EPA-670/2-74-040
December 1974
URBAN STORMWATER MANAGEMENT AND TECHNOLOGY:
An Assessment
By
John A. Lager
William G. Smith
Metcalf $ Eddy, Inc.
Palo Alto, California 94303
Program Element No. 1BB034
Contract No. 68-03-0179
Project Officer
Richard Field
Storm and Combined Sewer Section (Edison, N.J.)
Advanced Waste Treatment Research Laboratory
National Environmental Research Center
Cincinnati, Ohio 45268
NATIONAL ENVIRONMENTAL RESEARCH CENTER
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
For sale by the Superintendent of Document*, U.S. Government '
Printing Office, Washington, D.C. 20402
-------
FOREWORD
Man and his environment must be protected from the
adverse effects of pesticides, radiation, noise and other
forms of pollution, and the unwise management of solid
waste. Efforts to protect the environment require a focus
that recognizes the interplay between the components of our
physical environment--air, water, and land. The National
Environmental Research Centers provide this multidisciplinary
focus through programs engaged in
o studies on the effects of environmental
contaminants on man and the biosphere, and
o a search for ways to prevent contamination
and to recycle valuable resources.
Essentially every metropolitan area of the United States
has a stormwater pollution problem, whether served by a com-
bined sewer or a separate sewer system. This study provides
selected results of a comprehensive investigation assessment
of promising methods for the management of such pollution.
A. W. Breidenbach, Ph.D.
Director
National Environmental
Research Center, Cincinnati
11
-------
ABSTRACT
A comprehensive investigation and assessment of promising
completed and ongoing urban stormwater projects, representa
tive of the state-of-the-art in abatement theory and tech-
nology has been accomplished. The results, presented in
textbook format, provide a compendium of project information
on management and technology alternatives within a framework
of problem identification, evaluation procedures, and pro-
gram assessment and selection.
Essentially every metropolitan area of the United States has
a stormwater problem, whether served by a combined sewer
system (approximately 29 percent of the total sewered popu-
lation) or a separate sewer system. However, the tools for
reducing stormwater pollution, in the form of demonstrated
processes and devices, do exist providing many-faceted
approach techniques to individual situations. These tools
are constantly being increased in number and improved upon
as a part of a continuing nationwide research and develop-
ment effort. The most promising approaches to date involve
the integrated use of control and treatment systems with an
areawide, multidisciplinary perspective.
finnKWaS fub?ltJedJin fulfiHment of Contract Number
68-03-0179 by Metcalf § Eddy, Inc., Western Regional Office,
under the sponsorship of the Environmental Protection Agency
Work was completed as of December 1973.
REVIEW NOTICE
The National Environmental Research Center--Cincinnati has
reviewed this report and approved its publication. Approval
does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use
111
-------
When a city takes a bath, what
do you do with the dirty water?
PREFACE
The nationwide significance of pollution caused by storm-
generated discharges was first identified in the 1964 U.S.
Public Health Service's publication on the "Pollutional
Effects of Stormwater and Overflows from Combined Sewer
Systems.11 Congress, in recognizing this problem, authorized
funds under the Federal Water Pollution Control Act of 1965
and following legislation for the research, development, and
demonstration of techniques for controlling this source of
pollution.
The 1972 Amendments place new and stronger emphasis on
urban runoff as a source of pollution. "An accelerated
effort..." is stressed "...to develop, refine, and achieve
practical application of waste management methods applicable
to nonpoint sources of pollutants to eliminate the discharge
of pollutants including, but not limited to, elimination of
runoff of pollutants..." Construction grant applications
and areawide or basin wastewater treatment management plans
are being encouraged to include "...the necessary waste
water collection and urban storm water runoff systems..."
for the control and treatment of storm-generated pollution.
In the intervening years over 140 grants and contracts
totaling over $90 million have been awarded under the United
States Environmental Protection Agency (EPA) Storm and Com-
bined Sewer Technology Research, Development, and Demonstra-
tion Program. The federal government's share has been
approximately $46 million or 51.1 percent.
The objective of this text is to bring together the vast
amount of knowledge and technical data accumulated through
these projects and to present it in a format useful for
decision-makers, whether they be principally engineers,
administrators, or members of a concerned public.
To enhance professional and public awareness of the storm-
water problem and abatement programs, a 30-minute narrated
documentary color film has been produced as a part of
IV
-------
this project. The film, "Stormwater Pollution Control: A
New Technology," is available through the EPA for public
showings. F
It must be recognized that the research and demonstration
program is continuing and growing rapidly and that the
assessments made herein are based on present and past data
necessarily reflecting their limitations. It is believed '
however, that a valuable first step has been made in estab-
lishing a base upon which to direct future program emphasis,
to plan and conduct pollution evaluations, and to implement
effective abatement projects through construction and
management.
Representative complementary studies underway concurrently
include:
"Water Quality Management Planning for Urban Runoff"
(Contract 68-01-1846). This study will produce a guide
manual to assist planners in a "first-cut" assessment
ot the types and sources of urban runoff pollution, the
effects of land use, the applicability of abatement/
treatment measures, and potential environmental impacts
Ine use of nomographs, graphs, and tables is emphasized
"Engineering Aspects of Storm and Combined Sewer
Overflow Technology - A Manual of Instruction" (EPA
No. 801358). Under this project a manual, including
lecture outlines, for graduate course instruction in
an Environmental Engineering curriculum is being devel-
oped and tested as a model for program introduction at
universities throughout the United States.
Because several key terms are used extensively in this text
they are defined here. A complete glossary and list of
abbreviations is in Section XVII.
Combined sewage - A sewage containing both domestic sewage
and surface water or stormwater, with or without industrial
wastes. Includes flow in heavily infiltrated sanitary sewer
systems as well as combined sewer systems.
Combined sewer - A sewer receiving both intercepted surface
runoff and municipal sewage.
Combined sewer overflow - Flow from a combined sewer in ex-
cess oF tne interceptor capacity that is discharged into a
receiving water. 5
Intercepted surface runoff - That portion of surface runoff
that enters a sewer, either storm or combined, directly
through catch basins, inlets, etc.
-------
Municipal sewage - Sewage from a community which may be
composed of domestic sewage, industrial wastes, or both.
Nonsewered urban runoff - That part of the precipitation
which runs off the surface of an urban drainage area and
reaches a stream or other body of water without passing
through a sewer system.
Sanitary sewer - A sewer that carries liquid and water-
carried wastes from residences, commercial buildings, indus-
trial plants, and institutions, together with relatively low
quantities of ground, storm, and surface waters that are not
admitted intentionally.
Storm flow - Overland flow, sewer flow, or receiving stream
flow caused totally or partially by surface runoff.
Storm sewer - A sewer that carries intercepted surface run-
off,street wash and other wash waters, or drainage, but
excludes doemstic sewage and industrial wastes.
Storm sewer discharge - Flow from a storm sewer that is dis-
charged into a receiving water.
Stormwater - Water resulting from precipitation which either
percolates into the soil, runs off freely from the surface,
or is captured by storm sewer, combined sewer, and to a
limited degree sanitary sewer facilities.
Surface runoff - Precipitation that falls onto the surfaces
of roofs,streets, ground, etc., and is not absorbed or
retained by that surface, thereby collecting and running
off.
Urban runoff - Surface runoff from an urban drainage area
that reaches a stream or other body of water or a sewer.
VI
-------
CONTENTS
Abstract
Preface
Contents
Figures
Tables
Acknowledgments
Section
Part I - CONCLUSIONS, RECOMMENDATIONS,
AND INTRODUCTION
I CONCLUSIONS 3
The Stormwater Problem 3
Relative Magnitude 5
Mass Loadings §
Variability g
Design Considerations 3
Management Alternatives 9
Source Control 9
Collection System Control 10
Storage and Treatment H
Integrated (Complex) Systems 16
II RECOMMENDATIONS 18
Evaluations and Planning 18
Control 20
Treatment 20
Impact 22
vn
-------
CONTENTS (Continued)
Section Pa;
III INTRODUCTION 23
Objective and Scope 24
Method of Presentation 25
Background 26
1967 National Inventory 28
EPA Research, Development, and
Demonstration Program 29
Other Programs 30
Emerging Emphasis on the Stormwater Problem 31
1972 FWPCA Amendments 32
Comptroller General's Report 33
State and Local Requirements 33
Information Sources and Definition of Terms 33
Information Sources 33
Nomenclature and Terminology 34
Part II - FORMULATING AN APPROACH
IV GUIDE TO PROGRAM ASSESSMENT AND SELECTION 39
Preliminary Program Definition 39
Scope 40
Objectives 40
Problem Identification 41
Tributary Area 41
Sewerage Network 42
Characteristic Hydrology 42
Receiving Waters 50
Agencies 51
Data Assessment and Program Development 52
"First Cut" Analysis 52
Candidate Evaluation: Data Needs 55
Selection and Implementation of Final Plan 56
V THE STORMWATER PROBLEM 59
Emergence of the Stormwater Problem 59
Historical Development of Sewer Systems 59
The Stormwater Problem — Its Multiple
Facets 64
Characteristics of Stormwater 70
Quantities of Flow 70
Quality of Overflows and Discharges 73
Stormwater Pollution Loadings 83
Sources and Movement of Pollutants 88
Contaminant Sources 88
Collection and Transport 93
Vlll
-------
CONTENTS (Continued)
Section
Environmental Effects
Bucyrus, Ohio
Roanoke, Virginia
Beneficial Aspects of Wet-Weather
Discharges
VI EVALUTAION PROCEDURES AND CRITERIA
Data Collection 103
Sampling 104
Flow Measurement 112
Direct-Reading Quality Sensors 118
Analysis H9
Process and Equipment Evaluation 121
Method of Approach 123
Basic Design Data 124
Operational Results 125
Control Systems 126
System Monitoring 128
Remote Supervisory Control 128
Automatic Control 128
Costs 12Q
Data Sources 129
Updating and Transferability 130
Economies of Scale 131
Part III - MANAGEMENT ALTERNATIVES
AND TECHNOLOGY
VII SOURCE CONTROL 135
Quality Control Measures 135
Solid Waste Management 136
Street Cleaning 136
Use of Chemicals 138
Erosion Control 139
Quantity and/or Rate Control Measures 140
VIII COLLECTION SYSTEM CONTROL 145
Sewer Separation 145
General 145
Detailed Analysis 148
Conclusions 150
Infiltration/Inflow Control 152
Sources 152
Inflow Control 152
Infiltration Control 153
IX
-------
CONTENTS (Continued)
Section
Flushing and Polymer Injection 163
Flushing 163
Polymer Injection 165
Regulators 168
Conventional Designs 170
Improved Regulator Designs 175
Evaluation and Selection 183
Remote Monitoring and Control 184
System Components and Operations 186
System Control 186
Detailed Example . 189
IX STORAGE 196
Types of Storage Facilities 197
In-Line 197
Off-Line 200
Cost Data 213
X PHYSICAL TREATMENT WITH AND WITHOUT
CHEMICAL ADDITION 215
Sedimentation 215
Typical Combined Sewer Overflow
Sedimentation Facilities 217
Operation 218
Costs 219
Dissolved Air Flotation 220
Introduction 220
Design Criteria 222
Demonstration Projects 227
Advantages and Disadvantages 230
Costs 231
Screens 233
Introduction 233
Description of Screening Devices 234
Description of Demonstration Projects 248
Costs 252
Filtration 254
Introduction 254
Design Criteria 254
Demonstration Projects 254
Advantages and Disadvantages 256
Costs 256
Concentration Devices 258
-------
CONTENTS (Continued)
Section
Page
XI BIOLOGICAL TREATMENT 259
Introduction 259
Application to Combined Sewer Overflow
Treatment 260
Contact Stabilization 262
Description of the Process 262
Demonstration Project, Kenosha, Wisconsin 262
Advantages and Disadvantages 268
Trickling Filters 268
Description of Process 268
Demonstration Project, New Providence,
New Jersey 2?o
Advantages and Disadvantages 274
Rotating Biological Contactors 275
Description of Process 275
Efficiency 275
Operational Considerations 276
Design Parameters 279
Advantages and Disadvantages 279
Demonstration Project, Milwaukee, Wisconsin 281
Treatment Lagoons 282
Oxidation Ponds 287
Aerated Lagoons 290
Facultative Lagoons 296
Demonstration Projects
Costs
XII PHYSICAL -CHEMICAL SYSTEMS 307
Introduction 307
Unit Processes 308
Chemical Clarification 308
Chemical Recovery 3-j.O
Filtration 31Q
Carbon Adsorption 311
Activated Carbon Regeneration 314
Performance of Physical-Chemical Systems 315
South Lake Tahoe, California 317
Ewing-Lawrence Sewerage Authority,
Trenton, New Jersey 317
Rocky River, Ohio 317
Washington, D.C. 310
Dallas, Texas 318
Pomona, California 319
XI
-------
CONTENTS (Continued)
Section
Storm Flow Applications
Albany, New York
San Francisco, California
Kingman Lake, Washington, B.C.
Costs of Physical-Chemical Treatment
XIII DISINFECTION 326
Agents and Means 326
Chemical Agents 326
Physical Agents 331
Mechanical Means 332
Gamma Radiation 332
Combined Sewer Overflow and Storm Sewer
Discharge Disinfection 332
Chlorination 333
Alternatives to Chlorine Gas 336
Ozonation 342
Demonstration Projects 342
New Orleans, Louisiana 342
Ionics Hypochlorite Generator 343
Cottage Farm Storm Detention and Chlori-
nation Facility, Boston, Massachusetts 344
Onondaga County, Syracuse, New York 345
Philadelphia, Pennsylvania 346
Costs 346
Part IV - IMPLEMENTATION
XIV INTEGRATED SYSTEMS 353
Interfacing with Dry-Weather Facilities 353
Interfacing Unit Processes 356
Mathematical Modeling Techniques 357
Available Models 361
Application of Mathematical Models 363
Examples of Comprehensive Approach 363
Example of "First Cut" Approach 366
Master Plan Examples 367
San Francisco, California 367
Chicago, Illinois 370
Boston, Massachusetts 373
Seattle, Washington 374
Washington, D.C. 375
Environmental Compatibility 376
Satellite Facilities 377
Multiple-Use Facilities 379
Total Planned Communities 381
XII
-------
CONTENTS (Continued)
Section
Page
XV OPERATION AND MAINTENANCE 383
Operating Controls and Options 385
Sustaining (Dry-Weather) Maintenance 387
In-Line Facilities 388
Off-Line Facilities 391
Practices for Improved Operation and
Maintenance 393
Maintenance Costs 394
Support Facilities and Supplies 394
Safety 396
Solids Handling and Disposal 397
Estimating Solids Removal -zoo
Collection ^
Ultimate Disposal 402
Part V - REFERENCES, GLOSSARY,
AND CONVERSION FACTORS
XVI REFERENCES 4Q7
Cited References 4n7
Uncited References 423
XVII GLOSSARY, ABBREVIATIONS, AND CONVERSION FACTORS 436
Glossary
Abbreviations 436
Organizations 441
Symbols 441
Abbreviations 441
Conversion Factors 7,
445
XVIII LIST OF PUBLICATIONS
446
Xlll
-------
FIGURES
No. Page
1 Urban Stormwater Management Strategy 4
2 National Rainfall Frequency - Duration
Experience: 1-Year 1-Hour Rainfall 43
3 National Rainfall Frequency - Duration
Experience: 1-Year 24-Hour Rainfall 44
4 December 22, 1971 Frontal Storm Pattern,
San Francisco 46
5 Rainfall Isohyets for Heat Island
Showers, July 17-20, 1972, Washington, D.C. 47
6 Conceptual Computer Application for Master
Drainage Planning for Water Quality Control 57
7 Relative Use of Combined Sewers in the
United States 62
8 Relative Use of Combined Sewers by States 63
9 The Stormwater Problem - Combined Overflow
Residuals 65
10 The Stormwater Problem - Beach Degradation 66
11 Effectiveness of Interceptors of Different
Capacities for 1960, St. Louis, Missouri 72
12 Average Annual BOD5 Contributed to the
Roanoke River by Municipal Sewage 85
13 BOD5 Contributed to the Roanoke River by
Municipal Sewage During Maximum Yearly
Rainfall Event 86
14 Street Collapse Due to Infiltration 94
15 Typical Sampling Units and Installations 107
16 Model for Automatic Operational Control
System 127
xiv
-------
FIGURES (Continued)
No. n
Page
17 Construction Cost Trend 131
18 Stormwater Surface Detention Pond (Chicago) 142
19 Common Elements of an Interceptor and
Transport System 146
20 Packer-Sealer with Two Inflatable Sections 163
21 Polymer Injection Station for Sanitary
Sewer (Dallas) 167
22 Typical Early Type of Regulator 169
23 Typical Static Sewer Regulators 171
24 Typical Semiautomatic Dynamic Sewer Regulators 172
25 Typical Automatic Dynamic Sewer Regulator 174
26 Schematic Arrangement of a Fluidic Sewer
Regulator j76
27 Vortex Regulator 177
28 Solids Separation Action in the Swirl
Concentrator Hydraulic Model 179
29 Recommended Configuration for Swirl
Concentrator -^gQ
30 Spiral Flow Regulator 181
31 Elements of a Remote Control System (Seattle) 187
32 Typical Regulator Stations of In-Line Storage
Systems 5 199
33 Detention Basin Plan, Chippewa Falls 201
34 Schematic of Detection Facilities, Akron, Ohio 202
35 Jamaica Bay (Spring Creek Auxiliary Water
Pollution Control Plant) Retention Basin 204
36 Schematic of Humboldt Avenue Detention and
Chlorination Facility, Milwaukee, Wisconsin 205
37 Humboldt Ave. (Milwaukee) Retention Basin 206
38 Schematic of Cottage Farm Detention and
Chlorination Facility, Boston, Massachusetts 208
39 Cottage Farm (Boston) Detention and Chlori-
nation Facility 209
xv
-------
FIGURES (Continued)
No.
40 Deep Tunnel Concepts and Construction (Chicago) 211
41 Construction Cost Versus Design Capacity for
Sedimentation 221
42 Schematic of Dissolved Air Flotation Facilities
at Racine, Wisconsin 223
43 Dissolved Air Flotation Facilities (Racine) 224
44 Relationship Between Suspended Solids Removal
and Saturation Tank Pressure 227
45 Construction Cost Versus Design Capacity for
Dissolved Air Flotation, ENR 2000 232
46 Screenings from Mechanically Cleaned Bar Racks 235
47 Schematic of a Microstrainer or Drum Screen 236
48 SS Removal Versus Screen Opening 238
49 Rotary Fine Screen Schematic 244
50 Contact Stabilization Plant, Kenosha, Wisconsin 263
51 Combined Sewer Overflow Treatment by Contact
Stabilization (Kenosha) 264
52 Trickling Filter Plant Schematic, New Providence,
N.J. 272
53 Combined Sewer Overflow Treatment by High-Rate
Trickling Filtration (New Providence) 273
54 Rotating Biological Contactor 275
55 The Effect of Removals with Varying Contact
Times for Rotating Biological Contactors 278
56 Hydraulic Loading Versus Design 6005 Removals
for Rotating Biological Contactors 280
57 Relationship of Removal Efficiency to the
Organic Loading Rate for Rotating Biological
Contactors 283
58 Combined Sewer Overflow Treatment Lagoons 285
59 Typical Oxidation Pond Outlet with Three
Concentric Baffles to Reduce the Presence of
Algae in the Effluent 291
xvi
-------
FIGURES (Continued)
No. Pag(
60 Types of Aerators 292
61 Reactions Found in Facultative Lagoons 296
62 Two Pond Configurations to Promote Anaerobic
Zones and Decomposition of Settled Material 298
63 Schematic of Retention and Treatment Facilities,
Including Porposed Modifications, Mount Clemens,
Michigan 304
64 Flow Diagram of Clarification System 318
65 Physical-Chemical Treatment Pilot Plant,
Washington, D.C. 319
66 Process Flow Sheet for Physical-Chemical
Pilot Plant, Albany, New York 321
67 Ultraviolet Light Used as a Method of
Disinfection 332
68 Chlorination Control System, Oakwood Beach
Sewage Treatment Plant, New York City 33?
69 Stability of Sodium Hypochlorite Solution 339
70 Hypochlorite Generating Plant for Stormwater
Disinfection, New Orleans 340
71 Electrolytic Hypochlorite Generator 341
72 Schematic of Ozone Generation and Injection
for Microstrainer Facility 347
73 Methods of Interfacing Stormwater Facilities
with Existing Systems 355
74 San Francisco Master Plan Elements 369
75 Operational Schematic of the Recommended
Pollution Control Plan for Chicago 371
76 Baker Street Dissolved Air Flotation Facility,
San Francisco 373
77 Remote-Controlled Regulator Station, Seattle,
Washington 379
78 Conceptual Plot Plan of Kingman Lake Water
Reclamation Facility 380
79 Combined Sewer Overflow Collection and Treatment
Facility, Mount Clemens, Michigan 381
xvi i
-------
FIGURES (Continued)
No. Page
80 Stormwater Detention Facility (Boston): Before,
During, and After Storm Event 384
81 Control Panel at the Cottage Farm Detention and
Chlorination Facility (Boston) 385
82 Sludge Bank in Conner Street Sewer, Detroit 399
83 Flushing System, Red Run Drain, Detroit 400
XVlll
-------
TABLES
^ Page
1 Generalized Quality Comparisons of Wastewaters 6
2 Computed Annual Discharges to Receiving Water
for a Hypothetical City of 100,000 7
3 Summary of Storage Costs H
4 Comparison of Physical Treatment Alternatives 13
5 Comparison of Biological Treatment Alternatives 14
6 Management Alternatives Summary 53
7 Predominant Type of Sewer System in the 20
Largest U.S. Cities, 1900 and 1970 62
8 Comparison of Dry- and Wet-Weather Flows in
Separate Sanitary Sewers for Various Locations 68
9 Comparison of Annual Municipal Sewage Bypassed
or Overflowed for Various Cities 73
10 Comparison of Quality Characteristics from
First Flushes and Extended Overflows of a
Combined Sewer, Hawley Road Sewer, Milwaukee,
Wisconsin 76
11 Comparison of Quality of Combined Sewage for
Various Cities 78
12 Comparison of Quality of Storm Sewer Discharges
for Various Cities 80
13 Storm Sewer Discharge Quality from 15 Test Areas,
Tulsa, Oklahoma 81
14 Storm Sewer Discharge Quality from a 5-Square
Mile Urban Watershed, Castro Valley Creek,
California 82
15 Comparison of BOD5 from Combined Overflows,
Storm Discharges, and Snow Melt, Des Moines,
Iowa 83
xix
-------
TABLES (Continued)
No.
16 Average Annual BOD5 Contributed to the Roanoke
River by Municipal Sewage, Roanoke, Virginia 85
17 BOD5 Contributed to the Roanoke River by
Municipal Sewage During Maximum Yearly Rainfall
Event, Roanoke, Virginia 86
18 Estimated Annual Load of Pollutants Entering
the Area Receiving Streams, Tulsa, Oklahoma 87
19 Field Test Results of Catch Basin Source
Pollutants and Removals 92
20 Summary of Dry- and Wet-Weather Sandusky River
Analyses, Bucyrus, Ohio 99
21 Relative Concentrations of Pollutants During
Average Dry- and Average Wet-Weather Conditions,
Roanoke, Virginia 101
22 Representative Commercially Available Samplers 111
23 Typical Flow Measurement Applications 117
24 BOD5 Concentrations Versus Time from Start of
Overflow, District of Columbia 120
25 Street Sweeper Efficiency Versus Particle Size 137
26 Estimated Costs of Sewer Separation in
Various Cities 149
27 Sewer Separation Versus Conceptual Alternatives 151
28 Estimated Flushing Costs for Demonstration
Project, Detroit, Michigan 165
29 Installed Construction Costs and Annual
Operation and Maintenance Costs of Regulators 185
30 Typical CATAD Regulator Station Monitoring
Hourly Log 192
31 Typical CATAD Pump Station Monitoring Hourly
Log 193
32 Typical CATAD Water Quality Monitoring Hourly
Log 193
33 Typical CATAD Storm Log 195
xx
-------
TABLES (Continued)
Page
34 Summary of Storage Costs for Various Cities 214
35 Summary Data on Sedimentation Basins Combined
with Storage Facilities 216
36 Cost of Stormwater Sedimentation Facilities 219
37 Dissolved Air Flotation Design Parameters 225
38 Typical Removals Achieved with Screening/
Dissolved Air Flotation 228
39 Summary of Performance Characteristics, Baker
Street Dissolved Air Flotation Facility, San
Francisco, California 230
40 Dissolved Air Flotation Cost for 25 Mgd 231
41 Classification of Screens 233
42 Microstrainer and Drum Screen Installations
in the United States That Treat Combined
Sewer Overflows 237
43 Data Summary on Microstrainers and Drum
Screens . 242
44 Recommended Microstrainer Design Parameters
for Combined Sewer Overflow Treatment 244
45 Rotary Fine Screen Design Parameters 247
46 Characteristics of Various Types of Screens 249
47 Range and Level of Variables Tested 251
48 Cost of Microstrainers and Fine Screens for
25-Mgd Plants 253
49 Types of Filtration Processes Investigated
for Combined Sewer Overflows 255
50 Design Parameters for Filtration Mixed Media,
High Rate 257
51 Contact Stabilization Removals During 1972 265
52 Contact Stabilization Operating Parameters 267
53 Comparison of Low-Rate, High-Rate, and
Ultrahigh-Rate Trickling Filters 269
54 Trickling Filter Removals, New Providence,
New Jersey 274
xxi
-------
TABLES (Continued)
No. Page
55 Rotating Biological Contactors --Variations in
Removals Resulting from Flow Increases of
10 Times Due to Wet Weather 277
56 Rotating Biological Contactor Design
Parameters 280
57 Removals Reported in Pilot-Plant Studies of
Rotating Biological Contactors, Milwaukee,
Wisconsin 282
58 Comparison of Different Types of Lagoons
Treating Storm Flows for Various Cities 284
59 Removal Efficiencies of Treatment Lagoons
for Various Cities 286
60 Methods of Reducing Algae in the Oxidation
Pond Effluent 290
61 Oxidation Pond Design Parameters 291
62 Types of Aeration Equipment for Aerated
Lagoons 293
63 Aerated Lagoon Design Parameters 295
64 Facultative Lagoon Design Parameters 299
65 Capital and Operation and Maintenance Costs
for Biological Treatment 305
66 Cost for Oxidation Ponds to Treat 25 Mgd of
Overflow 306
67 Achievements of Chemical Clarification 309
68 Illustrations of Varying Media Design for
Various Types of Floe Removal 312
69 Examples of Filter Performance 312
70 Operating Data of Physical-Chemical Treatment
Plants for Various Cities 315
71 Performance Data of Physical-Chemical
Treatment Plants for Various Cities 316
72 Estimated Capital and Operation and
Maintenance Costs for Typical Physical-
Chemical Treatment Plant 324
xx 11
-------
TABLES (Continued)
No. D
Pag
73 Removal or Destruction of Bacteria by Different
Wastewater Treatment Processes 327
74 Comparison of Ideal and Actual Chemical
Disinfectant Characteristics 328
75 Comparison of Data on Disinfection Devices 335
76 Chemical Disinfection Agents and Sources Used
by Various Cities for Combined Sewer Overflows
and Storm Sewer Discharges 338
77 Cost Data on Chlorine Gas and Hypochlorite
Disinfection 34g
78 Estimated Costs of Tertiary Treatment Plants
Using Ozone 349
79 Comparison of Estimated Capital Costs for
3 Different Disinfection Methods 350
80 Interfacing Storm Flow Facilities with
Existing Sewerage System 353
81 General Interfacing Between Types of Storage
and Treatment Devices
82 General Data Requirements, Stormwater
Management Model
83 Comparison of Master Plans and Projects in
Various Cities 368
84 Examples of Combined Sewer System Operation
Controls 387
85 Sludge Production and Solids Disposal Methods
for Various Treatment Processes 401
XXlll
-------
ACKNOWLEDGMENTS
A great deal of cooperation has been received during the
conduct of this project. Metcalf § Eddy, Inc., gratefully
acknowledges the cooperation and direct participation of key
personnel from the headquarters and regional offices of the
EPA and of all the municipalities contacted and their
consultants.
Especially acknowledged is the leadership and assistance of
Richard Field, Chief of the Storm and Combined Sewer
Technology Branch, National Environmental Research Center--
Cincinnati, EPA, Edison, New Jersey, who was the Project
Officer; and of Anthony N. Tafuri, Staff Engineer.
Project assistance and report reviews were provided by
Dr. George Tchobanoglous, Consultant, and Marcella Tennant,
Technical Editor. Consultants Henry S. Brie-trose and Ronald
Alexander advised on the film contracting and general
content.
Film direction and production was by Veriation Films, Palo
Alto,-California (Robert Moore, Director; Stephen Longstreth,
Writer; and David Espar, Photography).
This project was conducted under the supervision and direc-
tion of John A. Lager, Project Director, and William G.
Smith, Project Engineer. Portions of the report were
written by Howard L. Hoffman, Michael K. Mullis, and
C. David Tonkin. Denny K. Tuan assisted in the literature
search.
xxiv
-------
Part I
CONCLUSIONS, RECOMMENDATIONS,
AND
INTRODUCTION
-------
-------
Section I
CONCLUSIONS
Control and/or treatment of storm sewer discharges and com-
bined sewer overflows is a major problem in the field of
water quality management. Over the past decade much re-
search effort has been expended and a large amount of data
!£% ?n*Pnieircte£* ?rimarily through the actions and sup-
port ot the U.S. Environmental Protection Agency's fEPAl
Storm and Combined Sewer Technology Research, Development
and Demonstration Program. '
A comprehensive investigation and assessment of promising
completed, and ongoing projects, representative of the
state-of-the-art in abatement theory and technology has
been accomplished in this study. The results, reported
herein, are presented as a compendium of project information
on management and technology alternatives within a framework
ot problem identification, evaluation procedures, and pro-
gram assessment and selection.
Critical elements in developing a strategy for urban storm-
water management are shown on Figure 1. Highlights of the
more important assessments and data with respect to these
elements developed or exposed during the course of the study
are presented in the following conclusions.
THE STORMWATER PROBLEM
• Essentially every metropolitan area of the United
States has a stormwater problem, whether served by a
combined sewer system (approximately 29 percent of the
total sewered population) or a separate sewer system.
• The problem is best quantified when discharges are com-
pared on the basis of mass loadings released over dis-
crete periods of time encompassing one or several
consecutive storm events. In many cases, however
aesthetics or beneficial uses (such as maintaining
-------
[=>
[=>
FORMULATE APPROACH
• OBJECTIVES
• PROBLEM IDENTIFICATION
• SYSTEM EVALUATION
• DATA BASE
V
SELECT MANAGEMENT ALTERNATIVE
• ESTABLISH CONTROL
• MONITORING
• STORAGE
• AUTOMATION
• TREAT TO OBJECTIVE LEVELS
• SOURCE
• SYSTEM
• OFF-LINE
• DUAL USE
V
IMPLEMENT AND ASSESS PERFORMANCE
• PILOT/PROTOTYPE
• OPERATION AND MAINTENANCE
• MONITORING AND FEEDBACK
OF INFORMATION
Figure 1. Urban stormwater
management strategy
-------
•
receiving water quality above body contact use stand-
ards J are of primary concern.
Each metropolitan area should therefore be directly in
volved in setting its goals and objectives for embark-
ing upon a stormwater management program.
De^top analyses alone necessarily have limited appli-
cability, and a locally developed real data base is
essential. Recognizing that the effort will require
£0m-,5nu t(? !everal years, sampling and monitoring
should be initiated at the earliest practicable time.
The tools for reducing stormwater pollution, in the
torm of demonstrated processes and devices, do exist
providing many-faceted approach techniques to individ-
ual situations. These tools are constantly being in-
creased in number and improved upon as a part of a
research and development program guided by
Relative Magnitude
• The stormwater problem involves three types of dis-
charges: (1) combined sewer overflows, (2) surface
runoff either collected separately or occurring as
nonsewered runoff, and (3) overflows of infiltrated
municipal sewage resulting from precipitation.
• Of the three types of discharges, the combined sewer
overflows and overflows of infiltrated municipal sewage
have similar characteristics, with 5-day biochemical
oxygen demand (BOD5) loadings averaging approximately
one-half the strength of untreated domestic sewage.
This contrasts with surface runoff from urban areas
which has a BOD5 approximately the strength of second-
ary effluent. A generalized comparison of the maior
quality parameters is summarized in Table 1.
t It has been suggested in several studies that chemical
oxygen demand (COD) is a more representative measure of
comparison, but its widespread adoption is handicapped
by its limited data base. COD has the advantages of
simpler testing procedures and greater resistance to
toxic materials found in urban runoff.
> The major constituent of street runoff has been found
to be suspended and settleable solids—primarily inor-
ganic, mineral-like matter, similar to sand and silt.
Along with this material there is organic matter algal
-------
Table 1. GENERALIZED QUALITY COMPARISONS
OF WASTEWATERS
Type
Untreated municipal
Treated municipal
Primary effluent
Secondary effluent
Combined sewage
Surface runoff
mg/1
200
135
25
115
30
mg/1
200
80
15
410
630
MPN/100 ml
5 x 107
2 x 107
1 x 103
5 x 106
4 * 105
mg/1 as N
40
35
30
11
3
mg/1 as P
10
8
5
4
1
Source: Data condensed from Tables 12 and 13, Section V. Values based upon flow-
weighted means in individual test areas.
nutrients, coliform bacteria, heavy metals, and
pesticides. Significantly, the latter are largely con
centrated in the very fine material (< 43 microns)
limiting the pollution abatement effectiveness of con-
ventional street sweeping operations and catch basins.
Bacterial contamination of combined sewage is
"typically" 1 order of magnitude lower, and surface
runoff 2 to 4 orders of magnitude lower, than that of
untreated municipal sewage. Significantly, however,
the concentrations are 2 to 5 orders of magnitude
higher than those considered safe for water contact
activities.
Mass Loadings
Mass loadings represent the product of quantity and
concentrations. They are of prime importance in storm
water analyses because of the high volume of storm in-
duced flows. For example, rates of urban runoff from
an average storm intensity of 0.25 cm/hr (0.10 in./hr)
may be expected to be about 5 to 10 times the dry-
weather flow contribution from the same area.
Similarly, a not uncommon rainfall intensity of 2.5
cm/hr (1.0 in./hr) will produce flow rates of 50 to
100 times the dry-weather flow.
-------
The relative impact of the stormwater problem in terms
of mass loadings (in this case on an annual basis If
illustrated m the hypothetical comparison in Table 2
The assumptions represent a hybrid of statistical data
on municipalities and are thought to be, in a general
sense, representative. In Case 1 the pollution as
measured by BOD5 is dominated by combined sewer over-
flows, assuming secondary treatment to be in effect for
dry-weather flows. On the other hand, if only primary
treatment were provided, the positions would bVre 7
versed with dry-weather discharges accounting for
approximately 65 percent of the annual loading. Case 2
is an illustration of what might be accomplished
through a sewer separation program: the reduction of
the annual BOD5 discharged by approximately 44 percent
(SS)Wdischaraed°mPanying increase in suspended solids
Table 2. COMPUTED ANNUAL DISCHARGES TO RECEIVTNr
WATER FOR A HYPOTHETICAL CITY OF 100,000
''« ^"ipUon J}0.,^ B0»s SS
1,000 Ib/yr
Case 1 - Sewered on combined plan
Municipal effluent
Dry weather 3,600
Intercepted wet weather 300
Combined sewer overflows 2,100
Total 6,000
Case 2 - Sewered on separate plan
Municipal effluent
Dry weather 3>600 75Q 45Q
Infiltrated municipal overflows 300 290 l 020
Storm sewer discharges 2,100 530 11,030
T°tal 6»°°0 1,570 12,500
Assumptions
Population density - 20 persons/acre
Degree of treatment of municipal flows - secondary flOO
Average annual rainfall - 33 in. during 760 hr (
Runoff coefficient - 0.5
Flow-weighted mean concentrations - from Table 1
Effective interceptor/treatment capacity = 2.0 x dry-weather flow
Note: mil gal. x 3.785 = Ml
Ib x 0.454 = kg
acre x 0.405 = ha
in. x 2.54 = cm
-------
• Since combined sewer overflows represent a comingling
of surface runoff and municipal sewage, the difference
between the BODs released in Case 1 and that in Case 2
represents (1) direct losses of municipal sewage and
(2) scouring of pollution from the combined collection
system. Assuming 90 percent of the municipal sewage is
lost to the receiving waters without treatment during
periods of rainfall, this loss would account for 7.8
percent (0.9 x 760/8,760 = 0.078) of the annual BODs
loading in the untreated municipal sewage, or approxi-
mately 216,000 kg/yr (475,000 Ib/yr). The remaining
347,000 kg/yr (765,000 Ib/yr) would therefore be
attributed to the scoured deposits, frequently termed
the "first flush" effect.
Variability
• Stormwater characteristics are highly variable, exhibit-
ing changes of 10 to 1 or more in a single storm, from
area to area and from one storm to the next.
• Higher concentrations of pollutants are generally ex-
pected under the following conditions: the early
stages of a storm (including so-called first flush
effects); in more densely settled and/or industrialized
areas; in response to high rainfall intensities; after
prolonged dry periods; and in areas with construction
activities (inorganic solids). Conversely, concentra-
tions tend to be lower as a storm progresses and in the
latter storms of a closely spaced series.
• The storm path and collection system configuration may
have an additional pronounced influence on the quality
of combined sewer overflows. On the basis of system
hydraulics and regulator efficiencies, discharge mix-
tures from neighboring outfalls may vary simultaneously
from totally raw municipal sewage to dilute surface
runoff.
DESIGN CONSIDERATIONS
• Because of the intermittency and extreme variability of
combined sewer overflows or storm sewer discharges in
both flows and loadings, there is no such thing as an
"average" design condition for stormwater treatment and
control facilities. Therefore, a unit operation or
process that performs only when conditions are right
may be too restrictive for practical applications.
• Simplified mathematical models based upon the general
storage equation and operated off real (continuous)
-------
rainfall data provide an excellent tool for equating
the effectiveness of alternative storage volumes and
treatment rates.
• Although precise characterization is virtually impossi-
ble, nominally accurate flow measurement (say 5 to 15
percent) and direct-reading remote sensing of quality
characteristics are necessary adjuncts to integrated
system design and management.
• The magnitude, debris content, and brute force of storm
flows are major design constraints that limit options
for centralization (because of high transmission costs)
and the applicability of sophisticated and complex
equipment. v
• In assessing the performance of past projects the
project scale (both overall capacity and equipment)
pretreatment controls, and location with respect to
maintenance and supervisory services are critical fac-
tors in the credibility, applicability, and reliability
ot the processes.
• Control and treatment of stormwater introduce many
unique operation and maintenance requirements These
include automated control, startup and shutdown proce-
dures, maintenance and surveillance between storms, and
solids handling and disposal.
• Control and treatment facilities should be designed to
permit total system test operations during dry-weather
periods without incurring adverse discharges to the
receiving waters.
• Programs must be scaled and implemented with full
recognition given to the limitations in the available
data and the relative immaturity of the state-of-
the-art.
MANAGEMENT ALTERNATIVES
Management alternatives for stormwater pollution abatement
in this text have been categorized into four application
areas: source control, collection system control, storage
and treatment, and integrated (complex) systems.
Source Control
• Source controls are defined as those measures for re-
ducing stormwater pollution that involve actions within
the urban drainage basin before runoff enters the
-------
sewer system. Examples include surface flow attenua-
tion, use of porous pavements, erosion control, chemi-
cal use restrictions, and improved sanitation practices
(street cleaning, more frequent refuse pickup, etc.).
• Benefits may include reduced costs per quantity of
pollutant removed, improved neighborhood cleanliness,
reduced flooding, area beautification, etc.
• In a recent survey, the practice of surface detention
for urban runoff control was concluded to be a very
effective means of handling runoff, reducing downstream
flooding and drainage costs, and, often, improving
aesthetics and recreational facilities.
• Demonstration of the effective incorporation of source
controls in urban planning is found in Columbia,
Maryland; Du Page County, Illinois; and a planned com-
munity being developed near Houston, Texas.
Collection System Control
• Collection system control, as used in this text, in-
cludes all alternatives pertaining to collection system
management or alteration. Examples include inflow/
infiltration control, the use of improved regulator
devices, temporarily increased line-carrying capacities
using polymer (friction reducing) flow additives, sewer
separation, and the use of remote monitoring/control
systems.
• Detailed knowledge of how collection systems respond to
wet-weather flows is almost universally lacking in
municipalities today. As a result, demonstration proj-
ects frequently reveal relief points and crossovers
critical to proper functioning that were not known to
exist prior to the project. Such conditions place
great emphasis on the need for early and intensive
monitoring and modeling for predictive responses.
• System controls using in-line storage represent promis-
ing alternatives in areas where conduits are large,
deep, and flat (i.e., backwater impoundments become
feasible), and interceptor capacity is high. Reported
costs for storage capacity gained in this manner range
from 10 to 50 percent of the cost of like off-line
facilities. Because system controls are directed
toward maximum utilization of existing facilities,
they rank among the first of alternatives to be
considered.
10
-------
It is reported that up to 45 percent flow augmentation
was achieved through the use of friction reducing addi-
tives in a closed conduit during periods of peak runoff,
thereby eliminating overflows of combined sewage.
The concept of constructing new sanitary sewers to
replace existing combined sewers largely has been aban-
doned because of the enormous cost, limited effective-
ness, inconvenience to the public, and extended time
required for implementation.
Storage and Treatment
Storage —
Storage is perhaps the most cost-effective method
available for reducing pollution resulting from over-
tlows of combined sewage and to improve management of
urban runoff. It is the best documented abatement
measure in present practice. Representative project
costs are listed in Table 3.
Table 3. SUMMARY OF STORAGE COSTS
Location
Oak Lawn, 111.
Seattle, Wash.
Chippewa Falls, Wis.
Jamaica Bay, N.Y.
Milwaukee, Wis.
Akron, Ohio
Boston, Mass.
Chicago, 111.
Type
Surface detention
In-line
Open, lined basin
Covered basin
Basin plus sewer
Buried basin
Buried-void space
Buried, short detention
Open quarry
Tunnels and appurtenances
Capacity ,
mil gal.
53.7
32.0
2.8
10.0
23.0
4.0
0.7
1.3
2,736.0
2,834.0
Cost
$/gal.
0.03
0.23
0.26
2.12
0.92
0.50
0.62
4.74b
0.21
0.27
a. ENR = 2000.
b. Includes influent pumping station, chlorination facilities
3. nd O
3. nd
Source: Table 34, Section IX.
Note: mil gal. x 3.785 = Ml
$/gal. x 0.264 = $/l
11
-------
• Storage facilities possess many of the favorable attri-
butes desired in stormwater treatment: (1) they may be
used to provide flow equalization or attenuation and,
in the case of tunnels, flow transmission; (2) they
respond without difficulty to intermittent and random
storm behavior; (3) they are relatively unaffected by
flow and quality changes; and (4) frequently, they can
be operated in concert with regional dry-weather flow
treatment plants for benefits during both dry- and wet-
weather conditions.
• Storage facility variations include concrete holding
tanks, open basins, tunnels, underground and underwater
containers, underground "silos," granular packed beds
(void space storage), and the use of abandoned facili-
ties and existing sewer lines.
• Disadvantages of storage facilities include their large
size, high cost, and dependency on other treatment
facilities for processing the retained water and
settled solids.
Physical Treatment With and Without Chemical Addition —
• Physical treatment processes are well suited to storm-
water applications in many ways, particularly with
respect to SS removal. These processes include sedi-
mentation, dissolved air flotation, screening (coarse,
fine mesh, and micro), filtration, and swirl
concentration. Representative cost, loading, and effi-
ciency data are shown in Table 4.
• Removal efficiencies for the physical processes tend to
vary directly with the influent contaminant concentra-
tions yielding relatively constant effluent results.
Thus, instantaneous removal efficiencies lose their
significance; an improved basis of comparison is mass
loadings applied versus those discharged.
• While the availability of full-scale field data for
storm flow applications is limited at the present time,
projects now operating or under construction will soon
greatly enhance the data base available to designers.
Notable are the dissolved air flotation facilities in
Racine, Wisconsin, and San Francisco, California; the
chemical clarification facilities in Dallas, Texas;
and two projects providing parallel full-scale testing
of alternative screening units in Fort Wayne, Indiana,
and Syracuse, New York. Screening devices to be
tested include fine screens, rotary fine screens, hy-
draulic sieves (static devices), and microstrainers.
12
-------
Table 4. COMPARISON OF PHYSICAL
TREATMENT ALTERNATIVES
Process
Estimated
removal, I
Design loading Capital cost
rate BODr SS
Dissolved air flotation15
(split flow) 3,200 gpd/sq ft 40 60 35,000
Dissolved air flotation13
(split flow) with
chemical addition 25 mg/1 52 78
Microstrainer
(20-60 microns) 30 gpm/sq ft 10-50 70 11,000
Fine screen
(100-600 microns) 50 gpm/sq ft 15 40 7,800
Ultrahigh rate^
filtration 24 gpm/sq ft 40 65 63,000°
Ultrahigh rate'3
filtration with poly-
electrolyte addition 1-2 mg/1 65 94
Chemical clarification
using waste lime sludge 2,570 gpd/sq ft 60 60 54,000
a. Based on 25-mgd facility.
b. Includes ultrafine screens as pretreatment.
c. Extrapolated from pilot scale data.
Note: gpd/sq ft x 0.283 = cu m/min/ha
gpm/sq ft x 0.679 = 1/sec/sq m
mgd x 3.785 = Ml/day
Concentration devices, typified by the swirl concen-
trator and helical or spiral flow devices, have intro-
duced an advanced form of sewer regulator--one capable
of controlling both quantity and quality. These de-
vices take advantage of secondary fluid motions and
natural liquid/solids separation in bends and other
forms of rotational flow to split storm flow into a
low volume concentrate and a high volume, relatively
clear stream. Prototype swirl regulators are under
construction in Lancaster, Pennsylvania, and Syracuse,
New York. A third unit has been placed into operation
±or sewage grit separation in Denver, Colorado. Set-
tleable solids removals ranging from 65 to more than
90 Percent, corresponding to chamber retention times
in the order of 5 to 15 seconds, have been predicted on
™*+ i SJoo?/ modetl tes?s' Indicated costs are approxi-
mately $285/cu m/sec ($6,500/mgd). Swirl devices are
being developed for primary treatment; having the poten-
tial for more effective treatment than conventional
sedimentation at only l/12th the detention time.
13
-------
Biological Treatment —
• Biological treatment of wastewater, used primarily for
domestic and industrial flows, produces an effluent of
high quality at comparatively low cost. For treatment
of storm flow, however, the following are serious draw-
backs: (1) the biomass used to assimilate the waste
constituents must either be kept alive during times of
dry weather or allowed to develop for each storm event;
and (2) once developed, the biomass is highly suscep-
tible to washout by hydraulic surges and organic
overload.
• Examples of biological treatment applications to storm-
water include (1) the contact stabilization modifica-
tion of activated sludge, (2) high-rate trickling
filtration, (3) bioadsorption using rotating biological
contactors, and (4) oxidation lagoons of various types.
The first three are operated conjunctively with dry-
weather flow plants to supply the biomass, and the
fourth use approaches total storage of the flows (rep-
resentative detention times of 1 to 10 days). A com-
parison of the alternatives is shown in Table 5.
Table 5. COMPARISON OF BIOLOGICAL TREATMENT
ALTERNATIVES
Process and design flow
Contact stabilization
modification of
secondary plant (20 mgd)
High rate trickling
filter modification
of secondary plant
(6 mgd)
Rotating biological
contactor^5 (10.4 mgd)
Treatment lagoons
(0.3 to 2.2 mgd)
Design loading
rate
108 Ib BOD5/
1,000 cf
Plastic media,
40 mgad; rock
media, 12 mgad
5-15 Ib BOD5/
day/1,000 sq ft
10 days'
retention
Estimated
removal , 1
BOD5 SS
83 92
65 65
54 70
40-90 50-60
Capital cost,
$/mgd
78,300
79,100
30,000
43,000 to
55,000
a. Data based on single installations.
b. Includes cost of final clarifier.
Note: lb/1,000 cf x 16.077 = g/cu m
mgad x 9.357 = 1/day/ha
lb/day/1,000 sq ft x 4.88 = kg/day/1,000 sq m
mgd x 3.785 = Ml/day
14
-------
• Biological process applications severely restrict
allowable flow and/or loading variations. For example
in the projects cited in Table 5, the contact stabili-
zation modification was limited to 2 to 3 times the
dry-weather flow; the trickling filter application, to
5 to 10 times the dry-weather flow; and the biological
contactor, to 5 to 10 times the dry-weather flow.
• Dual-use principles were effectively demonstrated in
the first two cases in Table 5 in that the wet-weather
plant additions were used to advantage during dry-
weather periods for upgrading treatment performance.
This capability should be exploited in all storm flow
management strategies.
Physical-Chemical Treatment -
• Physical-chemical treatment is becoming competitive in
cost with biological treatment, especially where sig-
nificant phosphorus removal is required. Physical-
chemical processes are of particular importance to
storm flow treatment because of their adaptability to
automated operation, rapid startup and shutdown charac-
teristics, and very good resistance to shockloads.
• The most promising combination of unit processes
appears to be chemical clarification followed by fil-
tration and adsorption on activated carbon.
• Drawbacks to physical-chemical treatment include high
initial costs, high rates of chemical consumption, and
increased sludge (by dry weight) to be disposed of.
• A unique variation of the usual coagulation-adsorption
process was applied in a pilot demonstration in Albany,
New York. Both powdered carbon and coagulants were
added in a static mixing-reaction pipeline, and the
resultant coagulated matter was flocculated downstream
separated by tube settlers, and polished by multi-media
filtration. BODs and SS removals of 94 and 99 percent
respectively, were achieved. A prototype facility was*
estimated to cost $39,000/M1 ($146,000/mgd) based upon
a capacity of 94 Ml (25 mgd).
Disinfection —
• Because the disinfectant and contact demands are great
in storm flow applications, current research centers on
(1) high-rate applications, (2) use of alternative dis-
infectants with reduced toxicity residuals and high
15
-------
reaction rates, and (3) on-site generation of
disinfectants.
• It was recently reported that satisfactory disinfection
of a combined sewer overflow was attainable with con-
tact times as short as 2 to 4 minutes (versus conven-
tional practice of 10 to 15 minutes) using a chlorine
concentration of 5 mg/1. These results were obtained
on microstrainer effluent in a specially designed con-
tact tank using parallel corrugated baffles to provide
high mixing intensities.
• Effluent bacteria requirements of 1,000 total coliforms/
100 ml and 200 fecal coliforms/100 ml were reached in
another bench-scale application on microstrainer efflu-
ent in contact times of 60 seconds or less using 25
mg/1 chlorine or 12 mg/1 chlorine dioxide in a single-
stage disinfection process, or 8 mg/1 chlorine and
2 mg/1 chlorine dioxide in a two-stage disinfection
process.
• Other disinfectants under study in the stormwater pro-
gram include sodium hypochlorite and ozone.
• An on-site sodium hypochlorite generation plant has
been constructed in New Orleans capable of producing
75,700 1 (20,000 gal.) of hypochlorite per 8-hour day.
Estimated production cost is $0.05/kg ($0.12/lb) of
available chlorine.
Integrated (Complex) Systems
• The most promising approaches to urban storm flow man-
agement involve the integrated use of control and
treatment systems with an areawide, multidisciplinary
(water use, land use, wet- and dry-period discharges,
etc.) perspective.
• Storm flow treatment processes can be most effectively
used following some form of storage (flow equalization).
This yields not only longer running periods, reduced
shock effects, and buffer flexibility for startup and
shutdown, but also, frequently, lower overall costs.
• Resort to mathematical model usage is ultimately essen-
tial to problem comprehension and effective program
implementation because of the myriad conditions and
alternatives encountered. Model use in the initial
program planning process is likewise highly beneficial.
16
-------
Mathematical models have been developed and success-
fully applied, at many levels of sophistication and
complexity, to the solution of stormwater management
problems. These models vary from simple "mass-diagram"
balances of runoff, storage, treatment, and overflows
to those capable of representing the whole gamut of
urban stormwater runoff phenomena (including both
quantity and quality).
Programs developed in the conception/design phases of
a project can form the basis for real-time decision-
making in prototype operations to maximize counter-
measures to abate pollution.
In San Francisco studies it was concluded that combined
sewer overflows occur during rainfall intensities
greater than 0.02 in./hr (90 percent of all storm time)
It was further concluded, however, that the concept of
constructing combined sewers should be retained and
that storage and treatment facilities should be con-
structed so that no more than 8 overflows will occur in
each year. By such action, control would be achieved
on up to 90 percent of the annual combined sewer over-
flow discharges at a cost of $42,400/ha ($17,ISO/acre).
This represents approximately 40 percent of the esti-
mated cost of separating existing combined sewers.
Integrated approaches are notably demonstrated by the
comprehensive master plans developed in Chicago and
San Francisco; by the remote monitoring and control
systems of Seattle, Minneapolis-St. Paul, and Detroit;
by the beneficial-use oriented programs of Mount
Clemens, Michigan; the Kingman Lake reclamation concept
for Washington, D. C.; and the planned community ap-
proaches at Woodlands, Texas, and Columbia, Maryland--
to name a few. All of these approaches demonstrate
clearly man's capability to meet the challenge of this
new technology and, in so doing, greatly improve his
habitat.
17
-------
Section II
RECOMMENDATIONS
Recommendations, developed during this study to guide future
actions, are grouped into four categories dealing with:
(1) evaluations and planning, (2) control, (3) treatment,
and (4) impact.
EVALUATIONS AND PLANNING
• Standardized procedures should be developed and imple-
mented in the characterization of wet-weather dis-
charges and control/treatment process evaluations.
• Greater emphasis now should be placed upon the total
impact of projects as opposed to optimizing devices for
a presumed set of criteria. In this regard, four pri-
mary guides are recommended for weighing future per-
formance evaluations:
1. Reliability/durability based upon the frequency
of total and partial unit operations to the
total number of continuous storm events.
2. Efficiency as measured by the total mass of
pollutants removed by the facility as a per-
centage of the total mass applied over the com-
plete storm event including upstream bypasses.
3. Effluent quality as measured by the average
and maximum concentrations measured in the
discharge.
4- Dual use as measured by the effective utiliza-
tion of the facility during non-storm periods.
• Development and application of direct-reading remote
sensors should be stressed, particularly with respect
to organic and solids concentrations. Correlations
18
-------
should be developed between remotely sensed parameters
and commonly used design and system operation
parameters.
Improved capability for remotely sensing flow rates in
large conduits in an economic manner should be
developed. Also, improved portable flowmeters and flow
proportional samplers for limited studies and infiltra-
tion analyses are a priority need. Units should be
capable of measuring both open channel and surcharged
flows and should not be damaged by sitting idle for ex-
tended periods when there is no discharge or by charac-
teristic debris and should not obstruct flow.
Planning processes for stormwater management should en-
compass the three-phase strategy program outlined in
Section I (Figure 1): (1) formulating an approach,
(2) selection of management alternative, and (3) imple-
mentation and assessment of performance.
Municipalities should move ahead now to assess and then
attack their stormwater problem. First-phase actions
should include at a minimum:
1. Initial consensus on objectives and goals.
2. Identification of the existing collection/
treatment system reaction to storm events,
singling out key indicator locations.
3. Installation and operation of monitoring, meas-
uring, and sampling equipment at selected key
locations with provisions for automatic start-
ing under wet-weather flows.
4. Characterizing rainfall behavior and receiving
water response based upon historical data and
initiating new programs where necessary.
5. Becoming directly involved in the new technol-
ogies surrounding urban stormwater management.
In management plan selection procedures, the many-
faceted approaches available should be weighed with
emphasis upon first establishing control over the flows
and then treating to the objective degree.
Program implementation should be staged so as to pro-
vide prototype flexibility and maximum feedback of real
experience as input to subsequent designs and planning.
19
-------
CONTROL
• Intensive demonstration projects should be undertaken
in typical urban areas to explore the maximum benefits
to be derived from the application of source controls.
At least one year's monitoring of unimproved conditions
should precede the implementation of new practices to
allow the development of adequate baseline criteria.
• Similar projects should be undertaken to identify the
benefits derived from the use of collection system con-
trols such as: periodic sewer flushing, improved
catch basin design and/or maintenance, inflow/
infiltration controls, improved regulators (fluidic
controls and swirl concentrators providing both quan-
tity and quality separation), etc.
• Studies should be undertaken to determine the effects
of short-and long-term storage of stormwater runoff
and combined sewage overflows with respect to hazards,
nuisance, and treatability. Both in-line and off-line
conditions should be studied.
• The degree and impact of control facilities should be
included in every predesign analysis for storm flow
treatment. Benefits gained may include increased relia
bility, greater flexibility, more sustained operations,
lower required treatment capacity, and lower overall
cost.
• Increased emphasis should be placed on the further de-
velopment and use of mathematical models as primary
planning and design aids. Models should be used to
maximize the total integrated system approach, in-
cluding basinwide planning and standards.
• Automatic controls, including backup systems, should be
used in most, if not all, urban storm flow management
systems to provide at a minimum for positive startup,
emergency shutdown, and monitoring.
TREATMENT
• Greater emphasis should be placed on the design and use
of storm flow treatment facilities to augment dry-
weather flow treatment when not required for storm
flows. For example, microstrainers and/or multi-media
filters could be used for storm flow treatment in wet
weather and for effluent polishing, increased capaci-
ties, and equipment backup in dry weather.
20
-------
Prototype projects on actual storm and/or combined
sewer overflows should be undertaken to demonstrate the
feasibility of the following key process alternatives:
1. Dual-media filtration with and without poly-
electrolyte feed with ultrafine screening as
pretreatment.
2. Physical-chemical treatment for stormwater
reclamation and reuse.
3. Rotating biological contactors followed by
clarification.
4. High energy gradient chlorine contact chambers.
5. Rapid oxidants for short period disinfection.
Prototype projects now underway should be carefully
screened and data published (annually) in the form of
a state-of-the-art supplement. Major processes in this
category include:
1. Swirl concentrators.
2. Biological processes (contact stabilization,
high-rate trickling filtration, treatment
lagoons).
3. Screening with dissolved air flotation.
4. Fine screening, static screens, rotating fine
screens.
5. Microstrainers.
6. Sedimentation.
7. Disinfection and on-site generation of oxidants.
Studies of the solids resulting from various storm flow
treatment methods should be undertaken. Such studies
should be aimed at characterizing and quantifying the
solids removed during treatment and determining the
handling and disposal alternatives available.
Particular emphasis needs to be placed on the potential
impact of solids derived from storm flow on dry-weather
flow facilities if these are to be used jointly.
21
-------
• Studies on the means and benefits of increasing the
diversion of wet-weather flows to dry-weather plants
and the impact of the increased flows on plant perform-
ance need to be undertaken on the basis of interim as
well as long-term solutions.
• Further demonstrations are required on the upgrading of
dry-weather plants to handle increased wet-weather
flows through temporary process modifications or
supplements.
• The practical application of multiple satellite storm
flow treatment plants and their non-storm beneficial
use should be further demonstrated.
IMPACT
The monitoring of receiving water quality, both during
and following storms, should be intensified and the
data used to improve gross evaluations of the impact
of wet-weather discharges. Again, continuous remote
sensing is highly desirable.
The development of stormwater abatement alternatives
should stress the concept of multiple-use facilities.
For example, the previously mentioned integrated use
of stormwater facilities with dry-weather plants as
backup and/or increased treatment capability, the
direct integration with recreational facilities, the
use of underground construction to reserve the surface
for public facilities, etc., would provide attendant
benefits.
Prototype projects demonstrating the reuse potential
of stormwater should be further emphasized and
implemented.
Through educational programs, increased awareness
should be fostered on the part of engineers, adminis-
trators, and the public on stormwater discharge and
combined sewer overflow problems, technology, and
abatement programs. To aid in this effort a 30-minute
narrated documentary color film has been produced as a
part of this study for distribution by EPA for public
showings before concerned groups.
22
-------
Section III
INTRODUCTION
Precipitation falling on urban areas becomes contaminated
as it enters and passes through or within the environment.
The sources of this contamination are the air (smog, dust
and particulate matter, vapors, gases, etc.), building
and/or ground surfaces, and residues deposited by previous
storm and/or municipal sewage flows within the storm or
combined sewers. After passing through the environment,
stormwater, with or without municipal sewage picked up in
transit, is usually discharged to a receiving water body,
such as a lake, a stream, an estuary, or an ocean. There
the pollutants in the stormwater are decomposed (nonconser-
vative), accumulated (conservative), or transported
downstream.
During the next decade, it is expected that billions of
dollars will be spent in the United States to combat the
degradation of streams and other water bodies by pollutants
released through storm discharges and combined sewer
overflows. The EPA is therefore directing or assisting in
multiple research and development programs and investiga-
tions to identify, control, and correct the causes of known
problems relating to these storm occurrences.
This text was designed, in part, to fulfill the present
need for a compilation and listing of the available litera-
ture concerning storm sewer discharge and combined sewer
overflow treatment and control. There has been a definite
lack of concise, comparative information on past project
results, presented in a single document, especially with
respect to (1) the cost of treatment and control systems,
(2) the performance of such systems, and (3) the assessment
of the various project criteria and goals. Rapid informa-
tion retrieval for evaluating past projects has been
severely hampered by the volume and availability of
literature.
23
-------
This text on the state-of-the-art contains the results of
an extensive literature review along with current updated
information and evaluations concerning the various EPA dem-
onstration projects and many others. This compendium of
stormwater technology should assist designers, administra-
tors, educators, and prospective extramural project appli-
cants as well as those who must review the projects.
OBJECTIVE AND SCOPE
The objective of this text is to identify and assess exist-
ing and available techniques of controlling and/or treating
urban runoff and combined sewer overflows in an environ-
mentally acceptable manner and to guide those responsible
for developing and implementing corrective action.
The scope of this text includes the following with respect
to management alternatives:
1. Sewer separation, its functions, purposes, limita-
tions, and true perspective based on modern
technology.
2. Control and/or treatment capabilities of facilities
intended to function as alternatives to sewer
separation.
3. New developments in sewer construction, repairs,
and usage.
4. Basis for design.
5. Levels of treatment efficiency to be expected from
unit processes or from typical combinations of
treatment and/or control processes.
6. Types and ranges of pollutant most amenable to
removal or conversion.
7. Mathematical modeling techniques for "predictive"
and "decision-making" purposes.
8. Flow measurement methods and sampling devices.
9. Economic evaluations.
10. Facilities and systems application assessment.
24
-------
METHOD OF PRESENTATION
This text has been arranged to facilitate its use as a
solution-oriented reference source. As a result, the indi-
vidual sections making up the main body have been sequenced
along the lines of a typical control/treatment implementa-
tion strategy.
The Conclusions, Recommendations, and Introduction are
presented in Sections I, II, and III, respectively.
Together they form Part I of the text. Background informa-
tion is provided within the introduction on the EPA Research,
Development, and Demonstration Program, the emerging nation-
wide emphasis on stormwater problems, and the importance of
uniform nomenclature and terminology.
The next three Sections are grouped as Part II, Formulating
an Approach. A guide to the assessment of the problem and
selection of the abatement or control methods and processes
is presented in Section IV. In this section, the intention
is to lead the reader, step by step, through the procedures
to be followed in defining the program, identifying the
problem, assessing available data, developing a data-gather-
ing program, selecting and developing a final plan, and
implementing the program.
Background information on the stormwater problem, including
details on the emergence of the problem, the characteristics
of stormwater, the sources and movement of pollutants, and
the environmental effects, is presented and discussed in
Section V.
Evaluation procedures and criteria are presented in
Section VI. The collection of data, with particular empha-
sis on quality sampling and flow measurement, and the anal-
ysis of data are discussed. Processes and equipment
evaluation, system control, and costs are also described.
Management Alternatives and Technology, Part III, is sub-
divided into seven sections according to each generalized
alternative category. Unit processes and operations are
described and each discussion covers their (1) useful appli-
cation, (2) value in meeting specific pollutant removal
requirements, (3) performance data, (4) advantages and limi-
tations, and (5) available cost data.
Part IV, Implementation, has two subsections. The first,
Section XIV, describes the combining of the unit processes
and operations into complete programs, the applications of
mathematical models, master plan examples, and environmental
aspects. General operation and maintenance considerations
25
-------
and experience drawn from all projects are summarized in
Section XV.
Cited references by Section and general references are
presented in Section XVI. Numbers enclosed in brackets
within the text correspond to cited reference listings.
A glossary, abbreviations, and conversion factors are
presented in Section XVII.
BACKGROUND
Recognition of the significance of stormwater-induced water
pollution has been slow. Current interest, limited though
it may be, is largely attributable to the efforts of the EPA
Storm and Combined Sewer Pollution Control Research,
Development, and Demonstration Program.
A 1964 U.S. Public Health Service in-depth study [7] was the
first federal appraisal of the problem. So little was known
about stormwater problems before that time that, in 1963
testimony before a Senate subcommittee investigating water
pollution, Secretary of Health, Education, and Welfare
Celebrezze stated in part:
No real knowledge exists today as to what a
national separation program might cost, although
estimates have been made in billions of dollars.
Even the extent of pollution caused by unsepa-
rated sewers is not known, although preliminary
studies suggest it is very great....Before insti-
tuting a federal program for assistance in the
separation of combined sewers, the ultimate cost
and duration of which are speculative, we need to
obtain realistic estimates of the costs of a sepa-
ration program....Once reasonably accurate infor-
mation as to total cost of a national separation
program is obtained and alternative methods have
been fully explored, we will be able to make in-
formed decisions among the alternatives and pre-
sent recommendations to Congress based thereon.
Consequently, I am unable to support this provi-
sion [funds for sewer separation] of the bill at
this time, because I do not think we have adequate
information. [7]
26
-------
The first federal legislative recognition of the signifi-
cance of stormwater problems is found in the Water Quality
Act of 1965, Section 6.(a) in which it is stated that:
The Secretary is authorized to make grants to any
State, municipality, or intermunicipal or inter-
state agency for the purpose of —
(1) assisting in the development of any project
which will demonstrate a new or improved
method of controlling the discharge into
any waters of untreated or inadequately
treated sewage or other wastes from sewers
which carry storm water or both storm water
and sewage or other wastes,...(Emphasis
added)
Gradually, sanitary districts and local and state govern-
ments have become more aware of stormwater problems. In
some cases awareness has been created by the surcharging of
overloaded sanitary and combined sewers, causing localized
wastewater flooding of streets and basements--events usu-
ally brought to the attention of City personnel.
The District of Columbia has been aware of the problems of
combined sewers for many years. After 1890, separate
sewers were installed in new areas of the city. In studies
conducted by consultants in 1933 and 1935, it was recommended
that certain areas within the older combined sewer portions
of the District be separated. In a 1957 Board of Engineers
study,it was recommended that a project be undertaken to
(1) separate 10 percent of the combined sewer area and
(2) construct new interceptors to convey more combined
wastewater to the water pollution control plant. This rec-
ommendation was only partially implemented because of
budgetary reasons. Although the Board recognized that com-
bined sewer overflows were a significant source of water
pollution, it had little information on the pollution result-
ing from storm sewer effluents alone. A 1970 EPA-sponsored
study of the District's stormwater problem recommended:
Discontinue the current sewer separation pro-
gram and develop pollution abatement programs
for both the combined and separated sewer areas
if the pollutional characteristics of storm
water determined in the current study are con-
firmed in other areas of the District. [2]
27
-------
1967 National Inventory
In 1967,the APWA conducted a national survey [9] to obtain
data on (1) the number of people served by different types
of sewer systems, (2) the extent of storm and combined sewer
overflows, (3) water use in waters receiving overflows,
(4) flooding related to combined sewers, (5) infiltration,
(6) combined sewer regulators, (7) treatment plant bypassing,
and (8) costs of sewer separation and other abatement
alternatives. The survey included all urban communities
with populations greater than 25,000 and representative
samples of smaller ones from each state. Among the exten-
sive findings of this survey were the following (costs
shown in brackets have been adjusted to ENR 2000):
If all jurisdictions with combined sewers were
to replace them by providing a separate conduit
for sanitary and industrial wastes, and another
for storm water, they would face an expenditure
of approximately [$56 billion]. Inclusion of
the expenses incurred in making the necessary
plumbing changes in and on private property to
effect total separation would increase this cost
to approximately [$90 billion]. However, the
report strongly implies that such a calculation
must be considered theoretical. Responses from
many municipalities, especially those with high
population densities, disclose that the possi-
bility of changing all combined sewers to sepa-
rate is remote.
Alternate methods of control and/or treatment
other than sewer separation are in use in some
jurisdictions and research is being conducted to
further evaluate the effectiveness and applica-
bility of various methods. There were insuffi-
cient data to make a detailed cost estimate of
the alternate methods. However, from the lim-
ited data available, it appears that the cost of
such methods would be about one-half the cost of
complete separation. In addition, the expense
of separating plumbing in and on private prop-
erty would not be necessary. If alternate means
of control and/or treatment were used exclu-
sively, which is unlikely, the total cost would
thus approximate [$28 billion].
28
-------
Of the 641 surveyed jurisdictions [1,329 juris-
dictions estimated nationwide total], 493 re-
ported 9,860 combined sewer overflows. In
addition, 237 of the 641 jurisdictions reported
759 overflows from combined sewer pumping sta-
tions and 160 communities reported 755 separate
sanitary sewage pumping station overflows. Re-
ported treatment plant overflows or bypasses
amounted to 532. In addition, 87 jurisdictions
listed 2,306 "other" overflow sources. In large
measure, these are unregulated discharges. The
grand total of 14,212 overflows in the surveyed
jurisdictions indicates the multiplicity of dis-
charges into watercourses, lakes and coastal
waters.
Infiltration was reported as a problem under dry
weather conditions by 14 percent of the juris-
dictions surveyed; under wet weather conditions,
53 percent reported infiltration as a problem.
Bypassing of untreated or partially treated
sewage occurs at water pollution contfbl plants
at frequent intervals, for various reasons, in-
cluding the overtaxing of plant capacities dur-
ing storm flow periods.
Engineering studies on overflow problems have
been made in 310 of the jurisdictions surveyed.
Of these, 220 jurisdictions have made estimates
of the cost for a project or projects to elimi-
nate all or part of their combined sewer over-
flow problems. In addition, 222 jurisdictions
surveyed reported current plans for sewer
separation projects and 52 jurisdictions have
plans for constructing alternate control or
treatment facilities. [9]
EPA Research, Development, and Demonstration Program
In 1965, Congress authorized a Research, Development, and
Demonstration Program to find lower cost remedial alterna-
tives to complete combined sewer separation. Today this
EPA program is under the jurisdiction of the-Storm and
Combined Sewer Technology Branch of the National Environ-
mental Research Center — Cincinnati. Major emphasis to date
29
-------
has been directed toward combined sewer overflows, although
it has been found that storm sewer discharges also carry high
pollutant loads (contrary to longstanding beliefs).
Thus far, over 140 projects have been carried out by means
of demonstration grants and contracts. These projects have
produced much information useful in defining the problem and
in the application of remedial techniques related primarily
to combined sewer overflows. As a result, new hardware has
been developed and is now available to those engaged in
planning and construction of remedial works. For the most
part, the funded projects have not been integrated into an
overall abatement program but have covered many individual
facets of the problem.
Valuable information on the characteristics of storm dis-
charges and combined sewer overflows, along with various
treatment and control processes and systems, has been de-
rived from these projects. On an individual basis, there
have been both successes and failures in the program, but
on the whole, many effective tools for combating the problem
have been developed.
Other Programs
In addition to the EPA Research, Development, and Demonstra-
tion Program, notable other national programs in the storm-
water field include those of the American Society of Civil
Engineers (ASCE), the National Science Foundation (NSF), and
the Office of Water Resources Research (OWRR).
ASCE — The American Society of Civil Engineers,1 Urban Water
Resources Research Program was initiated and developed by
the ASCE Urban Water Resources Research Council. The basic
purpose of the Program is to help establish coordinated
long-range research in urban water resources on a national
scale. The Program has developed over 3 phases: (1) 1967-
1969, research needs assessment; (2) 1969-1971, studies on
urban water problems under the theme of urban water manage-
ment; and (3) 1971-1973, translation of research findings
into practice. To date, 18 Technical Memoranda (see
Section XVI for listing) have been published and distributed.
NSF — Within the National Science Foundation, the Division
of Environmental Systems and Resources is one of four organi-
zational units administering the program of Research Applied
to National Needs. The purpose of the programs of this
Division is to support research to develop improved under-
standing of environmental issues and the impact of human
activity on the environment, and to improve the prospect of
30
-------
reconciling economic development with improved environmental
quality in the United States. The major program areas are
(1) regional environmental systems, (2) environmental
aspects of trace contaminants, and (3) weather modification.
QWRR - The Office of Water Resources Research, U.S. Depart-
ment of the Interior, in addition to direct research support,
compiles annually a summary of descriptions of current re-
search on water resources. The catalog series, initiated in
February 1965, makes readily available to all who are en-
gaged in water-related research, or otherwise concerned with
water resources problems, information on what is being done,
by whom, and where.
EMERGING EMPHASIS ON THE STORMWATER PROBLEM
As a result of these recent and ongoing programs, the magni-
tude of the stormwater problem is becoming more widely known.
Studies are underway to characterize more thoroughly the
quantity and quality of storm runoff, storm sewer discharges,
and combined sewer overflows. This state-of-the-art text
brings to the engineering world the first comprehensive
analysis of the problem and what can be done about it. Two
similar studies are proceeding concurrently, one in Canada
and one in France. It has become readily apparent during
this study that no single "ideal" solution to the problem
exists.
It was originally believed that the best answer to abatement
of the combined sewer overflow problem was the elimination
of combined sewers by constructing separate storm and sani-
tary sewers. More recently, in light of the costs and dis-
ruption involved in separating combined sewers, emphasis is
being placed on alternatives to sewer separation. Specific
attention is being paid to (1) source controls, (2) storage
and/or treatment of combined sewer overflows, (3) reduction
of infiltration and inflows to sanitary sewers, (4) more
complete utilization of existing combined sewer and treat-
ment systems, and (5) mathematical modeling techniques for
predictive and decision-making purposes in augmenting de-
sign, selection, and operation of abatement systems.
The EPA has sponsored numerous studies over the last few
years concerning the treatment of combined sewer overflows,
both with and without storage. Retention of combined sewer
flows makes it possible to provide treatment at existing
sanitary treatment facilities or at facilities specifically
designed for storm flows. If the level of treatment re-
quired for storm flows is less than that provided at the
water pollution control plant or to offset high transmission
31
-------
costs, adding the treatment at the storage locations may be
advantageous. Thus, several small satellite plants can be
used, possibly in conjunction with phased construction,
rather than one large plant.
A manual of practice on the prevention and correction of
excessive infiltration and inflow into sewer systems was
recently completed [8]. The objectives are (1) elimination
of untreated wastewater bypasses and overflows, (2) lower
total costs of treatment works, and (3) elimination of the
construction of unnecessary treatment works capacity.
1972 FWPCA Amendments
The Federal Water Pollution Control Act Amendments, passed
in October 1972, place new and stronger emphasis on storm-
water pollution. In the Act, Congress defined "treatment
works" to add, for the first time,
...any other method or system for preventing,
abating, reducing, storing, treating, sepa-
rating, or disposing of municipal waste, in-
cluding storm water runoff, or industrial
waste, including waste in combined storm water
and sanitary sewer systems. (Emphasis added)
Hence, projects for abatement of stormwater pollution (sub-
ject to guidelines set forth by the Administrator of the
EPA) will be eligible for general construction grants even
if new technology is not involved. This is consistent with
the purpose for the construction grant program to
...provide control or treatment of all point
and nonpoint sources of [water] pollution.
Further, it is consistent with the President's Environmental
Message of 1970 in which he stated:
A river cannot be polluted on its left bank and
clear on its right. In a given waterway,
abating some of the pollution is often little
better than doing nothing at all, and money
spent on such partial efforts is largely
wasted.
The Act also requires that grants for the construction of
treatment works will not be approved unless the tributary
sewer system is not subject to excessive infiltration.
32
-------
Comptroller General's Report
Because of the seriousness of the combined sewer overflow
problem, the Comptroller General of the United States, on
March 28, 1973, submitted a report to Congress on the "Need
to Control Discharges From Sewers Carrying Both Sewage and
Storm Runoff." Findings revealed that combined sewer over-
flows "...are a major pollution problem and prevent many
areas from attaining Federal and State water quality goals."
The report recommended that the EPA require states to iden-
tify, study, and submit abatement plans for combined sewer
overflow pollution and to consider the award of construction
grants for these abatement facilities. [3]
State and Local Requirements
On the state and local level, regulations concerning water
pollution abatement have been tightened. Illinois [5, 10,
14] and Georgia [4] have promulgated their requirements for
overflow control and treatment.
In an effort to curb channel erosion and downstream silta-
tion, the State of Maryland [13, 11, 1] requires the land
developer to attenuate runoff so as not to allow flow re-
leases (for up to two year storms) to occur any faster than
before site construction.
The City and County of San Francisco, California, has passed
a 1970 resolution [12] for a special time schedule to regu-
late discharges from combined sewers. A recent regulation
in Orange County, Florida [6], states that a complete storm-
water management system be provided to handle all stormwater
runoff in "prime recharge areas." It also states that treat-
ment is required for stormwater in all drainage systems.
This regulation is the first to deal so specifically with
the control and treatment of urban runoff as opposed to com-
bined storm and sanitary flows.
INFORMATION SOURCES AND DEFINITION OF TERMS
Information Sources
Information for this text was gathered from (1) the litera-
ture, (2) project documents, (3) demonstration project site
visits and interviews, and (4) previous experience. To
obtain information regarding the many ongoing and concurrent
studies of the EPA Research, Development, and Demonstration
Program, monthly progress reports for the various projects
were reviewed. In addition, reports from the EPA's Environ-
mental Protection Technology Series were studied for infor-
mation concerning recently completed projects. Information
33
-------
was also obtained from an extensive survey of other litera-
ture in the field. More than 600 references, including
books, magazine and journal articles, engineering reports,
progress reports of ongoing projects, and EPA reports were
collected and reviewed.
Visits were made to numerous EPA demonstration project sites,
both completed and ongoing, to view the facilities and to
obtain information on the size and scope of the various
projects, their function, and problems encountered. In con-
junction with these visits, discussions were held with plant
operators, design engineers, city officials, and EPA offi-
cials concerning the design, operation, and effectiveness of
the various projects. Construction costs, operating and
maintenance costs, and operational data were obtained for
the various plants where available.
Construction drawings of many of the plants were reviewed.
Particular attention was paid to unique construction fea-
tures, special application of mechanical equipment, unusual
design or construction problems, plant layout, physical
size, and mode of operation (automatic or manual).
Nomenclature and Terminology
Efforts to attain better levels of treatment or abatement
for storm discharges and combined sewer overflows must be
based on a uniform understanding of the "language" of the
field. During the investigations performed for this study,
it became apparent that no standardized nomenclature exists
among the jurisdictional authorities, state and local
agencies, and others who operate in the storm discharge and
combined sewer overflow field. To assist in unifying terms,
terminology, and nomenclature, a Glossary of Terms is in-
cluded in Section XVII of this report.
The units used to define design criteria, operating effi-
ciencies, pollutants, etc., were found to vary widely even
for similar facilities at different locations. An attempt
has been made in this text to standardize the units and to
suggest evaluation parameters for the various methods and
processes described.
In light of the proposed national policy to simplify units
of measurement by converting from the English system to the
metric system, the units used in this text are the metric
units. Since metric units are not commonly used for most
measurements nor are they readily recognized by most people
in the United States, the metric units are followed by the
English equivalent in parentheses throughout the text.
34
-------
Units presented in tables and figures are English, and the
metric conversion factors are listed below each table and
figure. A list of conversion factors is also presented in
Section XVII for the reader's convenience.
35
-------
-------
Part II
FORMULATING AN APPROACH
-------
-------
Section IV
GUIDE TO PROGRAM ASSESSMENT AND SELECTION
What constitutes an effective program for urban storm flow
management? How do the pieces fit together? Where does
one start? These questions are addressed briefly in this
section in the context of outlining a program for the urban
administrator and/or engineer. The purpose of presenting a
guide at this early point in the text is to provide a frame
work upon which to relate material presented in the subse-
quent sections.
The success of a program depends largely upon an early con-
sensus on objectives and allocation of available resources.
The work is motivated by problem acknowledgement and iden-
tified magnitude. Program development logically follows
using coarse (roughing out) and fine analyses, supplemented
by identification of data needs and data collection, until
alternative plans are evolved. Final stages include public
participation, selection of a final plan, program implemen-
tation, and continuing studies.
PRELIMINARY PROGRAM DEFINITION
Stated simply, the administrator/engineer must define the
scope and objectives to be achieved before methods of storm
and combined sewage treatment and control can be selected.
The task is not easy. For example, the objective of
"attaining a significant reduction of pollutants discharged
to the receiving waters" is rather meaningless unless it is
expanded by addressing such questions as:
Significant with respect to what?
Which pollutants?
Where in the receiving water?
Under what circumstances?
39
-------
Within what funding constraints?
Within what timing constraints?
Is the program to be limited to the reduction of pollu-
tion from urban runoff or are other alternatives to be
considered (e.g., upgrading existing treatment facili-
ties, removal of sludge banks or other benthic de-
posits, control of agricultural drainage, control of
water use and land use practices, etc.)?
To what extent is the program contingent on actions by
other jurisdictions?
Thus, as an initial step, the administrator/engineer should
identify what degree of improvement at what point in time
will constitute successful achievement of program objectives
and, conversely, what outcome will indicate failure.
The scope of a project as defined in this text includes
a definition of the magnitude or limits of the investiga-
tion, the required timing, and the level of funding avail-
able to support the work. The study area and interdepend-
encies, if any, with adjoining areas and/or systems should
be clearly defined in the initial scope. The purpose is to
determine to what extent the problem may be isolated, hence
simplifying the limits to be defined. Flows crossing the
study area boundaries should be identified (i.e., overflows
to receiving water, intercepted flows to treatment and from
upstream areas, overland flows at times of flooding, cross-
connections, etc.).
Next, the program timing should be selected. Are the im-
provements to be of an interim nature or are there to be
clear ties to a broader long-range program? Is there an
immediate (urgent?) problem which must be allowed for? Are
there deadlines to be met?
Finally, the potential project funding, both sources and
general amounts, must be realistically appraised. This
factor may greatly reduce the number of alternatives to be
studied.
Objectives
Objectives may be identified by legislative or regulatory
agency requirements, or they may be the result of strong
local concern to clean up the beaches, increase safe swim-
ming days, reduce flooding, improve aesthetic enjoyment of
40
-------
recreational waters, protect water supplies, or application
for a justified waiver, etc. Where specific problems exist
and are identified, direct approaches are facilitated with
resultant economy of effort. On the other hand, the objec-
tive may be one of prevention to avoid a potential future
problem. In the latter case, the study may be of a recon-
naissance nature to match instream quality with specific
point and nonpoint discharges. In water-short areas, the
objective may be to maximize the reclamation potential of
stormwater discharges and, as a result, a survey may be
needed not only of the sources but also the potential users.
The advantages of modest improvements in some areas should
be weighed against a planning overkill with the possible
danger of indefinitely postponing construction. Approaches
which integrate wet-weather solutions with dry-weather
system and treatment needs should be given priority
consideration.
Finally, the objectives should be listed in concise state-
ments, comments should be solicited, and a practice of
continual reassessment should be adopted. Surprisingly, but
effectively, the fundamental objectives of one of the larg-
est stormwater studies ever undertaken (Chicagoland Area)
numbered only two:
1. Prevention of backflow to Lake Michigan for
all storms of record.
2. Meet the applicable waterway standards estab-
lished by the State Pollution Control Board
and the Metropolitan Sanitary District of
Greater Chicago. [2]
PROBLEM IDENTIFICATION
Each storm flow management problem encountered may be ex-
pected to have the following base elements: (1) a geo-
graphic area, (2) a sewerage network, (3) a characteristic
hydrology, (4) a receiving body of water, and (5) agencies
with jurisdictional control. All of these elements must be
considered in identifying the problem and setting baseline
criteria. Baseline criteria (along with the scope and ob-
jectives) are essentially a part of the "ground rules," and
all alternatives must start with these items in common,
hence as a base.
Tributary Area
Of particular consequence in identifying a tributary area
are its boundaries, size, land use characteristics and popu-
lation (present and projected), political subdivisions,
41
-------
and topography. Maps, aerial photographs, census tract data,
planning reports, soil studies, and on-site inspection are
among the most used sources of information. Street litter,
construction activities, flooding complaints, land use,
property ownership, maintenance practices, pesticide and/or
fertilizer use, and the like can be assessed through direct
observation and through the appropriate city departments and
retail outlets.
These data are used primarily to estimate the expected sur-
face runoff quantity and quality, and to locate problem
areas and possible sites for abatement facilities. The
site conditions (e.g., the availability of land) may greatly
influence the selection of alternatives.
Sewerage Network
The type of system (combined, separate, or hybrid) and its
general condition (age, structural soundness, infiltration,
adequacy of slopes, etc.) must be known. Next, the basic
location, alignment, and capacities of the major trunks,
pumping stations, and all points of control, discharge, and
interconnections with other areas must be identified. Of
particular concern will be the availability of interceptor
and treatment capacities, measured flow and quality data,
records of surcharging or flooding, and basis of design.
Paths of overland flow and locations of depressed area sur-
face storage may be of interest.
The primary purpose in this aspect of the work is to know
how the system will respond under different storm occur-
rences and how in-system modifications may alter these
responses.
Characteristic Hydrology
Rainfall and stream flow are the driving forces behind
essentially all storm flow investigations. Because storm
patterns, intensity, and frequency vary markedly with geo-
graphic location (see Figures 2 and 3), alternative methods
of approach must be considered. For example, storms of high
intensity and short duration may be best countered with
storage, whereas storms of low intensity and long duration
may be more effectively controlled through increased treat-
ment capacity or runoff deterrents such as porous pavements.
Intervals between storms are significant in that they may
dictate dewatering requirements and, in turn, treatment
rates in system cleanup from one storm in preparation for
the next.
42
-------
CO /—N
>- v>
=3 CD
T3 .C
O
O
CD —
=3 CO
— co
CO CD
CO
CD
O
• c
CM CD
CD L-
i_ CD
Z3 C2.
00 X
— CD
43
-------
CD x—s
»- V)
=3 CD
•o jc
o
CD
CD
CD rj-
i- CM
CD CD
C CD
O >t
— I
•»-» *—
CD
CD
CD
O
CD
=3 CD
Dfl O.
— X
U_ CD
44
-------
The average annual rainfall recorded in 87 major cities in
the United States is 84 cm (33 inches). The range of
averages for these cities is from 5 to 178 cm/yr (2 to
70 in./yr). When dealing with storm flow problems, other
important characteristics are the intensity and type of
precipitation, and the magnitude, frequency, and duration
of storm events. Because rainfall events tend to repeat
themselves over long periods of time, considerable benefit
can be drawn from extensive historical records. Because
the repetition is random and not sequential, it is best to
make comparisons by arraying the data according to some
parameter of interest, such as magnitude or maximum
intensity. While the characteristics of tomorrow's storm
cannot be predicted, the probability of occurrence of a
certain number of storms of like characteristics in a cer-
tain period of time can be forecast reasonably.
Because many storms exhibit pronounced frontal patterns
and/or intensity cell behavior, the spatial and temporal
variance of rainfall within a single storm may be
substantial.
Frontal Patterns — In recent studies conducted in San
Francisco[1],it was found that 30 rain gages--average of
1 per 324 ha (800 acres)--were required for good representa-
tion of storm patterns in this particularly hilly and ex-
posed city. A time series selected from these rainfall data
(Figure 4) illustrates a typical storm pattern associated
with a weather front. The authors [3] found significant
spatial and temporal differences in rainfall, not only as
shown on the figure, but also when the same storm was de-
picted using alternatively 5-minute, 10-minute, and 15-
minute cumulative rainfall amounts. The primary signifi-
cance of such patterns is that while some areas are under
maximum storm stress, other areas nearby are relatively
unaffected. Thus, centralized facilities, coupled with
total system management and control, may be able to provide
better pollution abatement with smaller facilities by ser-
vicing the stressed areas preferentially.
Intensity Cells — Another frequently encountered storm
phenomenon is the "heat island" shower condition shown on
Figure 5 for Washington, D. C. In the observer's words:
These four days offer stronger evidence than any-
thing observed so far this year, that the heat
and perhaps pollution generated by a city can
assist nature in creating rain. During this
period the NCA was under a high pollution alert.
On each of these days Washington National Air-
port (WNA) reported prevailing southwest surface
45
-------
PACIFIC
OCEAN
S,F. BAY
APPROXIMATE
RAIN GAGE LOCATIONS
TIME: 19:48
TIME: 19:58
3 4
SCALE IN MILES
TIME: 20:08
TIME: 20:42
ooo
ooo
=*»=
LEGEND
SHAPINGS REPRESENT 10-MINUTE CUMULATIVE
RAINFALL AMOUNTS PRIOR TO TIMES DEPICTED
0.0-.03 IN.
.03-.05 IN.
.05-.08 IN.
.08-.10 IN,
. 10-.13 IN.
NOTE: IN. x 2.54 - CM
Ml. x 1.609- KM
Figure 4. December 22, 1971
frontal storm pattern, San Francisco [3]
46
-------
^1
:"'O
V''
NOTE; MI x i.eog- KM
IN. X 2.54 - CM
SCALE IN MILES
Figure 5. Rainfall isohyets (inches) for
heat island showers, July 17-20, 1972,
Washington, D. C. [6]
47
-------
winds averaging 5 mph or less, and winds up to
10,000 feet were less than 10 mph. This unusu-
ally slow air movement allowed the same air par-
cel to remain over the hot city long enough to
absorb plenty of heat. By late afternoon each
day (early afternoon in some areas), as those
heated parcels began rising, they cooled and
condensed their moisture. As a result, showers
and thunderstorms formed over the eastern suburbs
and interior city. Meanwhile, the northern,
western and most of the southern parts of the NCA
remained dry. Note that the rainfall patterns
(isohyetals) are circular, indicating little
movement of the clouds. On other days the iso-
hyetals were elliptical, showing more rapid
storm motion, and the air did not remain over
the city long enough to cause rain. [6]
In storm flow management systems, these intense but ex-
tremely localized showers could best be controlled if access
to centralized storage/treatment facilities were provided.
Conversely, remote satellite facilities would be easily
overtaxed by such downpours or, by accident of their loca-
tion, not even come into use.
Illustrative Comparisons — Each major urban area, and to
some extent,subdivisions thereof, must be considered sepa-
rately in plan development because each will have its own
characteristic hydrologic phenomena. This is illustrated by
a comparison of three distinctly different regimes --Chicago,
San Francisco, and the United Kingdom. Both Chicago and San
Francisco are large cities with old combined sewer systems.
The fact that the annual precipitation in Chicago averages
84.3 cm (33.2 inches), as compared to 47.5 cm (18.7 inches)
for San Francisco, is not as significant as the difference
in the nature of the storms which occur. Most of the precip
itation (90+ percent) in San Francisco occurs during the
months of October through April. So, for 5 months of the
year, there is almost no rain and, consequently, no storm
flow problem. During a typical rainy season month, however,
San Francisco receives as much precipitation as Chicago,
where the precipitation is distributed more evenly through
the year. Also, in Chicago, sudden, intense storms and sum-
mer thundershowers are common, whereas such events are rare
in San Francisco. This is observed in the data by comparing
the 1-hour rainfall with a 1-year return period in Chicago
of 3.0 cm (1.2 inches), with that of San Francisco of 1.5 cm
(0.6 inches). In other words, Chicago can expect to receive
3.0 cm (1.2 inches) of rain or more falling in 1 hour, on
the average, once a year. By comparison, on the average
48
-------
once a year, San Francisco would receive a like amount of
rain during a 5-hour period, and London receives the same
amount during a 24-hour period.
Everything else being equal, a system for treating combined
sewer overflows in Chicago would have to be sized to accom-
modate storms of greater intensity. For example, a 10-ha
(25-acre) drainage area with a runoff coefficient of 0.5
would require a 0.42 cu m/sec (9.5 mgd) treatment facility
designed for a 1-year, 1-hour storm in Chicago, but only
a 0.21 cu m/sec (4.7 mgd) facility for San Francisco. Note
that the size of a storm flow treatment system is primarily
intensity (rate) sensitive, whereas the capacity of a stor-
age system is intensity-duration (volume) sensitive.
In the United Kingdom, where most storms are of very low
intensity, many cities have provided primary treatment for
excess flow from combined sewer areas for many years. The
standard excess flow capacity is approximately 6 times dry
weather flow with greater flows discharged without treatment
[5]. Since few storms in the United Kingdom have very
long duration, other cities have provided storm standby
tanks to capture excess flow for treatment after the storm.
The notable success of these installations, some in operation
for over 50 years, is largely attributable to the uniformity
of the storm events.
Storm Event Definition — The best storm event definition
for a particular study should consider the timing and sepa-
ration of rainfall, the source data available, the scope
and objectives of the investigation, and the limitations of
the physical system.
Because storm events are to a large extent independent of
clock time, simple daily and, to some extent, hourly preci-
pitation tabulations and analyses may be misleading. For
example, a 10-hour storm starting at 8 p.m. would appear
as two storms on an arbitrary daily tabulation, and maximum
hourly precipitation based on clock hours could be signifi-
cantly less than the maximum 60-minute rainfall chosen
irrespective of starting time.
Further, to facilitate practical engineering analysis, rea-
sonable separation between "storm events" is considered
most desirable. Thus, on-again/off-again showers would be
classified as a single event rather than as multiple storms.
This improves the credibility of statistical analyses which
presume relative independence of events.
49
-------
In a recent study for the District of Columbia, storm
events were characterized as:
...starting with the first measurable rainfall
after a minimum of 6 consecutive hours of no
measured rainfall. A storm event ends at the
next 6 consecutive hours of no measured rain.
[4]
The study encompassed the entire District area and, on this
basis, a minimum 6-hour separation was considered appropri-
ate for initiating significant corrective action, such as
partial dewatering of storage basins for treatment.
Although the selection of the separation interval is some-
what arbitrary, it is necessary before breaking down the
continuous rainfall record into "discrete" events where the
events lose the identity of sequential occurrence.
