: 1 in/hr - 25.4 mm/hr; 1 acre
1 cfs = 0.283 m3/sec
0.405 hectares;
Assuming the "C" values calculated using SWMM flows are correct, the
peak rainfall intensities would have to be reduced to obtain the peak runoff
rates generated using the Rational Method as follows:
Rational Method
Flow Range (cfs)
Rainfall
Intensity
Range (in/hr)
626
646
887
920
384
399
576
609
2.44
2.41
1.75
1.64
1.50
1.50
1.14
1.08
Time of
Concentration
(minutes)
15.2
16.0
22.5
25.2
Extrapolating the above, using the appropriate intensity-duration-
frequency curves indicates that use of the Rational formula results in.
protection from storms having a return frequency of about two years if
maximum "C" values are used, to eight months if minimum "C" values are used.
Hence, drainage systems designed by the Rational Method may not provide the
anticipated degree of protection.
The magnitude by which the pipes are undersized can be estimated since
the pipe diameter for a given rainfall intensity is directly related to-
the 3/8 power of the associated "C" if the slopes are maintained equal.
The following summary indicates the actual pipe diameter changes:
Rational
"C" Range
.507-.311
.519-.321
.549-. 356
.562-.372
SWMM
Diameter (ft)
8.0
8.0
11.5
11.5
Rational Diameter
Range (ft)
7.0 - 5.8
7.0 - 5.9
9.8 - 8.2
9.8 - 8.4
This demonstrates a pipe diameter underestimation of from 12.5 to 29 percent.
This undersizing of pipes can result in reduced capital costs but
144
-------
offers less protection against flooding. The following summary of typical
costs for sewers installed in 18 to 20 foot trenches illustrates this appa-
rent savings. The diameters are based on flows as calculated above and the
costs are estimated using TABLE D-17.
SWMM
Diameter (ft)
8.0
8.0
11.5
11.5
SWMM
$/LF
420.
420.
800.
800.
Rational
Diameter (ft)
Rational
$/LF
Percent
Savings
7.0 •
7.0 -
10.0.-
10.0 •
6.0
6.0
8.5
8.5
350
350
612
612
298
298
452
452
17
17
3
3
29
29
44
44
The inclination to use-this method can be great since initial costs may
be up to 44 percent lower. But/use of the Rational Method to design storm
sewers proposed to provide protection for the 5-year storm can result in'
flooding with storms having as short a return frequency as eight months.
As a further indication of the underestimation of flows generated
by the Rational Method, reference is drawn to a drainage study in Louisville,
Kentucky by the U.S. Army Corps of Engineers in 1949 (33). In this study
report, a series of pumping station inflows were calculated, using both the
Rational and the Unit Hydrograph approaches. The results of this comparison
are summarized below:
Pumping Station
Seventeenth St.
Shawnee Park
State Fair Grounds
Lower Paddy Run
Rational
Flow (cfs)
99.9
939
1700
1495
Unit Hydrograph
Flow (cfs)
126
1359
2520
2630
The Rational flows are lower by 20 to 43 percent and the results are
comparable to that predicted using the Elizabeth data.
METHODOLOGY
SWMM, as used in this work represents the state-of-the-art, hydrauli-
cally and hydrologically sophisticated, single storm event simulator. It
uses a "micro" approach in that the rainfall-runoff process is divided into
its basic elements. It simulates the process, spatially and temporally, and
routes storm runoff and combined sewage quantity and quality from individual
catchments and subcatchments through a sewer network, interceptor system,
storage and treatment facilities, and finally \to receiving waters. SWMM
analyzes or designs a sewer system for an actual or synthetic storm event.
It requires detailed delineation of the subcatchments and sewer layout and is
an expensive procedure.
STORM is a continuous simulation model for overall planning. It ana-
lyzes hourly runoff quantity and quality for long-term hourly precipitation
145
-------
records, based on such parameters as percent imperviousness and land use and
evaluates the effectiveness of storage and interceptor/treatment facilities
for overflow pollution control. Typical to all continuous simulation modelsj
STORM uses a "macro" approach by combining and averaging the effects of the
basic elements. That is, rainfall-runoff processes are simulated for areas
of finite size and for coarse time intervals. It emphasizes the gross
pollution effect (total pounds) of pollutants discharged in a given storm
rather than the detailed time and spatial characteristics. To operate 0n
precipitation data over long time periods, the model contains a feedback
mechanism which does continuous accounting of water inputs (rainfall and/or
snowmelt) to and losses (depression storage and infiltration) from the
drainage basin. In dealing with' long-term real rainfall events, STORM was a
necessary tool in the project study.
The methodology used follows:
1. SWMM developed the alternative designs in District A for runoff from
a synthetic, intermediate pattern, 5-year return interval hyetograph
shown in Figure D-5. Quantity and quality characteristics, with
respect to combined sewage and storm runoff, were also determined
for smaller rainfalls.
2. STORM evaluated the effectiveness of storage and interceptor/
treatment facilities for storm runoff and combined sewage overflow
pollution control using continuous real rainfall events.
3. Calibration of the quantity parameters contained in STORM used SWMM
data generated from the detailed simulation of District A.
4. Quantity and quality and cost data, obtained from District A, were
projected to the entire City for the evaluation of sewer system
alternatives for both flood and pollution control.
146
-------
APPENDIX B
MODIFICATION OF SWMM AND STORM PROGRAMS
In addition to continuously updating the models as revisions became
available, both models were modified to meet the project requirements. The
modifications made to SWMM and STORM and the necessity of making them are
described.
SWMM
Both the RUNOFF and TRANSPORT Blocks in SWMM consider flow routing in
sewer pipes. The RUNOFF Block employs a simple storage routing scheme using
the continuity equation of flow in the gutters/pipes and Manning's equation.
This technique is sufficient as the RUNOFF Block considers runoff in overland
and gutter/pipes where sewer sizes are generally small and slopes are rela-
tively steep. The RUNOFF Block can handle up to 200 pipe elements.
The sewer routing in the TRANSPORT Block uses a more complicated' tech-
nique of a modified kinematic wave approach in accordance with 'Manning's
and continuity equations. Backwater effects within individual sewer segments
are considered. The TRANSPORT Block can handle up to a total of 160 sewer
elements (conduit and non-conduit), of which about half are non-conduit
elements. The TRANSPORT Block uses better routing techniques and has a sewer
design option that the RUNOFF Block lacks. In the design version, small
diameters are first assumed for the sewers that would ensure full pipe flow.
The size of the sewer is then increased and the computation is repeated until
free-surface flow occurs. In the previously available RUNOFF Block, pipes
are permitted to surcharge when full. The surcharged volume is assumed to be
temporarily stored at the upstream junction until sewer capacity becomes
available to drain it. Use of the RUNOFF Block for design requires trial
assumption of sewer sizes until stored surcharges are eliminated.
Because the SWMM RUNOFF Block (a) can handle more pipe elements than
the TRANSPORT Block; and (b) has an acceptable flow routing routine, it
appeared advantageous to use the RUNOFF Block for design of upstream lateral
sewers. Consequently, SWMM was modified to allow design capability with pipes
surcharged. As the capacity of the initially assumed pipe size is exceeded,
the pipe size is automatically increased to the next larger commercial size
until free-surface flow is attained. When a pipe is resized, the runoff
computations are repeated for the new (larger) conduit for the current time
step only. The runoff hydrograph up to the time when surcharging would have
occurred in the particular pipe is not recomputed since this would require a
147
-------
basic change of the internal computational sequence. The design version of
the TRANSPORT Block also does not recompute the previously computed hydro-
graph and pollutograph. The subroutines involved in the program changes
include RHYDRO, and GUTTER.
STORM
The available version of STORM was developed for simulation of urban
stormwater runoff quantity and quality. It did not model dry-weather
flow and consequently could not be used for simulation of combined sewage
overflows. Combined sewer overflow problems and their extent, and corrective
measures, and their effectiveness, were an essential part of the project
study. Hence, STORM was modified in this study to add the dry-weather
flow simulation capability.
Dry-weather flow input data provides for diurnal variation of quantity
and quality for various land uses. Dry-weather flow can include domestic,
commercial and industrial wastes and infiltration. Four options are provided
for computation of average daily dry-weather flpw quantity:
1. inputs of the total dry weather flow,
* 2. inputs of flow for each land use separately,
3. inputs of the coefficients to be applied to populations or
areas to estimate each component of the dry—weather flow.
Domestic flow is assumed to be linearly proportional to popu-
lation. Other component flows are linearly proportional to the
area, and
4. use of default values provided in the program.
The daily flow may also be varied for weekdays, Saturdays, Sundays
and Holidays and for hourly flow during the day by input of user-specified
ratios or by default values in the program. This option allows computation
of the diurnal flow variations.
For computation of dry-weather flow quality, coliform counts were added
to the five pollutants considered in the available STORM program. Compu-
tation of coliform organisms washoff is assumed similar to the computation
for other pollutants except that no contribution is assumed from the sus-
pended and settleable solids. The default values for the coliform counts in
dust and dirt are programmed. These default values, like those for other
pollutants, are used to compute the amount of pollutant washout in relation
to the amount of dust and dirt washout in a storm event. The options pro-
vided to compute average daily dry-weather flow pollutant loading for all six
constituents are similar to that provided for the dry-weather flow quantity
computation. Variations of pollutant load for the day of the week and hour
of the day are also incorporated.
148
-------
Changes to the STORM program include (a) the addition of Subroutine
DWFLOW which carries out most of the dry-weather flow computation, (b)
substantial modification' of the MAIN program and Subroutine OUTPUT which
reads input data and prints out computed results respectively, and (c) minor
modifications to Subroutines DEFINE, DIRT, EVENT, and ZERO. A copy of the
program changes was forwarded to the Hydrologic Engineering Center, of the
Army Corps of Engineers and was incorporated in the new version of STORM
released in July, 1976 (28).
In the original version of STORM, a storm event is.defined by the
properties of the individual rainfall events, such as duration and. intensity,
storm spacing, and time to empty any storage facilities. An event was
defined as beginning when storage was first required and continued until the
storage basin was emptied. Any number of overflows occurring during the
duration of the event was considered as the same overflow event. If a
rainfall event produces runoff that does not exceed the sewer capacity,
storage would not be utilized, and the rainfall would not register as an
event. In the case of zero storage, runoff from watershed would directly
drain to the interceptor sewer, and the number of events as defined above
would be the same as the number of overflow events.
With the inclusion of dry-weather flows, the treatment or interceptor
capacity should be at least adequate to handle peak dry-weather flows. Flows
exceeding this value would require additional control measures.
149
-------
APPENDIX C
QUANTITY AND QUALITY CONSIDERATIONS, SWMM VS. STORM
Difference between SWMM and STORM regarding storm runoff quantity
and quality are worth noting. Recognition of these differences is necessary
for interpretation and interchange of results generated by the two models.
QUANTITY
STORM accounts for losses of rainfall and/or snowmelt volume due to
infiltration in the watershed by the use of a runoff coefficient. The runoff
computation during a given hour is expressed as:
R = C;(P - f)
(C-l)
where R is runoff volume, P the rainfall volume, f the available depression
storage and C the composite runoff coefficient depending upon urban land
use. C -is derived from two basic coefficients, C, and C?. C, repre-
sents the runoff coefficient for pervious areas and C,, for impervious
areas. For a given watershed, knowing the area for each T.and use and the
associated percent of imperviousness, C can be calculated arithmetically from
C^ and C2. In the STORM Program, the default values of C^ and C^ are
respectively 0.15 and 0.90. These coefficients are a function of rainfall
characteristics (depth, duration, intensity) and antecedent dry conditions
and calibration is required. The parameter "f" represents the antecedent
conditions. There is no runoff from the watershed until depression storage,
which is uniformly applied to the entire watershed, is filled.
In SWMM, runoff volume over a short time interval, measured in minutes,
is determined by using a number of overland flow elements to simulate the
initial collection processes. The amount of runoff from pervious and imper-
vious areas is separately considered. Rainfall on certain impervious areas
results in immediate runoff without loss to depression storage. This is the
case in Elizabeth, since most dwellings have roof drains directly connected
to a street gutter.
SWMM uses a more detailed concept of the hydraulics of rainfall runoff
process than STORM. Parameters required can be reasonably estimated.
Consequently, in the absence of real time runoff data in the study area, data
generated by SWMM in District A were used to calibrate the runoff coeffi-
cients C, and C? in STORM.
150
-------
QUALITY
Neglecting land surface erosion and dry-weather flow, both SWMM and
STORM compute pollutant street washout in runoff according to the amount of
dust and dirt (DD) accumulated along the street curbs prior to occurrence of
a storm. From the total pounds of dust and dirt washout, the loadings of
pollutant components such as SS and BOD are computed either from the avail-
able local data or from specified default values. Where quality data is not
available for evaluating pollutant parameters, default values specified in
either program could be used. TABLE C-l shows the default values internally
specified in SWMM and STORM for computation of SS, BOD and coliform loading
from mass of dust and dirt in the street washout. Also shown in the table
are the rates of dust and dirt accumulation in each land use. These rates
were obtained from an APWA study in the Chicago urban area (34) and are used
in both programs as default values.
TABLE C-l. POLLUTANT DEFAULT VALUES FOR SWMM AND STORM
Land Use
Dust and Dirt (DD)*
lb/day/100 ft. of
curb
mg Pollutant/
gram of DD
SS BOD* Coliforms
SWMM STORM (MPN/gram)
Single Family
Multiple Family
Commercial
Industrial
Park
0.7
2.3
3.3
4.6
1.5
1000.
1000.
1000.
1000.
1000.
111.
80.
170.
67.
111.
5.0
3.6
7.7
3.0
5.0
1.3 x 10"
f.
2. 7 x 10°
1.7 x 10b
1.0 x 10b
0
* For both SWMM and STORM to compute soluble BOD
1 lb/day/100 ft = 1.486 kg/day/lOOm
The SWMM TRANSPORT Block only routes three pollutants; SS, BOD and
total coliforms. These are commonly used as indicators of effluent water
quality. In this study, only SS' and BOD were used.
SWMM and STORM use the same default values for pollutant loading calcu-
lations except for SS and BOD. SWMM assumes the quantity of SS as about ten
times greater than STORM. Both SWMM and STORM use the same values for
computing pounds of soluble BOD from pounds of dust and dirt. However in
the programmed calculations, SWMM adds five percent of the SS and STORM 10
percent to allow for the non-soluble BOD contribution. This results in BOD
loadings (or concentrations) computed using SWMM default values equal to
about five times that computed using STORM default values. The SS factors
used by STORM are based on calibrations from the Selby Street data in San
Francisco (35) rather than on physical analysis on dust and dirt. The SS
factors used by SWMM appear to be based on convenient choice.