The recorded observations of the U.S. Weather Bureau are
the principal data source, but these may be supplemented by
city and private gages forming a blanketing network.
Pertinent data suited for computer (statistical) analyses
and ranking include for each event:
The storm date, starting hour, duration, total
rainfall, maximum hourly rainfall, and the hour
after the start of the storm in which it
occurred, the days elapsed since the last storm,
and occurrences of excessive precipitation...
From the arrayed data, engineering judgments can be made as
to the significant events and event series for planning
purposes.
Receiving Waters
Just as the input hydrology and tributary area characteris-
tics provide the source data for testing storm flow alter-
natives, the receiving body of water represents the ultimate
receptor of the generated wastes. Thus, receiving water
conditions are of paramount importance in identifying the
problem. The problem may be in the form of a direct hazard,
such as infectious bacteria, toxicity, or flammables; a
biological degradation; an aesthetic or public nuisance;
or any combination thereof.
The nature of the receiving waters--lakes, estuaries, swift-
flowing streams, etc.--will largely determine the degree to
which wastes are assimilated (consumed), accumulated, or
transferred. Other considerations include the relative
50
-------
magnitudes, loadings, and locations of all discharges; back-
ground loads in the receiving water; uses and withdrawals.
Generally, the most difficult task is relating conditions
in the receiving waters to specific discharges, because
only the aggregated effect is seen.
Stream flow records should be analyzed with respect to
monthly minimum and average flow values, background impacts
(such as flow augmentation and water use commitments), and
historical quality data (such as zones of maximum stress,
recovery data, natural background levels, trends, and rela-
tionships to specific storm events). For example, the data
may indicate a noticeable increase in pollutants after
storms or, on the contrary, localized evidence of a bene-
ficial (flushing) effect. Other suggested alternatives are
the employment of mathematical (simulation) models, inten-
sive representative sampling and analyses of both the dis-
charges and receiving waters, the location and identifica-
tion of bottom deposits, and the use of conservative tracers
(to identify flow paths and dilutions).
Characterization of each waste stream (point discharge)
should include its flow rate, duration and total volume, and
the concentrations and total mass of pollutants. A deter-
mination of floatables may be facilitated by the temporary
construction of a floating log boom around the discharge
point (see Section V, Figure 9)- Selective grouping of dis
charges, maintaining a materials and flow balance, may
greatly simplify mathematical approaches and help relieve
data congestion. Nonpoint runoff, where significant, may be
handled by simulation models or approximated from tabulated
values.
The cumulative study effort is directed toward predicting
receiving water responses to any feasible waste load con-
figuration and the relative impact effect of different
systems (i.e., stormwater overflows versus normal dry-
weather treatment plant and industrial effluents versus
irrigation water returns, etc.). In special cases, it may
be necessary to measure or estimate flow quantities and
characteristics at selected upstream points (i.e., diver-
sions) in the collection system as well as at the point of
discharge. The latter will permit consideration of upstream
impoundments, alternative routings, and evaluation of in-
system controls.
Agencies
Agencies having jurisdictional control must be identified
so that they may be adequately represented in the planning
process. In essence, the conceptual approach must be
51
-------
areawide in scope to preclude actions of one agency nullify-
ing or hindering those of another. Again, the receiving
water response to the aggregated effect of the loadings,
both natural and induced, and the impact of isolated dis-
charges are nearly impossible to detect. Other important
reasons for agency identification are to distribute imple-
mentation responsibilities equitably and to encourage the
free flow of information and sharing of effort.
DATA ASSESSMENT AND PROGRAM DEVELOPMENT
With the problem(s) identified and the available resources
assessed, the selection of candidate alternatives is rather
straightforward. One of the first actions could be to list
the major strengths and weaknesses of the existing system.
Is it undersized? Oversized? Are there particular
bottlenecks? High incidence of flooding? Is excess treat-
ment capacity available at existing plants? Can the exist-
ing trunk combined sewers and interceptors handle increased
flow rates or temporary in-line storage? What feasible
sites are available for construction of storage-treatment
facilities? Which projects now underway will be affected?
Which commitments must be met? Who is best qualified to
conduct the study? How may the effort be shared and re-
sources put to most effective use? All of these questions
are typical of those to be considered in the preplanning
stage.
"First Cut" Analysis
The purpose of a "first cut" analysis is to reduce effec-
tively the number of alternatives requiring detailed anal-
ysis and to point out areas with high return prospects. In
this text, Part III, Management Alternatives and
Technology, has been organized by category to facilitate
this analysis: source control, collection system control,
storage, and treatment. Each alternative is presented as a
unit process. Integrated (complex) systems, the combining
of two or more unit processes, are discussed in Section XIV.
The results are summarized in Table 6. Comparisons are based
upon the type of treatment performed, the degree of contami-
nant removal, and the relative capital cost. A new manual
[8] is being prepared to normalize input data as an aid to
this level of analysis.
Source controls are designed to correct the problem before
it becomes a problem (i.e., controlled ponding to reduce
flow peaks within the system, runoff attenuation through the
use of porous pavements, improved maintenance of construc-
tion sites to minimize scour and washouts, better housekeep-
ing to clean up leaves, street litter, and refuse before
52
-------
Table 6. MANAGEMENT ALTERNATIVES SUMMARY'
Categories
SOURCE CONTROLS
Retention/detention
Enforced controls
Neighborhood sanitation
COLLECTION SYSTEM CONTROLSC
Sewer separation
Sewer flushing and
cleaning
Improved regulators
and tide gates
Remote monitoring with
supervisory control
Fully automated control
STORAGE AND TREATMENT
Storage
Physical treatment with
and without chemicals
Biological treatment
Physical -chemical
Disinfection
INTEGRATED SYSTEMS
(Varies)
Aesthetic benefits
»
•
.
*
•
•
•
•
c
T
3
0
u.
•
•
•
•
•
BOD5 reduction :
o
.
•
•
•
•
•
•
•
o
•
•
•
•
•
•
•
CD
0)
O
*
•
•
Suspended solids reduction:
o
•
•
•
•
»
•
•
•
•
o
•
•
•
•
•
•
•
CD
l-i
O
«
*
•
Total coliform reduction:
o
•
«
»
•
•
•
•
o
•
•
•
•
•
•
•
o
o
•
•
•
•
Nutrient reduction:
CD
•
»
»
,
*
•
•
•
O
•
«
»
•
•
•
CD
O
«
•
•
Capital cost range,0 $/acre
d
o
CD
0
0
o
.
•
•
•
•
•
o
o
o
O
0
O
0
0
,
•
•
0
o
CD
CD
-------
this material finds its way into storm runoff, etc.)- What
can be accomplished and what it will cost can be estimated
most effectively by reexamining the city's current practice
and a general site inspection.
Collection system controls utilizing in-system storage repre
sent promising alternatives in areas where conduits are
large, deep, and flat (i.e., backwater impoundments become
feasible) and where interceptor capacity is high. Where
interceptor capacity is available but trunk lines are steep
or slopes are broken up, the same effect may be feasible
with supplemental off-line storage. Upstream off-line stor-
age may have an added benefit of reducing line surcharging.
Improved regulators are generally essential, and remote
monitoring and control are highly desirable. Flexibility
and low capital cost are the major assets of collection sys-
tem controls, and sophistication and high maintenance are
the major drawbacks. Sewer separation has many limitations,
including effectiveness, inconvenience to the public, time
required for implementation, and cost.
Storage and treatment alternatives are strongly influenced
by input hydrology and the capacities and limitations of
available facilities--interceptors and treatment works.
Storms are generally an unpredictable series of events char-
acterized by hydrographs of rapidly changing peaks and
valleys extending over limited durations. For this reason,
storage facilities ahead of treatment can be used to level
the peaks and valleys into rates of flow more suitable for
treatment. Also, very large storage capacities may fully
capture one or more storm events, allowing treatment to be
deferred and/or the rate reduced to match available capaci-
ties most effectively.
To obtain a good balance between storage volume, treatment
rates, and overflow occurrences, simplified computer pro-
grams have been developed and applied [4, 1]. Typically,
continuous rainfall data on a daily or hourly basis are in-
put to these programs and they are used to compute runoff,
route it through a single specified storage volume with a
specified withdrawal (treatment) rate, and list the times,
durations, and quantities of overflow. In this manner,
many alternative storage/treatment combinations can be
rapidly compared and preliminarily assessed as to capaci-
ties required to meet selected overflow criteria (quantity
and frequency). Having a provisionally defined capacity,
the treatment process can be varied to satisfy quality
objectives and, perhaps, to relieve dry-weather flow
deficiencies.
54
-------
The final step of a "first cut" analysis should be in the
form of a summary tying together program objectives, system
strengths and weaknesses addressed, and alternatives studied
and accepted and/or discarded. Integrated approaches
coupling two or more categories should be fully explored,
and a listing of the best feasible alternatives should be
derived. Comparisons with programs undertaken in other
areas should be considered.
Candidate Evaluation: Data Needs
In this step, attention is directed to detailing the selected
alternative plans as to costs, efficiencies, objective
achievement, land requirements, environmental effects, oper-
ation feasibility, and social and political acceptance.
Of greatest importance is the assessment of additional data
needs and the programming of data acquisition. The latter
may require several years of effort, including flow measure-
ment and sampling, and may include pilot and prototype
storage/treatment operations. Considering the scale of the
projects involved, this time requirement is not unreason-
able; rather, it is a factor to be used to great advantage
in program implementation (i.e., optimizing the data bene-
fits of the test facilities). Assessment procedures are
discussed in Section VI, Evaluation Procedures and
Criteria.
Each candidate plan should be developed to identify the num-
ber, sizes, locations, and functions of facilities required.
Preliminary basic design decisions should be made, concep-
tual schematic drawings completed, land requirements de-
fined, and alternative sites selected. Subsurface and site
investigations should be sufficient to identify the ade-
quacy of the proposed construction and to permit preliminary
cost estimating. Operation and maintenance of storm flow
management facilities introduces new and perhaps unfamiliar
requirements, particularly in such areas as startup and
shutdown procedures, sustaining (non-storm) maintenance,
support supplies, and solids handling and disposal. Typical
problems and procedures, introduced in Section XV,
Operation and Maintenance, are among those which should
be addressed in the planning process. Ties to the existing
system and programs should be indicated, and functional
staging of the construction should be anticipated. Public
involvement should be encouraged, and close liaison should
be maintained with agencies having jurisdiction and other
concerned parties.
55
-------
Objectives should be reassessed and potential program per-
formance should be set forth. Mathematical model ap-
proaches, such as those described in Section XIV, intensive
demonstration project review, and results of local monitor-
ing and pilot testing should be considered, and further
candidate screening should be accomplished. A conceptual-
ized block drawing of the planning process is shown on
Figure 6. While the setup is specifically designed for
computer coding, the basic functions and sequence of opera-
tions are generally typical of any approach method.
SELECTION AND IMPLEMENTATION OF FINAL PLAN
Through the preceding analyses, sufficient documentation
should have been accumulated and systematically reported
upon to permit formal program recommendation and review,
and adoption by the public and jurisdictional agencies.
Points of issue may still remain; however, the concept and
initial program of approach must be approved, and funding
must be authorized. Logically, the plan approved will pro-
vide the best combination of the following:
• Satisfaction of program objectives
• Maximum use of existing facilities
• Earliest relief of existing problems
• Widest public acceptance
• Achievability within cost/funding constraints
• Suitability to effective staged implementation
• Flexibility to meet changing needs and technology
• Beneficial reuse/aesthetics
• Minimum adverse impact on environment both
during and after construction
• Coordination with other programs
• Auxiliary benefits
Recognizing the limitations in available data and the rela-
tive immaturity of the state-of-the-art, programs should be
scaled and implemented accordingly. That is, if the suc-
cess of a program is to hinge on the ability of a particular
process or facility to perform up to a certain standard or
56
-------
INPUT
HYDROLOGY
SYSTEM
INVENTORY
DEVELOP
SIMULATION
MODEL
HISTORICAL
PERFORMANCE
ESTABLISH
DESIGN
STANDARDS
AND
OBJECTIVES
SET UP
ALTERNATE
SOLUTIONS
NO
SUPPLY
UNIT
COSTS
REPEAT
FOR ALL
ESTABLISH
COST
BENEFIT
YES
RANK
ALTERNATIVES
SELECT
CAPITAL
IMPROVEMENT
PROGRAM
REPEAT
FOR BEST
RECOMMEND
IMPLEMENTATION
PROGRAM
Figure 6. Conceptual computer application for
master drainage planning for water quality control
57
-------
on the presumption of a specific degree of uniformity in
the wastewater flow and characteristics, then these key
principles should be put to physical tests as soon as pos-
sible in the program. Likewise, if the existing treatment
facilities will be subjected to increased loadings or sub-
stantially altered characteristics under the proposed pro-
gram, pilot testing or alternative means should be imple-
mented to determine the effects on plant performance and
operation. Further, if extended periods of wastewater,
chemical, sludge, or screenings storage is anticipated, the
limitations and consequences should be determined early in
the program. If a proposed system control scheme may sub-
ject certain portions of the collection system to temporary
surcharge loading conditions, then, of course, the piping
should be checked for these loadings or fail-safe safety
devices should be provided.
This text has been structured so that, as the reader pro-
gresses through the subsequent sections, he will become
meaningfully aware of the state-of-the-art in Urban Storm-
water Management and Technology in the context of problem
identification and practicable countermeasure application.
58
-------
Section V
THE STORMWATER PROBLEM
Water pollution from combined sewer overflows, surface run-
off either collected separately or occurring as nonsewered
runoff, and overflows of infiltrated municipal sewage result.
ing from precipitation are all aspects of the stormwater
problem. In this section the emergence of the problem, the
characteristics of stormwater discharges, the sources and
movements of pollutants, and the environmental effects of
stormwater discharges are presented and discussed.
EMERGENCE OF THE STORMWATER PROBLEM
Is there a stormwater problem? The main objective of this
section is to demonstrate that essentially every metropoli-
tan area of the United States has a stormwater problem.
Whether a city has a combined sewer system, or a separate
sewer system, the disposal of stormwater contributes large
quantities of pollutants to nearby receiving waters. Even
nonsewered urban runoff has been shown to be a significant
pollution source [1]. To understand clearly the various
aspects of the stormwater problem, it will be helpful to
review the development of sewer systems.
Historical Development of Sewer Systems
Even in ancient Rome the need to remove rainfall runoff was
recognized. Buildings, plazas, and streets of nonporous
materials interfered with natural percolation and runoff
processes. To remove these excess waters the Romans con-
structed large underground drains--the first storm sewers.
As cities developed in Europe and later in North America,
extensive drainage systems for stormwater runoff were
provided. These drains usually discharged to the nearest
water body. They were reserved for removing stormwater;
solid wastes and human excretion were excluded by ordinance.
Most of these early storm sewers were designed and con-
structed poorly. Because of stone and brick construction
59
-------
large volumes of water infiltrated into the sewers, thereby
reducing the capacity for their intended purpose. In many
cases, because the drains were not sloped properly stagnant
underground pools developed after storms.
The traditional methods utilizing privies and "night soil"
transport for the disposal of human wastes could no longer
be used with the intense population densities fostered by
the industrial revolution. In the slums, courtyards and
living areas became fouled with excrement and wastewater
[21]. Also, medical and civil authorities were beginning
to realize the interrelationship between filthy living
conditions and disease epidemics. To effect a remedy, in
London in 1815, the law was changed to allow disposal of
sanitary wastes via the storm sewers. Hence, what had orig-
inally been storm sewers became combined sewers, receiving
both storm drainage and municipal sewage. Similar actions
occurred in Boston in 1833 and Paris in 1880. However, most
houses in these cities had neither indoor sanitary facili-
ties nor connections to the sewers until many years later.
It was not until the latter part of the nineteenth century
that it was recognized fully that the problem of enveloping
filth had not been solved but merely transferred from the
land to the receiving waters. Those cities less favorably
situated with respect to their receiving waters initiated
the practice of treating municipal sewage prior to discharge.
In other cases, small waterways that became receiving waters,
such as Tiber Creek traversing the Mall in Washington, B.C.,
were totally enclosed and annexed into the combined systems.
To capture the municipal sewage, interceptor sewer systems
were developed to bring dry-weather flow to central loca-
tions for treatment. Since interceptor and treatment sys-
tems were usually (and most still are) designed to convey
or treat only dry-weather flows, relief points were con-
structed at junctions between trunk sewers and interceptor
sewers. These relief points divided the flow during storms
when the combined sewer might be carrying 5, 50, or even
more times the dry-weather flow. A portion of the combined
sewage was intercepted and treated (up to 3 times the dry-
weather flow rate intercepted on the average [27] with
approximately half of this receiving some treatment). The
balance, untreated, discharged directly to the receiving
water. Since the portion of the annual municipal sewage
which discharged without treatment was small (commonly esti-
mated between 2 and 8 percent), it was not thought to be
significant. In addition, relief points were also con-
structed on the interceptor system just ahead of the treat-
ment plant.
60
-------
In line with the more recent practice of installing sewer
systems for conveying separate municipal sewage, it is
surprising to note that, at first, this was not done to
protect local receiving waters. According to the 1928
text, American Sewerage Practice, by Leonard Metcalf and
Harrison P. Eddy [21]:
The construction of a system of separate sewers
without a system of storm drains, or with only a
partial one, has become common practice in small
communities, and is somewhat prevalent in the
larger cities. This has been due generally to
economic necessity, either real or fancied. The
small towns frequently consider it financially
impossible to finance an adequate system of com-
bined sewers, and it is often possible to allow
storm water to flow in gutters and in natural
water courses for many years after the necessity
for separate [sanitary] sewers has become pressing.
However, in some cases separate sewer systems were de-
signed "...where the river or creek is so small that even
diluted sewage from storm-water overflows would be objec-
tionable, especially when the water is to be used for
domestic purposes at no great distance below the town"
[21]. As American cities expanded, combined sewer over-
flows --even to large water bodies--became objectionable,
and the installation of separate sewer systems became
standard practice.
For the past 25 years most cities and towns have required
that all new buildings be provided with separate sewer
systems, one for sanitary wastes and one for storm drainage.
However, most of the largest and oldest cities in the nation
still have combined sewers in a major portion of their ser-
vice area (Table 7).
The proportions of the U.S. population served by combined
sewers, separate sewers, and no sewers, as projected from
1962 data, are indicated on Figure 7 [27] .
Generally, the urban areas which were settled earliest are
the ones with the greatest proportion of combined sewers.
Figure 8, based on the 1962 USPHS inventory, shows that
combined sewers are most extensively used in the northeast
and the Great Lakes region. The average population density
of combined sewer service areas is 13.1 persons per acre,
as compared to 7.4 persons per acre for separate sanitary
sewer service areas. The high average population density
of combined sewer areas is indicative of the problem with
61
-------
Table 7. PREDOMINANT TYPE OF SEWER SYSTEM IN THE
20 LARGEST U.S. CITIES, 1900 and 1970a
20 Largest
City
New York
Chicago
Philadelphia
St. Louis
Boston
Baltimore
Cleveland
Buffalo
San Francisco
Cincinnati
Pittsburgh
New Orleans
Detroit
Milwaukee
Washington
Newark
Jersey City
Louisville
Minneapolis
Providence
cities of 1900
Type of system
Separate Combined
sewer sewer
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
20 Largest
City
New York
Chicago
Los Angeles
Philadelphia
Detroit
Houston
Baltimore
Dallas
Washington
Cleveland
Indianapolis
Milwaukee
San Francisco
San Diego
San Antonio
Boston
Memphis
St. Louis
New Orleans
Phoenix
cities of
Type of
Separate
sewer
X
X
X
X
X
X
X
X
X
1970
system
Combined
sewer
X
X
X
X
X
X
X
X
X
X
X
a. The entire metropolitan area is not included.
SEPARATE
SANITARY
SEWERS
COMBINED SEWERS
/ / / '
,236,000
1962 TOTAL SEWERED POPULATION 125,770,000
Figure 7.
combined sewers
Relative use of
n the United States [21]
62
-------
RATIO OF PROJECTED POPULATION
SERVED BY COMBINED SEWERS TO
TOTAL SEWERED POPULATION, 1962
0-10%
1! -25%
26-50%
51-75%
OVER 75%
Figure 8. Relative use of combined sewers by states [21]
63
-------
sewer separation in many cities--it is costly in terms of
construction and mass inconvenience to excavate in heavily
built-up areas.
The Stormwater Problem — Its Multiple Facets
The stormwater problem is not peculiar to combined sewer
systems. The nation's demand for clean streams, lakes, bays,
and coastal waters requires an evaluation of the quality of
all waters that flow into them. These include rural and
agricultural area runoff, water pollution control plant
effluents, industrial wastewaters, combined sewer overflows,
storm sewer discharges, and overflows of infiltrated munici-
pal sewage. The latter three discharges constitute the
urban stormwater problem.
Combined Sewer Overflows — According to McKee [20] , once the
pavement is wetted, a rainfall of only 0.025 cm/hr (0.01
in./hr) will cause combined sewer overflows in Boston. The
difficulty in alleviating this overflow problem in Boston
is great. With a treatment system for combined wastewater
designed to handle up to 3 times the dry-weather flow,
73 percent of the sanitary wastewater would still overflow
during a 0.25 cm/hr (0.1 in./hr) storm. During the summer
this would produce an average of 5 to 6 overflows each
month.
One of the worst aspects of combined sewer overflows is
that, in spite of dilution by stormwater, the gross BODs
loading of the tremendous volume of wastewater is often
greater than that of the dry-weather flow. Large flows
during storms not only wash pollutants from the air and city
surfaces into the sewers, but also flush out deposited mate-
rial (Figure 9). During dry weather, deposits of silt and
sludge often accumulate at sewer joints and in poorly sloped
sewer lines only to be flushed out during a storm. The re-
sulting shockload may turn a healthy stream or lake into a
septic health hazard.
The weakest element in a combined sewer system is the
regulator control point. For example, through faulty oper-
ation, untreated discharges can be released to receiving
waters at any time. In tidal and flood zones, gross in-
fluxes of estuarine and fresh waters into the conveyance
and treatment systems may occur (Figure 10).
The adverse effects of combined sewer overflows on receiving
waters were largely responsible for the requirement that
separate sewer systems be installed in newly constructed
areas. In some cities programs of "separating" existing
64
-------
(e)
(f)
Figure 9. The stormwater problem-
combined overflow residuals
(a) Solids in overflow structure after intense storm, the next storm's 'first flush'
(b) Floating debris trapped by log boom across stream receiving combined overflows
(c) Overflow occurring at tide gate (d) Interior of gate structure after storm
(e,f) Exterior of gate structures after storm
65
-------
Figure 10. The stormwater problem - beach degradation
(a) Typical combined overflow site (b) Beach scour from force of overflow
(c) Residuals on final bar rack (d) Residuals on beach (e) Tide gate remnants
(i nfIow source)
66
-------
combined sewers have been started. Some of these programs
have been completed, others have been halted. In a 1967
APWA study [27], the cost of converting existing combined
sewers to separate sewers was estimated at $90 billion (re-
vised to ENR 2000) not including the inconvenience to the
local population while city streets are torn up during
construction. Even this would not eliminate the stormwater
problem because the pollution from untreated storm dis-
charges and the overflows from heavily infiltrated sanitary
sewer systems would not be corrected.
Storm Sewer Discharges - Within the past 20 years, it has
Deen recognized that waters discharged from separate storm
sewers contain pollutants. Even without the addition of
sanitary and industrial wastewaters, storm sewer discharges
are usually high in SS and on occasion may have BODr concen-
trations approaching those in municipal sewage. Rain
falling on an urban area picks up pollutants from the air
dusty roofs, littered and dirty streets and sidewalks
traffic byproducts (tire residuals, vehicular exhaust),
galvanic corrosion particulates, and chemicals applied for
fertilization, control of ice, rodents, insects, and weeds
Have you ever noticed the clean air after a rainfall? Air
pollutants have been returned to the earth. Erosion of
hillsides and construction sites caused by rainfall can
produce extremely high concentrations of inorganic SS--
frequently several times higher than those in municipal
sewage. Unauthorized or intentional cross-connections
with sanitary or industrial sewers are common.
Very few sanitary districts have considered storm sewer dis-
charge as a pollution problem. It was thought that sewer
separation would produce two flows--the polluted sanitary
flows and the clean storm flows. Studies by the EPA and
others have shown this not to be the case. This is not
to say that separate sewers are not advantageous in some
cases. In many situations, it will be advisable to give
different types or levels of treatment to the separate
wastewaters depending upon program objectives (see discus-
sion in Section IV).
Overflows of Infiltrated Municipal Sewage - Most sanitary
sewers in the United States are de facto combined sewers.
Stormwater enters these sewers through cracks, unauthorized
(and sometimes authorized) roof and area drains, submerged
manhole covers, improperly formed or deteriorated joints,
eroded mortar in brick sewers, basement and foundation
drains, and poorly constructed house connections. A compari
son of dry- and wet-weather flows in separate sanitary
sewers for various cities is presented in Table 8. The in-
crease in flow during wet weather ranged from 21 percent
67
-------
Table 8. COMPARISON OF DRY- AND WET-WEATHER FLOWS
IN SEPARATE SANITARY SEWERS FOR VARIOUS LOCATIONS [28]
Location
Duluth, Minn.
Denver, Colo.
Grand Island, Neb.
Leavenworth, Kan.
Bay City, Mich.
Dallas, Tex.
Mission, Township,
Main Sewer District
No . 1 , Johnson Co . ,
Kan.
Holland, Mich.
Austin, Tex.
Hays, Kan.
Paragould, Ark.
Osawatomie, Kan.
Sheridan, Wyo .
Billings, Mont.
Billings, Mont.
Year
1967
1956
1962
1967
1967
1960
1969
1969
1966
1965
1962
1966
1960
1960
1968
Popula-
tion
110,000
510,000
26,000
41,000
60,000
715,500
50,000
19,300
270,000
15,600
7,700
4,700
11,500
50,500
59,500
Dry-weather
mgd
11.
40.
3.
5.
16.
117.
17.
4.
19.
2.
1.
1.
4.
10.
10.
flowa
cu m/day
3
0*
9
0*
0
0
5
3
4*
5
1
2
3
8
5
42
151
14
18
60
442
66
16
73
9
4
4
16
40
39
,750
,400*
,750
,950*
,500
,500
,200
,300
,500*
,460
,160
,540
,300
,800
,700
Wet
-weather flow
mgd
30
57
6
10
70
226
104
5
46
4
6
2
17
17
14
.0
.0*
.7
.0*
.0
.0
.0
.22
.5*
.7
.0
.9
.0
.7
.0
cu m/day
113,500
216,000*
25,350
37,850*
265,000
855,000
394,000
19,750
176,000*
17,800
22,700
10,980
64,300
66,900
53,000
Remarks
Bypassed storm flow
not included
--
--
Much wet-weather flow
bypassed
Considerable flow
bypassed
Wet-weather flow
estimated
--
Wet- weather flow
estimated
--
--
Wet-weather flow
estimated
Wet-weather flow
estimated
Some seepage from
irrigation ditches
--
Following program to
reduce infiltration
a. All flows are peak, except those marked (*) which are average.
68
-------
in Holland, Michigan, to 495 percent for Mission Township,
Main Sewer District No. 1, Johnson County, Kansas. It
is noted that, in addition to the large wet-weather flows
recorded, considerable unmeasured amounts of the flow were
bypassed in several of the systems. This table includes
peak flows for some cities and average flows for others.
Water Pollution Control Plant Bypasses - A special category
of combined sewage overflow occurs just upstream of the
treatment facilities. Generally, all flows up to the system
transport capacity, unless purposely diverted, are carried
to the water pollution control plant. A portion of the
flows exceeding transport capacities pond in low areas until
system capacity is available or they are reduced by percola-
tion into the ground, and/or are diverted either to storm
drains or directly to receiving waters. Of the flows
arriving at the plant, not all may be treated.
Most water pollution control plants are designed to function
properly at flows up to some low multiple of the average dry-
weather flow. Typical multiples range from 1.5 to 3.0.
Incoming flows exceeding the plant or unit process capacity
must be bypassed, sometimes with partial treatment, to the
receiving water. This occurs at both wastewater treatment
plants connected to combined sewers and, in many cases,
those connected to technically "separate" sanitary sewers.
In addition to the need to bypass peak wet-weather flows
because of capacity limitations, it is sometimes necessary
to bypass because of detrimental changes in wastewater
quality. In a study of stormwater problems and control in
Oakland-Berkeley, California, sanitary sewers [34], it was
reported that bypassing was required at the regional water
pollution control plant under storm conditions at flows less
than 50 percent of the plant's hydraulic capacity. Typi-
cally, the large increase in fine solids (silt) concentration
in the wastewaters required bypassing to avoid damage to
the solids removal equipment in the primary sedimentation
tanks. This silt, too fine to be removed in the grit cham-
bers, settled in such quantities in the sedimentation tanks
that the chain and flight collectors were buried and unable
to function. Sources of the silt were traced to multiple
cracks and poor joints in the collection system. As storm-
waters saturated the soil, silt-laden flows would enter
the system and in the process compound their impact by
scouring out deposits of grit accumulated in the pipes
during dry-weather periods.
In San Francisco, with a hydrologic year characterized by
extended dry- and wet-weather periods, the initial seasonal
storms in the combined system carry debris, representing
69
-------
several months' accumulation, in such quantities as to
greatly overload headworks screening facilities. This cre-
ates a bypass condition generally not repeated in later
storms [15] .
In some cases it has been necessary to bypass prolonged
storm flows because low organic matter concentrations would
not support the biological treatment units [25] . Elsewhere
intense hydraulic loadings during storms may purge biologi-
cal process solids from final clarifiers and/or filters,
critically upsetting the process, if temporary bypasses are
not permitted. Still other plants bypass during wet weather
because increased inorganic concentrations in the sludge
pumped to anaerobic digesters greatly reduce their effec-
tive volume.
CHARACTERISTICS OF STORMWATER
Quantities of Flow
The potential problem that stormwater represents to all
wastewater disposal systems can be demonstrated with some
simple calculations. For example, in the Chicago metro-
politan area where the average daily dry-weather wastewater
flow is 48.2 cu m/sec (1,100 mgd), a storm with an inten-
sity of only 0.25 cm/hr (0.1 in./hr) over the 97,000 ha
(240,000 acres) within the area represents a flow of
0.25 cm/hr x 97,000 ha x 100 cu m/cm-ha
x 1/3,600 hr/sec = 686 cu m/sec (18,000 mgd)
Of course, large portions of the drainage basin are still
pervious so that much of the rainfall will permeate into the
soil. Also, there will be significant portions of nonsewered
direct runoff into creeks and waterways. Even so, if the
runoff coefficient is 0.50, under average conditions the
flow in the combined sewer system would increase 720 percent.
Consider what would happen if a large storm with an inten-
sity of 2.5 cm/hr (1 in./hr) covered the Chicago metropoli-
tan area! Summer thunderstorms of even greater intensity
occur every year, although they rarely cover the entire
basin at one time.
In fact, the stormwater problem in Chicago is severe for a
number of reasons. Sewer surcharging and widespread flood-
ing is common. In addition, as the imperviousness of the
area has increased with development, it occasionally has
become necessary to allow combined wastewater to backflow
70
-------
into Lake Michigan to minimize flooding. The lake, which is
used for water supply and for recreation, does not receive
dry-weather flow from Chicago. Most combined wastewaters
from Chicago discharge to several channels that eventually
empty into the Illinois River system. Chicago has under-
taken a series of studies to determine the best way to
alleviate the flooding and pollution. Current estimates for
wet-weather improvements are for eventual expenditures
of approximately $1.32 billion. The program is described
further in Section XIV.
To describe the proportion of precipitation that enters a
combined or separate sanitary sewer, an Infiltration ratio
may be computed. This is the ratio of rainfall entering
sewers to the total rainfall and thus includes both infiltra-
tion and inflow as defined in the 1972 Act Amendments. For
combined sewers, the infiltration ratio may be equivalent to
the runoff coefficient. In an EPA-sponsored study of the
wastewater collection/treatment area, 20,800 ha (51,400
acres), of the East Bay Municipal Utility District (Oakland/
Berkeley, California) [34] an overall infiltration ratio of
0.11 was computed. However, 30.6 percent of the stormwater
in the system was traced to the 4 percent of the study area
served by combined sewers. The measured infiltration ratios
for the combined areas were as high as 0.70 and the sep-
arated areas ranged from 0.01 to 0.25 with values of 0.06 to
0.14 predominanting. In general, high ratios were associ-
ated with old sewers, those with rigid joints, and land
areas having gentle ground slopes.
St. Louis, Missouri - One often-reported measurement of the
overflow from combined sewers is the proportion of the
annual dry-weather flow that either overflows or is bypassed
According to Shifrin and Homer, 2.23 to 3.09 percent of the'
annual municipal sewage of St. Louis is discharged without
treatment during combined sewer overflows [30]. The rela-
tionship between combined sewer overflow of untreated munic-
ipal sewage and interceptor capacity for St. Louis in 1960
is shown on Figure 11. Of course, if the hydraulic capacity
of the treatment plant is less than the interceptor capac-
ity, then significant additional amounts must be bypassed
at the treatment plant.
Washington, D C. - According to a study of the combined
sewer areas of Washington, D. C., during the years 1956-1958
an average of 3.3 percent of the annual municipal sewage was'
discharged to the Potomac River system through overflows
[30]. Of this quantity, 0.3 percent discharged during
dry weather because of an overloaded sewer system. Air
conditioning cooling water is in part responsible for these
dry-weather overflows.
71
-------
3.5
LLJ C3
CO ^
CJ
oo
2.5
2,0
1.5
100
150
200
250
NTERCEPTOR CAPACITY,
% OF AVERAGE FLOW
Figure 11. Effectiveness of interceptors
of different capacities for 1960 [30]
St. Louis, Missouri
Oakland, California, and Vicinity - The heavily infiltrated
sanitary sewer system of the East Bay Municipal Utilities
District required treatment plant bypasses of 1.7 percent of
the annual average municipal sewage flow during 1969-1970
[34]. In addition, intersystem overflows (occurring at
cross-connections between storm drains and sanitary sewers),
interceptor overflows, and intrasystem overflows (from man-
holes, other openings, and broken sewers) also contributed
to the discharge of untreated municipal sewage to San
Francisco Bay. As shown in Table 9, it has been found in
other studies of combined sewer systems that, typically, 2
to 5 percent of the annual municipal sewage is "lost" during
wet weather [11, 23, 9].
In the past it was thought that if only 2 to 5 percent of
the annual municipal sewage flow was discharged by waste-
water collection systems without treatment, then the system
was working well. After all, an efficiency of 95 to 98
percent on an annual basis seems impressive. Actually,
it is not impressive on two counts. First, by reporting
72
-------
Table 9. COMPARISON OF ANNUAL MUNICIPAL SEWAGE BYPASSED
OR OVERFLOWED FOR VARIOUS CITIES
City
Cleveland, Ohio
Washington, D.C.
Year
1960
1956-58
Sewer
system
Combined
Combined
Annual \
overflowed
2.23-3.09
3.3a
Oakland (and
vicinity), California 1968 961 Separate) b
4% Combined} 1>09
Roanoke, Virginia 1964-68 Separate 1-2
Minneapolis-St.Paul,
Minnesota 1966 Combined/ up to 6
separate
San Francisco,
California 1972 Combined 1.7
a. This is only for the combined sewer portion of the
system.
b. Only includes bypasses of treatment plant. Additional
flow is lost through sanitary overflows into storm
sewers.
municipal sewage lost as a percentage of annual municipal
sewage collected, the shockloading effect of stormwater
pollution is ignored. For instance, in some rivers suffi-
cient dissolved oxygen (DO) is present to support a wide
range of biota 90 percent of the time. During wet weather,
however, storm and combined sewer overflows exert such
an oxygen demand that the river becomes septic and fishkills
occur. In many parts of the country, summer thunderstorms
dump an inch or more of rain in an hour. Because these
overflows usually occur during low flow in the receiving
stream, this accentuates the shockloading effect.
The second reason that a 95 to 98 percent annual efficiency
for a wastewater collection system is not impressive re-
lates to the quality of overflows and discharges.
Quality of Overflows and Discharges
Historically, few people gave any consideration to the water
quality of urban runoff. It was believed that urban runoff
73
-------
finding its way into a combined sewer diluted the offensive
constituents of the dry-weather flow. In the case of sepa-
rate sewer systems, the urban runoff entered the storm
sewers. This storm wastewater was discharged without treat-
ment because people thought that it was relatively
uncontaminated. When the U.S. Public Health Service pub-
lished "Pollutional Effects of Stormwater and Overflows from
Combined Sewer Systems: A Preliminary Appraisal" in 1964,
very few studies of urban runoff quality had been performed.
The fact that the report title included "Pollutional Effects
of Stormwater" indicates that at least some officials were
aware of the water pollution potential of urban runoff.
In recent years, however, a considerable number of charac-
terization studies have been performed. The reported qual-
ity parameters (e.g., BOD5, SS, total P, total N, heavy
metals, and bacteria) vary considerably in concentration.
They vary not only with time as the storm progresses but
also from location-to-location during the same storm and
from storm-to-storm within the same location. Where the
runoff is mixed with municipal sewage, as in combined sewer
systems, the true picture is further obscured because the
sanitary flow and its quality are also highly time-dependent
(daily and weekly cycles).
Because of these multiple variations and the difficulties
associated with representative sampling, relationships be-
tween cause and effect are largely obscured, even though
a considerable amount of data is available.
In most studies of urban runoff it has been observed that
higher concentrations of pollutants may be expected under
the following conditions: the early stages of a storm
(including "first flush" effects); in more densely settled,
highly paved, or industrialized areas; in response to in-
tense rainfall periods; after prolonged dry periods; and in
areas with construction activities (inorganic solids).
Conversely, concentrations tend to decrease as a storm
progresses and in the latter storms of a closely spaced
series.
Along with summarizing the available data, selected examples
are presented in this section to illustrate various occur-
rences suggested by the data.
First Flush Effect - Very high concentrations of BODs, SS,
grease, and other pollutants are often found in overflow
samples collected during the earliest part of an overflow
event from combined sewers. This phenomenon also occurs
to a lesser extent with storm sewer discharges. Early inves
tigators often noted these high pollutant concentrations and
74
-------
the term "first flush" has been adopted to describe it.
During dry-weather periods, the flow in most combined sewers
is only a low percentage (generally less than 5 percent) of
the sewer capacity. At joints and along sections of sewer
with low or adverse slopes, solids tend to accumulate.
In particular, older sewers were often laid with too little
regard to necessary carrying velocities. The absence of a
self-scouring flow is the primary cause of the first flush
effect in combined sewers. Catch basins that are not
cleaned promptly after a storm often accumulate solids
which are discharged during the next storm. Other effects
include the surface buildup of debris and pollutants through
inattention or inadequate cleaning programs.
In storm sewers there is a tendency for solids to settle
out during the latter stages of a storm as the flow tapers
off and velocities are reduced. Also, large separate sewer
systems may have relief points that allow some surcharged
sanitary sewers, even on rare dry-weather occasions, to
overflow into the storm sewers. The solids in the municipal
sewage overflows may accumulate in storm sewers under these
conditions and contribute to a first flush effect in storm
sewers. A similar situation exists when storm sewers have
illicit direct industrial and sanitary connections.
In combined, sanitary, and storm sewers, accumulated solids
deposits may contain grease and other organic matter under-
going decay. When the sewer flow increases sharply during
a storm, this solid matter which will exert a high organic
loading will be discharged. Depending on the sewer system
the rainfall intensity, and the number of antecedent dry '
days, a first flush effect may result. If it does occur
it may last for a few minutes or even hours. Solids scoured
from the upper reaches of a large system may take a long
time to reach a distant overflow point. BOD5 concentrations
dur^ng th^s phenomenon may often exceed those of the normal
untreated dry-weather wastewater.
During 28 overflows of a large combined sewer in Milwaukee
Wisconsin, in 1969, a first flush was observed 25 percent '
of the time [7]. The period of high pollutant concentra-
tion lasted from 10 minutes to 1 hour. The difference in
water quality between the first flush and the later flow
("extended overflow") from this particular combined sewer
is reported in Table 10. These first flushes occurred
when the interval between overflows was greater than 4 days.
75
-------
Table 10. COMPARISON OF QUALITY CHARACTERISTICS
FROM FIRST FLUSHES AND EXTENDED OVERFLOWS OF A
COMBINED SEWER [7]
HAWLEY ROAD SEWER, MILWAUKEE, WISCONSIN
First flushes, Extended overflows,
Characteristic mg/la mg/la
COD
BOD5
ss
vss
Total N
500
170
330
221
17
-765
-182
-848
-495
-24
113
26
113
58
3
-166
-53
-174
-87
-6
Coliforms
a. At 95 percent confidence level.
b. Coliform 1.5 x 105 to 310 x 105/100 ml.
In an EPA-funded project in Cumberland, Maryland, it was
observed that:
On June 16, 1968, a rain storm producing 1/4 inch
rainfall occurred after a dry period of 8 days.
The Boiling Green Wastewater Treatment Plant in-
fluent was sampled according to the plan....The
initial sample obtained was gray, appeared to
be normal late night flow, but after only two
minutes the rate began to increase rapidly and
the sewage became black and gave off a very
strong odor, indicating septicity. This odor
did not disappear until the flow again returned
to nearly normal. The results of Sample No. 2
indicate the flushing of grit and putrescible
material undergoing anaerobic digestion from the
bottom of the sewers. [10]
Combined Sewage — Since heavily infiltrated sanitary sewers
are de facto combined sewers, the water quality information
presented in this discussion relates to both heavily infil-
trated municipal sewage and combined sewer overflows.
Combined sewage is a mixture of various proportions of
municipal sewage and storm runoff. For this reason, it
76
-------
would seem that the concentration of any single pollutant
in combined sewage would lie somewhere between that of
local municipal sewage and that of local urban runoff.
Even though it may seem so, it is often not the case, mainly
during the early portions of a storm. The previously de-
scribed first flush effect often results in short periods
when combined sewage exhibits greater strength than separate
municipal sewage. When deposition problems are severe and/
or the collection systems extensive, the high concentration
conditions may persist for several hours. In other in-
stances, strong industrial wastes may be discharged
(unauthorized) during storms, perhaps in the form of waste
lagoon spills, in the belief that the additional dilution
available will lessen the impact of the discharge on the
treatment plant.
Because of the variation over time, it is also somewhat mis-
leading to report mean characteristics of combined sewage
quality. This is also true, but to a lesser extent, for
separate storm runoff. Average values for the various
characteristics can be mean over time or mean over flow.
Furthermore, the first 10 minutes, the first 30 minutes,
or some other duration of the overflow may not be included
in the calculations of a quality characteristic mean. In
spite of the possible confusion, most investigators have
reported "mean" characteristics of combined sewage to give
a general indication of quality. Some of these findings and
values for typical untreated and treated municipal sewage
(primary and secondary effluent) are summarized in Table 11.
Unless otherwise noted, tabulated values are mean over
flow (also referred to as "flow-weighted mean"), the pre-
ferred definition of "mean."
In contrast to urban runoff quality, which may be highly
dependent upon runoff intensity, time since start of rain-
fall, and antecedent dry period, combined overflow quality
may be affected significantly by the hour of storm
occurrence. It is commonly known that dry-weather flows
and quality concentrations follow cyclical patterns over
daily and weekly periods. What is of particular interest
is that the cycle peaks for both flow and quality (e.g.,
BOD5) frequently occur together, resulting in a compound
effect. In a prior study of San Francisco data on dry-
weather flows, it was noted:
Measured values of both flow and quality varia-
tions showed extreme variation of pounds of [BO
released per 10-minute interval exceeded a ratio
of 50 to 1 (largest 10-minute release divided by
the smallest) in one day. [32]
77
-------
Table 11. COMPARISON OF QUALITY OF COMBINED
SEWAGE FOR VARIOUS CITIESa
BODc ,
Type of wastewater, mg/1
COD,
mg/1
DO,
mg/1
SS,
mg/1
Total
coliforms ,
MPN/100 ml
Total
nitrogen,
mg/1 as N
Total
phosphorus ,
mjj/1
as P
Avg Range Avg Range Avg Avg Range
Range
Typical untreated
municipal
Typical treated
municipal
Primary effluent
Secondary effluent
Selected combined
Atlanta, Ga.,
1969 [31]
Berkeley, Calif.
1968-69 [34]c
Brooklyn, N.Y. ,
1972 [8]
Bucyrus, Ohio
1968-69 [35]
Cincinnati, Ohio,
1970 [36]
Des Moines, Iowa,
1968-69 [6]
Detroit, Mich.,
1965 [2]
Kenosha, Wis.,
1970 [18]
Milwaukee , Wis. ,
1969 [7]
Northampton, U.K.,
1960-62 [22]
Racine, Wis. ,
1971 [18]
Roanoke, Va. ,
1969 [12]
Sacramento, Calif.,
1968-69 [37]
San Francisco, Calif.
1969-70 [3]
Washington, D.C.,
1969 [5]
200 100-300 500 250-750
135 70-200 330 165-500
25 15-45 56 25-80
100 48-540 --
60 18-300 200 20-600
180 86-428 --
220 11-560 400 13-920
200 80-380 250 190-410
115 29-158 --
1S3 74-685 115
129 -- 464
55 26-182 177 118-765
ISO 80-350
119
115
1SS 70-328 238 59-513
49 1.5-202 255 17-626
71 10-470 382 80-1,760
200 100-350
80 40-120
15 10-30
<20 1x10-1x10
lxlO"-l*10
200 40-150
2,052 132-8,759
470 20-2,440
2,200 SOO-1,800
2S5 155-1,166
274 120-804
458
244 113-848
400 200-800
439
78
125 56-502
68 4-426
622 35-2,000
2*20 2x10-5x10
2*10-3x10
5*20 7x10-9*10
2*10-2*10
3*10 4xl0-6*10
a. Data presented here are for general comparisons only. Since different sampling methods, number of samples, and other
procedures were used, the reader should consult the references before using the data for specific planning purposes.
b. Only orthophosphate.
c. Infiltrated sanitary sewer overflow.
d. Only ammonia plus organic nitrogen (total Kjeldahl).
e. Only ammonia.
f. Only fecal.
78
-------
The storm path and collection system configuration also have
pronounced influence on combined overflow quality. For ex-
ample, in a storm concentrated in the upper reaches of a
drainage basin or tracing a path from upstream to down-
stream, flow in the conduits in the upper reaches will have
greater depth and velocities as compared to the downstream
portions. This can create a flood wave within the primary
conduit that will accelerate the dry-weather flow in the
lower area as a plug until it is released at the point of
overflow. Thus, the early overflow may be totally municipal
sewage. In current studies by the Department of Public
Works of the City of San Francisco [4], it has been found
that on the basis of system hydraulics and static regulator
inefficiencies, discharge mixtures from neighboring outfalls
may vary simultaneously from totally raw municipal sewage to
dilute surface runoff. This is judged to be a direct conse-
quence of static control of regulators.
Storm Sewer Discharges - Findings from various studies are
summarized in Table 12 showing important characteristics of
various separate storm sewer discharges in comparison with
those for typical untreated and treated municipal sewage
(primary and secondary effluent). The most obvious conclu-
sion about the quality of storm sewer discharge is that it
varies greatly from one metropolitan area to another.
Furthermore, that a wide variation in quality can occur
within a single metropolitan area is shown clearly by the
data from Tulsa, Oklahoma, in Table 13 [33]. The greatest
variations in quality occur in the concentrations of bac-
terial and suspended solids. Note that the data presented
from Tulsa do not indicate the possible range of quality
from a single test area. Storm sewer discharge quality was
found to vary greatly from storm to storm and at different
times during a storm. The column in Table 13 "Range of the
Test Area Means" is based on the mean values of various
water quality characteristics derived from 15 test areas.
Castro Valley, California - In a 1971-1972 study [14] of a
"typical" San Francisco Bay Area urban watershed consisting
principally of residential with some light commercial
areas, the USGS, under contract to the Corps of Engineers,
monitored seven storm events. The watershed, which drains
about 13 sq km (5 sq mi) through a series of pipe networks
and creeks, is about 85 percent urbanized, and has an esti-
mated population density of 11 persons per acre. The events
sampled represent 25-30 percent of the total rainfall in
79
-------
Table 12. COMPARISON OF QUALITY OF STORM SEWER
DISCHARGES FOR VARIOUS CITIESa
Type of wastewater,
location, year,
Ref. No.
Typical untreated
munic ipal
Typical treated
municipal
Primary effluent
Secondary effluent
Storm sewer
discharges
Ann Arbor , Mich . ,
1965 [2]
BODs, COD,
mg/1 mg/1
Avg Range Avg Range
200 100-300 500 250-750
135 70-200 330 165-500
25 15-45 55 25-80
28 11-62
Total
DO , SS, coliforms ,
mg/1 mg/1 MPN/100 ml
Avg Avg Range Avg Range
200 100-350 5*10? 1*107-1*109
80 40-120 2*707 5xl06-5xl08
15 10-30 1*10Z 1*102-1*104
2,080 650-11,900
Total
nitrogen,
mg/1 as N
Avg
40
35
30
3. 5
Total
phosphorus ,
mg/1 as P
Avg
10
7 . 5
5. 0
1 . 7
Castro Valley,
Calif.,1971-72 [14]
Des Mo-ines , Iowa ,
1969 [6]
Durham , N .C .
1968 [1]
4-37
12-100
2-232
95-1,053
I.os Angeles ,
Calif.,1967-f
Madison, Wis.
1970-71 [17]
[19]
4*10-6*10
10-1 ,000
New Orleans, La.,
1967-69d [16]
Roanoke, Va.,
1969 [12]
Sacramento, Calif.
1968-69 [37]
Tulsa, Okla.,
1968-69 [33]
Washington , D.C. ,
1969 [5]
10-7*10
58 21-176
85 12-405
335 29-1,514
71 3-211
247 84-2,052
1,697 130-11,280
I*10-3xl0
a. Data presented here are for general comparisons only. Since different sampling methods, number of samples, and
other procedures were used, the reader should consult the references before using the data for specific planning
purposes.
b. Only ammonia plus nitrate.
c. Only fecal.
d. Median values from 1 sampling station.
e. Only organic (Kjeldahl) nitrogen.
f. Only soluble orthophosphate.
80
-------
Table 13. STORM SEWER DISCHARGE QUALITY FROM
15 TEST AREAS [33]
TULSA, OKLAHOMA
Parameter
Mean of the
test areas
Range of the test
area means
Bacterial, number/100 ml'
Total coliform
Fecal coliform
Fecal streptococcus
Organic, mg/1
BOD
COD
Total organic carbon
Nutrients, mg/1
Organic Kjeldahl
nitrogen
Soluble orthophosphate
Solids, mg/1
Total
Suspended
Dissolved
Other parameters
PH
Chloride, mg/1
Specific conductance,
micromhos/cm
87,000
420
6,000
11.8
85.5
31.8
0.85
1.15
545
367
178
7.4
11.5
108
5,000-400,000
10-18,000
700-30,000
8-18
42-138
15-48
0.36-1.48
0.54-3.49
199-2,242
84-2,052
89-400
6.8-8.4
2-46
56-220
a. Geometric mean.
this dry year. A summary of the data, reproduced in
Table 14, indicated that
...relatively high BOD can be expected from the
first runoff event of the rainy season and also
for other events during the year that occur
after a significant dry spell. The dissolved
oxygen (DO) data expressed as percent saturation
shows a significant corresponding decrease asso-
ciated with the high BOD. [14]
Des Moines, Iowa — In a third study [6] conducted in Des
Moines, Iowa, combined sewer overflows and storm sewer dis
charges were sampled over a 12-month period, 1968-1969.
81
-------
Table 14. STORM SEWER DISCHARGE QUALITY FROM A
5-SQUARE MILE URBAN WATERSHED [14]
CASTRO VALLEY CREEK, CALIFORNIA
San Irancisco Bava
19ftO-1964
Parameter
Temperature ,
°C
pll
DO, mg/1
DO , \
saturation
BODS, mg/1
Ammoni a
nitrogen ,
mg/1
Nitrate
nitrogen ,
mg/1
Dissolved
silica, mg/1
Total
col if orm
bacteria ,
MPN/100 ml
low
mean
high
low
mean
high
low
mean
high
low
mean
high
low
mean
high
low
mean
high
low
mean
high
low
mean
high
low
mean
high
South
Bay
9.3
16.3
24.0
7.6
8.0
0 . 7
3. 1
8 .3
9
55
92
0.5
10
48
3
11
0.05
0.35
1.1
2.3
8.7
16
10
20,000
3 x 108
Lower Upper Storm of
Bay Bay 11 Nov 71
10.
14.
21 .
8 .
7.
7 .
8 .
81
90
99
0.
0.
1 .
0.
0.
0.
0.
0.
0.
2.
5.
7 .
30,
7 14.5
8 14.5
0 15.0
8 b.7
9 7 . 7
1 7.0
0 4.4
4 7.9
5 5.1
43
77
50
4
8 .8 44
5
06 1.2
12 .1
21 2.3
08- 0
34 .3
55 1.4
9
4 6.5 7.1
7
10
500 1,000
000
Castro Valley Creek
Storm of Storm of Storm of Storm of Storm of
13 N'ov 71 2 Dec 71 9 Dec 71 22 Dec 71 27 Dec 71
9.5 8.
13.0
10.5 10.
6.7 5.4 6.
6.9 6.4 7 .
9.5 9.
8 .1
10.3 9.
84 79 85
90 86 90
4.0 10.
9.5 11 .
.4 .3
.6 .7
1.5 .6
1.7 1.2 3.
3.3 1.5 12.
4,
>16,000
41,
5 8.0
11 .0
5 9.0
6 6.0 6.6
4 6.4 6.9
0 9.4
10.4
6 10.0
88
0 1.7 1.7
0 5.0 2.2
1 .2 .1
4 .3 .2
3 1 .8
3 2.3
0 2.2 7
200 9,500 5,200
000 12,500 16,200
Storm of Storm of
25 Jan 72 5 Apr 72
7 .5
15.
8 .5
6.2
7 .
6 .8
6.
6.
63
76
68
4.7 2.
6.0 37 .
.3
.4 1.
.6 0
.9 4.
3.1 7.
16,
63,
0
2
4
9
6
0
3
0
2
6
000
000
a. From "Interim Water Quality Control Plan, San Francisco Bay Basin," California Regional Water Quality Control Board
San Francisco Bay Region, June 1971.
*Determination Pending.
With the exception of BOD5, which was considerably stronger,
the reported values were similar in range to those previ-
ously discussed. The results, however, were an aggregation
of composite and grab samples of both rainfall-runoff and
snow melt. A retabulation of the results for BODs taken
from the raw data, excluding the grab samples and separating
out the snow melt, is presented in Table 15. The revised
listings are consistent with the generally reported values,
and it is particularly interesting to note high concentra-
tions associated with urban snow melt.
Because of the wide range of quality of storm sewer dis-
charge, any generalizations must not be blindly accepted.
However, as mentioned previously, "typical" storm waste-
water is often characterized as having (1) SS concentration
82
-------
Table 15. COMPARISON OF BOD5 FROM COMBINED OVERFLOWS,
STORM DISCHARGES, AND SNOW MELT [6]
DES MOINES, IOWA
Type
of discharge
Mean of the
test areas ,
mg/1
Range of the
test area means ,
mg/1
Combined overflows 80 53-117
(5 areas)
Storm sewer discharge 32 23-46
(rain induced -
4 areas)
Storm sewer discharge 75 67-85
(snow melt - 4 areas)
equal to or greater than that of untreated sanitary waste -
water, and (2) BODs concentration approximately equal to
that of secondary effluent. The reasons for the great
variations in the quality of storm wastewater will be dis-
cussed in a later section dealing with contaminant sources.
Stormwater Pollution Loadings
So far, storm and combined sewer overflow quantities and
the pollutant concentrations have been considered
separately. To evaluate the impact of these flows on re-
ceiving waters, it will be necessary to consider quantity
and quality concurrently. Comparisons of the amounts of
untreated storm and combined wastewaters and amounts of
treated wastewaters received by a stream are only part of
the story. Similarly, comparing water quality characteris-
tics does not give a representative picture of the signifi-
cance of Stormwater problems. Rather, it is necessary
to evaluate pollutant loadings, the product of quantity and
concentration, which should be considered on both a storm
and an annual basis.
Roanoke, Virginia - In studies sponsored by the EPA Storm
and Combined Sewer Pollution Control Program, attempts
have been made to evaluate the relative significance of
Stormwater problems. An especially lucid example comparing
the impacts of dry- and wet-weather flows is in the investi-
gation of overflows and bypasses of the heavily infiltrated
sanitary sewer system of Roanoke, Virginia. The amounts
83
-------
of wastewater and pounds of BOD5 from the treatment plant,
treatment plant bypasses, and sanitary sewer overflows on
an annual basis are summarized in Table 16. Note that the
BOD5 from storm sewer discharges is not being considered.
Although only 1.6 percent of the annual municipal flow is
discharged through overflows or bypasses, 8.4 percent of the
annual BODs loading on the Roanoke River is contributed by
these sources. If storm sewer discharge BOD$ were also
considered, the stormwater BODs percentage would be even
greater. A visual comparison of BOD5 contributions is
given in Figure 12.
On an annual basis, the 8.4 percent BOD5 contributed by over
flows and bypasses from the Roanoke sanitary sewer system
does not seem especially significant. However, as indicated
in Table 17 and Figure 13, wastewater volume and BOD5 load-
ings during a maximum yearly rainfall event are significant.
During such an event, 33.1 percent of the flow is bypassed
or overflows, and this accounts for 75.7 percent of the
BODs loading. Hence, in Roanoke, with secondary treatment
of its sanitary wastewater, the BODs loading on the river
increases fivefold from 2,700 kg (6,000 Ib) during a dry-
weather day to 14,000 kg (31,000 Ib) during a 17-hour storm.
The impact of these loadings on the receiving waters is dis-
cussed under environmental effects. If it were necessary
for Roanoke to undertake advanced wastewater treatment
of dry-weather flow, the relative significance of stormwater
pollution would be even greater.
Bucyrus, Ohio — Overflows of the combined sewer system
of Bucyrus, Ohio, contribute approximately 14,000 kg (30,000
Ib) of phosphates (as P04) annually to the Sandusky River.
By comparison, effluent from the city activated sludge
plant contains 73,000 kg (160,000 Ib) each year, and up-
stream rural runoff contributes 50,000 kg (110,000 Ib)
each year. Unless the city undertakes phosphorus removal
of dry-weather flow, the phosphate from combined sewer over-
flows will not be especially significant on an annual basis.
However, a requirement of 90 percent phosphorus removal
would greatly increase the relative significance of the
overflows. Under these circumstances, the rural runoff
would become the most significant source of phosphates.
Tulsa, Oklahoma — In a recent EPA-sponsored study, the
pollutant loadings from storm sewer discharges as compared
to those from treatment plants were evaluated for Tulsa,
Oklahoma, which is serviced by separate sewer systems.
These loadings on an annual basis for BOD5, COD, SS, organic
(Kjeldahl) nitrogen, and soluble orthophosphate are summa-
rized in Table 18. It is important to note the difference
in pollutant loadings. For example, although the storm
84
-------
Table 16. AVERAGE ANNUAL BODc CONTRIBUTED TO THE
ROANOKE RIVER BY MUNICIPAL SEWAGE [27]
ROANOKE, VIRGINIA
Source
of BOD5
Sewage
mil gal
volume
% of
total
BOD5
1 of
Ib total
Treatment plant
effluent
Treatment plant
bypasses
Sanitary sewer
overflows
Total
7,300 98.4
45
79
0.6
1.0
7,424 100.0
2,192,000 91.6
90,000 3.8
111,000 4.6
2,393,000 100.0
Note: mil gal. x 3.785 = Ml
Ib x 0.454 = kg
PLANT EFFLUENT
2. 192,000 LB
NOTE: LB x 0.454
TREATMENT
PLANT BYPASS
90,000 LB
SANITARY SEWER
OVERFLOWS
111,000 LB
Figure 12. Average annual BOD5 contributed to
the Roanoke River by municipal sewage [27]
85
-------
Table 17. BOD5 CONTRIBUTED TO THE ROANOKE RIVER BY
MUNICIPAL SEWAGE DURING MAXIMUM YEARLY RAINFALL EVENT [27]
ROANOKE, VIRGINIA
Source of BOD5
Sewage volume
% of
mil gal. total
BOD5
% of
Ib total
Treatment plant
effluent
Treatment plant
bypasses
Sanitary sewer
27.5 66.9 7,570 24.3
7.4 18.0 14,810 47.5
overflows
Total
6.2
41.1
15.1
100.0
8,790
31,170
28.2
100.0
Note: mil gal. x 3.785 = Ml
Ib x 0.454 = kg
TREATMENT
PLANT EFFLUENT
7,570 LB
TREATMENT
y PLANT BYPASS
SANITARY SEWER
OVERFLOWS
8,790 LB
NOTE: LB x 0.454 = KG
Figure 13. BOD5 contributed to the Roanoke River by
municipal sewage during maximum yearly rainfall event [27]
86
-------
Table 18. ESTIMATED ANNUAL LOAD OF POLLUTANTS
ENTERING THE AREA RECEIVING STREAMS [33]
TULSA, OKLAHOMA
Pollutant
Soluble
orthophosphate
Average annual
storm sewer
discharge
pollution load,
Ib
Contribution of
storm sewer
discharges
to total load,
171,000
1968 Average
annual load
from treatment
plant effluents
Ib
BOD5
COD
SS
Organic
(Kjeldahl)
nitrogen
1,620,000
11,200,000
39,000,000
130,000
20
31
85
31
7,060,000
24,400,000
6,710,000
278,000
4,180,000
a. One primary (21 mgd) and three secondary (19 mgd, total) plants
Note: Ib x 0.454 = kg
mgd x 0.0438 = cu m/sec
sewer discharges contribute only 20 percent of the annual
BODs to the area receiving streams, this source contributes
85 percent of the annual SS. Notice that the storm sewer
discharge COD contribution is 31 percent as compared to
the BODs contribution of 20 percent. This indicates that
the material in the storm sewer discharges will be more
slowly oxidized than that in the treatment plant effluent.
In this section, the magnitudes of the contaminants found
in urban stormwater have been considered. Before discussing
corrective measures, it is also important to know something
about the sources of the pollutants in stormwater and their
movement. These topics are covered in the following
subsection.
-------
SOURCES AND MOVEMENT OF POLLUTANTS
Contaminant Sources
The various contaminants found in stormwater are derived
from a number of different sources. Pollutants are adsorbed
and absorbed from the air as rain- and snowfall over a city;
from the surfaces of buildings, streets, vacant land, con-
struction sites, parking lots, and yards in urban runoff;
and in the sewer system. In the following discussion,
the pickup of the contaminants will be traced from the
initiation of precipitation until the stormwater runoff
is discharged into receiving waters.
From the Ai:r — The effects of urban air pollution on water
pollution have received only limited study and attention.
In American cities there is a wide variation in the extent
and type of air pollution. Generally, the most significant
urban air pollutants are oxides of sulfur and nitrogen, fine
particulate matter (dust), carbon monoxide, and volatile
hydrocarbons. During rainfall, sulfur oxides and nitrogen
oxides dissolve in water droplets. Particulate matter will
adhere to droplets and snowflakes. Portions of particulate
matter will be dissolved in the droplets and melted snow.
Air pollutants picked up by precipitation in the atmosphere
are usually less significant than the particulates that
have settled to city surfaces and are washed away in urban
runoff. These particulates, or dust, fall at annual rates
of 170 to 320 metric tons/sq km (500 to 900 tons/sq mi) in
most metropolitan areas [39] . That portion not removed
in the sweeping of sidewalks and streets will run off dur-
ing wet-weather and snow-melt periods.
In the absence of better information, it can be concluded
that improvement of the air quality over an urban basin
will benefit the water quality, but the magnitude of such
benefit is, at present, difficult to estimate.
On City Surfaces — As precipitation runs off urban surfaces,
ITis exposed to contamination from a wide variety of
sources. Dust and dirt, street litter, deicing chemicals,
herbicides, dead leaves, pesticides, eroded materials,
traffic residuals, and animal droppings are some of the
more important substances that are carried away by storm-
water as it passes through a metropolitan area.
In one of the earliest studies in the Storm and Combined
Sewer Pollution Control Program, undertaken by the APWA,
the sources of urban runoff pollution were analyzed
88
-------
in detail. Street refuse or litter was defined in the study
report as:
...the accumulation of materials found on the
street, sidewalk, or along the curb and gutter
which can be removed by [conventional] sweeping.
All components of street litter contribute to
water pollution, in the form of floating mate-
rial, suspended or dissolved solids, and by
bacterial contamination. The amount and compo-
sition of street litter varies widely. However,
no systematic effort previously has been made
to determine its rate of accumulation and com-
position over a period of time. The sources of
street litter vary from community to community,
from season to season, and from area to area of
the same community. Street litter is the pro-
duct of both human and natural actions. Litter
(which is defined as waste scattered about--a
clutter), includes remnants resulting from care-
less public and private waste collection opera-
tions; animal and bird fecal droppings; soil
washed or eroded from land surfaces; construc-
tion debris; road surfacing materials ravelled
by travel, impact, frost action or other causes;
air pollution dustfall; wind-blown dirt from
open areas; and a host of subsidiary materials.
[39]
High concentrations of SS in combined sewer and storm
sewer overflows have been traced, in some cases, to erosion
at construction sites. Construction of streets, buildings,
and utility services often exposes large areas of bare
earth. When storm intensity exceeds the ability of the
bare earth to absorb the precipitation, the runoff often
carries away the more erodible material. Unpaved or poorly
paved areas present a similar problem. In a 1948-1950 in-
vestigation of an area with cobblestone streets in
Leningrad, USSR, 14,541 mg/1 of SS were discovered in storm-
water [26]. In a study in Tulsa, Oklahoma, for the EPA,
a maximum average total solids concentration of 2,242 mg/1
was found for storm runoff from one test area of the
basin:
The value for this site was eight to nine times
greater than the average for the other test
areas. This extremely high concentration can be
explained by exposed open land. Shortly after
the start of the project, construction began on
a large apartment house complex. The land was
89
-------
stripped of its ground cover, cuts were made for
streets, and water and sewer line trenches were
dug. Construction continued throughout the
project. Therefore, this test area is repre-
sentative of a drainage basin that is under
development. [33]
Chemicals are applied in metropolitan areas to encourage
growth of certain plants and to discourage growth of others
They are also applied to minimize populations of rodents,
certain insects, and birds. In northern portions of the
country, deicing salts and abrasives are widely used in
both urban and rural areas to make roads safer during the
winter. Even when properly used, a large portion of these
various chemicals will be washed off urban surfaces in
stormwater.
The most common highway and sidewalk deicing compound used
is sodium chloride. In addition, significant quantities of
calcium chloride are spread in cities. Typical rates of
spreading are 100 to 300 kg of salt per km (400 to 1,200 Ib
of salt per mile) of highway per application. Since this
salt is nondegradable, it eventually runs off in water
that will flow into streams or groundwater. In an EPA
survey of the environmental impact of highway deicing
it was concluded, in part:
Road deicing salts are found in high concentra-
tions in highway runoff. Large salt loads enter
municipal sewage treatment plants and surface
streams via combined and storm sewers, and di-
rect runoff. Concentrations of chlorides as
high as 25,000 milligrams per liter (mg/1) have
been found in street drainage, and up to 2,700
mg/1 in storm sewers. Surface streams along
highways and those in urban areas have been
found to contain up to 2,730 mg/1 chlorides.
Influence of highway salts upon major rivers in
the U.S. at this time appears relatively minor...
Materials storage sites are a frequent source of
salt pollution to groundwaters and surface
streams. Deicing salts are often stockpiled in
open areas without suitable protection against
inclement weather...Careful site selection and
properly-designed materials storage facilities
would serve to minimize incidents of water sup-
ply contamination and provide for better product
handling and quality control...
90
-------
The special additives found in most road deicers
cause considerable concern because of their
severe latent toxic properties and other poten-
tial side effects. Significantly, little is
known as to their fate and disposition, and
effects upon the environment...[13]
The reduction of the contamination of urban runoff is dis-
cussed in Section VII under Source Control.
In the Sewer System — Additional contaminants may be intro-
duced to storm and combined sewer overflows through openings
to the sewer system. In many cases these contaminants
find their way into the sewer system in violation of local
ordinances. Street drains to storm or combined sewers are
convenient receptacles of various wastes that are difficult
to dispose of. Examples of these wastes are used automobile
crankcase oil, dog droppings, leaves, and yard trimmings.
Excess lawn watering carries fertilizers, herbicides, and
pesticides along curbs to stormwater openings. In separate
storm sewers these flows may either discharge untreated or
collect until a storm washes them out. Rinse water from
home car washing, which is high in grease, cleaning agents,
and fine dirt, similarly enters storm drains.
In combined sewers and heavily infiltrated sanitary sewers,
the most significant source of organic and bacterial con-
tamination is usually the municipal sewage which is in
the system by design. Turbulence of wastewater flows pro-
vides complete mixing of infiltration water, storm runoff,
and municipal sewage. Grease and settled solids in the
sewers (a problem magnified by the widespread use of garbage
grinders) , which would be treated if they reached the treat-
ment plant during dry weather, often accumulate in low-flow
reaches only to be flushed out by storm flow influxes, thus
contributing to the wet-weather discharges.
Unless catch basins are cleaned between storms, a nearly im-
possible task, they may be a significant source of contami-
nation of wet-weather flows. Although catch basins are
intended primarily to trap grit and coarse debris during
storms and/or to provide water seals to impound sewage
odors, they may also trap significant amounts of leaves and
other organic matter. This material, unless removed, de-
composes over a period of time in the stagnant pools of
water after a storm, only to be flushed out during a fol-
lowing storm. Thus, catch basins may contribute quantities
of high BODs wastes to the first flush of a storm
(Table 19).
91
-------
Table 19. FIELD TEST RESULTS OF CATCH BASIN
SOURCE POLLUTANTS AND REMOVALS
Chlorides,
mg/1
location Range Avg Range Avg Range Avg Avg Avg Avg Removals'
BOD,, mg/1 mg/1 SS, mg/1
Test V - TS, mg/1 VS, mg/1 MPN
Chicago, 111. ;>0*- removal at
[2710 0.1 in. rain
Commercial area 35-225 126 -- -- -- -- -- -' " 80?° removal at
0.2 in. rain
Residential area 50-85 67.5 -- -- -- -- " " " 94*« removal at
0.3 in. rain
Washington, D.C. 6-625 126 11-160 42 26-36,250 2,100
[26]c
Detroit, Mich. -- 234 -- -- -- -- 660 299 930,000 2-1/2 hr after
[2410 start of rain:
95°. MPN removal
7.6°. TS increase
24°6 VS increase
59°« BODj removal
San Francisco,
Calif. [38]
First sampling 5-1,500 241
series0
Second sampling 15-420 156
series^
a. Percentage of prc-storm basin contents flushed into collection system after various accumulations of rainfall
or elapsed time.
b. APWA study of 7 catch basins.
c. Sampled from 11 catch hasins.
d. Study on 1 catch basin.
c. COD, mg/1: range, 153-37,"00; avg, ",4::. Total N, mg/1: range, 0.5-33.2; avg, 12.2. Total P, mg/1: '0.2.
Sampled from 12 catch hasins.
f. COD, mg/1: range 708-143,11(10; avg, 2",'.'74. Total N, mg/1: range, O.S-ld.5; avg, 6.5. Total P, mg/1: '0.2.
Sampled from 9 catch basins.
In overloaded separate sanitary sewer systems, it is not
uncommon to find cross-connections between sanitary and
storm sewers. Because an overloaded sanitary sewer may
cause untreated municipal sewage to back up into basements
or other low portions of buildings during periods of ex-
treme peak flows, city personnel will often alleviate the
problem with a stopgap solution by constructing an overflow
from the sanitary sewer to a nearby storm sewer (or stream)
which has excess capacity. Of course, such solutions "un-
separate" the sewer system. If the treatment plant has
sufficient capacity to treat the municipal flow, then the
overflow from sanitary sewer to storm sewer increases the
amount of untreated municipal sewage discharging to the
receiving water.
92
-------
Collection and Transport
Contaminants picked up by storm runoff pass through the
sewer inlets (combined or storm sewers) and must then be
transported to treatment plants or be discharged at
overflow or bypass points. The fate of these pollutants
depends on the design and condition of the sewer system.
Infiltration may dilute combined flows; however, by causing
an increase in the amount of overflow, this excess water,
possibly uncontaminated before entering the sewer, aggra-
vates the problem. For sanitary sewers, heavy infiltration
may require overflows and/or bypasses. Infiltration in
separate storm sewers may also be significant if treatment
of these flows is eventually required.
Sources of Infiltration - During wet weather, when the
upper layers of soil tend to become saturated with water,
infiltration into conduits adds to the amount of wastewater
that must be disposed of and reduces effective conduit
capacities. Infiltration enters sewers through cracks and
holes, joints, house connections, and manholes.
Cracks and holes are most likely to be found in older
sewers, but sometimes faulty materials and construction
methods result in defects in newly installed sewers.
Uncoated steel and iron sewers may be attacked from the in-
side by wastewater acids and salts and from the outside by
certain soils. Concrete pipe is often attacked by hydrogen
sulfide gas which is produced by the anaerobic decomposi-
tion of the deposited solids found in wastewater. Differ-
ential settling or poor preparation of bedding often sub-
jects sewers to loads greater than design loads. This can
cause sewer cracking, and under extreme conditions, collapse
Small cracks and holes are often penetrated by tree roots
that enlarge the opening and also obstruct flow in the sewer
A small crack may allow surrounding silt and sand to be car-
ried into the sewer along with the infiltration. This
removed silt can leave a cavity that may eventually under-
mine the sewer foundation and lead to a situation such as a
street collapse as shown on Figure 14.
Some of the oldest sewers in the United States were con-
structed of brick or stone. These were rarely watertight
even when they were new. Groundwater pressure might first
force out some mortar and later a brick or two. Although
these sewers are no longer being built, many continue to be
in service because they are located in the most heavily
built-up areas of cities where it is extremely expensive to
replace them.
93
-------
Figure 14. Street col lapse
due to inf iItration [29]
Defective sewer joints are usually a more significant cause
of infiltration than cracks or holes. In particular, older
sewer joints (pre-1950s) were rigid, and effective gasket
materials were not used. By not allowing for deflection,
joints would often separate or break as the sewer settled.
A slight break or deterioration in a gasket seal is an
invitation for roots to enter. In many sewers, joints
were not made properly during construction. Today's in-
creasing need for treatment of wet-weather flows requires
more care in the joining of sewers. This subject is dis-
cussed in greater detail in Section VIII.
Even if a local government ensures that joints are well
constructed on sewers for which it is directly responsible,
there is still the possibility of leaks on house connections
In some cities defective house connections have been deter-
mined to be the largest single source of infiltration.
Many cities now require that the joint between the house
94
-------
connection and the city sewer be made by, or under the super
vision of, city personnel.
A survey in New Orleans, Louisiana, indicated that manholes
contribute greatly to sewer infiltration [29]. Some of
the major causes of infiltration observed included:
(1) broken sewer pipe because of settlement of either the
manhole or the pipe; (2) improper construction or deteriora-
tion of cement mortar linings or brick manholes; (3) cracks
in the foundation, sidewalls, and castings; (4) improper
seals at sewer connections; (5) improper construction meth-
ods when manholes are raised or lowered; (6) dislodging of
castings from top of manhole by heavy equipment used for
land clearing, filling, or leveling ground.
In areas with a high groundwater table, the greatest cost
of infiltration may be in increased construction and opera-
tion expense for treatment plants. These cities often have
high levels of infiltration even during dry weather.
Combined Sewer Systems — Although new combined sewer systems
are infrequently designed in the United States, some cities
presently served by combined sewers continue to add new
combined sewers to existing systems. The practice of instal-
ling separate sewer systems has been institutionalized.
For example, to obtain federal funding on urban redevelop-
ment projects, combined sewers must be separated in the
project area. In a survey of municipal sewerage facilities
conducted by the APWA [27], it was found that 15,309 km
(9,515 miles) of new separate sanitary sewers were con-
structed from 1957 to 1967. During the same period, only
1,926 km (1,197 miles) of new combined sewers were built.
The surveyed communities contained 94 percent of the popu-
lation served by combined sewers, but only 17 percent of
the population served by separate sanitary sewers.
Obviously then, the national ratio of new sanitary sewer
miles to combined sewer miles must be considerably greater
than 8 to 1.
One of the most important components of a combined sewer
system is the interceptor sewer (or simply, the interceptor).
Large combined sewer systems may have a network of
interceptors. In the APWA survey, considerable variation
was found in the capacity of interceptors. In terms of the
ratio of peak flow to mean dry-weather flow, the capacity
ranged from 1:1 to 8:1, with 4:1 median.
The greater the capacity of the interceptor, the smaller
the potential amount of wastewater that will overflow at
the trunk sewer-interceptor overflow point. The poten-
tial is emphasized because most combined sewer regulators
95
-------
will begin to allow overflows before the flow reaches the
capacity of the interceptor. In particular, regulators
such as leaping weirs, side weirs, and other static devices,
will rarely utilize the complete available interceptor
capacity. Combined sewer systems with computer-controlled
regulators to optimize interceptor utilization are described
later in this text.
In many cities, excess wastewater treatment capacity is
limited. Therefore, even if more of the combined flow is
intercepted, it must be bypassed when it arrives at the
treatment plant. Typically, in the design of treatment
plants, capacity for treating 1.5 to 3.0 times the mean
dry-weather flow is provided. This additional capacity
is usually provided, not for wet-weather flows, but rather
for the normal peaking of dry-weather flow during a 24-hour
period.
Separate Sewer Systems - Most American cities built pri-
marily during the twentieth century have separate sewer
systems for sanitary wastewaters and storm runoff. Hence,
cities like Los Angeles, Dallas, and Denver have separate
systems, as compared with San Francisco, Chicago, and New
York, which have combined systems. Other cities, like
Washington, D. C., Atlanta, and Minneapolis, have hybrid
systems with large areas sewered by combined sewers and
more or less equal areas served by separate sewers. Some
cities started out with combined sewers but later separated
the sewer systems. In some cases, the existing sewers were
used to convey municipal sewage, and new storm sewers were
constructed. In other cases, the existing sewers became the
storm sewers, and new sanitary sewers were built. Suburban
areas with good natural drainage often have separate sani-
tary sewers and no storm sewers. For example, of the people
served by separate sanitary sewers in the communities sur-
veyed by APWA in 1967, only 60 percent were also served by
separate storm sewers.
Of particular concern, in studying the possible collection
and treatment of separate storm sewer discharges, is the
fact that storm sewers are normally designed to take the
most direct and economical route to the nearest waterway.
Thus, as one travels outward from the urban activity core,
the storm sewers become more and more fragmented and dis-
persed, and open channel flow in natural or "improved"
creeks becomes more commonplace. This makes interception
increasingly difficult and costly to a prohibitive degree.
Environmental concern over the preservation of remaining
natural creeks and streams largely precludes massive diver-
sion of these flows for storage and treatment systems inte-
grated with the municipal sewage flows.
96
-------
ENVIRONMENTAL EFFECTS
In previous parts of this section, the quantities, quali-
ties, and loadings of wet-weather overflows and bypasses
have been discussed. Of real concern in any water quality
study are the effects of any discharges, procedures, or
physical operations on water bodies. In the case of storm
sewer discharges and combined sewer overflows, the focus of
concern is on their effects on receiving waters. The prob-
lem is to isolate the effects of wet-weather discharges and
to compare these effects with the effects of other pollution
sources. The question that needs to be answered is: What
improvements (if any) in the receiving water will result
from the reduction of wet-weather pollution?
Except for a few cases, real water quality benefits can be
obtained through improved management of stormwater. The
nature and magnitude of receiving water degradation associ-
ated with wet-weather discharges are brought out in the
following two examples. Bucyrus, Ohio, has a combined
sewer system, and Roanoke, Virginia, has a predominantly
separate sewer system. Identification of the variations
between wet-weather and dry-weather conditions forms the
basis of evaluating the potential effectiveness of wet-
weather controls and treatment.
Conversely, a very heavily polluted stream may actually be
improved by the discharge of large amounts of relatively
dilute storm or combined wastewater, if only temporarily.
However, this is the exception; and in the future, as there
are fewer chronically polluted rivers, it will become an
increasingly rare exception.
Bucyrus, Ohio
Bucyrus, Ohio, is an incorporated city of 948 ha (2,340
acres) with a population of 13,000. The combined sewer
system carries an average dry-weather flow of 0.1 cu m/sec
(2.2 mgd). Secondary effluent and combined sewer overflows
are discharged to the upper reaches of the Sandusky River,
a tributary of Lake Erie. One of the aspects of Bucyrus,
which made it particularly useful as a study area, is that
there are no significant urban areas upstream. Hence, it
is possible to study the degradation of Sandusky water qual-
ity as the river passes from above Bucyrus to several miles
downstream, and to do so under varying weather and flow
conditions. Furthermore, since rain usually falls on
Bucyrus before it falls on the upstream basin, combined
sewer overflows occur before river flow increases
significantly. Under these conditions, scouring of benthal
97
-------
deposits is not as likely to occur and possibly interfere
with the determination of the effect of the overflows on the
Sandusky River. In the 1968-1969 EPA-sponsored study of
stormwater problems in Bucyrus, analyses were made of
Sandusky River samples. The results are summarized in
Table 20. From these samples, the investigators found
the following:
Upstream of the Urban Area
The major differences in the upstream water qual-
ity characteristics during dry and wet weather
are in the suspended solids, nitrates and bac-
teria counts. The average dry weather suspended
solids of 32 mg/1 increase to an average 465 mg/1
during wet weather. The average dry weather con-
centration of nitrates is 7.2 mg/1 as NC>3 and is
increased to an average 21.7 mg/1 during wet
weather. This increase in nitrates seems to be
due to agriculture runoff. The total coliform
count is reduced from a dry weather average of
59,000 per 100 ml to 3,400 per 100 ml during wet
weather. This reduction in bacteria is due to
the added dilution water from the upper drainage
area.
Through and Downstream of the Urban Area
The comparison between the dry and wet weather
river samples indicates that the waste loads
from the overflow affect the river quality as
far downstream as the fifth bridge, which is ap-
proximately seven miles downstream from the
wastewater treatment plant. During periods of
overflows the average BOD concentration at the
first bridge downstream from the wastewater"
treatment plant is increased from a dry weather
average of 6 mg/1 to 14 mg/l> the suspended
solids increase from 49 mg/1 to 192 mg/1, and
tTie total coliforms increase from a dry weather
average of 400,000 per 100 milliliters to 4.5
million per 100 milliliters.The average coli-
form count at the fifth bridge downstream from
the wastewater treatment plant is increased from
an average 4,500 per 100 milliliters to 86,000
per 100 milliliters. [35] (Emphasis added)
98
-------
Table 20. SUMMARY OF DRY- AND WET-WEATHER
SANDUSKY RIVER ANALYSES
BUCYRUS, OHIO
1.
2.
3.
4.
5.
BODs,
mg/1
Dry Wet
Location weather weather
Sandusky River - upstream
Number of analyses 33 22
Average 4 5
Range 1-14 2-13
Sandusky River - downstream
1st bridge downstream from
wastewater treatment plant
Number of analyses 27 43
Average 6 ]4
Range 2-12 4-51
Sandusky River - downstream
2nd bridge from
wastewater treatment plant
Number of analyses 9 g
Average 7 5
Range 3-22 3-8
Sandusky River - downstream
3rd bridge from
Number of analyses 12 17
Average 4 f,
Range 1-8 3-10
Sandusky River - downstream
5th bridge from
wastewater treatment plant
Number of analyses 13 19
Average 5 g
Range 2-13 2-12
*>S • Total coliforms
mg/1 /100 ml
Dry Wet Dry Wet
weather weather weather weather
20 13 3 4
32 465 59,000 3,400
5-160 20-1,960 23,000-95,000 1,200-6,300
14 38 8 11
49 192 0.4 x 106 4.5 x 106
8-190 5-960 2xl03-1.5xl06 0 . 05x1 O6- 8 . 8xl06
8 8
44 62
10-195 20-135
4 17 5 1
36 36 15,000 130,000
27-45 20-50 5,600-40,000
5 11 4 1
18 90 4,500 86,000
15-25 25-300 3,000-5,300
99
-------
In the case of Bucyrus, the detrimental effects of combined
sewer overflows are significant enough that it is not neces-
sary to even make a water quality analysis to realize that
a problem exists:
The overflow wastes have several effects on the
river. The most obvious one is the debris and
organic solids which settle in the river in and
below the city. These create odors and un-
sightly conditions long after the overflow is
past.
The results of the aquatic biology survey cor-
roborates the results of the water quality
studies of this report. The river upstream from
Bucyrus has a relatively undisturbed fauna of
the types normally found in unpolluted waters.
The river inside the City of Bucyrus shows indi-
cation of gross pollution and has sections com-
pletely devoid of life. The river downstream
from Bucyrus, during periods of low flow, is
biotically dead for six to eight miles below the
wastewater treatment plant. [35]
Roanoke, Virginia
The city of Roanoke, Virginia, covers 6,786 ha (16,756 acres)
of valley land between the Blue Ridge and Allegheny
mountains. The city population of 93,000 and the surround-
ing suburban population are served by separate sanitary
and storm sewer systems, except for a very small combined
service area downtown. The municipal sewage is treated
by a 1-cu m/sec (22-mgd) activated sludge plant. The Roanoke
River and several tributaries receive storm sewer discharges
and infiltrated municipal sewage overflows, in addition to
the treatment plant effluent.
In a 1965 study of the sanitary sewer system, it was con-
cluded that overflows "...were resulting in unsightly and
undesirable pollution of the watercourses in the City"
[12], Since a major regional recreation area, Smith
Mountain Lake, is located 16 km (10 miles) downstream, degra-
dation of the Roanoke River was of special concern. The
impact of wet-weather flows on the Roanoke River was quanti-
fied in a 1968-1969 investigation sponsored by the EPA.
The results of a stream sampling program which determined
the concentrations of various pollutants during both wet
and dry weather are reported in Table 21. Murray Run,
Trout Run, and 24th Street are three Roanoke tributaries
with average dry-weather flows of 176, 42, and 31 I/sec
(6.2, 1.5, and 1.1 cfs), respectively. As shown, all three
100
-------
Table 21. RELATIVE CONCENTRATIONS OF POLLUTANTS DURING
AVERAGE DRY- AND AVERAGE WET-WEATHER CONDITIONS
ROANOKE, VIRGINIA
BOD5, mg/1
TS, mg/1
TVS, mg/l
SS, mg/1
Stream
Dr/ Wet Dry Wet Dry Wet Dry Wet
weather weather weather weather weather weather weather weather
Murray
Run 8
Trout
Run 3
24th
Street 8
17
18
20
248
281
194
623
460
514
85
147
126
134
139
172
37
17
20
89
93
103
VSS, mg/1
Stream
Dry Wet
weather weather
Settleable
solids, ml/1
Dry Wet
weather weather
Flow, mgd
Dry Wet
weather weather
Murray
Run 12 25
Trout
Run 8 28
24th
Street 7 24
0 2 4.0 7.7
0 3 l.o 13.8
0 3 0.7 3.4
streams suffer a considerable degradation in water quality
during wet weather. SS and BOD5 increase by several times
in each stream.
It is not possible to describe in two examples all of the
adverse environmental effects of storm sewer discharges and
combined sewer overflows. However, by considering the sig-
nificance of the stormwater problem in two smaller cities,
it is not difficult to imagine the impact of wet-weather
discharges from large cities on receiving waters. In fact,
following a detailed examination of ten combined sewer
overflow systems surrounding Upper New York Bay and the
lower reaches of the Hudson and East rivers, it was
concluded:
...In view of the tremendous quantities of pol-
lutants bypassed during the rainfall from this
combined sewer system, it does not seem reason-
able to debate whether secondary treatment
plants should be designed for 80, 85, or 90% BOD
101
-------
or suspended solids removal when in fact the
small increments gained in this range are com-
pletely overshadowed by the bypassing occurring
at regulators during wet weather flow.
...The necessary improvement in the quality of
receiving waters and the reopening of beaches
will not be accomplished by the multi-billion
dollar treatment plant upgrading and expansion
program now going on within the District, and
the monies spent for this construction in large
part will be wasted if means of mitigating the
effects of combined sewers are not found. [8]
Beneficial Aspects of Wet-Weather Discharges
Discharges of wet-weather flows--particularly separate storm
sewer and/or treated discharges--to extremely foul streams
may have some beneficial effects on the receiving waters.
As noted earlier in presenting wastewater characteristics,
the DO of storm flows is generally high, and the organic
content--in all but the initial runoff stages — is relatively
low. Just as runoff tends to clean the air and land areas,
the temporary high flows may tend to flush pollutants from
creeks and normally stagnant backwater areas into river
zones of greater assimilative and recovery capacity. Storm
flows also provide a viable and major source of groundwater
replenishment not only by direct percolation in pervious
areas, but also by continuous releases through natural
stream beds and ponds. Finally, there is the plant and
wildlife subsisting in the flood plain environment which
must be considered in evaluating management alternatives.
As introductory information to alternative control and
treatment methods, a discussion of the development of
process and system evaluation criteria is presented in the
next section.
102
-------
Section VI
EVALUATION PROCEDURES AND CRITERIA
The ability to collect and/or evaluate data effectively and
to assess process and equipment applicability to stormwater
problems is essential to program development. It is in-
tended that this section provide a transition from an
awareness of the problem described in Section V to the
critical evaluation of basic alternatives and technology
presented in Part III. A brief introduction to sampling
parameters to consider, measurement, and data analysis is
presented, followed by a discussion of considerations perti-
nent to process and equipment evaluation. The link between
data and equipment operation and control systems is also
introduced. Finally, the handling of cost data estimates
for facilities is discussed.
DATA COLLECTION
Data collection is primarily concerned with the determina-
tion of the volume/flow rate and characteristics of the
wastewater and receiving waters before, during, and after
storm events. Locations of interest include points of
overflow (potential or real) or diversion, in-system sites
upstream and downstream of treatment facilities, and moni-
toring stations within the receiving waters. Logically the
data are intended to identify the nature of the problem'and
the effectiveness of the solution(s). Data analysis in-
volves the interpretation of the results and is discussed
later in this section. The goal of the analysis is to
weigh the credibility and significance of the data and
to provide feedback for guidance in improving the collection
program. To be of general value, analytical work must fol-
low standardized procedures (NPDES Guides recommended) and
the means of sample collection, storage, and examination
must be identified clearly.
103
-------
Sampling
A comprehensive state-of-the-art study for EPA on wastewater
flow sampling [12] has recently been completed. The study
has provided much of the basis for this summary, and the
reader is referred to it for more detailed information.
Two comments from the study are believed particularly en-
lightening in placing stormwater applications in proper
perspective:
Because of the variability in the character of
storm and/or combined sewage, and because of the
many physical difficulties in collecting samples
to characterize the sewage, precise characteri-
zation is not practicable, nor is it possible.
In recognition of this fact, one must guard
against embarking on an excessively detailed
sampling program, thus increasing costs, both
for sampling and for analyzing the samples,
beyond costs that can be considered sufficient
for conducting a program which is adequate for
the intended purpose.
Sampling programs should start long before in-
stallation of combined sewer overflow and storm-
water treatment/control facilities to establish
the objectives of the facilities and to provide
necessary design and operation criteria. A much
longer time period for sampling may be required
than anticipated because of the need to sample
during periods of storm runoff, which may be few
in drought years. [12]
Thus the scope and timing of sampling programs must be con-
sidered concurrently with the selection of site locations
and sampling devices.
Programs - Sampling programs are established for a variety
of purposes--primarily, problem identification, process and
equipment evaluations, and waste stream and receiving water
monitoring. The more common pollutants analyzed are BODs,
COD, chloride, nutrients (nitrogen and phosphorus deriva-
tives), pH, total solids, suspended solids, volatile solids,
oil and grease, and coliform groups. If industrial wastes
are included and/or occasionally for urban runoff, addi-
tional analyses for such pollutants as cyanide, fluoride,
heavy metals, pesticides, sulfate, and sulfide may be needed
104
-------
Requirements for intensive continuous monitoring by various
enforcing agencies appear to be imminent; thus, sampling
is becoming an increasingly larger part of pollution control.
Most important to any sampling program are (1) determina-
tion and definition of purpose, (2) program timing, (3) iden-
tification of specific data requirements, and (4) evaluation
of the costs involved with respect to the quantity and qual-
ity of data required. A careful study of costs, distinguish-
ing between sample collection, subsequent analyses, and re-
porting, should be made prior to commencing a program.
For example, where collection costs are high (because of
location or necessary coincidence with a particular event)
or where objectives are selectively stringent, additional
analyses may be justified.
If the number of samples is large and the program is to con-
tinue over a long period, consideration should be given to
the use of automatic analyzing equipment. Caution is needed,
however, in its selection. With some equipment, the time
required for making necessary adjustments between each of a
series of tests may counteract the rapidity of making single
parameter analyses.
The use of mathematical statistical analysis for determining
the probable errors in the data obtained by sewer sampling
is usually not practicable. For example, a single grab
sample of 1 liter, even in dry-weather flows, may not be
representative of the average character of the flow. The
grab sample is representative of only an instant in time
and, if the sewage is not thoroughly mixed in the pipe, of
one point in the cross-section of the flow. During storm
flows, sewage characteristics change rapidly, making a grab
sample even less representative. Thus, a large number of
samples taken over short time intervals is required to char-
acterize the sewage in a combined sewer adequately.
Compositing the samples in proportion to flow rate may yield
the average character of the sewage during that period of
compositing, but it does not describe the pattern of changes
occurring during that period.
As each program progresses, it may be possible to reduce the
number of samples collected or analyses performed on the
basis of a periodic review of the data obtained.
Site Selection - The adequacy of a sampling program depends
largely on the optimum selection of sampling sites. To
105
-------
ensure that the samples are representative, the following
general guides are recommended:
In sewers and in deep, narrow channels, samples
should be taken from a point one-third the
water depth from the bottom. The collection
point in wide channels should be rotated across
the channel. The velocity of flow at the sam-
ple point should, at all times, be sufficient
to prevent deposition of solids. When collect-
ing samples, care should be taken to avoid cre-
ating excessive turbulence which may liberate
dissolved gases and yield an unrepresentative
sample. [19]
Additional considerations are as follows: (1) the site
should provide maximum accessibility and safety; (2) it
should be a sufficient distance downstream from the nearest
tributary inlet to ensure complete mixing of the two flows;
and (3) there should be a straight length of pipe at least
6 sewer diameters upstream of the site. If the cross-
section of the sewage flow is homogeneous with respect to
the constituents being sampled, then a single point of
sample extraction will be adequate. If there is a spatial
variation in the concentration of the particular constitu-
ent, as is more often the case, then the sampler intake
must be designed to gather a sample that is--as nearly as
possible — representative of the actual flow.
Awareness of the general character of sewer flows and flow
modes in storm and combined sewers and knowledge of the
variability of pollutant concentration leads to an under-
standing of how best to select sampling sites.
Equipment — Samplers presently available operate on mechan-
ical (e.g. , dipper), suction lift (pump above flow level),
forced flow (submersed pump), or fluidic (differential pres-
sure actuated) principles. Typical sampling units and in-
stallations are shown on Figure 15.
In addition to gathering a representative sample, the sam-
pling equipment must also be capable of transporting the
sample, without precontamination or cross-contamination from
earlier samples or aliquots, and suitably storing the gath-
ered sample. Chemical preservation is required for certain
parameters that may be subject to later analyses, but re-
frigeration of the sample is also required and is considered
to be among the best means of preservation. Special precau-
tions and analysis techniques may be necessary, however, to
prevent data distortion of certain parameters obtained from
refrigerated samples [1].
106
-------
Figure 15. Typical sampling units and installations
(a) River sampler supporting a 1.3 I/sec (20 gpm) submersible pump (b) Sample
receiver, direct analyzer, and signal transmitter (c) Discrete sewage sampler,
time-and level-paced, with refrigerator (d) Sampling station at stormwater
pumping station forebay (e) Associated discrete sampler, solenoid operated
107
-------
The design should be such that maintenance and troubleshoot-
ing of the unit are relatively simple tasks. The unit
should have maximum inherent reliability. As a general rule,
complexity in design should be avoided, even at the sacri-
fice of a certain degree of flexibility of operation.
Factors that should be considered in the selection of a
sampler include the following:
• Range of wastewater conditions.
• Rate and frequency of change of wastewater
conditions.
• Periodicity or randomness of change of wastewater
conditions.
• Availability of recorded flow data.
• Need for determining instantaneous conditions,
average conditions, or both.
• Volume of sample required.
• Need for sample preservation.
• Estimated size of suspended matter.
• Need for automatic controls for starting and
stopping.
• Need for mobility or for a permanent installation.
Although it would be impossible for a single piece of equip-
ment to be ideal for all sampling programs in all storm and
combined sewer flows, samplers presently available offer a
variety of features and some general requirements can be
delineated. A sampler should be capable of (1) collecting
a representative sample of sufficient size for the necessary
analyses; (2) operating in the mode required (either auto-
matic or manual); (3) collecting either discrete or com-
posite samples; (4) unattended operation while remaining
in a standby condition for extended periods of time; and
(5) operating under a variety of hostile environmental
conditions and sewer flow ranges.
Assessments of the state-of-the-art study [12] on sampler
equipment included the following:
Intake design — For representative sampling, the preferred
orientation of the intake is into the sewage flow (facing
108
-------
upstream). Further, the intake velocity should equal or
exceed the velocity of the stream and the geometry of the
intake has little effect. "In the absence of some consider-
ation arising from the particular installation site, a regu-
lar distribution of sampling intakes across the flow, each
operating at the same velocity, would appear to suffice.
Since the intakes should be as non-invasive as possible in
order to minimize the obstruction to the flow and hence the
possibility of sewer line blockage, it seems desirable to
locate them around the periphery of the conduit."
Collection method - "...the suction lift gathering method
appears to offer more advantages and flexibility than either
of the others. The limitation on sample lift can be over-
come by designing the pumping portion of the unit so that it
can be separated from the rest of the sampler and thus posi-
tioned not more than 30 feet [9.2 meters] above the flow to
be sampled. For the majority of sites, however, even this
will not be necessary." The first flow of any suction lift
sampler should be considered nonrepresentative and returned
to waste.
Sample transport — The sampler conduit size must be large
enough to be free of plugging or clogging but small enough
to ensure velocities high enough to prevent settling of the
SS. Thus, the sample flow rate and line size are interde-
pendent and must be approached together from a design
consideration. The minimum line size should be 10- to 12-mm
(3/8- to 1/2-inch) inside diameter and the minimum veloci-
ties should not drop below 0.6 to 0.9 m/sec (2 to 3 fps).
Sample container - The sample capacity will depend upon the
subsequent analyses to which the sample is to be subjected
and the volumetric requirements for these analyses. As a
minimum, it is recommended that at least 1 pint and prefer-
ably 1 liter of fluid be collected for any discrete sample.
For composite samples at least 1 gallon, and preferably 2,
should be collected. The container itself should be either
easy to clean or disposable.
Controls and power - It is desirable that the equipment
start automatically upon signal from an external device in-
dicating the onset of a storm.
One of the most attractive techniques for automatically
starting stormwater samplers is through flow depth sensing.
A remote power source with automatic changeover to battery
operation on power outage is recommended. The controls for
an automatic sampler should allow some degree of freedom in
the operation and utilization of the equipment. A built-in
109
-------
timer is desirable to allow preprogrammed operation of the
equipment. Such operation is particularly useful, for exam-
ple, in characterizing the pollutant concentration increase
in the early stages of storm runoff. However, the equipment
should also be capable of flow proportional operation.
Representative sampler types, features, and costs are listed
in Table 22. Manufacturers' quoted 1972 equipment costs
ranged from $275 to $5,606 [12], depending upon the degree
of sophistication of the equipment with respect to (1) the
number of samples collected, (2) mode of operation, (3) type
of sample collected, (4) weatherproofing, (5) refrigeration,
etc.
Case Histories - Several municipalities have installed large
scale systems utilizing automatic samplers for identifying
the performance of stormwater overflow control or abatement
systems. Two are described below as representative examples,
METRO (the Municipality of Metropolitan Seattle) has con-
ducted a comprehensive water quality sampling program
throughout its entire metropolitan drainage area since 1963
[16]. At the inception of its computerized control demon-
stration grant from the EPA in 1967, additional specialized
water quality monitoring studies were added to the existing
program to concentrate on certain areas within the collec-
tion system that contributed to combined sewer overflows.
Six compositing and seven 24-bottle discrete samplers (each
sequentially programmed, refrigerated, automatic [Sirco
Controls Co., Model B/ST-VS]) were designed and built as
part of the demonstration grant. They are installed at
widely separated overflow stations in the project, and
each operates whenever the adjacent outfall gate is in
the open position. The suction lift samplers draw overflow
samples up as much as 5.2 meters (17 feet) to a bottle of
at least 1-liter volume. Samples are then refrigerated
until they are collected for analysis. Within each sampler,
a section containing programmers, timers, and other control
devices rests above the refrigerated section. The control
enclosure is heated and contains air circulation fans to
reduce interior corrosion from condensation.
According to METRO, the term "automatic" is somewhat de-
ceiving since considerable manual effort is involved in
collecting samples, replacing bottles, and testing and re-
pairing the various electrical components. Following an
initial 6-month break-in period, the performance record of
the units has been satisfactory.
In summary, METRO found that most sampler malfunctions were
simple to correct; rarely has the manufacturer been called
110
-------
Table 22. REPRESENTATIVE COMMERCIALLY
AVAILABLE SAMPLERS21
Operate Available options
Max- under
Model imum frcez- Auto- Hxplo- Win-
designa- sample Sub- ing matic Refri- sion ter-
tion lift, mer- condi- Port- start- ger- proof- izing Price
Manufacturer or No. Type Power ft siblc tions able up Timer ation ing kit $
N-Con
Inc.
N - Con
Inc.
N-Con
Inc.
Systems Co . ,
Systems Co . ,
Systems Co. ,
Surveyor
Scout
Sentry
Com./1'
Scq .
Com. /
Seq.
Com. /
Seq.
Halt ./Hxt 5
Bat t.
Bat t. /Hxt .
6
15
15
Kd
N
N
N
N
N
Y
N
N
N Y N
N Y N
N Y N
N
N
N
275
450
895
N-Con Systems Co., Trebler Com./ Hxt. •-0 N X N N Y Y N 995-1 560
Inc. Seq. '
Protech, Inc. CC-125 Com./ Refrigerant 32 N Y Y N Y N Y Y 583-795
Seq. or compressed
gas
Protech, Inc. CG-125S Com./ Batt./Hxt. 32 N Y N N Y Y Y Y 1,650-2,310
Scq .
a. Data from [12]. Listing does not constitute endorsement or recommendation for use
b. Com. = composite; Seq. = sequential.
c. Batt. = battery; Hxt. = external.
d. N - No; Y = Yes.
e. N/A = Not applicahlc.
Protech, Inc. CG-125FP Com./ Refrigerant 32 N Y Y N Y N Y Y 693-1,108
Scq. or compressed
gas or Batt./
Hxt .
Protech, Inc. CG-150 Com./ Rt-f r igerant 32 N Y Y Y Y Y Y Y 895-2 510
Scq. or compressed '
gas or Batt./
Hxt.
Protcch, Inc.
Protech, Inc.
Sigmamotor, Inc.
Sigmamotor, Inc.
Sigmamotor, Inc.
Sirco Controls
Co.
Sirco Controls
Co.
Sirco Controls
Co.
Sirco Controls
Co.
Sonford Products
Corp.
Sonford Products
Corp.
Sonford Products
Corp.
CHL-300
DHL-240S
WA-1
WUPP-2
WM-1-24
B/ST-VS
B/IH-VS
B/DP-VS
PI I -A
IIG-4
NW-3
TC-2
Com. /
Seq.
Com. /
Scq.
Com. /
Seq.
Com. /
Seq .
Com. /
Seq.
Com. /
Scq .
Com. /
Seq .
Com. /
Scq.
Com. /
Seq .
Com. /
Seq.
Com. /
Seq.
Com. /
Seq.
Hxt.
lixt .
Batt. /Hxt.
Batt./Hxt .
Batt. /Hxt .
Batt./Hxt.
Batt./Hxt.
Batt ./Hxt .
Batt.
Batt. /Hxt.
Spring
driver
clock
Ext.
32
32
22
22
22
22
200
N/AC
18
~ 0
8
N/A
N
N
N
N
N
N
N
N
N
N
N
N
Y
Y
N
N
N
Y
Y
Y
N
N
N
N
Y
\
Y
Y
Y
N
N
N
Y
Y
Y
N
N
Y
N
N
N
N
N
N
N
N
Y
N
Y
Y
Y
N
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
N
Y
Y
Y
Y
N
N
N
Y
Y Y
Y Y
N
N
N
Y
Y
Y
N
N
N
N
1 ,450-2 ,910
5,606
400-700
680-770
1 ,050-1 ,.525
1 ,760-2 ,950
1 ,380-2,850
1 ,550-2,640
1 ,387-1 ,832
325-495
920
2,495
111
-------
in to make major repairs. As a rule, a properly maintained
sampler, no matter how simple or complex, will work well
and will sample each overflow according to design
specifications.
DMWD (the Detroit Metropolitan Water Department) has also
used automatic samplers. Their experience was similar to
METRO'S, but DMWD was disappointed with the results of
their program [25] . DMWD used an automatic sampler to
collect samples every 1/2 hour. The sampling pump, located
inside the manhole, was limited to 5.5 meters (18 feet) of
suction lift. Debris in the combined sewer wrapped around
the sampling head and caused blockages. The plugging prob-
lem was considerably improved but not eliminated after
several months.
Another problem arose during a study of the variation of
pollutional load in the combined sewer during storm events.
Although variations in SS concentrations are to be expected,
the wide variations (ranging from 0 to 1,000 mg/1) occur-
ring at random time periods indicated that the samples were
not representative. The low pumping rates necessitated by
a 24-hour sampling period resulted in accumulation of solids
in the sample line. These were then released into the sam-
ple bottle as a "slug." Daily flushing and cleaning of
sample lines did not solve this problem.
DMWD suggests several modifications for improving the auto-
matic sampler's performance: (1) the sample line should
have a minimum inside diameter of 38 mm (1-1/2 inches);
(2) the pumps should have a minimum capacity of 1.6 I/sec
(25 gpm); (3) a primary grinder should be installed on
the sampler line; and (4) a flow-through system with a
take-off for the sampling device should be used. For sam-
pling combined sewer overflows, the system should be
equipped with the necessary sensors to detect overflows
and to start and stop the pumps. On-site power for the
pump and comminutor is required. Although this system would
have limitations as to location, it is believed that a large
flow must be maintained if truly representative samples are
to be obtained. With a flow-through system, the sampler
could be located away from the manhole and protected from
damage.
Flow Measurement
Accurate flow measurement is a basic requirement in waste-
water system design and operation. Methods used are classi-
fied as either direct discharge or velocity-area types
[19]. Direct discharge methods (typically a control section
with a differential head measurement) used in conventional,
112
-------
dry-weather facilities are generally simple, proven, and
well documented. The devices used include weirs, flumes,
venturi and magnetic flow meters, and flow tubes, all with
their associated sensing and recording appurtenances. Most
devices induce headlosses to the flow stream, generally in
proportion to the accuracy required, and in the case of the
cited meters and tubes, require full section flow (no free
water surface and minimum entrained gases).
Because of the volumes and consequent costs involved and
the extreme flow variability encountered, stormwater flows
are generally measured by the velocity-area method (typi-
cally a measured depth coupled with a "normal" flow velocity
assumption or measurement). Thus, instruments/operators
will couple depth probes (using floats and scows, bubbler
tubes, sonic or electrical devices, etc.) and known cross-
sectional geometry with velocity measurements (using tracers,
current meters, etc.) or approximating computations (using
conduit slopes and Manning's formula, gate losses, etc.).
Note that simple depth measurements are highly susceptible
to errors due to nonuniform and unsteady flow conditions.
Typical requirements for an effective stormwater conduit
gage, taken from a recent prototype design contract [21],
are as follows:
Necessary
1. It must be capable of functioning under all condi-
tions of flow, from partially full, open channel
type flow to full flow under varying surcharge
pressures.
2. Interference with pipeline hydraulics must be kept
at a minimum.
3. It must neither influence nor be influenced by any
contaminants in the liquid.
4. It must operate with satisfactory accuracy under
all flow conditions.
5. The number of moving parts must be kept at a mini-
mum for easy maintenance.
6. It must be capable of being instrumented for remote
readout.
7. It must be resistant to theft and vandalism.
113
-------
Desirable
1. It should be applicable to a wide range of conduit
sizes.
2. It should be rapidly installed through a standard
manhole or in other locations where the sewer is
accessible.
3. It should be readily installed in existing pipes.
4. Its power requirements should be kept at a minimum.
5. Its cost should be reasonable.
Devices closely approaching these ideals have yet to be
developed. Objects in the flow, grease buildup, corrosion,
and flow variability are the major difficulties encountered.
Once a device is installed, maintenance and recalibration
must be carefully attended to.
Typical Devices — The following is a brief summary of repre-
sentative wastewater measuring devices taken from Schontzler
[22] and Metcalf $ Eddy, Inc. [19].
Floats and scows — Floats are used to measure the change of
wastewater level in a small stilling well at the side of a
channel, as opposed to scows which are anchored in the
middle of the stream. The major problem encountered with
the float is that the stilling well tends to become clogged
and silted, requiring constant cleaning. On the other hand,
a scow, riding in midstream, becomes a depository for all
floating debris, rags, paper, etc., carried by the waste-
water flow. This accumulation tends to submerge the scow
and causes erroneous readings. Constant care and cleaning
are required. A representative cost of a simple float and
drum chart recorder would be under $500.
Gas bubblers — Gas bubbler devices, adapted from measuring
heights in tanks of liquid, involve immersion of a dip tube
[e.g., 6 mm (1/4-inch) diameter pipe] to a point near the
base of the flowing liquid. Air, or other gas, is injected
through the tube, and the resultant back pressure, which is
measured, is a direct function of liquid level, provided
that the liquid density is constant. One problem is the
tendency to collect crystallized solids inside the lower end
of the tube, reducing the diameter and resulting in a false
reading. A constant water purge may alleviate the problem.
Other problems include hangups of floating debris and
changes due to the aspirating effect of the flowing water.
114
-------
The latter induces errors as the flow rate varies from the
calibration rate. The system requires an accurately regu-
lated gas source, either a pressurized gas container or
compressor, to develop the reference back pressure. A
single installation, running on bottled nitrogen gas with
recorder and regulator, might cost on the order of $750,
exclusive of a control device (e.g., Palmer-Bowlus flume)
[7].
Surface-seeking probe — This new device has a thin, needle-
like probe which constantly positions itself at, or slightly
above, the surface of the liquid. The probe is lowered by
a precision motor until the tip just makes contact with the
surface of the water. Controlled by an electronic circuit,
it then retracts slightly, and lowers again checking the
new level of the water. Cycles are repeated continuously
in this manner. The signal transmitting unit may be mounted
up to 7.6 meters (25 feet) above the water surface. The
approximate cost of a probe and recording computer is $2,000
(plus installation). Records of direct operations on storm-
water are not presently available [22].
Magnetic flowmeter — In this device, an electromagnetic
field is generated by placing electric coils around the pipe
using the flowing liquid as a conductor. The induced volt-
age, measured by electrodes placed on either side of the
pipe, is proportional to the flow velocity. The pipe must
be flowing full. Because grease buildups interfere with
accurate readings, the electrodes generally are removable
for cleaning. High capital costs (e.g., the factory cost
for a 125-cm (48-inch) diameter unit with indicator recorder
is on the order of $20,000 to $25,000) usually limit appli-
cations to small pipe sizes.
Ultrasonic and sonic devices — Ultrasonic measurement, a
development from World War II sonar work, has the advantage
of avoiding direct contact between the device and the liquid,
The best applications appear to be larger installations
where experienced technicians are available to maintain
calibration and perform repairs. Typically, both fluid
depth and velocity measurements are made. In measuring
depths, for good accuracy and to avoid extraneous signals
bouncing off the walls of the channel, the sonic transducer
should be as close as practicable to the surface of the
liquid.
A demonstration project using this type of equipment for
stormwater flow measurement is underway in Milwaukee [28,
14]. Velocities along several horizontal planes within
each of two prototype conduits (one 1.5-meter [5-foot] diam-
eter, the other 3.8-meter [12-1/2-foot)] are computed and
115
-------
averaged utilizing ultrasonic transducers placed at various
elevations within the conduits. The flow depth is measured
by a separate sonic transducer mounted above the liquid
level in a manhole. The installation in the smaller conduit
is performing satisfactorily; however, readings from the
larger are erratic probably because of high air entrainment
(a deep drop manhole is located just upstream of the meter
site). The small air bubbles increase the attenuation of
sonic energy by orders of magnitude and eliminate operation
[15].
The basic depth/velocity unit, manufactured in Japan, re-
portedly costs on the order of $10,000 to $15,000. The unit
has a continuous readout display capability.
Electrical devices — Electrical devices involve the use of
equipment such as conductivity cells, capacitors, hot-wire
anemometers, and warm-film anemometers. To date this equip-
ment has been found, for the most part, unsuited for sewage
flow measurements because of interferences from floating
and suspended material and lack of uniformity of the medium
being tested.
Applications — Typical applications of flow measurement in
stormwater management systems are listed in Table 23. Many
installations are limited to only stage monitoring and the
most commonly used devices are of the gas bubbler type.
Flow measurement practice at treatment facilities, as ex-
pected, reflects the conventional alternatives of flumes,
flow tubes, and magnetic flowmeters with occasional attempts
at innovation.
Recent research and development attempts to devise new units
have been only partially successful. For example, in one
study on the application of thermal techniques [8], it
was found that: (1) utilization of flush-mounted hot-wire
or warm-film anemometers in a direct-reading mode was
not feasible because of shifts in calibration resulting
from contamination buildup and lack of equipment ruggedness
and reliability; and (2) the technique of measuring time-
of-flight thermal pulses lacked adequate precision. In a
second study [21], capacitor plates were used to sense the
change in liquid level in the sewer pipe and heat pulse
timing for velocity measurement. The overall accuracy
of the final prototypes, which had 20- and 61-cm (8- and
24-inch diameters, was reported as ±15 percent at best.
Scum deposits on the walls of the gage and the nonrepresent-
ativeness of point velocities at the boundary were noted as
significantly and adversely affecting the accuracy of
velocity readings.
116
-------
Table 23. TYPICAL FLOW MEASUREMENT APPLICATIONS
Location
Boston, Mass . ,
Cottage Farm Stormwater
Treatment Station
Dallas , Tex. ,
Stormwater Treatment
Facility
Des Moines , Iowa,
At overflow points
Detroit, Mich. ,
In-system conduits
Milwaukee, Wis . ,
In-system conduits
Type of device
Flow tube
(Dall)
Parshall flume
Stick gages
Float
Gas bubbler
Pressure
actuated air
bellows
Ultrasonic
Design
maximum Make and
flow, cfs model no.
135 BIF model
121 Dall
22 24-in.
throat
(ea.)
Stage only
Leopold §
Stevens
Type F,
Model 68
Foxboro
components
Stage only, Bristol
up to 40 ft Company
155 Tokyo
Keiki Co. ,
Ltd.
(Badger
Meter Co.)
UF-100
Remarks
Located on pump discharge lines
(4 un i t s ) .
3 units. Used also for flow
distribution.
ated with weirs. Velocities
were spot checked with current
meters [4] .
118 units: sensors and trans-
mitters. Installed in lines
10-ft diam and larger.
2 sites, 5-ft and 12-1/2 ft
diam conduits. Combined
sewage.
Minneapolis-St.Paul,
In-system conduits
New Providence, N.J.
New York, X.Y.
Spring Creek
Detention Tanks
Racine, Wisconsin,
Dissolved Air
Flotation Facility
San Francisco, Calif.
In-system conduits
Seattle, Wash. ,
In-system conduits
Washington, D.C. ,
Conduit flows
Gas bubbler
(air)
Magnetic flow
meters
Magnetic flow
meters and
gas bubbler
Parshall flume
Gas bubbler
Gas bubbler,
float-skow,
sonic
Tracer
solution and
gas bubbler
Stage only -- 40 units located in 5-12 ft
diam conduits. Generally 3 per
regulator: 1 upstream, 1 down-
stream on overflow, 1 on diver-
sion to interceptor.
10 Fischer 2 units (12-in. and 14-in.diam)
Porter tied to automated flow control.
—2,000 Fischer Small 8-in. units suspended in
Porter large conduits to measure flow
velocities.
62 48-in. High turbulence experienced due
throat to proximity of screw pump.
Stage only, -- 120 units: 116 in combined
0-15 ft sewer conduits, 4 for tidal
stage. All with remote readout
and logging at central console
[D611].
Stage only -- Computed flows from stage
differential and gate opening.
System actuated by increase in
flow level detected by gas
bubbler. Lithium chloride was
then injected into sewage at
metered rate. Samples were
collected at periodic time in-
tervals 400-800 ft downstream
and analyzed. Flows were com-
puted on the basis of the lithium
concentration detected [3].
BIF Transmitter Model 231-20, Bell § Gossett compressor 1/10 hp, Norgren pressure regulator
all in vandal resistant enclosure. Approximate unit cost of $1,400 excluding hookup [11].
Float-skow prefired where small head differentials are encountered; sonic device where pre-
cision justified the cost (approximately $11,000 per installation including sensor and trans-
mitter [16] .
cfs x 28.32
ft x .305 -
in. x 2. 54 =
I/sec
117
-------
Direct-Reading Quality Sensors
Another essential element in comprehensive stormwater man-
agement programs is direct-reading, remote sensing of
quality characteristics. While the necessary parameters
that can be remotely and continuously monitored are many in
number, the results are not always satisfactory. For ex-
ample, remote and continuous sensing attempts of BOD5 and
COD have not been very successful in the past. Unfortu-
nately, these two parameters are generally considered among
the most important gross water quality indicators.
Remote sensing in wastewaters to date is primarily re-
stricted to in situ measurement by means of electrochemical
transducers without altering the physical-chemical charac-
teristics of the medium being tested by the addition of
reagents, etc. These sensors are usually capable of meas-
uring temperature, electrical conductivity, DO, turbidity,
pH, solar radiation, chlorides, oxidation-reduction poten-
tial, and alpha and beta radioactivity. Electrochemical
transducers are categorized according to the types of meas-
urement (1) conductometric, (2) potentiometric, and
(3) voltammetric. A detailed explanation of the theory in-
volved is provided by Mancy [13, pp. 141-196],
The primary problem in the use of electrochemical trans-
ducer sensors is the fouling of the electrode probes, which
makes the sensing system inoperative and requires extensive
servicing and calibration. Few electrodes have a reliable
method of self-cleaning. In the case of electrodes that
rely on a chemical reaction at the probe-medium interface,
such as pH sensing, polarization can occur and result in
false readings.
Typical Application — The following parameters, which are
sensed remotely at Seattle, Washington, illustrate the state-
of-the-art, particularly in the case of a computer-controlled
combined sewer system [10]:
1. Weather--rainfall intensity, duration, volume,
and wind direction and speed.
2. Storm and combined sewer overflow volume--measured
by sensing tide level, trunk level, wet well level,
and regulator and tide gate opening.
3. Receiving water quality--DO, pH, temperature,
solar radiation, turbidity, and electrical
conductivity [Schneider Instrument Co., Model
RM25AT].
118
-------
4. In-system flows, storage, and potential problems--
level sensors and explosive or other hazardous gas
meters.
Other parameters that are remotely monitored less frequently
in water quality monitoring programs throughout the country
are oxidation-reduction potential, chloride, and fluoride.
Research — The recent emphasis on remote monitoring and
control, and the demonstrated effectiveness of such systems,
are encouraging the development of new devices.
One such device, a prototype unit using depolarization of
scattered light measurements and capable of monitoring
SS concentrations in wastewater (ranging from a few mg/1
to 5,000 mg/1 by weight), has been developed, and initial,
very limited testing has been completed [24]. In the device,
polarized light is directed into the flow stream via a
polarizer (e.g., prism) and backscatter is measured in terms
of a ratio (polarized to depolarized to light). During
measurements, the present instrument requires attention
for adjustments of the electronics. Further development
has been recommended.
Remote and continuous sensing of chemical and biological
parameters generally has not been possible in the past;
however, some automatic analyzers do exist, such as those
used for TOG determinations, and are continuously being
perfected into reliable and accurate instruments. In
sanitary engineering, these parameters must be known to
apply complete combined sewer system management and control.
The 5-day waiting period needed to obtain standard BODs
loadings is, of course, impractical for storm-flow routing
decisions. Instead, the present practice is to attempt
to collect large amounts of data and to construct prediction
models. Such programs may take years to complete and
the predicted result may prove to be disappointing. It is
readily apparent, then, that much more emphasis should be
placed on the research and development of acceptable remote
sensing of these parameters.
Analysis
Effective data analysis should include, as a minimum, the
definition .of: (1) flow extremes (e.g., the ratio of dry-
weather flow to maximum conduit capacity); (2) frequencies
of occurrence of flow rates and parameter loadings;
(3) types and frequencies of samples; (4) mean values and
ranges of characteristics; (5) rates of change patterns and
pre-storm impacts; (6) site and time dependency (e.g., size
of area, land use, time of day, and seasonal effects); and
119
-------
(7) special conditions or qualifications (e.g., construc-
tion impacts, plant bypasses, atypical flows or source
area management, snow melt versus rainfall-runoff).
Because a wide scattering of data points may be expected,
arraying the data, holding specific variables of interest
constant, may yield significant information. For example,
to determine the probable BODs concentration variation in
overflows within the District of Columbia, values were
arranged as a function of the time elapsed since the start
of overflow [18] . These data were based upon 35 storm
events, 3 independent sources, and both combined and sepa-
rate systems. The results, presented in Table 24, show
clearly a progressive reduction in concentrations with time
and a marked difference in the magnitude of the combined
versus separate systems.
Statistical analysis of hydrologic data is generally justi-
fied because of the vast amount of source data available
(e.g., U.S. Weather Bureau data); however, actual in-system
quantity and quality data are seldom that abundant. One
exception worth pursuing would be treatment plant influent
concentrations on wet days versus dry days. In any analysis,
Table 24. BOD5 CONCENTRATIONS VERSUS TIME
FROM START OF OVERFLOW [18]
DISTRICT OF COLUMBIA
a. Ranges indicate the average low values and the average high values for individual storms where available. First hour,
etc., is measured from start of overflow.
b. Reported range and mean of all (94) samples analyzed.
Sources :
Source 1 - Sampled data December 1968 to September 19h9 (2bJ.
Source 2 - Sampled data May to August 1'Jh'J (3|.
Source 7, - Completed data July to August 19<>9 |5|.
120
-------
care must be taken in precisely defining the array limits
and securing sufficient independent data points to obtain
credibility. Data correlations with values found in the
literature may facilitate expansion of minimum data resources
for a starting basis of design.
Analytical procedures and reporting of results should follow
those outlined in NPDES General Instructions, Appendix A —
Standard Analytical Methods [20] . A valuable reference
source is Methods for ChemTcal Analysis of Water and Wastes,
1971 Edition by EPA, with subsequent changes and errata.
In interpreting the data, not only must the uncertainties
of the sampling and sample preservation be questioned but
also the precision and accuracy with which the analytical
work can be performed. Using 6005--perhaps the most com-
monly measured parameter in wastewater--as an example, the
EPA 1971 text reports under "Precision and Accuracy" (using
a stock solution under ideal conditions'):
Seventy-seven analysts in fifty-three labora-
tories analyzed natural water samples plus an
exact increment of biodegradable organic com-
pounds. At a mean value of 194 mg/1 BOD, the
standard deviation was ±40 mg/1. There is no
acceptable procedure for determining the
accuracy of the BOD test.
PROCESS AND EQUIPMENT EVALUATION
Whereas conventional wastewater treatment is based on com-
paratively steady-state conditions, stormwater treatment
must adapt to intermittent and random occurrences. The
flows and quality characteristics, as identified earlier,
are subject to high variability over short periods of time.
Because of this variability, there is no such thing as an
"average" design condition.
How, then, does an engineer evaluate the performance effi-
ciency of particular processes or equipment units in storm-
water applications? First, the process or unit results
should be valued against the program objectives--that is,
should the treatment be oriented toward mass removals or
limiting concentrations in the effluent? Mass removals are
of primary concern when discharging into impoundments
(lakes), estuaries, or other locations where accumulations
may occur. Concentrations may be of primary interest where
shockloadings or threshold values may be limiting, but re-
ceiving water flows are sufficient to prevent accumulations
(i.e., each storm functions as an independent event).
Second, effective utilization, such as hours of operation
121
-------
per year, reduction in overflow occurrences, etc., is an
important criterion. Third, service reliability must be
evaluated (i.e., did the unit/process function as it should
have and when it should have). A unit that performs only
when conditions are right may be too restrictive for prac-
tical applications in stormwater management.
In addition to variability, the magnitude of the flows
alone is a major design and evaluation constraint. Peak
flow rates may equal or exceed 50 to 100 times dry-weather
flows from the same area; thus, facilities must be exorbi-
tantly large in size or supported by equalization storage.
The economic justification of mammoth treatment facilities
that are used only infrequently is difficult to ascertain.
For example, San Francisco has an average of 381 hours
of rainfall per year (4 percent of the total time) of which
only 5 hours exceed an hourly intensity of 0.76 cm/hr
(0.30 in./hr)--the design peak flow rate of the Baker Street
combined overflow treatment facility. As a result, the
facility would appear to be under-utilized 99.94 percent
of the time!
High debris content in urban runoff driven by high veloci-
ties and turbulence may render sophisticated and complex
equipment ineffectual or impossible to maintain. Because
of the high flow rates (e.g., peak runoff from a 5-year
storm), wastewater transmission costs are very high. This
constrains options for centralization and frequently forces
construction in prime real estate areas. Thus, not only
simplicity but also compactness and aesthetic appearance
of units are necessary considerations.
In summary, four primary guides are recommended for weighing
future design and performance evaluations:
1. Reliability/durability based upon the frequency of
total and partial unit operations to the total
number of continuous storm events.
2. Efficiency as measured by the total mass of pollu-
tants removed by the facility as a percentage of
the total mass applied over the complete storm
event including upstream bypasses.
3. Effluent quality as measured by the average and
maximum concentrations measured in the discharge.
4- Dual use as measured by the effective utilization
of the facility during non-storm periods. -
122
-------
Finally, it should be noted that there is no unique solu-
tion to all of the problems associated with stormwater
discharges. Even the extreme of total capture of the run-
off, while totally removing the pollutants, may have
adverse effects on the receiving water by hindering natural
flushing and replenishment.
Method of Approach
To assess process and equipment applicability to the storm-
water problem, the following sources/methods are generally
available: (1) site visits and interviews; (2) project re-
ports, seminar papers, and published articles; (3) progress
reports and records of post-construction performance;
(4) general background literature surveys; and (5) records
of construction periods and cost breakdowns. Having first
reviewed the material compiled herein, the reader may be
guided by this methodology to further his specific
investigations.
It must be recognized that essentially all projects reviewed
in the preparation of this text were designed primarily to
develop and/or demonstrate new or improved methods of con-
trolling the discharge into any waters of untreated or
inadequately treated sewage or other wastes from sewers
which carry stormwater or both stormwater and sewage or
other wastes [27]. Thus, to induce marked advances in the
state-of-the-art, projects were directed into generally
uncharted areas of design expertise (e.g., abnormal loading
rates, high equipment and material stresses for short peri-
ods, extreme turbulence, etc.). Therefore, the success
or failure of a particular project is not as important
as what was learned and can be applied to subsequent work.
For this reason, a major requirement of all projects is that
facilities be included to maximize the flexibility of pro-
totype operation (alternate sources of test flow and dis-
posal, alternate reduced flow and/or increased loading
configurations, capability of planned equipment shakedowns,
etc.). It was observed that projects without this built-in
flexibility were generally restricted severely in their
performance.
The most important factors in assessing process and equip-
ment performance are the project scale and pretreatment
controls. The scale at which the demonstration tests were
performed may tend to mask the credibility of the results
obtained. Likewise, pretreatment may so alter the charac-
teristics of the waste that the tests will be nonrepresenta-
tive of subsequent applications lacking such pretreatment.
123
-------
Common scales are bench, pilot, and prototype. Bench
scale is generally accomplished within a laboratory under
precise control; pilot scale is field testing at a reduced
scale and possibly using a synthetic flow; and prototype
is full-scale but stressing a demonstration configuration.
Pretreatment may include screening, flow equalization and/or
settling, and feed control. For example, a test facility
fed by a single constant-flow pump, or a restricting series
of pumps, likely will not reflect prototype instream perform
ance but rather prototype performance when balanced by up-
stream storage. Also, the pump suction may effectively
screen out certain solids, floatables, and debris.
Similarly, a test facility operated on a preferential basis,
say on an 8:00 a.m. to 5:00 p.m. shift and/or on selected
storms, will not provide the same data as a facility treat-
ing all storms at all hours (particularly with respect to
reliability). Obviously, the facility that best approaches
the prototype in size and operation is the most viable.
Another important factor is the location of the facility
and the nature of the study area in which it is located.
A facility located within or directly adjacent to a continu-
ously manned treatment plant may reflect better maintenance
and supervision, hence better operating results, than a
remotely located unit with limited access. The accepta-
bility of the unit to the public is a necessity. The
nature of the study area (intensity of development, use,
industrial flow component, climate, topography, etc.) may
significantly influence unit performance and the attention
it receives. Sampling methods, analysis, and control as
well as the study duration must also be considered.
In assessing and/or executing all projects, the study objec-
tives should be clearly identified, and a detailed program
should be developed and adhered to. The percentage of
total storms treated may be as important as the degree
of efficiency obtained in selected storms. Innovations,
repairs, and failures should be carefully logged to provide
a basis for succeeding work.
Basic Design Data
Basic design data, operational controls and procedures, and
simplified flowsheets should be sought for each project
studied. How were the unit and its appurtenances sized?
What were the available design modes (flow paths, recircu-
lation ratios, biological or chemical feed rates, automatic
or manual controls, sequential startup, etc.) and which
ones were utilized?
124
-------
Largely, the basic design data essentially are fixed at the
same time that the physical dimensions are fixed, and the
pumps and other support facilities are selected (exceptions
include screen mesh, filter depths, selective bypassing,
etc.). Flow controls, pretreatment applications, and chemi-
cal feed are then the basic variable tools used by the
operator or designer to optimize the unit's performance
for the individual storms. In practical applications,
without some form of available storage, flow control options
beyond the process unit loading flexibilities are minimal,
because lowering the flow to the unit to improve efficien-
cies in the treated fraction results in corresponding flow
increases in the untreated bypass fraction.
To establish a stormwater treatment process, in all but the
simplest devices, there appears to be a similar need for a
minimum level of prior storage to allow for startup, moder-
ation of flows, and sustaining the process. Depending on
the local hydrology and topography, this storage may be
obtainable within the system without resorting to off-line
units.
Finally, there must be provision for the ultimate disposal
of the residue (e.g., tank dewatering and solids disposal)
and plant deactivation between storms. Because storm events
are intermittent, all processes are necessarily operated on
a batch basis unless essentially total storage (capture) is
effected or recycling is feasible.
Operational Results
Operational results include a description of the maintenance
and operation activities completed, the volumes processed,
pollutants (solids) removed, power and chemicals expended,
and labor applied. Efficiencies vary from storm to storm.
Ideally, efficiencies should be based on the total measured
pollutants removed by treatment divided by the total pollu-
tants in the untreated storm flow including bypasses. More
commonly, however, they are the results of composite samples
taken before and after the units, and do not account for
bypasses. Normally, they are not flow weighted. For
selected analysis, discrete samples may be collected and
subsequently composited manually in proportion to the
recorded flow rates. This procedure is costly and must be
done with care. The multiple handling of the samples may
introduce errors, particularly in respect to floatables and
solids. To automate flow proportional samplers fully, the
pre-storm guesswork must be exceptionally good. For example,
with a not unusual 50 to 1 flow variation, the minimum size
and frequency of sampling must be such as to arrive at a
125
-------
total volume that is large enough for performance of all
of the analyses but not so large as to overflow the
container.
Thus, while no simple evaluation rules for assessment are
possible, a prior knowledge of desirable features and limi-
tations can provide effective guidelines.
CONTROL SYSTEMS
Because storms occur without respect for nights, holidays,
and weekends, and give little advance warning, some form of
automatic operational control is required. Treatment effec-
tiveness is largely dependent upon the facilities coming
on-line almost instantaneously and, to a varying degree,
self-adjusting to changes in flow and concentration.
Coordinating multiple facilities into a highly effective
management program requires both remote sensing and central-
ized control. An introduction to control systems is pre-
sented here. Discussions of particular applications and
features are presented in Sections VIII and XV.
System operational control spans a spectrum ranging from
monitoring alone to complete, or "hands-off," automatic
control. While a number of levels of development, each
with several variations, could be delineated, there appear
to be three basic stages: system monitoring, remote super-
visory control, and automatic control [17]. Advances in
control capability for urban water resources have evolved
mostly from systems for monitoring field variables.
The basic elements of a control system may include all or
combinations of the following: (1) remote sensors (rain
gages, flow level and selected quality monitors--such as
DO, TOG, SS, and/or pH probes, gate limit switches, etc.);
(2) signal transmission (leased telephone wires, pneumatic
circuits); (3) display and logging (central computer,
graphic panels, warning lights); (4) centralized control
capability (control of system gates and/or pumps from a
central location); and (5) in the case of fully automated
control, a computer program that makes decisions and exe-
cutes control options based upon current monitoring data
and memory instructions.
A block model for a complete automatic operational control
system is shown on Figure 16. This represents ^an ultimate
system and is a goal not yet achieved in major scale for
management of any basic urban water resource function [17].
Approaches to this goal, however, appear to be well underway
126
-------
ON-SITE
VARIABLES
DISPLAY
SIGNAL PROCESSING AT CONTROL CENTER
LOGGING
FIELD
PARAMETERS
DATA
ANALYSIS
DISPLAYS
CHARTING
FIELD
PARAMETERS
SURVEILLANCE BY HUMAN SUPERVISOR
ON-SITE
DISPLAY OF
RESPONSES
Figure 16. Model for automatic
operational control system [17]
127
-------
System Monitoring
The basic components start with field instrument sensing and
the transmission of the signals back to a central processor
and display. This is the basic remote monitoring function.
"On-site variables display," when included, is for checking
fidelity of instrument response, instrument malfunction,
etc., at individual instrument sites. In the absence of
this feature, instruments can be checked only by using
auxiliary radio or telephone communications with the moni-
toring center. The simpler monitoring systems feature
only data logging, visual display charting, and warning
and alarm displaying. Some type of computer capacity is
required for data analysis, but this can be off-line.
Completely missing is any direct field control function.
Remote Supervisory Control
Remote supervisory control adds a remote manual control
capability to the basic monitoring system. In this configu-
ration the supervisor observing the monitoring display may
initiate immediate corrective action by remotely opening or
closing gates, changing pump speeds, etc. "Control logic
queries" refer to supervisor interrogations of computer-
programmed logic, which can range from elaborate algorithms
incorporating prototype response simulation through simple
retrieval of command alternatives stored in the computer
memory. As in the situation for data analysis, computer
access can be off-line.
Actuation of field control elements is performed remotely by
manual supervisor command. In simpler systems, control
element reponse to an actuation command can be ascertained
indirectly by observing changes in affected field sensor
signals. On the other hand, the "response signals telem-
etry" to the control center provides direct control-loop
feedback, with an opportunity to damp system actuation
response instabilities by using guidelines preprogrammed
as part of the control logic.
«r
Automatic Control
The next and final step to complete automatic operational
control adds an "automatic control signals generator" and
converts the manual control function to a manual override.
Thus, a series of monitoring signals will prompt, through
a computer analysis program and memory, a particular con-
trol response or sequence of responses to monitored data.
In stormwater management applications, the automation may
be as simple as starting an auxiliary pumping unit when a
128
-------
conduit flow depth exceeds a specified level, or it may be
a complicated system involving dozens of rain gages, level
sensors, multiple storage/treatment options and devices,
all capable of intelligent and rapid manipulation to
counter any storm pattern or emergency condition. Analysis
of the systems requires an understanding of the components
and their interrelationships. As with processes and equip-
ment, evaluation depends upon performance.
COSTS
Costs of stormwater management facilities are highly depend-
ent upon the size, the process, location, time of construc-
tion, and provisions for solids handling and disposal. The
primary variables, exclusive of the process itself, are:
land, weather, foundations, groundwater, access, conditions
of the existing system, temporary services, operator facili-
ties, component life, and aesthetics. The operator facili-
ties include laboratories; showers and lockers; humidity,
temperature, and ventilation controls; degree of automation,
flexibility; etc. The sparsity and individuality of real
installations make generalizations difficult and their use
hazardous.
Of interest to the engineer/planner, with these limitations
in mind, are (1) data sources, (2) transferability and up-
dating, and (3) economies of scale.
Data Sources
Cost data were extracted from project reports, bid tabula-
tions, contractor and/or manufacturer quotations, and
published articles. Excellent general references based
upon conventional wastewater treatment processes and broken
down by unit function include the following: (1) estimates
for facilities costs and manpower requirements, 1971 [9];
(2) cost of conventional and advanced treatment, 1968
[23]; and (3) cost and performance estimates for tertiary
processes, 1969 [6]. Readers are cautioned that the
estimating data and methods presented cannot in any way
be used as a substitute for cost estimating based on de-
tailed knowledge of a particular installation.
Estimates of cost for well-defined structures at definite
locations for a current specific period, and involving
fairly well-known working conditions, can be made with a
reasonable degree of accuracy. A favorable comparison with
actual bid prices for similar work also is important in
that it adds a measure of certainty. The absence of any
one of these essential elements lessens the level of
reliability. The absence of all such elements reduces the
129
-------
value of cost estimates to something that is largely judg-
ment, supported wherever possible by adjusting available
cost data.
Updating and Transferability
In presenting construction cost data in this report, an
attempt has been made to adjust all data to a common base:
an assumed U.S. Average Engineering News-Record (ENR)
Construction Cost Index of 2000.This index, commonly used
in engineering evaluations, is updated monthly and described
in the magazine as follows:
ENR developed cost indexes for 22 individual
cities in the U.S. and Canada to show the trend
in basic costs--construction materials and wage
rates, in each major construction center. Ex-
cept for steel which is priced at three major
mill centers, the indexes use local prices and
wages. They are not intended to measure the
cost differential between cities. They include
no special adjustments for variations in produc-
tivity or design requirements.
Components and weighting: 200 hr common labor,
20-cities average; 25 cwt structural steel
shapes, mill price; 20-cities averages of 22.56
cwt (6 bbl) of Portland cement and 1.088 Mbfm
2 x 4s lumber.
The trend of the ENR U.S. Average Construction Cost Index,
plotted at semiannual intervals spanning the last 20 years,
is shown on Figure 17. The June 1973 U.S. Average is
1896, and the average for 20 cities varies from a low of
1427 at Birmingham, Alabama, to a high cost of 2344 at
New York.
The following relationship was used to adjust costs cited
in the literature to report (ENR 2000) values:
Report value = ENR CC Index (for city at time x cited cost
of construction)
If the month was not given for the year construction oc-
curred, the ENR Index for June of that year was used.
130
-------
2000
1800
00 —
CD OS
e_3
. 600
400
200
JUN DEC JUN DEC JUN DEC JUN
53 55 58 60 63 65 68
YEAR
DEC JUN
70 73
Figure 17. Construction cost trend
Similarly, by projecting cost trends into the future, order
of magnitude costs of deferred construction may be obtained
ENR CC Index (projected to the
met - time of future construction) . -
cost - - „, - 1— x report value
Economies of Scale
In conventional processes and equipment for domestic waste
water treatment, there are well defined economies of scale
Thus, it is normally expected that a unit of capacity in a
131
-------
large plant will cost less than a unit of capacity in a
small plant. The general relationship is of the form:
Where C. = cost of an item of capacity QA
A n
K = base cost factor and equals C^/Qg
M = measure of the economy of scale
The derivation, cost factors, and applications are dis-
cussed in detail by Berthouex [2]. It was reported that,
in general, M has a range of 0.5 to 0.9, with each type of
equipment and each type of processing plant having its
characteristic value. The resultant cost projections are
estimated to be within a range of ±20 to 30 percent.
Again, the sparsity of data limits a determination of the
degree of direct applicability of this concept within the
stormwater management field.
132
-------
Part III
MANAGEMENT ALTERNATIVES
AND
TECHNOLOGY
-------
-------
Section VII
SOURCE CONTROL
Source control includes all measures for reducing storm-
water pollution that involve actions within the urban drain-
age basin before urban runoff enters the sewer system.
Included are measures that affect both the quantity and
quality of urban runoff. Examples of source control meas-
ures for reducing the quantity and/or the rate of urban run-
off include (1) use of roof storage, (2) intentional ponding,
(3) disconnection of area and roof drains, and (4) use of
porous pavements. Examples of source control measures for
improving the quality of urban runoff include (1) decreasing
dustfall on the area by reducing air pollution, (2) erosion
control during construction of buildings and highways,
(3) placing berms around small lots, (4) improved street
sweeping practices, (5) removal of lead compounds from gaso-
line, and (6) improved methods for deicing pavements.
In the past, efforts in water pollution control have been
concerned primarily with point sources--houses, factories,
offices, etc. In the future, as the nation restores the
quality of its surface waters, it will be necessary to
pay greater attention to nonpoint (area) sources. In this
discussion attention will be focused on nonpoint sources in
urban areas.
QUALITY CONTROL MEASURES
In dealing with environmental problems it is often necessary
to make trade-offs. For example, by reducing the concentra-
tions of sulfur dioxide or particulate matter in the air,
urban runoff quality is improved. In determining the sig-
nificance of air pollution control, it is important to con-
sider possible benefits of improved quality of local
watercourses. In general, practices leading to urban clean-
liness will also lead to improvement in water quality of
urban runoff.
135
-------
It is extremely difficult to quantify the potential reduc-
tion in stormwater pollution resulting from improved air
pollution control. Contaminants scoured from the air would
probably be reduced in proportion to the reduction of their
airborne concentrations. Industrial stockpiles and open
storage of granular materials are sources of both air and
stormwater pollutants resulting from wind action and/or rain
falling directly on the material. However, most of the
contamination of urban runoff is caused by air pollution
particles previously settled on city surfaces--roofs, side-
walks, and streets. The amount of contaminants picked up by
urban runoff is dependent upon fallout rate, frequency and
efficiency of street cleaning, and the path that runoff
follows before entering a stormwater inlet.
Solid Waste Management
Although intentional disposal of waste materials on city
streets and sidewalks is generally prohibited, it is prac-
ticed commonly. Spent containers from food and drink,
cigarettes, newspapers, floor sweepings, and a multitude of
other materials carelessly discarded become street litter.
Unless removed by street cleaning equipment, these mate-
rials often end up in stormwater discharges. Enforcement
of antilitter laws, convenient location of sidewalk waste
disposal containers, and public education programs are all
source control measures that may provide significant water
pollution control benefits. It is difficult, however, to
measure the effects of such measures in economic terms.
Two benefits that do occur are aesthetic improvement of the
urban area and reduced pollution of the urban runoff.
Materials tossed through stormwater inlets are not part of
the street litter problem, but they are part of the storm-
water problem. Typical examples are leaves, garden clip-
pings, and used automobile crankcase oil. Public education
is the most valuable measure for reducing this type of
stormwater pollution. Most people have very little idea
of what happens to waste discharged to stormwater inlets.
By establishing a solid waste management program that en-
courages proper disposal of most solid wastes, benefits may
accrue to water pollution control. Again, it is difficult
to measure these benefits in economic terms.
Street Cleaning
Specially designed street sweeping vehicles are used by
most cities to prevent the accumulation of litter, dirt,
and dust on city streets. In some cities part of this work
is still done manually, but mechanized equipment is being
136
-------
relied upon increasingly for reasons of efficiency. Street
sweeping is done for aesthetic reasons, not for water pol-
lution control, although most material removed by street
sweeping would otherwise be included in storm sewer dis-
charges or combined sewer overflows during wet weather.
The effectiveness of street sweeping operations with respect
to stormwater pollution has been analyzed in two recent EPA-
sponsored studies [12, 8]. It was found that a "...great
portion of the overall pollutional potential is associated
with the fine solids fraction of the street surface
contaminants" [8], Current broom-type street sweepers, how-
ever, were relatively inefficient in removing fine material
smaller than 400 microns. From the data reported in
Table 25 it can be concluded that the smaller the particle
size, the lower the sweeper efficiency. The overall effi-
ciency of the street sweeping operation can be greatly
increased by scheduling streets for cleaning and posting
"no parking" signs for those hours and by improving such
schedules. Tests by APWA of vacuum cleaning equipment for
municipal street sweeping indicated removals of 95 percent
or higher for fine material.
Table 25. STREET SWEEPER EFFICIENCY
VERSUS PARTICLE SIZE [8]
Sweeper
Particle size, efficiency,
microns %
2,000 79
840-2,000 66
246-840 60
104-246 48
43-104 20
<43 15
Overall 50
137
-------
Use of Chemicals
Chemicals of a wide variety are spread in the urban environ-
ment for various purposes. The two most important groups
of chemicals are those used for control of snow and ice on
highways and those for the control of vegetation.
Street runoff from the melting of ice and snow, mixed with
chloride salts, finds its way to nearby receiving waters by
several different means: (1) through local sewage treatment
plants via combined and sanitary sewers, (2) through storm
sewers, and (3) by dumping of snow removed from streets into
the nearby waterways. According to Field:
The dumping of extremely large amounts of accu-
mulated snow and ice from streets and highways,
either directly or indirectly into nearby water
bodies, could constitute a serious pollution
problem. These deposits have been shown to
contain up to 10,000 mg/1 sodium chloride, 100
mg/1 oils, and 100 mg/1 lead. The latter two
constituents are attributable to automotive
exhaust. [13]
In terms of source control, several things can be done to
minimize the contamination of urban runoff by deicing chemi-
cals and abrasives [1, 13]. One possibility is simply to
prohibit the use of certain chemicals: in particular,
those highly toxic substances, such as cyanide and chromium
compounds, which have been added to deicing salts as anti-
caking agents and corrosion inhibitors. Although highway
departments may be willing to accept prohibition of addi-
tives, few desire the prohibition of the use of deicing
salts, such as sodium chloride and calcium chloride.
The easiest way to minimize adverse effects of deicing
salts is by using less of them. For example, in Ann Arbor,
Michigan, the following steps were found useful in reducing
the rates of salt application without sacrificing safety:
(1) no salt application on straight, flat sections;
(2) better training for operators of salt-spreading equip-
ment; and (3) keeping records of salt use [5]. In addition,
Ann Arbor is investigating new types of spreaders and snow-
plows, improved cab monitoring devices that will control
salt application more accurately, public information pro-
grams on winter driving, and new training programs for
equipment operators. In many other cities salt application
virtually replaces snowplowing. However, as adverse impacts
of excessive salt use are more widely recognized, more
138
-------
cities will develop programs to minimize salt use without
sacrificing highway safety.
Toward the goal of reducing road salt damage, the following
possibilities were investigated in an EPA-sponsored project:
(1) external in-slab thermal melting; (2) stationary/mobile
melters ; (3) substitute deicing compounds; (4) compressed
air types of snowplows; (5) adhesion reducing pavement mate-
rials; (6) solar energy storing pavement substances;
(7) electromagnetic ice shatterers; (8) improved drainage,
enhancing runoff, accident reduction, and snow melt control/
treatment; (9) salt retrieval/treatment; and (10) improved
tire/vehicular design [13]. A deicer users' manual and a
manual of design for storage facilities and recommended
methods of handling deicing materials throughout storage is
to be provided in another study [13].
An alternative to salt is the use of abrasives, such as
sand and cinders. Abrasives, however, do not fit into the
"bare pavement" policy which now prevails, and they can
also become stormwater pollutants. Although abrasives gen-
erally contribute only small amounts of dissolved solids to
stormwaters, they can contribute significant portions of
the SS. Furthermore, large quantities of sand and cinders
can clog storm and combined sewers. Most cities remove a
large portion of the abrasives during street cleaning, but
the added cost of collecting large amounts of sand and
cinders must be included in the cost of their use. In addi-
tion, abrasives are more expensive (by weight) than sodium
chloride. With salts, however, many of the costs, such
as those for corrosion damage, degradation of water supplies,
and damage to roadside vegetation, are indirect and often
ignored. Hence, in some instances, a more complete economic
comparison might favor abrasives over salt.
With regard to control of vegetation, fertilizers, pesti-
cides, herbicides, and other chemicals are widely used in
cities and have been found in urban runoff [11]. The amount
of these materials in urban runoff, taken all together, has
been found to be rather high [7]. Limited use of these sub-
stances consistent with their intended purpose and careful
attention to their storage and distribution should be
practiced.
Erosion Control
There are many methods for reducing sediment yield from
urban areas, especially for land undergoing development.
This is important because of the extreme sediment yields
often occurring on such land. For example, in a 1961-1964
139
-------
study of sediment movement in the Scott Run basin, Fairfax
County, Virginia, it was determined that "...highway con-
struction areas, varying from less than 1 to more than
10 percent of the basins at a given time, contributed 85
percent of the sediment" [9]. The sediment yield of the
highway construction area was found to be 10 times that
from cultivated land, 200 times that from grassland, and
2,000 times that from forest land. Scott Run is a suburban
tributary of the Potomac River. In a study of the entire
Potomac River it was estimated that the Washington Metro-
politan Area, which consists of 2 percent of the basin,
produced 25 percent of the sediment in the river.
For this reason the EPA sponsored a project to develop
"Guidelines for Erosion and Sediment Control Planning and
Implementation" [3]. Among the techniques described to
reduce sediment yields are: (1) proper selection of build-
ing and highway sites, (2) maintenance of native vegetation,
(3) use of mulches, (4) drainage channel protection modi-
fication, (5) careful backfilling after laying pipes,
(6) protection of stockpiles for removed earth, (7) sedi-
ment retention basins, (8) timing of clearing and grading
during season when erosion is less, (9) traffic control,
and (10) use of fences to protect trees. Four appendixes
of that report contain technical information on 42 sediment
and erosion control products and practices.
QUANTITY AND/OR RATE CONTROL MEASURES
For almost any system for abating storm sewer discharges and
combined sewer overflow pollution, the cost involved is very
sensitive to both the quantity and the rate of the flows
involved. For example, a combined sewer overflow screening
facility designed for a 1-year, 1-hour rainfall might treat
several million gallons within 1 or 2 hours. Any techniques
or devices to slow the flow of significant amounts of storm-
water to the stormwater inlets would increase the period
in which the runoff volume could be treated. Hence, a
smaller and less expensive treatment facility would afford
the same degree of pollution abatement. Slowing the flow of
stormwater to the inlets, especially from pervious areas,
allows additional recharge of the groundwater by percolation
of the stormwater into the ground.
By temporarily detaining (on-site) runoff from rainfall
directly falling on an impervious area, it is possible to
reduce the rate of flow into the sewer system. This results
in less flow to store and/or treat. If this is coupled with
retaining (on-site) runoff from pervious areas for percola-
tion into the ground, then the total volume of water enter-
ing the sewer system is reduced. Hence, considerable
140
-------
savings in both operation and initial construction costs of
stormwater facilities will result when the amount of storm-
water is attenuated. Methods used for slowing or dampening
the rate at which flows enter the sewer system by temporar-
ily holding runoff on an area are termed "detention"
measures. Methods used to prevent runoff from entering the
sewer system at all are termed "retention" measures.
In an article entitled "Storm Water for Fun and Profit,"
several urban detention/retention techniques are described
110], At an industrial facility in Bensenville, Illinois
regrading of the plant site produced multiple benefits:
(1) the plant is now above the flood plain; (2) borrow pits
are now lagoons to retain stormwater; (3) the stormwater
is used as industrial supply water at a cost of $0.04/kl
($0.015/1,000 gal.); and (4) lagoons were landscaped and
provide aesthetic and recreational benefits (picnicking,
fishing, etc.). In other projects utilization of rooftops
tennis courts, ponds, and plazas to detain and/or retain
precipitation in urban areas is described.
An example of a large scale municipal multipurpose detention
basin is the Melvina Ditch Detention Reservoir constructed
by the Metropolitan Sanitary District of Greater Chicago in
Oak Lawn, Illinois [6]. The reservoir, which has a capacity
of 204,000 cu m (165 acre-ft), was designed to serve as a
recreation facility in addition to its primary function of
reducing local flooding. Steps were constructed down the
basin side slope. Winter recreation activities include
tobogganing and skiing on a large earth mound formed in one
corner of the basin. A concrete paved area (anti-erosion
section at the inlet) is flooded during winter months to
serve as an ice-skating rink. During the summer months, it
is used for volleyball, basketball, and general play. The
reservoir is shown in operation as a stormwater detention
facility in Figure 18.
In a rather unique approach to urban runoff control, the EPA
has sponsored an investigation of porous pavements [4]
Pavement for streets, sidewalks, and parking lots make up a
large percentage of the impervious area of metropolitan
areas. If precipitation could pass through the pavement and
recharge the groundwater, stormwater would become a resource
rather than a substance to be disposed of. This would re-
quire that the precipitation would be sufficiently treated
by passage through the soil so that street surface contami-
nants would not pollute local groundwater. Although porous
pavements were originally developed for highway safety
purposes, they show considerable promise as a method to
attenuate runoff. In cold climates, porous pavements may
not be practical because of the danger of paving destruction
141
-------
Figure 18. Stormwater surface detention pond (Chicago)
142
-------
due to the alternate freezing and thawing of a semisaturated
subbase. It has been reported that frost-heave can be over-
come by providing a gravel base layer of sufficient thick-
ness to equal or exceed the gravel base reservoir capacity
required to accept the rainfall percolating through the
porous pavement [4]. In the study it was concluded that
"roads designed with porous asphaltic concrete were found to
be generally more economical than conventional roads with
storm sewers." Other benefits, including augmentation of
municipal water supplies, highway safety, relief of flash
flooding, preservation of vegetation, and prevention of
puddling, are described in the published report.
In most urban areas, considerable economic and intangible
benefits are available from improved planning and practices
for the disposal of stormwater. Here, the most difficult
problem may be in adding a constraint for architects,
planners, and departments of public works to consider in
their practices. As described in "Storm Water for Fun and
Profit," however, it is often worth the effort [10].
An example of a community where several different methods of
source control are being employed is The Woodlands north of
Houston, Texas. This will be an entirely new community situ-
ated on 73 sq km (18,000 acres) to be developed over the
next 20 years. Two major goals are (1) preserving as much
as possible the natural environment of its setting and
(2) minimizing increases in pollution loadings resulting
from increased stormwater flow rates [2].
Runoff will be recharged to the ground, as near as possible
to the point where the rain falls. This will be achieved by
a "natural drainage concept" which includes the following:
1. The existing natural drainage system will be uti-
lized to the extent possible in its unimproved
state. Existing drainage courses are grass covered,
thus slowing and reducing runoff through
infiltration.
2. Where drainage channels are required, wide, shallow
swales lined with existing native vegetation will
be used instead of cutting narrow, deep drainage
ditches.
3. Flow retarding devices, such as retention ponds and
recharge berms, will be used where practical to
minimize increases in runoff volume and peak flow
rate due to urbanization. A storage reservoir will
decrease the amount and rate of runoff and promote
143
-------
recharge of groundwater. Erosion control measures
in construction areas will minimize the increased
solids loadings in runoff from such areas.
4. Drainage pipes and other flood control structures
will be used only where the natural system is in-
adequate, such as at high density urban activity
centers. Plans presently call for the use of
porous pavements to reduce runoff from streets.
5. Control will be exercised over the type and amount
of fertilizers, pesticides, and herbicides to mini
mize pollution of the runoff.
It has been estimated that the drainage system will cost an
average of $243/ha ($600/acre) , compared with perhaps
$486/ha ($l,200/acre) for a conventional system [2].
144
-------
Section VIII
COLLECTION SYSTEM CONTROL
Collection system control, as presented in this text, per-
tains to those management alternatives concerned with the
interception and transport of the wastewaters. These alter-
natives include sewer separation, catch basin cleaning,
infiltration/inflow control, line flushing and polymer injec
tion, regulators, and remote monitoring and control. Each
of these alternatives is discussed in this section.
A definition drawing of the common elements of an intercep-
tion and transport system and their interrelationships,
adapted from [6, Plate IV-4], is shown on Figure 19. The
system shown represents a combined sewer system and illus-
trates, in particular, a major problem in sewer separation
projects: the dual-use house plumbing that carries both
roof and area drainage in addition to sanitary wastes.
SEWER SEPARATION
General
Sewer separation--the conversion of a combined sewer system
into two separate sanitary and storm sewer systems, in which
the sanitary sewer may be a gravity, pressure, or vacuum
system—has often been touted as the ultimate answer to
the problem of abating combined sewer overflow pollution of
receiving waters. In recent detailed studies, however, it
has been pointed out that sewer separation is not a panacea
and that, in most cases, it is a poor alternative. A review
of the history and current thinking and findings about sewer
separation is presented and discussed below.
Separation of domestic sewage and industrial wastewaters
from stormwater can be accomplished in three ways: (1) by
adding a new sanitary sewer and using the combined sewer as
145
-------
COMBINEO SE»ER OVERFLOW
MIXTURf OF tVHieiPAL »f«fttf
AMI STMWATII IISCMAIIIMI
INTI THE MCflflNG WATERS
Figure 19. Common elements of an interceptor
and transport system [6]
146
-------
a stormwater sewer; (2) by adding a new storm sewer and
using the combined sewer as a sanitary sewer; or (3) by add-
ing a "sewer within a sewer" pressure system. The latter
method, originated in 1965 by Professor Gordon M. Fair, is
designed to withdraw the domestic sewage fraction of flows
from existing plumbing systems and transmit it through a
sequence of added components. These components consist of
(1) a storage, grinding, and pumping unit within each build-
ing, (2) pressure tubing fished from the unit through the
existing building sewer line to the existing combined sewer,
and (3) pressure piping inserted in the existing combined
sewer extending to the existing interceptor sewers. From
there the sewage is conveyed by gravity to the treatment
plant. The remaining capacity in the existing combined
sewer is used to convey only stormwater to the receiving
waters. The selection of one of the three methods of sewer
separation for implementation depends on technical feasi-
bility and economic conditions [12].
The practice of sewer separation has been underway for many
years without the full benefit of recent technological in-
formation regarding the characteristics and quality of
storm sewer discharges and the possible alternatives. There
are two main reasons for reconsidering the once acceptable
practice of sewer separation. The first reason is the
change in physical conditions and quality standards from the
past which encompasses the following: (1) increases in
urban impervious areas and municipal water usage, causing
overflows of increased duration and quantity; (2) rapid in-
dustrial expansion, causing increased quantities of indus-
trial wastewaters in the overflows; (3) increasing environ-
mental concern for better water quality management; and
(4) the realization that the total amount of available
fresh water is limited and that complete reclamation of
substantial portions of the flow may be necessary in the
future. The second reason is that based on new findings
from extensive EPA-sponsored and other research during the
past decade sewer separation may be an unsuitable and un-
economical solution for combined sewer overflows. These new
findings are that (1) separated storm sewer discharges con-
tain pollutants that affect the receiving water and create
new problems [4, 35]; (2) storm sewer discharges occur more
frequently and last longer than combined sewer overflows
because combined sewer regulators prevent overflows during
minor rainfall events; and (3) many alternatives exist
that decrease the pollution at a cost of one-half or less
than that of sewer separation.
In a survey conducted by the AWPA, it was found that as of
1962 approximately one-fourth of the total U.S. population
147
-------
(50 million persons) was served in whole or in part by com-
bined sewer systems [42]. Furthermore, it was reported that
there were 14,212 overflows in the total 641 jurisdictions
surveyed; of these, 9,860 combined sewer overflows were
reported from 493 jurisdictions. Until 1967, the most com-
mon remedial method reported was sewer separation, and of
274 jurisdictions with plans for corrective facilities con-
struction, 222 indicated that some degree of sewer separa-
tion would be undertaken.
Detailed Analysis
Sewer separation will continue to be used to some degree in
the future and thus an investigation of the methods, their
advantages and disadvantages, and their costs is warranted.
There are three categories of sewer separation systems:
pressure, vacuum, and gravity.
The most comprehensive study of the pressure or "sewer with-
in a sewer" concept was published by the ASCE [12] in 1969.
The greatest disadvantage of pressure systems is generally
higher costs, as shown in a comparison of pressure and
gravity system costs in the cities of Boston, Milwaukee,
and San Francisco presented in Table 26. The ratios of pres-
sure to gravity costs are 1.4, 1.5, and 1.5, respectively.
The in-sewer pressure lines varied from 6.3 to 40.6 cm (2-1/2
to 16 inches) in diameter and pressure control valves limited
the line pressure to 2.11 kg/sq cm (30 psi) . A major portion
of the costs is the "in-house separation" which can be
as high as 82 percent of the total cost for separation
using a pressure system [12]. Besides the high costs, other
disadvantages of pressure systems are that (1) they are dif-
ficult to maintain; (2) they require complex controls; and
(3) they are dependent on electricity for operation. It is
important to realize that approximately 72 percent of all
combined sewers are less than 0.61 meters (2.0 feet) in
diameter, making it difficult to install the pressure pipe.
The advantages are that (1) as an alternative, they provide
an additional degree of latitude in sewer design, (2) there
is minimal construction interference to commerce and traffic
and (3) they are handy in low areas. '
Sewer separation of existing combined sewers has histori-
cally been accomplished by utilizing gravity systems. The
advantages of gravity sewer separation are that (1) all
sanitary sewage is treated prior to discharge; (2) treatment
plants operate more efficiently under the relatively stable
sanitary flows; (3) other alternatives are less reliable
148
-------
Table 26. ESTIMATED COSTS OF SEWER SEPARATION
IN VARIOUS CITIES
Location , [Ref . ]
Boston, Mass. [12]
Boston, Mass. [9]
Bucyrus, Ohio [49]
Chippewa Falls, Wis.
[9, 48]
Chicago, 111. [20]
Cleveland, Ohiod [18]
Des Moines, Iowa6 [8]
Milwaukee, Wis. [12]
Sandusky, Ohiod [51]
San Francisco, Calif.
Study area acreage
53
(along Summer St.)
12,000
2,340
90
240,000
13,000
1,836
157
(along Prospect Ave.)
2,205
303
Population,
density/acre
(year)
--
--
5.5 (1969)
--
--
11.7 (1968)
--
72.0 (1966)
16.5 (1969)
67.5 (1960)
Estimated cost^
Type lb Type 2C
(gravity) (gravity)
150,400
81,800
4,930 4,660
8,420
28,220
32,300
6,450
23,330
20,680
41,210
$/acre
Type 3
(pressure)
204,800
--
--
--
--
34,330
--
61,140
[12']
Seattle, Wash. [30]
Seattle, Wash. [30]
(along Laguna St.)
925
(Southwest District)
695
(East Central District)
9,740
9,950
Washington, D.C. [31]
Washington, D.C. [31]
Regional costsd'f [42]
New England
Middle Atlantic
South Atlantic
Southern
Midwest
West
National average [51]
.11,741
12,800
NAg
NA
NA
NA
NA
NA
NA
45.0 (1965) 66,250
45.0 (1965) 52,950
35,580
24,350
24,530
16,720
10,710
9,250
18,260
a. Adjusted to ENR = 2000.
b. Type 1 is constructing new sanitary sewers and using existing combined sewers for storm sewers
c. Type 2 is constructing new storm sewers and using existing combined sewers for sanitary sewers
d. Type 1 or Type 2 not identified in report.
e. Combination of Type 1 and Type 2.
f. Average costs.
g. NA = not available.
Note: acres x 0.405 = ha
149
-------
overall because of external power requirements; (4) no land
acquisition is necessary; (5) receiving water pollution
loads can be reduced by 50 percent (according to independent
studies [49, 30]); and (6) little increase in manpower is
required.
Disadvantages of gravity systems may be divided into three
categories: nonquantifiable, separation effectiveness, and
costs. Nonquantifiable disadvantages, which based on past
experience are the most important, are that (1) considerable
work is involved in in-house plumbing separation; (2) there
are business losses during construction; (3) traffic is
disrupted; (4) political and jurisdictional disputes must
be resolved; (5) extensive policing is necessary to ensure
complete and total separation; and (6) considerable time is
required for completion (e.g., in 1957 separation in
Washington, B.C., was estimated to take until sometime
after the year 2000 to complete) [24], Separation effec-
tiveness disadvantages are as follows: (1) there is only a
partial reduction of the pollutional effects of combined
sewer overflows [30]; (2) urban area stormwater runoff con-
tains significant contaminants [7, 4]; and (3) it is diffi-
cult to protect storm sewers from sanitary connections
(either authorized or unauthorized). Estimated costs for
gravity sewer separation are shown for various cities in
Table 26.
The cost disadvantages of separation, when compared to some
conceptive alternative solutions, are indicated in Table 27.
Again, the major reason for the higher costs of sewer sepa-
ration are in-house plumbing changes which can be as high as
82 percent of the total sewer separation costs [12].
Conclusions
On the basis of currently available information, it appears
that sewer separation of existing combined sewer systems is
not a practical and economical solution for combined sewer
overflow pollution abatement. Several cited alternatives
listed in Table 27 suggest other solutions, most of which
are considerably less expensive and should give better re-
sults with respect to receiving water pollution abatement.
In addition, storm sewer discharges may not be allowed at
all in the future, thus forcing collection and treatment of
all sewage and stormwater prior to discharge. In this case,
the argument for either separate or combined sewers is moot.
The choice between sewer separation and other alternatives
will be controlled by the uniqueness of each situation.
The examples cited in Table 27 leave no doubt that any alter-
native to sewer separation is the better choice. However,
150
-------
Table 27. SEWER SEPARATION VERSUS
CONCEPTUAL ALTERNATIVES
Capital costs , $
Location,[Ref. ] Separation
Alternative
Cost ratio
, b
Alternative
Boston, Mass.
[9]
Bucyrus, Ohio
[49]
Chicago, 111. 6,772,255,000 1,322,378,000
[20]
Cleveland,
Ohio [18]
Detroit, Mich. 2,859,185,000 2,859,000 1,000.0°
[33]
997,280,000 779,692,000 1.3
15,957,000 9,220,000 1.7
5.1
372,405,000 111,842,000 3.3
Seattle,
Wash. [30]
15,486,000 ^8,185,000 1.9(
Washington, 677,778,000 353,333,000 1.9
B.C. [7]
Deep tunnel storage
Lagoon system
Storage tunnels and
quarries
Offshore stabilization
ponds
Sewer monitoring and
remote control of
existing combined
sewer storage system
Computer controlled
in-sewer storage
system
Tunnels and mined
storage
a. Adjusted to ENR = 2000.
b. Ratio of separation cost to alternative cost.
c. Alternative costs are for first phases only and do not include future total
system.
d. Separation costs are only for southwest and east central Seattle, while
alternative costs are for the total combined sewer area.
local conditions elsewhere may very well dictate at least
partial separation as the best solution for combined sewer
overflow abatement. This is being done in Seattle,
Washington [30]. Seattle's partial separation program pro-
vides for removal of all street drainage from sanitary
sewers and for continued discharge of roof leaders and
foundation drains to sanitary sewers. In any event, if done
properly, the required feasibility studies of the various
possible methods for combined sewer overflow abatement may
reduce the unit cost significantly.
In any event, a thorough feasibility study of combined sewer
overflow abatement methods is required. The results of
such a study should indicate significant unit cost reduc-
tions for whatever method or combination of methods is
implemented.
151
-------
INFILTRATION/INFLOW CONTROL
A serious problem results from (1) excessive infiltration
into sewers from groundwater sources and (2) high inflow
rates into sewer systems through direct connections from
sources other than those which the sewers are intended to
serve. Inflow does not include, and is distinguished from,
infiltration. The sources and control of infiltration and
inflow are discussed in this subsection.
Sources
Infiltration is the volume of groundwater entering sewers
and building sewer connections from the soil through defec-
tive joints, broken, cracked, or eroded pipe, improper
connections, manhole walls, etc. Inflow is the volume of
any kind of water discharged into sewer lines from such
sources as roof leaders, cellar and yard drains, foundation
drains, commercial and industrial so-called "clean water"
discharges, drains from springs and swampy areas, depressed
manhole covers, cross connections, etc.
Inflow sources generally represent a deliberate connection
of a drain line to a sewerage system. These connections may
be authorized and permitted; or they may be illicit connec-
tions made for the convenience of property owners and for
the solution of on-property problems, without consideration
of their effects on public sewer systems.
The intrusion of these waters takes up flow capacity in the
sewers. Especially in the relatively small sanitary sewers,
these waters may cause flooding of street and road areas and
backflooding into properties. This flooding constitutes a
health hazard. Thus these sanitary sewers actually function
as combined sewers, and the resulting flooding becomes a
form of combined sewer overflow.
The two types of extraneous water, inflow and infiltration,
which intrude into sewers do not differ significantly in
quality, except for the pollutants unavoidably or deliber-
ately introduced into waters by commercial-industrial
operations [13]. Foundation inflow, for example, does not
vary greatly from the kind of water that infiltrates sewer
lines from groundwater sources. Basement drainage may
carry wastes and debris originating in homes, including
laundry wastewater.
Inflow Control
Correction of inflow conditions is dependent on regulatory
action on the part of city officials, rather than on public
152
-------
construction measures. If elimination of existing inflows
is deemed necessary because of adverse effects of these
flows on sewer systems, pumping stations, treatment plants,
or combined sewer regulator-overflow installations, new or
more restrictive sewer-use regulations may have to be
invoked.
The effects of inflows into sewers can be greatly reduced
by a variety of methods. Many authorities advocate the
discharge of roof water into street gutter areas--or onto
on-lot areas in the hope that it will percolate into the
soil [13]. Discharging roof or areaway drainage onto the
land or into street gutters reduces the immediate impact
on the sewer system by allowing reduction of the volume and
attenuation of the flow. The use of pervious drainage
swales and surface storage basins within urban areas allows
the stormwater to percolate into the ground.
Depressed manholes (those with vented covers in street areas
where runoff can pond over the cover) can be repaired or the
covers replaced with unvented covers.
Infiltration Control
Excessive infiltration is a serious problem in the design,
construction, operation, and maintenance of sewer systems.
Neither combined sewers nor separate sanitary sewers are
designed to accept large quantities of such infiltration
flows.
The problem of infiltration involves two basic areas of
concern: (1) prevention in new sewers by adequate design
construction, inspection, and testing practices, and '
(2) the elimination or cure of existing infiltration in old
sewers by proper survey, investigation, and corrective
measures. Control of infiltration in new sewer systems in-
volves engineering decisions and specification of the meth-
ods and materials of sewer construction, pipe, joints, and
laying procedures and techniques. Correction of existing
sewer infiltration can be accomplished by three basic
approaches: (1) replacing the defective component,
(2) sealing the existing openings, and (3) building within
the existing component.
Infiltration Control in New Construction - The types of
pipe and joints selected for use in sewer construction play
an important role in the prevention and cure of infiltration
The effectiveness of installation and conditions under which*
they function can have an equally great influence 'on the
watertightness of the resulting sewer structures and their
ability to resist excessive water entry while in service.
153
-------
Choice of sewer pipe — Improvements in pipe materials assure
the designer's ability to provide proper materials to meet
any rational infiltration allowances he wishes to specify.
The upgrading of pipe manufacture to meet rigid quality
standards and specifications has eliminated the basic ques-
tion of watertightness of pipe material. The important
issues to consider in pipe material selection are struc-
tural integrity, strength of the wastewater character, and
local soil or gradient conditions. Combinations of these
factors may make one material better suited than another or
preferable under certain special installation conditions.
In such situations, pipe materials are often chosen for
reasons other than their relative resistance to infiltration,
The cost of the pipe is usually a small part of the total
project cost. For rough estimating purposes, the cost of
installed sewer pipes (excluding manholes, laterals and
connections, appurtenances, etc.) ranges from $0.97 to $1.55
per cm diameter per linear meter ($1.25 to $2.00 per inch
diameter per linear foot).
Materials commonly used for sewer pipe construction include
(1) asbestos cement, (2) bituminous coated corrugated metal,
(3) brick, (4) cast iron or ductile iron, (5) concrete
(monolithic or plain), (6) plastic (including glass fiber
reinforced plastic, polyvinylchloride, ABS, and poly-
ethylene), (7) reinforced concrete, (8) steel, (9) vitrified
clay, and (10) aluminum. All of these materials, with the
possible exceptions of the plastics and aluminum, have been
used in sewer construction for many years.
Since sewer pipe made from the plastic materials is rela-
tively new, a brief description of the use of plastic pipes
is included below.
Solid wall plastic pipe usually refers to materials such as
polyvinylchloride (PVC), chlorinated polyvinylchloride
(CPVC) , polyvinyldichloride (PVDC), and polyethylene. These
materials are lightweight, have high tensile strength, have
excellent chemical resistance, and can be joined by solvent
welding, fusion welding, or threading. The PVC is probably
the most commonly used plastic pipe because it is stronger
and more rigid than most of the other thermoplastics; how-
ever, PVC is available only in diameters up to 30.5 cm
(12 inches).
Polyethylene pipe is finding major use as a liner for dete-
riorated existing sewer lines [26] . Several lengths of
polyethylene pipe can be joined by fusion welding into a
long, flexible tube. This tube is then pulled into the
existing sewer. When the existing house laterals have been
connected to this new pipe liner, the result is a watertight
154
-------
pipe with no joints. Polyethylene pipe is available in
diameters up to 121.9 cm (48 inches).
Acrylonitrile butadiene styrene (ABS) is most commonly used
for truss pipe. Truss pipe, deriving its name from its
cross-section configuration, was invented in 1946. It has
a good modulus of elasticity which, when under loading
contributes about equally with the soil envelope to resisting
deflection. This nonbrittle characteristic usually pre-
vents the pipe from breaking so long as continuous lateral
support is provided. Chemical weld sleeves are the most
common method of joining this pipe.
Glass fiber reinforced plastic (FRP) pipe differs from
those mentioned above in that the polymer resins are rein-
forced with glass fiber. This glass fiber reinforcement
results in an exceptionally high strength/weight ratio
FRP is available in diameters from 5.1 to 152.5 cm (2 to
60 inches) and in lengths of 3.05, 6.1, and 12.2 meters (10,
20, and 40 feet). '
Selection of sewer joints - For controlling infiltration
there is probably nothing as important as the sewer joints
No sewer system is better than its joints. A good joint
must be watertight, root penetration-tight, resistant to
the effects of soil, groundwater, and sewage, long-lasting
and flexible. &>
Until about 30 years ago, cement mortar was the standard
joint material. However, it was subject to shrinking and
cracking and tended to break loose from pipe bells and
spigots. To overcome these defects, various forms of
asphaltic compound joints were used. These were satisfac-
tory only so long as care and skill were used in their
preparation.
Several jointing methods in use today have proved to be a
vast improvement over both the cement mortar and asphaltic
compounds. These include PVC and polyurethane, compression
gaskets, and chemical weld joints.
PVC and polyurethane .joints - PVC and polyurethane joints
are most common for clay sewer pipes. Polyurethane has
been found satisfactory because of its high resilience.
Clay pipe manufacturers now use a polyester compound cast
on the spigot and into the bell to make the seal. A com-
pression gasket is used to make the seal between these
surfaces as the spigot is placed inside the bell.
Compression gasket joints - Compression gasket joints have
been used tor many years. The gaskets are most commonly
155
-------
made of natural rubber, synthetic rubber, or various other
elastomers. These joints are used on asbestos cement pipe,
cast iron pipe, concrete pipe, vitrified clay pipe, and
certain types of plastic pipes. Compression gasket joints
are most effective against infiltration while still pro-
viding for deflection of the pipe.
Chemical weld joints — Chemical weld joints are used to join
certain types of plastic pipes and glass fiber pipes. The
joints provide a watertight seal. It has been reported
that, on the basis of field tests, jointing under wet or
difficult-to-see conditions does not lend itself to precise
and careful workmanship. Thus special care is necessary
in preparing these joints in the field. More experience
with these pipes in sewer applications is necessary to
determine the longevity of this type of joint.
Heat shrinkable tubing — A new type of joint developed
recently is the heat shrinkable tubing (HST) [27] . The HST
material begins as an ordinary plastic or rubber compound
which is then extruded into sections of tubing. The tubing
is then heated and stretched in diameter but not in length.
After cooling it retains the expanded diameter. If a length
of 8-inch diameter tubing is expanded to 16 inches, it
will conform to any shape between 8 and 16 inches when
reheated. This characteristic gives the HST the ability to
form a tight fit around sewer pipe joints.
The material recommended for HST joints is a polyolefin
which has a high degree of chemical resistance and the
ability to resist scorching and burning, and is both eco-
nomical and easy to apply. To further assure HST joint
strength and resistance to internal pressure, a hot melt
adhesive is recommended as an inner surface sealant. The
adhesive material has a melting temperature close to that
of the HST and will bind the tubing and pipe materials to-
gether as the tubing cools to its final shape. Both pro-
pane torches and catalytic heaters can be used as the heat
source.
Physical properties of the HST reportedly were better than
those of currently used joint materials:
The coupling of commercial sewer pipe, both butt-
end and bell and spigot, with watertight joints
using heat shrinkable plastic tubing is feasible
and economically practical. Used in conjunction
with a hot melt adhesive it can surpass in phys-
ical and chemical strength any of the conven-
tional joints presently being used with clay,
concrete, and asbestos-cement nonpressure sewer
pipe. [27]
156
-------
Protection of concrete pipe - Corrosion exhibited inside
sewers is generally of two types: (1) that occurring above
the liquid level caused by the formation of sulfuric acid
from any hydrogen sulfide in the sewer atmosphere and
(2) that below the liquid level caused by high concentra-
tions of sulfates or other mineral and organic acids often
encountered in industrial wastes. Acidic soils can attack
the exterior of the pipe. Concrete pipe is generally the
most susceptible to these types of corrosion. The severity
of corrosion in concrete sewers has greatly increased in
recent years because of the rapid expansion of the popula-
tion, industrial growth, the use of garbage disposal units
and increasing control of inflow and infiltration into the'
sewer.
To control the corrosion of sewer pipe, especially concrete,
a method of impregnation has been developed recently The
corrosion control is achieved by impregnating the concrete
pipe with an inert or noncorrosive material that reduces
the permeability and porosity of the concrete by actually
sealing the interior surface of the concrete [28] This
prevents hydrogen sulfide from penetrating into the con-
crete and at the same time reinforces the physical and
mechanical properties of the concrete. The material used
as the impregnating liquid is a modified sulfur formulation
that was found to possess both impregnating ability and
corrosion resistance.
Other materials, such as vinyl vinylidene chloride, vinyl
acetate/acrylic, and reclaimed rubber emulsion, will resist
corrosion by actually forming a coating on the concrete
sunace.
Sections of concrete sewer pipe were given applications of
several different treatments (impregnation and various
coatings) and placed in severe corrosive sewer environments
at three sites in Texas during 1970 and 1971. Preliminary
results from these tests indicate that the various treat-
ments are functioning effectively. Additional long term
experience is necessary for complete evaluation.
Sewer design considerations - Factors to be considered dur-
ing design include allowance for infiltration, manholes and
covers, and maintainability.
Since it is virtually impossible to avoid some infiltration,
the added flow must be accommodated. Infiltration design
allowances vary tremendously from one municipality to
another [13, 41]. A distinction should be made between
infiltration design allowances and infiltration construction
allowances. The infiltration design allowance should
157
-------
include the maximum infiltration anticipated during the life
of the sewer, while the construction allowance should be
the maximum allowable infiltration at the time of
construction. The construction infiltration will increase
continuously throughout the life of the project. APWA has
recommended the establishment of a construction infiltration
allowance of 185 I/cm diameter/km/day (200 gal./inch
diameter/mile/day) or less. This is not unreasonable in
light of improvements in pipe and joint materials and con-
struction methods.
Average and peak design flows should be related to the
actual conditions for the area under design. Too often
flow criteria are taken from a standard textbook. Adequate
subsurface investigations should be undertaken to establish
conditions that may affect pipe and joint selection or
bedding requirements. Consideration should be given to the
constructability and maintainability of the sewer system.
This calls for direct communication between the designer
and maintenance personnel.
Manholes should be designed with as few construction joints
as possible. In recent years the development of custom-
made precast manholes with pipe stubs already cast in place
has reduced the problem of shearing and damage of connect-
ing pipes. The use of flexible connectors at all joints
adjacent to manholes reduces the possibility of differen-
tial settlement shearing the connecting pipes.
Manhole cover design is attracting serious attention in
view of evidence that even small perforations can produce
sizable contributions of extraneous inflow. A single
2.5-cm (1-inch) hole in a manhole top covered with 15.2 cm
(6 inches) of water may admit 0.5 I/sec (11,520 gpd) [41].
Solid sealed covers should be used for manholes in areas
subject to flooding. If solid covers are used, alternative
venting methods must be used to admit air or remove sewer
gases.
Construction considerations — The most critical factor rela-
tive to infiltration prevention is the act of construction.
The capability of currently manufactured pipes and joints
to be assembled allowing minimal infiltration must be
coupled with good workmanship and adequate inspection,
expecially at house connections.
Trenches should be made as narrow as possible but wide
enough to permit proper laying of pipe, inspection of
joints, and consolidation of backfill. Construction should
be accomplished in dry conditions. If water is encountered
158
-------
in the excavation, dewatering should be done by sump pump-
ing, use of well points, or deep wells.
Bedding and backfill material should be a combination of
both coarse and fine aggregate. The coarse aggregate pro-
vides good support (by having a wider column of support be-
neath the pipe), while the fine aggregate fills the voids
in the coarse material to retard the transmission of water
Crushed stone with a gradation of 1.9 to 0.6 cm (3/4 to
1/4 inch) is recommended. In the Gulf Coast area, where
infiltration is of particular concern, success has been
reported [46] using a bedding mixture composed of 46 5
percent coarse shell, 42.7 percent sand, 6.4 percent port-
land cement, and 4.4 percent bentonite. Bentonite is used
because it is a clay mineral with an expanding lattice
structure that enables it to swell with the addition of
water. Thus, it will act as a barrier to water flow by ex-
panding into and filling the voids in a sand/coarse aggre-
gate mixture. The selected backfill materials, preferably
the same as the bedding materials, should not be "dropped"
into the trench but should be carefully placed and com-
pacted in three separate lifts.
Backfill should be thoroughly compacted around the pipe
The backfill should be placed and packed by hand under and
around the pipe and compacted by light hand tampers.
Machines can be used for compacting as backfilling contin-
ues to the original ground surface.
Infiltration can be minimized by proper construction proce-
dures, rigid inspection of materials and methods of instal-
lation, and performance of soil and groundwater tests. In
addition, the integrity of all newly constructed sewers
should be checked by post-construction performance tests
(low-pressure air, infiltration measurement, or exfiltration
measurement) prior to acceptance.
Infiltration Control in Existing Sewers - The correction of
infiltration involves a lengthy7 systematic approach or
plan of action. The haphazard use of investigative devices
and sealing equipment is ineffectual and extremely costly.
An infiltration control plan should include the following*
steps: (1) identify the drainage system, (2) identify the
scope of the infiltration (flow measurement, rainfall simu-
lation), (3) physically survey the sewer system (soil con-
ditions, groundwater conditions, smoke tests), (4) perform
an economic and feasibility study (determine the most cost
effective locations for infiltration control), (5) clean
the sewer if necessary, (6) make a television and photo-
graphic inspection, and finally, (7) restore the sewer
system.
159
-------
Identification of the drainage system includes a review of
detailed maps of the sewer system; field checks of the line,
grade, and sizes; and identification of sections and man-
holes that are bottlenecks.
To identify the scope of the infiltration, it is necessary
to measure and record both dry- and wet-weather flows at
key manholes. Groundwater elevations should be obtained
simultaneously with sewer flow measurements.
A physical survey of the entire sewer system, or that por-
tion of major concern, where every manhole is entered and
sewers are examined visually to observe the degree and
nature of deposition, flows, pipe conditions, and manhole
condition should be made. Smoke testing may reveal infil-
tration sources only under low groundwater conditions. If
the groundwater table is above the pipe, the smoke may be
lost in the water. Soil conditions and groundwater condi-
tions should also be noted.
An economic and feasibility study is necessary to determine
the locations where the greatest amount of infiltration can
be eliminated for the least expenditure of money. In some
cases, it may be most cost effective to provide additional
treatment capacity at the sewage treatment plant for the
infiltration. Cost estimates can be developed for subse-
quent correctional stages as necessary.
Cleaning — A systematic program of sewer cleaning (1) can
restore the full hydraulic capacity and self-scouring
velocity of the sewer and its ability to convey infiltra-
tion without flooding; (2) can aid in the discovery of
trouble spots, such as areas with possible breaks, offset
joints, restrictions, and poor house taps, before any sub-
stantial damage is caused; and (3) is a necessary prerequi-
site to television and photographic inspection. It is
one of the most important and useful forms of preventive
maintenance. This type of program involves periodic
cleaning on a regular, recurring basis.
By frequent hydraulic flushing of the sewers, the interval
between mechanical cleanings of the sewer can be extended.
This will be discussed in more detail later in this section.
Equipment used in cleaning falls into three general classi-
fications: (1) redding machines, (2) bucket machines, and
(3) for small sewers, hydraulic devices. The rodding
machine, which is used most commonly, removes heavy conglom-
erations of grease and root intrusions. The bucket machine
utilizes two cables threaded between manholes. One cable
160
-------
pulls the bucket into the sewer line; the other withdraws
the bucket when it is full. It also allows the bucket to be
moved back and forth in case it becomes stuck in the sewer.
The bucket machine is effective in removing sand, gravel,
roots, and grease from large sewers. Hydraulic devices in-
clude both high velocity water jet and other hydraulically
propelled devices such as the "sewer ball." The high veloc-
ity water jet is very effective in removing sand, gravel,
and grease. Care is necessary in using high pressure hy-
draulic cleaning equipment. In sandy soil where the sewer
may be defective, creation of voids may cause collapse of
the pipe.
Inspection - The purpose of sewer inspection is to reveal
sewer restrictions; to detect cracks, broken joints, and im-
proper connections; and to locate sites of infiltration and
exfiltration. Modern inspection methods include both tele-
vision and photographic techniques. Hydraulic methods, such
as exfiltration tests, may also be used.
Television inspection is performed by pulling a closed cir-
cuit camera through the sewer. The picture is viewed on a
monitor and may be recorded on videotape for a permanent
record. Thus, the amount of infiltration can be estimated
and the condition of the sewer can be ascertained.
Television systems used for sewer inspection are constantly
being refined to reduce the size of the equipment, to in-
crease the clarity, color, and depth of image, and to re-
duce its operational cost. It is believed that the tele-
vision system will eventually be accepted as standard
equipment for sewer inspection.
Photographic inspection is similar in nature to television
inspection. It is more convenient and economical in opera-
tion, but no intimate and immediate knowledge is available.
The pictures may be on stereo slides, slides, or movie
film, and must be taken at predetermined distance intervals.
The major advantages of using these media are that sewer
problem areas can be readily located and diagnosed without
expensive excavating. Arbitrary use of these inspection
techniques without preplanning and budget review and with-
out precleaning is not recommended. Repair techniques can
be applied better and repairs can be made less expensively
in conjunction with inspection.
Restoration - On the basis of the results of an inspection
program, the engineer can decide on the appropriate methods
for correcting structural deficiencies and eliminating
infiltration. The correction alternatives include
161
-------
(1) replacement of broken sections, (2) insertion of various
types of sleeves or liners, (3) internal sealing of joints
and cracks with gels or slurries, and (4) external sealing
by soil injection grouting. Additional detailed information
is available in recent EPA reports on jointing materials
[13, 41, 27] and sealants [29, 13, 41, 25].
The method most commonly used to correct structural defects
and infiltration (in sections where major structural damage
is not present) is internal sealing with gels or slurries.
The use of a chemical blocking method to seal sewer cracks,
breaks, and bad joints is much more economical and feasible
than sewer replacement or the inadequate concrete flooding
method. With recent improvements in television and photo-
graphic inspection methods, sealing by chemical blocking
appears to be an even more encouraging method than
heretofore. Chemical blocking is accomplished by injecting
a chemical grout and catalyst into the crack or break. A
sealing packer is used to place the grout and catalyst.