151
-------
Test runs made by Huber (35) with Lancaster, Pennsylvania data, show
that use of the SWMM SS default.values and availability factor could produce
SS concentration levels corresponding roughly to that obtained using the
STORM default values. The availability factor varies with the rate of runoff
and measures the portion of the dust and dirt available in a given rainfall
event for the production of SS (6). Recently, Ganczarczyk (36) studied 99
hectares (245 acres) of catchment in Canada under developed conditions using
the STORM program and STORM default values. The predicted BOD was of the
same order of magnitude as reported in the literature for several cities both
in and out of the United States while predicted SS were underestimated.
As SWMM is an event simulator and STORM a tool for runoff simulation of
long-term precipitation records, the computations of street dust and dirt
accumulation with varying numbers of antecedent dry days and street sweeping
intervals are different. These differences are best illustrated by an
example. Figure C-l, developed for District A, compares the predicted SS
accumulation in streets using the SWMM and STORM programs with both SWMM and
STORM default values for an assumed seven-day street sweeping interval and a
75 percent sweeping efficiency. The daily SS accumulation in streets of
District A computed with the SWMM default values is 1,195 kg (2,638 pounds)
and with the STORM default values, 131 kg (290 pounds). The SS accumu-
lation computed in SWMM increases monotonically with the number of antecedent
dry days. Those calculated with STORM indicate periodical fluctuation of SS
accumulation at the street curb reflecting the effect of street cleaning
frequency and the number of dry days since the last street cleaning. In
SWMM, the additional accumulation of dust and dirt, consequently SS, on the
street curb is assumed to be equal to the maximum accumulation for the period
between successive street sweepings. It does not credit the cleaning effects
of street sweeping. Recognition of this difference is significant when
comparing runoff quality from a single storm event with two models.
In computing pounds of pollutants washed off streets in any given time
interval, both SWMM and STORM assume the washoff rate to be proportional to
pounds remaining on the ground and to the runoff rate. In SWMM, the rate is
the average from both pervious and impervious areas. STORM uses runoff from
impervious areas alone. For a given storm event, the runoff rate from an
impervious area is greater than the average rate from both the impervious and
pervious areas. A higher runoff rate means higher washoff and a greater
availability factor. Consequently, for the same storm event, pollutant
washout computed by STORM could be greater than that computed with SWMM.
Since STORM does not model sewers, it can not model solid deposition in
sewers during dry days or their resuspension during wet days. For this
reason, STORM can not simulate the "first flush" phenomenon in a combined
sewer system.
152
-------
28,-*
24
20
tn
Ml
O
-16
EC
h-
co
z
o
12
0
o
CO
CO
DRAINAGE AREA = 655 ACRES (DISTRICT A)
STREET SWEEPING INTERVAL = 7 DAYS
SWEEPING EFFICIENCY = 0.75
CURVE I - SWMM USING SWMM DEFAULT VALUES
CURVE 2 - STORM USING SWMM DEFAULT VALUES
CURVE 3 - SWMM USING STORM DEFAULT VALUES
CURVE 4 - STORM USING STORM DEFAULT VALUES
10 15 20
NUMBER OF DRY DAYS
25
30
Figure C-l. Street, SS accumulation, SWMM vs. STORM
153
-------
APPENDIX D
MODEL INPUT DATA
Input data to computer models utilized in the study includes (a) meteor-
ological data for long-term rainfall-runoff simulation, (b) a 5-year synthe-
tic storm hyetograph for runoff determination, (c) hyetographs of less
intense storms for planning a pollution control system, (d) the discrete
elements of the drainage system, (e) hydraulic properties and land use of
discrete elements, (f) area and land use of other drainage districts in the
urban area, and (g) unit cost data for sewers of varying size and depth and
other elements of the sewerage system.
LONG-TERM METEOROLOGICAL DATA
To use the STORM program to account for the rainfall-runoff character-
istics of the study area, the continuous record of the hourly precipitation
data at Newark International Airport over the 12-year period from 1963 to
1974 was used without modification. Rainfall was assumed to be uniform over
the seven-square mile study area.
Average daily air temperature data at Newark International Airport
over the same 12-year period was also used for snowfall and snowmelt compu-
tation. Precipitation occurring on the days with air temperatures at and
below freezing (°C or 32°F) were considered to be snowfall and no runoff
was assumed. Snow pack on the ground was assumed to start melting as
the recorded air temperature rose above freezing. TABLE D-l shows the
amount of annual precipitation, the number of annual precipitation events and
the estimated number of runoff events. A precipitation event is defined as
a period of continuous non-zero precipitation. Traces (rainfall or equiva-
lent water depth less than 0.25 mm/hour or 0.01 inches/hour are considered to
be non-zero precipitation. Precipitations interrupted by a period of one
hour or longer are separate events. To eliminate many inconsequential
precipitation events, those events with a total less than 0.76 mm (0.03
inches) are deleted and become part of the dry period. The estimated number
of runoff events is based on the depression storage capacity of 3.94 mm
(0.155 inches) computed for District A. No runoff occurs until the total
precipitation exceeds the depression storage capacity. The 12-year period
selected for evaluation of stormwater runoff and combined sewage overflows
included the severe drought years of the early 1960's and the very wet years
of the early 1970's. The analysis made with .the data collected during this
period, therefore, should be representative of the range to be expected in
future years.
154
-------
TABLE D-l. ANNUAL PRECIPITATION AND RUNOFF EVENTS
Year
Precipitation
(inches)
Number of
Precipitation
Events
Number of
Runoff
Events
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
29.48
33.45
26.02
37.86
44.09
36.89
40.74
34.38
49.78
48.30
46.29
39.86
96
118
108
94
132
89
132
113
130
135
125
111
Annual Average 38.89 115
Annual Average (1941-1970)=41.45 inches
Note: 1 inch = 25.4 mm
56
76
59
51
84
53
77
74
79
83
72
70
70
There are, on the average, about 115 annual precipitation events.
About 40 percent of these precipitation events have a total less than the
depression storage capacity for District A and would not result in runoff
using STORM. About 70 annual storm runoff events are estimated for
District A.
Figures D-l, D-2 and D-3 present a statistical analysis of the amount
and duration of each precipitation event, and the antecedent dry hours
for the 12—year period. Precipitation events can be expected on the average
of about every three days with an average duration of about 4.5 hours.
The median value of antecedent dry hours is 36, and precipitation duration,
three hours. Similar analysis was made for cumulative precipitations
exceeding 3.94 mm (0.155 inches) or greater than the assumed value of
depression storage. The average' interval between consecutive runoff events
is about six days with a median of four days. The average duration of each
runoff event is seven hours and the median duration is six hours. About 75
percent of the precipitation events with a cumulative rainfall of less than
3.94 mm (0.155 inches) were of. three hours or less duration, and about 90
percent were of four hours or less. During the 12-year period, the longest
duration rainfall was 34 hours (September, 1969), with a cumulative total of
126 mm (4.97 inches). The largest amount of precipitation was 143 mm (5.62
inches) over 15 hours recorded on August 29, 1971. In October, 1973, 26
consecutive dry days were recorded, the longest in the 12-year period.
Previous study (8) of rainfalls in the Elizabeth River drainage basin
indicated about 35 percent of the rainfalls were of the advanced pattern
155
-------
39ViN30H3d
a
I
a
o
•H
4-1
cd
4J
•H
&4
•H
O
CJ
S-l
g
•H
JJ
W
•H
•X)
O
0
0)
8*
I
Q
00
•H
S1N3A3 jo
156
-------
30VlN30d3d 3AllV~mi/\inO
3
o
(T
=>
Q
o
LJ
(T
Q_
^ o
in
a
o
•H
4-J
Cfl
S-l
a
o
•H
4-1
•H
ft
•H
a
a)
d
o
•H
4-1
4-1
CO
•H
O
a
0)
3
CN1
A
60
•H
S1N3A3 JO d39IAiriN
157
-------
3AiJL\nnwno
3
o
X!
a
a)
13
0)
o
-------
in which the most intense rainfalls occurred in the first quarter of the
storm period, about 45 percent were of the intermediate pattern in which
the peak rainfalls occurred in the middle half of the storm, and 20 percent
were of the delayed pattern in which peak rainfalls occurred in the last
quarter of the storm period. The 12-year data generally conforms to this
rainfall pattern. Peak runoff from an advanced pattern storm is less than
that from an intermediate or delayed pattern storm since the rainfall-
producing runoff, or rainfall excess at the time of the peak rate of pre-
cipitation is less due to depression storage and greater infiltration
losses. In this study, both the advanced and intermediate pattern storms are
considered in evaluating control of wet-weather overflows.
FIVE-YEAR SYNTHETIC STORM HYETOGRAPH
The sewer system was designed to accept runoff from an intermediate
pattern synthetic storm hyetograph. The hyetograph of the.synthetic storm
was developed using the intensity-duration curves shown in the U.S. Weather
Bureau Technical Paper No. 25 (37) for Sandy Hook, New Jersey. The inten-
sity-duration curves shown in Figure D-4 were considered to be applicable to
the study area. Twelve-year data at the Newark International Airport was
analyzed. The peak hourly rainfall with 5-year return interval was in
excellent agreement with that obtained from the Sandy Hook intensity-duration
curves. The rainfall intensity-duration curves at Sandy Hook were 'updated in
a more recent U.S. Weather Bureau publication, Technical Paper No. 40 (38).
Intensities obtained from this later publication were found to be about 10
percent greater than that obtained from the older Technical Paper No. 25.
The synthetic storm hyetograph was computed for the assumed duration
of three hours, with peak rainfall intensity assumed to occur at the begin-
ning of the middle third of the storm duration. This duration is long enough
to cover the time of travel from headwater to outlet in the largest district
in Elizabeth.
In developing the shape of the storm hyetograph, it was recognized
that real storm events can produce rainfall intensities equal to different
return frequencies during their history. In the storm of July 20, 1961
at the Newark International Airport Weather Station, the frequency approxi-
mated a 15-year return period storm at ten minutes duration, a. two-year
return period at one hour duration and less than this return period for
three hours duration. The measured data is shown in Figure D-4. Because
each reach of a sewer conveying urban drainage is tested by a different
duration storm for a given return frequency, the concept of an envelope
hyetograph has been developed and successfully applied. The intensity-dura-
tion relationships of this hyetograph matches the Weather Bureau intensity-
duration curves at all time intervals. This hyetograph can be used to
determine the design flows which test the entire system, and eliminates the
heed of using more than one hyetograph of varying durations in design.
Figure D-5 shows the 5-year synthetic storm hyetograph developed
for 2-minute intervals. The storm hyetograph was developed as follows.
The intensities were read from the intensity-duration curve for 5-year
159
-------
SYMBOLS REPRESENT MEASURED RAINFALL!
AT NEWARK AIRPORT ON JULY 20, 1961
RETURN PERIOD(YEARS)
1 1
1 1 1 1 — 1
34 5
DURATION (hours)
Figure D-4. Rainfall intensity-duration frequency curves
160
-------
a:
LJ
^
u
s
i
o: or
UJ LJ
Hl-
zz
CO
I
w
\
cr
tl
2|
Ss
O .
^^^
l^i
^5i
r
•
«J
n
I1'
CJ
<
cc
LU
H- r
z 1-
** ' f
«i 1 r
£o r1 J
1— -r 1 1.1
Sf j 1
X :
X '
X,
l r
i _j
r J
" / r
I
V
L
-i 7 1
U\J
2 1 j
it JW
*£" '/
_i<> Y^
r- CLir A|
^o
"5 °
c +-1
.- OJ
^
-g !JS g
S 0
p a
s
•H
CO
o ^
-CD ^
cd
CD
1
O
~<3" IT)
1
CU
3
60
•H
_0 P"<
^
\
00
CD
in
-O
^- TO
A1ISN31NI
161
-------
return frequency at durations which are multiples of two minutes. The
rainfall amounts for incremental duration were obtained by multiplying
the intensity and duration. The rainfall amount differential of a successive
duration represents a possible 2-minute interval rainfall in the hyeto-
graph. After all possible two-minute rainfalls have been obtained, they are
recorded in sequence beginning with the largest two-minute interval rain^-
fall. The most intense rainfall rate was placed at the beginning of the
second hour of the hyetograph, the second most intense rainfall at the next
later time interval and the third most intense rainfall at the next earlier
time interval. The remaining hyetograph was developed in a similar fashion.
From the intensity-duration relation used to develop the 5-year syn-
thetic storm hyetograph, the average 1-hour, 2-hour, and 3-hour rainfalls
have intensities of 40.6, 26.7 and 20.6 mm (1.6, 1.05 and 0.81 inches) per
hour respectively.
For input to calibrate the STORM program, which accepts hourly rainfall
data only, the 5-year synthetic storm hyetograph with hourly intervals
(Figure D-5) was used. The hourly intensities of the hyetograph are 12.7,
40.6 and 8.4 mm (0.50, 1.60 and 0.33 inches) per hour respectively. These
intensities were computed so as to conform to intensity-duration curve.
Storms of advanced pattern were considered in relation with the contain-
ment or interception of storm runoff for pollution control. The advanced
pattern storm was assumed to start with the most intense rainfall and then to
decrease monotonically as the storm progressed. The 5-year synthetic storm
of advanced pattern is shown in Figure D-5.
HYETOGRAPHS OF LESS INTENSE STORMS
The 5-year synthetic storm hyetograph was used for the design of
collection systems for flood relief. Optimum sizing of the pollution control
system components, such as interceptors, storage and treatment facilities,
are defined here by less intense but more frequent rainfalls. Consequently,
1—year and 1.3—month synthetic storms were developed and used as input
to SWMM to determine the character of runoff quantity and quality required
to develop pollution control strategy. The hourly rainfall intensities
of these 1-year and 1.3-month synthetic storms and storms of other return
intervals are shown in TABLE D-2.
The data were prepared from a partial duration series analysis of
hourly data for the 12-year period assuming hourly intense rainfalls were
independent of others regardless of whether they occurred in the same pre-
cipitation event or in a separate event. Monthly exceedance values of the
rainfall data were arranged according to magnitude. The values are plotted
using the Weilbull formula (39) on semilog paper with the rectangular scale
representing hourly rainfall intensities and the logarithmic scale repre-
senting the return interval in months. Figure D-6 shows such a plot. The
plotted values lie approximately on a straight line up to the return interval
162
-------
163
-------
of two years. To check the hourly rainfall of 40.6 mm (1.6, inches) of the
5_year return interval storm, an annual .exceedance series was prepared
and plotted on a Gumbel's extreme probability paper. The hourly rainfall
value corresponding to the 5-year return interval agreed well with 40.6 mm
(1.6 inches).
TABLE D-2. HOURLY RAINFALL VERSUS RETURN INTERVAL
Return Interval
6 - Month
3 - Month
1.3 - Month
1 - Month
Hourly Rainfall
(inches)
0.74
0.60
0.40
0.34
Note: 1 inch = 25.4 mm
The advanced and intermediate rainfall hyetographs of 1-year and
1.3-month storms were synthesized in the same manner as the 5-year synthetic
storm. The magnitude of these storm hyetographs were derived from the 5-year
synthetic storm hyetograph by multiplying the latter by the ratios of 0.58
(0.92/ 1.60) and 0.25 (0.40/1.60) (TABLE D-2) respectively for the 1-year and
1.3-month synthetic storms. TABLE D-3 summarizes the gross characteristics
of 5-year, 1-year and 1.3-month synthetic storms. The 1-year and 1.3-month
synthetic storms as developed represent one of many possible storm patterns.