The packer has inflatable elements to isolate the leak, an
air line for inflation, and two pipes for delivering the
chemical grout and catalyst to the packer. An example of a
packer is shown on Figure 20. During the repair the two
inflatable end sections isolate the leak and chemical grout
and catalyst are injected into the center section. Then
the center section is inflated to force the grout from
the annulus between the packer and the sewer wall into
the leak. When the repair is complete,the packer is de-
flated and moved to the next repair location.
The current use of acrylamid gels as chemical blocking
agents is restricted by their lack of strength and other
physical limitations. Recently, improved materials, such
as epoxy-based and polyurethane-based sealants, have been
developed [29] . These new sealants have exhibited suita-
bility even under conditions of erratic or intermittent
infiltration where acrylamid gels failed because of re-
peated dehydration. The only difficulty in applying the
new sealant materials has been that, because of the physi-
cal properties of the sealants, new application equipment
incorporating a mixing mechanism is required. The cost of
this new equipment is approximately the same as the exist-
ing equipment. Modification of existing packing equipment
to accept the new sealants has been found to be feasible.
Sewers may also be sealed by inserting sleeves or liners,
as discussed previously.
162
-------
(T) INFLATION AIR
© INFLATABLE SECTION
(a) SUPPLY PIPE - GROUT
0 SUPPLY PIPE - CATALYST
© PROTECTIVE SHOE
Figure 20. Packer - sealer with
two inflatable sections [29]
Flushing, the systematic scouring out of pipelines during
dry-weather periods, and polymer injection to increase
pipeline carrying capacities temporarily are examples of
innovative methods to counter specific transport system
deficiencies.
Flushing
In many cases the high pollutional load of combined sewer
overflows is the result of pipeline deposits being scoured
by the high-velocity storm flows. These deposits (see
Figure 82) are solids that settle out or are trapped within
the lines during antecedent dry-weather periods. Systematic
sewer flushing is designed to remove the material periodi-
cally as it accumulates and to convey it hydraulically to
the treatment facilities. Thus, peak flow rates at the down
stream point are limited to the regulator/interceptor capac-
ities prior to overflow.
The concept of sewer flushing is not new. It has been in
use for several decades primarily for small lateral sewers
163
-------
with grades too flat to be self-cleansing. However, such
applications are relatively uncommon today. Because of the
volume of flow required and the noted system limitations,
stormwater applications to date have been limited to rela-
tively small lateral sewers.
Cleansing deposited solids by flushing in combined sewer
laterals with mild slopes (0.001 to 0.008) was studied
using 30-cm (12-inch) and 46-cm (18-inch) clay sewer pipes,
each 244 meters (800 feet) long [22] . Experimental data
were then used to formulate a mathematical design model to
provide an efficient means of selecting the most economical
flushing system that would achieve a desired cleansing
efficiency within the constraints set by the engineer and
limitations of the design equations.
It was found that the cleansing efficiency of deposited
material by periodic flush waves is dependent upon flush
volume, flush discharge rate, sewer slope, sewer length,
sewer flow rate, and sewer diameter. Neither details of
the flush device inlet to the sewer nor slight irregulari-
ties in the sewer slope and alignment significantly affected
the percent cleaning efficiencies.
Using sewage instead of clean water for flushing was found
to cause a general, minor decrease in the efficiency of the
cleansing operation. The effect is relatively small and is
the result of the redeposition of solids by the trailing
edge of the flush wave.
The effects of flush wave sequencing were found to be in-
significant as long as the flush releases were made pro-
gressively from the upstream end of the sewer* Also, the
cleansing efficiencies obtained by using various combina-
tions of flush waves were found to be quite similar to
those obtained using single flushes of equivalent volumes
and similar release rates. However, both of these hypothe-
ses are based on the limited findings from tests run on
relatively short sewers. Therefore, further testing is
required to give a complete picture of the relative impor-
tance of these two factors on the overall performance of a
complete flushing system.
A prototype flush station developed during the study can be
inserted in a manhole to provide functions necessary to col-
lect sewage from the sewer, store it in a coated fabric
tank, and release the stored sewage as a flush wave upon
receipt of an external signal.
One prototype lateral flushing demonstration project was
considered for an 11-ha (27-acre) drainage area in Detroit
164
-------
but as yet has not been undertaken. Two system layouts
were considered: one designed for 61 percent daily depos-
ited solids removals and the other for 72 percent removals
depending on the number of flushing stations used [22]
Rough cost estimates were made as shown in Table 28 The
C° ij ?r *elemetlT and remote control of the flushing system
would be dependent on the degree of automation needed as
Center?1 ^^ °£ the SySt6m in relation
Polymer Injection
Polymer gelled slurry injections into sewage have resulted
in significant hydraulic friction reductions; hence a tem-
porary increase (up to 144 percent reported [4011 in line-
capacities [40, 21, 36]. This increase is significant
in stormwater applications because the sewer surcharges
Table 28. ESTIMATED FLUSHING COSTS FOR
DEMONSTRATION PROJECT21 [22]
DETROIT, MICHIGAN
Alternate -^ 2
Number of flush stations per lateral 2 4
Area per lateral, acres 9 g
Daily solids removal, percent 61 72
Installed cost of fabric flush tanks $6,380 $12,900
Cost of telemetry and controls --b _.b
Monthly power cost $ 2.24 $ 4.69
Monthly maintenance cost $ 115 $ 229
Capital cost per acre $ 708 $ 1,430
Monthly maintenance and power cost
Per acre $13.00 $ 26.05
a. ENR = 2000.
b. Not estimated.
Note: Acre x 0.405 = ha.
165
-------
associated with excess wet-weather flows are generally of
short duration; thus, a marginally inadequate line can be
bolstered by polymer injections at critical periods. In
effect, this increases the overall treatment efficiency by
allowing more of the flow to reach the treatment plant,
while flooding from sewer surcharges is decreased.
The polymers tested in Richardson, Texas, included Polyox
Coagulant-701, Polyox WSR-301, and Separan AP-30 [40]. The
latter showed the greatest resistance to shear degradation
(which may be important in very long channels) but was the
least effective hydraulically. Tests conducted indicated
that the polymers and nonsolvents were not detrimental
to bacteria growth and therefore would not disrupt the
biological treatment of sewage in wastewater treatment
plants. Tests conducted on algae in a polymer environment
indicated that the polymers have no toxic effects and only
nominal nutrient effects. Fish bioassays indicated that in
a polymer slurry concentration of 500 mg/1, some fish deaths
resulted but that, in practice, concentrations above 250
mg/1 would provide no additional flow benefits. It was re-
ported that polymer concentrations of between 35 and 100
mg/1 decreased flow resistance sufficiently to eliminate
surcharges of more than 1.8 meters (6 feet) [40].
The Dallas Water Utilities District, Dallas, Texas has con-
structed a prototype polymer injection station (see
Figure 21) for relief of surcharge-caused overflows at 15
points along a 2,440-meter (8,000-foot) stretch of the
Bachman Creek sewer [36]. During storms, the infiltration
ratio approaches 8 to 1. The sanitary sewer is 46 cm
(18 inches) in diameter for the first 1,220 meters (4,000
feet) and then joins another 46-cm (18-inch) diameter
sewer and continues on. The Dallas polymer injection
station was built as a semiportable unit so that it can be
removed and installed at other locations needing an in-
terim solution once a permanent solution has been imple-
mented at Bachman Creek.
The polymer injection unit is enclosed by a 1.3-cm (1/2-inch)
steel sheet, 3.1 meters (10 feet) in diameter by approxi-
mately 7.9 meters (26 feet) in height. The upper half pro-
vides storage for 6,364 kg (14,000 Ib) of dry polymer and
also contains dehumidification equipment. The lower half
contains a polymer transfer blower, a polymer hopper and
agitator for dry feeding, a volumetric feeder and eductor,
and appurtenances. The unit is entirely self-contained
with only external power and water hookup necessary for the
unit to be in operation.
166
-------
Figure 21. Polymer injection
station for sanitary sewer (Dallas)
injection facility (b) Station interior showing eductor and polymer
foreground (d)' tll^^ll! .How?n^oca'r """ manh°'e *° "ft °f "'««"'"
injection site C3 '"" ° metering vault and polymer
167
-------
The unit is set up for fully automatic operation and may be
started by any of three external level sensors located
458 meters (1,500 feet) upstream, at the injection site, and
458 meters (1,500 feet) downstream.
Several polymers were tested, and Polyox WSR-301 was chosen
to be used when the Bachman Creek unit becomes operational.
The polymer is expected to reduce the surcharge by 6.1 meters
(20 feet) over the first 1,220-meter (4,000-foot) length.
The maximum equipmental feed rate is expected to be 2.3
kg/min (5 Ib/min). The actual polymer feed rate will be
flow paced by the liquid level in the sewer to maintain a
polymer concentration of about 150 ppm in the sanitary sewer.
The unit is capable of supplying a dosage of 500 to 600 mg/1
if needed. It is expected that the unit will be in operation
about five times per year and that surcharge reduction will
be complete in approximately 7 minutes after start of polymer
injection (travel time in the affected reach of sewer).
The actual construction cost for the unit, including instal-
lation of the site, was $146,000 (ENR 2000). The unit was
scheduled to be operable by mid-1973 with operational per-
formance and data available one year thereafter. Maintenance
is expected to be limited to a site visit and unit exercise
every 2 months.
REGULATORS
Historically, combined sewer regulators represent an attempt
to intercept all dry-weather flows, yet automatically pro-
vide for the bypass of wet-weather flows beyond available
treatment/interceptor capacity. Initially, this was accom-
plished by constructing a low dam (weir) across the combined
sewer downstream from a vertical or horizontal orifice as
shown on Figure 22. Flows dropping through the orifices
were collected by the interceptor and conveyed to the treat-
ment facility (see Figure 19). The dams were designed to
divert up to approximately 3 times the average dry-weather
flow to the interceptor with the excess overflowing to the
receiving water. Eventually more sophisticated mechanical
regulators were developed in an attempt to improve control
over the diverted volumes. No specific consideration was
given to quality control.
Recent research, however, has resulted in several regula-
tors that appear capable of providing both quality and
quantity control via induced hydraulic flow patterns that
tend to separate and concentrate the solids from the main
flow [10, 50, 15]. Other new devices promise excellent
quantity control without troublesome sophisticated design.
168
-------
Figure 22. Typical early type of regulator
169
-------
Conventional Designs
Conventional regulators can be subdivided into three major
groups: (1) static, (2) semiautomatic dynamic, and
(3) automatic dynamic. The grouping reflects the general
trend toward the increasing degree of control and sophisti-
cation and, of course, the capital and operation and main-
tenance costs. Conventional regulator design, use,
advantages, and disadvantages are well covered in the
literature [10, 11, 32].
Static Regulators — Static regulators can be defined as
fixed-position devices allowing little or no adjustment
after construction.
Static regulators consist of horizontal or vertical fixed
orifices, manually operated vertical gates, leaping and
side-spill weirs and dams, and self-priming siphons. Typi-
cal static regulators are shown on Figure 23. With the ex-
ception of the vertical gate, which does not often move,
they have no moving parts. Thus, they provide only minimal
control, and they are least expensive to build, less costly
to operate, and somewhat less troublesome to maintain.
In view of the increasingly more stringent receiving water
discharge limitations and the rising need of providing storm
water capacity in treatment plants, it is expected that the
use of conventional static regulators will decline. System
control, to utilize maximum capacity in the interceptor,
requires flexibility virtually nonexistent with static
regulators. Maintenance, with the exception of the vertical
gate, is mostly limited to removal of debris blocking the
regulator opening.
Semiautomatic Dynamic Regulators — Semiautomatic dynamic
regulators can oe defined as those which are adjustable over
a limited range of diverted flow and contain moving parts
but are not adaptable to remote control.
The family of semiautomatic dynamic (having moving parts)
regulators consists of float-operated gates, mechanical
tipping gates, and cylindrical gates. Typical semiautomatic
dynamic regulators are shown on Figure 24. All require
separate chambers to allow access for adjustment and
maintenance. As a rule this group is more expensive to
construct and to maintain than static regulators. They are
more susceptible to malfunction from debris interfering
with the moving parts and are subject to failure due to the
corrosive environment. However, better flow control is
provided because they respond automatically to combined
sewer and interceptor flow variations. The adjustment of
170
-------
EPTOR
f 1
*
,
i
t
4 ,
r
Vn
ORIFICE CHAMBER
DIVERSION
CASTING^
N\
\
U^-ORIFICE PLATE
r, * « * •
r\~
4 *
"
».
1
If
\
'. k
1 vJ
DIVERSION
CHAMBER
• STORM IATER
_i OUTLET
\f "^\ ''
AV^rv' ^*
^^ r r'.
t
' •»
e
•
«*.*'' * ' * 4,4,-, f
.* * * 1 1 • * t •> . <
(a) FIXED-ORIFICE DIVERSION AND CONTROL STRUCTURE [32]
_ PLATFORM, I //////////.
(b) TYPICAL MANUALLY OPERATED GATE REGULATOR
PHILADELPHIA [| l]
(c) ADJUSTABLE LEAPING »EIR AT SEATTLE, WASH. [32]
LONGITUDINAL SECTION
(d) HEIR OVERFLOW STRUCTURE [32]
Figure 23. Typical static sewer regulate
rs
171
-------
(a) AUTOMATIC SEIER REGULATOR [32]
(BROVN »KO BROVN TTPE »).
STOP LINK
(b) TIPPING GATE REGULATOR [ll]
USED BY ALLEGHENY COUNTY
SEWAGE AUTHORITY
6»TE CHDIBER
(C) CYLINDRICAL GATE REGULATOR [ifl]
Figure 24. Typical semiautomatic
dynamic sewer regulators
172
-------
these regulators is limited and strictly manual; thus thev
are unadaptable for remote control. ' y
The cylindrical gate is a relatively new device consisting
of a horizontal circular orifice over which a cylindrical
gate is hung. The gate is counterbalanced on an articulated
themcioslne of'thf ""?• 'T1 ^ the gate ^a^ber controls
. cl°sing °f the orifice by the cylindrical gate without
are^t*^ ?r "% 6Xt6rnal ener^ sources. fheL gates
are suitable for flow ranges from less than 283 I/sec
CIO cfs) to S,660 I/sec (200 cfs) [10].
Automatic Dynamic Regulators - Automatic dynamic regulators
consist of cylinder-operated (hydraulic or pneumatic) and
motor-operated gates as shown on Figure 25. The distinction
arTf, [?m ** • Pre5?ding regulators is that thesl regulators
JUStble and are readil Ad
arf, * • r
controUL JUSt?ble and- are readily Adaptable to rmoely
controlled operation. Because they are more sophisticated
in design they are more expensive to build and to maintain
regu&f S?e ^^ Problem? as with all conventional
regulators, along with corrosion and jamming.
Cylinder-operated gates - This regulator device consists of
a weir perpendicular to flow constructed across the combined
aevert^r?rVWh,iCh ^erts peak ^/-weather ffow through
a vertical, fixed orifice with a variable opening to an
interceptor. A cylinder-operated sluice gate varies the
orifice opening. The cylinder-operator rfspondf to an
hnh^T °r downstream level sensor (usually a float or air
bubbler) or to remote signals that override the sensor!
The amount of combined sewage diverted to the interceptor
is controlled by the gate with the excess flow continuing
on to the receiving waters. This type of regulator is
cess°:fCU3yi/sS!d f2T a«om"ic regulation & flows in ex-
cess of 113 I/sec (4 cfs) . Such gates are considered very
effective when operating correctly but they have been per-
sistently subject to hydraulic malfunctions. P
The cylinders may be operated by water, air, or oil
pressure The size of sluice gates operated by cylinders
using water pressure is usually limited to 0.8 to iVl sq m
sure of 1S? £}/ Such cylinders operate at a minimum pr^s
??™ft A I Jig/S^ Cm (2S psl) whlle the maximum pressure is
i^ed by the Clty water Pressure. Care must be taken to
?H? ??-Cr°KS "nnections with the city water supply by
installing backflow prevention devices. Y
Preferred oil cylinder pressures of 53 kg/sq cm (750 psil
do not restrict the sluice gate size but do require a sepa-
rate structure for housing the oil pumping equipment P
173
-------
LOAT ft
FLOAT WELL
NTERCEPTOR
PLAN VIEW
CYLINDER - OPERATED GATE REGULATOR
PHILADELPHIA [ll]
Figure 25. Typical automatic dynamic
sewer regulator
174
-------
Pressures of 6.3 to 14 kg/sq cm (90 to 200 psi) are gener-
ally used in air-actuated cylinders. These also require
separate structures for air compressor equipment. In juris-
dictions using both oil- and air-actuated cylinders, the oil
type is preferred [10, 11]. '
Motor-operated gates - The application and operation of this
type of regulator is similar to the cylinder-operated gate
except that a motor is used in place of the cylinder
bpecial precautions and structures are required to protect
the motor and other electrical equipment. Motor-operated
regulators are not generally considered economical for
flows less than 113 I/sec (4 cfs) .
Improved Regulator Designs
Recent emphasis has resulted in the development of several
new and innovative regulators both in the United States and
in Europe [10]. Those showing the greatest promise are
undergoing prototype testing. Regulators included in this
group are fluidic devices, swirl concentrators, broad-
crested inflatable fabric dams (see Section IX Storage)
and automatic slide gates and tide gates. Improved regula-
tors developed in England include the vortex regulator! high
side-spill weir, stilling pond regulator, and the spiral
flow regulator. The spiral flow regulator is being de-
veloped for American practice.
Fluidic Regulator - Fluidic devices of two general types
act?nnP?*^ilty ?ePendin£ on the type of fluid flow inter-
action that takes place within them. These categories are
(1) wall attachment and (2) vortex amplifier, of which the
former forms the largest group [10, 11]. in the wall attach-
ment devices a high velocity jet emitted between two walls
attaches itself to one wall, attracted there by a lower
pressure area next to that wall caused by air entrapment
at the opposite wall. A typical installation is shown on
Figure 26. The City of Philadelphia, Pennsylvania, is now
planning a full-scale demonstration of this type of fluidic
regulator [19]. The vortex amplifier is in the develop-
ment stage of small-scale modeling and as such is a long
way from full-scale demonstration [39] . Flows in excess
of the hydraulic design capacity cannot pass through these
regulators and instead flow over the unit into the over-
flow channel.
vn ~ The V°rtex r*«ulator (in England called
vortex overflow or rotary vortex overflow) consists of
a circular channel in which rotary motion of the sewage
175
-------
CONTROL PORT
ELEVATED EXIT WEIR
COMBINED
SEWER
OUTFALL
COMBINED FLOW
WEIR
COMMUNICATION LINES
FIXED AREA ORIFICE
SIMPLE LEVEL SENSOR
AIR SLOT
INTERCEPTOR FLOW
Figure 26. Schematic arrangement
of a fluidic sewer regulator [10]
is induced by the kinetic energy of the sewage entering
the tank (see Figure 27). Flow to the treatment plant is
deflected, and discharges through a pipe at the bottom near
the center of the channel. Excess flow in storm periods
discharges over a circular weir around the center of the
tank and is conveyed to receiving waters. The rotary motion
causes the sewage to follow a long path through the channel
thus setting up secondary flow patterns which create an
interface between the fluid sludge mass and the clear liquid
The flow containing the concentrated solids is directed to
the interceptor. Using synthetic sewage in model studies at
Bristol, England, suspended solids removal efficiencies of
up to 98 percent were reported [47] . Another series of
experiments elsewhere on a model vortex regulator using raw
sewage indicated poor performance in removing screenable
solids under certain conditions [1]. This lack of overflow
176
-------
COMBINED SEWER
STORM SEWER
NLET
BAFFLE
BRANCH INTERCEPTOR
TO TREATMENT PLANT
PLAN
BAFFLE
SECTION A
Figure 27. Vortex regulator [11]
quality enhancement was due to the free, unimpeded vortex
flow field employed. It was later found that flow dampening
by a deflector prevented solids being swept into the over-
flow by the violent vortex action.
Swirl Regulator/Concentrator - The swirl regulator/
concentrator, after thorough hydraulic modeling, shows out-
standing potential for providing both quality and quantity
control. A full-scale installation is planned in Lancaster,
Pennsylvania. The swirl regulator/concentrator is similar
to, and is an outgrowth of, the vortex regulator. APWA
studies, working with much larger flows in minimum-sized
chambers showed that a vortex flow pattern must be avoided
[50]. A different hydraulic condition is developed which
enhances solids removals. Review of the literature points
out that the main difference between the English vortex
177
-------
regulator and the swirl regulator/concentrator is the flow
field pattern. Another major difference is that larger
flow rates can be handled in the prototype swirl regulator/
concentrator (at Lancaster, Pennsylvania, the estimated in-
crease is 4 to 6 times greater) than in the equivalent size
vortex regulator.
A hydraulic laboratory model was used to determine geometric
configuration and settleable solids removal efficiencies.
Figure 28 shows the hydraulic model in action. Note the
solids separation and concentration toward the underflow
pipe to the treatment plant.
As a result of both mathematical and hydraulic modeling,
the performance of the prototype has been predicted. Based
upon a peak storm flow to peak dry-weather flow ratio of
55 to 1, 90 percent of the solids (grit particles with a
specific gravity of 2.65, having a diameter greater than
0.3 mm and settleable solids with a specific gravity of 1.2,
having a diameter larger than 1.0 mm) are concentrated into
3 percent of the flow [50, 15]. Hydraulic testing indicates
that removal efficiency increases as the peak storm flow to
peak dry-weather flow decreases. The recommended configura-
tion for the swirl regulator/concentrator is shown on
Figure 29.
The foul-sewer channel in the bottom of the swirl concentra-
tor is sized for peak dry-weather flow. During wet-weather
flows the concentrated settleable solids are carried out
the foul-sewer into an interceptor.
There are no moving parts so maintenance and adjustment re-
quirements are minimal. Fine tuning control is provided via
a separate chamber with a cylinder gate on the "foul sewer"
outlet to the interceptor. Remote control, although not
readily adaptable, could be accomplished by providing a
larger-than-necessary foul sewer (also diminishes the
chances of clogging) and throttling this line with a re-
motely controlled gate.
Spiral Flow Regulator - The spiral flow regulator is based
on the concept of using the secondary helical motion im-
parted to fluids at bends in conduits to concentrate the
settleable solids in the flow. A bend with a total angle
between 60 and 90 degrees is employed. Hydraulic model
studies of this device, carried out at the University of
Surrey, England [44], indicated that this is a feasible
178
-------
SttlRL
tfHiP*-
TEST
«j&
1
1
Figure 28. Solids separation action in the
swirl concentrator hydraulic model
179
-------
B ^
FIMTAILES
DEFLECTOR
FLO! DEFLECTOR
A-A
FLOI DEFLECTOR
B -J
B-B
PLAN
SECTIONS
Figure 290 Recommended configuration
for swi rl concentrator [50]
means of separating solids from the overflow. The simplest
form of the regulator is shown on Figure 30.
The heavily polluted sewage is drawn to the inner wall. It
then passes to a semicircular channel situated at a lower
level leading to the treatment plant. The proportion of
the concentrated discharge will depend on the particular
design. The overflow passes over a side weir for discharge
to the receiving waters. Surface debris collects at the
end o.f the chamber and passes over a short flume to join
the sewer conveying the flow to the treatment plant.
The authors of the model study report that even the sim-
plest application of the spiral flow separator will produce
an inexpensive regulator that will be superior to many
existing types. They also stated that further research is
necessary to define the variables, the limits of applica-
tions, and the actual limitations of the spiral flow
180
-------
CROSS CONNECTION FOR OVERFLO
CONTROL PIPE
-FLUME INVERT
E
TO P
PIPE FOR NORMAL FLOW
PROFILE ALONG CENTER LINE
OVERFLOW WEIRS
WITH DIP PLATES
CHANNEL FOR NORMA
FLOW & HEAVY SOL IBS
PIPE FOR
NORMAL FLO
& SOLIDS
SECTION A - A
FLUME
CONTROL PIPE FOR
OVERFLOW CHAMBER
SECTION B - B
LUME FOR
FLOATING
ATERIAL
CONTROL
PIPE
-^
TO INTERCEPTOR-^
Figure 30. Spiral flow regulator adapted from [11]
181
-------
regulator [44]. A prototype regulator has been success-
fully evaluated at Nantwich, England. A third generation
device is being developed for American practice.
Stilling Pond Regulator - The stilling pond regulator, as
used in England, is a short length of widened channel from
which the settled solids are discharged to the interceptor
[1]. Flow to the interceptor is controlled by the discharge
pipe which is sized so that it will be surcharged during
wet-weather flows. Its discharge will depend on the sewage
level in the regulator. Excess flows during storms dis-
charge over a transverse weir and are conveyed to the re-
ceiving waters. The use of the stilling pond may provide
time for the solids to settle out when the first flush of
stormwater arrives at the regulator and before discharge
over the weir begins.
This type of regulator is considered suitable for overflows
up to 85 I/sec (30 cfs). If the stilling pond is to be suc-
cessful in separating solids, it is suggested that not less
than a 3-minute retention be provided at the maximum rate
of flow [34].
High Side-Spill Weirs — Unsatisfactory experience with
side-spill weirs in England has led to the development of a
high side-spill weir, referred to there as the high double
side-weir overflow. These weirs are made shorter and higher
than would be required for the normal side-spill weir. The
rate of flow to the treatment plant may be controlled by use
of a throttle pipe or a float-controlled mechanical gate.
The ratio of screenings in the overflow to screenings in the
sewage passed on to treatment was 0.5, the lowest of the
four types investigated in England. This device has the
best general performance when compared to the English vortex
and stilling pond regulators and the low side-overflow
weir [1].
Tide Gates - Tide gates, backwater gates, or flap gates are
used to protect the interceptors and collector sewers from
high water levels in receiving waters and are considered
a regulating appurtenance when used for this purpose.
Tide gates are intended to open and permit discharge at the
outfall when the flow line in the sewer system regulator
chamber produces a small differential head on the upstream
face of the gate. Some types of gates are sufficiently
heavy to close automatically, ahead of any water level rise
in the receiving body. With careful installation and bal-
ancing, coupled with an effective preventive maintenance
program, the ability of the gate to open during overflow
182
-------
periods is not impaired because of the additional weight.
Tide gates should be checked regularly to be sure that the
hinge arms, pivot points, and seats are in good condition.
The gates should also be free of trash, timber, or other
obstructions which lock the shutter in the partly open posi
tion, allowing inflow.
Tide gates are available in a wide variety of sizes. They
may be rectangular, square, or circular in shape depending
on the requirement.
Electrode Potential - Another possible form of automatic
regulation is to make use of electrode potential measure-
ments of the combined or storm sewer overflows and to modu-
late the discharge accordingly. Studies made on predomi-
nantly stale domestic sewage in the laboratory showed, upon
analysis of experimental results, a high degree of correla-
tion between the electrode potential of the sewage and
its strength. Linear correlation coefficients between
electrode potential and the various sewage parameters meas-
ured were found to be as follows [43]:
Linear correlation
Parameter coefficient
BOD5 0.873
COD 0.852
Sulfides 0.896
Total phosphorus 0.893
Nitrates 0.807
Chlorides 0.225
It was demonstrated that the potential decreases as the
sulfide concentration increases, except when a small amount
of DO is present exerting an attenuating effect. Thus, the
quality of combined sewer overflows could be controlled by
monitoring the electrode potential and releasing only that
flow which would not damage the oxygen balance in the re-
ceiving waters. Storm flows with quality below the satis-
factory range would be shunted to storage until the
potential rises to the acceptable range.
Evaluation and Selection
The process of selecting a regulator can no longer be based
solely on economics. Of increasing importance is quality
control of the overflow as well as quantity control, other-
wise known as the "two Q's" concept.
183
-------
Regulators and their appurtenant facilities
should be recognized as devices which have the
dual responsibility of controlling both quantity
and quality of overflow to receiving waters, in
the interest of more effective pollution
control. [50]
As mentioned previously, new regulator devices have been
developed that provide both quantity and quality control.
These include electrode potential along with the swirl
regulator/concentrator, spiral flow regulator, vortex regu-
lator, and high side-spill weir. Thus, in the future, the
choice of a regulator must be based on several factors in-
cluding: (1) quantity control, (2) quality control,
(3) economics, (4) reliability, (5) ease of maintenance, and
(6) the desired mode of operation (automatic or
semiautomatic).
Regulator Costs — Selected installed construction costs are
shown in Table 29. These costs are to be used for order-
of-magnitude reference only because of the wide variance
of construction problems, unit sizes, location, number
of units per installation, and special appurtenances.
The cost of maintaining sewer regulators as reported in a
recent national survey also vary widely [10] . In most
cases, the reported expenditures are probably not adequate
to maintain the regulators in completely satisfactory
condition. The annual cost per regulator required to con-
duct a minimal maintenance program is listed in Table 29.
REMOTE MONITORING AND CONTROL
One alternative to the tremendous cost and disruption caused
by sewer separation is to upgrade existing combined sewer
systems by installing effective regulators, level sensors,
tide gates, rain gage networks, sewage and receiving water
quality monitors, overflow detectors, and flowmeters and
then apply computerized collection system control. Such
system controls are being developed and applied in several
U.S. cities. The concepts of control systems have been in-
troduced in Section VI. As applied to collection system
control, they are intended to assist a dispatcher (super-
visor) in routing and staving combined sewer flows to make
the most effective use of interceptor and line capacities.
As the components become more advanced and operating experi-
ence grows, system control offers the key to total inte-
grated system management and optimization.
184
-------
Table 29. INSTALLED CONSTRUCTION COSTS AND ANNUAL
OPERATION AND MAINTENANCE COSTS
OF REGULATORS21
Type of regulator
Broad-crested inflatable fabric dam [45, 37]
Cylinder operated gate [30, 11, 10]
Cylindrical gate [11]
Float operated gate [11, 10]
Fluidic device [14]
High side-spill weir
Horizontal fixed orifice (drop inlets) [11, 10]
Internal self-priming siphon [10]
Leaping weir [11, 10]
Manually operated vertical gate [11, 10]
Motor operated gate [30, 11]
Polymer injection [40, 36]
Side-spill weir [11, 10]
Spiral flow separator
Stilling pond
Swirl concentrator [38]
Tipping gate [11, 10]
Vertical fixed orifice [11, 10]
Vortex
Installed
construction
cost, $
4,200-7,200
13,000-590,000
44,000-166,000
140,000-260,000
33,000-83,000
NA
1,800-3,600
NA
2,800-33,000
8,500-282,000
72,000-446,000
12,900-146,000
1,100-25,000
NA
NA
124,000
49,000-418,000
17,000-37,000
NA
Annual cost
per regulator,
1,500
1,600-1,800
NAb
1,500-1,600
NA
NA
1,600-2,100
800-1,100
1,000-1,200
1,200-1,500
NA
NA
600-700
NA
NA
NA
1,500-1,800
800-1,100
NA
a. ENR = 2000.
b. NA = not available.
185
-------
System Components and Operations
The components of a remote monitoring and control system
can be classified as either intelligence, central proces-
sing, or control.
The intelligence system is used to sense and report the
minute-to-minute system status and raw data for predictions.
Examples include flow levels, quantities, and (in some
cases) characteristics at significant locations throughout
the system; current treatment rates, pumping rates, and
gate (regulator) positions; rainfall intensities; tide
levels; and receiving water quality.
Quality observations and comparisons may assist in deter-
mining where necessary overflows can be discharged with the
least impact. The central processing system is used to com-
pile, record, and display the data. Also, on the basis of
prerecorded data and programming, the processer (computer)
may convert, for example, flow levels and gate positions
into estimates of volumes in storage, overflowing, and in-
tercepted and may compute and display remaining available
capacities to store, intercept, treat, or bypass additional
flows.
The control system provides the means of manipulating the
system to maximum advantage. The devices include remotely
operated gates, valves, inflatable dams, regulators, and
pumps. Reactions to actuated controls and changed condi-
tions (i.e., increased rainfall, pump failure, and blocked
gate), of course, are sensed by the intelligence system,
thus reinitiating the cycle.
Representative elements of a typical system are shown on
Figure 31.
Because of the frequency and repetitiveness of the sensing
and the short time span for decision-making, computers must
form the basis of the control system. The complexity of
the hydrology and hydraulics of combined systems also dic-
tates the need for extensive preprogramming to determine
cause-effect relationships accurately and to assist in eval-
uating alternative courses of action. To be most effective,
real-time operational control must be a part of an overall
management scheme included in what is sometimes called a
"systems approach."
System Control
Before storm flow collection system control can be imple-
mented, the direction, intensity, and duration of the storm
186
-------
(e)
(f)
Figure 31. Elements of a remote control system (Seattle)
(a) Central control and status board (b) Telemetry signal transmitter (c)
Automated sewage sampler (d) Emergency power generator (e) Motor-operators
for regulator gates (f) Tipping-bucket rain gage at remote regulator station
(g) Pneumatic cylinder operator for regulator gate
187
-------
should be known so that runoff quantities may be anticipated
Thus, the rain gage network forms an integral part of the
system. Once the storm starts affecting the collection sys-
tem, the flow quantity and movements must be known for
decision-making, control implementation, and checking out
the system response. The advantages of knowing whether or
not an overflow is occurring are obvious, but consider the
added advantage of knowing at the same time that the feeder
line is only half full and/or that the interceptor/treatment
works are operating at less than full capacity. By initi-
ating controls, say closing a gate, the control supervisor
can force the feeder line to store flows until its capacity
is approached, or can increase diversion to the interceptor,
or both. If he guesses wrong, the next system scan affords
him the opportunity to revise his strategy accordingly.
Thus, system control or management converts the combined
sewer system from an essentially static system to a dynamic
system where the elements can be manipulated or operated as
changing conditions dictate.
The degree of automatic control or computer intelligence
level varies among the different cities. For example, in
Cincinnati, monitoring to detect unusual or unnecessary
overflows is applied and has been evaluated as being
successful [5]. In Minneapolis-St. Paul, the Metropolitan
Sewer Board is utilizing a central computer that receives
telemetered data from rain gages, river level monitors,
sewer flow and level sensors, and mechanical gate diversion
points to assist a dispatcher in routing stormwater flows
to make effective use of in-line sewer storage capacity [2].
The use of rain gages, level sensors, overflow detectors,
and a central computer to control pump stations and selected
regulating gates is underway in Detroit [3], The Munici-
pality of Metropolitan Seattle (METRO) is incorporating the
main features of the above projects plus additional water
quality monitoring functions [30] . The City and County of
San Francisco have embarked on the initial phase of a system
control project for which the ultimate goal is complete
hands-off computerized automatic control. They are cur-
rently collecting data on rainfall and combined sewer flows
to aid in the formulation of a system control scheme. More
details of the San Francisco system are described in
Section XIII under Master Plan Examples. The main differ-
ence between the San Francisco and Seattle projects, besides
size, is hands-off versus hands-on automatic supervisory
control [16] .
As an example of a complex "systems approach" to collection
system control, various aspects of the Seattle master plan
are discussed in detail below.
188
-------
Detailed Example
The METRO approach to collection system control in Seattle,
known as the Computer Augmented Treatment and Disposal
(CATAD) system, is a computer-directed system for maximum
utilization of available storage in the trunk and inter-
ceptor sewers to reduce or completely eliminate combined
sewer overflows. The objectives and background of this pro-
gram are discussed in Section XIII under Master Plan
Examples.
Background Data Collection - To develop the computer control
program necessary to meet the objectives of the CATAD system,
environmental background data were necessary to establish
the baseline conditions. Toward this end, two major studies
were undertaken dealing with weather analysis and water
quality analysis.
Weather analysis - A series of weather analyses were begun
in late 1969 to determine what types of meteorological
quantities would provide the best information for predicting
storm intensities and actual wet-weather flows in the com-
bined sewer system. The study was divided into two main
sections. The first was based on precipitation data only.
The second section considered wind speed and direction data
in addition to precipitation.
The end conclusion was that the combination of wind direc-
tion and rain gaging from remote stations would provide
advance information to enable the CATAD program to deter-
mine optimum flow regulation and storage levels within the
sewage collection system.
Rather than duplicate much meteorological work being accom-
plished by the weather bureau, METRO reduced the weather-
sensing portion of the CATAD program to the three following
procedures:
1. Long-range precipitation forecasts would be
entered into the computer program by obtaining the
chance of rain prediction issued by the weather
bureau at 6-hour intervals.
2. Medium-range rain gage data would be provided by
rain gages at METRO stations located to the
farthest north or south extent of the collection
system. The first amount of rain detected by
these gages would signal the immediate release of
all stored sewage and draw down of those trunks
and interceptors.
189
-------
3. Short-term weather prediction would be obtained by
rain gages located throughout the METRO drainage
area.
Water quality studies — Since 1963, METRO has been engaged
in a comprehensive water quality monitoring program through-
out the entire metropolitan drainage area. Upon receipt
of the CATAD demonstration grant in 1967, additional spe-
cialized water quality monitoring studies were added to the
existing program to concentrate on certain areas that con-
tribute to combined sewer overflows.
The objectives of the demonstration grant water quality
studies were twofold. First, new water quality studies
were begun or old programs modified to show how receiving
water quality and other dynamic system parameters have
changed during the periods of expansion, interception, regu-
lation, and separation. Second, a base level for various
parameters was to be established to be used as a tool for
measuring the results of the CATAD demonstration project.
The studies have been divided into two general areas re-
lated to the collection system itself and the receiving
waters adjacent to the municipality. Weather and other
pertinent environmental factors are correlated with data
from the two main study categories.
Overflow sampling was divided into three categories: physi-
cal and chemical sampling, bacteriological sampling, and
overflow volume computation.
Examples of a typical sewer sampling station and receiving
water sampling and monitoring station are shown on
Figure 15 (a, b, c).
System Operation — The CATAD system controls comprise a
computer-based central facility for automatic control of
remote regulator and pumping stations. The control center
is located at the METRO office building with satellite
terminals at the West Point and Renton treatment plants.
The principal features of the control center include a
computer, its associated peripheral equipment, an operators
console, map display, and logging and events printers [23],
Remote monitoring and control units have been provided for
36 remote pumping and regulator stations. Twenty-four re-
mote control units have been installed at pumping and
regulator stations on the trunk and interceptor sewers
leading to the West Point sewage treatment plant and nine
remote control units have been installed at pumping stations
along the interceptor sewers transporting primarily sanitary
190
-------
sewage to the Renton treatment plant. One control unit of
each collection system is located at the treatment plant
influent pumping station. Three additional control units
are to be installed at new pumping and regulator installa-
tions during the next several years.
Precipitation is monitored at six remote regulator stations
strategically selected to be representative of the sub-
basins within the drainage area served by METRO. Precipi-
tation data are telemetered to the central computer for
processing along with other regulator station data. The
items monitored at each regulator station, as shown in
Table 30, include the sewage level in both the trunk and
interceptor sewers, the maximum level set points in the
trunk and interceptor, the tide level, the outfall and regu-
lator gate positions, the overflow rate, the diversion rate
to the interceptor, the trunk flow rate, the interceptor
upstream flow rate, the stored flow, the interceptor down-
stream flow rate, the unused storage volume, and the explo-
sion hazard. The items monitored at each pump station, as
shown in Table 31, include the wet well liquid level, the
liquid level set point, the station and force main discharge,
the inflow rate, the explosion hazard, the storage rate the
unused storage, and the speed of each pump. Water quality
parameters monitored in the receiving waters are temperature
conductivity, dissolved oxygen, pH, and solar radiation as '
shown in Table 32.
Regulators and pumping stations may be operated under three
modes of control [23]:
1. Local Automatic Control. Each station is operated
independently by controllers within the station in
response to signals from local sensing devices.
2. Remote Supervisory Control. Stations are remotely
controlled by operator-indicated commands from the
central terminal via the CATAD system computer.
3. Remote Automatic Control. Stations are operated
from the central terminal under program control by
the CATAD system computer.
Since the trunk sewers are sized to carry storm flows
there is a large volume available for storage during dry-
weather periods. The use of this storage capacity during
storms is effected by controlling the quantity of flow
diverted from combined trunks into the interceptor.
191
-------
Table 30. TYPICAL CATAD REGULATOR STATION MONITORING
HOURLY LOG
11/14/72 1000 W POINT SYS HOURLY LOG REGULATOR STATION
TRKLVL TRKSET TIDE OUTPOS OVRFLO TRKFLO STOFLO UNUSTO
INTLVL
LOG DENNY RS
100.76
0.00
LUN DENNY RS
KING RS
CONN RS
LANDER RS
2 HANFORD
100.02
94.73
105.23
97.70
101.24
97.46
102.27
98.91
RS
100.97
98.81
INTSET
96.56
109.88 105.89
96
102
106
101
106
102
105
102
.56
105.34
.40
.37 109.38
.35
.01 106.06
.75
.23 104.81
.75
REGPOS
0.9
100.9
-0
100
-0
99
-0
99
-0
99
0
100
.2
.5
.1
.3
.2
.8
.1
.4
.0
.0
DIVFLO
0.0
3.6
0.0
10.6
0.0
3.4
0.0
3.1
0.0
6.3
0.0
5.7
UPSFLO
3.7
10
32
3
1
3
31
6
28
6
20
.6
.6
.5
.4
.1
.6
.3
.6
.4
.8
DNSFLO
0.1
47
0
4
0
34
0
35
0
26
.0
.1
.9
.0
.7
.0
.1
.7
.6
EXPHAZ
0.14
0.03
-0.5
0.31
0.57
2.15
BRANDON RS
MICHIGAN
CHELAN RS
HARBOR RS
W MICH RS
8TH SOUTH
DEXTER RS
L CITY RS
1 HANFORD
102.37
98.96
RS
101.50
100.30
101.56
100.53
108.38
108.08
116.46
107.41
RS
100.49
98.12
136.56
134.28
150.36
114.33
RS
101.61
95.40
105
100
105
101
107
103
109
108
99
144
137
157
108
.93 105.59
.40
.69 105.35
.65
.98 105.61
.21
106.09
.13
.37
105.76
.58
.34
.75
.06
.05
-0
102
-0
102
0
100
-0
99
0
99
0
100
36
100
-0
.1
.9
.4
.9
.1
.3
.2
.5
.0
.9
.3
.4
.3
.1
.7
0.0
0.0
0.0
0.0
0.0
4.4
0.0
0.9
0.0
0.7
0.0
2.8
0.0
4.2
0.0
13.4
0.0
0
14
0
12
4
0
0
3
2
4
13
.0
.0
.0
.0
.4
.9
.7
.1
.8
.1
.4
14
12
2
0
3
2
-0
4
38
.0
.0
.7
.9
.8
.2
.1
.2
.9
1.09
-3.6
192
-------
Table 31. TYPICAL CATAD PUMP STATION MONITORING
HOURLY LOG
11/14/72 1000 W POINT SYS HOURLY LOG - PUMP STATION
LEVEL SETPNT STADIS INFLOW STORAT PUMP 1 PUMP 2 PUMP 3
W POINT PS FMDIS EXPHAZ UNUSTO PUMP 4 PUMP 5 PUMP 6
100.53
INTERBAY PS
93.32 93.29
DUWAMISH PS
E MARG PS
W MARG PS
30TH NE PS
90.92
94.31 94.29
94.30 93.77
118.83 119.05
BELVOIR PS
112.49 112.49
MATTHEWS PS
KENMORE PS
87.97 87.94
92.41
103.0
48.9
0.6
18.0
-0.4
11.2
4.5
2.2
-1.1
2.0
13.4
-1.2
2.1
151
0
0
201
0
0
490
0
0
0
0
327
0
355
0
0
0
510
0
0
0
48.9
0
487
0
3.4
373
0
826
-702.7
Table 32. TYPICAL CATAD WATER QUALITY MONITORING
HOURLY LOG
11/25/72 2100 WATER QUALITY HOURLY LOG
SURFACE
BOTTOM
RENTON JUNC.
E. MARGINAL
16TH AVE. S.
SPOKANE ST.
KENT
TEMP C2400 C24000 D.O. PH S.R.I.
TEMP C48000 D.O.
046.1 0120 00880 09.56 05.85
045.6 0386 00000 09.88 06.88 0.06
047.9 0000 22320 07.42 07.40
049.8
049.9
43360 06.98
043.7 0000 24200 08.38 07.40
41160 07.48
045.4 0095 00040 11.54 07.00
193
-------
The design of the METRO interceptor system provides a posi-
tive means for controlling these bypassed flows. A regula-
tor station (Figure 32) at each major trunk sewer controls
both the diversion of combined sewage into the interceptor
and the overflow from the trunk (sewage in excess of the
capacity of the interceptor) . The volume of flow diverted
to the interceptor is automatically controlled by modulating
the regulator gate position in response to changes in the
level of sewage in the interceptor. As the level in the
interceptor rises above a preset maximum, the regulator gate
closes to reduce the volume of diverted flow and maintain
the preset level. Storm flow in excess of the diverted flow
is stored in the trunk sewer and the level of the sewage in
the trunk commences to rise. When the level rises above a
preset maximum, the outfall gate will open automatically to
discharge the excess storm flow and modulate to maintain
the preset maximum level in the trunk.
Accomplishments — The most demonstrative method of pointing
out accomplishments is to show the results of interception
of an actual storm. Two days of CATAD printouts were ob-
tained from METRO, one set for the storm flow that occurred
on November 25, 1972, and the second set for the dry-weather
flow on November 14, 1972. The dry-weather flow data were
used to establish an approximate dry-weather flow base for
comparison purposes. The particular regulator station
analyzed is the Denny-Lake Union (identified as LUN DENNY RS
in the CATAD printouts). A sample storm log is shown in
Table 33. The data included in this log are the rainfall
occurring and the maximum rainfall rate during the hour,
the maximum overflow rate and the overflow volume occurring
during the hour, and the total overflow volume from the
start of the overflow. A 16-hour period from 0700 hours
to 2300 hours was used for the comparison. From the data,
hydrographs were generated which yielded a dry-weather
flow volume of 140,540 cu m (37.13 mil gal.) and a wet-
weather flow volume of 204,650 cu m (54.07 mil gal.). The
potential overflow volume then is the difference between
the two or 64,120 cu m (16.94 mil gal.). The amount of
actual overflow from the station allowed by the CATAD system
was 11,660 cu m (3.08 mil gal.). Thus the effective storm
runoff containment for this particular storm and regulator
station was approximately 82 percent.
Several improvements have been observed in Elliott Bay fol-
lowing the August 1970 interception and regulation of 12
major combined sewer overflows which are that reductions in
coliform levels range from 63 to 98 percent and that moni-
toring indicates an improvement of between 2 and 3 mg/1 of
dissolved oxygen in the bay.
194
-------
Table 33. TYPICAL CATAD STORM LOG
11/25/72 2100 STORM LOG
FLTIME RAINFL MAXRAT OVRMAX OVRVOL OVRTOT
LOG DENNY RS 0.0 0.00 0.02
0.0 0.00 0.02
LUN DENNY RS
0.00 0.00 7.7 0.02 3.08
KING RS
0.0 0.00 0.02
E MARG PS
0.01 0.06
MATTHEWS PS
0.01 0.06
KENMORE PS
0.01 0.06
RENTON PS
0.01 0.06
Other improvements have been observed in the Duwamish River
The dissolved oxygen levels have increased nearly 200 per-
cent from an average of 2.5 mg/1 to 4.5 mg/1 but the im-
provement cannot definitely be attributed to the major
combined sewer overflow's interception until an additional
summer's data can be compared. Improved trawl fish catches
indicate larger populations of certain fish species, in-
cluding English sole and Chinook salmon, in the lower por-
tion of the river following interception. Decreases in
the ammonia-nitrogen concentrations at certain stations
along the river can be attributed to improved nitrification
techniques being utilized at the Renton secondary treatment
plant discharging into the river upstream of these stations,
C°s*s 7 The CATAD project cost was $3.1 million, of which
$1.4 million was a demonstration grant from the EPA Storm
and Combined Sewer Technology Program. These costs do
not include regulators or pumping station costs. Computer
monitoring and control to each station costs about $15,000.
Construction costs for new regulators with motor-driven
gates and local controls average $250,000. The CATAD opera
tion costs about $200,000 per year including salaries, sup-
plies, amortized capital expenses, and maintenance costs
[17].
195
-------
Section IX
STORAGE
Storage is, perhaps, the most cost effective method avail-
able for reducing pollution resulting from combined sewer
overflows and to improve management of urban stormwater
runoff. As such, it is the best documented abatement meas-
ure in present practice. Storage, with the resulting sedi-
mentation that occurs, can also be thought of as a treatment
process.
Storage facilities possess many of the favorable attributes
desired in combined sewer overflow treatment: (1) they are
basically simple in design and operation; (2) they respond
without difficulty to intermittent and random storm behavior;
(3) they are relatively unaffected by flow and quality
changes; and (4) they are capable of providing flow equali-
zation and, in the case of tunnels, transmission.
Frequently they can be operated in concert with regional
dry-weather flow treatment plants for benefits during both
dry- and wet-weather conditions. Finally, storage facili-
ties are relatively fail-safe and adapt well to stage
construction. Drawbacks of such facilities are related pri-
marily to their large size (real estate requirements), cost,
and visual impact. Also, access to treatment plants or
processes for dewatering, washdown, and solids disposal is
required.
Storage facilities presently in operation have been sized
on the basis of one or more of several possible criteria.
The facilities should: (1) provide a specified detention
time for runoff from a storm of a given duration or return
frequency; (2) contain a given volume of runoff from the
tributary area, such as the first 1.27 cm (1/2 inch) of
runoff; (3) contain the runoff from a given volume of rain,
such as the runoff from 1.27 cm (1/2 inch) of rain; or
(4) contain a specified volume. Because storage facilities
are generally designed to also function as sedimentation
and/or disinfection tanks, a major advantage is the SS reduc-
tion of any overflows from the storage units. Particular
196
-------
design concerns are flow handling, washdown and solids
removal, and protection against odors and hazardous gases.
Air can be used to resuspend the settled solids prior to
dewatering and for odor and explosive gas control.
As with treatment systems, automated intelligence (data col-
lection and analysis) and control can play a significant
part in the management and optimal use of storage systems.
The objective of the control system is to optimize the con-
tainment and treatment of combined sewer overflows with
actions dependent upon the storm pattern, treatment and
storage availability, and projected storm and system
behavior. When overflows to receiving waters are necessary,
quality monitoring coupled with system controls will permit
the releases to occur in the least damaging manner. The
smaller the storage volume available (in tunnels, sewers,
and tanks) and the more variable the rainfall pattern, the
more critical the monitoring and control system becomes.
TYPES OF STORAGE FACILITIES
Storage facilities may be constructed in-line or off-line;
they may be open or closed; they may be constructed inland
(upstream) or on the shoreline; they may have auxiliary
functions, such as flood protection (sewer relief) and flow
transmission; and they may be used for hazardous spill con-
tainment during dry weather.
In-Line
Because combined sewers are designed to carry maximum flows
occurring, say, once in 5 years (50 to 100 times the average
dry-weather flow), during most storms there will be consid-
erable unused volume within the major conduits. In-line
storage is provided by damming, gating, or otherwise re-
stricting flow passage just downstream from the regulator
diversions to create additional storage by backing up the
water in these upstream lines. Essential to effective utili-
zation of this concept are sewers with flat sewer grades in
the vicinity of the interceptor, high interceptor capacity,
and extensive control and monitoring networks. To be safe,
the restriction must be easily and automatically removed
from the flow stream when critical flow levels are approached
or exceeded. Such systems have been successfully implemented
in Seattle, Minneapolis-St. Paul, and Detroit [14, 7, 15],
In-line storage has been utilized in other processes by
setting retention basin weirs sufficiently high to back flows
into the trunk sewer system before overflows occur. This is
done to ensure maximum utilization of both interceptor and
treatment plant capacities.
197
-------
Seattle, Washington — First operational in late 1971, the
system presently has 10 fully equipped regulator stations.
such as the one shown on Figure 32, with 3 more under design,
All stations are monitored and are designed so that they may
be operated by a supervisor from a central control console.
Fully automated control will be attempted in 1973. The
estimated maximum safe storage in the trunklines and inter-
ceptors is 121.1 Ml (32 mil gal.), or roughly equivalent to
0.13 cm (0.05 inches) of direct runoff from the combined
sewer and partially separated sewer areas. Interceptor
capacity is generally 3 times the estimated year 2000 dry-
weather flow. Under supervisory operation, overflows have
been reduced in volume by approximately 52 percent.
Minneapolis-St. Paul, Minnesota — This system, operational
since April 1969,is quite similar to that in Seattle, ex-
cept that inflatable Fabridams are used in place of the
motor-operated outfall gates, as also shown on Figure 32.
Fifteen Fabridams, operated by low pressure air, are
located in the major trunks, which are 1.52 to 3.66 meters
(5 to 12 feet) in diameter, immediately downstream of the
regulator gates. Normally, they are kept in a fully in-
flated condition forming a dam to approximately mid-height
of the conduits. When storm flows are sufficiently large
so as to threaten to surcharge the trunk sewers, as indi-
cated by the flow depth monitoring, the Fabridams may be
deflated remotely from the control center. On the trunks
where they are installed, the total overflow volume reduc-
tion has been estimated to range from 35 to 70 percent,
depending on the nature of the storm event [7]. Based upon
a comparison of pre- and post-project conditions, the number
of overflows was reduced 58 percent (from 281 to 117) and
the total overflow duration was reduced 88 percent (from
1,183 hours to 147 hours) from April 1969 to May 1970. A
major accomplishment of the plan has been the almost total
capture of the contaminated spring thaw runoff.
Detroit, Michigan — The Detroit Metropolitan Water Service
(DMWS) has installed the nucleus of a sewer monitoring and
remote control system for controlling combined sewer over-
flows from many small storms to the Detroit and Rouge rivers
[1]. This system includes telemeter-connected rain gages,
sewer level sensors, overflow detectors, a central computer,
a central data logger, and a central operating console for
pumping stations and selected regulating gates. The cost of
the system was slightly over $2.7 million. This system has
enabled DMWS to apply such pollution control techniques as
storm flow anticipation, first flush interception, selective
retention, and selective overflowing.
198
-------
LOCAL CONTROL STRUCTURE
TRUNK WATER LEVEL
CONTROLS OUTFALL GATE
EXISTING TRUNK SEWER-
DEPTH OF WATER IN
INTERCEPTOR CONTROLS
REGULATOR GATE
METRO INTERCEPTOR
REGULATOR GATE
(NORMALLY OPEN)
OUTFALL GATE
(NORMALLY CLOSED)
TIDE LEVEL SENSOR
MAY PREVENT GATE
OPENING
r^ S
RECEIVING WATER
(a) SEATTLE, WASHINGTON [M]
UNDERGROUND EQUIPMENT VAULT
POWER OPERATED GATE
INFLATAILE DAM
TO RIVER
TO INTERCEPTOR
(b) IINNEAPOLIS-ST.PAUL, MINNESOTA [7]
Figure 32. Typical regulator stations
of in-line storage systems
199
-------
The operator, upon receiving advance information on storms
from a remote rain gage, increases the treatment plant pump-
ing rate. This lowers the surcharged interceptor gradient
and allows for greater interceptor storage capacity and
conveyance. This practice has enabled DMWS to contain and
treat many intense spot storms entirely, in addition to many
scattered citywide rains.
Off-Line
Typical off-line storage devices can range from lagoons [18],
to huge primary settling tank-like structures [10, 2], to
underground silos [8] , to underwater bags [4], to void space
storage, to deep tunnels [5], and mine labyrinths. In
almost all cases, feedback of the retained flows to the
sanitary system for ultimate disposal is proposed or
practiced. The underground and offshore storage has been
proposed to meet the severe land area and premium cost
constraints.
Chippewa Falls, Wisconsin — A 36.45-ha (90-acre) combined
sewer area of this Wisconsin community has been served by
a 10.6-M1 (2.8-mil gal.) open storage lagoon since 1969 [18].
The storage volume is equivalent to 2.92 cm (1.15 inches) of
runoff from the tributary area. A plan of the retention
basin is shown on Figure 33. In the two-year period 1969-
1970, the lagoon was 93.7 percent effective in capturing
overflow volumes. During this period, the combined sewer
overflows from 59 of 62 storms were totally contained by the
basin. Flow storage in the basin up to 12 hours caused no
adverse odor problems. The basin was paved with 5.08 cm
(2 inches) of asphalt, and the most effective cleaning of
solids was through the use of conventional street sweepers.
The basin is dewatered to an existing activated sludge
plant after storms with no adverse effect on the biological
treatment process. Secondary clarifier capacity, however,
had to be doubled to avoid excessive loss of solids during
sustained high flows.
Akron, Ohio — An underground 2.7-M1 (0.7-mil gal.) capacity
storage facility has been constructed in Akron, Ohio (see
Figure 34), utilizing the concept of void space storage [17].
The basin is trapezoidal in cross section (3:1 side slopes)
with top dimensions of 61 meters (200 feet) by 61 meters
(200 feet) and a usable depth of 3.4 meters (11 feet). It
serves a 76-ha (188.5 acre) combined sewered area. The rock
fill material completely filling the basin in which the com-
bined sewage is to be temporarily stored is washed gravel,
graded from 6.3 to 8.9 cm (2-1/2 to 3-1/2 inches) in
diameter. The effective void space is approximately 33 per-
cent of the total volume. The fill is completely enclosed
200
-------
RIVER
SANITARY PUMPING
:
y=
V ST.
STATION ^
DIVERSION STRUCTURE
VEHICLE
GATE
STOIiiATER
PUMPING
STATION
CHAIN LINK
FENCE
Figure 33. Detention basin plan
Chippewa FalIs [18]
201
-------
DIVERSION \t COMBINED SEWER
LEAPING WEIR FOR
SANITARY FLOW T9
SANITARY SEWER
VERTICAL BAR SCREEN AND
S0LIDS HOLDING CHAMBER
TUBE
CLARIFIERS
OVERFLOW TO STORM SEWER
STORAGE BEB
PUMPED TO SANITARY SEWER
ME?
SLBBIE BRAIIFF P0MPED TB SANITARY SEWER
Figure 34.
Schematic of detention facilities,
Akron, Ohio [17]
202
-------
in a watertight plastic liner (30 mil sheeting) and covered
with 1 meter (3.28 feet) of earth.
Storm-flushed material flows from the combined sewer system
into a clarification (roughing) chamber where it is chlori-
nated and a large percentage of the solids are removed. The
chlorinated wastewater flows over a weir into the holding
basin where it is stored until it can be transported to the
treatment plant. The purpose of the rock fill is to save
cost and to maintain the ground-level surface area available
for other uses. The installation is expected to begin oper-
ating in late 1973. The system's susceptibility to clogging
and its treatment efficiency will be evaluated.
Jamaica Bay, New York City, New York - This large, covered
concrete storage basin, completed in 1972, intercepts com-
bined overflows from a 1,318.7-ha (3,256-acre) service area.
The total storage capacity of 87.1 Ml (23 mil gal., includ-
ing 10 mil gal. in basins and 13 mil gal. through backup in
the trunk system), is equivalent to 0.66 cm (0.26 inches) of
direct runoff. The basin is designed to retain fully the
runoff from up to 50 percent of the summer storm events, and
to provide primary treatment and chlorination for larger
storms (20 minutes detention minimum for intensities up to
1.3 cm/hr (0.50 in./hr) which covers 98 percent of all
storm time). About three-fourths of the retained volume in
the tanks and sewers are drained by gravity to a dry-weather
flow treatment plant following each storm. The remaining
volume and settled solids then are pumped to the treatment
plant.
Appurtenances include traveling bridge hydraulic sludge
collectors, mechanically cleaned bar racks, centrifugal
type grit separators, and sodium hypochlorite storage and
feed facilities. The unit is completely covered and has
6 parallel basins, each 15.25 meters wide by 145.2 meters
long by 3.5 meters deep (50 feet by 476 feet by 11.5 feet),
as shown on Figure 35. Forced air ventilation is provided
to prevent odor problems and explosion hazards. Twenty
overflows, all with the equivalent of primary treatment,
are anticipated to occur on the average each summer season.
Humboldt Avenue, Milwaukee, Wisconsin — This single covered
concrete storage tank, completed in 1969, serves a 230.9-ha
(570-acre) section of the combined sewer portion of
Milwaukee [3]. Its storage capacity of 15.1 Ml (4 mil gal.)
is equivalent to the runoff from 1.27 cm (0.5 inch) of rain-
fall over the tributary area. However, it has been contain-
ing the runoff from 2.0 to 2.3 cm (0.8 to 0.9 inches) of
rainfall. A schematic of the facility is shown on Figure 36.
Photographs of the facility are shown on Figure 37.
203
-------
(d)
Figure 35. Jamaica Bay (Spring Creek auxiliary water
pollution control plant) retention basin
(a) Exterior view of facility at discharge to receiving water (b) Traveling bridge
collector with drop pipe assemblies (c) Closeup of bridge showing typical drop
pipe and header for directing high pressure sprays on floor (d) One of six 50-ft
wide bays, looking at inlet ports
204
-------
COMBINED SEWER
OVERFLOW
TANK
DEWATERING
LINE
INTERCEPTOR
EFFLUENT DISCHARGE
RIVER
Figure 36.
Schematic of Humboldt Avenue detention
and chlorination faciIi ty,
MiIwaukee, Wi sconsin
205
-------
(d)
Figure 37. Humboldt Ave. (Milwaukee) retention basin
(a) Exterior view of the predominantly buried facility on the Milwaukee River
(b) Operations building housing ch I orination, screening, and pumping equipment
(c) Looking ovfir tanks from operations building (d) Walking above tanks
(note tank area did not require fencing
206
-------
The facility consists of a mechanically cleaned 3.8 cm (1.5
inch) bar screen, a single detention tank 22.9 meters wide
by 128.1 meters long by 6.1 meters deep (75 feet by 420 feet
by 20 feet), a dewatering pump station, and chlorination
facilities. The detention tank serves as a settling basin,
thereby providing primary treatment for combined sewer over-
flows from large storms. Chlorine injection facilities are
included at both the headworks, for odor control, and at the
midpoint of the tank, for disinfection during overflows.
Following storms, seven mechanical mixers within the tank
resuspend the settled solids as the tank, contents are pumped
back to the interceptor for treatment at a dry-weather plant
During the first year of tank operation, the mixers per-
formed satisfactorily so that it was not necessary to clean
the detention tank manually.
Boston, Massachusetts - The Cottage Farm Stormwater Treat-
ment Station was completed in May 1971. A maximum retention
capacity of 4.9 Ml (1.3 mil gal.) is provided in the 6 par-
allel channels, each 8.2 meters wide by 32.9 meters long by
3.1 meters deep (27 feet by 108 feet by 10 feet) [6]. The
facility is designed to provide primary treatment and
chlorination. It has a 10-minute detention time at a flow
rate of 10,200 I/sec (233 mgd), the difference between the
capacities of the incoming trunk and outgoing interceptors.
The facility consists of bar screens and coarse screens,
a pumping station with a 10.7-meter (35-foot) lift, hypo-
chlorination facilities, detention basins, and an outfall.
A schematic of the facility is shown on Figure 38 and photo-
graphs are shown on Figure 39. Additional photographs are
shown on Figure 80.
Screenings from the 8.9-cm (3-1/2-inch) clear opening
bar screens and the 1.27-cm (1/2-inch) clear opening coarse
screens are flushed directly to the downstream interceptor.
Flow is pumped to the detention tanks, each of which is
gated so it can be isolated. Hypochlorite solution is fed
to the pump discharges. Overflows from the detention tanks
pass through horizontal hinged screens with 0.51-cm (0.2-
inch) fine mesh openings, before discharge to the outfall.
These screens are used to trap additional suspended matter
and carryover. Approximately 2.5 cm (1 inch) of grit and
sludge accumulates in the tank and wet well during a storm.
Fire hoses and a sloped and troughed floor system are used
to aid the flushing of settled solids from the tank during
cleanup following each storm (approximately 8 hours are
required). These solids are discharged to the downstream
interceptor.
207
-------
COMBINED
SEWAGE FLOW
(10-12xOWF)
WET
WELLy
BAR COARSE
SCREENS7 SCREENS
HYPOCHLORITE
PUMPS vTUBESr--------S
1 FEED
CONTROL
GATES
OVERFLOW IN
EXCESS OF
INTERCEPTOR
CAPACITY
INTERCEPTOR
TO DRY WEATHER
TREATMENT
(4-5xDWF)
I—
TANK DRAIN (TYP)
CHLORINE RESIDUAL
ANALYZER
OUTFALL
TO RIVER
1 f CHLORINE
KbSIUUAL
ANALYZER *
^SOLIDS RETURN LOCATION ^
AND DRAIN CONTROL GATES-
-------
(b)
Figure 39. Cottage Farm (Boston)
detention and chlorination facility
(a) Exterior view from Charles River showing buried tanks, drain gate operators,and
control building (b) View from effluent trough of one of six parallel bays; hori-
zontal fine screens are seen in foreground
209
-------
Chicago, Illinois - Chicago has pioneered in the development
of abandoned mine storage, deep vertical drop shafts [11],
and deep tunnels in hard rock [5] for the interception, con-
veyance, and temporary storage of combined sewer overflows.
Three construction contracts partially funded by the EPA
are presently in progress, representing a total investment
in excess of $50 million (ENR 2000) providing over 17.7 km
(11 miles) of tunnel and appurtenant drop shafts and pumping.
Typical examples are shown on Figure 40. Under the recently
adopted master plan for the 97,200-ha (240,000-acre) Greater
Chicago service area, the following are to be accomplished:
(1) treatment of all wet-weather flows at a dry-weather flow
facility sized for 1.5 average dry-weather flow maximum rate;
(2) interception of all existing wet-weather outfalls by a
deep conveyance tunnel system; and (3) storage of all inter-
cepted flows above treatment capacity at one large and two
auxiliary reservoirs until they are absorbed by the dry-
weather flow treatment facilities operating at nearly a
constant maximum rate [9]. The cost for this plan has been
estimated to be $2,873 million as opposed to $5,521 million
for complete sewer separation, and separation would not meet
the proposed water quality or flood control objectives.
The major reservoir would be a quarry 100.7 meters deep by
152.5 to 366.0 meters wide by 4.02 km long (330 feet by
500 to 1,200 feet by 2.5 miles). The combined sewer over-
flows retained in the reservoir would be aerated continu-
ously, and accumulated solids would be removed periodically
by dredging. Most storms would be dewatered in 2 to 10 days;
the largest, in 50 days. The reservoir would be dewatered
to a dry-weather treatment plant at a rate of 19.8 cu m/sec
(450 mgd) or 0.5 times average dry-weather flow. The total
storage provided would be equivalent to 7.98 cm (3.14 inches)
of runoff, 70,309,500 cu m (57,000 acre-ft), or 9 percent of
the annual average rainfall. Based upon this plan, the fre-
quency of overflows to the Chicago River was estimated to be
4 times in 21 years.
The tunnels, a total of 193 km (120 miles) ranging from 3.05
to 12.81 meters (10 to 42 feet) in diameter, would be con-
structed in dolomite rock 45.8 to 88.5 meters (150 to 290
feet) below the surface. Drop shafts would be placed at
approximately 0.80-km (1/2-mile) intervals, and a forced
ventilation system providing one air change per hour during
dewatering would be operated to control odors and gases.
The tunnels would be constructed on self-cleansing gradients
with additional provisions for flushing by introducing river
water into the tunnel.
Washington, D. C. - At Washington, D. C., a pilot plant was
built and tested to assess the feasibility of treatment and
210
-------
(e)
Figure 40. Deep tunnel concepts and construction (Chicago)
(a) Conceptual
them to tunnel
(d) Conceptual
cep tor or river
drawing of drop shaft to intercept street level flows and conduct
(b) Drop shaft under construction (c) Tunnel under construction
drawing of terminal pumping station to raise flows back to inter-
level (e) "Mole" excavated tunnel under construction
211
-------
underwater storage of combined sewer overflows from a
12.15-ha (30-acre) drainage area [4], The plant consisted
of a treatment facility (grit removal, bar rack, and comminu-
tion), underwater storage tanks, and the associated instru-
mentation, pumping, and control systems. Two 0.38-Ml (0.1-
mil gal.) tanks were anchored underwater in the Anacostia
River. The tanks were standard pillow tanks made of nylon-
reinforced synthetic rubber.
Portions of the overflow from a combined sewer overflow line
were diverted to the pilot plant. The flow entered the grit
chamber of the pilot plant, then passed through a bar screen
followed by a comminution before entering the underwater
storage tanks. Material removed in the grit chambers was
returned to the interceptor. After the storm subsided and
during nonpeak hours, liquid was pumped from the tanks into
the interceptor for transport to the dry-weather treatment
plant. To prevent settlement of solids in the storage tanks,
compressed air was forced into the tanks to agitate any
settled sludge and to enable pumping out of all the contents
of the storage tanks to the interceptor.
Of the numerous operational problems encountered during this
project, the major one was clogging of the effluent port by
the flexible tank top membrane during dewatering.
Sandusky, Ohio - A 0.76-M1 (0.2-mil gal.) capacity under-
water storage facility was constructed and tested in
Sandusky, Ohio. The facility consisted of three basic sys-
tem components, the underwater storage tank with its associ-
ated piping and controls, a connecting structure to the
existing outfall, and a control building to house instru-
mentation and pump systems [19]. Two 0.38-M1 (0.1-mil gal.)
collapsible tanks serving a 6.0-ha (14.9-acre) area were
anchored underwater in Lake Erie. The tanks consist of a
steel pipe frame in the form of a modified octagon with a
nylon-reinforced synthetic rubber flexible membrane top and
bottom. A concrete pad was poured to fit the bottom con-
tours of both storage tanks. The tank top conforms to the
bottom contours when empty and the top rises upon filling.
Combined sewer overflow passes through a bar screen in a
connecting chamber to remove all trash from the overflow
before passing to the influent pipes to the tanks. A diver-
sion weir allows control of the filling of one tank or the
other. At high flow rates, both tanks fill simultaneously.
After tank capacity is reached, the flow backs up in the
connection chamber and passes out a safety overflow.
Combined sewer overflow reaching the underwater storage
tanks passes through a sedimentation control chamber over
212
-------
the inlet port of each tank. Most suspended material
settles out within this chamber. This chamber also supports
the top membrane during dewatering to prevent closing of the
tank effluent port. The combined sewer overflow is stored
until interceptor capacity is available to transport the
stored volume to the treatment plant. Emptying the storage
tanks by pumping can be accomplished in about one hour per
tank. Each tank is emptied separately. A tank flushing
system is operated in conjunction with the pumping of each
tank. In tests to date, the operation of the tanks has
proven successful.
Cost Data
Costs of storage structures are highly dependent upon the
location of construction; land, foundations, groundwater,
and aesthetics are primary considerations. The costs re-
ported in Table 34, adjusted to ENR 2000, are presented only
as preliminary guides.
213
-------
Table 34. SUMMARY OF STORAGE COSTS
FOR VARIOUS CITIESa
Location
Seattle, Wash. [14]
Control and monitoring system
Automated regulator stations
Minneapolis-St. Paul, Minn. [7]
Chippewa Falls, Wis. [18]
Storage
Treatment
Akron, Ohio [17]
Oak Lawn, 111. [16]
Melvina Ditch Detention
Reservoir
Jamaica Bay, New York City,
N.Y. [10,12]
Basin
Sewer
Humboldt Avenue, Milwaukee,
Wis. [3, 13]
Boston, Mass. [6]
Cottage Farm Stormwater
Treatment Station
Chicago, 111. [9]
Reservoirs
Collection, tunnel, and
pumping0
Reservoirs and tunnels
Treatment
Sandusky, Ohio [19]
Washington, D.C. [4]
Storage,
mil gal. Capital cost, $
3,500,000
3,900,000
32.0 7,400,000
3,000,000
2.8 744,000
186,000
2.8 950,000
0.7 441,000
53.7 1,388,000
10.0 21,200,000
13.0
23.0 21,200,000
4.0 2,010,000
1.3 6,200,000
2,736.0 568,000,000
2,834.0 755,000
5,570.0 1,323,000,000
1,550,000,0-00
5,570.0 2,873,000,000
0.2 535,000
0.2 883,000
Cost per
acre,
$/acre
—
5,550
8,260
2,070
10,330
2,340
540
6,530
6,530
3,560
2,370
3,150
5,500
6,460
11,960
36,000
29,430
Storage
cost ,
$/gal.
0.23
--
0.26
0.26
0.62
0.03
2.12
0.92
0.50
4.74b
0.21
0.27
0.24
..
0.24
2.67
4.41
Annual
operation and
maintenance
cost, $
250,000
--
2,500
7,200
9,700
23,300
'--
50,000
65,000
--
8,700,000
6,380
3,340
a. ENR = 2000.
b. Includes pumping station, chlorination facilities, and outfall.
c. Includes 193.1 km (120 miles) of tunnels.
Note: $/acre x 2.47 = $/hectare; $/gal. x 0.264 = $/l; mil gal. x 3.785 = Ml
214
-------
Section X
PHYSICAL TREATMENT WITH AND WITHOUT CHEMICAL ADDITION
Physical treatment operations are a means of treatment in
which the application of physical forces predominate.
Typical examples include screening, sedimentation, flotation,
and filtration. Physical treatment operations may or may
not include the addition of small concentrations of
chemicals.
Physical treatment operations in many ways are well suited
to combined sewer overflow pollution abatement. Suspended
solids characteristically found in large quantities in com-
bined sewage are especially amenable to removal by physical
treatment. Most physical treatment operations are easily
automated and operate at high efficiencies over a wide range
of flows. And, most important, the facilities can stand
idle for long periods of time without affecting treatment
efficiencies.
The physical treatment operations being used for combined
sewer overflow treatment range from commonly used, well
understood processes such as sedimentation and filtration to
new ones such as the swirl concentrator/regulator and fine
screens. The different operations are discussed here in the
following order: sedimentation, dissolved air flotation,
screens, and filtration. The swirl concentrator/regulator
was discussed previously in Section VIII.
SEDIMENTATION
The time-honored method for removing SS is sedimentation.
Removal efficiencies in primary clarifiers usually are
approximately 30 percent for BODs and 60 percent for SS [3,
20, 25]. The types of solids removed in primary clarifiers
are somewhat comparable to those in combined sewer overflows.
Both are represented by discrete, nonflocculating particle
settling. Removal efficiencies for BODs and SS at combined
sewer overflow storage/sedimentation facilities have been
reported to approximate those for primary clarifiers.
-------
Although sedimentation may be the preferred method for re-
moving SS from combined sewer overflows, it is not always
the most economical. The primary limitations are the large
size of sedimentation facilities, the long detention times,
and the low removal efficiency for colloidal matter [7].
Two solutions to these limitations are: (1) combining
the sedimentation process with storage facilities, which is
usually done simply by the nature of the storage configura-
tion, and (2) using tube settlers or separators to reduce
the detention time and improve SS removals. Several
storage/sedimentation facilities have been constructed and
operated with apparent success (see Table 35). Relatively
little operational information is available, however.
Further data will be available as some of the ongoing proj-
ects are completed.
Table 35. SUMMARY DATA ON SEDIMENTATION BASINS
COMBINED WITH STORAGE FACILITIES
Location of facility
Cottage Farm Detention
and Chlorination Facility,
Cambridge, Mass.
Chippewa Falls, Wis.
Columbus , Ohio
Whittier Street
Alum Creek
Humboldt Ave . ,
Milwaukee, Wis.
Spring Creek Jamaica
Bay, New York, N.Y.
Mount Clemens, Mich.
Lancaster, Pa.
Weiss Street,
Saginaw, Mich.
Size
mil
gal
1.3
2.8
4.0
0.9
4.0
10.0
b.
5.8
1.2
3.6
, Removal efficiency
Type of
storage facility SS BODj, *
a. In operation
Covered concrete 45 Erratic
tanks
Asphalt paved 18-70 22-74
storage basin
Open concrete 15-45 15-35
tanks
Covered concrete NA NA
tank
Covered concrete NA NA
tanks
Covered concrete NA NA
tanks
In planning or construction phase
Concrete tanks
Concrete silo
Concrete tanks
Type of solids
removal equipment
Manual washdown
Solids removal by
street cleaners
Mechanical wash-
down
Mechanical wash-
down
Resuspension of
solids by mixers
Traveling bridge
hydraulic mixers
Resuspension of
solids and mechan-
ical washdown by
eductors
Air agitation and
pumping
Mechanical and
manual washdown
a. All facilities store solids during storm event and clean sedimentation basin when flows to
the interceptor can handle the solid water and solids.
b. NA := not available.
Note: mil gal. x 3,785.0 = cu m
216
-------
Typical Combined Sewer Overflow Sedimentation Facilities
The description of the individual sedimentation facilities
is presented here except for combination storage/clarifiers
previously discussed in Section IX, Storage.
Columbus, Ohio — This was perhaps the first application of
combined sewage holding tanks built in the United States.
The Whittier Street Storm Standby Tanks were built in 1932.
The three tanks, as originally constructed, had no sludge
collection equipment. The floors were sloped to aid drain-
age and the washdown of collected solids following the storm
In 1966, the tanks were modified and mechanical sludge col-
lection equipment installed. Floor elevations of the storm
standby tanks are such that they are above the flow level in
the interceptor at 2 times dry-weather flow. To fill the
tanks, a downstream regulating gate is throttled automati-
cally to allow only 2 times dry-weather flow to pass on to
the treatment plant. The excess flow backs up, flowing into
the storm standby tanks. The tanks are filled sequentially
to reduce cleanup time when all three tanks are not needed
to contain the storm flow. Overflow occurs only after all
three tanks are filled. The Whittier Street facility has a
capacity of 11,040 I/sec (252 mgd), with a detention period
of 24 minutes [19]. The total tank volume is 15.9 Ml
(4.2 mil gal.) .
Dallas, Texas — The Bachman Stormwater Plant has three
settling tanks with a combined capacity of 1,225 I/sec (28
mgd). This plant was designed to test the effects of three
slightly different processes on heavily infiltrated munici-
pal sewage. The three processes are (1) flocculation fol-
lowed by sedimentation and additional clarification using
tube settlers, (2) flocculation followed by sedimentation,
and (3) sedimentation only. Waste lime sludge from a nearby
water treatment plant and polymer are used as flocculants.
All three tanks are rectangular with circular sludge col-
lectors only in the first half of each tank. Operational
tests are being run presently and no performance data are
yet available.
Mount Clemens, Michigan — Planning is currently underway for
a sedimentation/storage facility to work in conjunction with
the biological treatment system described in Section XI,
Biological Treatment. The Mount Clemens storage facilities
are divided into two basic parts. The first is a combina-
tion storage/clarifier portion consisting of three rectangu-
lar tanks in series with a total storage capacity of 21,950
cu m (5.8 mil gal.). The tanks are separated by weirs which
allow each tank to be filled sequentially. The bottoms of
the tanks are sloped toward a channel running the length of
217
-------
each tank. Settled solids are resuspended by using sub-
merged eductors along the walls of the tanks. Each eductor
is directed to the center channel and is activated after the
tanks have been partially drained.
Saginaw, Michigan - The Weiss Street facility, a storage/
clarifier combination with a 13,620 cu m (3.6 mil gal.)
storage capacity, is currently under construction. The first
of three tanks is designed to capture the grit and most of
the heavier suspended matter. It operates in series with the
remaining two tanks which operate in parallel. The tank
bottoms are sloped and troughed to aid in the removal of the
settled solids. The tanks are washed down under manual con-
trol after each storm using spray nozzles mounted along walk-
ways next to the tanks. The first tank also has a clamshell
bucket to remove grit. The last two tanks have horizontal
screens placed below the water level just in front of the
overflow weirs. A baffle is used to ensure water flows up-
ward through the screens before overflowing the weir.
Operation
It is interesting to note that in all the storage/
sedimentation projects, settled sludge is stored until
after the storm event. At this time, the contents in the
tanks, including the solids, are slowly drained back to the
interceptor. The notable exception to this procedure is at
Chippewa Falls, Wisconsin, where solids are removed from the
basin after dewatering using a front end loader or an ordi-
nary street sweeper for disposal at a sanitary landfill.
Several different methods for resuspending or removing the
settled solids are used at the various other storage/
sedimentation facilities. At the Humboldt Avenue facility
in Milwaukee, Wisconsin, mechanical mixers are used to re-
suspend the settled solids. Traveling bridges with hy-
draulic nozzles are used at the Spring Creek plant in New
York City. At the Cottage Farm facility, a fixed water
spray in conjunction with a sloped and troughed floor is
used to flush the solids out of the basins.
The use of tube settlers and separators has been limited
mainly to water treatment facilities and some secondary
clarifiers at municipal sewage treatment plants. Their use
in the storm overflow facilities is found presently only at
the Bachman Stormwater Plant in Dallas. To date, the oper-
ating data for this plant are insufficient for reaching any
conclusions regarding the effectiveness of tube settlers
for storm overflows [5, 4].
218
-------
The advantages of sedimentation are that (1) the process is
familiar to the design engineers and operators; (2) the
facilities can be made to operate automatically; (3) sludge
collection equipment can be added to storage facilities with
a very minimal incremental cost; (4) the process provides
for storage of at least part of the overflow; and (5) disin-
fection can be effected concurrently with sedimentation in
the same tank. The disadvantages are that (1) the land
requirement is high; (2) the cost when the process is used
alone is high; (3) only primary treatment is afforded the
storm overflow; and (4) some manual cleaning of most basins
is necessary after each storm event. It is recommended that
the primary use of sedimentation continue to be in conjunc-
tion with storage facilities.
Costs
The actual costs of sedimentation facilities are difficult
to assess, particularly when they are combined with storage
facilities. The cost of sedimentation in the latter case is
presented as the total cost, since a sedimentation basin
alone would cost approximately the same amount. The cost
data available for various storage/sedimentation facilities
are presented in Table 36.
Table 36. COST OF STORMWATER SEDIMENTATION FACILITIESa
, Capital cost,
Location of facility mgd $/mgd $/acre
Annual
operation and
lintenance cost,
$/mgd
Cambridge, Mass.
Cottage Farm Storm-
water Treatment
Station 62.4 100,000 -- 1,240
Columbus, Ohio
Whittier Street 192 32,000
Alum Creek 43 43,000
Milwaukee, Wis.
Humboldt Avenue 192 10,500 3,560 260
New York, N. Y.
Spring Creek-
Jamaica Bay 480 44,000 6,530
a. ENR = 2000.
b. Maximum capacity assuming 30-minute detention time.
c. Includes pump station and screening facilities.
Note: mgd x 43.808 - I/sec
219
-------
Other basic cost data for the sedimentation facilities have
been presented previously in Table 34 in Section IX, Storage
The data are scattered and no acceptable curve can be de-
rived from them. To assist in evaluating the data, four
estimates were made using cost curves from the literature
[18]. The resulting points form the curve shown on
Figure 41. The curve is intended to give only an indication
of the relative magnitude of construction costs for sedi-
mentation facilities.
DISSOLVED AIR FLOTATION
Introduction
Dissolved air flotation is a unit operation used to separate
solid particles or liquid droplets from a liquid phase.
Separation is brought about by introducing fine air bubbles
into the liquid phase. As the bubbles attach to the solid
particles or liquid droplets, the buoyant force of the com-
bined particle and air bubble is great enough to cause the
particle to rise. Once the particles have floated to the
surface, they are removed by skimming. The principal reason
for using dissolved air flotation is because the relative
difference between the specific gravity of the combined par-
ticle and air bubble and the effective specific gravity of
water is made significantly higher and is more controllable
than using plain sedimentation. Thus, according to Stokes'
Law, the velocity of the particle and air bubble is greater
than the particle itself because of the greater difference
in specific gravities. In engineering terms this means
higher overflow rates and shorter detention times can be
used for dissolved air flotation than for conventional
settling.
This process has a definite advantage over gravity sedimen-
tation when used on combined sewer overflows in that parti-
cles with densities both higher and lower than the liquid
can be removed in one skimming operation. Dissolved air
flotation also aids in the removal of oil and grease which
are not as readily removed during sedimentation.
There are two basic processes for forming the air bubbles:
(1) dissolve air into the waste stream under pressure,then
release the pressure to allow the air to come out of solu-
tion, and (2) saturate the waste with air at atmospheric
pressure,then apply a vacuum over the flotation tank to re-
duce the pressure allowing the bubbles to form. The first
process is used most commonly. There are three methods for
implementing the first process. The first method is the
full flow method where all the flow is pressurized and mixed
with air before entering the flotation tank. The second is
220
-------
CO
oc
«c
_J
—1
ca
<=» 10
CO
1
NSTRyCTION COST, M
S
0.1
: LEGEND:
'•
Q STORAGE/CLARIFIER COST (SEE TABLE 36)
: x ESTIMATED COST [l B] I
•
-
; /}
•
m .
/*
/' '
\
El
-
-
10 100
DESIGN CAPACITY, MGD
NOTC: BGD X 4J. 808-L/JEC
Figure 41. Construction cost versus
design capacity for sedimentation
1000
221
-------
split flow flotation where part of the incoming flow is
pressurized and mixed with air before being recombined with
the remaining flow and entering the flotation tank. And
the last is the recycle system in which a portion of the
effluent is pressurized before being returned and mixed with
the incoming flow. The last two methods are used for the
larger size units since they require only a portion of
the total flow to be pressurized. In combined sewer over-
flow treatment studies the split flow method has been used
because the flotation tank only needs to be designed for the
actual flow arriving at the plant and need not include any
recycled flow. However, subsequent laboratory studies have
indicated better removals may be achieved by using the re-
cycle type of dissolved air flotation [39].
Typical facilities consist of saturation tanks to dissolve
air into a portion of the flow, a small mixing chamber to
recombine the flow that has been pressurized with that which
has not, and flotation tanks or cells. In most flotation
cells, two sets of flight scrapers, top and bottom, are used.
These remove the accumulated float and settled sludge. At
two major combined sewer overflow study sites, however, the
bottom scrapers were not used. Instead, 50-mesh rotating
fine screens ahead of the dissolved air flotation units re-
moved the coarser material in the waste flows, thus elimi-
nating the majority of settleable material. A schematic of
the dissolved air flotation facilities at Racine, Wisconsin,
is shown on Figure 42. Photographs of a typical dissolved
air flotation facility are shown on Figure 43.
Design Criteria
The principal parameters that affect removal efficiencies
are (1) overflow rate, (2) amount of air dissolved in the
flows, and (3) chemical addition. Chemical addition has
been used to improve removals, and ferric chloride has been
the chemical most commonly added.
Any one of several methods may be used to size a dissolved
air flotation facility. Values for design parameters used
in the combined sewer overflow studies are listed in
Table 37. The most commonly used design equation is that
recommended by the American Petroleum Institute [1].
When designing dissolved air flotation, regardless of whether
by formulated equations found in the literature or by the
design parameters listed previously in Table 37, certain
items should be remembered. First, the saturation tank
should be a packless chamber to prevent solids plugging or
buildup and second, excess amounts of air should be supplied
222
-------
REGULATOR
TRUNK
SEWER
SPLIT FLOW AIR
DISSOLVING
PRESSURIZATION
TANKS
SLUDGE HOLDING
TANKS (TYP)
INTERCEPTOR
SEWER
Figure 42. Schematic of dissolved air
flotation facilities at Racine, Wisconsin [8]
223
-------
(f)
Figure 43. Dissolved air flotation facilities (Racine)
(a) Overview of site during construction (b) Overview of flotation tanks after
light roof addition (c) Fine drum screen pretreatment units (d) Air saturation
(pressurization) tanks (e) End of float drawoff (helical cross conveyor) (f)
Float holding tanks (for temporary storage before feedback to interceptor)
224
-------
Table 37. DISSOLVED AIR FLOTATION
DESIGN PARAMETERS3
Design parameter
Range
Overflow rates, gpd/sq ft
Horizontal velocity, fpm
Detention time
Flotation cell, min
Saturation tank, min
Mixing chamber, min
Ratio of pressurized flow
to total flow, \
Air to pressurized flow
ratio, scfm/100 gal.
Pressure in saturation tank,
psi
Float
Volume, % of total flow
Concentration of float, \
Settleable sludge handling
Without pretreatment
With pretreatment
2,000-4,500
1.3-3.7
15-20
1-3
- 1
20-30
1.0
50-60
0.75-1.4
1-2
Provide for sludge removal
equipment in flotation cells.
Pretreatment uses self-cleaning
50-meshfine screens;
sludge removal equipment not
required in flotation cells.
a. Values for split flow dissolved air flotation.
Note: gpd/sq ft x 1.698 x 10'3 = cu m/hr/sq m
fpm x 0.00508 = m/sec
scfm/100 gal. x 0.00747 = cu m/min/100 liters
psi x 0.0703 = kg/sq cm
225
-------
and bled off since oxygen has a higher solubility than
nitrogen. Finally, the pressure release valve and the
discharge line from the saturation tank should be designed
to induce good mixing with the remainder of the flow and
promote fine bubble formation [1].
Overflow Rate — The removals achieved by dissolved air flota
tion are governed by several factors. The most critical
design parameter is the surface overflow rate which can be
easily translated into the rise rate of the particle and
air bubble. To remove an air particle with a given rise
rate,the corresponding overflow rate must not be exceeded.
In rough terms, it has been reported that overflow rates
above 6.1 cu m/hr/sq m (3,600 gpd/sq ft) start to reduce
removal efficiencies. Below this value the removals remain
relatively constant.
Dissolved Air Requirements — Also important in affecting
removals is the amount of air dissolved. An insufficient
supply of dissolved air reduces the amount of air available
for each solid particle,and thus the difference between the
air-particle density and the density of water is not great
enough to meet the minimum rise rate. Also, the better the
atomization or bubble coverage over the plan area of the
tank, the better the chance for collision between the
bubbles and the suspended particles. The amount of air
supplied to a split flow flotation facility is dependent on
the percentage of flow saturated with air and the pressure.
In the studies using combined sewer overflows, the optimum
value for the percentage of flow pressurized averages around
20. In one study with a full flow system, removals were
found to be directly related to the pressure used in the
saturation tank, see Figure 44 [17], The optimum pressure
is 3.5 to 4.2 kg/sq cm (50 to 60 psi) which agrees with
other studies performed [40, 2].
Flotation Aids — Probably, the most controllable factor
affecting particle removals is the amount and type of chemi-
cals added. In all studies, some kinds of chemicals were
added to improve removals. In one case, small floating
beads were used in lieu of air to provide the flotation [12],
This proved to be unsuccessful. The majority of chemicals
added, however, were polyelectrolytes and ferric chloride.
Ferric chloride has been reported to be the most successful
and has improved SS removals by more than 30 percent. The
use of polyelectrolytes alone and in one case bentonite clay
with polyelectrolytes has not resulted in important in-
creases in removal efficiencies. Lime and alum have also
been used.
226
-------
70
60
50
40
ifc*
/
X
x^
-*— —
0 10 20 30 40 50
SATURATION TANK PRESSURE, PSI
NOTE: PSI \ 0.0703- KI/SQ CM
60
70
Figure 44. Relationship between suspended solids removal
and saturation tank pressure [17]
In the studies on treatment of combined sewer overflows, SS
removals of approximately 60 percent and BOD5 removals of
about 40 percent have been reported. The addition of 20 to
30 mg/1 of ferric chloride increased SS removals to approxi-
mately 70 percent and BODs to about 50 percent. Phosphorus
removals increased from near zero to 70 percent. The aver-
age reported removals from three studies are listed in
Table 38 [40, 17, 12] .
Demonstration Projects
There have been four studies using dissolved air flotation to
treat combined sewer overflows. These are at Milwaukee and
Racine, Wisconsin (the latter of which is just finishing
construction in mid-1973), Fort Smith, Arkansas, and San
Francisco, California. The Fort Smith and Milwaukee plants
were pilot operations.
Milwaukee, Wisconsin [40] - The 220-1/sec (5-mgd) pilot dem-
onstration plant was a prefabricated steel unit set on the
ground surface under an elevated section of highway. The
use of fine screens (297 micron openings) ahead of the flota-
tion chamber to eliminate the need for bottom scrapers
was initially tested at this facility. The screens removed
227
-------
Table 38. TYPICAL REMOVALS ACHIEVED WITH
SCREENING/DISSOLVED AIR FLOTATION
Without chemicals With chemicals
Constituents Effluent, mg/1 % Removal Effluent, mg/1 % Removal
ss
vss
BOD
COD
Total N
Total P
81-106
o
47a
29-102
123a
4.2-16.8
1.3-8.8
56
Q
53a
41
41a
14
16
42
18
12
46
4.2
0.5
-48
-29
-20
-83
-15.9
-5.6
77
70
57
45
17
69
a. Only one set of samples.
grit and most of the nonfloatable material successfully.
The system used the split flow method for dissolving air
into the flow. Approximately 20 percent of the total flow
was pressurized to 2.8 to 3.5 kg/sq cm (40 to 50 psi) in a
packless saturation tank,then remixed with the remainder
of the flow for one minute in a mixing chamber. Flow then
entered the flotation cell for flotation and removal of the
floating matter (float) by scrapers. The float was col-
lected in a holding tank for discharge back to the dry-
weather interceptor.
Racine, Wisconsin — The Racine prototype facilities are
essentially the same design as the one in Milwaukee. It,
however, is constructed partly underground out of concrete.
There are two plants: one 615 I/sec (14 mgd) and the other
1,925 I/sec (44 mgd) in size. Flow to each plant passes
through a 2.5 cm (1-inch) bar screen before being lifted to
the fine screens by screw pumps. Each plant is built with
multiple flotation tanks to accommodate a high flow
variation. A separate air saturation tank and pump serves
each flotation tank. Flow into the flotation tanks is
controlled by weirs which allow sequential filling of only
as many tanks as are necessary to handle the flow.
228
-------
Fort Smith, Arkansas [17] - The pilot-plant study at Fort
Smith, Arkansas, used a dual flotation cell tank preceded by
a circular vibrating screen and four hydrocyclones for gross
SS removal. Full flow pressurization was used with a pack-
less saturation tank. The flotation cells, with an effec-
tive surface area of 22 sq m (240 sq ft), had both float
scrapers and bottom sludge scrapers.
The dissolved air flotation system, with 12 minutes de-
tention time, removed suspended solids from combined sewage
as effectively as conventional clarifiers with 4 hours deten-
tion time. During rain events and without chemical aids,
the system removed an average of 69 percent of the suspended
solids passing a gyratory screen installed to remove gross
particles. Injection of alum and a polyelectrolyte into the
system increased the removal rate to an average of 84
percent. Alum alone was ineffective. Without chemical aids
BODs reduction averaged 26 percent. When chemical floccu-
lating aids were injected, BODs reduction increased to an
average of 42 percent. The float collected contained 5 to
7 percent dried solids of 70 percent volatility.
San Francisco, California - The facility at San Francisco
is a complete prototype plant. The plant is fully automated
and contains mechanically cleaned bar screen, chemical feed
equipment, pressurizing pumps, saturation tanks, two 525-
1/sec (12-mgd) dissolved air flotation tanks, and chlorina-
tion facilities. It can function on either the split flow
or recycle method of operation and accepts overflow from a
combined sewer. The flotation cells have both float scrapers
and bottom sludge scrapers. Wet-weather operational data to
date have been sparse. However, extensive testing has been
performed using raw and diluted raw sewage [16],
A summary of the performance characteristics for this facil-
ity is shown in Table 39. The removal of pollutants at
any given level of process variables, and with air/solids
ratio nonlimiting, was greatly dependent upon the type and
dosage of chemicals. Alum was more effective than polymer
for the chemical conversion of the influent solids to forms
susceptible to separation by dissolved air flotation.
Increasing process efficiency was observed with increasing
alum dosages and with decreasing liquid loading rates. The
character of the float/sludge, which for convenience was
combined into a single waste stream for analysis, was depend-
ent on the type and amount of chemical used for pretreatment.
Average characteristics ranged from 22 to 50 mg/1 for oil
and grease, 23 to 26 mg/1 for total nitrogen, 0.5 to 1.8
mg/1 for orthophosphate, and 98 to 584 mg/1 for total sus-
pended solids.
229
-------
Table 39. SUMMARY OF PERFORMANCE CHARACTERISTICS,
BAKER STREET DISSOLVED AIR FLOTATION FACILITY [16]
SAN FRANCISCO, CALIFORNIA
Effluent concentration,
mg/1
Constituent
BOD5
COD
Settleable solids
Oil and grease
Floatables
Total coliform
Fecal coliform
Total nitrogen
Orthophosphate
Color
Maximum
114.
205.
15.
26.
0.
2.4 x
2.4 x
20.
4.
22.
0
0
0
3
57
105
105*
1
45
0
Minimum
34.0
53.0
<0 . 1
3.3
<0.01
<30a
<30a
10.6
<0 . 07
2.0
Removal efficiency,
Maximum
70.
77.
93.
63.
100.
>99.
>99.
53.
99.
95.
5
0
5
2
0
99
99
0
0
0
Minimum
13
10
0
0
60
99
99
0
43
15
.5
.8
.0
.0
.0
.44
.44
.0
.4
.8
Average
46
44
47
29
95
99
99
18
80
57
.1
.4
.7
.1
.2
.92
.91
.4
.9
.3
a. MPN/100 ml
Advantages and Disadvantages
The advantages of dissolved air flotation are that (1) moder
ately good SS and BODs removals can be achieved; (2) the
separation rate can be controlled by adjusting the amount of
air supplied; (3) it is ideally suited for the high amount
of SS found in combined sewer overflows; (4) capital cost is
moderate owing to high separation rates, high surface load-
ing rates, and short detention times; and (5) the system can
be automated. Disadvantages of dissolved air flotation in-
clude: (1) dissolved material is not removed without the
use of chemical additions; (2) operating costs are rela-
tively high compared to other physical processes;
(3) greater operator skill is required; and (4) provisions
must be made to prevent wind and rain from disturbing the
float.
230
-------
Costs
The cost of dissolved air flotation facilities scaled to a
1,095 I/sec (25 mgd) capacity is presented in Table 40. The
wide spread between the San Francisco data and the remaining
two locations is attributed to the architectural treatment
given to San Francisco's facilities. In all cases extra
cost items such as cofferdams, special foundations, etc.,
were not included but pretreatment devices were. The avail-
able data were plotted to further compare the costs and to
help develop a cost equation (see Figure 45). The developed
equation, limited to facilities ranging from 219 to 43,810
I/sec (5 nigd to 1,000 mgd) is:
Ca = 53,000 (Qa)
0.84
where Ca is the capital cost of the facility and Qa is the
plant capacity in mgd. The curve defined by this equation
is shown on Figure 45 as well as a similar curve developed
for sedimentation facilities. A comparison between the two
cost curves shows dissolved air flotation costs at about
Table 40. DISSOLVED AIR FLOTATION COST FOR 25 MGD1
Milwaukee, Wisconsin
Fort Smith, Concrete Racine, San Francisco,
Arkansas Steel tank tank Wisconsin California
(Estimated (Actual (Estimated (Actual (Actual
cost) cost) cost) cost) bid cost)
Plant location
Average
cost
Construction cost
including pre- .
treatment devices $480,000
Operation and
maintenance
Total cost,
t/1,000 gal. 10.83
Chemical cost
alone,
-------
100
CO
a
C-3
10
0.1
LEGEND:
0 SAN FRANCISCO DATA [ 15]
0 MILWAUKEE DATA [4«]
A FORT SMITH DATA [17]
* DERIVED COST EQUATION
x COST CURVE FOR SEDIMENTATION
10 100
DESIGN CAPACITY, MGD
'iooo
NOTE; MBD x 43.808- I/SEC
Figure 45. Construction cost versus
design capacity for dissolved air flotation, ENR 2000
232
-------
one-half those for sedimentation with both yielding approxi-
mately 60 to 70 percent SS removal. The operation and main-
tenance costs are not so well defined. The average operation
and maintenance cost value reported was $0.017/1,000 1
($0.066/1,000 gal.).
SCREENS
Introduction
Screens of almost all sizes are effective in removing sus-
pended material. The range in sizes is from 3-inch clear
openings (bar screens) to openings as small as 15 microns
(stainless steel woven microscreens). To facilitate the
following discussion, screens have been divided into four
classifications, as shown in Table 41: (1) bar screens
(2) coarse screens, (3) fine screens, and (4) microscreens.
Bar Screens and Coarse Screens - No special studies have
been made to evaluate these two types of screens in relation
to combined sewer overflows. The bases for design should be
Table 41. CLASSIFICATION OF SCREENS
Opening
Classification
Mesh
Inches
Microns
Bar screens
(> 1 in.)
Coarse screens
(1 to 3/16 in.)
Fine screens
(3/16 to 1/250 in.)
Microscreens
(< 1/250 in.)
3
4
6
8
9
10
14
20
28
35
48
60
80
100
150
230
400
3.0
2.0
1.050
0.742
0.542
0.371
0.263
0.185
0.131
0.093
0.07«
0.065
0.046
0.0328
0.0232
0.0164
0.0116
0.0097
0.0075
0.0058
0.0041
0.0026
0.0015
0.0009
--
--
1,651
1,168
833
589
417
295
246
180
147
104
65
38
23
Note: in. x 2.54 = cm
233
-------
the same as for their use in dry-weather treatment
facilities. The reader is referred to the literature for
the necessary details [25]. Except for bar screens, their
use for combined sewer overflows may be limited. Coarse
screens are used as a pretreatment and protection device at
the Cottage Farm Detention and Chlorination Facility in
Boston. Bar screens are recommended for almost all storage
and treatment facilities and pump stations for protection of
downstream equipment. Typical screenings from a 1-inch bar
screen are shown on Figure 46.
Fine Screens and Microscreens — Fine screens and micro-
screens are discussed together because in most cases they
operate in a similar manner. The types of units found in
these classifications are rotating fine screens, hereinafter
referred to as drum screens; microscreens, commonly called
microstrainers; rotary fine screens; and hydraulic sieves
(static screens); vibrating screens; and gyratory screens.
To date, vibrating screens and gyratory screens have not
been used in prototype combined sewer overflow treatment
facilities.
Description of Screening Devices
Microstrainers and Drum Screens — The microstrainer and drum
screen are essentially the same device but with different
screen aperture sizes. A schematic of a typical unit is
shown on Figure 47. They are a mechanical filter using a
variable, low-speed (up to 4 to 7 rpm), continuously back-
washed, drum rotating about a horizontal axis and operating
under gravity conditions. The filter usually is a tightly
woven wire mesh fabric (called screen) fitted on the drum
periphery in paneled sections. The drum is placed in a tank,
and wastewater enters one end of the drum and flows outward
through the rotating screen. Seals at the ends of the drum
prevent water from escaping around the ends of the drum into
the tank. As the drum rotates, filtered solids, trapped on
the screen, are lifted above the water surface inside the
drum. At the top of the drum, the solids are backwashed off
the screen by high-pressure spray jets, collected in a trough,
and removed from the inside of the drum. In most cases,
both the rotational speed of the drum and the backwash rate
are adjustable. Backwash water is usually strained effluent.
The newer microstrainers use an ultraviolet light irradia-
tion source alongside the backwash jets to prevent growth of
organisms on the screens [36]. The drum is submerged from
approximately two-thirds to three-quarters of its diameter.
As noted previously, the usual flow pattern is radially out-
ward through the screen lining the drum; however, one drum
screen application used a reverse flow pattern [41].
234
-------
(c)
Figure 46. Screenings from mechanically cleaned bar racks
(a) View of screeninas from 1.9 cm (3/4-in.) clear opening rack and sluice trough for
(a) View of screenings from
return to interceptor (Boston)
c ear opening trough rack (b)
cieek
cm (3/4-in.) clear opening rack and
These screens are preceded by coarse
Screenings from 2.5 cm (1-in.) clear
10.1 cm
open! ng
(4-ln.)
rack at
ngs at
235
-------
BACKWASH
HOOD
Figure 47. Schematic of a
microstrainer or drum screen
The drum was completely submerged within an influent tank,
and flow passed inward through the circumference of the
drum. Submerged backwash jets were placed inside the drum.
Screen openings for microstrainers range from 15 to 65
microns and for drum screens, from 100 to 600 microns. The
various sizes of screen openings that have been tested on
combined sewer overflows, and other data, are listed in
Table 42.
Microstrainers and drum screens can be used in many differ-
ent treatment schemes. Their versatility comes from the
fact that the removal efficiency is adjustable by changing
the aperture size of the screen placed on the unit. The
primary use of microstrainers would be in lieu of a sedimen
tation basin to remove suspended matter. They can also be
used in conjunction with chemical treatment, such as ozone
or chlorine for chemically disinfecting/oxidizing both
organic and nonorganic oxidizable matter or microorganisms
236
-------
Table 42. MICROSTRAINER AND DRUM SCREEN INSTALLATIONS
IN THE UNITED STATES THAT TREAT
COMBINED SEWER OVERFLOWS
Location Type of device
Philadelphia, Pa. Microstrainer
[26, 27]
Mt. Clemens, Mich.b Microstrainer
[33, 32]
Milwaukee, Wis.a Drum screen
[40]
Cleveland, Ohio3 Drum screen
[22]
Size
Diameter, Length,
ft ft
5 3
6 6
7-1/2 6
4 1
Screen
opening,
microns
23, 35
21
297
420
Flow ,
mgd Use of strainer
0.43 Main treatment
1 Filters oxidation
pond effluent
5 Pretreatment to
dissolved air
flotation
^1.3 Pretreatment to
dual-media
filtration
East Providence, Drum screen
R.I.c [41]
40 sq in.d
150, 190, 0.0086 Main treatment
230
a. Pilot.
b. Prototype.
c. Bench scale.
d. Screening area.
Note: ft x 0.305 = m
mgd x 43.808 = I/sec
sq in. x 6.452 = sq cm
by removing excess solids prior to the disinfection/
oxidation step [26, 40, 27]. They have also been used as
polishers for treatment plant effluent. The drum screens
are used as pretreatment devices prior to other treatment
units. Two such examples are: (1) pretreatment for dis-
solved air flotation, removing coarser solids to reduce the
amount of SS settling out in the dissolved air flotation
units [40]; and (2) pretreatment for dual-media filtration
or other filtration processes [22] .
Many microstrainer and drum screen installations are in
operation. Although not all of these installations treat
combined sewer overflows, results of studies indicate that
the development of the microstrainer and drum screens is
fairly complete, and major problems with the units most
likely have been solved.
Efficiency - The removal efficiencies are affected by two
mechanisms: (1) straining by the screen and (2) filtering
of smaller particles by the mat deposited by the initial
straining [37, 13]. The governing mechanism is the
237
-------
size of the screen openings because this determines the
initial size of particles removed. The efficiencies of a
microstrainer and drum screens treating a waste with a nor-
mal distribution of particle sizes will increase as the size
of screen opening decreases. Suspended solids removals re-
ported in various studies within the United States bear this
out, as shown on Figure 48 [41, 26, 40, 22, 35, 27, 23]. In
reality, however, removals are based on the relative sizes
between the screen opening and the particle size. A drum
screen with a large screen opening can achieve high removals
if the majority of the solids in the waste flows are larger
than the screen opening. It appears important not to pump
ahead of microstrainers because this tends to break up frag-
ile particles and thereby reduce removal efficiencies. The
use of positive displacement pumps or spiral pumps may be
permissible.
The second most important condition affecting removal effi-
ciencies, especially for microstrainers, is the thickness of
filtered material on the screen. Whenever the thickness of
this filter mat is increased, the suspended matter removal
160
80
40
20
-o
8
100 200 300 400
SCREEN OPENING, MICRONS
500
Figure 48. SS removal versus screen opening
238
-------
will also increase because of the decrease in effective pore
size and the filtering action of the filtered mat [23, 26,
27, 13]. This condition was first observed when the influ-
ent waters had a high concentration of suspended matter [27].
To create a thicker filter mat on the screens, low drum
speeds are used so that the SS loading on the screen is
increased [26, 27, 23].
In recent studies using microstrainers, automatic drum speed
controls that are proportional to the headless across the
screen have been used [26, 27]. More sophisticated controls
are not usually warranted on combined sewer overflow
facilities. The high headloss is related to a thicker mat
and increased removal efficiencies. Backwash control also
is used sometimes [40] . Backwash is turned off during times
of low solids loading and activated when the loadings in-
crease the headloss. High head differentials may create
problems, however, by producing shear forces great enough to
break apart the fragile particles and flocculant material,
resulting in lower removal efficiencies by allowing the par-
ticles to pass through the screen [13, 6]. Also, the micro-
screens and fine screens must be designed to withstand these
increased head differentials [30].
Microstrainers and fine screens remove from 25 to 90 percent
of the SS and from 10 to 70 percent of the BOD5, depending
on the size of screens used and the type of wastewater being
treated. Microstrainers generally achieve approximately 70
and 60 percent SS and BODs removals, respectively. Fine
screens generally remove about 38 and 16 percent SS and BODs,
respectively. Most data reported in various studies con-
ducted throughout the United States indicate a broad range
of removal rates. Additional studies on combined sewer over-
flow strainers are warranted before removal efficiencies can
be predicted with any degree of accuracy, particularly with
respect to organic pollutants [26, 27].
Filter aids — Although coagulants have not been studied ex-
tensively in conjunction with microstrainers, it has been
reported that ferric chloride can improve removal
efficiencies [31]. In one study alum was used for phos-
phorus removal and to increase suspended matter removal by
producing a floe. This test was unsuccessful because the
alum floe was extremely small and very fragile, and as a
result it washed through the microstrainer [23]. In another
pilot-scale study presently underway, the use of alum-ferric
chloride appears to be proving unsuccessful. However, by
using approximately 2 mg/1 of cationic polymer (Betz 1150
and Atlasep 105C) the effluent quality is being improved
with respect to SS concentrations with an increased flux rate
of 97.7 cu m/hr/sq m (40 gpm/sq ft). These results are
239
-------
preliminary and more work is needed to verify them at a
larger scale and at the Philadelphia pilot plant site.
Screen cleaning — Of the several conditions which affect the
operation of the microstrainer and drum screen, the most
notable is proper cleaning of the screen. Spray jets,
located on the outside of the screen at the top of the drum,
are directed in a fan shape onto the screen. It has been
found that the pressure of this backwash spray is more im-
portant than the quantity of the backwash [13, 6]. There
does not seem to be any relationship between the volume of
backwash water applied and the hydraulic loading of the
microstrainer or drum screen. Thus, a constant backwash
rate can be applied regardless of the hydraulic loading [23],
Results of tests at Philadelphia have indicated no backwash-
ing problems.
Occasionally the microstrainer and, to a lesser degree, the
drum screen cannot be effectively cleaned by the backwash
jets. This condition, called "blinding" of the screen, is
generally associated with oil, grease, and bacterial growths
[13, 41, 23, 6]. Oil and grease cannot be removed effec-
tively without using a detergent or other chemical, such as
sodium hypochlorite, in the backwash water [6], Generally,
microstrainers and drum screens with the finer screen open-
ings (<147 microns) should not be used in situations where
excessive oil and grease concentrations are likely to be en-
countered from a particular drainage area. Bacterial growths
also have caused blinding problems on microstrainers, although
they have not been a major problem with drum screens. The use
of ultraviolet light is an effective means of control, as men-
tioned previously. It is important, however, to use an
ultraviolet light source of the proper frequency designed to
minimize the amount of ozone created [29]. With proper con-
struction of the microstrainer it is possible to reduce the
chances of the creation of ozone [26].
Screen life — In a wet environment, ozone is relatively cor-
rosive to the stainless steel screens. Since screens are
woven with very fine stainless steel wires, the amount of
corrosion needed to break through a strand of the wire is
small [29]. In fact, it has been reported that ozone in a
wet environment is more corrosive to the stainless steel
wires than chlorine in a wet environment [29]. Therefore,
it is important to reduce the concentration of ozone and/or
chlorine in and around the microstrainer. Both chlorine and
ozone have been used upstream of the microstrainer, but
enough detention time has been allowed so that concentra-
tions of these chemicals are relatively low. It is better
240
-------
to use post-chlorination or ozonation rather than prechlori-
nation or ozonation to prevent corrosion and for better dis-
infection, too. The screen life in an uncontrolled ozone
environment is 3 to 5 years assuming continuous use.
In properly designed units, stainless steel screens will
last 7 to 10 years. A monel screen may last three times
this period.
Design parameters — The design parameters for the micro-
strainer and drum screens are generally limited to screen
opening (aperture), flux rate (gpm/sq ft of submerged screen
area), headless across the screen, drum rotational speed,
volume and pressure of backwash water, and the type of auto-
matic controls available (Table 43). Recommended values for
these parameters are outlined in Table 44. For micro-
strainers the size of screen openings would normally be
either 23 or 35 microns. For drum screens operating as pre-
treatment to the main type of treatment, the reported screen
openings range from 150 to 420 microns.
It has been found, however, that designing a microstrainer
installation cannot be done by a simple rating term, such as
flux rate, to determine the number of square feet required
[6]. Instead, the screen size, volume and type of waste, SS
concentration, and other factors must also be taken into
consideration.
Rotary Fine Screens — Rotary fine screens are somewhat simi-
lar to the microstrainer and drum screen in that tightly
woven wire mesh fabric fitted around a drum is used to
strain the waste. However, the drum of the rotary fine
screen rotates about a vertical axis at high speeds--between
30 and 65 rpm. The influent, introduced into the center of
the rotating drum, is directed along horizontal baffles that
distribute the flow evenly to the entire surface of the
screen. Flow passing through the screen is discharged to
the receiving water or routed for further treatment. The
concentrate (the retained solids, plus the portion of the
inflow that does not pass through the screen) is returned to
the interceptor for further treatment. A schematic of a
rotary fine screen is shown on Figure 49. The reported
screen opening sizes ranged from 74 to 230 microns. Back-
washing to remove the retained solids is done at discrete
intervals and is by high-pressure spray jets, approximately
3.52 kg/sq cm (50 psi), on both sides of the screen. The
backwash water contains detergent or other cleansing chemi-
cals for improved cleaning. Problems have been reported
with grease blinding the screens at the Portland, Oregon
pilot plant. Screen life has been reported to be approxi-
mately 1,000 operating hours for combined sewer overflow
application.
241
-------
Table 43. DATA SUMMARY ON MICROSTRAINERS AND
DRUM SCREENS
Location
(1)
Philadelphia,
Pa.
Milwaukee, Wis.
Cleveland, Ohio
Lebanon, Ohio
Chicago, 111.
Letchworth,
England
Lebanon, Ohio
East Providence,
R.I.
Reference
number
(2)
[27]
[26]
[26]
[26]
[26]
[40]
[22]
[6]
[23]
[35]
[6]
[41]
[41]
[41]
Screen
opening, Total,
micron sq ft
(3) (4)
23 9.4
23 9.4
23 47.0
23 47.0
35 47.0
297 144.0
420 12.6
23 15.0
23 314.0
23 47.0
35 15.0
150 0.28
190 0.28
230 0.28
Screen area
Submerged, Submerged, Flux rate, Headloss,
sq ft * gpm/sq ft in.
(5) (6) (7) (8)
7.4 78 40 23
7.4 78 25 12a
28a 60° 9.1 4.7a
35a 74a 6.9 3.6a
35a 74a S.4 3.4a
72-90 51-64 40-50 12-14
max
NA NA 100 NA
9 60 ^7.0 6 max
NA NA 6.6 6 max
NA NA 3.1 NA
9 60 ^7.0 6 max
0.28 100f 18-25 NA
0.28 100f 18-25 NA
0.28 100f 18-25 NA
Backwash
Pressure,
psi
(9)
40
40
40
40
40
1*
NA
NA
20-55
NA
NA
NA
NA
NA
* of
total flow
(10)
<0.5
NAb
NA
NA
NA
0.856
NA
5.3
3.0
NA
5.3
28
28f
28
Table 43 continued on page 243.
242
-------
Table 43. (Continued)
Drum
Location
(1)
Philadelphia,
Pa.
Milwaukee, Wis.
number rp»
(2)
[27]
[26]
[26]
[26]
[26]
[40]
(11)
7 max
4.Sa
2.2a
3.0a
1.5a
O.S-5
Type of
controls
(12)
Drum speed pro-
portional to
headloss
Same as above
Same as above
Same as above
Speed control
to keep constant
headloss
Drum and back-
wash activated
when headloss
16 in.
(13)
No
No
No
No
No
NA
SS
t
(14)
66
7Sa
80a
64a
44
27
BODS
t
(15)
43
Oc
NA
69C
40C
27
Type of
waste
(16)
Overflow
Overflow
Overflow
Overflow
Overflow
Overflow
Uc A rtf
s e ox
screen unit
(17)
Main treatment
Main treatment
Main treatment
Main treatment
Main treatment
Primary treatment
to dissolved air
flotation
Cleveland, Ohio
Lebanon, Ohio
[22]
[6]
25
89
8
61
Overflow
Primary treatment
to dual media filter
Activated Effluent polisher
sludge
effluent
Chicago, 111.
Letchworth,
England
Lebanon, Ohio
East Providence,
R. I .
123]
[35]
[6]
[41]
[41]
[41]
^0.7 Idles, drum Some 71 74 Activated
speed increases sludge
when headloss effluent
exceeds set
value
NA NA NA 48 43 Activated
sludge
effluent
3.2 max None Some 73 81 Activated
sludge
effluent
16 NA Yes 54-56 12-21 Synthetic
sludge, raw
sludge
8 NA Yes 40-46 11-21 Synthetic
sludge, raw
sludge
8 NA Yes 33-47 11-13 Synthetic
sludge, raw
sludge
Effluent polisher
Effluent polisher
Effluent polisher
Main treatment
Main treatment
Main treatment
a. From Ref. [14] (Philadelphia).
b. NA » not available.
c. Questionable number.
d. Pressure on screen, average.
e. Backwash activated whenever headloss exceeded 6 in. if continuous 3 percent of total flow.
f. This unit operates totally submerged with flow from outside to inside opposite of the normal microstrainer.
Note: sq ft x 0.0929 » sq m
gpra/sq ft x 0.679 - 1/sec/sq m
in. x 2.54 » cm
psi x 0.0703 « kg/sq cm
243
-------
Table 44. RECOMMENDED MICROSTRAINER DESIGN
PARAMETERS FOR COMBINED SEWER OVERFLOW TREATMENT
Screen opening, microns
Main treatment
Pretreatment
Screen material
Drum speed, rpm
Maximum speed
Operating range
Flux rate of submerged
screen, gpm/sq ft
Low rate
High rate
Headless, in.
Submergence of drum, I
Backwash
Volume, % of inflow
Pressure, psi
Type of automatic
controls
23-35
150-420
Stainless steel
5-7
0-max speed
5-10
20-50
6-24
60-80
<0.5-3
40-50
Drum speed propor-
tional to headless
Note: gpm/sq ft x 0.679 = 1/sec/sq m
psi x 0.0703 = kg/sq cm
SCREEN
BACKWASH NOZZLES
NFLUENT FLOW
AUTOMATIC VALVE
-SCREENED EFFLUENT
Figure 49. Rotary fine screen schematic [11]
244
-------
The rotary fine screen was first introduced in the United
States in the late 1960s under the trade name of SWECO
Centrifugal Wastewater Concentrator unit. It was designed
initially as a primary treatment device for municipal sewage
and combined sewer overflows. An EPA report on the rotary
fine screens, published in 1970, contains a description of
their development from the initial concept to what is used
today [38] .
Efficiency — The removal efficiencies of rotary fine screens
are affected by the independent variables similar to those
for the microstrainer. The overall performance is a
function of screen opening size, rotational speed of the
screen, strength and durability of the screen material, and
backwash operation. Removal efficiencies generally increase
as the size of opening decreases,as with the microstrainer;
on the other hand, efficiencies decrease as the volume of
feed applied to the screen increases, which is opposite to
what would be expected with a microstrainer [38] . The de-
crease in efficiencies is probably caused by the higher
forces on the particles being removed by the rotary fine
screen. The thickness of the filtered mat apparently is not
a factor here. Although the removal efficiencies are not
significantly affected by the rotational speed of the screen,
the hydraulic efficiency increases as the rotational speed
increases, as the mesh of the collar screen becomes coarser,
and as the velocity of the feed approaching the screen
increases [38]. Thus, the screen rotational speed should
be as high as possible without incurring other detrimental
effects. This limiting speed is approximately 60 rpm for a
1.52-meter (5 foot) diameter drum or around 4.9 m/sec
(16 fps). At higher rotational speeds the screen material
ruptures because of overstressing.
Removal efficiencies ranged from 60 to 90 percent for set-
tleable solids, 30 to 32 percent for suspended solids, and
16 to 25 percent for COD [38, 11].
The initial hydraulic efficiency (i.e., the fraction of
influent water passing through the rotating screen) ranged
between 85 and 90 percent. The hydraulic efficiency is
highest immediately following backwashing, then slowly de-
creases until the concentrate flow rate reaches a preset
value at which time backwash is initiated. It was reported
that the average hydraulic efficiency of 85 percent at the
beginning of a screening run decreased to an average of 72
percent just prior to backwashing [11] . Thus the screened
solids removed (averaging 55 percent of the settleable
solids, 27 percent of the SS, and 16 percent of the COD)
were concentrated into approximately 20 to 25 percent of
the total flow.
245
-------
Screen cleaning - In the studies conducted on the rotary fine
screen [38, 11], blinding (clogging of the screen) has been a
problem. Blinding has been attributed to oil, grease, and
industrial waste from a paint manufacturer. This problem is
similar to that experienced during the development of micro-
strainers. The latest study at Shore Acres, California,
solved this problem by enforcing an industrial waste ordinance
prohibiting discharge of oil wastes to the sewer system.
To improve backwashing, a solution of hot water and liquid
solvent or detergent has been found necessary to obtain ef-
fective cleaning of the screens. This may have been neces-
sary only because of the nature of the common waste encoun-
tered in both studies [38, 11]. Of the solvents tested,
acidic and alcoholic agents did not adequately clean the
screens. Alkaline agents were reported not effective by
Portland [11], but Cornell, Rowland, Hayes $ Merryfield [38]
reported a caustic solution was the most efficient solvent.
Chloroform, solvent parts cleaner, soluble pine oil, ZIP,
Formula 409, and Vestal Eight offered limited effectiveness.
ZEP 9658 cleaned the screens effectively, but this cleaner
was not analyzed to determine its effect on effluent water
quality. The removal of paint was done effectively only by
hand cleaning using ZEP 9658.
Screen life — In the first study [38] , the average screen
life was approximately 4 hours. In a study conducted a year
later [11] using a similar unit incorporating a new screen
design and a rotational speed of 65 rpm, the average screen
life was 34 hours. Reducing the rotational speed to 55 rpm
increased the average screen life to 346 hours. Results of
a subsequent study at Shore Acres, California, indicate that
screen life may exceed 1 year. This extended life, however,
is most likely attributable to the much lower hydraulic
loading rate, 39.5 versus 123 I/sec (0.9 versus 2.8 mgd) or
30 versus 97 1/sec/sq m (44 versus 143 gpm/sq ft). The
present predicted screen life is 1,000 hours. Some screen
failures were attributed to punctures caused by objects
present in the feed waters.
Design parameters — The design and operating parameters of
the rotary fine screen are presented in Table 45. No mathe-
matical modeling of the rotary fine screen has been
performed. Further tests of the rotary fine screen are
needed to determine more accurately the life of the screens,
the removal efficiencies, and design parameters.
Two points should be remembered with respect to rotary fine
screens: (1) waste flows from the rotary fine screen range
from 10 to 20 percent of the total flow treated and may
contain solvents that may be difficult to treat downstream;
246
-------
Table 45. ROTARY FINE SCREEN DESIGN PARAMETERS
Screen opening, microns 74-167
Screen material Stainless steel
(tensile bolting cloth)
Peripheral speed of screen, fps 14.4-15.7
Drum speed, rpm 30-65
Flux rate, gpm/sq ft 100-122
Velocity of feed water striking 9-12
screen, fps
Hydraulic efficiency, I of inflow 80-93
Backwash
Volume, I of inflow 0.02-2.5
Pressure, psi 50
Temperature, deg C 77
Backwash solvents ZEP 9658
Solvent dilution in backwash water 800:1-10:1
Treatment cycle times 4-1/2 min ON,
1-2 min OFF for
backwashing
Note: fps x 0.305 = m/sec
gpm/sq ft x 0.679 = 1/sec/sq m
psi x 0.0703 = kg/sq cm
and (2) the rotary fine screen requires a nearly fixed rate
of flow. Thus a battery of many parallel units is required
to treat combined sewer overflows.
Hydraulic Sieve — The hydraulic sieve, ranging in screen
sizes from 8 to 60 mesh, consists of a fixed flat screen in-
clined at 25 to 35 degrees from the vertical and a header
box that directs the flow in a flat sheet down the width of
the screen. The liquid portion of the waste passes through
the screen, and the strained solids are allowed to roll down
to the base of the screen [21] . The solids are collected in
a relatively dry form. The wires making up the screen are
placed in the horizontal direction with a spacing between
290 to 1,600 microns.
The system was designed to remove relatively durable solids
larger than the screen opening. To prevent clogging, the
solids collected require some form of transportation other
247
-------
than pumping if resuspension in water is to be avoided.
This is one of the screening methods currently being tested
for combined storm overflows at Fort Wayne, Indiana [28, 43]
The installed units are to handle 767 I/sec (17.5 mgd)
using screens with openings of 1,525 microns (0.060 inch).
Advantages and Disadvantages - The four basic screening de-
vices have been developed to serve one of two types of
applications. The microstrainer is designed as a main
treatment device that can remove most of the suspended con-
taminants found in a combined sewer overflow. The other
three devices --drum screens, rotary fine screens, and hy-
draulic sieves — are basically pretreatment units designed
to remove the coarser material found in waste flows. The
advantages and disadvantages of each type are listed in
Table 46.
Description of Demonstration Projects
Philadelphia, Pennsylvania — The use of a microstrainer to
treat combined sewer overflows has been studied in
Philadelphia [26, 27]. The facility includes microstrain-
ing and disinfection. The microstrainer was a 5-foot diam-
eter by 3-foot long unit using either 35- or 23-micron
screen openings during the various tests conducted. The
drum was operated submerged at 2/3 of its depth. The com-
plete unit was equipped to automatically control drum speed
proportional to the headloss across the screen, with con-
tinuous backwash, and with an ultraviolet irradiation source
to prevent fouling of the screen by bacterial slimes. The
unit starts automatically whenever sufficient overflow
occurs. Because of the physical configuration of the sewer,
flow was pumped to the microstrainer. However, it is rec-
ommended that pumping be avoided whenever possible since
large solids that would be readily removed by microstraining
are broken up by the pumping. The study was conducted
in three phases: (1) operation of full screen area using
the 35 micron screen, (2) operation at full screen area
using 23 micron screen, and (3) operation at 20 percent
of the screen area using the 23 micron screen. The latter
was to test increased loading rates since the facility
had a limited pumping capacity. The facilities operated
approximately 40 times per year on combined sewer overflows.
Milwaukee and Racine, Wisconsin [40] — The use of fine
screens to remove most of the coarse solids at Milwaukee
and Racine has been briefly described previously under
Dissolved Air Flotation. One unit was used at Milwaukee
and six are used at Racine. They operate at a continuous
248
-------
Table 46. CHARACTERISTICS OF VARIOUS TYPES
OF SCREENS
Microstrainer
Drum
screen
Rotary fine
screen
Hydraulic
sieve3
Principal use
Approximate removal
efficiency, \
BOD
SS
Land requirements ,
sq ft/mgd
Cost, $/mgd
Can be used as a dry
weather flow polish-
ing device
Automatic operation
Able to treat highly
varying flows
Removes only par-
ticulate matter
Requires special
shutdown and
startup regimes
Screen life with
continuous use
Uses special sol-
vents in backwash
water
High solids concen-
trate volume, \ of
total flow
Main treatment Pretreatment Pretreatment Pretreatment to
to other de- to other de- other devices
vices and vices and
main treat- main treat-
ment ment
50
70
- 15-20
12,000
Yes
Possible
with con-
trols
Yes
Yes
Yes
7-10 yr
No
0.5-1.0
15
40
15-20
4,800
No
Possible
with con-
trols
Yes
Yes
Some
10 yr
No
0.5-1.0
15
35
24-62
8,000
No
20
5,600
No
Possible No controls
with con- needed
trols
Some limita- Yes
tion
Yes
Some
1,000 hr
Yes
10-20
Yes
No
20 yr
No
< 0.5
a. Information on hydraulic sieves is limited. Formal study on treatment of
combined sewer overflows is just beginning.
b. Based on a 25-mgd plant capacity.
Note: sq ft/mgd x 2.12 = sq m/cu m/sec
$/mgd x 0.38 = $/cu m/sec
249
-------
drum speed with backwashing activation whenever headless
exceeds 6 inches. Collected solids are discharged to the
float holding tanks. The screen size used in both cases
was 297 microns.
Cleveland, Ohio [221 - The Cleveland, Ohio, study on dual-
media filtration also included a fine screen as a pretreat-
ment unit to the filtration process. The 420 micron screen
was fitted over a 1.2 sq m (12.6 sq ft) drum unit. Drum
speed and backwash conditions were not reported. More de-
tails on the layout of the facilities are given in this
section under Filtration.
East Providence, Rhode Island [411 - This bench-scale study
was conducted to test the applicability of using a drum
screen and a diatomaceous earth filter in series to achieve
significant removals when operating on combined sewer
overflow. The study indicated good removals by the screen-
ing device in relation to other drum screens. The screening
unit, however, was of different configuration than other
drum screens. The device used was a small 259 sq cm (40
sq in.) unit consisting of a submerged rotating drum with
the flow passing through the screen from the outside to the
inside. Effluent was drawn off from the interior of the
rotating drum. The backwash system ran continuously using
submerged spray jets directed at the interior of the screen
dislodging strained solids and allowing them to pass through
ports separating dirty water from the rest of the influent
water. Synthetic sewage was used during the study. Screen
apertures tested were 150, 190, and 230 microns in size.
Portland, Oregon [38. Ill - A rotary fine screen unit was
tested in Portland, Oregon, on both dry-weather flow and
combined sewer overflows. The facility was constructed on
a 183 cm (72-inch) diameter trunk sewer serving a 10,000-ha
(25,000-acre) area. A portion of the flow was diverted to
a bypass line where it first flowed through a bar screen
before being lifted into the demonstration project by two
132 I/sec (2,100 gpm) turbine pumps. After passing through
the rotary fine screen, both the concentrated solids and
the effluent were returned to the Sullivan Gulch Pumping
Station wet well. In a typical installation on a combined
sewer overflow line, the effluent from the screens would
pass to a receiving stream after disinfection. The concen-
trated solids would be returned to an interceptor sewer
The screening unit used a 152 cm (60-inch) diameter drum
with 74, 105, and 167 micron screens. The units were oper-
ated at flow rates ranging from 43.2 to 126.2 I/sec (1 to
2.8 mgd). The range and levels of variables tested is
listed in Table 47.
250
-------
Table 47. RANGE AND LEVEL OF VARIABLES TESTED [38]
Level of best
Variable Range investigated performance
Screen material Dacron cloth, market grade Tensile bolting cloth
Stainless steel fabric
Tensile bolting cloth
Screen mesh 105 to 230 165 (105 micron opening)
(167 to 74 micron opening)
Screen rotational speed, rpm 30 to 60 60
Influent flow rate, gpm 700 to 2,000 1,700
Screen hydraulic loading, 50 to 143 122
gpm/sq ft
Velocity of feed water 3 to 12 11
striking screen, ft/sec
Type of operation Intermittent to continuous 4-1/2 min ON, 1/2 min
OFF for backwash
Backwash ratio (gal. back- 0.2 to 25.6 0.235
wash water/1,000 gal.
applied waste)
Note: gpm x 0.0631 = I/sec
gpm/sq ft x 0.679 = 1/sec/sq m
ft/sec x 0.305 = m/sec
gal. x 3.785 = 1
After several modifications of the screening units, the
present configuration was finalized. In the final form the
reported performance criteria were [38]:
Floatable solids removal 100%
Settleable solids removal 98%
Total suspended solids removal 34%
COD removal 27%
Screened effluent as % of influent 92%
Fort Wayne, Indiana [28] — Fort Wayne is a newly constructed
screening facility designed to compare three types of
screens: (1) fine screen, (2) rotary fine screen, and
(3) static screens. The three units operate in parallel
with first and last units each handling 767 I/sec (17.5
mgd) and the rotary fine screen handling 1,752 I/sec
(40 mgd) for a total of 3,286 I/sec (75 mgd). The fine
screen is a 3.7-meter by 3.7-meter (12-foot by 12-foot)
unit with a 147-micron screen. Some headloss controls are
provided. The rotary fine screen portion of the project
251
-------
has eight 152-cm (60-inch) diameter units operating in
parallel. They are to be operated sequentially to accommo-
date flow variation. The screen size is 105 microns.
Twelve static screens using 1,525-micron (0.06-inch) clear
opening screen represent the third portion of the facility.
These are the manufacturer's standard units that have been
used in industry to remove gross solids. A description of
a typical unit was presented above. The combined sewer over-
flow facilities are located across the Maumee River from
Fort Wayne's sewage treatment plant. Flows entering the
facilities are sewage treatment plant bypass and combined
sewer overflows. These flows are lifted to the screens by
pumps after passing through a bar screen. Chlorination and
a contact tank are provided.
Costs
Microstrainers and Drum Screens - The costs reported for
microstrainers vary considerably, as shown in Table 48. The
main reason is the variation in flux rates or loading coupled
with the type of waste treated (i.e., combined sewer over-
flows versus secondary effluent) [30]. With the exception
of the Philadelphia facility, all of the microstrainers are
used to treat sewage effluent at appreciably lower flux
rates which necessarily increased the cost. During the
Philadelphia study it was found possible to use a flux rate
of 73.3 cu m/hr/sq m (30 gpm/sq ft); therefore, the costs at
the three other locations listed in Table 48 have been modi-
fied to reflect this increase in loading rate. According to
the figures presented in Table 48, the average capital cost
is approximately $248/l/sec ($11,000/mgd) for treating com-
bined sewer overflows. The operation and maintenance costs
have not been adjusted. The approximate cost is $0.0013 to
$0.0026/1,000 1 ($0.005 to $0.01/1,000 gal.) for assuming 300
hours of operation per year. The single capital cost cited
for a fine screen is only the equipment cost and does not
include installation. Operation and maintenance costs
should be comparable to those for microstrainers.
Rotary Fine Screens - Cost data for rotary fine screens for
combined sewer overflows are based on a preliminary design
estimate for a screening facility in Seattle, Washington,
and actual construction costs at Fort Wayne, Indiana [38,
28]. The two costs were $700,000 and $250,000, for plants
of 1,095 I/sec (25 mgd) and 1,640 I/sec (37.5 mgd) ,
respectively. The differences in cost are due, in part, to
the fact that the Fort Wayne installation is a demonstration
prototype project where three types of screens operating in
parallel are treating a total flow of 3,285 I/sec (75 mgd)
in a single building. The cost for the rotary fine screen
252
-------
Table 48. COST OF MICROSTRAINERS AND
FINE SCREENS FOR 25-MGD PLANTSa
Loading Operation and
rate, Modified maintenance cost
Influent gpm/ Capital capital
source sq ft costb costc Annual
-------
FILTRATION
Introduction
In the physical treatment processes, filtration is one step
finer than screening. Solids are usually removed by one or
more of the following removal mechanisms: straining, im-
pingement, settling, and adhesion. Filtration has not been
used extensively in wastewater treatment, because of rapid
clogging which is principally due to compressible solids
being strained out at the surface and lodged within the
pores of the filter media. In stormwater runoff, however,
a large fraction of the solids are discrete, noncompressible
solids that are more readily filtered [30].
Effluents from primary or secondary treatment plants and
from physical-chemical treatment facilities contain com-
pressible solids.
The discussion on filters handling discrete, noncompressible
solids is presented here.
Design Criteria
Two factors affecting removal efficiency are flux rate and
the type of solids. As one would expect, the removals are
inversely proportional to the flux rate. At high flux
rates, solids are forced through the filters reducing solids
removal efficiency. Suspended solids removals were found
to be better for inert solids (discrete, noncompressible
solids) than for volatile solids (compressible solids).
This is the same conclusion found for microstrainers.
Loading Rates — The difference between filtering compres-
sible and noncompressible solids is basically the flux rate
used. High-rate filters handling compressible solids are
normally loaded at 12.2 to 24.5 cu m/hr/sq m (5 to 10 gpm/
sq ft), whereas those handling noncompressible solids will
filter at rates up to 73.4 cu m/hr/sq m (30 gpm/sq ft).
Chemicals — Many polyelectrolytes and some coagulants have
been tested. Some polyelectrolytes have been found which
increase removals of phosphorus and nitrogen. It is
cautioned, however, that polyelectrolytes are noted for
their unpredictability and the most effective polyelectro-
lyte must be determined for each wastewater.
Demonstration Projects
Studies have been made to investigate possible filtration
techniques for combined sewer overflows. The different
254
-------
methods attempted and general comments as to their success
are listed in Table 49. The most successful was dual-media
filtration using anthracite coal over sand with a fine
(420-micron) screen as a pretreatment device. The use of
the screen reduced the solids loading on the filter giving
longer runs. In effect, this was equivalent to a multi-
media filter.
Cleveland, Ohio — The successful combined sewer overflow
filtration study conducted at Cleveland, Ohio, was the first
to use a dual-media filter preceded by a fine screen on non-
industrial wastewaters [22]. The filters had a 1.22-meter
(4-foot) layer of anthracite coal with an effective size of
4 mm over a 0.92-meter (3-foot) layer of 2-mm effective size
sand. The three filters were constructed using 15.2-cm
(6-inch) diameter lucite tubes. The test procedure was to
pump from an outfall all combined sewer overflows that
occurred. The flows were treated by the 420-micron fine
Table 49. TYPES OF FILTRATION PROCESSES
INVESTIGATED FOR COMBINED SEWER OVERFLOWS
Filter description
Type of
Ref. No. facility
Remarks
Single-media filtration
Mixed-media filtration
With fine screen [22]
pretreatment
Without pretreatment [24]
Coal filtration [34]
Fiber glass plug [24]
filtration
Diatomaceous earth [41]
filtration
Upflow filtration with [24]
garnet sand
Ultrasonic filtration [42]
using fine screens that
are ultrasonically
cleaned
Crazed resin filtration
[10]
Easily clogged, acts
like a strainer.
Pilot plant Successful study with
50-904 removals.
Bench model Short runs, lower
removals.
Prototype Good for coarse solids.
Bench model Partly successful, needs
more work to verify the
results.
Bench model Effective but costly.
Bench model Unsuccessful.
Pilot plant Unsuccessful.
Pilot plant Unsuccessful.
255
-------
screen during the overflow event and stored in two 19-cu m
(5,000-gal.) tanks for the test filtration runs that
followed. Each tank had a mixer to keep solids in
suspension. Two pumps were then used to supply the filter
with screened water.
Removals for this filter were 65 percent for SS, 40 percent
for BODs, and 60 percent for COD [22]. The addition of
polyelectrolyte increased the SS removal to 94 percent, the
BODs removal to 65 percent, and the COD removal to 65 per-
cent. Inorganic coagulants, such as lime, alum, and ferric
chloride, did not prove as successful as polymers. Run
times averaged 6 hours at loading rates of 58.7 cu m/hr/sq m
(24 gpm/sq ft). Backwashing of the filters consisted of
alternately injecting air and water into the bottom of the
filter columns. Air volume was varied from 38.4 to
283 cu m/hr/sq m (2.1 to 15.5 scfm/sq ft) over 2.5 to 29
minutes. Backwash water volume used ranged from 1.9 to
8.6 percent of the total combined sewer overflow filtered,
with a median value of approximately 4 percent. The range
of backwash water rate used was 75.8 to 220 cu m/hr/sq m (31
to 90 gpm/sq ft) over 4 to 25 minutes.
A list of the basic design data is presented in Table 50.
Others - Two other filtration processes, fiber glass plug
filtration [24] and coal filtration [34], show some promise,
but additional research is necessary to perfect them. Other
methods,such as crazed resin filtration, upflow filtration
with garnet sand, and filtration using ultrasonically
cleaned fine screens,have not been successful and are not
considered worthy of further effort at the present time.
Advantages and Disadvantages
The advantages of dual-media filtration are that (1) rela-
tively good removals can be achieved; (2) process is versa-
tile enough to be used as an effluent polisher; (3) operation
is easily automated; and (4) small land area is necessary.
Disadvantages are that (1) costs are high; (2) dissolved
materials are not removed; and (3) storage of backwash water
is required.
Costs
Cost data were developed from a design estimate for 1.1, 2.2,
4.4, and 8.8 cu m/sec (25, 50, 100, and 200 mgd) filtration
plants at a satellite location [22], The basic plant as en-
visioned for the cost estimate includes a low lift pump
256
-------
Table 50. DESIGN PARAMETERS FOR FILTRATION MIXED
MEDIA, HIGH RATE [22]
Filtering media 4 ft anthracite coal
3 ft sand #612
Effective size, mm
Anthracite coal 4
Sand 2
Flux rate, gpm/sq ft
Design 24
Range 8.40
Headless, ft 5.30
Backwash
Volume, $ of inflow 4
Air (rate and time),
scfm/sq ft, min IQ, 10
Water (rate and time),
gpm/sq ft, min 50, 20
Filtering aid
Pol/electrolyte, type Anionic
Pretreatment 420 micron ultrafine
screen
Note: gpm x 0.679 = 1/sec/sq m
ft x 0.305 = m
scfm/sq ft x 0.305 = cu m/min/sq m
station, fine screens, chlorination facilities, and dual-
media filters with the same configuration as used at
Cleveland. Filters would be designed to operate at 58.6
cu m/hr/sq m (24 gpm/sq ft). Effluent water would be used
for backwashing. Collected solids from the screens and
filters would be returned to the interceptor leading to the
sewage treatment plant. The building housing the facilities
would be mostly above ground and would be designed to be
highly automated, capable of coming on-line automatically
with automatic backwash capability. Using this as the basis
for the cost estimate and an assumed figure of 300 hours of
operation per year to handle combined storm overflows, the
capital and operation and maintenance cost would be as shown
on page 258.
257
-------
Plant capacity Operation and
Capital maintenance
cu m/sec mgd cost, $ cost, $/yr
1.1 25 1,580,000 44,000
2.2 50 2,390,000 55,000
4.4 100 4,370,000 98,000
8.8 200 7,430,000 129,000
The cost data are based on an ENR of 2000.
The operating costs are estimated to be $0.0382/1,000 1
($0.141/1,000 gal.) for 300 hours of operating per year. Th©
high cost could easily be reduced, however, by designing the
system to serve also as a dry-weather effluent polisher dur-
ing periods with no storm flows.
CONCENTRATION DEVICES
Concentration devices, such as the swirl regulator/
concentrator and helical or spiral flow devices, have intro-
duced an advanced form of sewer regulator--one capable of
controlling both quantity and quality. These devices have
been previously described in Section VIII. A prototype
swirl regulator has recently been constructed in Syracuse,
New York. A second generation swirl concentrator has
been placed into operation as a treatment unit for municipal
sewage grit separation in Denver, Colorado. Settleable
solids removals ranging from 65 to more than 90 percent,
corresponding to chamber retention times of approximately 5
to 15 seconds, have been predicted on the basis of hydraulic
model tests. At the time of writing, no operational data
were available. Indicated costs are approximately $285/cu m/
sec ($6,500/mgd). A third generation swirl device has been
developed to take the place of conventional primary sedi-
mentation at 10 to 20 minute detention times.
258
-------
Section XI
BIOLOGICAL TREATMENT
INTRODUCTION
Contaminants in sewage and stormwater can be removed by
physical, chemical, and biological means. As noted previ-
ously, physical treatment involves the removal of contami-
nants through the controlled application of physical forces,
as in the case of gravity settling or centrifugation.
Unfortunately, because many of the contaminants in sewage
are colloidal in size or dissolved, they cannot be removed
by physical means. The removal of these contaminants is
normally accomplished by chemical or biological means.
In biological treatment, the objective is to remove the
nonsettleable colloidal solids and to stabilize the dis-
solved organic matter. This is normally accomplished by
biologically converting a portion of the organic matter
present in the sewage into cell tissue, which subsequently
can be removed readily by gravity settling. The energy re-
quired for the synthesis of cell tissue is obtained from the
oxidation of that portion of the organic matter not used for
the synthesis of cell tissue. In general, biological treat-
ment can be accomplished aerobically (in the presence of
oxygen) or anaerobically (in the absence of oxygen).
Typically, aerobic processes are used for the conversion of
the organic matter in sewage to cell tissue, and anaerobic
processes are used for the conversion of the cell tissue
produced to stabilized end-products. In the later case, the
cell tissue serves as the food source for the anaerobic
bacteria.
The kinetics of biological conversion and growth are gov-
erned by the following factors: (1) the rate of substrate
(food) utilization, (2) the growth rate of the organisms in
the system, (3) the mass of organisms in the system, (4) the
length of the contact time between the waste and the orga-
nisms, (5) types of organisms, and (6) environmental condi-
tions, such as temperature, pH, nutrients, etc.
259
-------
Operationally and from a design standpoint, the aforemen-
tioned factors are taken into account by considering (1) the
food-to-microorganism ratio, (2) the sludge retention time,
and (3) the hydraulic detention time.
The food-to-microorganism ratio is defined as the kilograms of
BOD5 (food) applied per unit time (often taken as the amount
consumed) per kilogram of organisms in the system. The sludge
age is defined by the kilograms of organisms wasted per day.
The hydraulic detention time is defined as the value, given in
units of time, obtained by dividing the volume of the reaction
vessel by the flow rate.
Because the food-to-microorganism ratio and the sludge re-
tention times are interrelated [17] , both are commonly used
in the design of biological systems. From field observa-
tions and laboratory studies, it has been found that as the
sludge age is increased and, correspondingly, the food-to-
microorganism ratio decreased, the settling characteristics
of the organisms in the system are enhanced, and they can be
removed easily by gravity settling. Typical values for the
food-to-microorganism ratio and sludge age are given in
reference [17].
As previously noted, the length of time the biomass is in
contact with the waste BODs is measured by the hydraulic de-
tention time. The minimum time to achieve a given removal
is dependent upon the food-to-microorganism ratio. Low
ratios (i.e., a high number of bacteria per kilogram of BODr)
allow faster utilization of a given amount of BODs. The
minimum time required may vary considerably, from 10 to 15
minutes in contact stabilization, or less for trickling
filters and rotating biological contactors, and up to 2 to
3 days for oxidation ponds. At the shorter contact times,
the biomass only removes the dissolved matter and possibly
some of the smaller colloidal matter [15]. At longer con-
tact times, suspended organic matter is utilized.
In any biological system, these factors control the process.
A mathematical model has been developed for the activated
sludge system [17, 14]. Models for trickling filters,
rotating biological contactors, and treatment lagoons have
not been formulated. Empirical designs and design param-
eters are used instead.
APPLICATION TO COMBINED SEWER OVERFLOW TREATMENT
Biological treatment of wastewater, used primarily for domes-
tic and industrial flows of organic nature, produces an
effluent of high quality and is generally the least costly
260
-------
of the process producing similar effluents. For combined
sewer overflows, however, it has one serious drawback: the
biomass used to assimilate the waste constituents found in
waste flows must be either (1) kept alive during times of
dry weather or (2) allowed to develop for each storm event.
Also, biological treatment processes are upset by erratic
loading conditions.
Two methods have been used to keep the biomass alive between
storms at combined sewer overflow treatment facilities. The
first method is to construct the wet-weather treatment facil
ities next to a dry-weather plant and to use the excess
biomass therefrom as required. Contact stabilization of
combined sewer overflows at Kenosha, Wisconsin, operates in
this manner. The second method is to have a treatment pro-
cess which can be used to treat wastewater with a high
variation in flow rate and strength. During dry weather, it
would be used for domestic flows and combined sewer over-
flows during wet weather. Trickling filters and rotating
biological contactors (at least in EPA demonstration proj-
ects) are in this category. Storage of the biomass either
in a tank in suspension or on a supporting medium by supply-
ing air and no substrate is not an effective method [24].
The other approach is to store the combined sewer overflows
for an extended period of time and allow the biomass to grow
large enough to treat the stored flows successfully.
Treatment lagoons are an example of this approach.
The^more sophisticated schemes, such as the contact stabili-
zation modification of activated sludge, trickling filters,
and rotating biological contactors, provide good treatment*
of the overflows, but all should include storage or equali-
zation ahead of them to prevent hydraulic overloading with
attendant biomass washout during wet weather. This adds to
the cost. Less sophisticated methods, such as oxidation
lagoons, aerated lagoons, and facultative lagoons, require
less attention, and the lagoons can act as storage units.
Although they may not need dry-weather flow to keep them in
operation, they do require more land.
Some other important considerations in the biological treat-
ment of combined sewer overflows are:
1. Most biological treatment processes are familiar
to designers and operators even though their appli-
cation to combined sewer overflow treatment is not.
2. Some of the treatment methods may require pretreat-
ment and entail a continuous cost even during dry
weather to keep the biomass alive.
261
-------
3. Shakedown runs are necessary to keep the units and
the usual large number of automatic controls in
operating order.
CONTACT STABILIZATION
Description of the Process
Contact stabilization is considered in lieu of other acti-
vated sludge process modifications for treating combined
sewer overflows, because it requires less tank volume to
provide essentially the same effluent quality. The over-
flow is mixed with returned activated sludge in an aerated
contact basin for approximately 20 minutes at the design
flow. Following the contact period, the activated sludge
is settled in a clarifier. The concentrated sludge then
flows to a stabilization basin where it is aerated for
several hours. During this period, the organics from the
overflow are utilized in growth and respiration and, as
a result, become "stabilized." The stabilized sludge is
then returned to the contact basin to be mixed with the in-
coming overflow. A schematic of a contact stabilization
plant for treating combined sewer overflows is shown on
Figure 50.
Demonstration Project, Kenosha, Wisconsin
A project sponsored by the EPA to evaluate the use of con-
tact stabilization for treatment of combined sewer overflows
from a 486-ha (1,200-acre) tributary area is presently under-
way at Kenosha, Wisconsin [23, 19]. It is an example of how
contact stabilization can be used to treat combined sewer
overflows using the waste activated sludge from a dry-weather
activated sludge plant. At the Kenosha municipal sewage
treatment plant, a 101-1/sec (23-mgd) facility, a new com-
bined sewer overflow treatment facility was constructed.
This facility consists of an aeration tank, a contacr sta-
bilization tank, and a new clarifier. The design capacity
of the new facility is 88 I/sec (20 mgd). The stabilization
tank, acting as the biosolids reservoir, receives the waste
activated sludge from the main plant. This sludge is held
for up to 7 days before final wasting. Thus, stabilized
activated sludge is kept in reserve ready to treat combined
sewer overflows when they occur. Photographs of the facil-
ity are shown on Figure 51.
Operation of the contact stabilization plant consists of
directing the combined sewer overflow to the contact tank
following comminution and grit removal, adding the reserve
activated sludge, and then conducting the waste flows to a
final clarifier for separation of the biosolids and other
262
-------
COMNINUTORED INTERCEPTOR FLOWS
*
PUMP STATION
COMBINED SEWER EXCESS FLOWS
DESIGN Q - 20 MGD
NORMAL FLOW TO
DWF FACILITIES
(23 MGD MAX)
GRIT CHAMBER
VELOCITY - 0.2 FPS
VOLUME - 33,000 CF
DETENTION TIME «
14 MIN @ DESIGN IF
Q+25« RECYCLE
TO SLUDGE THICKENER
FOR FINAL WASTING
SLUDGE PUMPS TO TRANSFER
STABILIZED SLUDGE
TO CONTACT TANK
CONTACT
TANK
TANK
OVERFLOW RATE -
1.280 GPD/SQ FT
DIAMETER-MO FT
WET
WEATHER
FINAL
CLARIFIER
TANK DIVIDED INTO
TWO COMPARTMENTS
OF 49,000 CF EACH
SLUDGE
RETURN
SLUDGE WASTING
LINE FROM DWF PLANT
EFFLUENT TO EXISTING
DWF CHLORINATION STATION
THEN TO LAKE MICHIGAN
NOTE:
MGD x 43.8 - L/SEC
FPS x 0.305 - M/SEC
CF x 28.3 - L
GPD/SQ FT x 0.183 —
FT x 0.305 - M
CU M/MIN/HA
Figure 50.
Contact stabilization plant [23]
Kenosha, Wi sconsin
263
-------
(0
(e)
Figure 51. Combined sewer overflow treatment
by contact stabilization (Kenosha)
(a) Contact tank with diffused air (b) Sludge stabilization tanks with floating
aerators (c) Floati.ng aerator anchoring and counterweight details (d) Closeup
of aerator operation (e) Final contact tank (peripherally fed) and effluent
264
-------
suspended matter from the final effluent. The solids are
returned to the stabilization tank where they are aerated
before being pumped back to the contact chamber. The plant
has been constructed so that additional activated sludge
may be obtained from the main plant, if necessary. During
dry weather, the final clarifier is used for added clarifi-
cation, and consideration has been given to using the
stabilization compartment for aerobic digestion of the waste
sludge.
Efficiency - The BOD5 and SS removals achieved during 1972
at Kenosha's combined sewer overflow treatment facilities
were 83 and 92 percent, respectively. The overall removals
are summarized in Table 51.
Removals under steady-state conditions are directly related
to sludge retention time (sludge age), food-to-microorganism
Table 51. CONTACT STABILIZATION REMOVALS
DURING 1972 [23]
Weighted mean a
Parameter
Total solids
Total volatile solids
Suspended solids
Suspended volatile
solids
Total BOD5
Dissolved BOD5
Total organic carbon
Dissolved organic
carbon
Total Kjeldahl - N
Total P04 - P
In, mg/i
704
270
314
121
102
24.1
113
21.8
11.0
4.8
Out, mg/l
455
140
26.4
15.2
17.8
7.6
22.8
15.3 -
5.5
2.4
$ Removal
35
48
92
87
83
68
80
30
50
50
Geometric mean
Total coliform
Fecal coliform
In, no. /ml
34,786
2,308
Out, no. /ml |
2,883
374
Removal
91
83
a. Data from 23 storm events.
265
-------
ratio, and detention times in the contact and stabilization
tanks [17, 24, 15, 14]. In the work at Kenosha, however, it
has not been possible to show any correlation between re-
movals and these items, although it has been shown that
operation based on an assumed uniform influent BODs is suf-
ficient for good BODs and SS removals (80 and 90 percent,
respectively).
Operating Parameters — With contact stabilization or any
other activated sludge process, operation is normally based
on the food-to-microorganism ratio or sludge retention time.
Because of this, difficulties may be encountered when using
an activated sludge process for treating a rapidly varying
and intermittent flow. The sludge retention time is particu-
larly difficult to control because overflows may not last
long enough for the plant to stabilize and for proper
wasting procedures to be instituted. Operating the plant on
stored overflows could reduce this problem The food-to-
microorganism ratio, which is interrelated to the sludge age,
can be used to control the operation of the plant; however,
it too is difficult to control since the concentration of
both the incoming BODs and the biological solids in the sys-
tem must be known. This is further complicated because the
BODs concentration in the combined sewer overflow may vary
significantly. Based on the results at Kenosha, it has been
found that exact control is not necessary for good operation.
The operating parameters used for the contact stabilization
plant at Kenosha are shown in Table 52. The values reported
are averages, and the range was generally within ±60 percent
of the value listed. For comparison, the design parameters
for sewage treatment by contact stabilization found in the
literature are also presented.
For units such as that at Kenosha, sophisticated design may
not be warranted because the system is operated for such
short periods that the biosolids and the kinetics of the
system do not have a chance to adjust to the incoming flow
before the storm is over. In this case, using the reported
design equations should be sufficient. Abatement plans that
include a contact stabilization process for the treatment of
stored overflows for periods of time greater than 5 to
10 days may warrant more sophisticated design to achieve
higher removal efficiencies. The use of the kinetic equa-
tions describing the metabolism of the bacteria, as formu-
lated by McCarty [14], Metcalf $ Eddy [17], and others, may
prove useful under such circumstances.
Results of Operational Tests — The work at Kenosha has not
been able to show any adverse condition that affects
removals. Based on the results of 23 storms studied, the
266
-------
Table 52. CONTACT STABILIZATION OPERATING PARAMETERS
Range of values
Kenosha, reported in
Parameter wis. [23] the literature
Sludge retention time,
day 4a 4_15
Food-to-microorganism ratio,
F/M, Ib BOD5/lb MLSS/day
Contact tank and stabili-
zation tank o.21 0.15-0.6
Contact tank alone 2.66
Loading rate,
Ib BOD5 /l.OOO cf 108 30-12S
MLSS, mg/1
Contact tank 3,000 1,000-3,000
Stabilization tank 12,000 4,000-10,000
Detention time, hr
Contact tank Q.25 0.25-1.0
Stabilization tank 3 3.5
Recycle ratio, Qr/Q .12-.58 0.25-1.0
Volume of air required,
cf/lb BOD5
Contact tank 470
Stabilization tank no 1,000-1,200
BOD5 removal, \ 83 80-90
a. Controlled principally by time between storms.
Note: Ib BOD5/lb MLSS/day x 1.0 = kg BOD5/kg MLSS/day
Ib BOD5/1,000 cf x!6.02= g BOD /I,000 cum
cf/lb BODS x 62.4 = I/kg BOD$
number of data points was not sufficient to produce signifi-
cant correlations between the many parameters tested (i.e.,
optimum food-to-microorganism ratio, sludge retention time,
detention times in contact and stabilization tanks, system
efficiencies as a function of time after rainfall, tempera-
ture, optimum startup and shutdown, etc.) [23]. The optimum
values for these parameters should become known as testing
continues.
Some degradation in the settling characteristics of the
sludge was noted when the sludge age reached 10 days. The
completion of the study should provide more definitive
answers to the possible correlation among the various param-
eters mentioned above.
The only operational consideration reported is the type of
aeration system. Diffused aeration proved to be better than
mechanical aeration primarily because of accumulation of ice
on the mechanical aerators during freezing conditions (see
the discussion of Treatment Lagoons later in this section
for additional information on this problem).
267
-------
Advantages and Disadvantages
Some advantages of the contact stabilization process for the
treatment of excess (combined sewer) flows in this applica-
tion are: (1) high degree of treatment; (2) central location
of maintenance personnel and equipment; and (3) reduction of
the loadings on dry-weather facilities, by dual use of
facilities, during normal operations and emergency shutdown
of the main plant, making the whole very versatile. Contact
stabilization shows definite promise as a method for treat-
ing combined sewer overflows when used in combination with
a dry-weather activated sludge treatment plant. Disadvan-
tages are: (1) high initial cost, (2) the facilities must
be located next to a dry-weather activated sludge plant,
(3) adequate interceptor capacity must exist to convey the
storm flow to the treatment plant, and (4) expansion of
major interceptors may be required.
TRICKLING FILTERS
Description of Process
Trickling filters are widely employed for the biological
treatment of municipal sewage. The filter is usually a
shallow, circular tank of large diameter filled with crushed
stone, drain rock, or other similar media. Settled sewage
is applied intermittently or continuously over the top sur-
face of the filter by means of a rotating distributor and is
collected and discharged at the bottom. Aerobic conditions
are maintained by a flow of air through the filter bed in-
duced by the difference in specific weights of the atmos-
phere inside and outside the bed.
The term "filter" is a misnomer, because the removal of
organic material is not accomplished with a filtering or
straining operation. Removal is the result of an adsorption
process occurring at the surface of biological slimes cover-
ing the filter media.
Classification — Trickling filters are classified by hy-
draulic or organic loading. Until recently, there were only
two flow classifications: low rate and high rate. A third
classification, ultrahigh rate, has been added since the ad-
vent of plastic medium filters. Although the distinctions
are based on hydraulic loading, they are centered in reality
around the organic loading that the filter can handle. A
comparison of the three classifications of trickling filters
is presented in Table 53.
The type of medium used varies considerably. Rock, slag,
hard coal, redwood slats, and corrugated plastic have been
used. Rock, slag, and hard coal have relatively low surface
268
-------
Table 53. COMPARISON OF LOW-RATE, HIGH-RATE, AND
ULTRAHIGH-RATE TRICKLING FILTERS
Factor
Low-rate
High-rate
Utrahigh-rate
Hydraulic loading, mgad 1 to 4
300 to 1,000
Organic loading,
Ib BOD5/acre-ft/day
Depth, ft
Recirculation
Medium
Power requirements
Filter flies
Sloughing
Operation
Dosing interval
Nitrification
6 to 10
None
Rocka
None
Many
Intermittent
Simple
Not more than
5 min (gener-
ally inter-
mittent)
Fully
nitrified
10 to 40
1,000 to 5,000
3 to 8
1:1 to 4:1
Rocka
10 to 50 hp/
mil gal.
Few, larvae
are washed
away
Continuous
Some skill
Not more than
15 sec
(continuous)
Nitrification
at low
loadings
40 to 120
2,000 to 10,000
(and above)
20 to 40
1:1 to 4:1
Plastic or
redwood slats
None
Continuous
Some skill
Generally
continuous
Nitrification
at low
loadings
a. Media could also be slag or hard coal.
Source: Adapted from [17].
Note: mgad x 0.108 = cu m/sec/ha
Ib/acre-ft/day x 0.368 = g/cu m/day
ft x 0.305 = m
hp x 0.746 = kw
mil gal. x 3.785 = Ml
269
-------
area per unit volume and are quite heavy, thus limiting the
depth of filter. Redwood slats and corrugated plastic are
much lighter and can be constructed with a larger surface
area per unit volume.
Operation — The operation of most high-rate and ultrahigh-
rate trickling filters is in series with a second or third
filter and/or with recirculation. The purpose is to provide
high removals by increasing the contact time of the waste
with the biomass attached to the filter material. When
operating alone without recirculation, trickling filters
used for treating domestic wastes remove between 50 and
75 percent of the BODs.
Under storm conditions, the trickling filter must handle
highly varying flows. Applying a varying organic load to
a filter does not produce optimum removals. It is gener-
ally thought that only sufficient biomass adheres to the
supporting medium to handle the normal organic load. As the
loading increases above this level, the maximum BODs utiliza-
tion rate of the biomass is reached. This is not a sharp
distinction because some excess biomass always adheres to
the medium and can accept some of the organic load.
A varying hydraulic load also affects removals. The in-
creased shearing action of high flows causes excess slough-
ing or washing off of the biomass. To help dampen this
effect, filters operating in series under dry-weather condi-
tions can be operated in parallel, thereby reducing some of
the increased hydraulic load on each filter. A maximum
overall flow variation (maximum/minimum) of 8 to 10 is
acceptable while still achieving significant removals [20].
Design — Trickling filter design has been based primarily on
empirical formulas. This does not imply that the basic bio-
logical kinetics are not operative; rather, it means that
mathematical description of the process has not been
formulated. There are several design equations in the
literature that may be used for the design of trickling
filters [17, 6]. In designing a trickling filter to treat
overflows, it must be remembered that dry-weather flow is
needed to keep the biomass active between storms. Generally,
two or more units should be used to provide high removals by
operating in series during dry weather and in parallel dur-
ing storm events to accommodate the flow variation needed.
Demonstration Project, New Providence, New Jersey
Trickling filters have been used extensively throughout the
United States to treat domestic flows, but only one facility
(at New Providence, New Jersey) has been designed to treat
270
-------
both dry-weather flow and combined sewer overflows from
heavily infiltrated sanitary sewers. This plant represents
one of the configurations that can be used to treat combined
sewer overflows. The plant, shown schematically on
Figure 52, is designed for a dry-weather flow of 26.3 I/sec
(0.6 mgd) and a maximum wet-weather flow of 263 I/sec (6.0
mgd). The plant includes a main pump station, primary and
secondary clarifiers, two trickling filters, chlorine con-
tact tank, administration building, and other miscellaneous
items. One of the filters, 11 meters in diameter by 4.4
meters high (36 feet by 14.4 feet), is packed with plastic
medium and the other, 19.8 meters in diameter by 1.8 meters
(65 feet by 6 feet), is packed with stone. Photographs of
the plant are presented on Figure 53.
The unique feature of the plant is its method of filter
operation designed to keep a live biomass available on both
filters all of the time. To do so, the filters operate in
series with the plastic medium .filter in lead position,for
treating all flows up to 123 I/sec (2.8 mgd). At this point,
an automatic transfer to parallel operation is accomplished
and maintained until flows again drop within the series
range. In parallel operation, the normal combined sewer
overflow treatment mode, both filters receive equal flow,
resulting in a much higher (3 to 1) unit loading on the
smaller plastic medium filter.
Efficiency — The removals have been reported to be 85 to
95 percent for both BODs and SS during dry-weather flow and
65 to 90 percent during wet-weather flow. There was no sig-
nificant removal of total nitrogen or phosphorus. The load-
ings and the average removals for the first two years of
operation are presented in Table 54.
During combined sewer overflows, it has been found that the
removal efficiencies dropped when the hydraulic loading in-
creased above 1.56 cu m/hr/sq m (40 mgad) for the plastic
medium and above 0.48 cu m/hr/sq m (12.2 mgad) for the rock
trickling filter. Also, the change from series operation to
parallel operation reduces removals as recirculation is
stopped and the waste flows pass through only one filter
before being discharged. Although this reduction occurred,
both filters recovered rapidly returning to dry-weather re-
moval rates at the end of the storm. It may also be possible
to improve removals by reducing the difference between dry-
weather flow loading and wet-weather flow loading. The plant
was originally designed for an increase of 10 to 15 times dry-
weather flow. An overflow rate in the secondary clarifier of
2.65 cu m/hr/sq m (1,560 gpd/sq ft) was found to be too high
during peak flows to achieve good removals.
271
-------
COMMINUTORED INTERCEPTOR FLOWS
SLUDGE GRINDER
PLASTIC
TRICKLING
FILTER
ROCK
TRICKLING
FILTER
EXCESS DRY-WEATHER
FLOW TO SUMMIT
FOR TREATMENT
2ND STAGE
WET WELL
s
/
^\
>
WEIR
CHLORINE CONTACT
TANKS
EFFLUENT TO
PASSAIC RIVER
Figure 52.
Trickling fiIter plant schematic
New Providence, N.J.
272
-------
(c)
Figure 53. Combined sewer overflow treatment by
high-rate trickling filtration (New Providence)
(a) Composite view of plant showing relative elevations of facilities (from left to
right: pumping station, primary cI arifier/storage tank, plastic media filter, rock
media filter) (b) High-rate application on rock media (c) Composite view from
top of primary cI arifier/storage tank (final clarifier is at right rear)
273
-------
Table 54. TRICKLING FILTER REMOVALS [20]
NEW PROVIDENCE, NEW JERSEY
treated elation
flow, rate
BOD
SS
mgd
mgd
- Trickling filters _ overall"
Influent, Effluent, Removal, removal,
mg/1 mg/1 % t
_ Trickling filters _ Overall3
Influent, Effluent, Removal, removal,
mg/1 mg/1 $ I
Dry weather flow
First year 0.54
Part of
second year
0.56
Wet weather flow
First year 3.96b
Part of
second year 1.72
O.to 0.8
86
94
86
87
93
a. Includes removals by primary sedimentation.
b. Average wet weather flow; average peak flows were 6.0 mgd with no recirculation.
c. Wet weather flow rate was reduced by approximately 1.5 mgd by pumping to another treatment plant.
Note: mgd x 43.8 = I/sec
In comparing the plastic medium and the rock filter, it was
noted that up to 2-1/2 times the BODs removal per unit vol-
ume was possible with the plastic medium. Also, on a capi-
tal cost basis, the plastic medium outperformed the rock
by 2 to 1 ($/kg BOD5 removed/1,000 cu m) .
Design Parameters — The average hydraulic and organic load-
ings applied to the New Providence facilities are slightly
above the recommended design values. The recommended values
are :
Plastic Medium
Rock
Hydraulic
loading
Organic
loading
2.73 cu m/hr/sq m
(70 mgad)
1.36 kg BODr/day cu m
(85 Ib BOD5/day/l , 000 cf or
3,700 Ib BOD5/acre-£t/day)
0.78 cu m/hr/sq m
(20 mgad)
0.64 kg BODr/day/cu m
(40 Ib BOD5/day/l ,000 cf or
1,742 Ib BOD5/acre-f t/day)
Additional design parameters were included previously in
Table 53.
Advantages and Disadvantages
Advantages of trickling filters include: (1) they handle
varying hydraulic and organic loads, (2) are simple to oper
ate, (3) have ability to withstand shockloads , and (4) have
274
-------
ability to recover rapidly from high flows. Trickling fil-
ters are not without disadvantages, however. They need a
continuous base flow to keep the biomass active, which is
most important in using them to treat combined sewer over-
flow, and removals decrease when high flow and BOD5 load-
ings are applied. Problems may be encountered when treating
more dilute combined sewer overflow or storm sewer discharge
ROTATING BIOLOGICAL CONTACTORS
Rotating biological contactors are a recent development in
biological treatment. The method was first developed in
Germany in 1955 to treat domestic wastes. Recently,
rotating biological contactors have been tried with combined
sewer overflows because of their reported ability to handle
highly varying flows. A description of an EPA-sponsored
demonstration project in Milwaukee, Wisconsin, is included
later in this section.
Description of Process
The rotating biological contactor is similar to a cross be-
tween a trickling filter and an activated sludge system. It
consists of a shaft supporting a set of rotating discs upon
which a biomass is grown and a shallow contact tank that
houses the shaft-disc assemblies, as shown on Figure 54.
ROTATING
DISCS •
BAFFLES
CONTACT TANK
TO FINAL
CLARIFIERS
Figure 54. Rotating biological contactor
275
-------
The rotating discs are partially submerged and baffles are
used between each shaft-disc unit to prevent short-
circuiting. The waste flow enters the contact tank at one
end and is allowed to flow either perpendicular to or par-
allel to one or more units for treatment. The removal of
organic matter from the waste flow, either municipal sewage
or combined sewer overflow, is accomplished by adsorption of
the organic matter at the surface of the biological growth
covering the rotating discs. Rotational shearing forces
cause sloughing of excess biomass. Secondary clarification
should follow the rotating biological contactor treatment to
remove sloughed biomass.
As in all biological systems, because microorganisms have a
maximum metabolism rate, only a given amount of substrate
can be removed with a given amount of biomass. Although
this is true generally in the rotating biological contactor,
excess biomass can be held on the disc and can effectively
act as a reserve for use at higher loadings. The effective-
ness, however, is somewhat limited by the oxygen transfer
rate and the substrate diffusion gradient through the layer
of biomass on each disc. This is similar to what happens in
trickling filters. In general, though, the reserve biomass
reduces the importance of maintaining a uniform loading
rate.
Efficiency
The reported BODs removal efficiencies range from 60 to
95 percent [7, 2, 3, 26]. The higher values are for more
recent installations treating dry-weather flow. Suspended
solids removals are also in this range. Removals for
settleable solids, nitrogen, and phosphorus have been re-
ported to be 80 to 90, 40, and 50 percent, respectively.
When treating combined sewage flows, controlled treatment
(70 percent or better COD removal efficiency) was report-
edly maintained up to 8 to 10 times dry-weather flow [7],
A linear reduction in COD removal efficiency from 70 down to
20 percent was reported for the flow range from 10 to 30
times dry-weather flow.
Operational Considerations
Conditions noted to affect the BODs and COD removals in a
rotating biological contactor are (1) organic loading rate,
(2) contact time, (3) effluent settling, (4) the number of
units in series, and (5) high flow rates. The most impor-
tant condition is high flow rates which affects the first
three of the conditions just enumerated. The maximum allow-
able variation in flow is approximately 10 times the base
276
-------
The rotating discs are partially submerged and baffles are
used between each shaft-disc unit to prevent short-
circuiting. The waste flow enters the contact tank at one
end and is allowed to flow either perpendicular to or par-
allel to one or more units for treatment. The removal of
organic matter from the waste flow, either municipal sewage
or combined sewer overflow, is accomplished by adsorption of
the organic matter at the surface of the biological growth
covering the rotating discs. Rotational shearing forces
cause sloughing of excess biomass. Secondary clarification
should follow the rotating biological contactor treatment to
remove sloughed biomass.
As in all biological systems, because microorganisms have a
maximum metabolism rate, only a given amount of substrate
can be removed with a given amount of biomass. Although
this is true generally in the rotating biological contactor,
excess biomass can be held on the disc and can effectively
act as a reserve for use at higher loadings. The effective-
ness, however, is somewhat limited by the oxygen transfer
rate and the substrate diffusion gradient through the layer
of biomass on each disc. This is similar to what happens in
trickling filters. In general, though, the reserve biomass
reduces the importance of maintaining a uniform loading
rate.
Efficiency
The reported BOD5 removal efficiencies range from 60 to
95 percent [7, 2, 3, 26]. The higher values are for more
recent installations treating dry-weather flow. Suspended
solids removals are also in this range. Removals for
settleable solids, nitrogen, and phosphorus have been re-
ported to be 80 to 90, 40, and 50 percent, respectively.
When treating combined sewage flows, controlled treatment
(70 percent or better COD removal efficiency) was report-
edly maintained up to 8 to 10 times dry-weather flow [7].
A linear reduction in COD removal efficiency from 70 down to
20 percent was reported for the flow range from 10 to 30
times dry-weather flow.
Operational Considerations
Conditions noted to affect the BODs and COD removals in a
rotating biological contactor are (1) organic loading rate,
(2) contact time, (3) effluent settling, (4) the number of
units in series, and (5) high flow rates. The most impor-
tant condition is high flow rates which affects the first
three of the conditions just enumerated. The maximum allow-
able variation in flow is approximately 10 times the base
276
-------
Figure 18. Stormwater surface detention pond (Chicago)
142
-------
Table 54. TRICKLING FILTER REMOVALS [20]
NEW PROVIDENCE, NEW JERSEY
BOD SS
treated culatlon Trickling fitters Overall a Trickling filters Overalla
flow, rate, Influent, Effluent, Removal, removal, Influent, Effluent, Removal, removal,
Condition mgd ' mgd mg/1 mg/1 % t mg/1 mg/1 $ t
Dry weather flow
First year 0.54 0.8 104 23 78 86 86 20 77 87
Part of
second year 0.56 -- -- 9 -- 94 -- 12 -- 93
Wet weather flow
First year 3.96b O.to 0.8 86 39 53 64 64 36 42 67
Part of
second year 1.72C -- -- 17 -- 87 -- 20 -- 86
a. Includes removals by primary sedimentation.
b. Average wet weather flow; average peak flows were 6.0 mgd with no recirculation.
c. Wet weather flow rate was reduced by approximately 1.5 mgd by pumping to another treatment plant.
Note: mgd x 43.8 = I/sec
In comparing the plastic medium and the rock filter, it was
noted that up to 2-1/2 times the BOD5 removal per unit vol-
ume was possible with the plastic medium. Also, on a capi-
tal cost basis, the plastic medium outperformed the rock
by 2 to 1 ($/kg BODs removed/1,000 cu m).
Design Parameters — The average hydraulic and organic load-
ings applied to the New Providence facilities are slightly
above the recommended design values. The recommended values
are:
Plastic Medium Rock
Hydraulic 2.73 cu m/hr/sq m 0.78 cu m/hr/sq m
loading (70 mgad) (20 mgad)
Organic 1.36 kg BODr/day cu m 0.64 kg BODr/day/cu m
loading (85 Ib BOD5/day/l,000 cf or (40 Ib BODr/day/1,000 cf or
3,700 Ib BOD5/acre-£t/day) 1,742 Ib BOD5/acre-ft/day)
Additional design parameters were included previously in
Table 53.
Advantages and Disadvantages
Advantages of trickling filters include: (1) they handle
varying hydraulic and organic loads, (2) are simple to oper
ate, (3) have ability to withstand shockloads, and (4) have
274
-------
recharge of groundwater. Erosion control measures
in construction areas will minimize the increased
solids loadings in runoff from such areas.
4. Drainage pipes and other flood control structures
will be used only where the natural system is in-
adequate, such as at high density urban activity
centers. Plans presently call for the use of
porous pavements to reduce runoff from streets.
5. Control will be exercised over the type and amount
of fertilizers, pesticides, and herbicides to mini
mize pollution of the runoff.
It has been estimated that the drainage system will cost an
average of $243/ha ($600/acre) , compared with perhaps
$486/ha ($l,200/acre) for a conventional system [2].
144
-------
COMMINUTORED INTERCEPTOR FLOWS
SLUDGE GRINDER
PLASTIC
TRICKLING
FILTER
ROCK
TRICKLING
FILTER
EXCESS DRY-WEATHER
FLOW TO SUMMIT
FOR TREATMENT
-*-
2ND
WET
PUMP
1
STAGE ,
WELL '
^
>
i
WEIR
CHLORINE CONTACT
TANKS
EFFLUENT TO
PASSAIC RIVER
Figure 52.
Trickling filter plant schematic,
New Providence, N.J.
272
-------
HOUSE
WASTE
(INCLUDING
DRAINAGE x
FROM ROOF) -' ' SURFACE
CATCHBAS.N MMF
BUILDING
SEWER
WNBINED SEWEl
SANITARY INTERCEPTOR
TO WATER POLLUTION
PJ.ANTS FOR
TREATMENT
RECEIVING WATERS
COMBINED SE«CR OVERFLOW
MIXTURC OF HHHCIML *(ff*tf
AMI STIIMWATCR BISCNAMINI
INTI THE MCCCIVING WATERS
Figure 19. Common elements of an Interceptor
and transport system [6]
146
-------
area per unit volume and are quite heavy, thus limiting the
depth of filter. Redwood slats and corrugated plastic are
much lighter and can be constructed with a larger surface
area per unit volume.
Operation — The operation of most high-rate and ultrahigh-
rate trickling filters is in series with a second or third
filter and/or with recirculation. The purpose is to provide
high removals by increasing the contact time of the waste
with the biomass attached to the filter material. When
operating alone without recirculation, trickling filters
used for treating domestic wastes remove between 50 and
75 percent of the BODs.
Under storm conditions, the trickling filter must handle
highly varying flows. Applying a varying organic load to
a filter does not produce optimum removals. It is gener-
ally thought that only sufficient biomass adheres to the
supporting medium to handle the normal organic load. As the
loading increases above this level, the maximum BODs utiliza-
tion rate of the biomass is reached. This is not a sharp
distinction because some excess biomass always adheres to
the medium and can accept some of the organic load.
A varying hydraulic load also affects removals. The in-
creased shearing action of high flows causes excess slough-
ing or washing off of the biomass. To help dampen this
effect, filters operating in series under dry-weather condi-
tions can be operated in parallel, thereby reducing some of
the increased hydraulic load on each filter. A maximum
overall flow variation (maximum/minimum) of 8 to 10 is
acceptable while still achieving significant removals [20].
Design — Trickling filter design has been based primarily on
empirical formulas. This does not imply that the basic bio-
logical kinetics are not operative; rather, it means that
mathematical description of the process has not been
formulated. There are several design equations in the
literature that may be used for the design of trickling
filters [17, 6]. In designing a trickling filter to treat
overflows, it must be remembered that dry-weather flow is
needed to keep the biomass active between storms. Generally,
two or more units should be used to provide high removals by
operating in series during dry weather and in parallel dur-
ing storm events to accommodate the flow variation needed.
Demonstration Project, New Providence, New Jersey
Trickling filters have been used extensively throughout the
United States to treat domestic flows, but only one facility
(at New Providence, New Jersey) has been designed to treat
270
-------
(50 million persons) was served in whole or in part by com-
bined sewer systems [42]. Furthermore, it was reported that
there were 14,212 overflows in the total 641 jurisdictions
surveyed; of these, 9,860 combined sewer overflows were
reported from 493 jurisdictions. Until 1967, the most com-
mon remedial method reported was sewer separation, and of
274 jurisdictions with plans for corrective facilities con-
struction, 222 indicated that some degree of sewer separa-
tion would be undertaken.
Detailed Analysis
Sewer separation will continue to be used to some degree in
the future and thus an investigation of the methods, their
advantages and disadvantages, and their costs is warranted.
There are three categories of sewer separation systems:
pressure, vacuum, and gravity.
The most comprehensive study of the pressure or "sewer with-
in a sewer" concept was published by the ASCE [12] in 1969.
The greatest disadvantage of pressure systems is generally
higher costs, as shown in a comparison of pressure and
gravity system costs in the cities of Boston, Milwaukee,
and San Francisco presented in Table 26. The ratios of pres-
sure to gravity costs are 1.4, 1.5, and 1.5, respectively.
The in-sewer pressure lines varied from 6.3 to 40.6 cm (2-1/2
to 16 inches) in diameter and pressure control valves limited
the line pressure to 2.11 kg/sq cm (30 psi) . A major portion
of the costs is the "in-house separation" which can be
as high as 82 percent of the total cost for separation
using a pressure system [12]. Besides the high costs, other
disadvantages of pressure systems are that (1) they are dif-
ficult to maintain; (2) they require complex controls; and
(3) they are dependent on electricity for operation. It is
important to realize that approximately 72 percent of all
combined sewers are less than 0.61 meters (2.0 feet) in
diameter, making it difficult to install the pressure pipe.
The advantages are that (1) as an alternative, they provide
an additional degree of latitude in sewer design, (2) there
is minimal construction interference to commerce and traffic,
and (3) they are handy in low areas.
Sewer separation of existing combined sewers has histori-
cally been accomplished by utilizing gravity systems. The
advantages of gravity sewer separation are that (1) all
sanitary sewage is treated prior to discharge; (2) treatment
plants operate more efficiently under the relatively stable
sanitary flows; (3) other alternatives are less reliable
148
-------
Advantages and Disadvantages
Some advantages of the contact stabilization process for the
treatment of excess (combined sewer) flows in this applica-
tion are: (1) high degree of treatment; (2) central location
of maintenance personnel and equipment; and (3) reduction of
the loadings on dry-weather facilities, by dual use of
facilities, during normal operations and emergency shutdown
of the main plant, making the whole very versatile. Contact
stabilization shows definite promise as a method for treat-
ing combined sewer overflows when used in combination with
a dry-weather activated sludge treatment plant. Disadvan-
tages are: (1) high initial cost, (2) the facilities must
be located next to a dry-weather activated sludge plant,
(3) adequate interceptor capacity must exist to convey the
storm flow to the treatment plant, and (4) expansion of
major interceptors may be required.
TRICKLING FILTERS
Description of Process
Trickling filters are widely employed for the biological
treatment of municipal sewage. The filter is usually a
shallow, circular tank of large diameter filled with crushed
stone, drain rock, or other similar media. Settled sewage
is applied intermittently or continuously over the top sur-
face of the filter by means of a rotating distributor and is
collected and discharged at the bottom. Aerobic conditions
are maintained by a flow of air through the filter bed in-
duced by the difference in specific weights of the atmos-
phere inside and outside the bed.
The term "filter" is a misnomer, because the removal of
organic material is not accomplished with a filtering or
straining operation. Removal is the result of an adsorption
process occurring at the surface of biological slimes cover-
ing the filter media.
Classification — Trickling filters are classified by hy-
draulic or organic loading. Until recently, there were only
two flow classifications: low rate and high rate. A third
classification, ultrahigh rate, has been added since the ad-
vent of plastic medium filters. Although the distinctions
are based on hydraulic loading, they are centered in reality
around the organic loading that the filter can handle. A
comparison of the three classifications of trickling filters
is presented in Table 53.
The type of medium used varies considerably. Rock, slag,
hard coal, redwood slats, and corrugated plastic have been
used. Rock, slag, and hard coal have relatively low surface
268
-------
overall because of external power requirements; (4) no land
acquisition is necessary; (5) receiving water pollution
loads can be reduced by 50 percent (according to independent
studies [49, 30]); and (6) little increase in manpower is
required.
Disadvantages of gravity systems may be divided into three
categories: nonquantifiable, separation effectiveness, and
costs. Nonquantifiable disadvantages, which based on past
experience are the most important, are that (1) considerable
work is involved in in-house plumbing separation; (2) there
are business losses during construction; (3) traffic is
disrupted; (4) political and jurisdictional disputes must
be resolved; (5) extensive policing is necessary to ensure
complete and total separation; and (6) considerable time is
required for completion (e.g., in 1957 separation in
Washington, B.C., was estimated to take until sometime
after the year 2000 to complete) [24]. Separation effec-
tiveness disadvantages are as follows: (1) there is only a
partial reduction of the pollutional effects of combined
sewer overflows [30] ; (2) urban area stormwater runoff con-
tains significant contaminants [7, 4]; and (3) it is diffi-
cult to protect storm sewers from sanitary connections
(either authorized or unauthorized). Estimated costs for
gravity sewer separation are shown for various cities in
Table 26.
The cost disadvantages of separation, when compared to some
conceptive alternative solutions, are indicated in Table 27.
Again, the major reason for the higher costs of sewer sepa-
ration are in-house plumbing changes which can be as high as
82 percent of the total sewer separation costs [12].
Conclusions
On the basis of currently available information, it appears
that sewer separation of existing combined sewer systems is
not a practical and economical solution for combined sewer
overflow pollution abatement. Several cited alternatives
listed in Table 27 suggest other solutions, most of which
are considerably less expensive and should give better re-
sults with respect to receiving water pollution abatement.
In addition, storm sewer discharges may not be allowed at
all in the future, thus forcing collection and treatment of
all sewage and stormwater prior to discharge. In this case,
the argument for either separate or combined sewers is moot.
The choice between sewer separation and other alternatives
will be controlled by the uniqueness of each situation.
The examples cited in Table 27 leave no doubt that any alter-
native to sewer separation is the better choice. However,
150
-------
ratio, and detention times in the contact and stabilization
tanks [17, 24, 15, 14]. In the work at Kenosha, however, it
has not been possible to show any correlation between re-
movals and these items, although it has been shown that
operation based on an assumed uniform influent BODs is suf-
ficient for good BODs and SS removals (80 and 90 percent,
respectively).
Operating Parameters — With contact stabilization or any
other activated sludge process, operation is normally based
on the food-to-microorganism ratio or sludge retention time.
Because of this, difficulties may be encountered when using
an activated sludge process for treating a rapidly varying
and intermittent flow. The sludge retention time is particu-
larly difficult to control because overflows may not last
long enough for the plant to stabilize and for proper
wasting procedures to be instituted. Operating the plant on
stored overflows could reduce this problem The food-to-
microorganism ratio, which is interrelated to the sludge age,
can be used to control the operation of the plant; however,
it too is difficult to control since the concentration of
both the incoming BODs and the biological solids in the sys-
tem must be known. This is further complicated because the
BODs concentration in the combined sewer overflow may vary
significantly. Based on the results at Kenosha, it has been
found that e*act control is not necessary for good operation.
The operating parameters used for the contact stabilization
plant at Kenosha are shown in Table 52. The values reported
are averages, and the range was generally within ±60 percent
of the value listed. For comparison, the design parameters
for sewage treatment by contact stabilization found in the
literature are also presented.
For units such as that at Kenosha, sophisticated design may
not be warranted because the system is operated for such
short periods that the biosolids and the kinetics of the
system do not have a chance to adjust to the incoming flow
before the storm is over. In this case, using the reported
design equations should be sufficient. Abatement plans that
include a contact stabilization process for the treatment of
stored overflows for periods of time greater than 5 to
10 days may warrant more sophisticated design to achieve
higher removal efficiencies. The use of the kinetic equa-
tions describing the metabolism of the bacteria, as formu-
lated by McCarty [14], Metcalf § Eddy [17], and others, may
prove useful under such circumstances.
Results of Operational Tests — The work at Kenosha has not
been able to show any adverse condition that affects
removals. Based on the results of 23 storms studied, the
266
-------
INFILTRATION/INFLOW CONTROL
A serious problem results from (1) excessive infiltration
into sewers from groundwater sources and (2) high inflow
rates into sewer systems through direct connections from
sources other than those which the sewers are intended to
serve. Inflow does not include, and is distinguished from,
infiltration. The sources and control of infiltration and
inflow are discussed in this subsection.
Sources
Infiltration is the volume of groundwater entering sewers
and building sewer connections from the soil through defec-
tive joints, broken, cracked, or eroded pipe, improper
connections, manhole walls, etc. Inflow is the volume of
any kind of water discharged into sewer lines from such
sources as roof leaders, cellar and yard drains, foundation
drains, commercial and industrial so-called "clean water"
discharges, drains from springs and swampy areas, depressed
manhole covers, cross connections, etc.
Inflow sources generally represent a deliberate connection
of a drain line to a sewerage system. These connections may
be authorized and permitted; or they may be illicit connec-
tions made for the convenience of property owners and for
the solution of on-property problems, without consideration
of their effects on public sewer systems.
The intrusion of these waters takes up flow capacity in the
sewers. Especially in the relatively small sanitary sewers,
these waters may cause flooding of street and road areas and
backflooding into properties. This flooding constitutes a
health hazard. Thus these sanitary sewers actually function
as combined sewers, and the resulting flooding becomes a
form of combined sewer overflow.
The two types of extraneous water, inflow and infiltration,
which intrude into sewers do not differ significantly in
quality, except for the pollutants unavoidably or deliber-
ately introduced into waters by commercial-industrial
operations [13]. Foundation inflow, for example, does not
vary greatly from the kind of water that infiltrates sewer
lines from groundwater sources. Basement drainage may
carry wastes and debris originating in homes, including
laundry wastewater.