However, they provide a consistent tool for the derivation of runoff quantity
and quality information for planning level decisions of pollution control
strategy..
TABLE D-3. SYNTHETIC STORM CHARACTERISTICS
Average Rainfall Intensity (in/hr)
Return Interval 1-hour 2 hour 3-hour
Total Rainfall
(inches)
5-Year
1-Year
1.3-Month
1.6
0.92
0.40
1.08
0.63
0.27
0.81
0.47
0.20
2.43
1.41
0.60
Note: 1 in/hr = 25.4 mm/hr
164
-------
DEFINITION OF DISCRETE ELEMENTS
For use of SWMM, District A was divided into five major subareas, each
having relatively homogeneous land use. These five subareas are shown j^n
Figure D-7. Subareas I and II are almost entirely single family residential
housing, which contain approximately 43 percent of impervious surface. About:
half of Subarea III consists of commercial and industrial development which
contains about 80 percent impervious area. Subarea IV includes a park with
about 19 percent impervious area, while Subarea V is generally high-density
residential with 50 percent impervious area. TABLE D-4 summarizes the land
use distribution for each subarea as well as for all of District A.
TABLE D-4. AREA OF LAND USE DISTRIBUTION IN EACH SUBAREA (ACRES)
Percent
Single Multiple Total Imper-
Subarea Family Family Commercial Industrial Parks Area viousness
I 175.73
II 164.59
III 9.18
IV 115.36
V 5.33
Total 470.19
-
11.60
43.03
17.76
46 .,83
119.22
-
0.33
27. 07
4.67
2.63
34.70
- - 175.73
- 176.52
18.0.2 - 97.29
13.03 150.83
54.79
18.02 13.03 655.16
43.0
43.5
63.2
42.9
50.8
46.8
Note: 1 acre = 0.405 hectares ,
The five subareas were further subdivided into 279 subcatchments
areas ranging from 0.08 to 4.05 hectares (0.2 to 10 acres) with an average
of 0.93 hectares (2.3 acres). Overland drainage was routed to the nearest
street. The surface runoff drained to 139 lateral sewers (analyzed w|th
the SWMM RUNOFF Block) and subsequently to 31 trunk sewers (analyzed with
the SWMM TRANSPORT Block). The downstream end of the sewer system in
District "A connects to the Westerly Interceptor. Flows exceeding the capa-
city of the interceptor discharge to the Elizabeth River. .
The schematics of the combined sewer layout for each subarea are shpwr)i
separately |_n Figures D-8 to D-12. The combined sewer system was extended
from the separate storm sewer system shown in Figure D-13. It includes a
pipe in every street but excludes drainage at the upstream end of each sewer
branch. The separation of sewers in District A was studied previously (J.1-)
in an effort to alleviate frequent street and cellar flooding. In the
separate system, overland flow was allowed to the maximum extent practical,
TABLE D-5 compares the total length of separate sanitary, separate storm and'
the combined systems.
165
-------
Figure D-7. Subareas of drainage District A
166
-------
LEGEND
JUNCTION MANHOLE
TRUNK INLET (ELEMENT NUMBERS ASSIGNED)
PIPE WJTH COMBINED FLOW (ELEMENT NUMBERS ASSIGNED)"
— PIPE WITH SANITARY FLOW ONLY
Figure D-8. Combined sewer system layout, subarea I
167
-------
SUBAREA HL
SEE FIGURE D-8 FOR LEGEND
341
Figure D-9. Combined sewer system layout, subarea II
168
-------
SUBAREA
• .201
346
SEE FIGURE D-8 FOR LEGEND
SUBAREA XZ
Figure D—10. Combined sewer system Xayout, subarea III
/
SUBAREA 33C
356
SEE FIGURE D-8 FOR LEGEND
Figure D—11. Combined sewer system layout, subarea V
169
-------
N
SUBAREA HI
r i\
f*\ 2?
>
\, .
A .
i I
\i 249
X
7 ^
,
253
252
250
248
247
226 j
228 „
_j
235
' '
J
246 T
245 |
225
230 (
229
1
234
244
243,
1 ^~^"^ 3
1 ' ^"^
224 ,c,
232 |222 35^
1
233
T obo
231 i
236 ^
* 356L
r>-z~?
. 242 ^ 241 _ 240 . 239 ^
1 T
"^352
8^354 uj
/J SUBAF
A
\
\
\
^
/
I , I
Figure D-12. Combined sewer system layout, subarea IV
170
-------
LEGEND
JUNCTION MANHOLE
PIPE ELEMENT-(ASSIGNED NUMBER SHOWN)
Figure D-13. Separate storm sewer system layout
171
-------
TABLE D-5. TOTAL SEWER LENGTH COMPARISON
System
Length (ft)
Separate Sanitary
Separate Storm
Combined
106,000
29,000
63,640*
* Length of sewers carrying combined
sewage only. The total length of
the combined sewer system, including
extensions at extremities, is
106,000 feet.
Note: 1 foot = 0.3048 meters
HYDRAULIC PROPERTIES AND LAND USE
Delineation of subcatchment boundaries were prepared from a street map
of scale 1:4,800 and field checked. Subcatchment areas were planimetered.
Data prepared for each subcatchment included catchment width, area, imper-
viousness, slope, overland resistance factors, surface storage and infiltra-
tion factors. The subcatchment data are shown in TABLE D-6.
Definition of pipe elements includes length, slope, tributary area,
upstream and downstream connecting pipes and roughness coefficients. TABLE
D-7 lists the pipe element and its corresponding slope, length and tributary
drainage area. The trunk sewer elements were assigned numbers starting
at 300 and the lateral sewer elements, numbers below 300.
Since the SWMM RUNOFF (modified) and TRANSPORT Blocks have sewer
design capability, the sewer diameters required as input to the program
were generally assigned as 0.457 m (18 inches), the minimum diameter used in
practice for combined sewers. The Manning "n" was assumed equal to 0.013 for
diameters less than 0.762 m (30 inches) and 0.011 for diameters equal to or
greater than 0.762 m (30 inches), respectively. The modified version of the
SWMM RUNOFF program has incorporated these changes of roughness coeffi-
cient with pipe sizes. Data on the land use and curb length in each sub-
catchment was prepared and is presented in TABLE D-8. Dry-weather flows
were entered at designated inlet manholes to trunk sewers.
DISTRICT A DATA FOR STORM
STORM is a "macro" model and is intended to consider a large drainage
basin. It processes the composited information of all subcatchments rather
172
-------
o«
OK>rt
oo ooooooooooooooaooooooooaooooooooaooooooooooooaoooooaoooooooooooooooooo
>->* C3GOOOOOOOOOOOOOOOC>C3OOOOGOOOOOOOOOOOOOOOOOOOOOO»OoaoOC»OC5OC>OOC»C>OOOaoC>OOO
OOOOOC3OOOC3OOOOOOOoOCiO0OOaOOC>OaOODGOOOC3OOOOOOOOC3OOOOOaOOOOOO
Ul
Q
^^r>aixcoa3Kccxx«xocx-cccx3Xcccca:cocccococcaDoDco=c.aCcco3coa:a;cocOcocoffi
oa:a~oj«<\jwc\jowc\iC\j(\jcvcMCOojojrjCNjojcsic\jojc\JcyojOwOjojcN^
•••••••«••••••••••••••••••••«•••••
>OOOOOCJOOOOOoOOOOOOOOOOOOOOOOoOOGO
ooooooooc?oooooooooQOooooooc>aoGooo
5OOOOe>OC>GOOOOGOOOOOOC>C>OeSooC>OOOOOOe>
••••••*••••••••*••••••••*•••••••••
*-< x:fOfOfOroKifOrOK>roroMK^Mrt««tortr'>rOKiiororoiorOK)K)KiiOtO^ioioK)Ki«ro«
OC1OOOOOOC3OOOOC3OOOOOOOC>OC>^OC3OOOC1UC3OOOOC^OOOOOC}OO^OOOOC>OOOC1OOO
>-i> intou^ifitnif>tnintnu^u^intr)intotntnu^intfiu^inininu^iAiriiou^tf^u^intr)inLOu^u^U)u^intr«i^
—a: (MwcycMCvc\)pjcyc^c\'ojoJcye^evwc\jcyMcgoic«ojc\icM«ojwc\jcjc^c>jcycjc^csi^
l*!UJ *••*••»§•••••#••• ••••*••••• *••••••••••••*••••••'•••••••••••• ••••••»•»«•
OO_ OOOC3OOOOCJC'OCiC»OOCtC3C>e>OC3OOC3^OO^^OOOOOC3OOOOOOOC>C>OOC3OO^^.OCDOC3^O^^^C'>O^C>O^OO^^
<
D;
o
h-
tfi •
*S > eNic\c-jOucvwc-.cve\jccc«(^e\jcvjC\)C^ojevc\,c^c^oJcjcjCsicwcv«-jcs,c\wcvjo:e\jc\jc\iw
K UU! <30OC»C3OOC^OOOOC>OOCiOOOCJOOOOe>OCDeSOOC3OCJCJOOC^>C5OC5OC3tr)*ac>OC^OCiOC5OC>O«3CJC>OCJC>C>OC>ClC3C>C>C>CJO
^ , OOCJOCJC5OCjOOOO^OOC''OO^OCJOOO^O^C>OC.TOOOOO
Pd S
o
rM or • oooc>ooc^c^ClOoc1c~oc>cc)c>c^ocJCJOc?c7C)oocJ)OOooc^c^oc:>ciC)c^o^c>oc^c;c^c^c^c>C7Oc^oc)coo
o u'imu i»")u"iir.ti'U'ir.iru'ifiirnr.u^ir'if)irinu.u'iirir,ir)tr>tr)tf inu~iinu't^ir>inu^uiu>t^u"iuj(^(\jf\jc\je\jrjCJwc^
2 <-"-! «•*«•• * • ...••*•*«•••••••«•* • •
C~* *Cl. CJOC>OC*OOC>C5Ot3OO.OOC>C3OOOOoOCJOC3OOOOC>C>C>eao^C>ClOOOOC3OC3C3OOOOOC3OOC>OC>CJOOC-OC'C>OC'>C>er>C3C3CDoOOOC(OCVCr>C*^C>^e3tiC^^".^&OC1'^C3dOC1>OJO^OO^O^CJO^C>OO t) CD CD CD
j~{ *•/>?" oof-rnriC'Oo >r >e'C>c3|—>c.«oc".oo^s^'csc «&cioc. oooc-oc^Ok^c'so <'Jc^r.'CiYOoc^c;'c3c. oooc^odoc.joc'iocic'. cac;oc>c~>oCJo
O a"
Si .
o **
. -~. UJh- CJOOQC3C>OOOOC.>C3C.>OO^OOe3OC>CiOCJOCJOC>C>^ClClC>C>OOCJOtSC3OC3C>OCJOOaocaC>OOC3e>C3C>C3OO^C>OC3OCJC>e>OC»O
P-f a U. (JO^JCiC>c.ClC>cac^oc^c^ocJ^^c^c^CJC>c>c^l^c^e^<^l^c^cs^t^c^caocJC*t^c^c^CJOc^c>CD^>^tD^^e^^c^o^oc>c•^ooc>^OClOC>o^oc)O
r/s _j^ oociooooocjOc'OCionooooooocjo^caoor^ooo^ouoocaooooo^ooc^^c^ooo^^^canoOf^a^oocsoooo^
UJ wu. ••••,«•••••••••••••••••••••••«••«••«•««••••••••«•••••••••••••••••••••••
•- OOC-OOOnOOOOOC'C.i.^C3C)OOC?C>OC.>OOOCiOOC3Cr.C>C-'C30eDr'JOC'C30C?C»oOOOGOOC>C>C>OOC'
O
I K •
(^ 2T> t.»OC'it3C>CJOC>C3C^(~,^C3C3C' C3 C'>e3CT*C"'O O<.>OOOC'C1>OOC>OOOOC3 O OC3 C»C'OOOOOC.iOC>OO OOCJCJC3 O OC.>O O OOOO C>CJOC>O
1—1 IJJ.T «»««»»«»««*«»»#«««»»««»o«««»»»»*»«»««»«»*#»«»«»»««««««*»««»***«««»«««»
uu: alf*u^a^asccK»roro^o^)fOtO«KJlo^^rol0^^fOlOKlr^fOlOfo^oK)«lOlO^^^OrOK)rO(o^OK)^^ro
l ""Ziiziiiiiizziizziiriziiiiiiiziiiziiii
H o<\jc'\imr^invDr*toa^r-4*oir)vn«^jo^rvjeMc^»o.^-TX)n^tv*''nooinc>vDX[nc^^^r^oin.jaoi^xui%J3^eo^*D^
OC>C>C'OC>OOC)C3OC^OOOOC3C>OOOC>OOC3OOOOe^oC300OOC>OCTOC3C3C>OO£-JC>c>OOOC>C)OC3tl>OCrfC3
K-^ ••o««*«««**««««««*»««***«««»«f*»«««*««««««««»«««**««0«««*«**««o«*«««4«
•H - *-< *H<-l*-4iH fH *-< fMCMi-t
h-
O-L.