Inflow Control
Correction of inflow conditions is dependent on regulatory
action on the part of city officials, rather than on public
152
-------
(c)
(e)
Figure 51. Combined sewer overflow treatment
by contact stabilization (Kenosha)
(a) Contact tank with diffused air (b) Sludge stabilization tanks with floating
aerators (c) Floating aerator anchoring and counterweight details (d) Closeup
of aerator operation (e) Final contact tank (peripherally fed) and effluent
264
-------
Choice of sewer pipe - Improvements in pipe materials assure
the designer's ability to provide proper materials to meet
any rational infiltration allowances he wishes to specify.
The upgrading of pipe manufacture to meet rigid quality
standards and specifications has eliminated the basic ques-
tion of watertightness of pipe material. The important
issues to consider in pipe material selection are struc-
tural integrity, strength of the wastewater character, and
local soil or gradient conditions. Combinations of these
factors may make one material better suited than another or
preferable under certain special installation conditions.
In such situations, pipe materials are often chosen for
reasons other than their relative resistance to infiltration
The cost of the pipe is usually a small part of the total
project cost. For rough estimating purposes, the cost of
installed sewer pipes (excluding manholes, laterals and
connections, appurtenances, etc.) ranges from $0.97 to $1.55
per cm diameter per linear meter ($1.25 to $2.00 per inch
diameter per linear foot).
Materials commonly used for sewer pipe construction include
(1) asbestos cement, (2) bituminous coated corrugated metal,
(3) brick, (4) cast iron or ductile iron, (5) concrete
(monolithic or plain), (6) plastic (including glass fiber
reinforced plastic, polyvinylchloride, ABS, and poly-
ethylene), (7) reinforced concrete, (8) steel, (9) vitrified
clay, and (10) aluminum. All of these materials, with the
possible exceptions of the plastics and aluminum, have been
used in sewer construction for many years.
Since sewer pipe made from the plastic materials is rela-
tively new, a brief description of the use of plastic pipes
is included below.
Solid wall plastic pipe usually refers to materials such as
polyvinylchloride (PVC), chlorinated polyvinylchloride
(CPVC), polyvinyldichloride (PVDC), and polyethylene. These
materials are lightweight, have high tensile strength, have
excellent chemical resistance, and can be joined by solvent
welding, fusion welding, or threading. The PVC is probably
the most commonly used plastic pipe because it is stronger
and more rigid than most of the other thermoplastics; how-
ever, PVC is available only in diameters up to 30.5 cm
(12 inches).
Polyethylene pipe is finding major use as a liner for dete-
riorated existing sewer lines [26], Several lengths of
polyethylene pipe can be joined by fusion welding into a
long, flexible tube. This tube is then pulled into the
existing sewer. When the existing house laterals have been
connected to this new pipe liner, the result is a watertight
154
-------
3. Shakedown runs are necessary to keep the units and
the usual large number of automatic controls in
operating order.
CONTACT STABILIZATION
Description of the Process
Contact stabilization is considered in lieu of other acti-
vated sludge process modifications for treating combined
sewer overflows, because it requires less tank volume to
provide essentially the same effluent quality. The over-
flow is mixed with returned activated sludge in an aerated
contact basin for approximately 20 minutes at the design
flow. Following the contact period, the activated sludge
is settled in a clarifier. The concentrated sludge then
flows to a stabilization basin where it is aerated for
several hours. During this period, the organics from the
overflow are utilized in growth and respiration and, as
a result, become "stabilized." The stabilized sludge is
then returned to the contact basin to be mixed with the in-
coming overflow. A schematic of a contact stabilization
plant for treating combined sewer overflows is shown on
Figure 50.
Demonstration Project, Kenosha, Wisconsin
A project sponsored by the EPA to evaluate the use of con-
tact stabilization for treatment of combined sewer overflows
from a 486-ha (1,200-acre) tributary area is presently under-
way at Kenosha, Wisconsin [23, 19]. It is an example of how
contact stabilization can be used to treat combined sewer
overflows using the waste activated sludge from a dry-weather
activated sludge plant. At the Kenosha municipal sewage
treatment plant, a 101-1/sec (23-mgd) facility, a new com-
bined sewer overflow treatment facility was constructed.
This facility consists of an aeration tank, a contact sta-
bilization tank, and a new clarifier. The design capacity
of the new facility is 88 I/sec (20 mgd). The stabilization
tank, acting as the biosolids reservoir, receives the waste
activated sludge from the main plant. This sludge is held
for up to 7 days before final wasting. Thus, stabiliz-ed
activated sludge is kept in reserve ready to treat combined
sewer overflows when they occur. Photographs of the facil-
ity are shown on Figure 51.
Operation of the contact stabilization plant consists of
directing the combined sewer overflow to the contact tank
following comminution and grit removal, adding the reserve
activated sludge, and then conducting the waste flows to a
final clarifier for separation of the biosolids and other
262
-------
made of natural rubber, synthetic rubber, or various other
elastomers. These joints are used on asbestos cement pipe,
cast iron pipe, concrete pipe, vitrified clay pipe, and
certain types of plastic pipes. Compression gasket joints
are most effective against infiltration while still pro-
viding for deflection of the pipe.
Chemical weld joints — Chemical weld joints are used to join
certain types of plastic pipes and glass fiber pipes. The
joints provide a watertight seal. It has been reported
that, on the basis of field tests, jointing under wet or
difficult-to-see conditions does not lend itself to precise
and careful workmanship. Thus special care is necessary
in preparing these joints in the field. More experience
with these pipes in sewer applications is necessary to
determine the longevity of this type of joint.
Heat shrinkable tubing — A new type of joint developed
recently is the heat shrinkable tubing (HST) [27]. The HST
material begins as an ordinary plastic or rubber compound
which is then extruded into sections of tubing. The tubing
is then heated and stretched in diameter but not in length.
After cooling it retains the expanded diameter. If a length
of 8-inch diameter tubing is expanded to 16 inches, it
will conform to any shape between 8 and 16 inches when
reheated. This characteristic gives the HST the ability to
form a tight fit around sewer pipe joints.
The material recommended for HST joints is a polyolefin
which has a high degree of chemical resistance and the
ability to resist scorching and burning, and is both eco-
nomical and easy to apply. To further assure HST joint
strength and resistance to internal pressure, a hot melt
adhesive is recommended as an inner surface sealant. The
adhesive material has a melting temperature close to that
of the HST and will bind the tubing and pipe materials to-
gether as the tubing cools to its final shape. Both pro-
pane torches and catalytic heaters can be used as the heat
source.
Physical properties of the HST reportedly were better than
those of currently used joint materials:
The coupling of commercial sewer pipe, both butt-
end and bell and spigot, with watertight joints
using heat shrinkable plastic tubing is feasible
and economically practical. Used in conjunction
with a hot melt adhesive it can surpass in phys-
ical and chemical strength any of the conven-
tional joints presently being used with clay,
concrete, and asbestos-cement nonpressure sewer
pipe. [27]
156
-------
Operationally and from a design standpoint, the aforemen-
tioned factors are taken into account by considering (1) the
food-to-microorganism ratio, (2) the sludge retention time,
and (3) the hydraulic detention time.
The food-to-microorganism ratio is defined as the kilograms of
BODr (food) applied per unit time (often taken as the amount
consumed) per kilogram of organisms in the system. The sludge
age is defined by the kilograms of organisms wasted per day.
The hydraulic detention time is defined as the value, given in
units of time, obtained by dividing the volume of the reaction
vessel by the flow rate.
Because the food-to-microorganism ratio and the sludge re-
tention times are interrelated [17], both are commonly used
in the design of biological systems. From field observa-
tions and laboratory studies, it has been found that as the
sludge age is increased and, correspondingly, the food-to-
microorganism ratio decreased, the settling characteristics
of the organisms in the system are enhanced, and they can be
removed easily by gravity settling. Typical values for the
food-to-microorganism ratio and sludge age are given in
reference [17].
As previously noted, the length of time the biomass is in
contact with the waste BODs is measured by the hydraulic de-
tention time. The minimum time to achieve a given removal
is dependent upon the food-to-microorganism ratio. Low
ratios (i.e., a high number of bacteria per kilogram of BOD,-)
allow faster utilization of a given amount of BODs. The
minimum time required may vary considerably, from 10 to 15
minutes in contact stabilization, or less for trickling
filters and rotating biological contactors, and up to 2 to
3 days for oxidation ponds. At the shorter contact times,
the biomass only removes the dissolved matter and possibly
some of the smaller colloidal matter [15] . At longer con-
tact times, suspended organic matter is utilized.
In any biological system, these factors control the process.
A mathematical model has been developed for the activated
sludge system [17, 14]. Models for trickling filters,
rotating biological contactors, and treatment lagoons have
not been formulated. Empirical designs and design param-
eters are used instead.
APPLICATION TO COMBINED SEWER OVERFLOW TREATMENT
Biological treatment of wastewater, used primarily for domes-
tic and industrial flows of organic nature, produces an
effluent of high quality and is generally the least costly
260
-------
include the maximum infiltration anticipated during the life
of the sewer, while the construction allowance should be
the maximum allowable infiltration at the time of
construction. The construction infiltration will increase
continuously throughout the life of the project. APWA has
recommended the establishment of a construction infiltration
allowance of 185 I/cm diameter/km/day (200 gal./inch
diameter/mile/day) or less. This is not unreasonable in
light of improvements in pipe and joint materials and con-
struction methods.
Average and peak design flows should be related to the
actual conditions for the area under design. Too often
flow criteria are taken from a standard textbook. Adequate
subsurface investigations should be undertaken to establish
conditions that may affect pipe and joint selection or
bedding requirements. Consideration should be given to the
constructability and maintainability of the sewer system.
This calls for direct communication between the designer
and maintenance personnel.
Manholes should be designed with as few construction joints
as possible. In recent years the development of custom-
made precast manholes with pipe stubs already cast in place
has reduced the problem of shearing and damage of connect-
ing pipes. The use of flexible connectors at all joints
adjacent to manholes reduces the possibility of differen-
tial settlement shearing the connecting pipes.
Manhole cover design is attracting serious attention in
view of evidence that even small perforations can produce
sizable contributions of extraneous inflow. A single
2.5-cm (1-inch) hole in a manhole top covered with 15.2 cm
(6 inches) of water may admit 0.5 I/sec (11,520 gpd) [41],
Solid sealed covers should be used for manholes in areas
subject to flooding. If solid covers are used, alternative
venting methods must be used to admit air or remove sewer
gases.
Construction considerations — The most critical factor rela-
tive to infiltration prevention is the act of construction.
The capability of currently manufactured pipes and joints
to be assembled allowing minimal infiltration must be
coupled with good workmanship and adequate inspection,
expecially at house connections.
Trenches should be made as narrow as possible but wide
enough to permit proper laying of pipe, inspection of
joints, and consolidation of backfill. Construction should
be accomplished in dry conditions. If water is encountered
158
-------
Plant capacity Operation and
Capital maintenance
cu m/sec mgd cost, $ cost, $/yr
1.1 25 1,580,000 44,000
2.2 50 2,390,000 55,000
4.4 100 4,370,000 98,000
8.8 200 7,430,000 129,000
The cost data are based on an ENR of 2000.
The operating costs are estimated to be $0.0382/1,000 1
($0.141/1,000 gal.) for 300 hours of operating per year. The
high cost could easily be reduced, however, by designing the
system to serve also as a dry-weather effluent polisher dur-
ing periods with no storm flows.
CONCENTRATION DEVICES
Concentration devices, such as the swirl regulator/
concentrator and helical or spiral flow devices, have intro-
duced an advanced form of sewer regulator--one capable of
controlling both quantity and quality. These devices have
been previously described in Section VIII. A prototype
swirl regulator has recently been constructed in Syracuse,
New York. A second generation swirl concentrator has
been placed into operation as a treatment unit for municipal
sewage grit separation in Denver, Colorado. Settleable
solids removals ranging from 65 to more than 90 percent,
corresponding to chamber retention times of approximately 5
to 15 seconds, have been predicted on the basis of hydraulic
model tests. At the time of writing, no operational data
were available. Indicated costs are approximately $285/cu m/
sec ($6,500/mgd). A third generation swirl device has been
developed to take the place of conventional primary sedi-
mentation at 10 to 20 minute detention times.
258
-------
Identification of the drainage system includes a review of
detailed maps of the sewer system; field checks of the line,
grade, and sizes; and identification of sections and man-
holes that are bottlenecks.
To identify the scope of the infiltration, it is necessary
to measure and record both dry- and wet-weather flows at
key manholes. Groundwater elevations should be obtained
simultaneously with sewer flow measurements.
A physical survey of the entire sewer system, or that por-
tion of major concern, where every manhole is entered and
sewers are examined visually to observe the degree and
nature of deposition, flows, pipe conditions, and manhole
condition should be made. Smoke testing may reveal infil-
tration sources only under low groundwater conditions. If
the groundwater table is above the pipe, the smoke may be
lost in the water. Soil conditions and groundwater condi-
tions should also be noted.
An economic and feasibility study is necessary to determine
the locations where the greatest amount of infiltration can
be eliminated for the least expenditure of money. In some
cases, it may be most cost effective to provide additional
treatment capacity at the sewage treatment plant for the
infiltration. Cost estimates can be developed for subse-
quent correctional stages as necessary.
Cleaning — A systematic program of sewer cleaning (1) can
restore the full hydraulic capacity and self-scouring
velocity of the sewer and its ability to convey infiltra-
tion without flooding; (2) can aid in the discovery of
trouble spots, such as areas with possible breaks, offset
joints, restrictions, and poor house taps, before any sub-
stantial damage is caused; and (3) is a necessary prerequi-
site to television and photographic inspection. It is
one of the most important and useful forms of preventive
maintenance. This type of program involves periodic
cleaning on a regular, recurring basis.
By frequent hydraulic flushing of the sewers, the interval
between mechanical cleanings of the sewer can be extended.
This will be discussed in more detail later in this section.
Equipment used in cleaning falls into three general classi-
fications: (1) rodding machines, (2) bucket machines, and
(3) for small sewers, hydraulic devices. The rodding
machine, which is used most commonly, removes heavy conglom-
erations of grease and root intrusions. The bucket machine
utilizes two cables threaded between manholes. One cable
160
-------
screen during the overflow event and st9red in two 19-cu m
(5,000-gal.) tanks for the test filtration runs that
followed. Each tank had a mixer to keep solids in
suspension. Two pumps were then used to supply the filter
with screened water.
Removals for this filter were 65 percent for SS, 40 percent
for BODs, and 60 percent for COD [22]. The addition of
polyelectrolyte increased the SS removal to 94 percent, the
BODs removal to 65 percent, and the COD removal to 65 per-
cent. Inorganic coagulants, such as lime, alum, and ferric
chloride, did not prove as successful as polymers. Run
times averaged 6 hours at loading rates of 58.7 cu m/hr/sq m
(24 gpm/sq ft). Backwashing of the filters consisted of
alternately injecting air and water into the bottom of the
filter columns. Air volume was varied from 38.4 to
283 cu m/hr/sq m (2.1 to 15.5 scfm/sq ft) over 2.5 to 29
minutes. Backwash water volume used ranged from 1.9 to
8.6 percent of the total combined sewer overflow filtered,
with a median value of approximately 4 percent. The range
of backwash water rate used was 75.8 to 220 cu m/hr/sq m (31
to 90 gpm/sq ft) over 4 to 25 minutes.
A list of the basic design data is presented in Table 50.
Others — Two other filtration processes, fiber glass plug
filtration [24] and coal filtration [34], show some promise,
but additional research is necessary to perfect them. Other
methods, such as crazed resin filtration, upflow filtration
with garnet sand, and filtration using ultrasonically
cleaned fine screens,have not been successful and are not
considered worthy of further effort at the present time.
Advantages and Disadvantages
The advantages of dual-media filtration are that (1) rela-
tively good removals can be achieved; (2) process is versa-
tile enough to be used as an effluent polisher; (3) operation
is easily automated; and (4) small land area is necessary.
Disadvantages are that (1) costs are high; (2) dissolved
materials are not removed; and (3) storage of backwash water
is required.
Costs
Cost data were developed from a design estimate for 1.1, 2.2,
4.4, and 8.8 cu m/sec (25, 50, 100, and 200 mgd) filtration
plants at a satellite location [22]. The basic plant as en-
visioned for the cost estimate includes a low lift pump
256
-------
(1) replacement of broken sections, (2) insertion of various
types of sleeves or liners, (3) internal sealing of joints
and cracks with gels or slurries, and (4) external sealing
by soil injection grouting. Additional detailed information
is available in recent EPA reports on jointing materials
[13, 41, 27] and sealants [29, 13, 41, 25].
The method most commonly used to correct structural defects
and infiltration (in sections where major structural damage
is not present) is internal sealing with gels or slurries.
The use of a chemical blocking method to seal sewer cracks,
breaks, and bad joints is much more economical and feasible
than sewer replacement or the inadequate concrete flooding
method. With recent improvements in television and photo-
graphic inspection methods, sealing by chemical blocking
appears to be an even more encouraging method than
heretofore. Chemical blocking is accomplished by injecting
a chemical grout and catalyst into the crack or break. A
sealing packer is used to place the grout and catalyst.
The packer has inflatable elements to isolate the leak, an
air line for inflation, and two pipes for delivering the
chemical grout and catalyst to the packer. An example of a
packer is shown on Figure 20. During the repair the two
inflatable end sections isolate the leak and chemical grout
and catalyst are injected into the center section. Then
the center section is inflated to force the grout from
the annulus between the packer and the sewer wall into
the leak. When the repair is complete,the packer is de-
flated and moved to the next repair location.
The current use of acrylamid gels as chemical blocking
agents is restricted by their lack of strength and other
physical limitations. Recently, improved materials, such
as epoxy-based and polyurethane-based sealants, have been
developed [29]. These new sealants have exhibited suita-
bility even under conditions of erratic or intermittent
infiltration where acrylamid gels failed because of re-
peated dehydration. The only difficulty in applying the
new sealant materials has been that, because of the physi-
cal properties of the sealants, new application equipment
incorporating a mixing mechanism is required. The cost of
this new equipment is approximately the same as the exist-
ing equipment. Modification of existing packing equipment
to accept the new sealants has been found to be feasible.
Sewers may also be sealed by inserting sleeves or liners,
as discussed previously.
162
-------
FILTRATION
Introduction
In the physical treatment processes, filtration is one step
finer than screening. Solids are usually removed by one or
more of the following removal mechanisms: straining, im-
pingement, settling, and adhesion. Filtration has not been
used extensively in wastewater treatment, because of rapid
clogging which is principally due to compressible solids
being strained out at the surface and lodged within the
pores of the filter media. In stormwater runoff, however,
a large fraction of the solids are discrete, noncompressible
solids that are more readily filtered [30].
Effluents from primary or secondary treatment plants and
from physical-chemical treatment facilities contain com-
pressible solids.
The discussion on filters handling discrete, noncompressible
solids is presented here.
Design Criteria
Two factors affecting removal efficiency are flux rate and
the type of solids. As one would expect, the removals are
inversely proportional to the flux rate. At high flux
rates, solids are forced through the filters reducing solids
removal efficiency. Suspended solids removals were found
to be better for inert solids (discrete, noncompressible
solids) than for volatile solids (compressible solids).
This is the same conclusion found for microstrainers.
Loading Rates — The difference between filtering compres-
sible and noncompressible solids is basically the flux rate
used. High-rate filters handling compressible solids are
normally loaded at 12.2 to 24.5 cu m/hr/sq m (5 to 10 gpm/
sq ft), whereas those handling noncompressible solids will
filter at rates up to 73.4 cu m/hr/sq m (30 gpm/sq ft).
Chemicals — Many polyelectrolytes and some coagulants have
been tested. Some polyelectrolytes have been found which
increase removals of phosphorus and nitrogen. It is
cautioned, however, that polyelectrolytes are noted for
their unpredictability and the most effective polyelectro-
lyte must be determined for each wastewater.
Demonstration Projects
Studies have been made to investigate possible filtration
techniques for combined sewer overflows. The different
254
-------
with grades too flat to be self-cleansing. However, such
applications are relatively uncommon today. Because of the
volume of flow required and the noted system limitations,
stormwater applications to date have been limited to rela-
tively small lateral sewers.
Cleansing deposited solids by flushing in combined sewer
laterals with mild slopes (0.001 to 0.008) was studied
using 30-cm (12-inch) and 46-cm (18-inch) clay sewer pipes,
each 244 meters (800 feet) long [22] . Experimental data
were then used to formulate a mathematical design model to
provide an efficient means of selecting the most economical
flushing system that would achieve a desired cleansing
efficiency within the constraints set by the engineer and
limitations of the design equations.
It was found that the cleansing efficiency of deposited
material by periodic flush waves is dependent upon flush
volume, flush discharge rate, sewer slope, sewer length,
sewer flow rate, and sewer diameter. Neither details of
the flush device inlet to the sewer nor slight irregulari-
ties in the sewer slope and alignment significantly affected
the percent cleaning efficiencies.
Using sewage instead of clean water for flushing was found
to cause a general, minor decrease in the efficiency of the
cleansing operation. The effect is relatively small and is
the result of the redeposition of solids by the trailing
edge of the flush wave.
The effects of flush wave sequencing were found to be in-
significant as long as the flush releases were made pro-
gressively from the upstream end of the sewer* Also, the
cleansing efficiencies obtained by using various combina-
tions of flush waves were found to be quite similar to
those obtained using single flushes of equivalent volumes
and similar release rates. However, both of these hypothe-
ses are based on the limited findings from tests run on
relatively short sewers. Therefore, further testing is
required to give a complete picture of the relative impor-
tance of these two factors on the overall performance of a
complete flushing system.
A prototype flush station developed during the study can be
inserted in a manhole to provide functions necessary to col-
lect sewage from the sewer, store it in a coated fabric
tank, and release the stored sewage as a flush wave upon
receipt of an external signal.
One prototype lateral flushing demonstration project was
considered for an 11-ha (27-acre) drainage area in Detroit
164
-------
has eight 152-cm (60-inch) diameter units operating in
parallel. They are to be operated sequentially to accommo-
date flow variation. The screen size is 105 microns.
Twelve static screens using 1,525-micron (0.06-inch) clear
opening screen represent the third portion of the facility.
These are the manufacturer's standard units that have been
used in industry to remove gross solids. A description of
a typical unit was presented above. The combined sewer over-
flow facilities are located across the Maumee River from
Fort Wayne's sewage treatment plant. Flows entering the
facilities are sewage treatment plant bypass and combined
sewer overflows. These flows are lifted to the screens by
pumps after passing through a bar screen. Chlorination and
a contact tank are provided.
Costs
Microstrainers and Drum Screens — The costs reported for
microstrainers vary considerably, as shown in Table 48. The
main reason is the variation in flux rates or loading coupled
with the type of waste treated (i.e., combined sewer over-
flows versus secondary effluent) [30]. With the exception
of the Philadelphia facility, all of the microstrainers are
used to treat sewage effluent at appreciably lower flux
rates which necessarily increased the cost. During the
Philadelphia study it was found possible to use a flux rate
of 73.3 cu m/hr/sq m (30 gpm/sq ft); therefore, the costs at
the three other locations listed in Table 48 have been modi-
fied to reflect this increase in loading rate. According to
the figures presented in Table 48, the average capital cost
is approximately $248/l/sec ($11,000/mgd) for treating com-
bined sewer overflows. The operation and maintenance costs
have not been adjusted. The approximate cost is $0.0013 to
$0.0026/1,000 1 ($0.005 to $0.01/1,000 gal.) for assuming 300
hours of operation per year. The single capital cost cited
for a fine screen is only the equipment cost and does not
include installation. Operation and maintenance costs
should be comparable to those for microstrainers.
Rotary Fine Screens — Cost data for rotary fine screens for
combined sewer overflows are based on a preliminary design
estimate for a screening facility in Seattle, Washington,
and actual construction costs at Fort Wayne, Indiana [38,
28]. The two costs were $700,000 and $250,000, for plants
of 1,095 I/sec (25 mgd) and 1,640 I/sec (37.5 mgd),
respectively. The differences in cost are due, in part, to
the fact that the Fort Wayne installation is a demonstration
prototype project where three types of screens operating in
parallel are treating a total flow of 3,285 I/sec (75 mgd)
in a single building. The cost for the rotary fine screen
252
-------
associated with excess wet-weather flows are generally of
short duration; thus, a marginally inadequate line can be
bolstered by polymer injections at critical periods. In
effect, this increases the overall treatment efficiency by
allowing more of the flow to reach the treatment plant,
while flooding from sewer surcharges is decreased.
The polymers tested in Richardson, Texas, included Polyox
Coagulant-701, Polyox WSR-301, and Separan AP-30 [40]. The
latter showed the greatest resistance to shear degradation
(which may be important in very long channels) but was the
least effective hydraulically. Tests conducted indicated
that the polymers and nonsolvents were not detrimental
to bacteria growth and therefore would not disrupt the
biological treatment of sewage in wastewater treatment
plants. Tests conducted on algae in a polymer environment
indicated that the polymers have no toxic effects and only
nominal nutrient effects. Fish bioassays indicated that in
a polymer slurry concentration of 500 mg/1, some fish deaths
resulted but that, in practice, concentrations above 250
mg/1 would provide no additional flow benefits. It was re-
ported that polymer concentrations of between 35 and 100
mg/1 decreased flow resistance sufficiently to eliminate
surcharges of more than 1.8 meters (6 feet) [40].
The Dallas Water Utilities District, Dallas, Texas has con-
structed a prototype polymer injection station (see
Figure 21) for relief of surcharge-caused overflows at 15
points along a 2,440-meter (8,000-foot) stretch of the
Bachman Creek sewer [36]. During storms, the infiltration
ratio approaches 8 to 1. The sanitary sewer is 46 cm
(18 inches) in diameter for the first 1,220 meters (4,000
feet) and then joins another 46-cm (18-inch) diameter
sewer and continues on. The Dallas polymer injection
station was built as a semiportable unit so that it can be
removed and installed at other locations needing an in-
terim solution once a permanent solution has been imple-
mented at Bachman Creek.
The polymer injection unit is enclosed by a 1.3-cm (1/2-inch)
steel sheet, 3.1 meters (10 feet) in diameter by approxi-
mately 7.9 meters (26 feet) in height. The upper half pro-
vides storage for 6,364 kg (14,000 Ib) of dry polymer and
also contains dehumidification equipment. The lower half
contains a polymer transfer blower, a polymer hopper and
agitator for dry feeding, a volumetric feeder and eductor,
and appurtenances. The unit is entirely self-contained
with only external power and water hookup necessary for the
unit to be in operation.
166
-------
drum speed with backwashing activation whenever headless
exceeds 6 inches. Collected solids are discharged to the
float holding tanks. The screen size used in both cases
was 297 microns.
Cleveland, Ohio [22] - The Cleveland, Ohio, study on dual-
media filtration also included a fine screen as a pretreat-
ment unit to the filtration process. The 420 micron screen
was fitted over a 1.2 sq m (12.6 sq ft) drum unit. Drum
speed and backwash conditions were not reported. More de-
tails on the layout of the facilities are given in this
section under Filtration.
East Providence, Rhode Island [41] - This bench-scale study
was conducted to test the applicability of using a:drum
screen and a diatomaceous earth filter in series to achieve
significant removals when operating on combined sewer
overflow. The study indicated good removals by the screen-
ing device in relation to other drum screens. The screening
unit, however, was of different configuration than other
drum screens. The device used was a small 259 sq cm (40
sq in.) unit consisting of a submerged rotating drum with
the flow passing through the screen from the outside to the
inside. Effluent was drawn off from the interior of the
rotating drum. The backwash system ran continuously using
submerged spray jets directed at the interior of the screen
dislodging strained solids and allowing them to pass through
ports separating dirty water from the rest of the influent
water. Synthetic sewage was used during the study. Screen
apertures tested were 150, 190, and 230 microns in size.
Portland, Oregon [58, 11] - A rotary fine screen unit was
tested in Portland, Oregon, on both dry-weather flow and
combined sewer overflows. The facility was constructed on
a 183 cm (72-inch) diameter trunk sewer serving a 10,000-ha
(25,000-acre) area. A portion of the flow was diverted to
a bypass line where it first flowed through a bar screen
before being lifted into the demonstration project by two
132 I/sec (2,100 gpm) turbine pumps. After passing through
the rotary fine screen, both the concentrated solids and
the effluent were returned to the Sullivan Gulch Pumping
Station wet well. In a typical installation on a combined
sewer overflow line, the effluent from the screens would
pass to a receiving stream after disinfection. The concen-
trated solids would be returned to an interceptor sewer.
The screening unit used a 152 cm (60-inch) diameter drum
with 74, 105, and 167 micron screens. The units were oper-
ated at flow rates ranging from 43.2 to 126.2 I/sec (1 to
2.8 mgd). The range and levels of variables tested is
listed in Table 47.
250
-------
The unit is set up for fully automatic operation and may be
started by any of three external level sensors located
458 meters (1,500 feet) upstream, at the injection site, and
458 meters (1,500 feet) downstream.
Several polymers were tested, and Polyox WSR-301 was chosen
to be used when the Bachman Creek unit becomes operational.
The polymer is expected to reduce the surcharge by 6.1 meters
(20 feet) over the first 1,220-meter (4,000-foot) length.
The maximum equipmental feed rate is expected to be 2.3
kg/min (5 Ib/min). The actual polymer feed rate will be
flow paced by the liquid level in the sewer to maintain a
polymer concentration of about 150 ppm in the sanitary sewer.
The unit is capable of supplying a dosage of 500 to 600 mg/1
if needed. It is expected that the unit will be in operation
about five times per year and that surcharge reduction will
be complete in approximately 7 minutes after start of polymer
injection (travel time in the affected reach of sewer).
The actual construction cost for the unit, including instal-
lation of the site, was $146,000 (ENR 2000). The unit was
scheduled to be operable by mid-1973 with operational per-
formance and data available one year thereafter. Maintenance
is expected to be limited to a site visit and unit exercise
every 2 months.
REGULATORS
Historically, combined sewer regulators represent an attempt
to intercept all dry-weather flows, yet automatically pro-
vide for the bypass of wet-weather flows beyond available
treatment/interceptor capacity. Initially, this was accom-
plished by constructing a low dam (weir) across the combined
sewer downstream from a vertical or horizontal orifice as
shown on Figure 22. Flows dropping through the orifices
were collected by the interceptor and conveyed to the treat-
ment facility (see Figure 19). The dams were designed to
divert up to approximately 3 times the average dry-weather
flow to the interceptor with the excess overflowing to the
receiving water. Eventually more sophisticated mechanical
regulators were developed in an attempt to improve control
over the diverted volumes. No specific consideration was
given to quality control.
Recent research, however, has resulted in several regula-
tors that appear capable of providing both quality and
quantity control via induced hydraulic flow patterns that
tend to separate and concentrate the solids from the main
flow [10, 50, 15]. Other new devices promise excellent
quantity control without troublesome sophisticated design.
168
-------
than pumping if resuspension in water is to be avoided.
This is one of the screening methods currently being tested
for combined storm overflows at Fort Wayne, Indiana [28, 43]
The installed units are to handle 767 I/sec (17.5 mgd)
using screens with openings of 1,525 microns (0.060 inch).
Advantages and Disadvantages — The four basic screening de-
vices have been developed to serve one of two types of
applications. The microstrainer is designed as a main
treatment device that can remove most of the suspended con-
taminants found in a combined sewer overflow. The other
three devices --drum screens, rotary fine screens, and hy-
draulic sieves—are basically pretreatment units designed
to remove the coarser material found in waste flows. The
advantages and disadvantages of each type are listed in
Table 46.
Description of Demonstration Projects
Philadelphia, Pennsylvania — The use of a microstrainer to
treat combined sewer overflows has been studied in
Philadelphia [26, 27]. The facility includes microstrain-
ing and disinfection. The microstrainer was a 5-foot diam-
eter by 3-foot long unit using either 35- or 23-micron
screen openings during the various tests conducted. The
drum was operated submerged at 2/3 of its depth. The com-
plete unit was equipped to automatically control drum speed
proportional to the headloss across the screen, with con-
tinuous backwash, and with an ultraviolet irradiation source
to prevent fouling of the screen by bacterial slimes. The
unit starts automatically whenever sufficient overflow
occurs. Because of the physical configuration of the sewer,
flow was pumped to the microstrainer. However, it is rec-
ommended that pumping be avoided whenever possible since
large solids that would be readily removed by microstraining
are broken up by the pumping. The study was conducted
in three phases: (1) operation of full screen area using
the 35 micron screen, (2) operation at full screen area
using 23 micron screen, and (3) operation at 20 percent
of the screen area using the 23 micron screen. The latter
was to test increased loading rates since the facility
had a limited pumping capacity. The facilities operated
approximately 40 times per year on combined sewer overflows.
Milwaukee and Racine, Wisconsin [40] — The use of fine
screens to remove most of the coarse solids at Milwaukee
and Racine has been briefly described previously under
Dissolved Air Flotation. One unit was used at Milwaukee
and six are used at Racine. They operate at a continuous
248
-------
Conventional Designs
Conventional regulators can be subdivided into three major
groups: (1) static, (2) semiautomatic dynamic, and
(3) automatic dynamic. The grouping reflects the general
trend toward the increasing degree of control and sophisti-
cation and, of course, the capital and operation and main-
tenance costs. Conventional regulator design, use,
advantages, and disadvantages are well covered in the
literature [10, 11, 32].
Static Regulators — Static regulators can be defined as
fixed-position devices allowing little or no adjustment
after construction.
Static regulators consist of horizontal or vertical fixed
orifices, manually operated vertical gates, leaping and
side-spill weirs and dams, and self-priming siphons. Typi-
cal static regulators are shown on Figure 23. With the ex-
ception of the vertical gate, which does not often move,
they have no moving parts. Thus, they provide only minimal
control, and they are least expensive to build, less costly
to operate, and somewhat less troublesome to maintain.
In view of the increasingly more stringent receiving water
discharge limitations and the rising need of providing storm
water capacity in treatment plants, it is expected that the
use of conventional static regulators will decline. System
control, to utilize maximum capacity in the interceptor,
requires flexibility virtually nonexistent with static
regulators. Maintenance, with the exception of the vertical
gate, is mostly limited to removal of debris blocking the
regulator opening.
Semiautomatic Dynamic Regulators — Semiautomatic dynamic
regulators can oe defined as those which are adjustable over
a limited range of diverted flow and contain moving parts
but are not adaptable to remote control.
The family of semiautomatic dynamic (having moving parts)
regulators consists of float-operated gates, mechanical
tipping gates, and cylindrical gates. Typical semiautomatic
dynamic regulators are shown on Figure 24. All require
separate chambers to allow access for adjustment and
maintenance. As a rule this group is more expensive to
construct and to maintain than static regulators. They are
more susceptible to malfunction from debris interfering
with the moving parts and are subject to failure due to the
corrosive environment. However, better flow control is
provided because they respond automatically to combined
sewer and interceptor flow variations. The adjustment of
170
-------
Screen cleaning — In the studies conducted on the rotary fine
screen [38, 11J, blinding (clogging of the screen) has been a
problem. Blinding has been attributed to oil, grease, and
industrial waste from a paint manufacturer. This problem is
similar to that experienced during the development of micro-
strainers. The latest study at Shore Acres, California,
solved this problem by enforcing an industrial waste ordinance
prohibiting discharge of oil wastes to the sewer system.
To improve backwashing, a solution of hot water and liquid
solvent or detergent has been found necessary to obtain ef-
fective cleaning of the screens. This may have been neces-
sary only because of the nature of the common waste encoun-
tered in both studies [38, 11]. Of the solvents tested,
acidic and alcoholic agents did not adequately clean the
screens. Alkaline agents were reported not effective by
Portland [11], but Cornell, Howland, Hayes § Merryfield [38]
reported a caustic solution was the most efficient solvent.
Chloroform, solvent parts cleaner, soluble pine oil, ZIP,
Formula 409, and Vestal Eight offered limited effectiveness.
ZEP 9658 cleaned the screens effectively, but this cleaner
was not analyzed to determine its effect on effluent water
quality. The removal of paint was done effectively only by
hand cleaning using ZEP 9658.
Screen life — In the first study [38], the average screen
life was approximately 4 hours. In a study conducted a year
later [11] using a similar unit incorporating a new screen
design and a rotational speed of 65 rpm, the average screen
life was 34 hours. Reducing the rotational speed to 55 rpm
increased the average screen life to 346 hours. Results of
a subsequent study at Shore Acres, California, indicate that
screen life may exceed 1 year. This extended life, however,
is most likely attributable to the much lower hydraulic
loading rate, 39.5 versus 123 I/sec (0.9 versus 2.8 mgd) or
30 versus 97 1/sec/sq m (44 versus 143 gpm/sq ft). The
present predicted screen life is 1,000 hours. Some screen
failures were attributed to punctures caused by objects
present in the feed waters.
Design parameters — The design and operating parameters of
the rotary fine screen are presented in Table 45. No mathe-
matical modeling of the rotary fine screen has been
performed. Further tests of the rotary fine screen are
needed to determine more accurately the life of the screens,
the removal efficiencies, and design parameters.
Two points should be remembered with respect to rotary fine
screens: (1) waste flows from the rotary fine screen range
from 10 to 20 percent of the total flow treated and may
contain solvents that may be difficult to treat downstream;
246
-------
(a) AUTOMATIC SEVER REGULATOR [32]
(BRO»K AND BROWN TTPE I).
STOP DISC BOLT
STOP LINK
(b) TIPPING GATE REGULATOR [ll]
USED BY HLEGHENT COUNTY
SEWAGE AUTHORITT
6»TE CHAIBER
(c) CYLINDRICAL GATE REGULATOR [l(j]
Figure 24. Typical semiautomatic
dynamic sewer regulators
172
-------
Table 44. RECOMMENDED MICROSTRAINER DESIGN
PARAMETERS FOR COMBINED SJEWER OVERFLOW TREATMENT
Screen opening, microns
Main treatment
Pretreatment
Screen material
Drum speed, rpm
Maximum speed
Operating range
Flux rate of submerged
screen, gpm/sq ft
Low rate
High rate
Headless, in.
Submergence of drum, I
Backwash
Volume, % of inflow
Pressure, psi
Type of automatic
controls
23-35
150-420
Stainless steel
5-7
0-max speed
5-10
20-50
6-24
60-80
<0.5-3
40-50
Drum speed propor-
tional to headless
Note: gpm/sq ft x 0.679 = 1/sec/sq m
psi x 0.0703 = kg/sq cm
SCREEN
-BACKWASH NOZZLES
NFLUENT FLOW
AUTOMATIC VALVE
^SCREENED EFFLUENT
Figure 49. Rotary fine screen schematic [11]
244
-------
6ALV. PIPE TO
FLOAT WELL
LOAT i
FLOAT WELL
NTERCEPTOR
PLAN VIEW
CYLINDER - OPERATED GATE REGULATOR
PHILADELPHIA [ll]
Figure 25. Typical automatic dynamic
sewer regulator
174
-------
Table 43. DATA SUMMARY ON MICROSTRAINERS AND
DRUM SCREENS
Location
(1)
Philadelphia,
Pa.
Milwaukee, Wis.
Cleveland, Ohio
Lebanon, Ohio
Chicago, 111.
Letchworth,
England
Lebanon, Ohio
East Providence,
R.I.
Reference
number
(2)
[27]
[26]
[26]
[26]
[26]
[40]
[22]
[6]
[23]
[35]
[6]
[41]
[41]
[41]
Screen
opening,
micron
(3)
23
23
23
23
35
297
420
23
23
23
35
150
190
230
Total,
sq ft
(4)
9.4
9.4
47.0
47.0
47.0
144.0
12.6
15.0
314.0
47. 0
15.0
0.28
0.28
0.28
Screen area
Submerged,
sq ft
(5)
7.4
7.4
28a
35a
35a
72-90
NA
9
NA
NA
9
0.28
0.28
0.28
Backwash
Submerged,
t
(6)
78
78
60a
74a
74a
51-64
NA
60
NA
NA
60
ioof
ioof
ioof
Flux rate,
gpm/sq ft
(7)
40
25
9.1
6.9
5.4
40-50
100
•\-7.0
6.6
3.1
-w.o
18-25
18-25
18-25
Headloss ,
in.
(8)
23
12a
4.7a
3.6a
3.4a
12-14
max
NA
6 max
6 max
NA
6 max
NA
NA
NA
Pressure,
psi
(9)
40
40
40
40
40
1*
NA
NA
20-55
NA
NA
NA
NA
NA
\ of
total flow
(10)
<0.5
NAb
NA
NA
NA
0.8SB
NA
5.3
3.0
NA
5.3
28
28f
28
Table 43 continued on page 243.
242
-------
CONTROL PORT
ELEVATED EXIT WEIR
COMBINED
SEWER
OUTFALL
COMBINED FLOW
WEIR
COMMUNICATION LINES
FIXED AREA ORIFICE
SIMPLE LEVEL SENSOR
AIR SLOT
INTERCEPTOR FLOW
Figure 26. Schematic arrangement
of a fluidic sewer regulator [10]
is induced by the kinetic energy of the sewage entering
the tank (see Figure 27). Flow to the treatment plant is
deflected, and discharges through a pipe at the bottom near
the center of the channel. Excess flow in storm periods
discharges over a circular weir around the center of the
tank and is conveyed to receiving waters. The rotary motion
causes the sewage to follow a long path through the channel
thus setting up secondary flow patterns which create an
interface between the fluid sludge mass and the clear liquid
The flow containing the concentrated solids is directed to
the interceptor. Using synthetic sewage in model studies at
Bristol, England, suspended solids removal efficiencies of
up to 98 percent were reported [47] . Another series of
experiments elsewhere on a model vortex regulator using raw
sewage indicated poor performance in removing screenable
solids under certain conditions [1]. This lack of overflow
176
-------
preliminary and more work is needed to verify them at a
larger scale and at the Philadelphia pilot plant site.
Screen cleaning — Of the several conditions which affect the
operation of the microstrainer and drum screen, the most
notable is proper cleaning of the screen. Spray jets,
located on the outside of the screen at the top of the drum,
are directed in a fan shape onto the screen. It has been
found that the pressure of this backwash spray is more im-
portant than the quantity of the backwash [13, 6], There
does not seem to be any relationship between the volume of
backwash water applied and the hydraulic loading of the
microstrainer or drum screen. Thus, a constant backwash
rate can be applied regardless of the hydraulic loading [23].
Results of tests at Philadelphia have indicated no backwash-
ing problems.
Occasionally the microstrainer and, to a lesser degree, the
drum screen cannot be effectively cleaned by the backwash
jets. This condition, called "blinding" of the screen, is
generally associated with oil, grease, and bacterial growths
[13, 41, 23, 6]. Oil and grease cannot be removed effec-
tively without using a detergent or other chemical, such as
sodium hypochlorite, in the backwash water [6], Generally,
microstrainers and drum screens with the finer screen open-
ings (<147 microns) should not be used in situations where
excessive oil and grease concentrations are likely to be en-
countered from a particular drainage area. Bacterial growths
also have caused blinding problems on microstrainers, although
they have not been a major problem with drum screens. The use
of ultraviolet light is an effective means of control, as men-
tioned previously. It is important, however, to use an
ultraviolet light source of the proper frequency designed to
minimize the amount of ozone created [29] . With proper con-
struction of the microstrainer it is possible to reduce the
chances of the creation of ozone [26] .
Screen life — In a wet environment, ozone is relatively cor-
rosive to the stainless steel screens. Since screens are
woven with very fine stainless steel wires, the amount of
corrosion needed to break through a strand of the wire is
small [29]. In fact, it has been reported that ozone in a
wet environment is more corrosive to the stainless steel
wires than chlorine in a wet environment [29]. Therefore,
it is important to reduce the concentration of ozone and/or
chlorine in and around the microstrainer. Both chlorine and
ozone have been used upstream of the microstrainer, but
enough detention time has been allowed so that concentra-
tions of these chemicals are relatively low. It is better
240
-------
regulator and the swirl regulator/concentrator is the flow
field pattern. Another major difference is that larger
flow rates can be handled in the prototype swirl regulator/
concentrator (at Lancaster, Pennsylvania, the estimated in-
crease is 4 to 6 times greater) than in the equivalent size
vortex regulator.
A hydraulic laboratory model was used to determine geometric
configuration and settleable solids removal efficiencies.
Figure 28 shows the hydraulic model in action. Note the
solids separation and concentration toward the underflow
pipe to the treatment plant.
As a result of both mathematical and hydraulic modeling,
the performance of the prototype has been predicted. Based
upon a peak storm flow to peak dry-weather flow ratio of
55 to 1, 90 percent of the solids (grit particles with a
specific gravity of 2.65, having a diameter greater than
0.3 mm and settleable solids with a specific gravity of 1.2,
having a diameter larger than 1.0 mm) are concentrated into
3 percent of the flow [50, 15]. Hydraulic testing indicates
that removal efficiency increases as the peak storm flow to
peak dry-weather flow decreases. The recommended configura-
tion for the swirl regulator/concentrator is shown on
Figure 29.
The foul-sewer channel in the bottom of the swirl concentra-
tor is sized for peak dry-weather flow. During wet-weather
flows the concentrated settleable solids are carried out
the foul-sewer into an interceptor.
There are no moving parts so maintenance and adjustment re-
quirements are minimal. Fine tuning control is provided via
a separate chamber with a cylinder gate on the "foul sewer"
outlet to the interceptor. Remote control, although not
readily adaptable, could be accomplished by providing a
larger-than-necessary foul sewer (also diminishes the
chances of clogging) and throttling this line with a re-
motely controlled gate.
Spiral Flow Regulator — The spiral flow regulator is based
on the concept of using the secondary helical motion im-
parted to fluids at bends in conduits to concentrate the
settleable solids in the flow. A bend with a total angle
between 60 and 90 degrees is employed. Hydraulic model
studies of this device, carried out at the University of
Surrey, England [44], indicated that this is a feasible
178
-------
size of the screen openings because this determines the
initial size of particles removed. The efficiencies of a
microstrainer and drum screens treating a waste with a nor-
mal distribution of particle sizes will increase as the size
of screen opening decreases. Suspended solids removals re-
ported in various studies within the United States bear this
out, as shown on Figure 48 [41, 26, 40, 22, 35, 27, 23]. In
reality, however, removals are based on the relative sizes
between the screen opening and the particle size. A drum
screen with a large screen opening can achieve high removals
if the majority of the solids in the waste flows are larger
than the screen opening. It appears important not to pump
ahead of microstrainers because this tends to break up frag-
ile particles and thereby reduce removal efficiencies. The
use of positive displacement pumps or spiral pumps may be
permissible.
The second most important condition affecting removal effi-
ciencies, especially for microstrainers, is the thickness of
filtered material on the screen. Whenever the thickness of
this filter mat is increased, the suspended matter removal
160
so
-T 60
40
20
-o
8
100 200 300 400
SCREEN OPENING, MICRONS
500
Figure 48. SS removal versus screen opening
238
-------
B ^
FLIITIILE:
DEFLECTOR. i 270.
FLOW DEFLECTOR
A-A
FLOW DEFLECTOR
B -J
B-B
PLAN
SECTIONS
Figure 29, Recommended configuration
for swi rl concentrator [50]
means of separating solids from the overflow. The simplest
form of the regulator is shown on Figure 30.
The heavily polluted sewage is drawn to the inner wall. It
then passes to a semicircular channel situated at a lower
level leading to the treatment plant. The proportion of
the concentrated discharge will depend on the particular
design. The overflow passes over a side weir for discharge
to the receiving waters. Surface debris collects at the
end of the chamber and passes over a short flume to join
the sewer conveying the flow to the treatment plant.
The authors of the model study report that even the sim-
plest application of the spiral flow separator will produce
an inexpensive regulator that will be superior to many
existing types. They also stated that further research is
necessary to define the variables, the limits of applica-
tions, and the actual limitations of the spiral flow
180
-------
BACKWASH
HOOD
Figure 47. Schemati c of a
microstrainer or drum screen
The drum was completely submerged within an influent tank,
and flow passed inward through the circumference of the
drum. Submerged backwash jets were placed inside the drum.
Screen openings for microstrainers range from 15 to 65
microns and for drum screens, from 100 to 600 microns. The
various sizes of screen openings that have been tested on
combined sewer overflows, and other data, are listed in
Table 42.
Microstrainers and drum screens can be used in many differ-
ent treatment schemes. Their versatility comes from the
fact that the removal efficiency is adjustable by changing
the aperture size of the screen placed on the unit. The
primary use of microstrainers would be in lieu of a sedimen
tation basin to remove suspended matter. They can also be
used in conjunction with chemical treatment, such as ozone
or chlorine for chemically disinfecting/oxidizing both
organic and nonorganic oxidizable matter or microorganisms
236
-------
regulator [44]. A prototype regulator has been success-
fully evaluated at Nantwich, England. A third generation
device is being developed for American practice.
Stilling Pond Regulator - The stilling pond regulator, as
used in England, is a short length of widened channel from
which the settled solids are discharged to the interceptor
[1]. Flow to the interceptor is controlled by the discharge
pipe which is sized so that it will be surcharged during
wet-weather flows. Its discharge will depend on the sewage
level in the regulator. Excess flows during storms dis-
charge over a transverse weir and are conveyed to the re-
ceiving waters. The use of the stilling pond may provide
time for the solids to settle out when the first flush of
stormwater arrives at the regulator and before discharge
over the weir begins.
This type of regulator is considered suitable for overflows
up to 85 I/sec (30 cfs). If the stilling pond is to be suc-
cessful in separating solids, it is suggested that not less
than a 3-minute retention be provided at the maximum rate
of flow [34].
High Side-Spill Weirs - Unsatisfactory experience with
side-spill weirs in England has led to the development of a
high side-spill weir, referred to there as the high double
side-weir overflow. These weirs are made shorter and higher
than would be required for the normal side-spill weir. The
rate of flow to the treatment plant may be controlled by use
of a throttle pipe or a float-controlled mechanical gate.
The ratio of screenings in the overflow to screenings in the
sewage passed on to treatment was 0.5, the lowest of the
four types investigated in England. This device has the
best general performance when compared to the English vortex
and stilling pond regulators and the low side-overflow
weir [1].
Tide Gates - Tide gates, backwater gates, or flap gates are
used to protect the interceptors and collector sewers from
high water levels in receiving waters and are considered
a regulating appurtenance when used for this purpose.
Tide gates are intended to open and permit discharge at the
outfall when the flow line in the sewer system regulator
chamber produces a small differential head on the upstream
face of the gate. Some types of gates are sufficiently
heavy to close automatically, ahead of any water level rise
in the receiving body. With careful installation and bal-
ancing, coupled with an effective preventive maintenance
program, the ability of the gate to open during overflow
182
-------
the same as for their use in dry-weather treatment
facilities. The reader is referred to the literature for
the necessary details [25]. Except for bar screens, their
use for combined sewer overflows may be limited. Coarse
screens are used as a pretreatment and protection device at
the Cottage Farm Detention and Chlorination Facility in
Boston. Bar screens are recommended for almost all storage
and treatment facilities and pump stations for protection of
downstream equipment. Typical screenings from a 1-inch bar
screen are shown on Figure 46.
Fine Screens and Microscreens — Fine screens and micro-
screens are discussed together because in most cases they
operate in a similar manner. The types of units found in
these classifications are rotating fine screens, hereinafter
referred to as drum screens; microscreens, commonly called
microstrainers; rotary fine screens; and hydraulic sieves
(static screens); vibrating screens; and gyratory screens.
To date, vibrating screens and gyratory screens have not
been used in prototype combined sewer overflow treatment
facilities.
Description of Screening Devices
Microstrainers and Drum Screens - The microstrainer and drum
screen are essentially the same device but with different
screen aperture sizes. A schematic of a typical unit is
shown on Figure 47. They are a mechanical filter using a
variable, low-speed (up to 4 to 7 rpm), continuously back-
washed, drum rotating about a horizontal axis and operating
under gravity conditions. The filter usually is a tightly
woven wire mesh fabric (called screen) fitted on the drum
periphery in paneled sections. The drum is placed in a tank,
and wastewater enters one end of the drum and flows outward
through the rotating screen. Seals at the ends of the drum
prevent water from escaping around the ends of the drum into
the tank. As the drum rotates, filtered solids, trapped on
the screen, are lifted above the water surface inside the
drum. At the top of the drum, the solids are backwashed off
the screen by high-pressure spray jets, collected in a trough,
and removed from the inside of the drum. In most cases,
both the rotational speed of the drum and the backwash rate
are adjustable. Backwash water is usually strained effluent.
The newer microstrainers use an ultraviolet light irradia-
tion source alongside the backwash jets to prevent growth of
organisms on the screens [36] . The drum is submerged from
approximately two-thirds to three-quarters of its diameter.
As noted previously, the usual flow pattern is radially out-
ward through the screen lining the drum; however, one drum
screen application used a reverse flow pattern [41].
234
-------
Regulators and their appurtenant facilities
should be recognized as devices which have the
dual responsibility of controlling both quantity
and quality of overflow to receiving waters, in
the interest of more effective pollution
control. [50]
As mentioned previously, new regulator devices have been
developed that provide both quantity and quality control.
These include electrode potential along with the swirl
regulator/concentrator, spiral flow regulator, vortex regu-
lator, and high side-spill weir. Thus, in the future, the
choice of a regulator must be based on several factors in-
cluding: (1) quantity control, (2) quality control,
(3) economics, (4) reliability, (5) ease of maintenance, and
(6) the desired mode of operation (automatic or
semiautomatic).
Regulator Costs - Selected installed construction costs are
shown in Table 29. These costs are to be used for order-
of-magnitude reference only because of the wide variance
of construction problems, unit sizes, location, number
of units per installation, and special appurtenances.
The cost of maintaining sewer regulators as reported in a
recent national survey also vary widely [10] . In most
cases, the reported expenditures are probably not adequate
to maintain the regulators in completely satisfactory
condition. The annual cost per regulator required to con-
duct a minimal maintenance program is listed in Table 29.
REMOTE MONITORING AND CONTROL
One alternative to the tremendous cost and disruption caused
by sewer separation is to upgrade existing combined sewer
systems by installing effective regulators, level sensors,
tide gates, rain gage networks, sewage and receiving water
quality monitors, overflow detectors, and flowmeters and
then apply computerized collection system control. Such
system controls are being developed and applied in several
U.S. cities. The concepts of control systems have been in-
troduced in Section VI. As applied to collection system
control, they are intended to assist a dispatcher (super-
visor) in routing and storing combined sewer flows to make
the most effective use of interceptor and line capacities.
As the components become more advanced and operating experi-
ence grows, system control offers the key to total inte-
grated system management and optimization.
184
-------
100
10
0.1
LEGEND:
0 SAN FRANCISCO DATA [ 15]
0 MILWAUKEE DATA [40]
A FORT SMITH DATA [17]
• DERIVED COST EQUATION
x COST CURVE FOR SEDIMENTATION
O
10 100
DESIGN CAPACITY, MGD
NOTE; MQD x 43.BOB- I/SEC
Figure 45. Construction cost versus
design capacity for dissolved air flotation, ENR 2000
232
-------
System Components and Operations
The components of a remote monitoring and control system
can be classified as either intelligence, central proces-
sing, or control.
The intelligence system is used to sense and report the
minute-to-minute system status and raw data for predictions.
Examples include flow levels, quantities, and (in some
cases) characteristics at significant locations throughout
the system; current treatment rates, pumping rates, and
gate (regulator) positions; rainfall intensities; tide
levels; and receiving water quality.
Quality observations and comparisons may assist in deter-
mining where necessary overflows can be discharged with the
least impact. The central processing system is used to com-
pile, record, and display the data. Also, on the basis of
prerecorded data and programming, the processer (computer)
may convert, for example, flow levels and gate positions
into estimates of volumes in storage, overflowing, and in-
tercepted and may compute and display remaining available
capacities to store, intercept, treat, or bypass additional
flows.
The control system provides the means of manipulating the
system to maximum advantage. The devices include remotely
operated gates, valves, inflatable dams, regulators, and
pumps. Reactions to actuated controls and changed condi-
tions (i.e., increased rainfall, pump failure, and blocked
gate), of course, are sensed by the intelligence system,
thus reinitiating the cycle.
Representative elements of a typical system are shown on
Figure 31.
Because of the frequency and repetitiveness of the sensing
and the short time span for decision-making, computers must
form the basis of the control system. The complexity of
the hydrology and hydraulics of combined systems also dic-
tates the need for extensive preprogramming to determine
cause-effect relationships accurately and to assist in eval-
uating alternative courses of action. To be most effective,
real-time operational control must be a part of an overall
management scheme included in what is sometimes called a
"systems approach."
System Control
Before storm flow collection system control can be imple-
mented, the direction, intensity, and duration of the storm
186
-------
Table 39. SUMMARY OF PERFORMANCE CHARACTERISTICS,
BAKER STREET DISSOLVED AIR FLOTATION FACILITY [16]
SAN FRANCISCO, CALIFORNIA
Effluent concentration,
mg/1
Constituent
BOD5
COD
Settleable solids
Oil and grease
Floatables
Total coliform
Fecal coliform
Total nitrogen
Orthophosphate
Color
Maximum
114.
205.
15.
26.
0.
2.4 x
2.4 x
20.
4.
22.
0
0
0
3
57
105
105*
1
45
0
Minimum
34.0
53.0
<0.1
3.3
<0.01
<30a
<30a
10.6
<0.07
2.0
Removal efficiency,
Maximum
70
77
93
63
100
>99
>99
53
99
95
.5
.0
.5
.2
.0
.99
.99
.0
.0
.0
Minimum
13
10
0
0
60
99
99
0
43
15
.5
.8
.0
.0
.0
.44
.44
.0
.4
.8
Average
46
44
47
29
95
99
99
18
80
57
.1
.4
.7
.1
.2
.92
.91
.4
.9
.3
a. MPN/100 ml
Advantages and Disadvantages
The advantages of dissolved air flotation are that (1) moder-
ately good SS and BODs removals can be achieved; (2) the
separation rate can be controlled by adjusting the amount of
air supplied; (3) it is ideally suited for the high amount
of SS found in combined sewer overflows; (4) capital cost is
moderate owing to high separation rates, high surface load-
ing rates, and short detention times; and (5) the system can
be automated. Disadvantages of dissolved air flotation in-
clude: (1) dissolved material is not removed without the
use of chemical additions; (2) operating costs are rela-
tively high compared to other physical processes;
(3) greater operator skill is required; and (4) provisions
must be made to prevent wind and rain from disturbing the
float.
230
-------
should be known so that runoff quantities may be anticipated
Thus, the rain gage network forms an integral part of the
system. Once the storm starts affecting the collection sys-
tem, the flow quantity and movements must be known for
decision-making, control implementation, and checking out
the system response. The advantages of knowing whether or
not an overflow is occurring are obvious, but consider the
added advantage of knowing at the same time that the feeder
line is only half full and/or that the interceptor/treatment
works are operating at less than full capacity. By initi-
ating controls, say closing a gate, the control supervisor
can force the feeder line to store flows until its capacity
is approached, or can increase diversion to the interceptor,
or both. If he guesses wrong, the next system scan affords
him the opportunity to revise his strategy accordingly.
Thus, system control or management converts the combined
sewer system from an essentially static system to a dynamic
system where the elements can be manipulated or operated as
changing conditions dictate.
The degree of automatic control or computer intelligence
level varies among the different cities. For example, in
Cincinnati, monitoring to detect unusual or unnecessary
overflows is applied and has been evaluated as being
successful [5]. In Minneapolis-St. Paul, the Metropolitan
Sewer Board is utilizing a central computer that receives
telemetered data from rain gages, river level monitors,
sewer flow and level sensors, and mechanical gate diversion
points to assist a dispatcher in routing stormwater flows
to make effective use of in-line sewer storage capacity [2].
The use of rain gages, level sensors, overflow detectors,
and a central computer to control pump stations and selected
regulating gates is underway in Detroit [3]. The Munici-
pality of Metropolitan Seattle (METRO) is incorporating the
main features of the above projects plus additional water
quality monitoring functions [30] . The City and County of
San Francisco have embarked on the initial phase of a system
control project for which the ultimate goal is complete
hands-off computerized automatic control. They are cur-
rently collecting data on rainfall and combined sewer flows
to aid in the formulation of a system control scheme. More
details of the San Francisco system are described in
Section XIII under Master Plan Examples. The main differ-
ence between the San Francisco and Seattle projects, besides
size, is hands-off versus hands-on automatic supervisory
control [16].
As an example of a complex "systems approach" to collection
system control, various aspects of the Seattle master plan
are discussed in detail below.
188
-------
Table 38. TYPICAL REMOVALS ACHIEVED WITH
SCREENING/DISSOLVED AIR FLOTATION
Without chemicals With chemicals
Constituents Effluent, mg/1 % Removal Effluent, mg/1 % Removal
ss
vss
BOD
COD
Total N
Total P
81-106
47a
29-102
123a
4.2-16.8
1.3-8.8
56
53a
41
41a
14
16
42
18
12
46
4.2
0.5
-48
-29
-20
-83
-15.9
-5.6
77
70
57
45
17
69
a. Only one set of samples.
grit and most of the nonfloatable material successfully.
The system used the split flow method for dissolving air
into the flow. Approximately 20 percent of the total flow
was pressurized to 2.8 to 3.5 kg/sq cm (40 to 50 psi) in a
packless saturation tank,then remixed with the remainder
of the flow for one minute in a mixing chamber. Flow then
entered the flotation cell for flotation and removal of the
floating matter (float) by scrapers. The float was col-
lected in a holding tank for discharge back to the dry-
weather interceptor.
Racine, Wisconsin — The Racine prototype facilities are
essentially the same design as the one in Milwaukee. It,
however, is constructed partly underground out of concrete.
There are two plants: one 615 I/sec (14 mgd) and the other
1,925 I/sec (44 mgd) in size. Flow to each plant passes
through a 2.5 cm (1-inch) bar screen before being lifted to
the fine screens by screw pumps. Each plant is built with
multiple flotation tanks to accommodate a high flow
variation. A separate air saturation tank and pump serves
each flotation tank. Flow into the flotation tanks is
controlled by weirs which allow sequential filling of only
as many tanks as are necessary to handle the flow.
228
-------
3. Short-term weather prediction would be obtained by
rain gages located throughout the METRO drainage
area.
Water quality studies — Since 1963, METRO has been engaged
in a comprehensive water quality monitoring program through-
out the entire metropolitan drainage area. Upon receipt
of the CATAD demonstration grant in 1967, additional spe-
cialized water quality monitoring studies were added to the
existing program to concentrate on certain areas that con-
tribute to combined sewer overflows.
The objectives of the demonstration grant water quality
studies were twofold. First, new water quality studies
were begun or old programs modified to show how receiving
water quality and other dynamic system parameters have
changed during the periods of expansion, interception, regu-
lation, and separation. Second, a base level for various
parameters was to be established to be used as a tool for
measuring the results of the CATAD demonstration project.
The studies have been divided into two general areas re-
lated to the collection system itself and the receiving
waters adjacent to the municipality. Weather and other
pertinent environmental factors are correlated with data
from the two main study categories.
Overflow sampling was divided into three categories: physi-
cal and chemical sampling, bacteriological sampling, and
overflow volume computation.
Examples of a typical sewer sampling station and receiving
water sampling and monitoring station are shown on
Figure 15 (a, b, c).
System Operation — The CATAD system controls comprise a
computer-based central facility for automatic control of
remote regulator and pumping stations. The control center
is located at the METRO office building with satellite
terminals at the West Point and Renton treatment plants.
The principal features of the control center include a
computer, its associated peripheral equipment, an operators
console, map display, and logging and events printers [23].
Remote monitoring and control units have been provided for
36 remote pumping and regulator stations. Twenty-four re-
mote control units have been installed at pumping and
regulator stations on the trunk and interceptor sewers
leading to the West Point sewage treatment plant and nine
remote control units have been installed at pumping stations
along the interceptor sewers transporting primarily sanitary
190
-------
and bled off since oxygen has a higher solubility than
nitrogen. Finally, the pressure release valve and the
discharge line from the saturation tank should be designed
to induce good mixing with the remainder of the flow and
promote fine bubble formation [1].
Overflow Rate — The removals achieved by dissolved air flota
tion are governed by several factors. The most critical
design parameter is the surface overflow rate which can be
easily translated into the rise rate of the particle and
air bubble. To remove an air particle with a given rise
rate,the corresponding overflow rate must not be exceeded.
In rough terms, it has been reported that overflow rates
above 6.1 cu m/hr/sq m (3,600 gpd/sq ft) start to reduce
removal efficiencies. Below this value the removals remain
relatively constant.
Dissolved Air Requirements — Also important in affecting
removals is the amount of air dissolved. An insufficient
supply of dissolved air reduces the amount of air available
for each solid particle,and thus the difference between the
air-particle density and the density of water is not great
enough to meet the minimum rise rate. Also, the better the
atomization or bubble coverage over the plan area of the
tank, the better the chance for collision between the
bubbles and the suspended particles. The amount of air
supplied to a split flow flotation facility is dependent on
the percentage of flow saturated with air and the pressure.
In the studies using combined sewer overflows, the optimum
value for the percentage of flow pressurized averages around
20. In one study with a full flow system, removals were
found to be directly related to the pressure used in the
saturation tank, see Figure 44 [17] . The optimum pressure
is 3.5 to 4.2 kg/sq cm (50 to 60 psi) which agrees with
other studies performed [40, 2].
Flotation Aids — Probably, the most controllable factor
affecting particle removals is the amount and type of chemi-
cals added. In all studies, some kinds of chemicals were
added to improve removals. In one case, small floating
beads were used in lieu of air to provide the flotation [12]
This proved to be unsuccessful. The majority of chemicals
added, however, were polyelectrolytes and ferric chloride.
Ferric chloride has been reported to be the most successful
and has improved SS removals by more than 30 percent. The
use of polyelectrolytes alone and in one case bentonite clay
with polyelectrolytes has not resulted in important in-
creases in removal efficiencies. Lime and alum have also
been used.
226
-------
Table 30. TYPICAL CATAD REGULATOR STATION MONITORING
HOURLY LOG
11/14/72 1000 W POINT SYS HOURLY LOG REGULATOR STATION
TRKLVL TRKSET TIDE OUTPOS OVRFLO TRKFLO STOFLO UNUSTO
INTLVL INTSET REGPOS DIVFLO UPSFLO DNSFLO EXPHAZ
LOG DENNY RS
LUN DENNY
KING RS
CONN RS
LANDER RS
2 HANFORD
100.
0.
RS
100.
94.
105.
97.
101.
97.
102.
98.
RS
100.
98.
76
00
02
73
23
70
24
46
27
91
97
81
96.
109.
96.
102.
106.
101.
106.
102.
105.
102.
56
88 105.89
56
105.34
40
37 109.38
35
01 106.06
75
23 104.81
75
0
100
-0
100
-0
99
-0
99
-0
99
0
100
.9
.9
.2
.5
.1
.3
.2
.8
.1
.4
.0
.0
0.0
3.6
0.0
10.6
0.0
3.4
0.0
3.1
0.0
6.3
0.0
5.7
3
10
32
3
1
3
31
6
28
6
20
.7
.6
.6
.5
.4
.1
.6
.3
.6
.4
.8
0.1
47.0
0.1
4.9
0.0
34.7
0.0
35.1
0.7
26.6
0.14
0.03
-0.5
0.31
0.57
2.15
BRANDON RS
MICHIGAN
CHELAN RS
HARBOR RS
W MICH RS
8TH SOUTH
DEXTER RS
L CITY RS
1 HANFORD
102.
98.
RS
101.
100.
101.
100.
108.
108.
116.
107.
RS
100.
98.
136.
134.
150.
114.
RS
101.
95.
37
96
50
30
56
53
38
08
46
41
49
12
56
28
36
33
61
40
105.
100.
105.
101.
107.
103.
109.
108.
99.
144.
137.
157.
108.
93 105.59
40
69 105.35
65
98 105.61
21
106.09
13
37
105.76
58
34
75
06
05
-0
102
-0
102
0
100
-0
99
0
99
0
100
36
100
-0
.1
.9
.4
.9
.1
.3
.2
.5
.0
.9
.3
.4
.3
.1
.7
0.0
0.0
0.0
0.0
0.0
4.4
0.0
0.9
0.0
0.7
0.0
2.8
0.0
4.2
0.0
13.4
0.0
0
14
0
12
4
0
0
3
2
4
13
.0
.0
.0
.0
.4
.9
.7
.1
.8
.1
.4
14.0
12.0
2.7
0.9
3.8
2.2
-0.1
4.2
38.9
1.09
-3.6
192
-------
(f)
Figure 43. Dissolved air flotation facilities (Racine)
(a) Overview of
Iight roof add i t
(p ressu r i za t i on)
site during construction (b) Overview of flotation tanks after
on (c) Fine drum screen pretreatment units (d) Air saturation
tanks (e) End of float drawoff (helical cross conveyor) (f)
Float holding tanks (for temporary storage before feedback to interceptor)
224
-------
The design of the METRO interceptor system provides a posi-
tive means for controlling these bypassed flows. A regula-
tor station (Figure 32) at each major trunk sewer controls
both the diversion of combined sewage into the interceptor
and the overflow from the trunk (sewage in excess of the
capacity of the interceptor). The volume of flow diverted
to the interceptor is automatically controlled by modulating
the regulator gate position in response to changes in the
level of sewage in the interceptor. As the level in the
interceptor rises above a preset maximum, the regulator gate
closes to reduce the volume of diverted flow and maintain
the preset level. Storm flow in excess of the diverted flow
is stored in the trunk sewer and the level of the sewage in
the trunk commences to rise. When the level rises above a
preset maximum, the outfall gate will open automatically to
discharge the excess storm flow and modulate to maintain
the preset maximum level in the trunk.
Accomplishments — The most demonstrative method of pointing
out accomplishments is to show the results of interception
of an actual storm. Two days of CATAD printouts were ob-
tained from METRO, one set for the storm flow that occurred
on November 25, 1972, and the second set for the dry-weather
flow on November 14, 1972. The dry-weather flow data were
used to establish an approximate dry-weather flow base for
comparison purposes. The particular regulator station
analyzed is the Denny-Lake Union (identified as LUN DENNY RS
in the CATAD printouts). A sample storm log is shown in
Table 33. The data included in this log are the rainfall
occurring and the maximum rainfall rate during the hour,
the maximum overflow rate and the overflow volume occurring
during the hour, and the total overflow volume from the
start of the overflow. A 16-hour period from 0700 hours
to 2300 hours was used for the comparison. From the data,
hydrographs were generated which yielded a dry-weather
flow volume of 140,540 cu m (37.13 mil gal.) and a wet-
weather flow volume of 204,650 cu m (54.07 mil gal.). The
potential overflow volume then is the difference between
the two or 64,120 cu m (16.94 mil gal.). The amount of
actual overflow from the station allowed by the CATAD system
was 11,660 cu m (3.08 mil gal.). Thus the effective storm
runoff containment for this particular storm and regulator
station was approximately 82 percent.
Several improvements have been observed in Elliott Bay fol-
lowing the August 1970 interception and regulation of 12
major combined sewer overflows which are that reductions in
coliform levels range from 63 to 98 percent and that moni-
toring indicates an improvement of between 2 and 3 mg/1 of
dissolved oxygen in the bay.
194
-------
split flow flotation where part of the incoming flow is
pressurized and mixed with air before being recombined with
the remaining flow and entering the flotation tank. And
the last is the recycle system in which a portion of the
effluent is pressurized before being returned and mixed with
the incoming flow. The last two methods are used for the
larger size units since they require only a portion of
the total flow to be pressurized. In combined sewer over-
flow treatment studies the split flow method has been used
because the flotation tank only needs to be designed for the
actual flow arriving at the plant and need not include any
recycled flow. However, subsequent laboratory studies have
indicated better removals may be achieved by using the re-
cycle type of dissolved air flotation [39].
Typical facilities consist of saturation tanks to dissolve
air into a portion of the flow, a small mixing chamber to
recombine the flow that has been pressurized with that which
has not, and flotation tanks or cells. In most flotation
cells, two sets of flight scrapers, top and bottom, are used.
These remove the accumulated float and settled sludge. At
two major combined sewer overflow study sites, however, the
bottom scrapers were not used. Instead, 50-mesh rotating
fine screens ahead of the dissolved air flotation units re-
moved the coarser material in the waste flows, thus elimi-
nating the majority of settleable material. A schematic of
the dissolved air flotation facilities at Racine, Wisconsin,
is shown on Figure 42. Photographs of a typical dissolved
air flotation facility are shown on Figure 43.
Design Criteria
The principal parameters that affect removal efficiencies
are (1) overflow rate, (2) amount of air dissolved in the
flows, and (3) chemical addition. Chemical addition has
been used to improve removals, and ferric chloride has been
the chemical most commonly added.
Any one of several methods may be used to size a dissolved
air flotation facility. Values for design parameters used
in the combined sewer overflow studies are listed in
Table 37. The most commonly used design equation is that
recommended by the American Petroleum Institute [1].
When designing dissolved air flotation, regardless of whether
by formulated equations found in the literature or by the
design parameters listed previously in Table 37, certain
items should be remembered. First, the saturation tank
should be a packless chamber to prevent solids plugging or
buildup and second, excess amounts of air should be supplied
222
-------
Section IX
STORAGE
Storage is, perhaps, the most cost effective method avail-
able for reducing pollution resulting from combined sewer
overflows and to improve management of urban stormwater
runoff. As such, it is the best documented abatement meas-
ure in present practice. Storage, with the resulting sedi-
mentation that occurs, can also be thought of as a treatment
process.
Storage facilities possess many of the favorable attributes
desired in combined sewer overflow treatment: (1) they are
basically simple in design and operation; (2) they respond
without difficulty to intermittent and random storm behavior;
(3) they are relatively unaffected by flow and quality
changes; and (4) they are capable of providing flow equali-
zation and, in the case of tunnels, transmission.
Frequently they can be operated in concert with regional
dry-weather flow treatment plants for benefits during both
dry- and wet-weather conditions. Finally, storage facili-
ties are relatively fail-safe and adapt well to stage
construction. Drawbacks of such facilities are related pri-
marily to their large size (real estate requirements), cost,
and visual impact. Also, access to treatment plants or
processes for dewatering, washdown, and solids disposal is
required.
Storage facilities presently in operation have been sized
on the basis of one or more of several possible criteria.
The facilities should: (1) provide a specified detention
time for runoff from a storm of a given duration or return
frequency; (2) contain a given volume of runoff from the
tributary area, such as the first 1.27 cm (1/2 inch) of
runoff; (3) contain the runoff from a given volume of rain,
such as the runoff from 1.27 cm (1/2 inch) of rain; or
(4) contain a specified volume. Because storage facilities
are generally designed to also function as sedimentation
and/or disinfection tanks, a major advantage is the SS reduc-
tion of any overflows from the storage units. Particular
196
-------
Other basic cost data for the sedimentation facilities have
been presented previously in Table 34 in Section IX, Storage
The data are scattered and no acceptable curve can be de-
rived from them. To assist in evaluating the data, four
estimates were made using cost curves from the literature
[18]. The resulting points form the curve shown on
Figure 41. The curve is intended to give only an indication
of the relative magnitude of construction costs for sedi-
mentation facilities.
DISSOLVED AIR FLOTATION
Introduction
Dissolved air flotation is a unit operation used to separate
solid particles or liquid droplets from a liquid phase.
Separation is brought about by introducing fine air bubbles
into the liquid phase. As the bubbles attach to the solid
particles or liquid droplets, the buoyant force of the com-
bined particle and air bubble is great enough to cause the
particle to rise. Once the particles have floated to the
surface, they are removed by skimming. The principal reason
for using dissolved air flotation is because the relative
difference between the specific gravity of the combined par-
ticle and air bubble and the effective specific gravity of
water is made significantly higher and is more controllable
than using plain sedimentation. Thus, according to Stokes'
Law, the velocity of the particle and air bubble is greater
than the particle itself because of the greater difference
in specific gravities. In engineering terms this means
higher overflow rates and shorter detention times can be
used for dissolved air flotation than for conventional
settling.
This process has a definite advantage over gravity sedimen-
tation when used on combined sewer overflows in that parti-
cles with densities both higher and lower than the liquid
can be removed in one skimming operation. Dissolved air
flotation also aids in the removal of oil and grease which
are not as readily removed during sedimentation.
There are two basic processes for forming the air bubbles:
(1) dissolve air into the waste stream under pressure,then
release the pressure to allow the air to come out of solu-
tion, and (2) saturate the waste with air at atmospheric
pressure,then apply a vacuum over the flotation tank to re-
duce the pressure allowing the bubbles to form. The first
process is used most commonly. There are three methods for
implementing the first process. The first method is the
full flow method where all the flow is pressurized and mixed
with air before entering the flotation tank. The second is
220
-------
Seattle, Washington - First operational in late 1971, the
system presently has 10 fully equipped regulator stations.
such as the one shown on Figure 32, with 3 more under design.
All stations are monitored and are designed so that they may
be operated by a supervisor from a central control console.
Fully automated control will be attempted in 1973. The
estimated maximum safe storage in the trunklines and inter-
ceptors is 121.1 Ml (32 mil gal.), or roughly equivalent to
0.13 cm (0.05 inches) of direct runoff from the combined
sewer and partially separated sewer areas. Interceptor
capacity is generally 3 times the estimated year 2000 dry-
weather flow. Under supervisory operation, overflows have
been reduced in volume by approximately 52 percent.
Minneapolis-St..Paul, Minnesota — This system, operational
since April 1969, is quite similar to that in Seattle, ex-
cept that inflatable Fabridams are used in place of the
motor-operated outfall gates, as also shown on Figure 32.
Fifteen Fabridams, operated by low pressure air, are
located in the major trunks, which are 1.52 to 3.66 meters
(5 to 12 feet) in diameter, immediately downstream of the
regulator gates. Normally, they are kept in a fully in-
flated condition forming a dam to approximately mid-height
of the conduits. When storm flows are sufficiently large
so as to threaten to surcharge the trunk sewers, as indi-
cated by the flow depth monitoring, the Fabridams may be
deflated remotely from the control center. On the trunks
where they are installed, the total overflow volume reduc-
tion has been estimated to range from 35 to 70 percent,
depending on the nature of the storm event [7]. Based upon
a comparison of pre- and post-project conditions, the number
of overflows was reduced 58 percent (from 281 to 117) and
the total overflow duration was reduced 88 percent (from
1,183 hours to 147 hours) from April 1969 to May 1970. A
major accomplishment of the plan has been the almost total
capture of the contaminated spring thaw runoff.
Detroit, Michigan - The Detroit Metropolitan Water Service
(DMWS) has installed the nucleus of a sewer monitoring and
remote control system for controlling combined sewer over-
flows from many small storms to the Detroit and Rouge rivers
[1]. This system includes telemeter-connected rain gages,
sewer level sensors, overflow detectors, a central computer,
a central data logger, and a central operating console for
pumping stations and selected regulating gates. The cost of
the system was slightly over $2.7 million. This system has
enabled DMWS to apply such pollution control techniques as
storm flow anticipation, first flush interception, selective
retention, and selective overflowing.
198
-------
each tank. Settled solids are resuspendcd by using sub-
merged eductors along the walls of the tanks. Each eductor
is directed to the center channel and is activated after the
tanks have been partially drained.
Saginaw, Michigan - The Weiss Street facility, a storage/
clarifier combination with a 13,620 cu m (3.6 mil gal.)
storage capacity, is currently under construction. The first
of three tanks is designed to capture the grit and most of
the heavier suspended matter. It operates in series with the
remaining two tanks which operate in parallel. The tank
bottoms are sloped and troughed to aid in the removal of the
settled solids. The tanks are washed down under manual con-
trol after each storm using spray nozzles mounted along walk-
ways next to the tanks. The first tank also has a clamshell
bucket to remove grit. The last two tanks have horizontal
screens placed below the water level just in front of the
overflow weirs. A baffle is used to ensure water flows up-
ward through the screens before overflowing the weir.
Operation
It is interesting to note that in all the storage/
sedimentation projects, settled sludge is stored until
after the storm event. At this time, the contents in the
tanks, including the solids, are slowly drained back to the
interceptor. The notable exception to this procedure is at
Chippewa Falls, Wisconsin, where solids are removed from the
basin after dewatering using a front end loader or an ordi-
nary street sweeper for disposal at a sanitary landfill.
Several different methods for resuspending or removing the
settled solids are used at the various other storage/
sedimentation facilities. At the Humboldt Avenue facility
in Milwaukee, Wisconsin, mechanical mixers are used to re-
suspend the settled solids. Traveling bridges with hy-
draulic nozzles are used at the Spring Creek plant in New
York City. At the Cottage Farm facility, a fixed water
spray in conjunction with a sloped and troughed floor is
used to flush the solids out of the basins.
The use of tube settlers and separators has been limited
mainly to water treatment facilities and some secondary
clarifiers at municipal sewage treatment plants. Their use
in the storm overflow facilities is found presently only at
the Bachman Stormwater Plant in Dallas. To date, the oper-
ating data for this plant are insufficient for reaching any
conclusions regarding the effectiveness of tube settlers
for storm overflows [5, 4].
218
-------
The operator, upon receiving advance information on storms
from a remote rain gage, increases the treatment plant pump-
ing rate. This lowers the surcharged interceptor gradient
and allows for greater interceptor storage capacity and
conveyance. This practice has enabled DMWS to contain and
treat many intense spot storms entirely, in addition to many
scattered citywide rains.
Off-Line
Typical off-line storage devices can range from lagoons [18],
to huge primary settling tank-like structures [10, 2], to
underground silos [8], to underwater bags [4], to void space
storage, to deep tunnels [5], and mine labyrinths. In
almost all cases, feedback of the retained flows to the
sanitary system for ultimate disposal is proposed or
practiced. The underground and offshore storage has been
proposed to meet the severe land area and premium cost
constraints.
Chippewa Falls, Wisconsin — A 36.45-ha (90-acre) combined
sewer area of this Wisconsin community has been served by
a 10.6-M1 (2.8-mil gal.) open storage lagoon since 1969 [18],
The storage volume is equivalent to 2.92 cm (1.15 inches) of
runoff from the tributary area. A plan of the retention
basin is shown on Figure 33. In the two-year period 1969-
1970, the lagoon was 93.7 percent effective in capturing
overflow volumes. During this period, the combined sewer
overflows from 59 of 62 storms were totally contained by the
basin. Flow storage in the basin up to 12 hours caused no
adverse odor problems. The basin was paved with 5.08 cm
(2 inches) of asphalt, and the most effective cleaning of
solids was through the use of conventional street sweepers.
The basin is dewatered to an existing activated sludge
plant after storms with no adverse effect on the biological
treatment process. Secondary clarifier capacity, however,
had to be doubled to avoid excessive loss of solids during
sustained high flows.
Akron, Ohio — An underground 2.7-M1 (0.7-mil gal.) capacity
storage facility has been constructed in Akron, Ohio (see
Figure 34), utilizing the concept of void space storage [17].
The basin is trapezoidal in cross section (3:1 side slopes)
with top dimensions of 61 meters (200 feet) by 61 meters
(200 feet) and a usable depth of 3.4 meters (11 feet). It
serves a 76-ha (188.5 acre) combined sewered area. The rock
fill material completely filling the basin in which the com-
bined sewage is to be temporarily stored is washed gravel,
graded from 6.3 to 8.9 cm (2-1/2 to 3-1/2 inches) in
diameter. The effective void space is approximately 33 per-
cent of the total volume. The fill is completely enclosed
200
-------
Although sedimentation may be the preferred method for re-
moving SS from combined sewer overflows, it is not always
the most economical. The primary limitations are the large
size of sedimentation facilities, the long detention times,
and the low removal efficiency for colloidal matter [7].
Two solutions to these limitations are: (1) combining
the sedimentation process with storage facilities, which is
usually done simply by the nature of the storage configura-
tion, and (2) using tube settlers or separators to reduce
the detention time and improve SS removals. Several
storage/sedimentation facilities have been constructed and
operated with apparent success (see Table 35). Relatively
little operational information is available, however.
Further data will be available as some of the ongoing proj-
ects are completed.
Table 35. SUMMARY DATA ON SEDIMENTATION BASINS
COMBINED WITH STORAGE FACILITIES
Location of facility
Cottage Farm Detention
and Chlorination Facility,
Cambridge, Mass.
Chippewa Falls, Wis.
Columbus , Ohio
Whittier Street
Alum Creek
Humboldt Ave. ,
Milwaukee, Wis.
Spring Creek Jamaica
Bay, New York, N.Y.
Mount Clemens, Mich.
Lancaster, Pa.
Weiss Street,
Saginaw, Mich.
Size
mil
gal
1.3
2.8
4.0
0.9
4.0
10.0
b.
5.8
1.2
3.6
, Removal efficiency
Type of
storage facility SS 8005, I
a. In operation
Covered concrete 45 Erratic
tanks
Asphalt paved 18-70 22-74
storage basin
Open concrete 15-45 15-35
tanks
Covered concrete NA NA
tank
Covered concrete NA NA
tanks
Covered concrete NA NA
tanks
In planning or construction phase
Concrete tanks
Concrete silo
Concrete tanks
Type of solids
removal equipment
Manual washdown
Solids removal by
street cleaners
Mechanical wash-
down
Mechanical wash-
down
Resuspension of
solids by mixers
Traveling bridge
hydraulic mixers
Resuspension of
solids and mechan-
ical washdown by
eductors
Air agitation and
pumping
Mechanical and
manual washdown
a. All facilities store solids during storm event and clean sedimentation basin when flows to
the interceptor can handle the solid water and solids.
b. NA = not available.
Note: mil gal. x 3,785.0 = cu m
216
-------
DIVERSION L COMBINED SEWER
VERTICAL BAR SCREEN AND
S8LIDS HOLDING CHAMBER
OVERFLOW TO STORM SEWER
STORAGE BEB
ROMPED TO SANITARY SEWER
Figure 34.
Schematic of detention facilities,
Akron, Ohio [17]
202
-------
Table 34. SUMMARY OF STORAGE COSTS
FOR VARIOUS CITIESa
Location
Seattle, Wash. [14]
Control and monitoring system
Automated regulator stations
Minneapolis-St. Paul, Minn. [7]
Chippewa Falls, Wis. [18]
Storage
Treatment
Akron, Ohio [17]
Oak Lawn, 111. [16]
Melvina Ditch Detention
Reservoir
Jamaica Bay, New York City,
N.Y. [10,12]
Basin
Sewer
Humboldt Avenue, Milwaukee,
Wis. [3, 13]
Boston, Mass. [6]
Cottage Farm Stormwater
Treatment Station
Chicago, 111. [9]
Reservoirs
Collection, tunnel, and
pumpingc
Reservoirs and tunnels
Treatment
Sandusky, Ohio [19]
Washington, D.C. [4]
Storage,
mil gal. Capital cost, $
3,500,000
3,900,000
32.0 7,400,000
3,000,000
2.8 744,000
186,000
2.8 950,000
0.7 441,000
53.7 1,388,000
10.0 21,200,000
13.0
23.0 21,200,000
4.0 2,010,000
1.3 6,200,000
2,736.0 568,000,000
2,834,0 755,000
5,570.0 1,323,000,000
1,550,000,0-00
5,570.0 2,873,000,000
0.2 535,000
0.2 883,000
Cost per
acre,
$/acre
5,550
8,260
2,070
10,330
2,340
540
6,530
6,530
3,560
2,370
3,150
5,500
6,460
11,960
36,000
29,430
Storage
cost ,
$/gal.
0.23
--
0.26
0.26
0.62
0.03
2.12
0.92
0.50
4.74b
0.21
0.27
0.24
0.24
2.67
4.41
Annual
operation and
maintenance
cost, $
250,000
--
2,500
7,200
9,700
23,300
--
50,000
65,000
--
8,700,000
6,380
3,340
a. ENR = 2000.
b. Includes pumping station, chlorination facilities, and outfall.
c. Includes 193.1 km (120 miles) of tunnels.
Note: $/acre x 2.47 = $/hectare; $/gal. x 0.264 = $/l; mil gal. x 3.785 = Ml
214
-------
(d)
Figure 35. Jamaica Bay (Spring Creek auxiliary water
pollution control plant) retention basin
(a) Exterior view of facility at discharge to receiving water (b) Traveling bridge
collector with drop pipe assemblies (c) Closeup of bridge showing typical drop
pipe and header for directing high pressure sprays on floor (d) One of six 50-ft
wide bays, looking at inlet ports
204
-------
underwater storage of combined sewer overflows from a
12.15-ha (30-acre) drainage area [4]. The plant consisted
of a treatment facility (grit removal, bar rack, and comminu-
tion), underwater storage tanks, and the associated instru-
mentation, pumping, and control systems. Two 0.38-Ml (0.1-
mil gal.) tanks were anchored underwater in the Anacostia
River. The tanks were standard pillow tanks made of nylon-
reinforced synthetic rubber.
Portions of the overflow from a combined sewer overflow line
were diverted to the pilot plant. The flow entered the grit
chamber of the pilot plant, then passed through a bar screen
followed by a comminution before entering the underwater
storage tanks. Material removed in the grit chambers was
returned to the interceptor. After the storm subsided and
during nonpeak hours, liquid was pumped from the tanks into
the interceptor for transport to the dry-weather treatment
plant. To prevent settlement of solids in the storage tanks,
compressed air was forced into the tanks to agitate any
settled sludge and to enable pumping out of all the contents
of the storage tanks to the interceptor.
Of the numerous operational problems encountered during this
project, the major one was clogging of the effluent port by
the flexible tank top membrane during dewatering.
Sandusky, Ohio - A 0.76-M1 (0.2-mil gal.) capacity under-
water storage facility was constructed and tested in
Sandusky, Ohio. The facility consisted of three basic sys-
tem components, the underwater storage tank with its associ-
ated piping and controls, a connecting structure to the
existing outfall, and a control building to house instru-
mentation and pump systems [19]. Two 0.38-M1 (0.1-mil gal.)
collapsible tanks serving a 6.0-ha (14.9-acre) area were
anchored underwater in Lake Erie. The tanks consist of a
steel pipe frame in the form of a modified octagon with a
nylon-reinforced synthetic rubber flexible membrane top and
bottom. A concrete pad was poured to fit the bottom con-
tours of both storage tanks. The tank top conforms to the
bottom contours when empty and the top rises upon filling.
Combined sewer overflow passes through a bar screen in a
connecting chamber to remove all trash from the overflow
before passing to the influent pipes to the tanks. A diver-
sion weir allows control of the filling of one tank or the
other. At high flow rates, both tanks fill simultaneously.
After tank capacity is reached, the flow backs up in the
connection chamber and passes out a safety overflow.
Combined sewer overflow reaching the underwater storage
tanks passes through a sedimentation control chamber over
212
-------
Figure 37. Humboldt Ave. (Milwaukee) retention basin
(a) Exterior view of the predominantly buried facility on the Milwaukee River
(b) Operations building housing ch I orination, screening, and pumping equipment
(c) Looking over tanks from operations building (d) Walking above tanks
(note tank area did not require fencing
206
-------
Chicago, Illinois - Chicago has pioneered in the development
of abandoned mine storage, deep vertical drop shafts [11],
and deep tunnels in hard rock [5] for the interception, con-
veyance, and temporary storage of combined sewer overflows.
Three construction contracts partially funded by the EPA
are presently in progress, representing a total investment
in excess of $50 million (ENR 2000) providing over 17.7 km
(11 miles) of tunnel and appurtenant drop shafts and pumping.
Typical examples are shown on Figure 40. Under the recently
adopted master plan for the 97,200-ha (240,000-acre) Greater
Chicago service area, the following are to be accomplished:
(1) treatment of all wet-weather flows at a dry-weather flow
facility sized for 1.5 average dry-weather flow maximum rate;
(2) interception of all existing wet-weather outfalls by a
deep conveyance tunnel system; and (3) storage of all inter-
cepted flows above treatment capacity at one large and two
auxiliary reservoirs until they are absorbed by the dry-
weather flow treatment facilities operating at nearly a
constant maximum rate [9]. The cost for this plan has been
estimated to be $2,873 million as opposed to $5,521 million
for complete sewer separation, and separation would not meet
the proposed water quality or flood control objectives.
The major reservoir would be a quarry 100.7 meters deep by
152.5 to 366.0 meters wide by 4.02 km long (330 feet by
500 to 1,200 feet by 2.5 miles). The combined sewer over-
flows retained in the reservoir would be aerated continu-
ously, and accumulated solids would be removed periodically
by dredging. Most storms would be dewatered in 2 to 10 days;
the largest, in 50 days. The reservoir would be dewatered
to a dry-weather treatment plant at a rate of 19.8 cu m/sec
(450 mgd) or 0.5 times average dry-weather flow. The total
storage provided would be equivalent to 7.98 cm (3.14 inches)
of runoff, 70,309,500 cu m (57,000 acre-ft), or 9 percent of
the annual average rainfall. Based upon this plan, the fre-
quency of overflows to the Chicago River was estimated to be
4 times in 21 years.
The tunnels, a total of 193 km (120 miles) ranging from 3.05
to 12.81 meters (10 to 42 feet) in diameter, would be con-
structed in dolomite rock 45.8 to 88.5 meters (150 to 290
feet) below the surface. Drop shafts would be placed at
approximately 0.80-km (1/2-mile) intervals, and a forced
ventilation system providing one air change per hour during
dewatering would be operated to control odors and gases.
The tunnels would be constructed on self-cleansing gradients
with additional provisions for flushing by introducing river
water into the tunnel.
Washington, D. C. - At Washington, D. C., a pilot plant was
built and tested to assess the feasibility of treatment and
210
-------
COMBINED
SEWAGE FLOW
(10-12xDWF)
t
WET
WELL
BAR COARSE
SCREENS? SCREENS
HYPOCHLORITE
-PUMPS VTUBES| S
x ' SOLUTION
FEED
CONTROL
GATES
Sn
X)—
OVERFLOW IN
EXCESS OF
INTERCEPTOR
CAPACITY
ALTERNATE
INTERCEPTOR
TO DRY WEATHER
TREATMENT
(4-5xDWF)
—X>—
TANK DRAIN (TYP)
| J CHLORINE
KtSIUUAL
ANALYZER ,1
^SOLIDS RETURN LOCATION ^_
AND DRAIN CONTROL GATES-<^
C3
_0
t—
f
\
H-
UJ
=3
U.