u_i
f—J" o>*4-inK)tHrHa^coocoae\;oQ'^w*J)^^^yc\itNic\ic\icoa^5Ccoc3OccrOrOiOJ^*r*r^o^o»CNjoevjryrocw^^*
our
o
I
<_'O •^T^^^P^tHrHrH*H«HCV]CgCV]^C4^C\tC^CijC\JrOrOnrOK)K)tOK3^^«^^ir)inu^ininLr>vO\^
173
-------
•a
Q)
I
4-1
d
o
o
VO
I
*-<«* cocccoeoaicococccoeQcococococccoc£cOKcocooococoeooQGOa>oococottaococoaoK>iOMioKJMK5ioKJWfOfOK5rtioK>Kj(OrofOwtopoto»OK)K3K)toMK)^
vi oooooooooQoooooooooooooooooooooooooaooooaooooooooooooooooooooooooooooo
>*X. OOOOOOOOOOOOOOOOOOOOOOOOOOOOOGOOOODOOOOOOOOOOOOOOOOOOOOOOOOOOOOaOOOOOO
<>-* • ••••••••••••••••••••••••••••^•••••••••••••••••••••••(••••••••••••••••v
3oOOO^OOOOC3OC9^O^^C30CJOC3OCr)OOOOOC>O^OOO^OOOOO^OOOOOOOOOC3OC3OOUOO^OOOOOOOO^OO
U.«rX*OOOCJOOOOO^OOC>OO^^OC3C3C>OOOOOOOOOOOOOOC9OOOOOOOC3OOOOOOOOOOOOOOOOOOOOOOOOOO
ZCC*-i •••••••••••••••••••••••••••••o««*««««*»«»««««**a««««««««*«*«c*««««*««*
i-« XtOKJIOMrOrOK7«(OrOK>fO«IOfOKlK>IOfOtOrOMfOK)lOfOMK>lOfOrortlOK)rOKIKlK:K)MK)tOM'O«
a
Z • OOOOPOOOOOOOOODOOOOCJOOOOOOOOOOOOODOOOOOOOOOOOODOOOOOOOOOOOOOOOOOOOOOOO
*-•> u^aiuitnu^tPu^u^ininuiinininuiininintninu^ininintnintou)inutinu^tniOinintninintntfi(nu7iointninin^
«^a.' e\icye\JCjNWWCVWcycvjC\ieNjWWf\iwc\!^cywcvcviCjcacyf\iwc^tvcuryweMWcye>icwN
WUJ •••••••••••• ••••••••*•••••••• ••••••••••••••••••••••« • • ••••••«•••••••*•
t5a CtOCSCJOOOOOCIOOOOOOOt3OOOOC,OOOOOOOOOeO^C^OOOCDOOOOOCDOC3OOOoeOOOOOOOC»OOOCDOOOOO
^
or
O
h- '
vt •
> jcvicye'fw^e\jcyw
CJLI aoe>ooocTOOooooctocr*oooc)ooooooooooc3c->ooooooac>ooaooooooooooGoooooooc}ooooooao
-*O. •••••••••••••••••••••••••••••o*c«*c»*»*««««««*«««**««*«»«*«««»*c»«««c«
U-> none oooooooo(^00000000000000ooooooc*oooooooooooaoc3ooc>t>ooooc=>ooooooooc>oooc3
a-^
3
W
Q* • OnOOOOOC'OOOODOOCT'C'OOoOOOOCiOOOOOOOOOOOCJOOOOOOO O^OOCJCJ^OOOC3OOOC>O^O^OC^CIOOOO
c> uiu'tu"iirtuu^tnu*inininiouJir)uTUTu7inininuiir)u^iPLnuiini^ui:irju^inu'>inu' tf}trfu^u)ininu>u^mtnu}tn
h-rc o*c>ic*jc\j.y^c\jWPJWc\:e\j(\ji>jv\jc^oJtMcsicv<\jcycNjcgcy<\ii\)we\jcy«cje\jCNjCNjc^
UUI •••••••••*••••••••• ••••«••••••!••••••••€••••••••••••••••••••••••
4CL OOOOO^OOOO^OOOi^OOC7OOOOOOO^OOOOC3OOO^OOO^CJOOOOC3C3OOOOOOC3O{3OOOOOOOOOOC2^OOO
U.
^J
U
O r^Kt^Crtpor- ^KK^K,fOK}rotr,»«iroKiKjf«*f^rtr^rof^K>fOKKffoWM
WUJ OO&r>>^OOO«3OOOOOL*OC30C'OOO<^C)«=>OC'>OOOOC.JOOOoOOOOOOOOC/OOOOOOcnOOOOC^C)OOOOOOOOOOOO
t-C. •••••••••••••••••••••••••••••o«**«»»«»««4*««»«««««*««»»««a**«t»a««**»a
t/!f- UjcLt »k™>c:»C''' •i^»c»e^o^oo'^cr'r°tr»;' to.zjor1 ot_'Or.r^oc» rsocjd OOC/OOOCJOC.>AC •ocic3'~>eioo'oc>c3OC'><*'>oraoC'O^c>c3c'' oooo
"^.HI
oe
t= I— ^OOOC'?OCJOOOOOOor.)OOC7OOOOOOC-OOe?OOOC.*OOOOOC'^OOOOOC30OOOO^OOOC3OOO^OOOOOOC9OC'}OO
O.U. OOOOOOOO(TlCtOC>C>OOOOOOOOOC.tOOOOOOC:C.OOOC>OOC>OOOOC3OOOO
_]*— OOOOClOOC3OUe3C7OOOOCtOOOC3OOOOC3OOOOOC-)OOOOOOOOOO^OOOOOOOOOOOOOOOOOOO^OOOOOOO
V7U. •••.•*.•.*••»•••••..•*••*•*•.«•..••••••.*•*•••**...«•**•••..*.....•.•.
*- , j«c^r»u.o"-ci*"jCjoo^ooooocjooooocfoc>e,t3C'o.Jooocjc3C'OoooOtL)ooooc:.ioooc>cc3Ciooc.*oocrjarioooooc'
*— *
2?J" t*CJC»O«.i&C"Cnl>ClC3OO'T*OC*OOOOOOC'OOOe'C3OOOCJOCiOOOOOOOOOOe.>(7>OC'' OtaciCJ OCJ£3CiOC3OC..>OOOOOC1 Owl Of)
ltl " ' -.'.'*'
&M
«;•— ••••••••••••••••••*••••••••••••*•••••*•• •""• •»••••••••••*••• »''• ••••••••••
uJO e3C\j»i*^HtyofOo.1Me\i^H'OfOo—«e*j^cv-»o^K»Hto*3-ofOo>-«cv; icc\i^jtncvie\j»HOJ»-t(\jo
otOOtiOOOC.^C3OO^3OC>OOO<_)C»OC>OOtriC'iC;oOOOOOC3Ot3C.OOOOOOOOOtJOOCjOOC3CJC30CJOOOC>OOOOO
H-«^ ••••*•••*••••»•••••€••••••••••*•••••••••*•••••••••••••••••••••••*•«•••
^H- .t*cutMOr-o^^aou3^c«h--n(T>iJ^^«o«uiaror^o^^-*Di3u')oirt—o-OLTO'-sDa»
tH rHtHv^i-lrH «-t -^ *H«H r-t »H
h-
act..
u-J
3 <*( K5 ~O ~t K i (O f"rf CM CJ CM CM CM CM OJCMC\K\JCVjC\I \\JC\J IN vX CM C\-fMW C%J <\l CM r\| (\j CM CVJ <.M iM f\J f\< CM W tN CM CV J.N ,M
O
OO tHt-«»H^^i-l(\J^JtM.NJC\K\JCJ'^OJ(.'yK3rOrOK»rO'OfO»OM»O^-r»,£)!*- «-4C5OCOcOCO 30CO(T«OMr>O.7^Cr«*3ff'0'*h*COCOoOoOff\ff»OC3*3ODOC^OCirH?HCyciJCMe\J
h=^ r^i-HtH«^^iHnt^rH^tHt^iHrH^rH^r^fH^iH*^iH<^THr4<^<^^IOrOrOC^C\JC\iCN.^i\:C\IOJrOWCMtO^
174
-------
'«co^
^i
ss«c
••••••••••••••••••••••••
••••••••
uiin
evw
•••••••••••••••••••••••••••
0)
I
4-J
a
o
o
• ••••••••••••••••••.
CJOOOOOOOOOOOOOOOOOOO
looooooociooeiinoocjocaciooo&oc'oooooooooooooc.-oooooooooooooooooooooeioooocjooooo
£|^u^u^tf>tninuiu^u-,u>ouir)Uiu}u^inuiui^uiu'JinirC3C>^C»C>O^C>C»OOCaOC>OC>CaOC>C5C>OC>CpC>OOC>OOOpC>OC-/e^C»OC>OC3OOOC3OCJC>OOO^C5e3C>O^C>C>C>C3OCiC>
I
Q
OK>r')^K>rorotOiOKj^iOroiOh>WKjfOM^'OH^«roK)(CK>KlMW«iOrOfOf'j«ro
3OOOOOC3OOOOOc>C;OOC3OC3C3OCiOOOOOClOOOOOOCJCTOOOOe>OOOC3OCJOC>^C»OOC3OOOOO^OOCJO
OOCiOOOOCJ
«•*•«.•••
•**•••
••.•«...
•«•••• ••••*•••
•••••••
••••»••••.*•.•.••
MtH^^pHCVjnjCVlC^Or^CMOi-tCNJAjnjr-tCMfHOr-tCUOCNi
3000000000000000000000000000000000000000000000
• •..•••..•*..*•.•••»•.........................
OK>CO CM CDuT^-^fo^in \DfOro •^•1/1 in vo«~*tocc vOfo^^»^cc:off>f»-K5rofOLnfvjinincoffc*uicOf'>''5ro*ocMoc
175
-------
W
OrOM
ooOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOaGOOOOOOOoOOOOOOOOOOOO
o *•: e\
o«a3eoM«cocoaoco^
>i<^
ooaoooaooooo
aooooQaOQOoooooooooooooaoooo
OOOC3OOC}OOOOOOOOC3OO^OOOOO
«•••••••••••••••••••••••
ooooooooaoociooooooooo
OOOOOOOOOOOC3OOOCJOCJOOO C3 OOOOOOO
;we\)e\jc>jojeN.e\jaO^OOOOC3aOC3OC3OOOC3OO
••••••••••••••••••••••••••••••••••••••••••••••••••••t*«
ooooaoooaooooooooooooooooooooooooooaooooooooooooooooocDoooo
O
UCI
r)ir.uiiru^
^
o
o
VO
Q
^>OOctC3OOCJOC30OOO"OCjOOOOOOOOOCJOOC
• •••••••••••••••••••••••^••••<
jooooeroi='t.i>c>cjooooc:ooe.oc:oc->oc oc-Ci.oc
I
OOOOOOOOOC
^
OOOOOOOOOCPOOOOOOOOOOOO OOOOO OoOO OOOO OOCJOOOCJOCIOOO OOO C2OOOOOOOCJ OCJO OOOO OOO
i -
^c\jf^olL"/^c\J*c\/rHrsiiy^<\)CNjcsitO*K)^ *C-a-vD
^
t^^t^i^^
lk
176
-------
TABLE D-7, PIPE ELEMENT DATA FOR COMBINED SEWER SYSTEM
SUBAREA III SUBAREA V
Pipe
Number
201
202
203
204
205
206
207
208
212
213
214
215
216
209
210
211
219
220
221
227
226
225
228
229
230
233
222
232
235
234
231
224
223
236
238
246
249
251
253
252
250
248
247
245
244
243
242
241
240
239
237
Slope
.0050
.0013
.0043
.0043
.0054
.0054
.0290
.0025
.0100
.0100
.0100
.0025
.0025
.0193
.0092
.0248
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0070
.0030
.0025
.0040
.0020
.0025
.0025
.0025
.0025
.0025
.0012
.0010
.0010
.0020
.0010
.0042
.0050
.0050
.0050
.0040
Length
(Ft)
220
440
200
250
300
230
320
480
850
450
470
300
750
350
250
300
1250
SUBAREA
400
350
650
280
330
350
280
260
290
165
500
320
300
580
180
400
540
250
250
340
300
300
250
250
250
360
320
250
320
320
300
300
300
400
Tributary
Area
(Acres)
8.88
4.00
13.31
1.28
5.17
2.20
1.61
5.28
16.52
2.39
8.99
5.10
5.10
2.09
1.84
1.36
9.69
IV
6.02
8.70
2.90
3.08
4.59
3.22
5.22
1.14
5.94
4.20
4.95
2.72
1.06
10.50
0
2.86
4.51
5.84
2.20
3.66
3.63
3.53
3.27
1.43
0.59
3.60
3.38
2.13
3.78
4.63
4.88
5.22
2.02
5.43
Surface
Elevation
(ft. MSL)
56.80
52.00
53.00
53.50
51.00
49.00
38.50
38.50
42.00
33.00
28.00
40.00
38.50
32.80
30.00
28.00
37.50
31.20
31.20
32.80
31.00
29.0Q
30.50
30.50
30.00
31.00
30.50
30.00
34.00
33.00
30.00
29.00
28.00
33.00
35.00
36.40
33.20
34.00 '
35.00
34.00
32.00
33.20
36.40
38.00
38.00
36.00
34.00
32.00
33.00
35.00
33.00
Pipe
Number
260
261
262
263
264
301
303
305
307
309
311
313
315
317
319
321
323
325
327
329
331
333
335
337
339
341
343
345
347
349
351
353
357
359
361
363
Slope
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0090
.0031
.0017
.0020
.0099
.0048
.0019
.0133
.0089
.0091
.0034
.0019
.0019
.0019
.0004
.0004
.0004
.0013
.0030
.0030
.0030
.0030
Length
(Ft)
480
540
490
350
360
TRUNK
130.0
250.0
250.0
270.0
290.0
300.0
800.0
420.0
390.0
330.0
240.0
250.0
260.0
320.0
320.0
390.0
390.0
350.0
340.0
340.0
220.0
430.0
320.0
870.0
430.0
220.0
710.0
340.0
470.0
900.0
750.0
Tributary
Area
(Acres)
8.63
7.78
6.42
2.09
7.63
SEWER
0
1.87
0
0.66
0.81
0.99
1.10
4.07
1.69
6.49
1.76
1.84
0
0
1.28
4.40
9.42
1.69
5.65
1.50
2.24
8.20
4.10
2.83
2.20
1.50
0.77
3.38
8.70
0
13.54
Surface
Elevation
(ft. MSL)
38.00
38.00
37.50
37.50
32.00
71.69
70.55
69.89
69.09
69.44
67,95
61.30
62.20
59.00
57.20
56.00
56.50
54.50
51.80
49.50
46.80
44.10
42.00
39.60
38.10
36.50
36.00
32.00
31.20
37.00
37.50
38.00
30.50
37.50
32.00
30.00
177
-------
TABLE D-7. (continued)
SUBAREA I
SUBAREA II
Pipe
Number
1
3
4
7
8
10
13
14
15
20
19
24
25
29
30
33
31
35
42
40
38
36
47
49
56
52
54
55
114
117
59
72
74
75
67
105
109
129
77
76
71
69
64
62
60
58
Slope
.0025
.0032
.0025
.0098
.0098
.0098
.0025
.0025
.0092
.0018
.0025
.0025
.0053
.0025
.0025
.0025
.0025
.0212
.0071
.0071
.0071
.0013
.0091
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0025
.0065
.0065
.0065
.0065
.0038
.0036
.0063
.0184
Length
(Ft)
280
240
280
220
420
300
240
570
310
310
570
570
310
230
300
220
235
300
275
260
280
620
480
600
360
400
280
240
220
370
680
420
260
540
290
150
150
690
690
220
300
300
650
290
290
270
Tributary
Area
(Acres)
3.71
3.60
4.66
1.65
3.05
3.78
3.23
3.38
3.05
3.01
6.42
6.79
2.86
2.75
5.65
3.56
3.45
0
2.82
3.12
3.08
5.14
4.58
3.56
4.15
6.26
1.87
2.02
0
4.11
4.15
4.95
4.98
3.12
3.47
0
0
1.60
0
1.36
1.14
5.50
7.16
2.28
4.44
1.28
Surface
Elevation
(ft. MSL)
73.24
72.49
72.49
77.62
73.30
71.69
77.00
70.55
70.55
69.89
69.89
69.09
69.09
76.00
74.81
76.00
74.81
68.44
72.52
70.63
68.95
68.44
67.95
63.50
62.00
64.00
63.00
62.00
61.30
63.00
62.00
71.00
72.00
71.00
68.50
62.00
61.30
73.00
72.00
71.00
69.00
68.50
68.00
65.00
62.00
62.00
Pipe
Number
82
101
99
98
95
94
92
91
189
187
85
104
103
108
112
119
78
80
83
121
123
125
126
84
128
148
149
147
146
143
141
139
138
135
87
89
133
Slope
.0025
.0025
.0075
.0025
.0062
.0025
.0141
.0025
.0104
.0425
.0025
.0217
.0025
.0025
.0025
.0025
.0025
.0025
.0055
.0025
.0025
.0025
.0054
.0059
.0233
.0025
.0388
.0413
.0118
.0025
.0058
.0026
.0025
.0085
.0085
.0051
.0051
Length
(Ft)
170
550
280
320
250
860
230
860
230
150
660
270
610
640
570
390
270
300
160
350
250
450
200
200
580
650
250
250
380
360
350
310
250.
150
150
220
310
Tributary
Area
(Acres)
4.51
.6.69
3.26
4.26
0.92
6.83
2.49
6.79
2.31
4.36
7.56
3.63
4.55
7.68
6.68
5.02
1.36
2.54
0
4.18
2.17
3.52
0
4.44
8.55
3.27
0.88
2.83
7.68
6.35
7.07
5.98
1.46
2.39
0.73
3.01
1.39
Surface
Elevation
(ft. MSL)
57.70
64.30
62.00
62.00
61.00
62.00
59.40
59.40
56.70
56.50
56.50
54.50
54.50
51.80
49.50
56.00
55.00
54.60"
55.00
56.00
55.00
55.00
54.90
54.80
44.10
69.00
60.10
50.50
48.30
46.20
44.10
43.10
43.10
42.80
42.20
41.00
39.60
178
-------
TABLE D-8. SUBCATCHMENT WATER QUALITY INPUT DATA FOR SWMM
SUBAREA LAND USE
NUMBER CUASS.
•1
2
3
4
5
I
8
9
11
12
13
14
IS
i ^
18
19
20
21
23
24
25
'26
27
28
29
i?
i2
33
I?
38
39
1
II
55
56
59
60
14
63
64
II
165
164
163
162
161
160
- 159
15ft
157
156
155
15*
153
152
150
149
148
147
146
145
144
143
148
141
171
35
36
45
46
47
46
49
50
51
53
66
2
2
2
2
2
2
J
J^
1
1
J
J_
1
J
J_
|
1
J_
J_
J
X
^
^
^
1
J^
J^
1
J
^
1
i
i
1
|
j^
^
i
1
i
i
I
1
1
^
^
i
i
i
i
i
i
i
^
i
j
^
1
^
i
i
TOTAL GUTTER MUHBER OF
UEN6TH*10**2 FT. CATCHBASINS
10.00
4.00
5.00
8.00
5.00
8.00
4.00
8.00
4.00
5.00
3.00
8.00
15.00
12.00
8.00
7.00
7.00
7.00
8.00
11.00
a. oo
11.00
8.00
6.00
11.00
14400
6.00
13.00
6.00
10.00
6.00
lO'.OO
7. -00
16400
20.00
16.00
14.00
3. .00
5.0.0
t. 00
.00
.5.30
13.00
4.00
7.00
4.50
27.00
27.00
15.00
6.00
5.00
8.00
6.00
7.30
6.50
8.00
4.00
5.00
s.oo
3.50
13.00
3.00
7*00
3.00
5.00
17.00
6.50
9.00
7.00
8.00
7.50
5.50
7.50
15.00
7.50
8.00
5.00
•9.00
5.00
13.50
11.00
6.00
8.00
9.00
5.00
3.00
6.00
4.50
,9.50
'8.00
7.50
12.00
8.00
0.0
0*0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
oTo1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
8.0
.0
§:§
O.G
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
o.c
0.0
0.0
0.0
0.0
0.0
0.0
040
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
o.o
0.0
0.0
0.0
0.0
0.0
S.o
.0
0.0
?«0
.0
0.0
g.o
0.0
0.0
o.c
179
-------
TABLE D-8. (continued)
SUBAREA
NUMBER
67
68
69
7P
71
72
73
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97,
98
99
100
101
102
103
104
105
106
107
10B
109
110
111
112
113
114
115
116
in
119
120
121
122
123
124
125
126
127
128
III
131
132
133
134
135
136
137
138
139
140
172
166
170.
310
309
306
283
284
282
286
285
294
295
296
300
298
299
§97
93
LAND USE
CLASS.
1
1
1
1
1
1
1
. 1
1
1
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 ,
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
\
1
1
1
1
1
1
1
1
1
i
1
1
1
1
1
1
1
1
1
1
1
1
1.
3
3
3
1
1
1
2
2
4
2
2
2
2
2
2
2
2
TOTAL GUTTER
LENGTH«10»*2 FT.
6.50
6.00
6.00
10.50
11.50
14.00
13.50
11.00
17.00
5.00
7.00
7.00
7.00
6.00
6.00
6.00
§. 00
.00
11.00
24.00
6.00
11.50
9.00
5.00
15.50
25.00
6.50
9.50
10.00
11.00
4.50
5.00
6.00
9.50
5.00
7.00
NUMBER OF
CATCH6ASINS
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
o.o •
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
O.D
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
o.o
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
180
-------
TABLE D-8. (continued)
SUB AREA :
NUMBER
274
287
288
302
303
304
305
291
292
201
202
203
204
205
206
207
208
209
218
219
220
III
223
224
225
226
227
228
229
230
231
232
233
^34
235
236
237
244
245
246
247
248
249
250
251
252 •
253
254
255
256
257
258
259
260
262
264
268
269
270
271
m
275
276
277
278
279
260
281
261
289
307
308
266
210
211
212
213
214
215
216
217
238
239
240
241
242
243
263
265
290
267
Lcf5i
3
2
: T
3
4
4
4
4
3
1
1
1
1
1
1
i
5
5
5
5
1
1
1
1
1
1
1
1
1
1
1
1
5
5
1
1
1
1
1
1
1
1
1
1
1
1
I
1
1
1
1
1
1
2
1
1
1
1
3
f
1
1
1
1
1
1
1
1
2
2
1
2
2
2
2
2
2
2
2
2
2
2
2
2
3
2
T.OTA
LENGT
L ,GUXT£R.
H*la**2 Ft.
6.00
20.00
9.00
9.00
17.00
17.00
6.00
17.00
1:88
9.00
9.00
9.00
9.00
10.00
10.00
10.00
15.00
10.00
6.00
6.00
6.00
7.00
10.00
!:88
6. DO
2.00
6.00
£.00
9.00
4.00
6.00
7. 00
7.00
13.00
12.00
11.00
11.00
5.00
13.00
11.00
11.00
11.00
6.00
6.00
9.00
12.00
5.00
9.00
5. CO
7.00
5.00
10.00
3.00
25.00
*:H-
8.00
6.00
9.00
8.00
9.90
6.00
13.00
12.00
11.00
12.00
.4 ..0.0
13.00
5.00
9.00
6.00
6.00
17.00
16.00
§.00
.00
12.00
18.00
8.00
9.00
7.00
11.00
10,00
9.00
10.00
10.00
9.00
8.00
9.00
3.00
16.00
jyaBnfi
0.0
0.0
0.0
0.0
0.0
0.0
p. o
6*0
0.0
0.0
0.0
0.0
0.0
0
p
0.0
0.0
0.0
0.0
0.0
0.0
0.0
o.o
0.0
0.0
1:1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
.0.0.
0.0
0.0
0.0
0.0
0.0
0.0
0.0
8:8
0.0
0.0
0.0
0.0
8.0
.0
O.Q
0.8
0.0
0.0
0.0
0.0
0.0
0.0
2 '2
0.0
O.Q
0.0
.0.0
0.0
0.0
0.0
0,0
0.0
0.0
O.Q
S.Q
.8
0.0
0.0
0.0
181
-------
than each separately. The pervious and impervious areas of each subarea
in District A were computed. The data on curb length to develop criteria as
to dust and dirt accumulation is included in TABLE D-9 for each major
subarea. The data were developed from an integration of subcatchment quality
input parameters shown in TABLE D-8. TABLE D-10 summarizes the District A
land use data which are used as direct input to the STORM program. The curb
length in single family residential is the greatest among all types of land
uses because, in Elizabeth, each single family lot is quite small, generally
0.08 hectares (0.20 acres) or less, and the street arrangement is compact.
TABLE D-9. SUBAREA CURB LENGTH (FEET) VERSUS LAND USE, DISTRICT A
Subarea
I
II
III
IV
V
Total
Single
Family
77,000
67,300
3,100
44,385
1,795
193,680
Multiple
Family
r-
2,480 •;
11,920 .
5,540
15,585
35,5-25
Commercial
-
120
6,980
1,810
920
9,830 '
Industrial Parks
_
;
3,900
3,685
_ • _
3,900 3,685
Note: 1 foot = 0.3048 meters
TABLE D-10. SUMMARY OF DRAINAGE DISTRICT A LAND USE DATA
Land
Use
Single Family
Multiple Family
Commercial
Industrial
Open Space
% of
Area
71.7
18.2
5.3 :
2.8
2.0
% Imper-
vious
43
50
80
80
19
Curb Length
(ft/acre)
413.
298. '
283.
216.
296.
1 ft/acre = .753 m/hectare
; 182
-------
The depression storage capacity as required by the STORM program is
a single parameter average over the entire drainage basin being considered.
With 47 percent impervious area, 1.59 mm (0.0625 inches) depression storage
capacity for impervious areas and 6.35 mm (0.25 inches) for pervious areas,
the average depression storage capacity for District A is computed as 3.94
mm (0.155 inches). In STORM, no runoff from a watershed takes place if
the total rainfall is less than available depression storage. STORM allows
depression storage to recover to its maximum capacity at a constant rate to
account for evaporation. The recovery rates are assumed constant for a given
month but may vary from month to month. The recovery of depression storage
is the mechanism in STORM to account for antecedent dry conditions and
enables the model to process continuous rainfall data.
The recovery rates are computed using the Meyer formula (40) with
the following expression:
E = C(e - e )
s a
(D-l)
10
where:
E = the rate of evaporation in inches per month,
eg = the saturation vapor pressure in inches of mercury (Hg) corresond-
ing to monthly mean air temperature,
e = the vapor pressure of air in inches of Hg, based upon monthly mean
air temperature and relative humidity,
W = the monthly mean wind speed in miles per hour obtained at 30 feet
above general level of surrounding roofs of building,
C = a coefficient depending on various factors affecting evaporation
and a value of 0.15 is suggested for small bodies of shallow waters
such as wet soil surfaces and small puddles.
. The monthly variation in evaporation rates for the study area were
computed and are shown in TABLE D-ll. The inputs to the Meyer formula
for computing the rate of evaporation are also shown. The inputs were
prepared using the climatological data recorded at Newark International
Airport during the period of 1963-1974. The computed evaporation rates agree
with the average annual lake evaporation in the New York area (41). Since
the evaporation rates affect the available depression storage prior to a
rainfall event, they have a more pronounced effect on runoff from low inten-
sity rainfall than from high intensity rainfall of the same duration.
183
-------
TABLE D-ll. MONTHLY EVAPORATION RATE
Mean Air
Temperature
Month (°F)
January
February
March
April
May
June
July
August
September
October
November
December
31.4
31.1
41.5
51.9
61.9
71.9
77.0
76.0
68.5
57.4
46.9
36.6
Mean
Relative
Humidity
(%)
65.3
59.3
63.3
57.0
.60.7
61.5
61.1
66.7
66.9
66.2
66.9
68.0
Mean
Wind
Speed
(mph)
11.3
12.2
12.1
11.6
10.5
9.6
9.2
9.2
9.6
9.6
10.5
11.1
Evaporation
Rate
(inches /day)
.065
.078
.105
.186
.232
.302
.322
.288
.233
.159
.112
.078
Note: 1 mph= 1.61 km per hour
1 inch/day - 25.4 mm/day
Pollutant washoff coefficient K is determined by the equation:
- P = Po(l-e-krAt)
(D-2)
where:
P - P
P°
P°
At
r
pounds of washoff in a time interval, At,
pounds of pollutant in the street at the start of rainfall,
pounds of pollutant remaining at the end of the time interval,
the time interval, and
runoff rate in inches per hour.
In SWMM, K is internally set equal to a value of 4.6, based on the assumption
that a uniform runoff rate of 12.7 mm (0.5 inches) per hour would wash
away 90 percent of the pollutant in one hour. For consistent comparision of
the results obtained with STORM and SWMM, the same washoff coefficient is
used for the STORM. There are studies in which other K values are used
(42).
Data for dry-weather flow quantity and quality required for simulation
of combined sewage overflow are discussed elsewhere.
AREA AND LAND USES FOR OTHER DISTRICTS
The City has been divided into 25 drainage districts. These districts,
184
-------
lettered A through Y, are shown in Figure D-14, which also shows the Easterly
and Westerly Interceptors and their corresponding inlet and sewer pipe
elements. *
Only composite data were required for drainage districts other than
District A. TABLE D-12 presents for each district the area, the various
kinds of land use, and the percent imperviousness. The areas of the planned
land use were taken from the approved master plan (43). Where planned
ultimate land use differed from the existing use, the use resulting in the
greater population was adopted. The percent imperviousness in each district
was the weighted average based on the percent imperviousness for various land
uses as shown in TABLE D-10. Two-family houses were considered as part of
the multiple family residential area. District A is the largest district in
the City and includes all land uses and, therefore, permits developing
relevant parameters for City—wide application.
The developed population of each district are shown in TABLE D-13.
The populations estimated are based on densities of 55.8 and 121.3 persons
per hectare (22.6 and 49.1 persons per acres) respectively for single-family
and multiple-family residential areas. These densities were obtained from
census data at presently developed areas for each land use. These estimated
populations were used to compute dry-weather flow. The estimated developed
population for District A is about 16,500 and for the City, 124,000. The
present population of the City is about 115,000.
The curb lengths for the various land uses were developed from data
presented in TABLE D-10 for District A. The curb length is used to estimate
the accumulation of dust and dirt on the streets and to define the amount
of pollutant washoff. ' " . *
DRY-WEATHER FLOW QUANTITY AND QUALITY
Use of SWMM and STORM as modified, for the analysis of combined sewage
quantity and quality requires ' input of dry—weather.flows, including domestic
sewage, commercial and industrial discharges and infiltration. Domestic
sewage flows were estimated from daily per capita water use and commercial
and industrial flows from average contributions per area. unit. Both commer-
cial and industrial wastewater quantities vary widely depending upon the type
and size- of the commercial and industrial development (44). Quality compo-
nents in the commercial and industrial wastewaters also vary widely (45,46).