LU
DETENTION
AND
Hllf[l CONTACT TANKS
&M\ (TYPICAL TANK)
fcxrtll
^HORIZ. INLET BAFFLED
FINE SCREENS
CHLORINE RESIDUAL
ANALYZER
OUTFALL
TO RIVER
(CONTROL TO SOLUTION FEED PUMPS)
Figure 38. Schematic of Cottage Farm
detention and chlorination facility,
Boston, Massachusetts
208
-------
COMBINED
SEWAGE FLOW
(10-12xDWF)
t
WET
WELL
HYPOCHLORITE
BAR COARSE
SCREENS^ SCREENS
CONTROL
GATES
uz
CXh-
OVERFLOW IN
EXCESS OF
INTERCEPTOR
CAPACITY
'INTERCEPTOR
TO DRY WEATHER
TREATMENT
(4-5xDWF)
TANK DRAIN (TYP)
| J CHLORINE
KbSIUUAL
ANALYZER ,
^SOLIDS RETURN LOCATION—^
AND DRAIN CONTROL GATES-cC^"
C3
=3
t—
f
t—
LU
=3
__l
Ub.
LU
F^
DETENTION
Akin
AND
SA^LEO CONTACT TANKS
^*f (TYPICAL TANK)
fcv^fl
^HORIZ. INLET BAFFLED
FINE SCREENS
CHLORINE RESIDUAL
ANALYZER
•OUTFALL
TO RIVER
(CONTROL TO SOLUTION FEED PUMPS)
Figure 38. Schematic of Cottage Farm
detention and chlorination facility,
Boston, Massachusetts
208
-------
Chicago, Illinois - Chicago has pioneered in the development
of abandoned mine storage, deep vertical drop shafts [11],
and deep tunnels in hard rock [5] for the interception, con-
veyance, and temporary storage of combined sewer overflows.
Three construction contracts partially funded by the EPA
are presently in progress, representing a total investment
in excess of $50 million (ENR 2000) providing over 17.7 km
(11 miles) of tunnel and appurtenant drop shafts and pumping.
Typical examples are shown on Figure 40. Under the recently
adopted master plan for the 97,200-ha (240,000-acre) Greater
Chicago service area, the following are to be accomplished:
(1) treatment of all wet-weather flows at a dry-weather flow
facility sized for 1.5 average dry-weather flow maximum rate;
(2) interception of all existing wet-weather outfalls by a
deep conveyance tunnel system; and (3) storage of all inter-
cepted flows above treatment capacity at one large and two
auxiliary reservoirs until they are absorbed by the dry-
weather flow treatment facilities operating at nearly a
constant maximum rate [9]. The cost for this plan has been
estimated to be $2,873 million as opposed to $5,521 million
for complete sewer separation, and separation would not meet
the proposed water quality or flood control objectives.
The major reservoir would be a quarry 100.7 meters deep by
152.5 to 366.0 meters wide by 4.02 km long (330 feet by
500 to 1,200 feet by 2.5 miles). The combined sewer over-
flows retained in the reservoir would be aerated continu-
ously, and accumulated solids would be removed periodically
by dredging. Most storms would be dewatered in 2 to 10 days;
the largest, in 50 days. The reservoir would be dewatered
to a dry-weather treatment plant at a rate of 19.8 cu m/sec
(450 mgd) or 0.5 times average dry-weather flow. The total
storage provided would be equivalent to 7.98 cm (3.14 inches)
of runoff, 70,309,500 cu m (57,000 acre-ft), or 9 percent of
the annual average rainfall. Based upon this plan, the fre-
quency of overflows to the Chicago River was estimated to be
4 times in 21 years.
The tunnels, a total of 193 km (120 miles) ranging from 3.05
to 12.81 meters (10 to 42 feet) in diameter, would be con-
structed in dolomite rock 45.8 to 88.5 meters (150 to 290
feet) below the surface. Drop shafts would be placed at
approximately 0.80-km (1/2-mile) intervals, and a forced
ventilation system providing one air change per hour during
dewatering would be operated to control odors and gases.
The tunnels would be constructed on self-cleansing gradients
with additional provisions for flushing by introducing river
water into the tunnel.
Washington, D. C. - At Washington, D. C., a pilot plant was
built and tested to assess the feasibility of treatment and
210
-------
(d)
Figure 37. Humboldt Ave. (Milwaukee) retention basin
(a) Exterior view of the predominantly buried facility on the Milwaukee River
(b) Operations building housing ch I orination, screening, and pumping equipment
(c) Looking over tanks from operations building (d) Walking above tanks
(note tank area did not require fencing
206
-------
underwater storage of combined sewer overflows from a
12.15-ha (30-acre) drainage area [4]. The plant consisted
of a treatment facility (grit removal, bar rack, and comminu-
tion), underwater storage tanks, and the associated instru-
mentation, pumping, and control systems. Two 0.38-Ml (0.1-
mil gal.) tanks were anchored underwater in the Anacostia
River. The tanks were standard pillow tanks made of nylon-
reinforced synthetic rubber.
Portions of the overflow from a combined sewer overflow line
were diverted to the pilot plant. The flow entered the grit
chamber of the pilot plant, then passed through a bar screen
followed by a comminution before entering the underwater
storage tanks. Material removed in the grit chambers was
returned to the interceptor. After the storm subsided and
during nonpeak hours, liquid was pumped from the tanks into
the interceptor for transport to the dry-weather treatment
plant. To prevent settlement of solids in the storage tanks,
compressed air was forced into the tanks to agitate any
settled sludge and to enable pumping out of all the contents
of the storage tanks to the interceptor.
Of the numerous operational problems encountered during this
project, the major one was clogging of the effluent port by
the flexible tank top membrane during dewatering.
Sandusky, Ohio - A 0.76-M1 (0.2-mil gal.) capacity under-
water storage facility was constructed and tested in
Sandusky, Ohio. The facility consisted of three basic sys-
tem components, the underwater storage tank with its associ-
ated piping and controls, a connecting structure to the
existing outfall, and a control building to house instru-
mentation and pump systems [19]. Two 0.38-M1 (0.1-mil gal.)
collapsible tanks serving a 6.0-ha (14.9-acre) area were
anchored underwater in Lake Erie. The tanks consist of a
steel pipe frame in the form of a modified octagon with a
nylon-reinforced synthetic rubber flexible membrane top and
bottom. A concrete pad was poured to fit the bottom con-
tours of both storage tanks. The tank top conforms to the
bottom contours when empty and the top rises upon filling.
Combined sewer overflow passes through a bar screen in a
connecting chamber to remove all trash from the overflow
before passing to the influent pipes to the tanks. A diver-
sion weir allows control of the filling of one tank or the
other. At high flow rates, both tanks fill simultaneously.
After tank capacity is reached, the flow backs up in the
connection chamber and passes out a safety overflow.
Combined sewer overflow reaching the underwater storage
tanks passes through a sedimentation control chamber over
212
-------
(d)
Figure 35. Jamaica Bay (Spring Creek auxiliary water
pollution control plant) retention basin
(a) Exterior view of facility at discharge to receiving water (b) Traveling bridge
collector with drop pipe assemblies (c) Closeup of bridge showing typical drop
pipe and header for directing high pressure sprays on floor (d) One of six 50-ft
wide bays, looking at inlet ports
204
-------
Table 34. SUMMARY OF STORAGE COSTS
FOR VARIOUS CITIESa
Location
Seattle, Wash. [14]
Control and monitoring system
Automated regulator stations
Minneapolis-St. Paul, Minn. [7]
Chippewa Falls, Wis. [18]
Storage
Treatment
Akron, Ohio [17]
Oak Lawn, 111. [16]
Melvina Ditch Detention
Reservoir
Jamaica Bay, New York City,
N.Y. [10,12]
Basin
Sewer
Humboldt Avenue, Milwaukee,
Wis. [3, 13]
Boston, Mass. [6]
Cottage Farm Stormwater
Treatment Station
Chicago, 111. [9]
Reservoirs
Collection, tunnel, and
pumpingc
Reservoirs and tunnels
Treatment
Sandusky, Ohio [19]
Washington, D.C. [4]
Storage,
mil gal. Capital cost, $
3,500,000
3,900,000
32.0 7,400,000
3,000,000
2.8 744,000
186,000
2.8 950,000
0.7 441,000
53.7 1,388,000
10.0 21,200,000
13.0
23.0 21,200,000
4.0 2,010,000
1.3 6,200,000
2,736.0 568,000,000
2,834.0 755,000
5,570.0 1,323,000,000
1,550,000,0-00
5,570.0 2,873,000,000
0.2 535,000
0.2 883,000
Cost per
acre,
$/acre
5,550
8,260
2,070
10,330
2,340
540
6,530
6,530
3,560
2,370
3.150
5,500
6,460
11,960
36,000
29,430
Storage
cost ,
$/gal.
0.23
--
0.26
0.26
0.62
0.03
2.12
0.92
0.50
4.74b
0.21
0.27
0.24
0.24
2.67
4.41
Annual
operation and
maintenance
cost, $
250,000
--
2,500
7,200
9,700
23,300
--
--
50,000
65,000
--
8,700,000
6,380
3,340
a. ENR = 2000.
b. Includes pumping station, chlorination facilities, and outfall.
c. Includes 193.1 km (120 miles) of tunnels.
Note: $/acre x 2.47 = S/hectare; $/gal. x 0.264 = $/l; mil gal. x 3.785 = Ml
214
-------
CQMIINEB SEWER
LEAPING WEIR FOR
SANITARY FLOW T8
SANITARY SEWER
VERTICAL BAR SCREEN AND
SOLIDS HOLDING CHAMBER
TUBE
CLARIFIERS
*-€?
OVERFLOW TO STORM SEWER
STORAGE BEB
PUMPED TO SANITARY SEWER
Figure 34.
Schematic of detention facilities,
Akron, Ohio [17]
202
-------
Although sedimentation may be the preferred method for re-
moving SS from combined sewer overflows, it is not always
the most economical. The primary limitations are the large
size of sedimentation facilities, the long detention times,
and the low removal efficiency for colloidal matter [7].
Two solutions to these limitations are: (1) combining
the sedimentation process with storage facilities, which is
usually done simply by the nature of the storage configura-
tion, and (2) using tube settlers or separators to reduce
the detention time and improve SS removals. Several
storage/sedimentation facilities have been constructed and
operated with apparent success (see Table 35). Relatively
little operational information is available, however.
Further data will be available as some of the ongoing proj-
ects are completed.
Table 35. SUMMARY DATA ON SEDIMENTATION BASINS
COMBINED WITH STORAGE FACILITIES
Location of facility
Cottage Farm Detention
and Chlorination Facility,
Cambridge, Mass.
Chippewa Falls, Wis.
Columbus , Ohio
Whittier Street
Alum Creek
Humboldt Ave. ,
Milwaukee, Wis.
Spring Creek Jamaica
Bay, New York, N.Y.
Mount Clemens, Mich.
Lancaster, Pa.
Weiss Street,
Saginaw, Mich.
Size, Removal efficienc)
gal. storage facility SS 8005, %
a. In operation
1.3 Covered concrete 45 Erratic
tanks
2.8 Asphalt paved 18-70 22-74
storage basin
4.0 Open concrete 15-45 15-35
tanks
0.9 Covered concrete NAb NA
tank
4.0 Covered concrete NA NA
tanks
10.0 Covered concrete NA NA
tanks
b. In planning or construction phase
5.8 Concrete tanks
1.2 Concrete silo
3.6 Concrete tanks
Type of solids
removal equipment
Manual washdown
Solids removal by
street cleaners
Mechanical wash-
down
Mechanical wash-
down
Resuspension of
solids by mixers
Traveling bridge
hydraulic mixers
Resuspension of
solids and mechan-
ical washdown by
eductors
Air agitation and
pumping
Mechanical and
manual washdown
a. All facilities store solids during storm event and clean sedimentation basin when flows to
the interceptor can handle the solid water and solids.
b. NA = not available.
Note: mil gal. x 3,785.0 = cu m
216
-------
The operator, upon receiving advance information on storms
from a remote rain gage, increases the treatment plant pump-
ing rate. This lowers the surcharged interceptor gradient
and allows for greater interceptor storage capacity and
conveyance. This practice has enabled DMWS to contain and
treat many intense spot storms entirely, in addition to many
scattered citywide rains.
Off-Line
Typical off-line storage devices can range from lagoons [18],
to huge primary settling tank-like structures [10, 2], to
underground silos [8] , to underwater bags [4], to void space
storage, to deep tunnels [5], and mine labyrinths. In
almost all cases, feedback of the retained flows to the
sanitary system for ultimate disposal is proposed or
practiced. The underground and offshore storage has been
proposed to meet the severe land area and premium cost
constraints.
Chippewa Falls, Wisconsin - A 36.45-ha (90-acre) combined
sewer area of this Wisconsin community has been served by
a 10.6-M1 (2.8-mil gal.) open storage lagoon since 1969 [18].
The storage volume is equivalent to 2.92 cm (1.15 inches) of
runoff from the tributary area. A plan of the retention
basin is shown on Figure 33. In the two-year period 1969-
1970, the lagoon was 93.7 percent effective in capturing
overflow volumes. During this period, the combined sewer
overflows from 59 of 62 storms were totally contained by the
basin. Flow storage in the basin up to 12 hours caused no
adverse odor problems. The basin was paved with 5.08 cm
(2 inches) of asphalt, and the most effective cleaning of
solids was through the use of conventional street sweepers.
The basin is dewatered to an existing activated sludge
plant after storms with no adverse effect on the biological
treatment process. Secondary clarifier capacity, however,
had to be doubled to avoid excessive loss of solids during
sustained high flows.
Akron, Ohio — An underground 2.7-M1 (0.7-mil gal.) capacity
storage facility has been constructed in Akron, Ohio (see
Figure 34), utilizing the concept of void space storage [17],
The basin is trapezoidal in cross section (3:1 side slopes)
with top dimensions of 61 meters (200 feet) by 61 meters
(200 feet) and a usable depth of 3.4 meters (11 feet). It
serves a 76-ha (188.5 acre) combined sewered area. The rock
fill material completely filling the basin in which the com-
bined sewage is to be temporarily stored is washed gravel,
graded from 6.3 to 8.9 cm (2-1/2 to 3-1/2 inches) in
diameter. The effective void space is approximately 33 per-
cent of the total volume. The fill is completely enclosed
200
-------
each tank. Settled solids are resuspended by using sub-
merged eductors along the walls of the tanks. Each eductor
is directed to the center channel and is activated after the
tanks have been partially drained.
Saginaw, Michigan - The Weiss Street facility, a storage/
clarifier combination with a 13,620 cu m (3.6 mil gal.)
storage capacity, is currently under construction. The first
of three tanks is designed to capture the grit and most of
the heavier suspended matter. It operates in series with the
remaining two tanks which operate in parallel. The tank
bottoms are sloped and troughed to aid in the removal of the
settled solids. The tanks are washed down under manual con-
trol after each storm using spray nozzles mounted along walk-
ways next to the tanks. The first tank also has a clamshell
bucket to remove grit. The last two tanks have horizontal
screens placed below the water level just in front of the
overflow weirs. A baffle is used to ensure water flows up-
ward through the screens before overflowing the weir.
Operation
It is interesting to note that in all the storage/
sedimentation projects, settled sludge is stored until
after the storm event. At this time, the contents in the
tanks, including the solids, are slowly drained back to the
interceptor. The notable exception to this procedure is at
Chippewa Falls, Wisconsin, where solids are removed from the
basin after dewatering using a front end loader or an ordi-
nary street sweeper for disposal at a sanitary landfill.
Several different methods for resuspending or removing the
settled solids are used at the various other storage/
sedimentation facilities. At the Humboldt Avenue facility
in Milwaukee, Wisconsin, mechanical mixers are used to re-
suspend the settled solids. Traveling bridges with hy-
draulic nozzles are used at the Spring Creek plant in New
York City. At the Cottage Farm facility, a fixed water
spray in conjunction with a sloped and troughed floor is
used to flush the solids out of the basins.
The use of tube settlers and separators has been limited
mainly to water treatment facilities and some secondary
clarifiers at municipal sewage treatment plants. Their use
in the storm overflow facilities is found presently only at
the Bachman Stormwater Plant in Dallas. To date, the oper-
ating data for this plant are insufficient for reaching any
conclusions regarding the effectiveness of tube settlers
for storm overflows [5, 4].
218
-------
Seattle, Washington - First operational in late 1971, the
system presently has 10 fully equipped regulator stations.
such as the one shown on Figure 32, with 3 more under design.
All stations are monitored and are designed so that they may
be operated by a supervisor from a central control console.
Fully automated control will be attempted in 1973. The
estimated maximum safe storage in the trunklines and inter-
ceptors is 121.1 Ml (32 mil gal.), or roughly equivalent to
0.13 cm (0.05 inches) of direct runoff from the combined
sewer and partially separated sewer areas. Interceptor
capacity is generally 3 times the estimated year 2000 dry-
weather flow. Under supervisory operation, overflows have
been reduced in volume by approximately 52 percent.
Minneapolis-St.Paul, Minnesota — This system, operational
since April 1969, is quite similar to that in Seattle, ex-
cept that inflatable Fabridams are used in place of the
motor-operated outfall gates, as also shown on Figure 32.
Fifteen Fabridams, operated by low pressure air, are
located in the major trunks, which are 1.52 to 3.66 meters
(5 to 12 feet) in diameter, immediately downstream of the
regulator gates. Normally, they are kept in a fully in-
flated condition forming a dam to approximately mid-height
of the conduits. When storm flows are sufficiently large
so as to threaten to surcharge the trunk sewers, as indi-
cated by the flow depth monitoring, the Fabridams may be
deflated remotely from the control center. On the trunks
where they are installed, the total overflow volume reduc-
tion has been estimated to range from 35 to 70 percent,
depending on the nature of the storm event [7]. Based upon
a comparison of pre- and post-project conditions, the number
of overflows was reduced 58 percent (from 281 to 117) and
the total overflow duration was reduced 88 percent (from
1,183 hours to 147 hours) from April 1969 to May 1970. A
major accomplishment of the plan has been the almost total
capture of the contaminated spring thaw runoff.
Detroit, M i ch igan - The Detroit Metropolitan Water Service
(DMWS) has installed the nucleus of a sewer monitoring and
remote control system for controlling combined sewer over-
flows from many small storms to the Detroit and Rouge rivers
[1]. This system includes telemeter-connected rain gages,
sewer level sensors, overflow detectors, a central computer,
a central data logger, and a central operating console for
pumping stations and selected regulating gates. The cost of
the system was slightly over $2.7 million. This system has
enabled DMWS to apply such pollution control techniques as
storm flow anticipation, first flush interception, selective
retention, and selective overflowing.
198
-------
Other basic cost data for the sedimentation facilities have
been presented previously in Table 34 in Section IX, Storage
The data are scattered and no acceptable curve can be de-
rived from them. To assist in evaluating the data, four
estimates were made using cost curves from the literature
[18]. The resulting points form the curve shown on
Figure 41. The curve is intended to give only an indication
of the relative magnitude of construction costs for sedi-
mentation facilities.
DISSOLVED AIR FLOTATION
Introduction
Dissolved air flotation is a unit operation used to separate
solid particles or liquid droplets from a liquid phase.
Separation is brought about by introducing fine air bubbles
into the liquid phase. As the bubbles attach to the solid
particles or liquid droplets, the buoyant force of the com-
bined particle and air bubble is great enough to cause the
particle to rise. Once the particles have floated to the
surface, they are removed by skimming. The principal reason
for using dissolved air flotation is because the relative
difference between the specific gravity of the combined par-
ticle and air bubble and the effective specific gravity of
water is made significantly higher and is more controllable
than using plain sedimentation. Thus, according to Stokes'
Law, the velocity of the particle and air bubble is greater
than the particle itself because of the greater difference
in specific gravities. In engineering terms this means
higher overflow rates and shorter detention times can be
used for dissolved air flotation than for conventional
settling.
This process has a definite advantage over gravity sedimen-
tation when used on combined sewer overflows in that parti-
cles with densities both higher and lower than the liquid
can be removed in one skimming operation. Dissolved air
flotation also aids in the removal of oil and grease which
are not as readily removed during sedimentation.
There are two basic processes for forming the air bubbles:
(1) dissolve air into the waste stream under pressure,then
release the pressure to allow the air to come out of solu-
tion, and (2) saturate the waste with air at atmospheric
pressure,then apply a vacuum over the flotation tank to re-
duce the pressure allowing the bubbles to form. The first
process is used most commonly. There are three methods for
implementing the first process. The first method is the
full flow method where all the flow is pressurized and mixed
with air before entering the flotation tank. The second is
220
-------
Section IX
STORAGE
Storage is, perhaps, the most cost effective method avail-
able for reducing pollution resulting from combined sewer
overflows and to improve management of urban stormwater
runoff. As such, it is the best documented abatement meas-
ure in present practice. Storage, with the resulting sedi-
mentation that occurs, can also be thought of as a treatment
process.
Storage facilities possess many of the favorable attributes
desired in combined sewer overflow treatment: (1) they are
basically simple in design and operation; (2) they respond
without difficulty to intermittent and random storm behavior;
(3) they are relatively unaffected by flow and quality
changes; and (4) they are capable of providing flow equali-
zation and, in the case of tunnels, transmission.
Frequently they can be operated in concert with regional
dry-weather flow treatment plants for benefits during both
dry- and wet-weather conditions. Finally, storage facili-
ties are relatively fail-safe and adapt well to stage
construction. Drawbacks of such facilities are related pri-
marily to their large size (real estate requirements), cost,
and visual impact. Also, access to treatment plants or
processes for dewatering, washdown, and solids disposal is
required.
Storage facilities presently in operation have been sized
on the basis of one or more of several possible criteria.
The facilities should: (1) provide a specified detention
time for runoff from a storm of a given duration or return
frequency; (2) contain a given volume of runoff from the
tributary area, such as the first 1.27 cm (1/2 inch) of
runoff; (3) contain the runoff from a given volume of rain,
such as the runoff from 1.27 cm (1/2 inch) of rain; or
(4) contain a specified volume. Because storage facilities
are generally designed to also function as sedimentation
and/or disinfection tanks, a major advantage is the SS reduc-
tion of any overflows from the storage units. Particular
196
-------
split flow flotation where part of the incoming flow is
pressurized and mixed with air before being recombined with
the remaining flow and entering the flotation tank. And
the last is the recycle system in which a portion of the
effluent is pressurized before being returned and mixed with
the incoming flow. The last two methods are used for the
larger size units since they require only a portion of
the total flow to be pressurized. In combined sewer over-
flow treatment studies the split flow method has been used
because the flotation tank only needs to be designed for the
actual flow arriving at the plant and need not include any
recycled flow. However, subsequent laboratory studies have
indicated better removals may be achieved by using the re-
cycle type of dissolved air flotation [39].
Typical facilities consist of saturation tanks to dissolve
air into a portion of the flow, a small mixing chamber to
recombine the flow that has been pressurized with that which
has not, and flotation tanks or cells. In most flotation
cells, two sets of flight scrapers, top and bottom, are used.
These remove the accumulated float and settled sludge. At
two major combined sewer overflow study sites, however, the
bottom scrapers were not used. Instead, 50-mesh rotating
fine screens ahead of the dissolved air flotation units re-
moved the coarser material in the waste flows, thus elimi-
nating the majority of settleable material. A schematic of
the dissolved air flotation facilities at Racine, Wisconsin,
is shown on Figure 42. Photographs of a typical dissolved
air flotation facility are shown on Figure 43.
Design Criteria
The principal parameters that affect removal efficiencies
are (1) overflow rate, (2) amount of air dissolved in the
flows, and (3) chemical addition. Chemical addition has
been used to improve removals, and ferric chloride has been
the chemical most commonly added.
Any one of several methods may be used to size a dissolved
air flotation facility. Values for design parameters used
in the combined sewer overflow studies are listed in
Table 37. The most commonly used design equation is that
recommended by the American Petroleum Institute [1].
When designing dissolved air flotation, regardless of whether
by formulated equations found in the literature or by the
design parameters listed previously in Table 37, certain
items should be remembered. First, the saturation tank
should be a packless chamber to prevent solids plugging or
buildup and second, excess amounts of air should be supplied
222
-------
The design of the METRO interceptor system provides a posi-
tive means for controlling these bypassed flows. A regula-
tor station (Figure 32) at each major trunk sewer controls
both the diversion of combined sewage into the interceptor
and the overflow from the trunk (sewage in excess of the
capacity of the interceptor) . The volume of flow diverted
to the interceptor is automatically controlled by modulating
the regulator gate position in response to changes in the
level of sewage in the interceptor. As the level in the
interceptor rises above a preset maximum, the regulator gate
closes to reduce the volume of diverted flow and maintain
the preset level. Storm flow in excess of the diverted flow
is stored in the trunk sewer and the level of the sewage in
the trunk commences to rise. When the level rises above a
preset maximum, the outfall gate will open automatically to
discharge the excess storm flow and modulate to maintain
the preset maximum level in the trunk.
Accomplishments — The most demonstrative method of pointing
out accomplishments is to show the results of interception
of an actual storm. Two days of CATAD printouts were ob-
tained from METRO, one set for the storm flow that occurred
on November 25, 1972, and the second set for the dry-weather
flow on November 14, 1972. The dry-weather flow data were
used to establish an approximate dry-weather flow base for
comparison purposes. The particular regulator station
analyzed is the Denny-Lake Union (identified as LUN DENNY RS
in the CATAD printouts). A sample storm log is shown in
Table 33. The data included in this log are the rainfall
occurring and the maximum rainfall rate during the hour,
the maximum overflow rate and the overflow volume occurring
during the hour, and the total overflow volume from the
start of the overflow. A 16-hour period from 0700 hours
to 2300 hours was used for the comparison. From the data,
hydrographs were generated which yielded a dry-weather
flow volume of 140,540 cu m (37.13 mil gal.) and a wet-
weather flow volume of 204,650 cu m (54.07 mil gal.). The
potential overflow volume then is the difference between
the two or 64,120 cu m (16.94 mil gal.). The amount of
actual overflow from the station allowed by the CATAD system
was 11,660 cu m (3.08 mil gal.). Thus the effective storm
runoff containment for this particular storm and regulator
station was approximately 82 percent.
Several improvements have been observed in Elliott Bay fol-
lowing the August 1970 interception and regulation of 12
major combined sewer overflows which are that reductions in
coliform levels range from 63 to 98 percent and that moni-
toring indicates an improvement of between 2 and 3 mg/1 of
dissolved oxygen in the bay.
194
-------
(f)
Figure 43. Dissolved air flotation facilities (Racine)
(a) Overview of site during construction (b) Overview of flotation tanks after
light roof addition (c) Fine drum screen pretreatment units (d) Air saturation
(pressurization) tanks (e) End of float drawoff (helical cross conveyor) (f)
Float holding tanks (for temporary storage before feedback to interceptor)
224
-------
Table 30. TYPICAL CATAD REGULATOR STATION MONITORING
HOURLY LOG
11/14/72 1000 W POINT SYS HOURLY LOG REGULATOR STATION
TRKLVL TRKSET TIDE OUTPOS OVRFLO TRKFLO STOFLO UNUSTO
INTLVL INTSET REGPOS DIVFLO UPSFLO DNSFLO EXPHAZ
LOG DENNY RS
100.
0.
LUN DENNY RS
KING RS
CONN RS
LANDER RS
2 HANFORD
100.
94.
105.
97.
101.
97.
102.
98.
RS
100.
98.
76
00
02
73
23
70
24
46
27
91
97
81
96.
109.
96.
102.
106.
101.
106.
102.
105.
102.
56
88 105.89
56
105.34
40
37 109.38
35
01 106.06
75
23 104.81
75
0
100
-0
100
-0
99
-0
99
-0
99
0
100
.9
.9
.2
.5
.1
.3
.2
.8
.1
.4
.0
.0
0.0
3.6
0.0
10.6
0.0
3.4
0.0
3.1
0.0
6.3
0.0
5.7
3
10
32
3
1
3
31
6
28
6
20
.7
.6
.6
.5
.4
.1
.6
.3
.6
.4
.8
0.1
47.0
0.1
4.9
0.0
34.7
0.0
35.1
0.7
26.6
0.14
0.03
-0.5
0.31
0.57
2.15
BRANDON RS
MICHIGAN
CHELAN RS
HARBOR RS
W MICH RS
8TH SOUTH
DEXTER RS
L CITY RS
1 HANFORD
102.
98.
RS
101.
100.
101.
100.
108.
108.
116.
107.
RS
100.
98.
136.
134.
150.
114.
RS
101.
95.
37
96
50
30
56
53
38
08
46
41
49
12
56
28
36
33
61
40
105.
100.
105.
101.
107.
103.
109.
108.
99.
144.
137.
157.
108.
93 105.59
40
69 105.35
65
98 105.61
21
106.09
13
37
105.76
58
34
75
06
05
-0
102
-0
102
0
100
-0
99
0
99
0
100
36
100
-0
.1
.9
.4
.9
.1
.3
.2
.5
.0
.9
.3
.4
.3
.1
.7
0.0
0.0
0.0
0.0
0.0
4.4
0.0
0.9
0.0
0.7
0.0
2.8
0.0
4.2
0.0
13.4
0.0
0
14
0
12
4
0
0
3
2
4
13
.0
.0
.0
.0
.4
.9
.7
.1
.8
.1
.4
14.0
12.0
2.7
0.9
3.8
2.2
-0.1
4.2
38.9
1.09
-3.6
192
-------
and bled off since oxygen has a higher solubility than
nitrogen. Finally, the pressure release valve and the
discharge line from the saturation tank should be designed
to induce good mixing with the remainder of the flow and
promote fine bubble formation [1].
Overflow Rate — The removals achieved by dissolved air flota
tion are governed by several factors. The most critical
design parameter is the surface overflow rate which can be
easily translated into the rise rate of the particle and
air bubble. To remove an air particle with a given rise
rate,the corresponding overflow rate must not be exceeded.
In rough terms, it has been reported that overflow rates
above 6.1 cu m/hr/sq m (3,600 gpd/sq ft) start to reduce
removal efficiencies. Below this value the removals remain
relatively constant.
Dissolved Air Requirements — Also important in affecting
removals is the amount of air dissolved. An insufficient
supply of dissolved air reduces the amount of air available
for each solid particle,and thus the difference between the
air-particle density and the density of water is not great
enough to meet the minimum rise rate. Also, the better the
atomization or bubble coverage over the plan area of the
tank, the better the chance for collision between the
bubbles and the suspended particles. The amount of air
supplied to a split flow flotation facility is dependent on
the percentage of flow saturated with air and the pressure.
In the studies using combined sewer overflows, the optimum
value for the percentage of flow pressurized averages around
20. In one study with a full flow system, removals were
found to be directly related to the pressure used in the
saturation tank, see Figure 44 [17] . The optimum pressure
is 3.5 to 4.2 kg/sq cm (50 to 60 psi) which agrees with
other studies performed [40, 2].
Flotation Aids — Probably, the most controllable factor
affecting particle removals is the amount and type of chemi-
cals added. In all studies, some kinds of chemicals were
added to improve removals. In one case, small floating
beads were used in lieu of air to provide the flotation [12],
This proved to be unsuccessful. The majority of chemicals
added, however, were polyelectrolytes and ferric chloride.
Ferric chloride has been reported to be the most successful
and has improved SS removals by more than 30 percent. The
use of polyelectrolytes alone and in one case bentonite clay
with polyelectrolytes has not resulted in important in-
creases in removal efficiencies. Lime and alum have also
been used.
226
-------
3. Short-term weather prediction would be obtained by
rain gages located throughout the METRO drainage
area.
Water quality studies — Since 1963, METRO has been engaged
in a comprehensive water quality monitoring program through-
out the entire metropolitan drainage area. Upon receipt
of the CATAD demonstration grant in 1967, additional spe-
cialized water quality monitoring studies were added to the
existing program to concentrate on certain areas that con-
tribute to combined sewer overflows.
The objectives of the demonstration grant water quality
studies were twofold. First, new water quality studies
were begun or old programs modified to show how receiving
water quality and other dynamic system parameters have
changed during the periods of expansion, interception, regu-
lation, and separation. Second, a base level for various
parameters was to be established to be used as a tool for
measuring the results of the CATAD demonstration project.
The studies have been divided into two general areas re-
lated to the collection system itself and the receiving
waters adjacent to the municipality. Weather and other
pertinent environmental factors are correlated with data
from the two main study categories.
Overflow sampling was divided into three categories: physi-
cal and chemical sampling, bacteriological sampling, and
overflow volume computation.
Examples of a typical sewer sampling station and receiving
water sampling and monitoring station are shown on
Figure 15 (a, b, c).
System Operation — The CATAD system controls comprise a
computer-based central facility for automatic control of
remote regulator and pumping stations. The control center
is located at the METRO office building with satellite
terminals at the West Point and Renton treatment plants.
The principal features of the control center include a
computer, its associated peripheral equipment, an operators
console, map display, and logging and events printers [23].
Remote monitoring and control units have been provided for
36 remote pumping and regulator stations. Twenty-four re-
mote control units have been installed at pumping and
regulator stations on the trunk and interceptor sewers
leading to the West Point sewage treatment plant and nine
remote control units have been installed at pumping stations
along the interceptor sewers transporting primarily sanitary
190
-------
Table 38. TYPICAL REMOVALS ACHIEVED WITH
SCREENING/DISSOLVED AIR FLOTATION
Without chemicals With chemicals
Constituents Effluent, mg/1 I Removal Effluent, mg/1 \ Removal
ss
vss
BOD
COD
Total N
Total P
81-106
o
47a
29-102
a
123a
4.2-16.8
1.3-8.8
56
a
53E
41
0
41a
14
16
42
18
12
46
4.2
0.5
-48
-29
-20
-83
-15.9
-5.6
77
70
57
45
17
69
a. Only one set of samples.
grit and most of the nonfloatable material successfully.
The system used the split flow method for dissolving air
into the flow. Approximately 20 percent of the total flow
was pressurized to 2.8 to 3.5 kg/sq cm (40 to 50 psi) in a
packless saturation tank,then remixed with the remainder
of the flow for one minute in a mixing chamber. Flow then
entered the flotation cell for flotation and removal of the
floating matter (float) by scrapers. The float was col-
lected in a holding tank for discharge back to the dry-
weather interceptor.
Racine, Wisconsin — The Racine prototype facilities are
essentially the same design as the one in Milwaukee. It,
however, is constructed partly underground out of concrete.
There are two plants: one 615 I/sec (14 mgd) and the other
1,925 I/sec (44 mgd) in size. Flow to each plant passes
through a 2.5 cm (1-inch) bar screen before being lifted to
the fine scre.ens by screw pumps. Each plant is built with
multiple flotation tanks to accommodate a high flow
variation. A separate air saturation tank and pump serves
each flotation tank. Flow into the flotation tanks is
controlled by weirs which allow sequential filling of only
as many tanks as are necessary to handle the flow.
228
-------
should be known so that runoff quantities may be anticipated
Thus, the rain gage network forms an integral part of the
system. Once the storm starts affecting the collection sys-
tem, the flow quantity and movements must be known for
decision-making, control implementation, and checking out
the system response. The advantages of knowing whether or
not an overflow is occurring are obvious, but consider the
added advantage of knowing at the same time that the feeder
line is only half full and/or that the interceptor/treatment
works are operating at less than full capacity. By initi-
ating controls, say closing a gate, the control supervisor
can force the feeder line to store flows until its capacity
is approached, or can increase diversion to the interceptor,
or both. If he guesses wrong, the next system scan affords
him the opportunity to revise his strategy accordingly.
Thus, system control or management converts the combined
sewer system from an essentially static system to a dynamic
system where the elements can be manipulated or operated as
changing conditions dictate.
The degree of automatic control or computer intelligence
level varies among the different cities. For example, in
Cincinnati, monitoring to detect unusual or unnecessary
overflows is applied and has been evaluated as being
successful [5], In Minneapolis-St. Paul, the Metropolitan
Sewer Board is utilizing a central computer that receives
telemetered data from rain gages, river level monitors,
sewer flow and level sensors, and mechanical gate diversion
points to assist a dispatcher in routing stormwater flows
to make effective use of in-line sewer storage capacity [2].
The use of rain gages, level sensors, overflow detectors,
and a central computer to control pump stations and selected
regulating gates is underway in Detroit [3], The Munici-
pality of Metropolitan Seattle (METRO) is incorporating the
main features of the above projects plus additional water
quality monitoring functions [30] . The City and County of
San Francisco have embarked on the initial phase of a system
control project for which the ultimate goal is complete
hands-off computerized automatic control. They are cur-
rently collecting data on rainfall and combined sewer flows
to aid in the formulation of a system control scheme. More
details of the San Francisco system are described in
Section XIII under Master Plan Examples. The main differ-
ence between the San Francisco and Seattle projects, besides
size, is hands-off versus hands-on automatic supervisory
control [16] .
As an example of a complex "systems approach" to collection
system control, various aspects of the Seattle master plan
are discussed in detail below.
188
-------
Table 39. SUMMARY OF PERFORMANCE CHARACTERISTICS,
BAKER STREET DISSOLVED AIR FLOTATION FACILITY [16]
SAN FRANCISCO, CALIFORNIA
Effluent concentration,
mg/1
Constituent
BOD5
COD
Settleable solids
Oil and grease
Floatables
Total coliform
Fecal coliform
Total nitrogen
Orthophosphate
Color
Maximum
114.0
205.0
15.0
26.3
0.57
c-a
2.4 x 105
2.4 x 105a
20.1
4.45
22.0
Minimum
34.0
53.0
<0.1
3.3
<0.01
<30a
<30a
10.6
<0.07
2.0
Removal efficiency,
%
Maximum
70.5
77.0
93.5
63.2
100.0
>99.99
>99.99
53.0
99.0
95.0
Minimum
13.5
10.8
0.0
0.0
60.0
99.44
99.44
0.0
43.4
15.8
Average
46.1
44.4
47.7
29.1
95.2
99.92
99.91
18.4
80.9
57.3
a. MPN/100 ml
Advantages and Disadvantages
The advantages of dissolved air flotation are that (1) moder
ately good SS and BODs removals can be achieved; (2) the
separation rate can be controlled by adjusting the amount of
air supplied; (3) it is ideally suited for the high amount
of SS found in combined sewer overflows; (4) capital cost is
moderate owing to high separation rates, high surface load-
ing rates, and short detention times; and (5) the system can
be automated. Disadvantages of dissolved air flotation in-
clude: (1) dissolved material is not removed without the
use of chemical additions; (2) operating costs are rela-
tively high compared to other physical processes;
(3) greater operator skill is required; and (4) provisions
must be made to prevent wind and rain from disturbing the
float.
230
-------
System Components and Operations
The components of a remote monitoring and control system
can be classified as either intelligence, central proces-
sing, or control.
The intelligence system is used to sense and report the
minute-to-minute system status and raw data for predictions.
Examples include flow levels, quantities, and (in some
cases) characteristics at significant locations throughout
the system; current treatment rates, pumping rates, and
gate (regulator) positions; rainfall intensities; tide
levels; and receiving water quality.
Quality observations and comparisons may assist in deter-
mining where necessary overflows can be discharged with the
least impact. The central processing system is used to com-
pile, record, and display the data. Also, on the basis of
prerecorded data and programming, the processer (computer)
may convert, for example, flow levels and gate positions
into estimates of volumes in storage, overflowing, and in-
tercepted and may compute and display remaining available
capacities to store, intercept, treat, or bypass additional
flows.
The control system provides the means of manipulating the
system to maximum advantage. The devices include remotely
operated gates, valves, inflatable dams, regulators, and
pumps. Reactions to actuated controls and changed condi-
tions (i.e., increased rainfall, pump failure, and blocked
gate), of course, are sensed by the intelligence system,
thus reinitiating the cycle.
Representative elements of a typical system are shown on
Figure 31.
Because of the frequency and repetitiveness of the sensing
and the short time span for decision-making, computers must
form the basis of the control system. The complexity of
the hydrology and hydraulics of combined systems also dic-
tates the need for extensive preprogramming to determine
cause-effect relationships accurately and to assist in eval-
uating alternative courses of action. To be most effective,
real-time operational control must be a part of an overall
management scheme included in what is sometimes called a
"systems approach."
System Control
Before storm flow collection system control can be imple-
mented, the direction, intensity, and duration of the storm
186
-------
100
LEGEND:
0 SAN FRANCISCO DATA [ 15]
0 MILWAUKEE DATA [40]
A FORT SMITH DATA [17]
* DERIVED COST EQUATION
x COST CURVE FOR SEDIMENTATION
10 100
DESIGN CAPACITY, MGD
NOTE; USD x 43.808- L/SEC
Figure 45. Construction cost versus
design capacity for dissolved air flotation, ENR 2000
232
-------
Regulators and their appurtenant facilities
should be recognized as devices which have the
dual responsibility of controlling both quantity
and quality of overflow to receiving waters, in
the interest of more effective pollution
control. [50]
As mentioned previously, new regulator devices have been
developed that provide both quantity and quality control.
These include electrode potential along with the swirl
regulator/concentrator, spiral flow regulator, vortex regu-
lator, and high side-spill weir. Thus, in the future, the
choice of a regulator must be based on several factors in-
cluding: (1) quantity control, (2) quality control,
(3) economics, (4) reliability, (5) ease of maintenance, and
(6) the desired mode of operation (automatic or
semiautomatic).
Regulator Costs — Selected installed construction costs are
shown in Table 29. These costs are to be used for order-
of-magnitude reference only because of the wide variance
of construction problems, unit sizes, location, number
of units per installation, and special appurtenances.
The cost of maintaining sewer regulators as reported in a
recent national survey also vary widely [10] . In most
cases, the reported expenditures are probably not adequate
to maintain the regulators in completely satisfactory
condition. The annual cost per regulator required to con-
duct a minimal maintenance program is listed in Table 29.
REMOTE MONITORING AND CONTROL
One alternative to the tremendous cost and disruption caused
by sewer separation is to upgrade existing combined sewer
systems by installing effective regulators, level sensors,
tide gates, rain gage networks, sewage and receiving water
quality monitors, overflow detectors, and flowmeters and
then apply computerized collection system control. Such
system controls are being developed and applied in several
U.S. cities. The concepts of control systems have been in-
troduced in Section VI. As applied to collection system
control, they are intended to assist a dispatcher (super-
visor) in routing and storing combined sewer flows to make
the most effective use of interceptor and line capacities.
As the components become more advanced and operating experi-
ence grows, system control offers the key to total inte-
grated system management and optimization.
184
-------
the same as for their use in dry-weather treatment
facilities. The reader is referred to the literature for
the necessary details [25]. Except for bar screens, their
use for combined sewer overflows may be limited. Coarse
screens are used as a pretreatment and protection device at
the Cottage Farm Detention and Chlorination Facility in
Boston. Bar screens are recommended for almost all storage
and treatment facilities and pump stations for protection of
downstream equipment. Typical screenings from a 1-inch bar
screen are shown on Figure 46.
Fine Screens and Microscreens — Fine screens and micro-
screens are discussed together because in most cases they
operate in a similar manner. The types of units found in
these classifications are rotating fine screens, hereinafter
referred to as drum screens; microscreens, commonly called
microstrainers; rotary fine screens; and hydraulic sieves
(static screens); vibrating screens; and gyratory screens.
To date, vibrating screens and gyratory screens have not
been used in prototype combined sewer overflow treatment
facilities.
Description of Screening Devices
Microstrainers and Drum Screens — The microstrainer and drum
screen are essentially the same device but with different
screen aperture sizes. A schematic of a typical unit is
shown on Figure 47. They are a mechanical filter using a
variable, low-speed (up to 4 to 7 rpm), continuously back-
washed, drum rotating about a horizontal axis and operating
under gravity conditions. The filter usually is a tightly
woven wire mesh fabric (called screen) fitted on the drum
periphery in paneled sections. The drum is placed in a tank,
and wastewater enters one end of the drum and flows outward
through the rotating screen. Seals at the ends of the drum
prevent water from escaping around the ends of the drum into
the tank. As the drum rotates, filtered solids, trapped on
the screen, are lifted above the water surface inside the
drum. At the top of the drum, the solids are backwashed off
the screen by high-pressure spray jets, collected in a trough,
and removed from the inside of the drum. In most cases,
both the rotational speed of the drum and the backwash rate
are adjustable. Backwash water is usually strained effluent.
The newer microstrainers use an ultraviolet light irradia-
tion source alongside the backwash jets to prevent growth of
organisms on the screens [36]. The drum is submerged from
approximately two-thirds to three-quarters of its diameter.
As noted previously, the usual flow pattern is radially out-
ward through the screen lining the drum; however, one drum
screen application used a reverse flow pattern [41].
234
-------
regulator [44]. A prototype regulator has been success-
fully evaluated at Nantwich, England. A third generation
device is being developed for American practice.
Stilling Pond Regulator - The stilling pond regulator, as
used in England, is a short length of widened channel from
which the settled solids are discharged to the interceptor
[1]. Flow to the interceptor is controlled by the discharge
pipe which is sized so that it will be surcharged during
wet-weather flows. Its discharge will depend on the sewage
level in the regulator. Excess flows during storms dis-
charge over a transverse weir and are conveyed to the re-
ceiving waters. The use of the stilling pond may provide
time for the solids to settle out when the first flush of
stormwater arrives at the regulator and before discharge
over the weir begins.
This type of regulator is considered suitable for overflows
up to 85 I/sec (30 cfs). If the stilling pond is to be suc-
cessful in separating solids, it is suggested that not less
than a 3-minute retention be provided at the maximum rate
of flow [34].
High Side-Spill Weirs — Unsatisfactory experience with
side-spill weirs in England has led to the development of a
high side-spill weir, referred to there as the high double
side-weir overflow. These weirs are made shorter and higher
than would be required for the normal side-spill weir. The
rate of flow to the treatment plant may be controlled by use
of a throttle pipe or a float-controlled mechanical gate.
The ratio of screenings in the overflow to screenings in the
sewage passed on to treatment was 0.5, the lowest of the
four types investigated in England. This device has the
best general performance when compared to the English vortex
and stilling pond regulators and the low side-overflow
weir [1].
Tide Gates — Tide gates, backwater gates, or flap gates are
used to protect the interceptors and collector sewers from
high water levels in receiving waters and are considered
a regulating appurtenance when used for this purpose.
Tide gates are intended to open and permit discharge at the
outfall when the flow line in the sewer system regulator
chamber produces a small differential head on the upstream
face of the gate. Some types of gates are sufficiently
heavy to close automatically, ahead of any water level rise
in the receiving body. With careful installation and bal-
ancing, coupled with an effective preventive maintenance
program, the ability of the gate to open during overflow
182
-------
BACKWASH
HOOD
Figure 47. Schematic of a
microstrainer or drum screen
The drum was completely submerged within an influent tank,
and flow passed inward through the circumference of the
drum. Submerged backwash jets were placed inside the drum.
Screen openings for microstrainers range from 15 to 65
microns and for drum screens, from 100 to 600 microns. The
various sizes of screen openings that have been tested on
combined sewer overflows, and other data., are listed in
Table 42.
Microstrainers and drum screens can be used in many differ-
ent treatment schemes. Their versatility comes from the
fact that the removal efficiency is adjustable by changing
the aperture size of the screen placed on the unit. The
primary use of microstrainers would be in lieu of a sedimen
tation basin to remove suspended matter. They can also be
used in conjunction with chemical treatment, such as ozone
or chlorine for chemically disinfecting/oxidizing both
organic and nonorganic oxidizable matter or microorganisms
236
-------
B ^
FLIATAILES
DEFLECTOR
B -J
FLOW DEFLECTOR
A-A
FLOW DEFLECTOR
B-B
PLAN
SECTIONS
Figure 29. Recommended configuration
for swi rl concentrator [50]
means of separating solids from the overflow.
form of the regulator is shown on Figure 30.
The simplest
The heavily polluted sewage is drawn to the inner wall. It
then passes to a semicircular channel situated at a lower
level leading to the treatment plant. The proportion of
the concentrated discharge will depend on the particular
design. The overflow passes over a side weir for discharge
to the receiving waters. Surface debris collects at the
end o.f the chamber and passes over a short flume to join
the sewer conveying the flow to the treatment plant.
The authors of the model study report that even the sim-
plest application of the spiral flow separator will produce
an inexpensive regulator that will be superior to many
existing types. They also stated that further research is
necessary to define the variables, the limits of applica-
tions, and the actual limitations of the spiral flow
180
-------
size of the screen openings because this determines the
initial size of particles removed. The efficiencies of a
microstrainer and drum screens treating a waste with a nor-
mal distribution of particle sizes will increase as the size
of screen opening decreases. Suspended solids removals re-
ported in various studies within the United States bear this
out, as shown on Figure 48 [41, 26, 40, 22, 35, 27, 23]. In
reality, however, removals are based on the relative sizes
between the screen opening and the particle size. A drum
screen with a large screen opening can achieve high removals
if the majority of the solids in the waste flows are larger
than the screen opening. It appears important not to pump
ahead of microstrainers because this tends to break up frag-
ile particles and thereby reduce removal efficiencies. The
use of positive displacement pumps or spiral pumps may be
permissible.
The second most important condition affecting removal effi-
ciencies, especially for microstrainers, is the thickness of
filtered material on the screen. Whenever the thickness of
this filter mat is increased, the suspended matter removal
100
80
-T 80
40
20
-o
8
500
Figure 48. SS removal versus screen opening
238
-------
regulator and the swirl regulator/concentrator is the flow
field pattern. Another major difference is that larger
flow rates can be handled in the prototype swirl regulator/
concentrator (at Lancaster, Pennsylvania, the estimated in-
crease is 4 to 6 times greater) than in the equivalent size
vortex regulator.
A hydraulic laboratory model was used to determine geometric
configuration and settleable solids removal efficiencies.
Figure 28 shows the hydraulic model in action. Note the
solids separation and concentration toward the underflow
pipe to the treatment plant.
As a result of both mathematical and hydraulic modeling,
the performance of the prototype has been predicted. Based
upon a peak storm flow to peak dry-weather flow ratio of
55 to 1, 90 percent of the solids (grit particles with a
specific gravity of 2.65, having a diameter greater than
0.3 mm and settleable solids with a specific gravity of 1.2,
having a diameter larger than 1.0 mm) are concentrated into
3 percent of the flow [50, 15]. Hydraulic testing indicates
that removal efficiency increases as the peak storm flow to
peak dry-weather flow decreases. The recommended configura-
tion for the swirl regulator/concentrator is shown on
Figure 29.
The foul-sewer channel in the bottom of the swirl concentra-
tor is sized for peak dry-weather flow. During wet-weather
flows the concentrated settleable solids are carried out
the foul-sewer into an interceptor.
There are no moving parts so maintenance and adjustment re-
quirements are minimal. Fine tuning control is provided via
a separate chamber with a cylinder gate on the "foul sewer"
outlet to the interceptor. Remote control, although not
readily adaptable, could be accomplished by providing a
larger-than-necessary foul sewer (also diminishes the
chances of clogging) and throttling this line with a re-
motely controlled gate.
Spiral Flow Regulator - The spiral flow regulator is based
on the concept of using the secondary helical motion im-
parted to fluids at bends in conduits to concentrate the
settleable solids in the flow. A bend with a total angle
between 60 and 90 degrees is employed. Hydraulic model
studies of this device, carried out at the University of
Surrey, England [44], indicated that this is a feasible
178
-------
preliminary and more work is needed to verify them at a
larger scale and at the Philadelphia pilot plant site.
Screen cleaning - Of the several conditions which affect the
operation of the microstrainer and drum screen, the most
notable is proper cleaning of the screen. Spray jets,
located on the outside of the screen at the top of the drum,
are directed in a fan shape onto the screen. It has been
found that the pressure of this backwash spray is more im-
portant than the quantity of the backwash [13, 6], There
does not seem to be any relationship between the volume of
backwash water applied and the hydraulic loading of the
microstrainer or drum screen. Thus, a constant backwash
rate can be applied regardless of the hydraulic loading [23].
Results of tests at Philadelphia have indicated no backwash-
ing problems.
Occasionally the microstrainer and, to a lesser degree, the
drum screen cannot be effectively cleaned by the backwash
jets. This condition, called "blinding" of the screen, is
generally associated with oil, grease, and bacterial growths
[13, 41, 23, 6]. Oil and grease cannot be removed effec-
tively without using a detergent or other chemical, such as
sodium hypochlorite, in the backwash water [6], Generally
microstrainers and drum screens with the finer screen open-
ings (<147 microns) should not be used in situations where
excessive oil and grease concentrations are likely to be en-
countered from a particular drainage area. Bacterial growths
also have caused blinding problems on microstrainers, although
they have not been a major problem with drum screens. The use
of ultraviolet light is an effective means of control, as men-
tioned previously. It is important, however, to use an
ultraviolet light source of the proper frequency designed to
minimize the amount of ozone created [29]. With proper con-
struction of the microstrainer it is possible to reduce the
chances of the creation of ozone [26].
Screen life — In a wet environment, ozone is relatively cor-
rosive to the stainless steel screens. Since screens are
woven with very fine stainless steel wires, the amount of
corrosion needed to break through a strand of the wire is
small [29]. In fact, it has been reported that ozone in a
wet environment is more corrosive to the stainless steel
wires than chlorine in a wet environment [29]. Therefore,
it is important to reduce the concentration of ozone and/or
chlorine in and around the microstrainer. Both chlorine and
ozone have been used upstream of the microstrainer, but
enough detention time has been allowed so that concentra-
tions of these chemicals are relatively low. It is better
240
-------
CONTROL PORT
ELEVATED EXIT WEIR
COMBINED
SEWER
OUTFALL
COMBINED FLOW
WEIR
COMMUNICATION LINES
FIXED AREA ORIFICE
SIMPLE
AIR
LEVEL SENSOR
SLOT
INTERCEPTOR FLOW
Figure 26. Schematic arrangement
of a fluidic sewer regulator [10]
is induced by the kinetic energy of the sewage entering
the tank (see Figure 27). Flow to the treatment plant is
deflected, and discharges through a pipe at the bottom near
the center of the channel. Excess flow in storm periods
discharges over a circular weir around the center of the
tank and is conveyed to receiving waters. The rotary motion
causes the sewage to follow a long path through the channel
thus setting up secondary flow patterns which create an
interface between the fluid sludge mass and the clear liquid
The flow containing the concentrated solids is directed to
the interceptor. Using synthetic sewage in model studies at
Bristol, England, suspended solids removal efficiencies of
up to 98 percent were reported [47] . Another series of
experiments elsewhere on a model vortex regulator using raw
sewage indicated poor performance in removing screenable
solids under certain conditions [1]. This lack of overflow
176
-------
Table 43. DATA SUMMARY ON MICROSTRAINERS AND
DRUM SCREENS
Location
(1)
Philadelphia,
Pa.
Milwaukee, Wis.
Cleveland, Ohio
Lebanon, Ohio
Chicago, 111.
Letchworth,
England
Lebanon, Ohio
East Providence,
R.I.
Reference
number
(2)
[27]
[26]
[26]
[26]
[26]
[40]
[22]
[6]
[23]
[35]
[6]
[41]
[41]
[41]
Screen
opening, Total,
micron sq ft
(3) (4)
23 9.4
23 9.4
23 47.0
23 47.0
35 47.0
297 144.0
420 12.6
23 15.0
23 314.0
23 47.0
35 15.0
150 0.28
190 0.28
230 0.28
Screen area
Submerged, Submerged, Flux rate, Headloss,
S<1 £t » gpm/sq ft in.
(5) (6) (7) (8)
7.4 78 40 23
7.4 78 25 12a
28a 60a 9.1 4.7a
3Sa 74a 6.9 3.6a
35a 74a 5.4 3.4a
72-90 51-64 40-50 12-14
max
NA NA 100 NA
9 60 -\,7.0 6 max
NA NA 6.6 6 max
NA NA 3.1 NA
9 60
-------
6ALV. PIPE TO
FLOAT WELL
LOAT I
FLOAT WELL
NTERCEPTOR
PLAN VIEW
CYLINDER - OPERATED GATE REGULATOR
PHILADELPHIA [ll]
Figure 25. Typical automatic dynamic
sewer regulator
174
-------
Table 44. RECOMMENDED MICROSTRAINER DESIGN
PARAMETERS FOR COMBINED SEWER OVERFLOW TREATMENT
Screen opening, microns
Main treatment
Pretreatment
Screen material
Drum speed, rpm
Maximum speed
Operating range
Flux rate of submerged
screen, gpm/sq ft
Low rate
High rate
Headless, in.
Submergence of drum, I
Backwash
Volume, I of inflow
Pressure, psi
Type of automatic
controls
23-35
150-420
Stainless steel
5-7
0-max speed
5-10
20-50
6-24
60-80
<0.5-3
40-50
Drum speed propor-
tional to headless
Note: gpm/sq ft x 0.679 = 1/sec/sq m
psi x 0.0703 = kg/sq cm
SCREEN
-BACKWASH NOZZLES
NFLUENT FLOW
AUTOMATIC VALVE
CONTROL
PANEL
NFLOW
-SCREENED EFFLUENT
Figure 49. Rotary fine screen schematic [11]
244
-------
(a) AUTOMATIC SENER REGULATOR [32]
(BROUN AND BROIN TTPE »).
STOP DISC BOLT
STOP LINK
(b) TIPPING GATE REGULATOR [ll]
USED Bt ALLEGHENY COUNTY
SEVA6E AUTHORITY
GATE CHAMBER
(C) CYLINDRICAL GATE REGULATOR [to]
Figure 24. Typical semiautomatic
dynamic sewer regulators
172
-------
Screen cleaning - In the studies conducted on the rotary fine
screen [38, 11], blinding (clogging of the screen) has been a
problem. Blinding has been attributed to oil, grease, and
industrial waste from a paint manufacturer. This problem is
similar to that experienced during the development of micro-
strainers. The latest study at Shore Acres, California,
solved this problem by enforcing an industrial waste ordinance
prohibiting discharge of oil wastes to the sewer system.
To improve backwashing, a solution of hot water and liquid
solvent or detergent has been found necessary to obtain ef-
fective cleaning of the screens. This may have been neces-
sary only because of the nature of the common waste encoun-
tered in both studies [38, 11]. Of the solvents tested,
acidic and alcoholic agents did not adequately clean the
screens. Alkaline agents were reported not effective by
Portland [11], but Cornell, Rowland, Hayes $ Merryfield [38]
reported a caustic solution was the most efficient solvent.
Chloroform, solvent parts cleaner, soluble pine oil, ZIP,
Formula 409, and Vestal Eight offered limited effectiveness.
ZEP 9658 cleaned the screens effectively, but this cleaner
was not analyzed to determine its effect on effluent water
quality. The removal of paint was done effectively only by
hand cleaning using ZEP 9658.
Screen life - In the first study [38] , the average screen
life was approximately 4 hours. In a study conducted a year
later [11] using a similar unit incorporating a new screen
design and a rotational speed of 65 rpm, the average screen
life was 34 hours. Reducing the rotational speed to 55 rpm
increased the average screen life to 346 hours. Results of
a subsequent study at Shore Acres, California, indicate that
screen life may exceed 1 year. This extended life, however,
is most likely attributable to the much lower hydraulic
loading rate, 39.5 versus 123 I/sec (0.9 versus 2.8 mgd) or
30 versus 97 1/sec/sq m (44 versus 143 gpm/sq ft). The
present predicted screen life is 1,000 hours. Some screen
failures were attributed to punctures caused by objects
present in the feed waters.
Design parameters - The design and operating parameters of
the rotary fine screen are presented in Table 45. No mathe-
matical modeling of the rotary fine screen has been
performed. Further tests of the rotary fine screen are
needed to determine more accurately the life of the screens,
the removal efficiencies, and design parameters.
Two points should be remembered with respect to rotary fine
screens: (1) waste flows from the rotary fine screen range
from 10 to 20 percent of the total flow treated and may
contain solvents that may be difficult to treat downstream;
246
-------
Conventional Designs
Conventional regulators can be subdivided into three major
groups: (1) static, (2) semiautomatic dynamic, and
(3) automatic dynamic. The grouping reflects the general
trend toward the increasing degree of control and sophisti-
cation and, of course, the capital and operation and main-
tenance costs. Conventional regulator design, use,
advantages, and disadvantages are well covered in the
literature [10, 11, 32].
Static Regulators — Static regulators can be defined as
fixed-position devices allowing little or no adjustment
after construction.
Static regulators consist of horizontal or vertical fixed
orifices, manually operated vertical gates, leaping and
side-spill weirs and dams, and self-priming siphons. Typi-
cal static regulators are shown on Figure 23. With the ex-
ception of the vertical gate, which does not often move,
they have no moving parts. Thus, they provide only minimal
control, and they are least expensive to build, less costly
to operate, and somewhat less troublesome to maintain.
In view of the increasingly more stringent receiving water
discharge limitations and the rising need of providing storm
water capacity in treatment plants, it is expected that the
use of conventional static regulators will decline. System
control, to utilize maximum capacity in the interceptor,
requires flexibility virtually nonexistent with static
regulators. Maintenance, with the exception of the vertical
gate, is mostly limited to removal of debris blocking the
regulator opening.
Semiautomatic Dynamic Regulators — Semiautomatic dynamic
regulators can fee defined as those which are adjustable over
a limited range of diverted flow and contain moving parts
but are not adaptable to remote control.
The family of semiautomatic dynamic (having moving parts)
regulators consists of float-operated gates, mechanical
tipping gates, and cylindrical gates. Typical semiautomatic
dynamic regulators are shown on Figure 24. All require
separate chambers to allow access for adjustment and
maintenance. As a rule this group is more expensive to
construct and to maintain than static regulators. They are
more susceptible to malfunction from debris interfering
with the moving parts and are subject to failure due to the
corrosive environment. However, better flow control is
provided because they respond automatically to combined
sewer and interceptor flow variations. The adjustment of
170
-------
than pumping if resuspension in water is to be avoided.
This is one of the screening methods currently being tested
for combined storm overflows at Fort Wayne, Indiana [28, 43]
The installed units are to handle 767 I/sec (17.5 mgd)
using screens with openings of 1,525 microns (o!o60 inch).
Advantages and Disadvantages - The four basic screening de-
vices have been developed to serve one of two types of
applications. The microstrainer is designed as a main
treatment device that can remove most of the suspended con-
taminants found in a combined sewer overflow. The other
three devices--drum screens, rotary fine screens, and hy-
draulic sieves—are basically pretreatment units designed
to remove the coarser material found in waste flows. The
advantages and disadvantages of each type are listed in
Table 46.
Description of Demonstration Projects
Philadelphia, Pennsylvania - The use of a microstrainer to
treat combined sewer overflows has been studied in
Philadelphia [26, 27]. The facility includes microstrain-
ing and disinfection. The microstrainer was a 5-foot diam-
eter by 3-foot long unit using either 35- or 23-micron
screen openings during the various tests conducted. The
drum was operated submerged at 2/3 of its depth. The com-
plete unit was equipped to automatically control drum speed
proportional to the headloss across the screen, with con-
tinuous backwash, and with an ultraviolet irradiation source
to prevent fouling of the screen by bacterial slimes. The
unit starts automatically whenever sufficient overflow
occurs. Because of the physical configuration of the sewer,
flow was pumped to the microstrainer. However, it is rec-
ommended that pumping be avoided whenever possible since
large solids that would be readily removed by microstraining
are broken up by the pumping. The study was conducted
in three phases: (1) operation of full screen area using
the 35 micron screen, (2) operation at full screen area
using 23 micron screen, and (3) operation at 20 percent
of the screen area using the 23 micron screen. The latter
was to test increased loading rates since the facility
had a limited pumping capacity. The facilities operated
approximately 40 times per year on combined sewer overflows.
Milwaukee and Racine, Wisconsin [40] - The use of fine
screens to remove most of the coarse solids at Milwaukee
and Racine has been briefly described previously under
Dissolved Air Flotation. One unit was used at Milwaukee
and six are used at Racine. They operate at a continuous
248
-------
The unit is set up for fully automatic operation and may be
started by any of three external level sensors located
458 meters (1,500 feet) upstream, at the injection site, and
458 meters (1,500 feet) downstream.
Several polymers were tested, and Polyox WSR-301 was chosen
to be used when the Bachman Creek unit becomes operational.
The polymer is expected to reduce the surcharge by 6.1 meters
(20 feet) over the first 1,220-meter (4,000-foot) length.
The maximum equipmental feed rate is expected to be 2.3
kg/min (5 Ib/min). The actual polymer feed rate will be
flow paced by the liquid level in the sewer to maintain a
polymer concentration of about 150 ppm in the sanitary sewer.
The unit is capable of supplying a dosage of 500 to 600 mg/1
if needed. It is expected that the unit will be in operation
about five times per year and that surcharge reduction will
be complete in approximately 7 minutes after start of polymer
injection (travel time in the affected reach of sewer).
The actual construction cost for the unit, including instal-
lation of the site, was $146,000 (ENR 2000). The unit was
scheduled to be operable by mid-1973 with operational per-
formance and data available one year thereafter. Maintenance
is expected to be limited to a site visit and unit exercise
every 2 months.
REGULATORS
Historically, combined sewer regulators represent an attempt
to intercept all dry-weather flows, yet automatically pro-
vide for the bypass of wet-weather flows beyond available
treatment/interceptor capacity. Initially, this was accom-
plished by constructing a low dam (weir) across the combined
sewer downstream from a vertical or horizontal orifice as
shown on Figure 22. Flows dropping through the orifices
were collected by the interceptor and conveyed to the treat-
ment facility (see Figure 19). The dams were designed to
divert up to approximately 3 times the average dry-weather
flow to the interceptor with the excess overflowing to the
receiving water. Eventually more sophisticated mechanical
regulators were developed in an attempt to improve control
over the diverted volumes. No specific consideration was
given to quality control.
Recent research, however, has resulted in several regula-
tors that appear capable of providing both quality and
quantity control via induced hydraulic flow patterns that
tend to separate and concentrate the solids from the main
flow [10, 50, 15]. Other new devices promise excellent
quantity control without troublesome sophisticated design.
168
-------
drum speed with backwashing activation whenever headless
exceeds 6 inches. Collected solids are discharged to the
float holding tanks. The screen size used in both cases
was 297 microns.
Cleveland, Ohio [221 - The Cleveland, Ohio, study on dual-
media filtration also included a fine screen as a pretreat-
ment unit to the filtration process. The 420 micron screen
was fitted over a 1.2 sq m (12.6 sq ft) drum unit. Drum
speed and backwash conditions were not reported. More de-
tails on the layout of the facilities are given in this
section under Filtration.
East Providence, Rhode Island [411 - This bench-scale study
was conducted to test the applicability of using a:drum
screen and a diatomaceous earth filter in series to achieve
significant removals when operating on combined sewer
overflow. The study indicated good removals by the screen-
ing device in relation to other drum screens. The screening
unit, however, was of different configuration than other
drum screens. The device used was a small 259 sq cm (40
sq in.) unit consisting of a submerged rotating drum with
the flow passing through the screen from the outside to the
inside. Effluent was drawn off from the interior of the
rotating drum. The backwash system ran continuously using
submerged spray jets directed at the interior of the screen
dislodging strained solids and allowing them to pass through
ports separating dirty water from the rest of the influent
water. Synthetic sewage was used during the study. Screen
apertures tested were 150, 190, and 230 microns in size.
Portland. Oregon [58, 111 - A rotary fine screen unit was
tested in Portland, Oregon, on both dry-weather flow and
combined sewer overflows. The facility was constructed on
a 183 cm (72-inch) diameter trunk sewer serving a 10,000-ha
(25,000-acre) area. A portion of the flow was diverted to
a bypass line where it first flowed through a bar screen
before being lifted into the demonstration project by two
132 I/sec (2,100 gpm) turbine pumps. After passing through
the rotary fine screen, both the concentrated solids and
the effluent were returned to the Sullivan Gulch Pumping
Station wet well. In a typical installation on a combined
sewer overflow line, the effluent from the screens would
pass to a receiving stream after disinfection. The concen-
trated solids would be returned to an interceptor sewer
The screening unit used a 152 cm (60-inch) diameter drum
with 74, 105, and 167 micron screens. The units were oper-
ated at flow rates ranging from 43.2 to 126.2 I/sec (1 to
2.8 mgd). The range and levels of variables tested is
listed in Table 47.
250
-------
associated with excess wet-weather flows are generally of
short duration; thus, a marginally inadequate line can be
bolstered by polymer injections at critical periods. In
effect, this increases the overall treatment efficiency by
allowing more of the flow to reach the treatment plant,
while flooding from sewer surcharges is decreased.
The polymers tested in Richardson, Texas, included Polyox
Coagulant-701, Polyox WSR-301, and Separan AP-30 [40]. The
latter showed the greatest resistance to shear degradation
(which may be important in very long channels) but was the
least effective hydraulically. Tests conducted indicated
that the polymers and nonsolvents were not detrimental
to bacteria growth and therefore would not disrupt the
biological treatment of sewage in wastewater treatment
plants. Tests conducted on algae in a polymer environment
indicated that the polymers have no toxic effects and only
nominal nutrient effects. Fish bioassays indicated that in
a polymer slurry concentration of 500 mg/1, some fish deaths
resulted but that, in practice, concentrations above 250
mg/1 would provide no additional flow benefits. It was re-
ported that polymer concentrations of between 35 and 100
mg/1 decreased flow resistance sufficiently to eliminate
surcharges of more than 1.8 meters (6 feet) [40],
The Dallas Water Utilities District, Dallas, Texas has con-
structed a prototype polymer injection station (see
Figure 21) for relief of surcharge-caused overflows at 15
points along a 2,440-meter (8,000-foot) stretch of the
Bachman Creek sewer [36]. During storms, the infiltration
ratio approaches 8 to 1. The sanitary sewer is 46 cm
(18 inches) in diameter for the first 1,220 meters (4,000
feet) and then joins another 46-cm (18-inch) diameter
sewer and continues on. The Dallas polymer injection
station was built as a semiportable unit so that it can be
removed and installed at other locations needing an in-
terim solution once a permanent solution has been imple-
mented at Bachman Creek.
The polymer injection unit is enclosed by a 1.3-cm (1/2-inch)
steel sheet, 3.1 meters (10 feet) in diameter by approxi-
mately 7.9 meters (26 feet) in height. The upper half pro-
vides storage for 6,364 kg (14,000 Ib) of dry polymer and
also contains dehumidification equipment. The lower half
contains a polymer transfer blower, a polymer hopper and
agitator for dry feeding, a volumetric feeder and eductor,
and appurtenances. The unit is entirely self-contained
with only external power and water hookup necessary for the
unit to be in operation.
166
-------
has eight 152-cm (60-inch) diameter units operating in
parallel. They are to be operated sequentially to accommo-
date flow variation. The screen size is 105 microns.
Twelve static screens using 1,525-micron (0.06-inch) clear
opening screen represent the third portion of the facility
These are the manufacturer's standard units that have been
used in industry to remove gross solids. A description of
a typical unit was presented above. The combined sewer over
flow facilities are located across the Maumee River from
Fort Wayne's sewage treatment plant. Flows entering the
facilities are sewage treatment plant bypass and combined
sewer overflows. These flows are lifted to the screens by
pumps after passing through a bar screen. Chlorination and
a contact tank are provided.
Costs
Microstrainers and Drum Screens - The costs reported for
microstrainers vary considerably, as shown in Table 48. The
main reason is the variation in flux rates or loading coupled
with the type of waste treated (i.e., combined sewer over-
flows versus secondary effluent) [30]. With the exception
of the Philadelphia facility, all of the microstrainers are
used to treat sewage effluent at appreciably lower flux
rates which necessarily increased the cost. During the
Philadelphia study it was found possible to use a flux rate
of 73 3 cu m/hr/sq m (30 gpm/sq ft); therefore, the costs at
the three other locations listed in Table 48 have been modi-
tied to reflect this increase in loading rate. According to
the figures presented in Table 48, the average capital cost
is approximately $248/l/sec ($11,000/mgd) for treating com-
bined sewer overflows. The operation and maintenance costs
nave not been adjusted. The approximate cost is $0.0013 to
$0.0026/1,000 1 ($0.005 to $0.01/1,000 gal.) for assuming ?00
hours of operation per year. The single capital cost cited
tor a fine screen is only the equipment cost and does not
include installation. Operation and maintenance costs
should be comparable to those for microstrainers.
Rotary Fine Screens - Cost data for rotary fine screens for
combined sewer overflows are based on a preliminary design
estimate for a screening facility in Seattle, Washington,
and actual construction costs at Fort Wayne, Indiana [38,
28J. The two costs were $700,000 and $250,000, for plants
of 1,095 I/sec (25 mgd) and 1,640 I/sec (37.5 mgd) ,
respectively. The differences in cost are due, in part, to
the fact that the Fort Wayne installation is a demonstration
prototype project where three types of screens operating in
parallel are treating a total flow of 3,285 I/sec (75 mgd)
in a single building. The cost for the rotary fine screen
252
-------
with grades too flat to be self-cleansing. However, such
applications are relatively uncommon today. Because of the
volume of flow required and the noted system limitations,
stormwater applications to date have been limited to rela-
tively small lateral sewers.
Cleansing deposited solids by flushing in combined sewer
laterals with mild slopes (0.001 to 0.008) was studied
using 30-cm (12-inch) and 46-cm (18-inch) clay sewer pipes,
each 244 meters (800 feet) long [22]. Experimental data
were then used to formulate a mathematical design model to
provide an efficient means of selecting the most economical
flushing system that would achieve a desired cleansing
efficiency within the constraints set by the engineer and
limitations of the design equations.
It was found that the cleansing efficiency of deposited
material by periodic flush waves is dependent upon flush
volume, flush discharge rate, sewer slope, sewer length,
sewer flow rate, and sewer diameter. Neither details of
the flush device inlet to the sewer nor slight irregulari-
ties in the sewer slope and alignment significantly affected
the percent cleaning efficiencies.
Using sewage instead of clean water for flushing was found
to cause a general, minor decrease in the efficiency of the
cleansing operation. The effect is relatively small and is
the result of the redeposition of solids by the trailing
edge of the flush wave.
The effects of flush wave sequencing were found to be in-
significant as long as the flush releases were made pro-
gressively from the upstream end of the sewer* Also, the
cleansing efficiencies obtained by using various combina-
tions of flush waves were found to be quite similar to
those obtained using single flushes of equivalent volumes
and similar release rates. However, both of these hypothe-
ses are based on the limited findings from tests run on
relatively short sewers. Therefore, further testing is
required to give a complete picture of the relative impor-
tance of these two factors on the overall performance of a
complete flushing system.
A prototype flush station developed during the study can be
inserted in a manhole to provide functions necessary to col-
lect sewage from the sewer, store it in a coated fabric
tank, and release the stored sewage as a flush wave upon
receipt of an external signal.
One prototype lateral flushing demonstration project was
considered for an 11-ha (27-acre) drainage area in Detroit
164
-------
FILTRATION
Introduction
In the physical treatment processes, filtration is one step
finer than screening. Solids are usually removed by one or
more of the following removal mechanisms: straining, im-
pingement, settling, and adhesion. Filtration has not been
used extensively in wastewater treatment, because of rapid
clogging which is principally due to compressible solids
being strained out at the surface and lodged within the
pores of the filter media. In stormwater runoff, however,
a large fraction of the solids are discrete, noncompressible
solids that are more readily filtered [30],
Effluents from primary or secondary treatment plants and
from physical-chemical treatment facilities contain com-
pressible solids.
The discussion on filters handling discrete, noncompressible
solids is presented here.
Design Criteria
Two factors affecting removal efficiency are flux rate and
the type of solids. As one would expect, the removals are
inversely proportional to the flux rate. At high flux
rates, solids are forced through the filters reducing solids
removal efficiency. Suspended solids removals were found
to be better for inert solids (discrete, noncompressible
solids) than for volatile solids (compressible solids).
This is the same conclusion found for microstrainers.
Loading Rates - The difference between filtering compres-
sible and noncompressible solids is basically the flux rate
used. High-rate filters handling compressible solids are
normally loaded at 12.2 to 24.5 cu m/hr/sq m (5 to 10 gpm/
sq ft), whereas those handling noncompressible solids will
filter at rates up to 73.4 cu m/hr/sq m (30 gpm/sq ft).
Chemicals - Many polyelectrolytes and some coagulants have
been tested. Some polyelectrolytes have been found which
increase removals of phosphorus and nitrogen. It is
cautioned, however, that polyelectrolytes are noted for
their unpredictability and the most effective polyelectro-
lyte must be determined for each wastewater.
Demonstration Projects
Studies have been made to investigate possible filtration
techniques for combined sewer overflows. The different
254
-------
(1) replacement of broken sections, (2) insertion of various
types of sleeves or liners, (3) internal sealing of joints
and cracks with gels or slurries, and (4) external sealing
by soil injection grouting. Additional detailed information
is available in recent EPA reports on jointing materials
[13, 41, 27] and sealants [29, 13, 41, 25].
The method most commonly used to correct structural defects
and infiltration (in sections where major structural damage
is not present) is internal sealing with gels or slurries.
The use of a chemical blocking method to seal sewer cracks,
breaks, and bad joints is much more economical and feasible
than sewer replacement or the inadequate concrete flooding
method. With recent improvements in television and photo-
graphic inspection methods, sealing by chemical blocking
appears to be an even more encouraging method than
heretofore. Chemical blocking is accomplished by injecting
a chemical grout and catalyst into the crack or break. A
sealing packer is used to place the grout and catalyst.
The packer has inflatable elements to isolate the leak, an
air line for inflation, and two pipes for delivering the
chemical grout and catalyst to the packer. An example of a
packer is shown on Figure 20. During the repair the two
inflatable end sections isolate the leak and chemical grout
and catalyst are injected into the center section. Then
the center section is inflated to force the grout from
the annulus between the packer and the sewer wall into
the leak. When the repair is complete,the packer is de-
flated and moved to the next repair location.
The current use of acrylamid gels as chemical blocking
agents is restricted by their lack of strength and other
physical limitations. Recently, improved materials, such
as epoxy-based and polyurethane-based sealants, have been
developed [29] . These new sealants have exhibited suita-
bility even under conditions of erratic or intermittent
infiltration where acrylamid gels failed because of re-
peated dehydration. The only difficulty in applying the
new sealant materials has been that, because of the physi-
cal properties of the sealants, new application equipment
incorporating a mixing mechanism is required. The cost of
this new equipment is approximately the same as the exist-
ing equipment. Modification of existing packing equipment
to accept the new sealants has been found to be feasible.
Sewers may also be sealed by inserting sleeves or liners,
as discussed previously.
162
-------
screen during the overflow event and st9red in two 19-cu m
(5,000-gal.) tanks for the test filtration runs that
followed. Each tank had a mixer to keep solids in
suspension. Two pumps were then used to supply the filter
with screened water.
Removals for this filter were 65 percent for SS, 40 percent
for BODs, and 60 percent for COD [22]. The addition of
polyelectrolyte increased the SS removal to 94 percent, the
BODs removal to 65 percent, and the COD removal to 65 per-
cent. Inorganic coagulants, such as lime, alum, and ferric
chloride, did not prove as successful as polymers. Run
times averaged 6 hours at loading rates of 58.7 cu m/hr/sq m
(24 gpm/sq ft). Backwashing of the filters consisted of
alternately injecting air and water into the bottom of the
filter columns. Air volume was varied from 38.4 to
283 cu m/hr/sq m (2.1 to 15.5 scfm/sq ft) over 2.5 to 29
minutes. Backwash water volume used ranged from 1.9 to
8.6 percent of the total combined sewer overflow filtered,
with a median value of approximately 4 percent. The range
of backwash water rate used was 75.8 to 220 cu m/hr/sq m (31
to 90 gpm/sq ft) over 4 to 25 minutes.
A list of the basic design data is presented in Table 50.
Others - Two other filtration processes, fiber glass plug
filtration [24] and coal filtration [34], show some promise,
but additional research is necessary to perfect them. Other
methods, such as crazed resin filtration, upflow filtration
with garnet sand, and filtration using ultrasonically
cleaned fine screens, have not been successful and are not
considered worthy of further effort at the present time.
Advantages and Disadvantages
The advantages of dual-media filtration are that (1) rela-
tively good removals can be achieved; (2) process is versa-
tile enough to be used as an effluent polisher; (3) operation
is easily automated; and (4) small land area is necessary.
Disadvantages are that (1) costs are high; (2) dissolved
materials are not removed; and (3) storage of backwash water
is required.
Costs
Cost data were developed from a design estimate for 1.1, 2.2,
4.4, and 8.8 cu m/sec (25, 50, 100, and 200 mgd) filtration
plants at a satellite location [22]. The basic plant as en-
visioned for the cost estimate includes a low lift pump
256
-------
Identification of the drainage system includes a review of
detailed maps of the sewer system; field checks of the line,
grade, and sizes; and identification of sections and man-
holes that are bottlenecks.
To identify the scope of the infiltration, it is necessary
to measure and record both dry- and wet-weather flows at
key manholes. Groundwater elevations should be obtained
simultaneously with sewer flow measurements.
A physical survey of the entire sewer system, or that por-
tion of major concern, where every manhole is entered and
sewers are examined visually to observe the degree and
nature of deposition, flows, pipe conditions, and manhole
condition should be made. Smoke testing may reveal infil-
tration sources only under low groundwater conditions. If
the groundwater table is above the pipe, the smoke may be
lost in the water. Soil conditions and groundwater condi-
tions should also be noted.
An economic and feasibility study is necessary to determine
the locations where the greatest amount of infiltration can
be eliminated for the least expenditure of money. In some
cases, it may be most cost effective to provide additional
treatment capacity at the sewage treatment plant for the
infiltration. Cost estimates can be developed for subse-
quent correctional stages as necessary.
Cleaning — A systematic program of sewer cleaning (1) can
restore the full hydraulic capacity and self-scouring
velocity of the sewer and its ability to convey infiltra-
tion without flooding; (2) can aid in the discovery of
trouble spots, such as areas with possible breaks, offset
joints, restrictions, and poor house taps, before any sub-
stantial damage is caused; and (3) is a necessary prerequi-
site to television and photographic inspection. It is
one of the most important and useful forms of preventive
maintenance. This type of program involves periodic
cleaning on a regular, recurring basis.
By frequent hydraulic flushing of the sewers, the interval
between mechanical cleanings of the sewer can be extended.
This will be discussed in more detail later in this section.
Equipment used in cleaning falls into three general classi-
fications: (1) rodding machines, (2) bucket machines, and
(3) for small sewers, hydraulic devices. The rodding
machine, which is used most commonly, removes heavy conglom-
erations of grease and root intrusions. The bucket machine
utilizes two cables threaded between manholes. One cable
160
-------
Plant capacity Operation and
Capital maintenance
cu m/sec mgd cost, $ cost, $/yr
1.1 25 1,580,000 44,000
2.2 50 2,390,000 55,000
4.4 100 4,370,000 98,000
8.8 200 7,430,000 129,000
The cost data are based on an ENR of 2000.
The operating costs are estimated to be $0.0382/1,000 1
($0.141/1,000 gal.) for 300 hours of operating per year. The
high cost could easily be reduced, however, by designing the
system to serve also as a dry-weather effluent polisher dur-
ing periods with no storm flows.
CONCENTRATION DEVICES
Concentration devices, such as the swirl regulator/
concentrator and helical or spiral flow devices, have intro-
duced an advanced form of sewer regulator--one capable of
controlling both quantity and quality. These devices have
been previously described in Section VIII. A prototype
swirl regulator has recently been constructed in Syracuse,
New York. A second generation swirl concentrator has
been placed into operation as a treatment unit for municipal
sewage grit separation in Denver, Colorado. Settleable
solids removals ranging from 65 to more than 90 percent,
corresponding to chamber retention times of approximately 5
to 15 seconds, have been predicted on the basis of hydraulic
model tests. At the time of writing, no operational data
were available. Indicated costs are approximately $285/cu m/
sec ($6,500/mgd). A third generation swirl device has been
developed to take the place of conventional primary sedi-
mentation at 10 to 20 minute detention times.
258
-------
include the maximum infiltration anticipated during the life
of the sewer, while the construction allowance should be
the maximum allowable infiltration at the time of
construction. The construction infiltration will increase
continuously throughout the life of the project. APWA has
recommended the establishment of a construction infiltration
allowance of 185 I/cm diameter/km/day (200 gal./inch
diameter/mile/day) or less. This is not unreasonable in
light of improvements in pipe and joint materials and con-
struction methods.
Average and peak design flows should be related to the
actual conditions for the area under design. Too often
flow criteria are taken from a standard textbook. Adequate
subsurface investigations should be undertaken to establish
conditions that may affect pipe and joint selection or
bedding requirements. Consideration should be given to the
constructability and maintainability of the sewer system.
This calls for direct communication between the designer
and maintenance personnel.
Manholes should be designed with as few construction joints
as possible. In recent years the development of custom-
made precast manholes with pipe stubs already cast in place
has reduced the problem of shearing and damage of connect-
ing pipes. The use of flexible connectors at all joints
adjacent to manholes reduces the possibility of differen-
tial settlement shearing the connecting pipes.
Manhole cover design is attracting serious attention in
view of evidence that even small perforations can produce
sizable contributions of extraneous inflow. A single
2.5-cm (1-inch) hole in a manhole top covered with 15.2 cm
(6 inches) of water may admit 0.5 I/sec (11,520 gpd) [41].
Solid sealed covers should be used for manholes in areas
subject to flooding. If solid covers are used, alternative
venting methods must be used to admit air or remove sewer
gases.
Construction considerations — The most critical factor rela-
tive to infiltration prevention is the act of construction.
The capability of currently manufactured pipes and joints
to be assembled allowing minimal infiltration must be
coupled with good workmanship and adequate inspection,
expecially at house connections.
Trenches should be made as narrow as possible but wide
enough to permit proper laying of pipe, inspection of
joints, and consolidation of backfill. Construction should
be accomplished in dry conditions. If water is encountered
158
-------
Operationally and from a design standpoint, the aforemen-
tioned factors are taken into account by considering (1) the
food-to-microorganism ratio, (2) the sludge retention time,
and (3) the hydraulic detention time.
The food-to-microorganism ratio is defined as the kilograms of
BOD5 (food) applied per unit time (often taken as the amount
consumed) per kilogram of organisms in the system. The sludge
age is defined by the kilograms of organisms wasted per day.
The hydraulic detention time is defined as the value, given in
units of time, obtained by dividing the volume of the reaction
vessel by the flow rate.
Because the food-to-microorganism ratio and the sludge re-
tention times are interrelated [17], both are commonly used
in the design of biological systems. From field observa-
tions and laboratory studies, it has been found that as the
sludge age is increased and, correspondingly, the food-to-
microorganism ratio decreased, the settling characteristics
of the organisms in the system are enhanced, and they can be
removed easily by gravity settling. Typical values for the
food-to-microorganism ratio and sludge age are given in
reference [17].
As previously noted, the length of time the biomass is in
contact with the waste BODs is measured by the hydraulic de-
tention time. The minimum time to achieve a given removal
is dependent upon the food-to-microorganism ratio. Low
ratios (i.e., a high number of bacteria per kilogram of BOD,.)
allow faster utilization of a given amount of BODs. The 5
minimum time required may vary considerably, from 10 to 15
minutes in contact stabilization, or less for trickling
filters and rotating biological contactors, and up to 2 to
3 days for oxidation ponds. At the shorter contact times,
the biomass only removes the dissolved matter and possibly
some of the smaller colloidal matter [15] . At longer con-
tact times, suspended organic matter is utilized.
In any biological system, these factors control the process.
A mathematical model has been developed for the activated
sludge system [17, 14]. Models for trickling filters,
rotating biological contactors, and treatment lagoons have
not been formulated. Empirical designs and design param-
eters are used instead.
APPLICATION TO COMBINED SEWER OVERFLOW TREATMENT
Biological treatment of wastewater, used primarily for domes-
tic and industrial flows of organic nature, produces an
effluent of high quality and is generally the least costly
260
-------
made of natural rubber, synthetic rubber, or various other
elastomers. These joints are used on asbestos cement pipe,
cast iron pipe, concrete pipe, vitrified clay pipe, and
certain types of plastic pipes. Compression gasket joints
are most effective against infiltration while still pro-
viding for deflection of the pipe.
Chemical weld joints — Chemical weld joints are used to join
certain types of plastic pipes and glass fiber pipes. The
joints provide a watertight seal. It has been reported
that, on the basis of field tests, jointing under wet or
difficult-to-see conditions does not lend itself to precise
and careful workmanship. Thus special care is necessary
in preparing these joints in the field. More experience
with these pipes in sewer applications is necessary to
determine the longevity of this type of joint.
Heat shrinkable tubing — A new type of joint developed
recently is the heat shrinkable tubing (HST) [27]. The HST
material begins as an ordinary plastic or rubber compound
which is then extruded into sections of tubing. The tubing
is then heated and stretched in diameter but not in length.
After cooling it retains the expanded diameter. If a length
of 8-inch diameter tubing is expanded to 16 inches, it
will conform to any shape between 8 and 16 inches when
reheated. This characteristic gives the HST the ability to
form a tight fit around sewer pipe joints.
The material recommended for HST joints is a polyolefin
which has a high degree of chemical resistance and the
ability to resist scorching and burning, and is both eco-
nomical and easy to apply. To further assure HST joint
strength and resistance to internal pressure, a hot melt
adhesive is recommended as an inner surface sealant. The
adhesive material has a melting temperature close to that
of the HST and will bind the tubing and pipe materials to-
gether as the tubing cools to its final shape. Both pro-
pane torches and catalytic heaters can be used as the heat
source.
Physical properties of the HST reportedly were better than
those of currently used joint materials:
The coupling of commercial sewer pipe, both butt-
end and bell and spigot, with watertight joints
using heat shrinkable plastic tubing is feasible
and economically practical. Used in conjunction
with a hot melt adhesive it can surpass in phys-
ical and chemical strength any of the conven-
tional joints presently being used with clay,
concrete, and asbestos-cement nonpressure sewer
pipe. [27]
156
-------
3. Shakedown runs are necessary to keep the units and
the usual large number of automatic controls in
operating order.
CONTACT STABILIZATION
Description of the Process
Contact stabilization is considered in lieu of other acti-
vated sludge process modifications for treating combined
sewer overflows, because it requires less tank volume to
provide essentially the same effluent quality. The over-
flow is mixed with returned activated sludge in an aerated
contact basin for approximately 20 minutes at the design
flow. Following the contact period, the activated sludge
is settled in a clarifier. The concentrated sludge then
flows to a stabilization basin where it is aerated for
several hours. During this period, the organics from the
overflow are utilized in growth and respiration and, as
a result, become "stabilized." The stabilized sludge is
then returned to the contact basin to be mixed with the in-
coming overflow. A schematic of a contact stabilization
plant for treating combined sewer overflows is shown on
Figure 50.
Demonstration Project, Kenosha, Wisconsin
A project sponsored by the EPA to evaluate the use of con-
tact stabilization for treatment of combined sewer overflows
from a 486-ha (1,200-acre) tributary area is presently under-
way at Kenosha, Wisconsin [23, 19]. It is an example of how
contact stabilization can be used to treat combined sewer
overflows using the waste activated sludge from a dry-weather
activated sludge plant. At the Kenosha municipal sewage
treatment plant, a 101-1/sec (23-mgd) facility, a new com-
bined sewer overflow treatment facility was constructed.
This facility consists of an aeration tank, a contact sta-
bilization tank, and a new clarifier. The design capacity
of the new facility is 88 I/sec (20 mgd). The stabilization
tank, acting as the biosolids reservoir, receives the waste
activated sludge from the main plant. This sludge is held
for up to 7 days before final wasting. Thus, stabiliz-ed
activated sludge is kept in reserve ready to treat combined
sewer overflows when they occur. Photographs of the facil-
ity are shown on Figure 51.
Operation of the contact stabilization plant consists of
directing the combined sewer overflow to the contact tank
following comminution and grit removal, adding the reserve
activated sludge, and then conducting the waste flows to a
final clarifier for separation of the biosolids and other
262
-------
Choice of sewer pipe - Improvements in pipe materials assure
the designer's ability to provide proper materials to meet
any rational infiltration allowances he wishes to specify.
The upgrading of pipe manufacture to meet rigid quality
standards and specifications has eliminated the basic ques-
tion of watertightness of pipe material. The important
issues to consider in pipe material selection are struc-
tural integrity, strength of the wastewater character, and
local soil or gradient conditions. Combinations of these
factors may make one material better suited than another or
preferable under certain special installation conditions.
In such situations, pipe materials are often chosen for
reasons other than their relative resistance to infiltration
The cost of the pipe is usually a small part of the total
project cost. For rough estimating purposes, the cost of
installed sewer pipes (excluding manholes, laterals and
connections, appurtenances, etc.) ranges from $0.97 to $1.55
per cm diameter per linear meter ($1.25 to $2.00 per inch
diameter per linear foot).
Materials commonly used for sewer pipe construction include
(1) asbestos cement, (2) bituminous coated corrugated metal,
(3) brick, (4) cast iron or ductile iron, (5) concrete
(monolithic or plain), (6) plastic (including glass fiber
reinforced plastic, polyvinylchloride, ABS, and poly-
ethylene), (7) reinforced concrete, (8) steel, (9) vitrified
clay, and (10) aluminum. All of these materials, with the
possible exceptions of the plastics and aluminum, have been
used in sewer construction for many years.
Since sewer pipe made from the plastic materials is rela-
tively new, a brief description of the use of plastic pipes
is included below.
Solid wall plastic pipe usually refers to materials such as
polyvinylchloride (PVC), chlorinated polyvinylchloride
(CPVC), polyvinyldichloride (PVDC), and polyethylene. These
materials are lightweight, have high tensile strength, have
excellent chemical resistance, and can be joined by solvent
welding, fusion welding, or threading. The PVC is probably
the most commonly used plastic pipe because it is stronger
and more rigid than most of the other thermoplastics; how-
ever, PVC is available only in diameters up to 30.5 cm
(12 inches).
Polyethylene pipe is finding major use as a liner for dete-
riorated existing sewer lines [26] . Several lengths of
polyethylene pipe can be joined by fusion welding into a
long, flexible tube. This tube is then pulled into the
existing sewer. When the existing house laterals have been
connected to this new pipe liner, the result is a watertight
154
-------
(c)
(e)
Figure 51. Combined sewer overflow treatment
by contact stabilization (Kenosha)
(a) Contact tank with diffused air (b) Sludge stabilization tanks with floating
aerators (c) Floati.ng aerator anchoring and counterweight details (d) Closeup
of aerator operation (e) Final contact tank (peripherally fed) and effluent
264
-------
INFILTRATION/INFLOW CONTROL
A serious problem results from (1) excessive infiltration
into sewers from groundwater sources and (2) high inflow
rates into sewer systems through direct connections from
sources other than those which the sewers are intended to
serve. Inflow does not include, and is distinguished from,
infiltration. The sources and control of infiltration and
inflow are discussed in this subsection.