Both the quantity and quality of dry-weather flow vary with the hour of the
day and the . day of the week. While dry-weather flows make a significant
contribution to the quality of combined sewage, their average quantity is
small, generally less than 0.2 percent of the sewer design flow computed,
using SWMM, with the 5-year return interval, intermediate pattern, synthetic
storm hyetograph (Figure D-5).
With the assumed per capita contribution of 379 liters (100 gallons)
per day from all sources, the average daily wastewater quantity from District
A is 6,245 m (1.65 million gallons). The average daily concentrations
of pollutants in the wastewaters adopted are shown in TABLE D-14.
185
-------
BOROUGH V,
OF
ROSELLE \>
PARK II
BOROUGH
OF
ROSELLE
WESTERLY '
INTERCEPTOR
EASTERLY
INTERCEPTOR
LINDEN
TO
JOINT MEETING
TREATMENT
PLANT
NOTES;
I. LETTERS DESIGNATE
DRAINAGE DISTRICTS.
2. NUMBERS ARE
INTERCEPTOR SEWER
ELEMENTS.
Figure D-14.
Definition of drainage, districts and interceptor system in
Elizabeth '
186
-------
TABLE D-12. SUMMARY OF LAND USE IN THE CITY OF ELIZABETH
Land Use (%)
District
A
B
C
D
E
F
G
H
I
J
K
L
M
N
0
P
Q
R
S
T
U
V
W
X
Y
Area
(acres)
655
112
111
117
229
88
34
122
70
420
65
66
67
79
103
37
83
303
207
62
331
440
415
138
56
Single
Family
72
90
20
10
35
0
0
0
0
10
0
0
0
0
5
0
10
0
0
0
0
0
0
15
0
Multiple
Family
18
10
55
60
35
70
0
25
85
70
60
50
90
100
0
50
40
0
75
50
50
40
90
65
100
Commercial
5
0
25
30
30
30
100
75
15
15
40
10
5
0
0
50
25
0
5
0
25
27
10
0
0
Industrial
3
0
0
0
0
0
0
0
0
5
0
40
0
0
50
0
25
100
15
50
25
30
0
0
0
Park
2
0
0
0
0
0
0
0
0
0
0
0
5
0
45
0
0
0
5
0
0
3
0
20
0
District
Percent
Impervi-
ousness
47
44
56
58
57
59
80
73
55
55
62
65
50
50
49
65
64
80
54
65
65
66
53
43
50
City
4410
17.3
47.5
16.3
16.3
2.6
58
NOTE: 1 acre = 0.405 hectares
Acres of Land Use indicated are based on approved Master Plan of the
City (43)
187
-------
TABLE D-13. POPULATION AT SATURATION DEVELOPMENT
District
A
B
C
D
E
P
G
H
I
J
K
L
M
N
0
P
Q
R
S
T
U
V
W
X
Y
Area
(Acres)
655
112
111
117
229
88
34
122
70
420
65
66
67
79
103
37
83
303
207
62
331
440
415
138
56
Single*
family
10,658
2,278
502
264
1,811
0
0
0
0
949
0
0
0
0
116
0
188
0
0
0
0
0
0
468
0
Multiple**
Family
5,789
550
2,998
3,447
3,935
3,025
0
1,498
2,921
18,560
1,915
1,620
2,961
3,879
0
908
750
0
7,623
1,522
8,126
8,642
18,339
4,404
2,750
Total
16,447
2,828
3,500
3,711
5,746
3,025
0
1,498
2,921
19,509
1,915
1,620
2,961
3,879
116
908
938
0
7,623
1,522
8,126
8,642
18,339
4,872
2,750
Total
4,410
17,234 106,162
123,396
* Based on 22.6 persons per acre (55.8 persons per hectare)
** Based on 49.1 persons per acre(121.3 persons per hectare)
Population Densities for indicated land uses are derived
using census data from presently developed areas.
188
-------
TABLE D-14. AVERAGE DAILY POLLUTANT CONCENTRATION OF DOMESTIC WASTEWATER
Pollutant Concentration
Suspended solids
BOD
Settleable solid
Total nitrogen
Total phosphate
Total coliform
240 mg/1
200 mg/1
6 mg/1
30 mg/1
10 mg/1
5.27 x 107 MPN/100ml
These pollutant concentrations are classified as "medium" according to an EPA
study report (45). Diurnal variations of sanitary wastewater quantity and
quality are shown in TABLE D-15. These variations are based on experience in
cities with population and urbanization comparable to Elizabeth. The peak
hourly rate occurs at 10 a.m. and equals 1.45 times the daily average flow
rate. For District A, the peak hourly wastewater flow rate is 0.105 m /sec
(2.4 mgd). Possible variations in quality and quantity of wastes by day of
the week were not considered.
The hourly DWF pollutant loadings for SS, BOD, and total coliform
in TABLE D-15 are based on the daily average concentrations shown in TABLE
D-14. It was assumed that the same concentration ratio variation applies to
all pollutants. During the preparation of input data to STORMS pollutant
concentration ratios in TABLE D-15 were used as pollutant mass loading. This
artifically increased the pollutant mass loading in the early hours of the
day and decreased it in the later hours. As a result, the amount of un-
treated sanitary sewage in the overflow may either increase or decrease for
individual storm events, depending upon the time of the day the rainfall
starts. On the average, it may increase the computed amounts of annual
pollutants discharged by about 10 percent. This deviation in computing
.annual amount of pollutants in the overflow would be the same for all alter-
natives tested and would not change the relative proportions of pollutants
discharged which- were used as a measure for determining the desirable
systems. Based on the assumed SS contribution from commercial and industrial
areas to be 0.06 and 0.08 kg per hectare (0.33 and 0.44 pounds per acre) per
day respectively, the total non-domestic portion of SS in the dry-weather
flow would be about 9 kg (20 pounds), or less than one percent of the assumed
wastewater loading. The diurnal variations shown in TABLE D-15 are DWF input
data to STORM.
In SWMM, consideration of DWF is made in the TRANSPORT Block. Thus,
for District A, DWF enters the trunk sewers directly. The district was
divided into 25 sub.areas. TABLE D-16 presents, for each subarea, the inlet
manhole, area, predominant land use, population density, average daily
flow and daily SS and BOD loading. Other than daily pollutant loadings,
189
-------
TABLE D-15. DIURNAL VARIATION OF DOMESTIC WASTEWATER
QUANTITY AND QUALITY FOR DISTRICT A
Hour of
the day
Flow
Ratio*
Flow
Rate
(mgd)
Concen-
tration
Ratio*
SS
(lbs/hr)
BOD
(lbs/hr)
Total
Colifortn
MPN x 1015/hr
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
.9
.7
.5
.35
.30
.30
.40
.85
1.35
1.45
1.35
1.30
1.25
1.20
1.15
1.05
1.05
1.15
1.49
•1.16
.83
.58
.50
.50
.66
1.40
2.23
2.40
2.23
2.15
2.
1.
1.
1.
1.
,25
,35
1.35
1.25
1.15
1.05
.06
.98
.90
1.73
1.73
1.90
2.06
2.23
2.23
2.06
1.90
1.73
.90
.70
.60
.50
.45
.40
.45
.50
.60
.70
.85
1.15
1.25
1.50
1.55
1.55
1.50
1.40
1.35
1.30
1.25
1.20
1.20
1.15
123
96
82
69
62
55
62
69
82
96
116
158
171
206
212
212
206
172
185
178
171
164
164
158
Daily Average = 1.65 Daily Total =3289
103
80
68
.57
51
46
51
57
68
80
97
131
143
171
177
177
171
160
154
148
143
137
137
131
2738
.124
.096
.083
.069
.062
.055
.062
.069
.083
.096
.117
.158
.172
.206
.213
.213
.206
.193
.186
.179
.172
.165
.165
.158
3.3
* Ratios are with respect to the corresponding average values.
Note: 1 mgd = 0.0438 m3/sec
1 Ib = 0.453 kg
190
-------
TABLE D-16.
SWMM INPUT DATA FOR DOMESTIC WASTEWATER COMPUTATION,
DISTRICT A
Subarea
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Inlet
Manhole
360
358
356
352
352
353
350
244
342
338
334
330
328
326
324
316
316
314
312
310
308
306
304 ,
302
300
Area
(Acres)
14.53
31.56
8.70
67.10
135.35
10.46
21.32
4.10
8.20
46.78
37.32
17.75
10.46
11.48
39.14
17.64
54.08
24.35
8.65
32.69
10.64
10.24
10.32
10.35
11.97
Land*
Use
2
2
2
1
1
1
1
3
3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Population
Density
(persons
per acre)
37.0
47.8
43.1
22.6
23.4
22.6
30.7
15.7
22.6
22.6
30.6
22.6
22.6
22.6
22.6
22.6
22.6
22.6
22.6
22.6
22.6
22.6
22.6
22.6
22.6
Average Average
Daily Daily Loading
Flow
(cfs)
.083
.233
.058
.237
.469
.036
.101
.010
.028
.164
.177
.062
.037
.040
.137
.062
.181
.085
.030
.115
.037
.036
.036
.036
.043
ss
(Ibs/day)
107.4
301.5
75.0
306.6
641.8
46.6
130.7
12.9
36.2
212.2
229.0
80.2
47.8
51.8
177.3
80.2
234.2
110.0
38.8
148.8
42.9
46.6
46.6
46.6
55.6
BOD
(Ibs /day)
89.5
251.2
62.5
255.5
534.8
38.8
108.9
10.8
30.2
176.8
190.8
66.8
39.9
43.1
147.7
66.8
195.2
91.6
32.3
124.0
39.9
38.8
38.8
38.8
46.4
2.56
3307.3
2760.
Predominant Land Use
1 - Single Family Residential
2 - Multiple Family Residential
3 - Commercial
Note: 1 person/acre = 2.471 persons/hectare
1 cfs = 0.0283 m3/sec
1 Ib/day = 0.453 kg/day
191
-------
which are computed internally in the program, additional data are required
for the computation of combined sewage. The slight differences for daily SS
and BOD loadings between TABLE D-15 (for STORM) and TABLE D-16 (for SWMM) are
due mainly to the significant number contained in the SWMM print out and are
inconsequential. The same diurnal quantity and quality variations shown in
TABLE D-15 were applied to SWMM. In SWMM computations, the simulations were
assumed to begin at 10 a.m. when the hourly DWF quantity was greatest.
UNIT COST DATA
To compare the cost—effectiveness of the alternative sewer systems,
cost data were developed for sewers, storage facilities, pumping stations
and treatment works. These costs are affected by variables such as land,
weather, subsurface conditions, groundwater, access, size of facilities,
process involved, etc. The cost data used are applicable to work in
Elizabeth, New Jersey and consolidate many factors that affect cost into
a single parameter (such as costs per foot for each size of sewer for various
depths of excavation, per gallon of storage capacity, per mgd of pumping
capacity, etc.). These parameters are adequate for comparative purposes but
would require refinement for definitive projects at specific sites.
Sewer Costs
Figure D-15 and TABLE D-17 present pipe cost per linear foot for various
depths of trench developed from actual cost data. The unit cost includes
the cost of pipe, pipe laying, excavation, dewatering, temporary tight
sheeting and backfill, and is based on the ENR construction cost index of
1,800. Pipe sizes smaller than 0.457 m (18 inches) are used only for sani-
tary sewers. Circular pipe sewers were assumed. The unit price increases
substantially at about a 6.1 m (20 fe,et) trench depth due to the additional
bracing and sheeting required and greater excavation costs.
Storage Costs
Costs of storage facilities presently in operation have been summarized
(18, 19). The cost of in-pipe storage, as used in Seattle, Washington,
was estimated at about $0.06 per liter- ($0.23 per gallon) of storage, mostly
for control and monitoring systems and automated regulating stations in
existing pipes. The cost of off-line storage depends upon the type and can
include open lagoons, concrete covered reservoirs or deep tunnels. The open
lagoon, such as the one used in Chippewa Falls, Wisconsin, costs about $.07
per liter ($0.26 per gallon) of storage. The cost range for covered concrete
storage is from $0.13 per liter ($0.50 per gallon) for Humboldt Avenue,
Milwaukee to $0.56 per liter ($2.12 per gallon) for a more recently built
storage basin at Jamaica Bay, New York City. For the deep tunnel storage
system in the City of Chicago, the cost was estimated as $0.06 per liter
($0.24 per gallon). The above costs are based on the ENR construction cost
index of 2,000. In a sewerage improvement program recently developed for the
City of Trenton, New Jersey (16), the cost of a sewage detention basin and
appurtenances was estimated at about $0.07 per liter ($0«25 per gallon)
192
-------
3000
2000
1000
900
800
700
600
x- 500
in
t_
o
= 400
o
•o
I- 300
O
O
u.
< 200
b
o
100
90
80
70
60
50
40
30
20
I. COST INCLUDES TRAFFIC, EXCAVATION,
OEWATERING, MAINTENANCE, UTILITIES,
TEMPORARY SHEETING, PIPE, BACKFILL,
TESTING AND RESTORATION.
(BASED ON ENR CONSTRUCTION COST
INDEX 1800)
2. NUMBERS ARE SEWER DIAMETERS IN INCHES.
6-8 8-10 10-12 12-14 14-16 16-18 18-20 20-22 22-24 24-26 26-28 28-30
EXCAVATION DEPTH (ft.)
Figure D-15. Unit cost of sewers
193
-------
•K
CO
Pi
w
CO
H
CO
o
o
H
M
25
r--
H
H
H
OOOOOOCMOOirtOOOOOOOOOOOOOOOOOOOOOO
----
d>
CO
o
cr>P9?t^'. cr» o rH CM co prHcMcoirir—cni I I I I i i i i i i i i i i i i
3
CO
CO
I rH CM CO tO I
VO - -- -
u
01
4J
QJ r^ co in in m
J/ -------
A.R-
a)
-a
C
o
ro
194
-------
of storage, based on the ENR construction cost index of 2,800. This cost
included an improvement of the existing 64,345 m (17 million gallons)
detention basin, a 8.3 m /sec (190 mgd) capacity pumping station and over
610 m (2,000 feet) of inlet and outlet sewers.
For the purpose of this study it appears reasonable to assume, from the
above data, a cost of $0.08 per liter ($0.30 per gallon) of storage. Higher
storage costs, up to $0.26 per liter ($1.00 per gallon), were also considered
to analyze the tradeoff between various amounts of storage facilities and
varying collection sewer, interceptor, and treatment plant capacities for
joint flood and pollution control where local conditions would result in
higher storage construction costs.
Pumping Station Costs
Pumping is generally required to deliver to or return from storage
wet weather flows. In very steep terrain, gravity flow into and out of the
storage facility may be possible. The cost of the pumping station proposed
for Tnenton is estimated at about $1,710,000 or an average of about $206,000
per m /sec ($9,000 per mgd) based on the ENR construction cost index of
2,800. This cost has been used in this study without adjustment for cost
index difference.
Treatment Costs
The alternative treatment processes for stormwater runoff and combined
sewage and their associated treatment costs were presented in Lager and
Smith (18) and later summarized by Field and Lager (19). The cost data
plotted in Figure D-16 are taken from this latter reference. Capital cost
per mgd of treatment capacity appears to increase nearly linearly with
BOD removal efficiency. The microstrainer, which is relatively economical
but relatively inefficient in removing BOD, is effective in SS removal with
a removal rate of 70 percent (19). Biological treatment processes are
impractical for satellite plants because of the problem of maintaining an
effective biomass with intermittent flows. However, it is practical and may
offer benefit to treat combined sewage or urban runoff at a central plant.
3
Elizabeth has been allocated a peak treatment rate of 1.75 m /sec (40
mgd) at the Joint Meeting Plant (about two times average flow). The plant
is being upgraded to provide secondary treatment. Hence, one approach
investigated provided external storage near the treatment plant for flow
equalization to make effective use of the allocated capacity and as a result,
the treatment costs for the alternative sewer systems were practically the
same. A second approach assumed varying peak treatment rates (up to 17.5
m /sec or 400 mgd) and appropriately modified storage quantities for flow
modulation. The SWMM STORAGE/TREATMENT Block was used to estimate costs under
this approach. All costs were adjusted to the ENR construction cost index
of 1,800.
195
-------
90
80
70
60
rx 5
o
o
o
•o
o>
E
O
o
o
50
40
30
20
/x 4 .
10 -
SECONDARY TREATMENT PROCESS
I. DISSOLVED AIR FLOTATION
2. MICROSTRAINER
3. ULTRAFINE SCREEN
4. CHEMICAL CLARIFICATION
USING WASTE LIME SLUDGE
5. CONTACT STABILIZATION
MODIFICATION OF
SECONDARY PLANT
6. HIGH RATE TRICKLING FILTER
MODIFICATION OF SECONDARY
PLANT
X DATA OBTAINED FROM
REFERENCE 19 (BASED ON ENR
CONSTRUCTION COST INDEX OF 2000)
10 20 30 40 50 60 70 80 90 100
% REMOVAL OF BOD
Figure D-16. Cost of treatment vs. % of BOD removal
196
-------
APPENDIX E
CALIBRATION OF STORM RUNOFF COEFFICIENTS
Two runoff coefficients are specified in STORM: C, and C~, respec-
tively, for pervious and impervious areas. Infiltration Tosses are limited
to pervious areas. Consequently, C, is variable while C~ is essentially
constant.
Infiltration in pervious areas is generally the most significant
rainfall loss. Infiltration rates may vary from several inches per hour in
extremely pervious deep sands to zero on tight clays. The infiltration
rate is dependent on soil characteristics, the terrain, the rainfall rate,
pattern and duration, and on the water content of the soil.
Because of the variability in the factors affecting the infiltration
rate for a given area, C, is not susceptible to precise determination.
It is higher for less frequent, more intense storms because infiltration
and other losses have a proportionally smaller effect on runoff.
Runoff coefficients for STORM were calibrated using runoff data gener-
ated by SWMM in District A. These coefficients were applied to other
drainage districts in the City to obtain the runoff volumes for synthetic
or real storms. The following assumptions were made:
1. C~ was set equal to 1.0, since there is no infiltration loss in
impervious areas and the depression storage is accounted for
separately.
2. Surface runoff data were generated by adapting the quantity
portion of the SWMM RUNOFF Block to exclude sewer routing.
This is consistent with STORM, in which the effect of sewer
routing is not considered. All subcatchments in District A
were used.
3. The depression storage capacities for pervious and impervious
areas used in SWMM were 6.35 and ,1.57 mm (0.25 and 0.062 inches)
respectively. Twenty five percent of the impervious area was
assumed to have no depression storage. The equivalent depression
storage for District A was 3.94 mm (0.155 inches), based on 47
percent of the area being impervious.
4. The infiltration capacity curve shown in Figure D-5 was used
in SWMM to account for the infiltration loss. Maximum and
197
-------
5.
minimum infiltration rates of 76.2 and 7.11 mm (3.0 and 0.28
inches) per hour, respectively, and a decay rate of 0.00138/
sec was used. The antecedent conditions for rainfall events
were such that the infiltration curve specified applied.
The typical rainfall event was assumed to have a 3-hour duration
and an intermediate pattern similar to the 5-year synthetic storm
pattern shown in Figure D-5.
C, was calibrated so that the 3-hour storm runoff volume computed with the
computer program adapted from SWMM was the same as that computed with the
STORM program.
Sensitivity runs were made using the adapted SWMM program to investi-
gate the effect on the runoff volume of (a) using the hourly hyetograph as
opposed to 2-minute hyetograph, and (b) the integration interval. These runs
were made for the 5-year synthetic storm in District A. The results are
presented in TABLE E-l.
TABLE E-l. SENSITIVITY OF HYETOGRAPH AND
INTEGRATION INTERVAL ON RUNOFF VOLUME
Case
No.
Hyetograph*
Integration
Interval
(min.)
5-Year Storm
Rainfall Runoff
(inches) (inches)
Percent of
Rainfall
Runoff
1
2
3
2 - min
Hourly
Hourly
2
2
60
2.43
2.43
2.43
1.75
1.69
1.72
0.72
0.70
0.71
* See Figure D-5 for intermediate pattern storm hyetograph
Note: 1 inch = 25.4 mm
There is insignificant difference in runoff volume obtained by using the
hourly hyetograph with a one-hour integration interval and that obtained by
using the 2-minute hyetograph with a 2-minute integration interval. This
finding does not apply to peak flow rates, however, where the hyetograph
interval makes a substantial difference. The savings in computer costs in
using hourly intervals to determine runoff volume are substantial being only
25 percent of those using 2-minute intervals. Hence, data generated with
the adapted SWMM program for calibration purposes use the hourly hyetograph
and integration interval.
The calibrated C, values are shown both in TABLE E-2 and Figure E-l as
198
-------
a function of 3-hour rainfalls. C, increases with rainfall intensity.
TABLE E-2. CALIBRATION OF RUNOFF COEFFICIENT C
Case
No.
1
2
3
4
3-hour
Rainfall
(inches)
4.86
2.43
1.38
0.61
Average
Return Intensity
Interval (In/hour)
> 100-Year
5-Year
1-Year
1.3 -Month
1.62
0.81
0.46
0.20
Runoff
Volume
(inches)
4.15
1.72
0.76
0.27
STORM
Cl
0.78
0.55
0.30
0.25
Note: District A data: Area = 655.1 A.cres
Impervious ness == 47%
85 percent of the rainfalls during the period
studied (1963-1974) totaled less than 0.6 inches
1 inch/hour = 25.4 mm/hour
The effect of storm duration on the runoff coefficient was also studied.
Computer runs were made with assumed 6-hour storm duration maintaining the
same average storm intensities as for the 3-hour storm. The increase in storm
duration increases the total amount of infiltration. Hence, for the same
average intensity rainfall, coefficients for longer duration storms are more
affected than those for shorter duration storms. For instance, the GI
value for a 3—hour storm of average intensity equal to 5.08 mm (0.20 inches;
per hour is computed to be 0.23. For the same intensity storm with a
6-hour duration, C. equalled 0.11. For storms with average intensity equal
to 20.6 mm (0.81 inches) per hour (5-year return frequency for the 3-hour
storm), the difference in runoff coefficient C. was insignificant (0.55
versus 0.58). However, the return frequency of a 6-hour storm with an
average intensity of 20.66 mm (0.81 inches) per hour is about 100 years or
longer.
Considering that C. varies with the storm severity (depth, duration)
and STORM assumes a constant runoff coefficient for the entire time span of
199
-------
-waois •yx-ooi
WU01S 'HA-9
o
II
tu
o:
CO
ID
o
tr
LU
Q.
a:
o
z u_
O UJ
"•8
< u-
Q 2
WU01S H1NOW-SM-
I
C7>
O
'CD
o
in
10 _2
ro* o
CO
to
00 Is- <0 IO
• • • ,
('o) vayv snoiAyad yod
ro
cvj
d
•1-1
n)
CO
>
0)
•rl
a 5
=1 1
j£. o
Z IH
2 2
o
H
H
cu
S-l
60
•H
ft)
200
-------
records to be simulated regardless of rainfall volume or intensity, a C,
value of 0.25 was used for the simulation of long-term records for evaluation
of overflow pollutional effect. Reducing C, to 0.15 would result in a nine
percent reduction in the-runoff volume from District A. In addition,
the amount of pollutant washout from streets is independent of C, in
analyses using STORM as previously discussed. The selected C, value of
0.25 should provide sufficiently consistent results when using STORM to
permit valid evaluation of the effectiveness of pollution control measures.
To estimate the 5-year synthetic storm runoff volume from other districts in
the City, however, C, value of 0.55 was used.
201
-------
SECTION XIV
REFERENCES
1. Field, R. and E.J. Struzeski, Jr., "Management and Control of Combined
Sewer Overflows", J. Water Pollution Control Federation, Vol. 44, No.
7, 1972. pp. 1393-1415.
2. Bryan, E.H., "Quality of Stormwater Drainage from Urban-Land Areas in
North Carolina", Report No. 37, Water Resources Research Institute of
The University of North Carolina, Raleigh, N.C., 1970.
3. Field, R. and A. N. Tafuri, "Stormflow Pollution Control in the U.S.",
in: Combined Sewer Overflow Seminar Papers, U.S. EPA Report,
EPA-670/2-73-007, NTIS-PB-231 836, November 1973.
4- Nebolsine, R. and G. Vercelli, "Is the Separation of Sewers Desir-
able?" in: Proceedings, National Symposium on Urban Rainfall and
Runoff and Sediment Control, Lexington, Kentucky, July 29-31, 1974.
5. Tholin, A.L. and C.J. Keifer, "Hydrology of Urban Runoff", Trans-
actions of the American Society of Civil Engineers, Vol. 125, 1960,
pp. 1308-1379.
6. Metcalf & Eddy, Inc., University of Florida, and Water Resources
Engineers, Inc., "Storm Water Management Model", U.S. EPA Report No.
11024 DOC 07-10/71, NTIS-PB 203 289-292, Vol. 1-4, 1971.
7. Sevuk, A. S., B. C. Yen, and G. E. Peterson, Illinois Storm Sewer
System Simulation Model; User's Manual, Water Resources Center
Research Report No. 73, University of Illinois, Urbana-Champaign,
Illinois, October 1973. .
8. Parsons, Brinckerhoff, Quade & Douglas, "Sewerage, Drainage and Flood
Control Improvement Program", Report prepared for the City of
Elizabeth, New Jersey, April 1962.
9. U.S. Weather Bureau, Department of Commerce, Technical Paper No. 29,
"Rainfall Intensity-Frequency Regime, Part 3 - The Middle Atlantic
Region", July, 1958.
202
-------
10. Sonnen, M.B., "Abatement of Deposition and Scour in Sewers", U.S.
, EPA Report, Contract No. 68-03-2205, to be published by NTIS.
11. Clinton Bogert Associates, "Sewerage and Drainage Improvement Program,
Districts A-l and A-2", report prepared for the City of Elizabeth, New
Jersey, January 1972.
12. Harmon, W.G., "Forecasting Sewage System Discharge at Toledo",
Engineering News-Record, 1918.
13. Materials obtained from Carl Maegaard, Hydro Storm Sewage Corporation,
919 Third Avenue, New York, New York 10022.
14. Janson, L.E., S. Bendixen and A. Harlaut, "Equalization of Flow
Variations in Combined Sewers", Journal of the Environmental Engineering
Division, Proceedings of the American Society of Civil Engineers, Vol.
102, No. EE6, December 1976.
15. Pisano, W.C. and C.S. Queiroz, "Procedures for Estimating Dry Weather
Pollutant Deposition in Sewerage Systems", U.S. EPA Report, EPA-600/
2-77-120, NTIS-PB 270 695, July, 1977.
16. Clinton Bogert Associates, "Facilities Plan", report prepared for the
City of Trenton, New Jersey, January 1976.
17. Yen, B.'C., and A.S. Sevuk, "Design of Storm Sewer Networks", Journal
of Environmental Engineering Division, Proceedings of the American
Society of Civil Engineers, Vol. 101, No. EE4, August 1975.
18. Lager, J.A. and W.G. Smith, "Urban Stormwater Management and Tech-
nology, An Assessment", U.S. EPA Report EPA-670/2-74-040, NTIS-PB 240
697, October 1974.
19. Field, R. and J.A. Lager, "Urban Runoff Pollution Control - State-of-
the-Art", Journal of the .Environmental Engineering Division, Pro-
ceedings of the American Society of Civil Engineers, Vol. 101, No.
EE1, February 1975.
20. Brandstetter, A., R. Field and H.C. Torno, "Evaluation of Mathematical
Models for the Simulation of Time-varying Runoff and Water Quality in
Storm and Combined Sewerage Systems", Proceedings of the Conference on
Environmental Modeling and Simulation, Cincinnati, Ohio, April 19-22,
1976.
21. Henderson, F.M., Open Channel Flow, The MacMillan Company, New York,
1966.
22. Chow, V.T. Open Channel Hydraulics, McGraw-Hill Book Company, New
York, 1959.
203
-------
23. Schaake, J.C., Jr., "Synthesis of the Inlet Hydrograph", Ph.D. Thesis,
The Johns Hopkins University, Baltimore, Maryland, 1965.
24. Eagleson, P.S., Dynamic Hydrology, McGraw-Hill Book Company, New York,
1970.
25. Brandstetter, A., "Comparative Analysis of Urban Stormwater Models",
in: "Short Course Proceedings, Application of Stormwater Management
Models", U.S. EPA Report, EPA-670/2-75-065, NTIS-PB 247 163, June
1975.
26. Brandstetter, A., "Assessment of Mathematical Models for Storm and
Combined Sewer Management", U.S. EPA Report, EPA-600/2-76-175a, August
1976.
27. Hydrologic Engineering Center, U.S. Army Corps of Engineers, "Urban
Storm Water Runoff: STORM", Generalized Computer Program 723-S8-
L2520, May 1974.
28. Hydrologic Engineering Center, U.S. Army Corps of Engineers, "Storage,
Treatment, Overflow, Runoff Model (STORM)" Generalized Computer Program
723-S8-L7520, July 1976.
29. Huber, W.C., J.P. Heaney, M.A. Medina, W.A. Peltz, H. Sheikh, and G.
Smith, "Storm Water Management Model User's Manual, Version II", U..S.
EPA Report, EPA-670/2-75-017,.March 1975.
30. "Modifications to the EPA Storm Water Management Model Documentation
(Version of March, 1975)," prepared by Water Resources Engineers, Inc.,
September 1975 and updated by University of Florida, May 1976.
31. Chow, T. and Ben Chie Yen, "Urban Stormwater Runoff: Determination
of Volumes and Flow rates", U.S. EPA Report EPA-600/2-76-116, NTIS-PB
252 410, May 1976.
32. American Society of Civil Engineers, "Design and Construction of
Sanitary and Storm Sewers", Manual and Report on Engineering Practice
- No. 37, 1970.
33. U.S. Army Corps of Engineers, Louisville District, Kentucky, "Supple-
mental Report, Definite Project for Local Flood Projection,
Louisville, Kentucky" June 1949.
34. American Public Works Association, "Water Pollution Aspects of Urban
Runoff", U.S. EPA Water Pollution Control Research Series, WP-20-15,
January 1969.
35. Huber, W.C., "Difference Between Old and New Runoff Quality Models",
Project Memo to N682 File, Grant R802411, University of Florida,
Gainesvile, Florida, May 28, 1974.
204
-------
36. Ganczarczyk, J.J., "Storage for Stormwater Ouallty Control", Paper
presented at the SWMM Users Group Meeting, Toronto, Ontario, April 1,
1976.
37. U.S. Weather Bureau, Department of Commerce, Technical Paper No. 25,
"Rainfall Intensity - Duration - Frequency Curves", December 1955.
38. U.S. Weather Bureau, Department of Commerce, Technical Paper No. 40,
"Rainfall Frequency Atlas of the United States for Durations from 30
Minutes to 24 Hours and Return Periods from 1 to 100 Years", May 1961.
39. Chow, V.T., "Statistical and Probability Analysis of Hydrologic Data",
in Handbook of Applied Hydrology, V.T. Chow, Editor, McGraw-Hill Book
Company, New York, 1964.
40. Meyer, A.F., "Computing Run-Off from Rainfall and Other Physical Data",
Transactions of the American Society of Civil Engineers, Vol. 79,,1915,
pp. 1056-1155.
41. Veihmeyer, F.J., "Evapotranspiration", in Handbook of Applied Hydrology,
V.T. Chow, Editor, McGraw-Hill Book Company, New York,,1964.
42. Roesner, L.A., H.M. Nichandros and R.P. Shubinski, Water Resources
Engineers; A-D. Feldman and J.W. Abbott, The Hydrologic Engineering
Center; A.O. Friedland, Department of Public Works, City and County of
San Francisco, " A Model for Evaluating Runoff Quality in Metropolitan
Master Planning", ASCE Urban Water Resources Research Program, Technical
Memorandum No. 23, 345 East 47th Street, New York, New York, April
1974.
43. Raymond & May Associates, "Analysis & Plan, Volume II, A Master Plan
Report, City of Elizabeth, New Jersey", October 1968.
44. Metcalf & Eddy, Inc., Wastewater Engineering; Collection, Treatment,
Disposal, McGraw-Hill Book Company, New York, 1972.
45. U.S. Environmental Protection Agency, "Water Quality Studies", Water
Program Operations Training Program, ^NTIS-PB 237 586, May 1974.
46. McGauhey, P.H., Engineering Management of Water Quality, McGraw-Hill
Book Company, New York, 1968.
47. Heaney, J.P., W.C. Huber, and S.J. Nix, "Storm Water Management Model:
Level I - Perliminary Screening Procedures", U.S. EPA Report, EPA -
600/2-76-275, October 1976.
205
-------
SECTION XV
GLOSSARY, ABBREVIATIONS, AND SYMBOLS
GLOSSARY
Advanced combined sewer system - A combined sewer system which incorporates
flow routing and storage as well as flow conveyance in the design.
Types of storage that could be considered include in-pipe storage
and satellite storage.
Advanced pattern storms - Storms with peak rainfall occurring in the first
quarter of the rainfall period.
Base sewage flow - Average rate of dry-weather flow in the sewer system
after subtracting infiltration.
Combined sewer - A sewer receiving both storm runoff and municipal sewage.
Combined sewer overflow - Flow from a combined sewer in excess of the inter-
ceptor capacity that is discharged into a receiving
water.
Concentration pollutograph - A graph plotting pollutant concentration per
unit of time versus time.
Conventional combined sewer system - A combined sewer system without flow
control devices for routing combined sewage flows. Within the
context of this report, "conventional" implies the design of sewers
using flows derived from a synthetic hyetograph.
Conventional separate storm and sanitary sewer system - A dual sewer system,
one conveying stormwater only and the second conveying only
municipal sewage. Also see storm sewer and sanitary sewer.
Depression storage - Storm water retained in surface depressions' after
infiltration capacity has been exceeded.
District A - A drainage district of about 265 hectares (655 acres) in the
City of Elizabeth. A detailed study was made for this district
to develop runoff quantity and quality .characteristics using
SWMM and STORM for application to the entire City.
Diurnal variation - Hourly variation in the dry-weather flow quantity and
quality.
206
-------
Drainage district - The area which is drained by a collection system, and,
generally, to one outlet to the interceptor.
Dry-weather flow - That flow in sanitary or combined sewers that contains no
direct rainwater runoff.
Evapotranspiration - Water withdrawn from soil by evaporation and plant trans-
piration.
First flush - The initial rainwater runoff flows which may be expected
to contain high concentrations of the pollutants that were de-
posited in the sewer during the period antecedent to the rainfall.
5-year synthetic rainfall hyetograph - A synthetically generated hyetograph
which matches at all time intervals the intensity-duration curve
for a 5-year return interval prepared by the U.S. National Weather
Service. In the report, it is also called "5-year synthetic
storm".
5-year synthetic storm - See 5-year synthetic rainfall hyetograph.
Froude number - A numerical quantity used as an index to characterize the
type of flow in a hydraulic structure that has the force of gravity (as the
only force producing motion) acting in conjunction with the resisting force
of inertia. It is equal to the square of a characteristic velocity (the
mean, surface, or maximum velocity) of the system, divided by the product of
a characteristic linear dimension, such as diameter or depth, and the gravity
constant or acceleration due to gravity-all expressed in consistent units so
that the combinations will be dimensionless. The number is used in open-
channel flow studies or in cases in which the free surface plays an essential
role in influencing motion.
Hydrograph - A graph plotting flow rates versus time.
Hyetograph - A graph plotting rainfall intensities versus time.
Infiltration - The water entering a sewer system and service connections
from the ground, through such means as, but not limited to, de-
fective pipes, pipe joints, connections, or manholes. Infiltration
does not include, and is distinguished from, inflow.
In-pipe storage - Storage in sewer pipes to temporarily contain combined
sewage or urban runoff. In the design of an in-pipe storage
system, sewers are increased in size to provide volumetric storage
in addition to flow conveyance. Regulators are installed to permit
use of this storage to reduce peak flow rates reaching treatment
facilities.
In-system - Within the physical confines of the area served by the collection
system.
207
-------
Mass rate pollutograph - A graph plotting pollutant mass loading per unit
of time versus time.
Off-line storage - Storage so situated as to reduce peak flow rates and
pollutant concentrations entering or flowing in interceptor and
treatment facilities from the collection systems. Construction is
similar to that for satellite storage.
1.3-month synthetic storm - Derived from the 5-year synthetic storm in the
same manner as 1-year synthetic storm except the ratio of hourly
rainfalls with 1.3 months and five years return intervals is
used.
1—year synthetic storm — Derived from the 5-year synthetic storm by multiply-
ing the.5-year synthetic hyetograph rainfall rates by the ratio of
hourly rainfall rates for one year and five year return intervals.
Overflow event — Overflow occurs during a storm event. A storm event
is defined by the properties of the individual rainfall events,
such as duration and intensity, storm spacing, and time to empty
any storage facilities. An event was defined as beginning when
storage was first required and continued until the storage basin
was emptied. Any number of overflows occurring during the duration
of the event was considered as the same overflow event. If a
rainfall produces runoff that does not exceed the sewer capacity,
storage would not be utilized, and the rainfall would not register
as an event. In the case of zero storage, runoff from the water-
shed would directly drain to the intercepting sewer, and the number
of events as defined above would be the same as the number of
overflow events.
Pollutograph - A graph plotting pollutional concentrations or pollutional
mass per unit of time versus time.
Routing - Computing the downstream outflow hydrograph of an open channel
or a sewer from known values of upstream flow, and diversions to,
and withdrawals from storage.
Sanitary sewer - A sewer that carries wastewater from residences, commercial
buildings, industrial plants, and institutions, together with a
relatively low quantity of ground, storm, and surface waters that
are not admitted intentionally.
Satellite storage - Storage in the collection sewer system for peak flow
rate reduction and flow mixing. Satellite storage basins could be
open, covered or tunneled structures.
STORM default values - Numerical values assigned in the STORM computer
program. These values are used when they are not specified in the
input data. In this report, the default values refer, in par*-
ticular, to the factors which compute SS, BOD and the most probable
208
-------
number (MPN) of coliforms from the weight of dust and dirt washed
off the street.
Storm runoff - That stofmWater Which flows overland and enters
a receiving water body*
a sewer or
Storm sewer - A sewer that carries intercepted surface runoff, street wash
and other wash waters or drainage, but excludes domestic sewage
and industrial wastes.
Stormwater - The water resulting from a precipitation event which may stay on
the land surface, percolate into the ground, run off into a body of
watery enter a storm sewer or a combined sewer, infiltrate a
sanitary sewer, or evaporate.
Subarea - A subdivision of a drainage district based upon a single or one
predominant land use.
Subcatchment - A subdivision of a subarea generally by topography and pipe
network configuration and usually draining to one reach of pipe.
Surcharge - The flow condition occurring in closed conduits when the hy-
draulic grade line is above the crown of the sewer.
SWMM default values - Same meaning as STORM default values except that
the values are specified in the SWMM program*
Trunk sewer - A sewer that receives many tributary lateral sewer branches
and serves a large drainage area.
Weir-orifice regulator - A flow-regulating structure proposed for sewer
installation. The structure consists of a weir at an angle to the
main sewer line and a small branch sewer pipe to pass the normal
dry-weather flow and to control the wet-weather flow rates diverted
downstream. The weir provides a damming effect to store storm flow
in large trunk sewers or purposely enlarged lateral sewers.
209
-------
ABBREVIATIONS
Ac
APWA
BOD
cfs
cfs/acre
cm
DWF
ENR
EPA
ft
ft3
ft. MSL
ft/sec2
in/hr
kg
km
kg/day/lOOm
Ibs
Ibs/day
lbs/day/100 ft
Ibs/min
m
m3
o
m /min.
acre (s)
American Public Works Assocation
5-day biochemical oxygen demand
cubic feet per second
cubic feet per second per acre
centimeter (s)
Dry-weather flow
Engineering-News Record
U.S. Environmental Protection Agency
feet
cubic feet
feet above mean sea level
feet per second square
inch (es).per hour
kilogram (s)
kilometer (s)
kilograms per day per 100 meters
pounds
pounds per day
pounds per day per 100 feet
pounds per minute
meter (s)
cubic meter (s)
cubic meter (s) per minute
210
-------
m /sec
mg
MG
mgd
mg/1
min.
mm/hr
N
P04
SS
STORM
SWMM
cubic meter (s) per second
milligram (s)
million gallons
million gallons per day
milligram (s) per liter
minute (s)
millimeter (s) per hour
nitrogen
orthophosphates
Suspended solids
Storage, Treatment, Overflow,
Runoff Model as released by the
Hydrologic Engineering Center,
U.S. Army Corps of Enginners
Storm Water Management Model as
released by the U.S. Environmental
Protection Agency
211
-------
SYMBOLS
%
$
°F
8"
d/dt
percent
dollar
degree (s) Fahrenheit
8 inches
less than
greater than
partial differential
first order differential with respect
to time, t
second order differential with respect to time, t
degree (s) Centigrade
STORM runoff coefficient for pervious area
STORM runoff coefficient for impervious area
212
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-78-090
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND-SUBTITLE
CONVENTIONAL AND ADVANCED SEWER DESIGN CONCEPTS
FOR DUAL PURPOSE FLOOD AND POLLUTION CONTROL
A Preliminary Case Study, Elizabeth, New Jersey
5. REPORT DATE
May 1978 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
Herbert L. Kaufman and Fu-Hsiung Lai
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Clinton Bogert Associates
2125 Center Avenue
Fort Lee, New Jersey 07024
10. PROGRAM ELEMENT NO.
1BC611
11. QCXKDEDSKX/GRANT NO.
S-802971
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory—Gin.,OH
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
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer: Anthony N. Tafuri, Phone: 201-321-6679
16. ABSTRACT
Alternatives for pollution abatement from combined sewer overflows and stormwater
discharges were evaluated. Separate storm and sanitary, conventional combined, and
advanced combined systems with varying amounts of in-pipe and/or- satellite storage
and controlled flow routing were compared. Cost-effectiveness assuming a desired
effluent quality and new sewer system was determined. The effects on pollution
abatement and cost of changing various elements (collection system, interceptors,
storage and treatment works) of the system were investigated. SWMM and STORM were
employed to design sewers, analyze the quantity and quality of combined sewage and
stormwater runoff, and analyze a continuous 12-year, real rainfall record. The
overflow frequency, pollutants, and volume for 59 alternatives were determined. The
runoff and overflow characteristics developed for 265 hectares, were used to deter-
mine the characteristics from the remaining 1785 hectares to plan a citywide sewer
system. The study evaluated the effects of (a) differing rainfall patterns and
intensities on pollutant concentrations, (b) varying interceptor and treatment
capacity on pollutants discharged, and (c) peak flow equalizing for different levels
of treatment. The long-term pollutional loads from combined sewage overflows were
quantified as well as the effect on overflow control of varying the amount and
location of storage.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
*Rainfall, *Runoff, *Storm sewers, Com-
bined sewer, *Sanitary sewers, ^Overflows,
*Hydraulics, Hydrology, *Cost effective-
ness, *Mathematical models, Computer
programs, Methodology, *Surface water
runoff, *Water pollution, Water quality,
Water storage
Combined sewer overflow
management, In-system
storage, In-pipe storage,
Satellite storage, Off-
line storage, Hydrographs,
Pollutographs, Frequency
analysis
13B
13. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
229
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
213
*U.S. GOVERNMENT PRINTING OFFICE 1978— 757-140/1362
-------
-------
-------
-------
-------
CD
01
c: o
S- -~>
8
•ti
cr>
O
to
I
CO
O
VD
O
I o- 3
N §• 5 %
3: 5. f Sr
rr» *j .3 __
=o * -* 5'
ns°8
I \r> ^- s
-. O cb ct)
^* 3. cj 5-
i>. &. *
aj g |-S.
O ^t> (t) Cb
a -^ o. 01
.
*
* %
5 *
^ •^
3 2
3 3-
CO Q)
•"3
CD
§
Ql
?
Z
Ql
C3-
CB
Z T)
m
O -n
H 0)
c < c
z >
H m Z
< ._ m
O co
< °
m °
0)
n
CD
o
Tl
03 o
o c
o 3.
¥ O m
< =S-Z
o
O CD
13' 3-
O 03
(0
CD
QJ
CD
O o
5'
en
» JJ
i^. g
§° o m
CD O
1 ^^ O3 < |
00 O ;r O
O ? ^
CD ^ >
= ® ^
CD -* m
o
c
0)
O
------- |