Sources
Infiltration is the volume of groundwater entering sewers
and building sewer connections from the soil through defec-
tive joints, broken, cracked, or eroded pipe, improper
connections, manhole walls, etc. Inflow is the volume of
any kind of water discharged into sewer lines from such
sources as roof leaders, cellar and yard drains, foundation
drains, commercial and industrial so-called "clean water"
discharges, drains from springs and swampy areas, depressed
manhole covers, cross connections, etc.
Inflow sources generally represent a deliberate connection
of a drain line to a sewerage system. These connections may
be authorized and permitted; or they may be illicit connec-
tions made for the convenience of property owners and for
the solution of on-property problems, without consideration
of their effects on public sewer systems.
The intrusion of these waters takes up flow capacity in the
sewers. Especially in the relatively small sanitary sewers,
these waters may cause flooding of street and road areas and
backflooding into properties. This flooding constitutes a
health hazard. Thus these sanitary sewers actually function
as combined sewers, and the resulting flooding becomes a
form of combined sewer overflow.
The two types of extraneous water, inflow and infiltration,
which intrude into sewers do not differ significantly in
quality, except for the pollutants unavoidably or deliber-
ately introduced into waters by commercial-industrial
operations [13]. Foundation inflow, for example, does not
vary greatly from the kind of water that infiltrates sewer
lines from groundwater sources. Basement drainage may
carry wastes and debris originating in homes, including
laundry wastewater.
Inflow Control
Correction of inflow conditions is dependent on regulatory
action on the part of city officials, rather than on public
152
-------
ratio, and detention times in the contact and stabilization
tanks [17, 24, 15, 14]. In the work at Kenosha, however, it
has not been possible to show any correlation between re-
movals and these items, although it has been shown that
operation based on an assumed uniform influent BODs is suf-
ficient for good BODs and SS removals (80 and 90 percent,
respectively).
Operating Parameters — With contact stabilization or any
other activated sludge process, operation is normally based
on the food-to-microorganism ratio or sludge retention time.
Because of this, difficulties may be encountered when using
an activated sludge process for treating a rapidly varying
and intermittent flow. The sludge retention time is particu-
larly difficult to control because overflows may not last
long enough for the plant to stabilize and for proper
wasting procedures to be instituted. Operating the plant on
stored overflows could reduce this problem The food-to-
microorganism ratio, which is interrelated to the sludge age,
can be used to control the operation of the plant; however,
it too is difficult to control since the concentration of
both the incoming BODs and tne biological solids in the sys-
tem must be known. This is further complicated because the
BODs concentration in the combined sewer overflow may vary
significantly. Based on the results at Kenosha, it has been
found that exact control is not necessary for good operation.
The operating parameters used for the contact stabilization
plant at Kenosha are shown in Table 52. The values reported
are averages, and the range was generally within ±60 percent
of the value listed. For comparison, the design parameters
for sewage treatment by contact stabilization found in the
literature are also presented.
For units such as that at Kenosha, sophisticated design may
not be warranted because the system is operated for such
short periods that the biosolids and the kinetics of the
system do not have a chance to adjust to the incoming flow
before the storm is over. In this case, using the reported
design equations should be sufficient. Abatement plans that
include a contact stabilization process for the treatment of
stored overflows for periods of time greater than 5 to
10 days may warrant more sophisticated design to achieve
higher removal efficiencies. The use of the kinetic equa-
tions describing the metabolism of the bacteria, as formu-
lated by McCarty [14], Metcalf $ Eddy [17], and others, may
prove useful under such circumstances.
Results of Operational Tests — The work at Kenosha has not
been able to show any adverse condition that affects
removals. Based on the results of 23 storms studied, the
266
-------
overall because of external power requirements; (4) no land
acquisition is necessary; (5) receiving water pollution
loads can be reduced by 50 percent (according to independent
studies [49, 30]); and (6) little increase in manpower is
required.
Disadvantages of gravity systems may be divided into three
categories: nonquantifiable, separation effectiveness, and
costs. Nonquantifiable disadvantages, which based on past
experience are the most important, are that (1) considerable
work is involved in in-house plumbing separation; (2) there
are business losses during construction; (3) traffic is
disrupted; (4) political and jurisdictional disputes must
be resolved; (5) extensive policing is necessary to ensure
complete and total separation; and (6) considerable time is
required for completion (e.g., in 1957 separation in
Washington, B.C., was estimated to take until sometime
after the year 2000 to complete) [24]. Separation effec-
tiveness disadvantages are as follows: (1) there is only a
partial reduction of the pollutional effects of combined
sewer overflows [30] ; (2) urban area stormwater runoff con-
tains significant contaminants [7, 4]; and (3) it is diffi-
cult to protect storm sewers from sanitary connections
(either authorized or unauthorized). Estimated costs for
gravity sewer separation are shown for various cities in
Table 26.
The cost disadvantages of separation, when compared to some
conceptive alternative solutions, are indicated in Table 27.
Again, the major reason for the higher costs of sewer sepa-
ration are in-house plumbing changes which can be as high as
82 percent of the total sewer separation costs [12],
Conclusions
On the basis of currently available information, it appears
that sewer separation of existing combined sewer systems is
not a practical and economical solution for combined sewer
overflow pollution abatement. Several cited alternatives
listed in Table 27 suggest other solutions, most of which
are considerably less expensive and should give better re-
sults with respect to receiving water pollution abatement.
In addition, storm sewer discharges may not be allowed at
all in the future, thus forcing collection and treatment of
all sewage and stormwater prior to discharge. In this case,
the argument for either separate or combined sewers is moot.
The choice between sewer separation and other alternatives
will be controlled by the uniqueness of each situation.
The examples cited in Table 27 leave no doubt that any alter-
native to sewer separation is the better choice. However,
150
-------
Advantages and Disadvantages
Some advantages of the contact stabilization process for the
treatment of excess (combined sewer) flows in this applica-
tion are: (1) high degree of treatment; (2) central location
of maintenance personnel and equipment; and (3) reduction of
the loadings on dry-weather facilities, by dual use of
facilities, during normal operations and emergency shutdown
of the main plant, making the whole very versatile. Contact
stabilization shows definite promise as a method for treat-
ing combined sewer overflows when used in combination with
a dry-weather activated sludge treatment plant. Disadvan-
tages are: (1) high initial cost, (2) the facilities must
be located next to a dry-weather activated sludge plant,
(3) adequate interceptor capacity must exist to convey the
storm flow to the treatment plant, and (4) expansion of
major interceptors may be required.
TRICKLING FILTERS
Description of Process
Trickling filters are widely employed for the biological
treatment of municipal sewage. The filter is usually a
shallow, circular tank of large diameter filled with crushed
stone, drain rock, or other similar media. Settled sewage
is applied intermittently or continuously over the top sur-
face of the filter by means of a rotating distributor and is
collected and discharged at the bottom. Aerobic conditions
are maintained by a flow of air through the filter bed in-
duced by the difference in specific weights of the atmos-
phere inside and outside the bed.
The term "filter" is a misnomer, because the removal of
organic material is not accomplished with a filtering or
straining operation. Removal is the result of an adsorption
process occurring at the surface of biological slimes cover-
ing the filter media.
Classification - Trickling filters are classified by hy-
draulic or organic loading. Until recently, there were only
two flow classifications: low rate and high rate. A third
classification, ultrahigh rate, has been added since the ad-
vent of plastic medium filters. Although the distinctions
are based on hydraulic loading, they are centered in reality
around the organic loading that the filter can handle. A
comparison of the three classifications of trickling filters
is presented in Table 53.
The type of medium used varies considerably. Rock, slag,
hard coal, redwood slats, and corrugated plastic have been
used. Rock, slag, and hard coal have relatively low surface
268
-------
(50 million persons) was served in whole or in part by com-
bined sewer systems [42]. Furthermore, it was reported that
there were 14,212 overflows in the total 641 jurisdictions
surveyed; of these, 9,860 combined sewer overflows were
reported from 493 jurisdictions. Until 1967, the most com-
mon remedial method reported was sewer separation, and of
274 jurisdictions with plans for corrective facilities con-
struction, 222 indicated that some degree of sewer separa-
tion would be undertaken.
Detailed Analysis
Sewer separation will continue to be used to some degree in
the future and thus an investigation of the methods, their
advantages and disadvantages, and their costs is warranted.
There are three categories of sewer separation systems:
pressure, vacuum, and gravity.
The most comprehensive study of the pressure or "sewer with-
in a sewer" concept was published by the ASCE [12] in 1969.
The greatest disadvantage of pressure systems is generally
higher costs, as shown in a comparison of pressure and
gravity system costs in the cities of Boston, Milwaukee,
and San Francisco presented in Table 26. The ratios of pres-
sure to gravity costs are 1.4, 1.5, and 1.5, respectively.
The in-sewer pressure lines varied from 6.3 to 40.6 cm (2-1/2
to 16 inches) in diameter and pressure control valves limited
the line pressure to 2.11 kg/sq cm (30 psi) . A major portion
of the costs is the "in-house separation" which can be
as high as 82 percent of the total cost for separation
using a pressure system [12]. Besides the high costs, other
disadvantages of pressure systems are that (1) they are dif-
ficult to maintain; (2) they require complex controls; and
(3) they are dependent on electricity for operation. It is
important to realize that approximately 72 percent of all
combined sewers are less than 0.61 meters (2.0 feet) in
diameter, making it difficult to install the pressure pipe.
The advantages are that (1) as an alternative, they provide
an additional degree of latitude in sewer design, (2) there
is minimal construction interference to commerce and traffic,
and (3) they are handy in low areas.
Sewer separation of existing combined sewers has histori-
cally been accomplished by utilizing gravity systems. The
advantages of gravity sewer separation are that (1) all
sanitary sewage is treated prior to discharge; (2) treatment
plants operate more efficiently under the relatively stable
sanitary flows; (3) other alternatives are less reliable
148
-------
area per unit volume and are quite heavy, thus limiting the
depth of filter. Redwood slats and corrugated plastic are
much lighter and can be constructed with a larger surface
area per unit volume.
Operation - The operation of most high-rate and ultrahigh-
rate trickling filters is in series with a second or third
filter and/or with recirculation. The purpose is to provide
high removals by increasing the contact time of the waste
with the biomass attached to the filter material. When
operating alone without recirculation, trickling filters
used for treating domestic wastes remove between 50 and
75 percent of the BODs.
Under storm conditions, the trickling filter must handle
highly varying flows. Applying a varying organic load to
a filter does not produce optimum removals. It is gener-
ally thought that only sufficient biomass adheres to the
supporting medium to handle the normal organic load. As the
loading increases above this level, the maximum BODs utiliza-
tion rate of the biomass is reached. This is not a sharp
distinction because some excess biomass always adheres to
the medium and can accept some of the organic load.
A varying hydraulic load also affects removals. The in-
creased shearing action of high flows causes excess slough-
ing or washing off of the biomass. To help dampen this
effect, filters operating in series under dry-weather condi-
tions can be operated in parallel, thereby reducing some of
the increased hydraulic load on each filter. A maximum
overall flow variation (maximum/minimum) of 8 to 10 is
acceptable while still achieving significant removals [20].
Design - Trickling filter design has been based primarily on
empirical formulas. This does not imply that the basic bio-
logical kinetics are not operative; rather, it means that
mathematical description of the process has not been
formulated. There are several design equations in the
literature that may be used for the design of trickling
filters [17, 6]. In designing a trickling filter to treat
overflows, it must be remembered that dry-weather flow is
needed to keep the biomass active between storms. Generally
two or more units should be used to provide high removals by
operating in series during dry weather and in parallel dur-
ing storm events to accommodate the flow variation needed.
Demonstration Project, New Providence, New Jersey
Trickling filters have been used extensively throughout the
United States to treat domestic flows, but only one facility
(at New Providence, New Jersey) has been designed to treat
270
-------
INTERCEPTOR
POLLUTION
kANTS FOR
TREATMENT
COMBINED SEWER OVERFLOW
MIXTURF OF HJHieiPA
AMI STIIMATCR 11 SOI All I Ml
INTI TNC MCCIVING WATERS
Figure 19. Common elements of an interceptor
and transport system [6]
146
-------
COMMINUTORED INTERCEPTOR FLOWS
SLUDGE GRINDER
PLASTIC
TRICKLING
FILTER
ROCK
TRICKLING
FILTER
EXCESS DRY-WEATHER
FLOW TO SUMMIT
FOR TREATMENT
2ND STAGE ^
WET WELL /•
^
>
WEIR
CHLORINE CONTACT
TANKS
EFFLUENT TO
PASSAIC RIVER
Figure 52.
Trickling filter plant schematic,
New Providence, N.J.
272
-------
recharge of groundwater. Erosion control measures
in construction areas will minimize the increased
solids loadings in runoff from such areas.
4. Drainage pipes and other flood control structures
will be used only where the natural system is in-
adequate, such as at high density urban activity
centers. Plans presently call for the use of
porous pavements to reduce runoff from streets.
5. Control will be exercised over the type and amount
of fertilizers, pesticides, and herbicides to mini
mize pollution of the runoff.
It has been estimated that the drainage system will cost an
average of $243/ha ($600/acre), compared with perhaps
$486/ha ($l,200/acre) for a conventional system [2].
144
-------
Table 54. TRICKLING FILTER REMOVALS [20]
NEW PROVIDENCE, NEW JERSEY
Average Recir- I~ 7~T. 77~ ~~ — ——
treated culation Trickling filters Overall'1 Trickling filters Overall3
flow, rate, Influent, Effluent, Removal, removal, Influent, Effluent, Removal, removal,
m8d mgd "»g/l mg/1 * t rag/1 mg/i $ ^
Dry weather flow
First year 0.54 0.8 104 23 78 86
Part of
second year 0.56 -- -- 9 -- 94
86 20 77 87
12 -- 93
Wet weather flow
First year 3.96 O.to 0.8
Part of
second year 1.72C
a. Includes removals by primary sedimentation.
b. Average wet weather flow; average peak flows were 6.0 mgd with no recirculation.
c. Wet weather flow rate was reduced by approximately 1.5 mgd by pumping to another treatment plant.
Note: mgd x 43.8 = I/sec
In comparing the plastic medium and the rock filter, it was
noted that up to 2-1/2 times the BODs removal per unit vol-
ume was possible with the plastic medium. Also, on a capi-
tal cost basis, the plastic medium outperformed the rock
by 2 to 1 ($/kg BOD5 removed/1,000 cu m).
Design Parameters - The average hydraulic and organic load-
ings applied to the New Providence facilities are slightly
above the recommended design values. The recommended values
are:
Plastic Medium Rock
Hydraulic 2.73 cu m/hr/sq m 0.78 cu m/hr/sq m
loading (70 mgad) (20 mgad)
Organic 1.36 kg BODc/day cu m 0.64 kg BODr/day/cu m
loading (85 Ib BOD5/day/l,000 cf or (40 Ib BODr/day/1,000 cf or
3,700 Ib BOD5/acre-£t/day) 1,742 Ib BOD5/acre-ft/day)
Additional design parameters were included previously in
Table 53.
Advantages and Disadvantages
Advantages of trickling filters include: (1) they handle
varying hydraulic and organic loads, (2) are simple to oper
ate, (3) have ability to withstand shockloads, and (4) have
274
-------
Figure 18. Stormwater surface detention pond (Chicago)
142
-------
The rotating discs are partially submerged and baffles are
used between each shaft-disc unit to prevent short-
circuiting. The waste flow enters the contact tank at one
end and is allowed to flow either perpendicular to or par-
allel to one or more units for treatment. The removal of
organic matter from the waste flow, either municipal sewage
or combined sewer overflow, is accomplished by adsorption of
the organic matter at the surface of the biological growth
covering the rotating discs. Rotational shearing forces
cause sloughing of excess biomass. Secondary clarification
should follow the rotating biological contactor treatment to
remove sloughed biomass.
As in all biological systems, because microorganisms have a
maximum metabolism rate, only a given amount of substrate
can be removed with a given amount of biomass. Although
this is true generally in the rotating biological contactor,
excess biomass can be held on the disc and can effectively
act as a reserve for use at higher loadings. The effective-
ness, however, is somewhat limited by the oxygen transfer
rate and the substrate diffusion gradient through the layer
of biomass on each disc. This is similar to what happens in
trickling filters. In general, though, the reserve biomass
reduces the importance of maintaining a uniform loading
rate. 6
Efficiency
The reported BOD5 removal efficiencies range from 60 to
95 percent [7, 2, 3, 26]. The higher values are for more
recent installations treating dry-weather flow. Suspended
solids removals are also in this range. Removals for
settleable solids, nitrogen, and phosphorus have been re-
ported to be 80 to 90, 40, and 50 percent, respectively.
When treating combined sewage flows, controlled treatment
(70 percent or better COD removal efficiency) was report-
edly maintained up to 8 to 10 times dry-weather flow [7],
A linear reduction in COD removal efficiency from 70 down to
20 percent was reported for the flow range from 10 to 30
times dry-weather flow.
Operational Considerations
Conditions noted to affect the BODs and COD removals in a
rotating biological contactor are (1) organic loading rate,
(2) contact time, (3) effluent settling, (4) the number of
units in series, and (5) high flow rates. The most impor-
tant condition is high flow rates which affects the first
three of the conditions just enumerated. The maximum allow-
able variation in flow is approximately 10 times the base
276
-------
dry-weather flow without excessive sloughing of the biomass
from the discs, and to retain good BODs and SS removals, it
may have to be much less [7].
Large increases in flow rates, at least up to tenfold, can
be tolerated, with removals of about 60 percent if the de-
tention times are 30 minutes or greater during the higher
flow [2], This is shown in Table 55. Continuous operation
under this condition eventually affects removal rates. In
a study by Marki, as reported by Antonie, it was noted that
the system took 1 day to return to steady-state removal
efficiency after being operated at high hydraulic loadings
[2]. The same conclusion was also reached at Milwaukee [5].
Contact times (hydraulic detention times) significantly
affected removals. Generally, the longer the contact time,
the higher the removals. Contact times of less than 15 to
30 minutes result in removal efficiencies of less than 60 to
70 percent. The results of several studies on this subject
are shown on Figure 55.
Another important aspect is that a continuous base flow is
required to keep the biomass alive between storms. There-
fore, either the unit is used as the dry-weather facility,
as an auxiliary dry-weather facility, or it is placed up-
stream of the main plant to treat domestic flows from a
nearby interceptor during dry weather and storm flows in
wet weather.
Table 55. ROTATING BIOLOGICAL CONTACTORS--VARIATION
IN REMOVALS RESULTING FROM FLOW INCREASES
OF 10 TIMES DUE TO WET WEATHER [2]
Condition A Condition B
Type of
operation
Steady state
For 8 hr/wk
Following day
Residence
time , min
120
12
120
BOD5
reduction, %
92.5
46.9
85.2
Residence
time , min
300
30
300
BOD5
reduction, 1
94.6
65.0
77.6
277
-------
• BOD5[28J
O COO [26]
• COD [ 2]
+ COD FIGURE 21 , [7]
A BOD5BY MARKt AS REPORTED
BY ANTONIE [ 2]
EXPECTED RANGE
OF REMOVALS
80 -
100
10
30 40 50
DETENTION TIME, MIN
Figure 55. The effect of removals with
varying contact times for rotating biological contactors
Several studies of methods for keeping the microorganisms
alive between storm events have been made. In one test,
the rotating biological contactor was operated for 8 hours
a day and remained idle for 16 hours. The discs were
rotated in the contacting tank, although there was no flow.
The biomass remained active, but 4 hours of operation under
normal conditions were required to bring removal rates back
up to steady-state values. This 4-hour lag to reach peak
efficiency would be too long for most combined sewer over-
flow conditions. A second condition was then tested. The
facility was operated at the design flow and loadings for
8 hours and then for 16 hours at 25 percent of its design
flow and organic loading. In almost all cases, removals
during the entire 8-hour period were at peak efficiencies.
The little lag time noticed during the first part of the
8-hour period was considered insignificant. Further study
is needed to determine whether the rotating biological
contactor can be operated during dry-weather flow conditions,
with less than 25 percent of the flow and organic loading,
278
-------
and still achieve this rapid startup. It is interesting to
note, however, that this situation is reminiscent of an
over-designed plant running at one-quarter load for two-
thirds of the time. In a full-scale system, this would be
the same as designing a rotating biological contactor treat-
ment plant to handle the combined sewer overflows and to
utilize it during dry-weather flows at much lower flow
rates and organic loadings.
Another finding in research studies was that several units
in series produce better removals than a single unit [26] ,
and that the percentage of removal versus the number of
units in series follows a curve similar to the contact time
curve shown on Figure 55. In other words, the greater the
number of units in series, the higher the removals [26],
The maximum number of units in series tested to date are the
12 shaft-disc assemblies in the Milwaukee prototype plant
[7].
Design Parameters
The only formal method for designing a rotating biological
contactor is that reported in manufacturers' catalogs.
General design parameters reported from various sources, in-
cluding the demonstration project at Milwaukee funded by
EPA, are reported in Table 56. Three major design param-
eters are: (1) the biomass required, which is a function of
the surface area upon which it will adhere; (2) the contact
time; and (3) the number of shaft-disc assemblies in series
needed to achieve the desired removal efficiency. The
design procedure would be first to determine the disc area
required from the curves on Figure 56, and then to determine
the minimum detention time needed to achieve the design re-
moval from the curves on Figure 55. The detention time
should be based on the expected maximum flow. One particu-
lar point in any design of a rotating biological contactor,
for obvious reasons, is the need for enclosing the unit
in areas subject to freezing ambient temperatures.
Advantages and Disadvantages
The principal advantages of the rotating biological con-
tactor are: (1) it has low power requirements, (2) a fair
degree of flow variation can be handled, (3) shockloads are
handled effectively, and (4) there are no fly and odor
problems. On the other hand, the disadvantages are that
(1) it requires a base flow to keep the biomass active,
(2) there is little control of biological process, (3) more
work is needed to define its capabilities, and (4) it must
be enclosed in freezing climates.
279
-------
Table 56. ROTATING BIOLOGICAL CONTACTOR
DESIGN PARAMETERS [7, 4, 26, 2]
Design parameter
Range
Hydraulic loading rate, gpd/sq ft
Organic loading rate, Ib BOD5/day/l,000 sq ft
Detention time, min
Number of shaft-discs in series
Peripheral speed of discs, fpm
Approximate required power, kwh/1 ,000 Ib BODC
2-8
5-15
15-20
4-10
60
4-5
Note: gpd/sq ft x 0.2828 = cu m/min/ha
Ib BOD /day/1,000 sq ft x 4.88 =
kg BODg/day/1,000 sq m
fpm x 0.0051 = m/sec
kwh/1,000 Ib BOD x 2.2 = kwh/1,000 kg BOD5
100
95
90
85
80
75
BODgCONCENTRATION (MG/I)
350
250
- 150
100
80
DOMESTIC iASTEiATER
TEMPERATURE ^ 13 DEC C
<55 DEi F)
I
3 4567
HYDRAULIC LOADING GPD/SQ FT
NOTE: GPD/SQ FT x 0.283 - cu M/MIN/HA
Figure 56. Hydraulic loading versus design
BOD5 removals for rotating biological contactors [4]
280
-------
The major use of the rotating biological contactor is ex-
pected to be in treating both dry- and wet-weather flows in
smaller communities. It fits their requirements because it
is relatively easy to operate and can handle fairly wide
variations in flow rates.
Demonstration Project, Milwaukee, Wisconsin
In Milwaukee, the application of the rotating biological con-
tactor to combined sewer overflows was studied in three
phases: (1) bench model testing, (2) pilot-plant testing,
and (3) prototype facilities [9, 7], The first two phases
simulated combined sewer overflows, and the prototype was
constructed on a 30-inch combined sewer to test the concept
under actual conditions. Average dry-weather flow from the
14-ha (35-acre) tributary area was 2.2 I/sec (0.05 mgd).
The prototype facility was designed for dry-weather flow
conditions but had a hydraulic capacity of approximately
30 times the dry-weather flow, or 66 I/sec (1.5 mgd). The
facility consisted of a pump station, a grit chamber that
also acted as a small surge chamber, and 2 bays with 12
shaft-disc assemblies in each bay. The bottom was contoured
to the discs, and a screw auger was placed down the center
of each bay to remove settled sloughed biomass. The initial
concept was to use this tank as a combination contact tank
and settling chamber, but this proved unworkable.
In the Milwaukee study, overall BOD5 removals were cited in
the 70 to 90 percent range at design average dry-weather load-
ings; however, organic loadings in excess of 3 to 4 times dry-
weather loadings reduced BODs removal efficiencies to below
50 percent. Pilot-plant studies showed that the removals for
COD were approximately 33 percent during these high loadings.
The removal efficiencies calculated from the reported raw data
obtained during the pilot-plant operation are presented in
Table 57. Controlled treatment during wet-weather flows (70
percent or better COD removal efficiency) was maintained at
hydraulic loadings up to 8 to 10 times dry-weather flow.
The variation in removal efficiencies can be attributed to
changes in (1) organic loading rate, (2) contact time,
(3) number of units in series, (4) effluent settling, and
(5) high flow rates. At Milwaukee, BOD5 removals decreased
when loadings increased above 0.032 kg BOD5/day/sq m (6.6 Ib
BOD5/day/l,000 sq ft) of disc area. Below this loading rate,
removals remained in the 90 percent range. When the loading
rate was between 0.027 and 0.062 kg BOD5/day/sq m (5.5 and
12.7 Ib BODs/day/1,000 sq ft) of disc area, BODs removals
281
-------
Table 57. REMOVALS REPORTED IN PILOT-PLANT STUDIES
OF ROTATING BIOLOGICAL CONTACTORS
MILWAUKEE, WISCONSIN
Dry-weather
hydraulic loading,
2-4 gpd/sq ft
Wet-weather
hydraulic loading,
11-12 gpd/sq ft
Parameter
BOD5
COD
Settleable
solids
Suspended
solids
Nitrogen
Phosphorus
Influent ,
mg/1
345
550
•9.2
349
40.5
8.5
Effluent,
mg/1
79
165
0.9
68
24.7
4.0
% Influent,
removal mg/1
77 395a
70 617
90 10.2
77 315b
38
53
Effluent, %
mg/1 removal
181a 54
394 33
1.7 82
96b 70b
_-
--
a. Derived from BOD:COD ratios. A single BOD^ test was run with 80 percent
removal. ^
b. Only two data points.
Source: Derived from data reported in [7].
Note: gpd/sf x 0.283 = cu m/min/ha
dropped from 90 to 70 percent. The relationship between
BODs removal and BODs loading rate is shown on Figure 57.
As a result of the study at Milwaukee, clarification is rec-
ommended following the rotating biological contactor. At
times, the prototype plant ran for several days discharging
enough sloughed biomass to make the BOD5, COD, SS, etc.,
higher in the effluent than in the influent. It was also
reported that pretreatment by sedimentation improved
removals.
TREATMENT LAGOONS
Since treatment lagoons are based on biological processes,
three basic systems are available: anaerobic, aerobic, and
facultative. Several different types of lagoons have been
developed, using either one or more of these systems. These
types are referred to as oxidation ponds, aerated lagoons,
facultative lagoons, or anaerobic lagoons.
282
-------
100
90
80
70
uj 60
50 -
1
10 20 30
WAX 1C LOMINfi, LB WMyi*Y/1. OSO SQ FT
40
NOTE: LB BOD./DAY/I.OOO so. FT x 4.88
-------
Table 58. COMPARISON OF DIFFERENT TYPES OF LAGOONS
TREATING STORM FLOWS FOR VARIOUS CITIES
Size, Volume, Detention Design flow
Location Type of lagoon acres mil gal. time, days rate, mgda
Springfield, 111,
Shelbyville, 111.
Southeast site
Southwest site
Mt.Clemens, Mich.
Equalization- 10 22.4
oxidation pond
Storage- 1 1.9
oxidation pond
Storage basin 3.9 4.0
Facultative pond 10.8 13.0
Facultative pond 2.1 2.1
Storage-aerated 1.5 5.6
lagoon
0.3
5.0
2.8
9.0
1.5
5.6
67
0.3'
1.4'
1.41
1.41
64.
East Chicago, Ind.
Oxidation pond
Aerated lagoon
Aerated
facultative
lagoon
2.8
2.3
^30
8.2
7.0
185
8.2
7.0
1.0
1.0
1.0
185
a. Designed outflow rate; inflow can be much greater.
b. Storm flow rate; the ponds also treat 0.3 mgd of trickling filter
effluent.
c. Design storm flow rate; outflow is 1.0 mgd.
Note: acre x 0.405 = ha
mil gal. x 3,785.0 = cu m
mgd x 3,785.0 = cu m/day
shown in Figure 58. In most cases, these treatment lagoons
offer multiple uses and benefits. They can be used to
supplement dry-weather treatment plants by acting as effluent
polishers and inflow equalization basins (as in Rohnert Park,
California); they offer some storage and settling capacity
and flow attenuation for wet-weather flows; and they can be
used as a central component of a community recreational area.
The average removals reported at the various study sites are
listed in Table 59. The reported pollutant removal effi-
ciencies for lagoons treating combined sewer overflows are
somewhat varied, but the overall quality of the effluent is
good. No attempt is made here to distinguish the difference
between lagoon types.
The main factor affecting removal efficiencies for all types
of lagoons is carryover of algae and other microorganisms in
the effluent. This was true to some degree at all demon-
stration sites. At Mount Clemens, Michigan, in anticipation
of this problem, a two-stage pressure sand filter and a
microstrainer were installed to polish the effluent.
284
-------
(e)
Figure 58. Combined sewer overflow treatment lagoons
(a) View of oxidation pond detailing grass embankment (She I byvi I I e)
pond following primary clarification (Springfield) (c) Facultative
riprap embankment (E. Chicago) (d) Aerated lagoon, pond no. 1 (Mt0
Compound overflow weir outlet (Springfield)
(b) Oxidation
Iagoon wi th
C I emens) (e)
285
-------
Table 59. REMOVAL EFFICIENCIES OF TREATMENT LAGOONS
FOR VARIOUS CITIES
(PERCENT)
Shelbyville,
Parameter
Springfield, -
Ill.a Southeast Southwest
Mount East
Clemens, Chicago,
Mich. Ind.c
BOD5
ss
vss
DO
Phosphorus
Nitrogen
Coliforms
27
20
Increased
Increased
22
NA
72
47
57
30
NA
40
56
86
91
-4
28
NA
69
62
96
91
92
NAd
NA
NA
NA
NA
,50
,50
NA
NA
NA
NA
NA
a.
b.
c.
d.
Based on daily samples, not necessarily coincident with storm
flows.
Two sites at Shelbyville, both of which are oxidation ponds.
Figures presented are for dry-weather flow conditions and do
not represent removals achieved during storm overflows.
Not available.
Two major operational problems were experienced. During
freezing weather, floating aerators proved to be troublesome
The spray from floating aerators freezes on the motors,
causing the units to overturn. Thus, their use was limited
to periods when freezing of the spray did not occur. The
second problem was inadequate provision for sludge storage
in some of the ponds. The use of sedimentation tanks or
basins ahead of the lagoons is necessary to reduce the SS
load to the ponds.
Design criteria for lagoons used to treat combined sewer
overflows have not yet been established. Conventionally,
lagoon design is based upon the BODs loading rate and flow
rate. However, the wide variation of the 300$ loading rate
for combined sewer overflows with respect to time simply
means design cannot be based only on this parameter.
286
-------
The advantages, disadvantages, and possible uses of the
treatment lagoon are varied. Its major advantages are:
(1) low capital costs, (2) virtually unattended operation,
(3) low operation and maintenance costs, (4) capability of
being easily modified to act also as a storage unit, and
(5) ability to act as a polishing lagoon during dry weather.
Some disadvantages include: (1) large land areas are re-
quired; (2) proper design of discharge facilities is neces-
sary to prevent discharge of algae and other microorganisms
to the receiving water; (3) the degree of treatment is diffi
cult to predict; (4) there are potential nuisance problems,
such as mosquitoes, flies, and odors; and (5) sludge
deposits reduce treatment capacity.
The following discussion of the various types of treatment
lagoons begins with oxidation ponds and continues with aer-
ated lagoons and facultative lagoons.
Oxidation Ponds
An oxidation pond generally is a shallow earthen basin de-
signed to promote a symbiotic existence between algae and
bacteria. Its depth is such that oxygen from algae and
surface regeneration can keep the total volume of the pond
aerobic. Several ponds are normally used together, most
often in series operation, which helps reduce the amount of
suspended solids in the effluent. Design of oxidation ponds
usually includes consideration for sludge storage within the
pond. The digestion of sludge deposits is sometimes anaero-
bic; however, this is not considered a limit in the defini-
tion of the oxidation pond as an aerobic system. Oxidation
ponds that are used to treat municipal wastes have very long
detention times, ranging from 20 to 120 days. The trend in
treating combined sewer overflows is to use detention times
of less than 20 days.
Removal Efficiencies — In oxidation ponds, removal efficien-
cies for suspended solids and BODs can vary tremendously
from positive to negative numbers. The reported ranges in
removals are 60 to -50 percent for suspended solids and 70
to -10 percent for BODs [17]. The reason for the high vari-
ation is that most of the influent BODs is converted into
suspended algae mass. This mass exerts a BODs demand and
creates suspended solids which are sometimes carried over
in the effluent.
Factors Affecting Removal Efficiencies - Many factors affect
the removal rates and efficiencies in oxidation ponds. Some
287
-------
of the more significant ones (listed in general order of
importance) are as follows:
1. Detention time.
2. Sufficient supply of oxygen either from algae,
surface reaeration, or some mechanical means.
3. Mixing of the pond contents by wind or mechanical
means.
4. Organic loading rate.
5. Removal of microorganisms and algae from the
effluent.
6. Temperature.
7. Sludge storage capacity which, if insufficient,
can reduce detention time and cause carryover in
the effluent.
Most of these factors are also relevant to aerated and
facultative lagoons and will be referred to in the subsec-
tions which follow.
The most important condition affecting removals is the de-
tention time. Pollutant removals are obtained both by
bacteria metabolizing the waste organics and by
sedimentation. Thus, the detention time is based on the
metabolism rate, the amount of biomass present in the ponds,
and the temperature. A 10-degree Celsius drop in tempera-
ture reduces the metabolism rate by about one-half. It
has been shown that with a detention time of 1 day up to
85 percent of the BOD5 can be removed [5]. In the kinetic
equations proposed by McCarty, a 1-day detention time may be
too close to the minimum to be used in design [14] . In a
recent study of treatment lagoons, it was reported that
detention times of 2 to 3 days were required for conversion
of all the biodegradable matter to new cells, and an addi-
tional 18 days were required for bacterial predators to
consume these new cells [29] . This would reduce the need
for effluent clarification and sludge-handling equipment.
Detention time can also affect removals by modifying the
settling characteristics of the biomass. In a basin with
plug flow or complete mix with no recycle, as is the case
in treatment lagoons, the sludge age is equal to hydraulic
detention time. Longer sludge ages result in better
settling [15]. Poor settling means low BODs removals due
to carryover of cell tissue in the effluent.
288
-------
The next two most important conditions affecting removals
are (1) enough algae to supply the needed oxygen for bac-
terial assimilation of the waste organics and (2) proper
mixing. The amount of algae in a pond is dependent upon
the quantity of sunlight and usable carbon for algae
synthesis. The quantity of sunlight varies with the lati-
tude of the site and the amount of normal cloud cover
[16]. In general, the more northern latitudes have less
sunlight. During the winter in northern areas, the amount
of sunlight for algae growth is affected also by ice cover.
Carbon sources for algae include carbon dioxide, carbonates,
and bicarbonates found in the influent water and carbon
dioxide given off during respiration by the bacteria and
algae. These two factors must therefore be considered when
designing oxidation ponds.
Sufficient mixing is necessary to (1) ensure dispersion of
the waste and oxygen throughout the pond volume, (2) ensure
good bacterial action, (3) eliminate areas deficient in DO,
and (4) prevent short-circuiting. Mixing can be accom-
plished by allowing the wind to mix the pond, using surface
aerators, or recirculating.
Good mixing and high algae production, however, have limits.
An oxidation pond can be overloaded with biodegradables.
An overloaded pond results in septic conditions with slower
anaerobic stabilization of the waste. Also, operational
problems occur, such as odors (hydrogen sulfide,
etc.). It is important, therefore, to keep the loadings
within the mixing and oxygen production capabilities of the
ponds.
Operational Considerations — Algae removal has been one of
the most common problems in oxidation ponds [27, 15].
Several methods have been tried to reduce algae concentra-
tions in the effluent. Those mentioned in the literature
are long detention times, ponds in series, rock filters,
sand filters, and chlorination, to name a few. A more de-
tailed explanation of these methods is given in Table 60.
Some of these methods can be used together to produce the
desired results. A typical oxidation pond outlet with three
concentric baffles to reduce the presence of algae is shown
on Figure 59. Microstraining of pond effluent has not yet
proven successful when using screen openings of 21 microns
in size or larger [27, 10].
Design Parameters - Many design procedures for oxidation
ponds are cited in the literature [17, 29, 13, 16], The
recommended design parameters for treatment of waste flows
are presented in Table 61.
289
-------
Table 60. METHODS OF REDUCING ALGAE IN THE
OXIDATION POND EFFLUENT
Method
Description
Long detention times
Series of ponds
Settling
Rock filtration
Chlorination
Chemical precipitation
Filtration
Macroorganisms
Use of long detention times of 120 to
160 days allows endogenous respiration
to reduce algae concentration [29].
A series of ponds with outlets from
each pond designed to reduce the carry-
over of algal mass (see Figure 59 for
a typical outlet) [29].
Prevention of wind action or other
mixing action to allow settling of
algae, bacteria, and other settleable
material to improve the effluent.
Settling should be done in the second
or third pond in a series [29].
Use of 1- to 2-inch rock around the
outlet pipe to act in a manner similar
to tube settlers [29].
Use of chlorine dosages high enough to
kill the algae; provide a small pond
for settling of the algae before dis-
charge [29] .
Use of chemical coagulants to remove
algae [29].
Use of dual-media high rate filtration
to remove algae [27, 8, 21].
Use of crustaceans, such as Daphnie ,
in a final pond to feed upon the algae
and other suspended organic matter
[11].
Note: inch x 2.54 = cm
Aerated Lagoons
There are two types of aerated lagoons: (1) the complete
mix system, simulating a modification of activated sludge
treatment; and (2) the aerated oxidation pond in which com-
plete mixing is not achieved but enough oxygen is supplied
for bacterial activity. Aerated lagoons are usually earthen
basins, lined or unlined. Unlike oxidation ponds, oxygen
is supplied by mechanical or diffused aeration equipment.
Aeration Equipment — At least six different types of aera-
tion equipment can be used in an aerated lagoon, as shown on
Figure 60. Floating mechanical aerators are used most
widely. The floating rotor aeration unit, sometimes called
a Kessener Brush or cage rotor, is a relatively recent
development. Weighted plastic tubing with openings along
290
-------
MAX WATER SURFACE
CONCENTRIC BAFFLES
i
UN fATIR SURFACE
•ATER
4
6 IN
!' 1 FT
OUTLET PIPEii* fr_
POND BOTTOM
r
Figure 59. Typical oxidation pond outlet
with three concentric baffles to reduce
the presence of algae in the effluent
Table 61. OXIDATION POND DESIGN
PARAMETERS [17, 29, 5, 16]
Organic loading rate
Ib BOD5/acre/day
Detention time, days
Overall
Per pond
Optimum
Depth, ft
Optimum
Number of ponds
Pond configuration:
Shape
Inlet
Outlet
21-50
30-160
10-40
20
2-5
4
2-6
Adaptable to terrain
Center generally pre-
ferred (anything that
reduces short-
circuiting)
Designed to vary pond
level from 2 to 5 ft
in-6 in. increments
and to reduce the
amount of algae trans-
fer
Note: Ib BOD5/acre/day x 1.12
feet x 0.505 = m
inch x 2.54 = cm
kg BOD5/ha/day
291
-------
PROPELLER
(a) FLOATING (MECHANICAL) AERATOR
(b) AERATION ROTOR
(c) PLASTIC TUBING AERATOR
o opp
LAGOON BOTTOM LAGOON BOTTOM
(d) AIR GUN AERATOR
o 2 °S.
AIR
SUPPLY
TUBING
CONCRETE
BASE
V r
AERATOR -~-^-
WATER
FLOW--^ L
LAGOON
. C
J
°j'
00°
^
)n n
fe-r-
ouo
C) .
°.f
o o
o»o
NKA AERATOR
Figure 60. Types of aerators
292
-------
its length has been used to supply extra oxygen to over-
loaded oxidation ponds. The tubing is used extensively in
the northern climates where freezing of the ponds is a com-
mon occurrence. The remaining three types of aerators, also
diffused aeration systems, are the helical diffuser, the
air gun, and the INKA system. The air gun, used for deeper
ponds, operates in a pulsating manner, discharging air up
through a vertical tubing and drawing in behind it a column
of water. It not only aerates the water but also has good
pumping action. The helical diffuser is made of extruded
plastic formed into a cylinder with a spiral interior
baffle designed to lengthen the flow pattern of air bubbles
and water. In the diffuser, air is injected at the bottom
of the vertical extruded plastic tube, which sits on the
bottom of the pond. The air bubbles rise in a spiral manner
up the plastic pipe drawing in water behind them. The INKA
system consists of an 8-foot by 5-foot diffuser plate placed
in between vertical baffles to create an airlift pump effect,
thus circulating the water past the diffusers. The advan-
tages and disadvantages of each type are listed in Table 62.
Table 62. TYPES OF AERATION EQUIPMENT FOR
AERATED LAGOONS
Types
Advantages
Disadvantages
Depths at which
commonly used,
ft
Floating mechanical
aerator
Floating rotor
aeration unit
Plastic tubing
diffuser
Air guns
INKA system
Helical diffuser
Good mixing and aeration
capabilities; not affected
by sludge deposits; easily
removed for maintenance.
Probably unaffected by
ice; not affected by sludge
deposits.
Not affected by floating
debris or ice; no ragging
problems.
Not affected by ice; good
mixing; good for deep
lagoons.
Not affected by ice or
sludge deposits; good
mixing.
Not affected by ice;
relatively good mixing.
Ice problems cause turning 10-15
over during freezing
weather; ragging problems
without a weedless im-
peller.
Possible ragging problem. ^3-10
Calcium carbonate buildup 3-10!
blocks air diffusion
holes; is affected by
sludge deposits.
Calcium carbonate buildup 12-20
blocks air holes; poten-
tial ragging problems; is
affected by sludge
deposits.
Potential ragging 8-15
problems.
Potential ragging 8-15
problems; is affected by
sludge deposits.
a. Has also been used in lakes at much deeper depths.
Note: ft x 0.305 = m
293
-------
Removal Efficiencies — Regardless of the type of aerator,
most aerated lagoons are constructed in multiple ponds. The
ponds may be used in either series or parallel operation.
Final clarification is required to achieve good BOD5 and SS
removals. Removals range from 75 to 95 percent for both
BODs and SS when sufficient settling is provided [29, 17,
13]. With detention times of less than 3 to 5 days and no
settling, the removals are much lower.
As previously discussed, DO concentration, adequate mixing,
control of biological solids carryover, short-circuiting,
and temperature all have an effect on removal efficiency.
In most cases, detention times longer than 2 days provide
sufficient treatment for the dissolved portion of the waste.
For good settling, detention times should be more than 3 to
4 days. In most designs, neither DO nor mixing are problems
Problems generally occur in providing adequate mixing in a
lagoon designed as a complete mix system. When mixing is
incomplete, settling occurs. This removes some of the bio-
mass from intimate contact with the waste material and
leads to a reduction in efficiency.
Design Considerations — The method of design for an aerated
lagoon depends on the type to be constructed. Aerated
oxidation ponds are normally designed by using empirical
design parameters. The kinetic design approach is not
used because settling is difficult to account for and the
flow regime is difficult to analyze. Typical design param-
eters found in the literature are listed in Table 63.
Factors that must be considered include (1) BOD5 removal,
(2) effluent characteristics, (3) oxygen requirements,
(4) temperature effects, and (5) energy requirements for
mixing. For actual design methods, the reader is directed
to the many presented in the literature [17, 29, 25, 5].
Complete mix aerated lagoons often are designed using em-
pirical parameters. It is also possible, however, to use
kinetic equations to design a proposed pond. A complete
description of the kinetic equation method for designing an
aerated lagoon can be found in reference 17.
All aerated lagoons should include some process for removing
biosolids from the effluent. The most common process is
final settling, either in a concrete tank or in a small non-
aerated final pond. In aerated oxidation ponds, settling
normally occurs within the ponds, and a final pond or
settling tank is not required. Other forms of removing
biosolids, such as sand filters, can be used. The design
of ponds used for solids removal should include volume for
storage of the accumulated sludges. No annual sludge volume
accumulation in ponds for treating combined sewer overflows
has been reported.
294
-------
Table 63. AERATED LAGOON DESIGN PARAMETERS
Description
Aerated Complete
From the oxidation mix aerated
literature pond lagoon
Organic loading,
Ib BOD5/acre day
Detention time, days
Optimum, days
Number of ponds
Depth, ft
100
2
2
8
-1,000
-10
--
-6
-15
100
5
2
6
-500
-11
--
-6
-10
500
1
4
1
10
-1,000
-8
-6
-4
-15
Mixing requirements for a
complete mix regime
hp/1,000 cf
Minimum pond velocity, fps
Type of aeration and amount
of oxygen transfera
Floating aerators, Ib 02/hp/hr 1.8-4.5
Floating rotor aeration unit,
Ib 02/hp/hr
^3-4
0.2-0.7
1-2
0.8-1.6
3-18
1.2-4.2
@ air supplied, scfm/ aerator 8-30
Plastic tubing aerator,
Ib 02/hr/100 ft
@ air supplied, scfm/100 ft
Air gun, Ib 02/hr/gun
@ air supplied, scfm/gun
Helical aerator,
Ib 02/hr/aerator
0.2-0.5
0.5
INKA aerator
Unknown
a. All of the oxygen transfer figures are general and should be
verified for design.
Note :
Ib BOD5/acre/day x 1.1208 = kg BODs/ha/day
ft x 0.3048 = m
fps x 0.3048 = m/sec
Ib 02/hp/hr x 0.6083 = kg 02/kwh
Ib 02/hr/100 ft x 1.4882 = kg 02/hr/100 m
Ib 02/hr/unit (gun, aerator) x 0.4536 = kg 0?/hr/unit
scfm/100 ft x 0.9284 = cu m/min/1,000 m
scfm/aerator x 0.0283 = cu m/min/aerator
295
-------
Facultative Lagoons
Facultative lagoons contain three zones or layers of biologi
cal activity: (1) the upper or aerobic layer, (2) the mid-
dle or facultative layer, and (3) the lower or anaerobic
layer. Facultative lagoons are deeper than those previously
discussed, and intermixing of the layers is minimized. Such
systems utilize the good features of both the anaerobic and
the aerobic lagoons. Settled materials, along with some of
the dissolved and suspended material, are allowed to stabi-
lize anaerobically into gaseous end-products and humus. In
the upper layer, the dissolved and suspended matter is
oxidized. The presence of this zone also helps to eliminate
the undesirable gaseous end-products given off in the an-
aerobic zone. The facultative zone acts as a transition
zone with both aerobic and anaerobic metabolism of the waste
products occurring. The chemical reactions within each zone
are shown on Figure 61.
\
-------
Originally, facultative ponds used algae to keep the upper
layer aerobic. Recently, aeration equipment has been added
to some ponds to ensure sufficient DO to keep them from
becoming completely anaerobic. However, the aerator size
must be limited to prevent the pond from becoming a com-
pletely mixed system.
Removal Efficiencies - BOD5 removals up to 90 to 95 percent
have been reported for facultative ponds treating municipal
sewage. The most common figures, however, are between 70
and 85 percent [29, 5, 12, 16]. Higher bacterial removals
have been reported for facultative lagoons than for aerated
lagoons or oxidation ponds [16, 5]. For facultative ponds
treating combined sewer overflows, BODs removal efficiencies
ranging from 50 to 90 percent and SS removals of approxi-
mately 50 percent have been reported.
Operational Considerations - In most cases, the factors
affecting oxidation ponds and aerated lagoons also control
the performance of facultative ponds. Detention time,
proper supply of oxygen either by algae or aeration equip-
ment, sludge age for good settling characteristics, algae
removal in the effluent, temperature, and short-circuiting
all affect removals [16, 17, 12, 18]. The notable exception
is the need for mixing. In facultative ponds, some mixing
of the upper and lower layers is important to promote good
distribution of pollutants throughout the individual layers.
Complete intermixing should be avoided to prevent odors from
escaping through the aerated zones without being oxidized.
This is usually handled by making the ponds deep enough
to create a thermocline, which limits mixing between layers,
and by using multiple small ponds to limit wind mixing.
Care, with respect to mixing, must be exercised when using
mechanical aeration equipment. Isolating the anaerobic
zone has also been accomplished by pond configuration,
as shown on Figure 62 [16, 18].
The principal conditions affecting operation of facultative
lagoons are the type of aeration equipment (if used) and
adequate sludge storage. When facultative lagoons use aera-
tion equipment to supply additional oxygen, care is needed
in selecting the proper type, with mixing restrictions and
cold-weather problems in mind. The lack of sludge storage
capacity may necessitate excessive maintenance in removing
accumulating sludges and also can prevent proper operation
of the lagoon.
Design Parameters - From a survey of the literature, a lack
of specified design parameters is evident. Most facultative
lagoons are designed for 56 kg BODs/ha/day (50 Ib BODs/acre
/day) or less, depending on the geographical latitude of the
297
-------
V
3 TO 5 FT
INLET PIPE
ANAEROBIC
ZONE
NOTE: FT x 0.305-
INLET PIPE
Figure 62. Two pond configurations to promote
anaerobic zones and decomposition of settled material [18]
site location. Generally, loadings are reduced in the more
northerly latitudes. In areas of possible ice cover during
the winter, the normal loading is 17 kg BOD5/ha/day (15 Ib
BODs/acre/day). Predicting the final performance of facul-
tative lagoons is difficult, regardless of the design proce-
dure used [12]. Design parameters reported in the litera-
ture are listed in Table 64.
Demonstration Projects
Springfield, Illinois [22] — Combined sewer overflows from
a pumping station in Springfield to a natural drainage
channel reportedly were responsible for repeating instances
of fish kills in Sugar Creek during the 1960s. About 95 per
cent of the flow in the channel during storm periods origi-
nates at the pumping station. The pump station serves a
tributary area of 894 ha (2,210 acres) and has a maximum
capacity of 856 I/sec (19.5 mgd) or approximately 5 times
the average daily flow. The combined sewers entering the
298
-------
Table 64. FACULTATIVE LAGOON DESIGN
PARAMETERS [17, 16, 29]
Organic loading in summer,
Ib BOD5/acre/day 15-80
Detention time, days 7-120a
Depth
Total, ft 6-12
Anaerobic zone, ft 3-4
Number of ponds 2-10
Pond configuration:
Shape Not important
Inlet Center inlet superior,
near bottom
Outlet Designed to reduce amount
of algae transfer when
not using mechanical
aeration
a. 120 days for complete treatment with good coliform
removals [28].
Note: Ib BOD5/acre/day x 1.12 = kg BOD5/ha/day
ft x 0.305 = m
pump station have a combined capacity of 23,000 I/sec
(494 mgd).
To abate this combined sewer overflow pollution problem, a
storage/oxidation pond was constructed in 1967. The main
purpose of the pond was to provide flow attenuation of the
combined sewer overflows. The 4.1-ha (10-acre) pond was
designed to provide 8 hours of detention at a flow rate of
2,940 I/sec (67 mgd). It was estimated that a 5-hour deten-
tion time would account for approximately 70 percent removal
of SS and 30 percent removal of BODs from the combined sewer
overflows by sedimentation alone. Further improvements
would occur as the pond functioned as an oxidation pond be-
tween storm events.
Performance of the pond during a 20-month period of observa-
tion indicated that it was successful in preventing severe
deterioration of downstream water quality due to combined
299
-------
sewer overflows. Average annual BOD5 reduction was 27
percent. Best evidence of the efficiency of the facility
was afforded by the fact that incidence of fish kills was
limited following construction of the storage/oxidation
pond.
Plans are presently underway to expand the facility to 3,280
I/sec (75 mgd), including the construction of a clarifier
and chlorination facility ahead of the existing pond. The
new clarifier is necessary to reduce the sediment buildup in
the existing pond. Approximately 21 percent of the pond had
been filled with sediment after 2-1/2 years of operation.
Shelbyville, Illinois [1] — Three different combined sewer
overflow treatment units were constructed during 1968-69.
Two units included storage/oxidation ponds and one unit was
a primary sedimentation tank. The two storage/oxidation
pond treatment schemes are described below.
Southeast site — A single-cell oxidation pond serves as a
combined sewer overflow treatment facility for a 17.8-ha
(44-acre) tributary area. The pond was designed and con-
structed by building an earthen embankment across the lower
reaches of a natural ravine downstream from an existing com-
bined sewer overflow. At the maximum water depth of 2.4
meters (8 feet), the pond has a volume of 7,230 cu m
(255,600 cu ft) and a surface area of 0.41 ha (1 acre). The
volume is sufficient to contain the 10-year storm of 1-hour
duration.
The liquid level in the pond is controlled by two 30.5-cm
(12-inch) square sluice gates mounted in an effluent
structure. One gate is located at an elevation correspond-
ing to the pond bottom, and the second gate is mounted
0.6 meter (2 feet) higher. The lower gate is normally
closed, and the upper one is set to drain a full pond to
the 0.6-meter (2-foot) level in five days.
Normal pond operation is a flow-through pattern, and during
dry weather, the average water depth is 0.6 meter (2 feet).
During wet weather, the pond discharge increases as the
depth of water over the effluent sluice gate increases.
There is always about 0.5 I/sec (12,000 gal./day) of dry-
weather flow into the pond from undetermined sources. The
pond, at the 0.6-meter (2-foot) level, provides a detention
time of 6 days at this flow. The average BOD5 loading for
the dry-weather flow is 35.5 kg/ha/day (31.7 Ib/acre/day).
During dry weather, the effluent contained 15.7 mg/1 BOD5
and 31 mg/1 SS. This was the result of removal efficiencies
300
-------
of 47 and 57 percent for BODs and SS, respectively. During
periods following combined sewer overflows, the pond was
found to discharge a very low BODs effluent (6 mg/1).
Suspended solids in the effluent were considerably higher
(50 to 75 mg/1) with only a moderate average decrease with
time. Both SS and BOD5 were discharged in greater concen-
trations during periods of high influent flows because of
carryover and short-circuiting. Fecal coliforms were also
reduced across the pond.
Southwest site — To provide treatment for combined sewer
overflows from a tributary area of 183 ha (450 acres) during
wet weather, as well as tertiary treatment for a dry-weather
trickling filter plant effluent, a storm-holding lagoon and
a two-cell facultative pond were constructed. The first
cell is designated as the stabilization pond and the second
cell is the polishing pond. The storm-holding lagoon has a
capacity of 15,140 cu m (535,000 cu ft) at a maximum depth
of 2.4 meters (8 feet). This lagoon was sized to contain
the expected runoff from a 10-year storm of 1-hour duration.
The maximum flow anticipated is 4,820 I/sec (110 mgd).
Effluent from the storm-holding lagoon is pumped to the
stabilization pond. The pumps can empty the lagoon in about
5 days. Effluent from the trickling filter plant is pumped
to the stabilization pond. The effluent from the stabiliza-
tion pond flows by gravity to the polishing pond. Effluent
from the polishing pond is chlorinated before discharge to
the receiving water stream.
BOD5 and SS removal efficiencies for the storm-holding
lagoon were reported to be 73 and 86 percent, respectively.
Based on dry-weather performance, the two-cell facultative
pond and chlorination system following the storm-holding
lagoon will provide significant reductions in soluble nutri-
ents and complete elimination of measurable fecal coliform
organisms.
East Chicago, Indiana - The recently completed combined
sewer overflow treatment facility at East Chicago consists
of a pumping station and a 12.1-ha (30-acre) facultative
lagoon that is 12.2 meters (40 feet) deep. During storm
conditions, the combined sewer overflow is pumped to a di-
version structure at the outlet weir where excessive flows
can be bypassed directly to the Little Calumet River.
During normal storm conditions, the flow enters the lagoon
for treatment. Nine surface aerators provide oxygen to the
top few feet of the water in the lagoon for aerobic oxida-
tion of the organic matter in the combined sewer overflow.
The sludge that settles to the bottom is decomposed
anaerobically. Effluent from the lagoon is discharged
301
-------
through a slotted weir having a maximum capacity of 876
I/sec (20 mgd).
This is a dual-purpose lagoon, since it also provides efflu-
ent polishing for a municipal sewage treatment plant.
Evaluation of the lagoon is underway currently. Thus, no
operational data are available at present.
Mount Clemens, Michigan [27, 8, 21] — At Mount Clemens, com-
bined sewer overflows from an 86-ha (212-acre) test area are
treated in three lagoons in series. The first lagoon acts
as a combination storage basin and aeration lagoon, the
second as an oxidation pond, and the third as another aera-
tion lagoon. The average depth of the lagoons is 2.4 to
3.1 meters (8 to 10 feet). Overflows up to a design maximum
rate of 2,800 I/sec (65 mgd) are directed to the first
lagoon for storage and partial treatment. They are then
pumped at a constant 43.8-1/sec (1-mgd) rate through a micro
strainer to the second pond, and then flow by gravity to the
third lagoon. The design retention times in the ponds are
4 days, 8 days, and 7 days, respectively. Effluent from the
final lagoon is discharged to the Clinton River following
high-rate pressure filtration and chlorination. During dry
weather, water in the last two ponds is recycled through the
filters and reclaimed for recreational purposes (fishing and
boating).
The lagoons or "lakelets" are free-form in shape and set
within an attractive 9.7-ha (24-acre) park site with scenic
walks, benches, and picnic sites. The variable-level lake-
.let 1 is obscured from public view, but lakelets 2 and 3 are
to be opened for small-boat sailing, canoeing, and fishing.
In 28 combined sewer overflow operations in the period
August to December 1972, BODs and SS removals averaged bet-
ter than 90 percent, with average effluent concentrations of
5.2 mg/1 and 14.5 mg/1, respectively. The main factor af-
fecting removal efficiencies was carryover of algae and
other microorganisms in the effluent. At Mount Clemens, in
anticipation of this problem, a two-stage pressure sand
filter and a microstrainer were installed to polish the
effluent. Sand filtration proved the most successful.
Two major operational problems were experienced. During
freezing weather, spray freezes on the motors, causing the
units to overturn; thus, their use was limited to periods
when freezing of the spray did not occur. The second prob-
lem was inadequate provision for sludge storage in some of
the lagoons.
302
-------
In response to highly favorable public reaction and to off-
set the problems experienced, plans are underway to greatly
extend and expand the present facilities. A schematic dia-
gram of the proposed modified facilities is shown on
Figure 63. The plan is to construct some new interceptors
to drain combined sewer overflows from about 4/5 of the city
to diversion structures. The rest of the city is on a sepa-
rated system. The diverted combined sewer overflows would
be pumped to a new retention structure. This structure
would provide some sedimentation in a prestorage unit and
aeration in the retention basin. Stormwater would be pumped
from here after equalization to a new clarifier and dual-bed
filtration unit and then to the existing lagoons. The first
lagoon will use air guns for aeration, and the last lagoon
will use tube-diffused aeration. The lagoons will be stone
lined and will have shallow side slopes to improve their
appearance. It is interesting to note that the land around
the pond site is being built up with high rise apartment
buildings.
COSTS
Costs for the various EPA demonstration grant projects have
been collected for comparison purposes. Construction costs
and the available operation and maintenance costs are pre-
sented in Table 65. The costs are also given on the basis
of capacity and tributary area. It should be remembered
that those costs are generally derived from a single proto-
type facility and costs for future facilities could vary
tremendously from those presented.
The cost of the Kenosha contact stabilization plant is based
on the cost of an aeration chamber, secondary clarifier,
some site piping, pumps, and controls. The capital cost
does not include the cost for headworks and chlorination
facilities. The operation and maintenance cost is based on
limited data, since the study is still underway.
At New Providence, New Jersey, the total construction costs
of $1,410,000 (ENR 2000) were for a complete treatment plant,
including modifications of an existing main pump station and
new primary and secondary clarifiers, two trickling filters,
chlorine contact tank, administration building, and miscel-
laneous items. Land costs were not reported. Since it is
unrealistic to charge all of these costs to storm facilities
alone, the costs for the plastic medium filter and the final
clarifier, one-half the costs for the electrical equipment
and site piping, the chemical feed equipment, and one-half
the cost of the construction site work were combined to form
the cost of the storm facilities. The remaining costs were
303
-------
COMBINED SEWER OVERFLOW
EMERGENCY OVERFLOW
TO RIVER
CONTROLLED OUTLET
TO RIVER
TO RIVER
LAKELET DRAIN
TO RIVER
OVERFLOW
TO RIVER
PUMP STATION
SEDIMENTATION
RETENTION BASIN
BYPASS
CLAR FIER
_____ ......... „.,
BYPASS
POTENTIAL CHEM. ADDITION
SETTLED SOLIDS
]** TO SANITARY
I
I
BACKWASH
t
L-i I ^
BACKWASH
"I
RECIRCULATION
BETWEEN
LAKELET NO. 1
i
LAKELET NO. 2
i i
LAKELET NO. 3
>
IY IRRIG.
r
STORMS 1
1
1
t
1
1
1
1
nanviviou .
SAND FILTERS "^
TO
RIVER
Figure 63.
Schematic of retention and treatment facilities,
including proposed modifications,
Mount Clemens, Michigan
304
-------
Table 65. CAPITAL AND OPERATION AND MAINTENANCE
COSTS FOR BIOLOGICAL TREATMENTa
Capital cost
(construction cost excluding land) Operation and maintenance
Plant — — -- : - : — cost (annual cost assuming
Type of plant capacity, ^/tributary- 250 hr/yr of operation),
and location mgd $ $/mgd acre
-------
Additional cost data for oxidation ponds to treat 1,095
I/sec (25 mgd) of overflow are presented in Table 66. The
original cost data for treating combined sewer overflows by
means of oxidation ponds varied tremendously because of the
variable pond volumes and detention times. To eliminate
this difference, ponds at all sites listed in the table were
adjusted to a 10-day detention time at a capacity of 1,095
I/sec (25 mgd). Both the results of this modification and
the original data are shown. The modified capital costs are
within 30 percent of each other; the average cost is
$1,731,100. Probably the best way to express cost is by
dollars per cubic meter. The average for this value is
$1.83/cu m ($2,256/acre-ft).
Cost data for aerated lagoons treating combined sewer over-
flows should be similar to costs for oxidation ponds except
for the added capital and operation and maintenance costs
for the aeration equipment.
Table 66. COST FOR OXIDATION PONDS TO TREAT
25 MGD OF OVERFLOWa
Shelbyville
Item Springfield Southeast Southwest
Original data adjusted
to 25-mgd flow rates
Capital cost,
excluding land cost: $ 188,000 $ 689,000 $1,814,000
Detention time, days 0.9 5.0 10.5
Volume of ponds at
design level, acre-ft 6.9 384 806
Operation and maintenance
cost for 250 hr/yr of
operation $ 2,200
Cost to treat 1,000 gal.
of overflow for 250 hr/yr,
*/l,000 gal. 0.02
Modified data
Capital cost, excluding
land cost at a dentention
time of 10 days $2,090,700 $1,376,400 $1,726,200
Capital cost per acre-ft
of pond volume $ 2,725 $ 1,794 $ 2,250
a. ENR = 2000.
Note: acre-ft x 1,233.4 = cu m
-------
Section XII
PHYSICAL-CHEMICAL SYSTEMS
When a high-quality effluent is required, such as may be ex-
pected in stormwater reclamation and reuse, physical-chemical
treatment systems may become both feasible and desirable.
As used in this text, physical-chemical systems imply a
means of treatment in which the removal of pollutants is
brought about primarily by chemical clarification in con-
junction with physical processes. The process string gener-
ally includes preliminary treatment, chemical clarification,
filtration, carbon adsorption, and disinfection.
In this section, the basic unit processes are discussed and
examples are cited for the treatment of municipal waste-
water and potential applications on storm sewer discharges
and combined sewer overflows. The inclusion of the munici-
pal wastewater examples is considered necessary because of
the very limited data base for operating installations.
INTRODUCTION
Physical-chemical treatment is not a new technology. In
fact, chemical precipitation, discovered in 1762, was a well-
established method of sewage treatment in England as early
as 1870 [18] . Chemical treatment was also used extensively
in the United States in the 1890s and early 1900s, but
with the development of biological treatment, the use of
chemicals was largely abandoned [18], In the 1930s, a num-
ber of physical-chemical systems were evaluated. These sys-
tems produced results that were superior to primary sedimen-
tation followed by activated sludge but at a cost of 1.5 to
2 times that of conventional treatment [10].
During the last 12 years, EPA-supported research has ad-
vanced physical-chemical treatment technology to the point
where it is becoming competitive in cost with biological
treatment, especially for situations where significant
phosphorus removal is required [14].
307
-------
Physical-chemical processes are of particular importance in
storm flow treatment because of their adaptability to auto-
matic operation (including almost instantaneous startup and
shutdown), excellent resistance to shockloads, nonsuscepti-
bility to biological upsets or toxicity, and ability to
consistently produce a high-quality effluent.
The most promising combination of unit processes that will
produce the desired effluent quality at a reasonable cost
appears to be chemical clarification followed by adsorption
on activated carbon. Several variations of this scheme,
which have been proposed and tested on a pilot scale, are
discussed in this section. Other processes that are in the
developmental stage but are far from the comparative evalu-
ation stage, are ultrafiltration, sorbent resins, magnetic
separation, ultrasonic flocculation, and reverse osmosis
[4, 5].
UNIT PROCESSES
The major unit processes utilized in physical-chemical treat
ment are discussed below. It is emphasized that (1) most of
these studies were at pilot-plant scale and (2) all of the
investigations were performed under unique local conditions.
Even with all of the testing that has been done to date, not
enough data are available from prototype facilities to have
developed standardized design criteria. Thus, pilot studies
should still be performed prior to designing a full-scale
installation.
Chemical Clarification
Chemical clarification provides the major portion of the
pollutant removal achieved by physical-chemical processes.
Raw wastewater, after coarse screening and grit removal, is
treated with a coagulating chemical. Chemicals commonly
used are lime, iron or aluminum salts, polyelectrolytes, and
combinations thereof* Following chemical addition, the
wastewater is then allowed to flocculate and settle. The
resulting sludge can be dewatered and either disposed of or
processed for recovery of chemicals. Lime sludges can be
recalcined for lime recovery if this proves economical.
At the present time, there is no rational method for pre-
dicting the dose of chemical required. Jar tests are most
commonly used for planning purposes. Fortunately, field
control of the coagulant dose is possible. The suggested
method for controlling chemical dosages is monitoring the
turbidity of the clarifier effluent [14] . With lime as the
coagulant, however, excellent control is achieved with
308
-------
pH measurement. Monitoring of phosphorus also appears prom-
ising because good clarification is usually obtained when
sufficient chemical is added to provide phosphorus removal.
Treatment efficiencies obtained with chemical clarification
in pilot plants at various locations are presented in
Table 67. In addition to, the expected high removals of SS
and phosphorus, significant organic removals were obtained.
On the basis of these and other data, it can be conserva-
tively estimated that chemical clarification of raw sewage
will consistently provide 65 to 75 percent removal of the
organics. With this degree of organic removal, the carbon
adsorption need provide only a small increment of additional
removal to match the performance of a good secondary treat-
ment system [13].
Design criteria for the coagulation equipment are similar to
those used in water treatment plants. Generally, these con-
sist of a flash mix of 1 minute, flocculation for 15 to 30
minutes, and sedimentation at upflow rates of 20.35 to
40.7 1/min/sq m (0.5 to 1.0 gpm/sq ft) [14].
Table 67. ACHIEVEMENTS OF CHEMICAL
CLARIFICATION [14]
Plant
Chemical
BOD5 SS Phosphorus
removal, removal, removal,
Blue Plains ,
Washington, B.C.
Ewing-Lawrence
Sewage Authority,
Trenton, N.J.
New Rochelle,
N.Y.
Westgate, Va.
Salt Lake City,
Utah
Lime pH 11.5 80
170 mg/1 80
ferric
chloride
Lime pH 11.5 80
125 mg/1 70
ferric
chloride
80-100 mg/1 75
ferric
chloride
90
95
98
95
90
98
80
309
-------
Sludge disposal is a major consideration in the economics of
any chemical clarification system. Only limited data are
available on the characteristics of sludges resulting from
chemical treatment of raw sewage. Typically, the dry weight
of sludge solids produced by physical-chemical treatment is
much greater than that produced by activated sludge--almost
twice as much. However, because of better thickening charac
teristics, the volume (wet) may be greater than or less than
sludge from conventional processes.
Chemical Recovery
In the case where lime is used as the coagulant, lime re-
covery by recalcination may be economical. This serves two
purposes: (1) the cost of makeup lime is reduced because
of partial recovery through recalcination, and (2) ultimate
disposal of the organic solids is accomplished because the
organics are destroyed by combustion along with the regen-
eration of lime. This has been successfully demonstrated
at the tertiary plant at South Lake Tahoe, California. It
was found that no significant saving in chemical cost was
accomplished by recalcining the lime sludge but that sludge
disposal costs were reduced because only a portion of the
lime sludge must be disposed of to prevent buildup of inerts
[24].
The recovery of alum has been investigated in one pilot
plant study [22]. Alum is recovered by acidifying the
thermally regenerated carbon-alumina slurry to a pH of 2.0
prior to recycle for raw sewage treatment. Further studies
may prove alum recovery competitive with lime recovery by
recalcination.
Recovery schemes for iron sludges have not yet been
developed. In the event that an iron recovery system is
developed involving incineration, it should be pointed out
that the use of fluidized bed furnaces on high iron content
sludges has resulted in some operational problems in at
least one installation [1]. The primary problems encoun-
tered were manifested in failures of the scrubber material
and in depositions of solid materials in the reactor duct
work and other sections of the scrubber system. It was con-
cluded by the investigators that the presence of high con-
centrations of iron and chloride in the plant influent due
to infiltration (brackish) was the major source of the
problems.
Filtration
The unit process of filtration has been previously discussed
under physical treatment in Section X. From the standpoint
310
-------
of a physical-chemical plant, two additional major consider-
ations must be resolved: (1) the location of the filter in
the process train and (2) the types and depths of media.
Where granular carbon columns are used, usually it is neces-
sary to locate the filter ahead of packed bed columns to
prevent rapid clogging of the column and after expanded bed
columns to capture column carryover solids. In powdered
carbon systems, a filter usually has to be located at the
end of the process to capture solids carryover from the
clarifier.
To offset the limitations of surface filtration by single-
medium filters (which usually capture the particles within
the first inch of filter medium [25]), in-depth filtration
by dual-media or tri-media filters has been developed. The
intent again is to utilize the full depth of the filter in
separating out the solids; thus, the desired gradation in
direction of flow is from coarse to fine. To maintain this
gradation during backwashing operations, media of different
densities as well as sizes are required.
It should be recognized that there is no one mixed-media
design that will be optimum for all wastewater filtration
applications. Small quantities of high-strength biological
floe, typically found in activated sludge effluents, may be
satisfactorily removed by a good dual-media design, while a
weak floe or increased solids loading may best be removed by
a mixed, tri-media bed filter. The marked effect that the
quality and quantity of the floe to be removed can have on
media selection is clearly indicated by the data in Table 68
The data are for raw water treatment plants. From this it
can be seen that, in most cases, pilot test of various media
designs can be more than justified by improved plant
performance.
Data on the performance of several filter systems on munici-
pal sewage are shown in Table 69.
Carbon Adsorption
The role of the carbon adsorption step is to remove soluble
organics from the wastewater. Organics removal capacity of
carbon is usually expressed in terms of pounds of organics
removed (either as COD or TOG) per pound of carbon. For
general purposes, a capacity of 0.5 kg of COD per kg (0.5 Ib
of COD per Ib) of granular carbon is reasonable [14]. This
is approximately equal to a requirement of 60 mg/1 activated
carbon (500 Ib of activated carbon per million gallons of
sewage treated). It must be remembered that this value is
311
-------
Table 68. ILLUSTRATIONS OF VARYING MEDIA DESIGN
FOR VARIOUS TYPES OF FLOG REMOVAL [7]
Garnet Silica sand
Location
Fort St. John, Ore.
Buffalo Pound, Ore.
Peoria, Ore.
Type of a
application Size
Very heavy loading -40x80
of fragile floe
Moderate loading -20x40
of very strong
floe
Moderate loading -40x80
of fragile floe
Depth, a Depth,
in. Size in.
8 -20x40 12
3 -10x20 12
' 3 -20x40 9
Coal
Depth,
Size in.
-10x20 22
-10x16 15
-10x20 18
a. Size: -40x80 = passing No.40 and retained on No.80 U.S.sieves.
Note: in. x 2.54 = cm
Table 69. EXAMPLES OF FILTER PERFORMANCE
Scale of
instal-
Location lation
Stanford Pilot
University
[25]
Lake Tahoe Full
[9]
Bernards Pilot
Township
[20]
Bernards Pilot
Township
[20]
Bernards Pilot
Township
[20]
Bernards Pilot
Township
[20]
Washington, Pilot
D.C. [20]
BODS, mg/1 COD, mg/1
Typ c o if
Feed filter Inf. Eff. Inf. Eff.
secondary
effluent
Chemically Tri-media 9 5 23 15
treated
secondary
effluent
Settled Moving bedb 65 12
trickling
filter
effluent
Unsettled Moving bed 55 3.8
trickling
filter
effluent
Primary Moving bed 67 12
effluent
Raw Moving bed 115 19
wastewater
clarified
raw
sewage
Total
phosphorus, Turbidity,
mg/1 JTUa SS, mg/1
Inf. Eff. Inf. Eff. Inf. Eff.
0.65 0.05 7.0 0.3 15 0
9.37 0.51 33 7 50 15
19.1 0.99 39 3.4 86 7.1
14.6 1.3 53 3.7 77 11
21.5 2.16 123 16.7 156 27
a. Jackson turbidity units.
b. A single-medium filter operating essentially continuously and in a closed loop. See [20].
312
-------
suggested only as a starting point, since experience shows
that the actual requirement varies considerably.
Carbon contacting systems generally employ granular acti-
vated carbon. The wastewater is passed either downward or
upward through the columns containing the carbon. Downflow
columns function as packed beds and additionally accomplish
filtration of the wastewater. Flow rates of 1.4 to 5.4
1/sec/sq m (2 to 8 gpm/sq ft) have been used. In this flow
range, essentially equivalent adsorption efficiency is ob-
tained from either packed or expanded (upflow) beds when the
same contact time is used. At flow rates below 2 gpm/sq ft,
adsorption efficiency is reduced, while at flow rates above
8 gpm/sq ft, excessive pressure drop takes place. Contact
times ranged from 30 to 60 minutes on an empty bed basis.
In general, increases of contact time up to about 30 minutes
yield proportional increases in organic removal. At contact
times above 30 minutes, the rate of increase falls off;
beyond 60 minutes, additional contact time yields no appre-
ciable increase in adsorption [14]. Gravity flow carbon
beds are usually designed for operation at the upper end of
the flow rate range. A pressure vessel carbon bed is more
expensive to construct but requires less land area and pro-
vides a greater ability to handle fluctuating flow rates.
Periodic backwashing of the downflow bed must be provided,
even if prefiltration is utilized, because suspended solids
will accumulate in the bed. Bacterial growth also occurs on
the carbon granules and tends to reduce the adsorptive
capacity and plug the bed. To assure removal of the gelati-
nous biological growth, surface wash and air scour should be
included in the backwash system.
Backwashing downflow carbon beds effectively relieves
clogging but does not completely remove biological growth.
Consequently, a significant amount of biological activity
is present in the beds at all times leading to the develop-
ment of anaerobic conditions and the production of sulfides.
Aeration of the column feed has been utilized, but this
leads to excessive biological growth and subsequent exces-
sive backwash requirements.
To overcome these difficulties, upflow carbon columns have
been developed which are operated in a slightly expanded
mode (10 to 15 percent expansion). This allows for signifi-
cant accumulation of biological activity on the carbon
granules with little increase in pressure loss. Thus,
aerobic conditions which eliminate sulfide generation can be
maintained.
313
-------
Backwash facilities must also be provided for expanded bed
systems to remove excess biological growth. The flow rate
range used for expanded bed columns is somewhat more restric-
tive than with packed bed columns. Flow rates above 162.8
to 203.5 1/sec/sq m (4 to 5 gpm/sq ft) are required for the
proper degree of expansion of the carbon granule sizes com-
mercially available (2,380 by 545 microns or 1,000 by 370
microns [8 by 30 mesh or 16 by 40 mesh]). In addition,
care must be exercised to prevent hydraulic surges that will
carry carbon out of the system. When the expanded bed sys-
tem is used, it must be followed by filtration.
The latest development in activated carbon contacting sys-
tems applied to wastewater treatment involves the utiliza-
tion of powdered activated carbon (particle size less than
74 microns [200 mesh]). This method provides for a mixture
of carbon slurry and wastewater in a reactor clarifier.
Polymer addition is usually required to achieve a good
gravity separation of the carbon from the wastewater follow-
ing contact. A demonstration application on combined sewer
overflows is described later in this section.
Activated Carbon Regeneration
For activated carbon to be more economical, in situ regener-
ation and reuse of the spent carbon should be included in
the systems. There are few options available to the de-
signer in the area of regeneration techniques. Regeneration
by several means, particularly chemical, has been attempted
in the past, but it now appears that the only presently
feasible method for destroying the adsorbed organics is
thermal regeneration [22, 9, 21]. The usual equipment for
thermal regeneration of granular or powdered carbon is
either a multiple hearth or the fluidized-bed furnace (as
yet^untried at full-scale), respectively. Unfortunately,
capital costs for thermal regeneration are relatively high.
In the case of smaller plants, this may leave only two
choices: (1) do not regenerate and use the carbon only once,
or (2) construct the furnace and share its use with another
physical-chemical plant in the immediate area.
Overall carbon losses of 5 to 10 percent per regeneration
cycle are typical for granular activated carbon regeneration
utilizing multiple hearth furnaces. At a 10 percent carbon
loss per cycle, approximately 5 percent of the original car-
bon remains after 30 cycles. Because virgin granular carbon
costs 53 to 66£ per kg (24 to 30
-------
In a pilot-scale powdered activated carbon regeneration
process utilizing a fluidized-bed furnace, overall carbon
losses of 2 to 10 percent per regeneration cycle were
reported [22]. Even though virgin powdered carbon costs
less than granular carbon, 18 to 22$ per kg (8 to 10$ per
pound), considerable savings are possible with regeneration.
It has also been demonstrated that thermally regenerated
activated carbon (both granular and powdered) has essen-
tially the same adsorptive capacity as the virgin carbon.
PERFORMANCE OF PHYSICAL-CHEMICAL SYSTEMS
Operational and design data are available on one full-scale
and several pilot-scale physical-chemical treatment systems.
One pilot study, in Albany, New York, operated both on com-
bined sewer overflows and municipal sewage. This study is
described in detail in the next subsection. The remaining
systems operated exclusively on municipal sewage feed with
various levels of pretreatment. Average operating and per-
formance data for each of the systems are shown in Tables 70
and 71, respectively. A brief description of the major
systems follows.
Table 70, OPERATING DATA OF PHYSICAL-CHEMICAL
TREATMENT PLANTS FOR VARIOUS CITIES
Location
South Lake Tahoe,
Calif. [9]
Ewing-L'awrence
Sewage Authority,
Trenton, N.J.
[12]
Rocky River,
Ohio [23]
Washington, D.C.
[2]
Dallas, Tex.
[13]
Pomona, Calif.
r i n i
Scale
of
instal-
lation
Full
Pilot
Pilot
Pilot
Pilot
Pilot
Type
of
feed
Settled
secondary
effluent
Primary
effluent
Raw
sewage
Weak raw
sewage
Primary
effluent
Settled
Design
flow
7.5 mgd
0.08 mgd
0.007 mgd
.1 mgd
1 mgd
0.3 mgd
Chemicals added
Coagulant, Filter aid,
mg/1 mg/1
Lime, 400; Alum, 1-20;
polymer, polymer,
0.25 0.01-0.10
Ferric
chloride,
170
Polymer,
0.3
Lime, 350;
ferric
hydroxide ,
5
Lime, 350
Filters
Loading
Type gpm/sq ft
Tri- 5.0
media
Dual- 1.8
media
Dual- 3.0
media
Multi- 3.0
media
.-
Type
Granular ,
pressure
upf low
Granular,
expanded -
bed
Granular ,
gravity
packed-bed
Granular ,
gravity
packed-bed
Granular
Granular
Activated carbon
Carbon
dosage Loading
Ib/mil gal. gpm/sq ft
250 6.4
190 5.0
500 4.0
7.0
350 7.0
[19] secondary
effluent
Albany, N.Y. Pilot Combined 0.1 mgd Alum, 200; Polymer Tri- 5.0 Powdered 1,700-
[22]a sewer lime, 190 media 5,000
overflow
a. Alum recovered by acidification with sulfuric acid at a dosage of 0.6 Ib/lb carbon.
Note: Ib/mil gal. x 0.120 - mg/1
gpm/sq ft x 0.679 « 1/sec/sq m
mgd x 43.8 - I/sec
315
-------
Table 71. PERFORMANCE DATA OF PHYSICAL-CHEMICAL
TREATMENT PLANTS FOR VARIOUS CITIES
Location
South Lake Tahoe,
Calif. [9]a
Ew ing -Lawrence
Sewage Authority, K
Trenton, N.J. [12]D
Rocky River,
Ohio [23]c
Washington, D.C.
[2]d
Dallas, Tex.
[13]e
Pomona, Calif.
[19]f
Albany, N.Y.
[22]g
South Lake Tahoe,
Calif. [9]a
Ewing-Lawrence
Sewage Authority, ,
Trenton, N.J. [12]°
Rocky River ,
Ohio [23]c
Washington, D.C.
[2]d
Dallas, Tex.
[13]e
Pomona, Calif.
[19]f
Albany, N.Y.
[22]g
a. Nitrogen removed
b. No filter after
c. No filtration or
d. Nitrogen removed
Total organic
BODs, mg/1 COD, mg/1 carbon, mg/1 SS, mg/1 Phosphorus, mg/1
Inf. Eff. 1 Inf. Eff. % Inf. Eff. * Inf. Eff. * Inf. Eff. *
30 1 97 40 10 75 -- -- -- 26 0 100 6 0.1 98
52 5 90 -- -- -- 46.3 4.3 91 62 9 85 30 3 90
118 8 93 235 44 81 52 13 75 107 7 94
127 5.4 96 308 12.8 96 100 6.0 94 161 4.3 97 8.5 0.14 98
45 3 93 117 29 75 34 9 74 41 5 88 11.9 0.1 99
47 9.5 80 13.0 2.5 85 10 1 90
100 6.0 94 383 23 94 -- -- -- 420 4.2 99
Methylene blue Coliforms, Turbidity,
Nitrogen, mg/1 active substance MPN/100 ml JTUh
Inf. Eff. * Inf. Eff. * Inf. Eff. * Inf. Eff. 4
25.5 0.6 98 2.0 0.1 95 2.5 x 106 50 99+ 50 0.3 99*
-- -- 35 1.5 96
21.2 4.5 79 -- -- -- -- -- ..
6.4 2.5 61 -- -- -- 60 x 105 9 99+ 9 0.8 91
6.7 3.7 45 -- 0.4 -- -- -- -- 10.3 1.6 85
-- -- -- 1.0
as ammonia by air stripping.
expanded-bed carbon columns, phosphorus removed as phosphate.
phosphorus removal .
as total nitrogen by ion exchange.
Nitrogen removed as ammonia; phosphorus removed as total phosphorus
Nitrogen removed as nitrate by volunteer denitrification.
No phosphorus removed (see text for explanation).
Jackson turbidity unit.
316
-------
South Lake Tahoe, California [9]
This^is a full-scale 328.5 I/sec (7.5 mgd) plant utilizing
physical-chemical treatment as a tertiary step for upgrading
the biological plant effluent. The secondary effluent is
first dosed with recalcined lime (sometimes alum) at rates
up to 400 mg/1 and rapid-mixed for 1 minute, followed by
5 minutes of air-agitated flocculation. Polymer is then
added prior to clarification (0.25 mg/1). The chemically
clarified effluent passes through an ammonia stripping tower,
is recarbonated, and then is filtered by tri-media filters.
After filtration, the effluent is run through pressure up-
flow granular carbon columns followed by chlorination prior
to discharge. Carbon and lime are regenerated and recal-
cined, respectively, in separate multiple hearth furnaces.
Ewing-Lawrence Sewerage Authority, Trenton, New Jersey [12]
This pilot plant was established primarily to compare the
performance of expanded bed and packed bed activated carbon
adsorption systems in the direct physical-chemical treatment
of raw sewage and/or primary effluent.
The feed wastewater used throughout the study was first
dosed with ferric chloride, followed by rapid mixing, fol-
lowed by two-stage slow mixing flocculation. The floccu-
lated effluent then was clarified in an upflow clarifier
followed by dual-media gravity filtration. The filtered
wastewater was then pumped to a storage tank for feeding to
the carbon adsorbers. A flow diagram of the pilot-plant
clarification system used in this study is shown on
Figure 64.
One important aspect of this pilot study was that the entire
system was designed for essentially automatic operation. A
technician visited the plant daily to take samples, adjust
flows, and perform routine maintenance. Few problems were
reported after shakedown, demonstrating that ease of auto-
mation is one strong argument in favor of physical-chemical
plants over biological plants, particularly in the case of
treating storm flows.
Rocky River, Ohio [25]
This pilot-plant study was performed to provide design data
for a 438-1/sec (10-mgd) physical-chemical plant for
Cuyahoga County, Ohio. The decision to upgrade the existing
primary plant by addition of a phyical-chemical plant,
rather than by a conventional secondary plant, was prompted
by extremely limited land space and the fact that the con-
struction costs were estimated to be substantially lower.
317
-------
.RAPID MIX
CHAMBER
COAGULANT
PUMP
U
Fed
PRIMARY ^^
EFFLUENT^,
PUMP
UNT^HjSj-L
UPFLOW CLARIFIER
STRAIGHT SIDE WITH
60 DEG CONE BOTTOM
TO CARBON
ADSORPTION
WATER
FOR BACKWASH
DUAL MEDIA
FILTER
9-IN. COAL
9-IN. SAND
6-IN. COARSE SAND
GRAVEL
CLARIFIED
FEED
RESERVOIR
-*• TO CARBON
ADSORPTION
Figure 64. Flow diagram of clarification system [12]
Raw sewage was chemically clarified using a polymer. No
additional chemical treatment for further coagulation or
phosphorus removal was applied. The clarified wastewater
(approximately 50 mg/1 SS) was then fed to a four-stage
gravity flow granular activated carbon column system. The
carbon columns served the dual purpose of filtration and
adsorption. Overall system average removals were better
than those from most secondary treatment plants.
Washington, D. C. [2]
This study of physical-chemical treatment of raw sanitary
wastewater was conducted at a pilot plant in Washington,
D. C. The plant was operated jointly by the EPA and the
District of Columbia. Figure 65 is a flow diagram of the
pilot plant used in this study. The automated pilot system
consists of cyclone degritting, two-stage (high-pH) lime
precipitation (for high phosphorus removals) with inter-
mediate recarbonation, dual-media filtration, pH control,
selective ion exchange (for ammonia removal), and downflow
granular activated carbon adsorption. Overall, the physical
chemical treatment produced very consistent removals of
organics and demonstrated high operational stability.
Dallas, Texas [15, 15]
This 43-1/sec (1-mgd) pilot plant, located at the Dallas
Water Reclamation Research Center, is being jointly studied
318
-------
RAW WATER
LIME PRECIPITATION
CaO
RECYCL
l~T
co2
••^•MiWM
E
}
t
i
FeCI3
*
3
k
[T
IT
T)
I]
U
U
U
U
U
CARBON
lj*J
'
V V
ION EXCHANGE
FILTERS
Figure 65. Physical-chemical treatment pilot plant [2]
Washington, D.C.
at the present time by the EPA, the Dallas Water Utilities
Department, and the Civil Engineering Department of Texas
A§M University. The plant consists of two complete mix
activated sludge systems (aeration basins and final clari-
fier) , a chemical treatment module, two filters, two acti-
vated carbon contactors, two chlorine contact basins, two
small oxidation ponds, and the necessary chemical feed
equipment. The plant is quite flexible and offers the oppor-
tunity to evaluate many different sequences of unit
processes. Influent to the pilot plant comes from a waste-
water treatment plant (two-stage high-rate trickling filter)
and is available from any point in the process sequence.
Pomona, California [19]
The Pomona study was one of the first field applications of
carbon adsorption in the United States and has been on-line
since 1965. The plant is jointly operated by the EPA and
the Sanitation District of Los Angeles County. The purpose
of the pilot-plant study was to provide a simple and reliable
adsorption system that would be capable of receiving and
319
-------
removing suspended solids from an activated sludge plant
effluent. The design features a four-stage packed bed pres-
sure downflow granular activated carbon column and includes
a thermal carbon regeneration process.
STORM FLOW APPLICATIONS
Whereas several combined sewage treatment demonstration pro-
jects have evaluated the benefits of chemical aids to process
operations, only one pilot operation representing a complete
physical-chemical treatment system has been implemented.
This project and two representative planning studies are
discussed in this subsection.
Again, the feasibility of multiprocess physical-chemical
systems in storm flow applications will largely .depend on
the effluent quality objectives, the degree of preunit flow
attenuation, and the ability to maximize the use of the
facilities between storms. These aspects are well demon-
strated in the following projects.
In this project, raw municipal sewage and combined sewer
overflows were contacted with powdered activated carbon to
remove dissolved organics. Alum was then used to aid in
subsequent clarification. Addition of polyelectrolyte was
followed by a short flocculation period. Solids were sepa-
rated from the liquid stream by gravity settling, and the
effluent was then disinfected and discharged or filtered
prior to discharge. A process flow sheet is shown on
Figure 66.
The pilot plant was composed of two major systems: a liquid
treatment system and a carbon regeneration facility. The
liquid treatment system was housed almost entirely in a
40-foot trailer van. The major components are: (1) a surge
tank, (2) a pipe reactor and static mixers, (3) chemical
addition equipment, (4) flocculation chambers, (5) tube
settler, (6) a tri-media filter, and (7) a centrifuge for
sludge dewatering.
A solid bowl centrifuge is used to dewater sludge which is
then stored in a holding tank for subsequent pumping to the
carbon regeneration facility. The carbon sludge was readily
dewaterable. After some operation experience, it was found
that (1) rapid mixing of the dewatered sludge reduced its
viscosity rendering it more pumpable, and (2) the addition
of the same polymer used in the waste treatment process to
the sludge at a dosage of 1 kg/1,000 kg (2 Ib/ton) of dry
320
-------
MAKE-UP
ALUM &
POWDERED
ACTIVATED PH
CARBON ADJUSTMENT
(LIME)
COMBINED
SEVER
OVERFLOW
RECYCLED
ALUM t
POWDERED
CARBON
J
r
FLUIDIZED
BED
FURNACE
SLUDGE
DEWATERING
INERT
ASH
SLOWDOWN
TO IISCHARiE
Figure 68. Process flow sheet
for physical-chemical pilot plant [22]
Albany, New York
solids increased the solids capture from 20 to 95 percent.
It was also found that sludge more than 2 to 3 days old
required polymer doses as high as 2 kg/1,000 kg (4 Ib/ton)
to achieve the same 95 percent solids capture in the
centrifuge.
The carbon-alumina sludge is thermally regenerated in a
fluidized inert sand bed [26] . The regenerated carbon-
alumina slurry is collected in storage tanks. After collec-
tion, the slurry is acidified with sulfuric acid to reclaim
the alum before reuse in the system. Makeup alum and carbon
are added at the same time. Average carbon losses per
regeneration cycle were 9.7 percent. Approximately 91 per-
cent of the alum can be recovered by acidification of the
carbon-alumina slurry after thermal regeneration of the
carbon sludge. Even after seven regeneration cycles, the
adsorption performance of the carbon was essentially that
of virgin carbon.
Two types of sewage were studied, combined sewer overflows
and municipal sewage, but only the combined sewer overflow
portion of the study will be of concern in this text.
321
-------
Nine storm events occurred during the pilot-plant study
period. These ranged in duration from 2 to 7 hours. The
amounts of total rainfall during a single storm ranged from
0.13 to 2.87 cm (0.05 to 1.13 inches).
Powdered carbon dosages ranged from 500 to 1,300 mg/1 during
the storm events. When compared to carbon dosages for
treating primary or secondary effluent, these values at
first appear to be extremely high. Rough calculations of
pounds of COD removed per pound of carbon used for overall
treatment show an averager of 1.1 kg of COD removed per
kilogram of carbon used (1.1 Ib COD per pound of carbon),
which is considerably higher than that for granular carbon.
Attempts to match the pilot-plant flow variation to the com-
bined sewer flow variations were not completely successful.
However, during a storm, the pilot-plant flow, initially
2.2 I/sec (35 gpm), was increased to 4.7 I/sec (75 gpm) in
a period of less than 2 minutes during the peak storm load-
ing, with no observable effluent quality degradation or
operational upsets. Thus, it appears that plant performance
is highly insensitive to rapid changes in flow rates if
chemical feed rates are rapidly adjusted to correspond to
the increased flow. This is a strong point in favor of
physical-chemical treatment of storm flows. During the peak
pollutant loadings of these storms, the effluent quality re-
mained essentially unaffected.
Some advantages of a powdered carbon system over a granular
carbon system are: (1) the carbon dosage can be varied with
the strength of the influent waste stream; (2) fluidized bed
furnaces can be used for regeneration and they are simpler
to operate and maintain than multiple hearth furnaces;
(3) granular carbon columns and their attendant sophisti-
cated piping systems, including backwashing facilities, are
eliminated; and (4) overall, a powdered carbon system is
simpler to operate and maintain.
One major disadvantage of a powdered carbon system is that
phosphorus removal is greatly reduced when compared to gran-
ular carbon systems. This results from the acidification
step for alum recovery. Acidification redissolves the
aluminum hydroxide releasing the alum for reuse in
coagulation. Unfortunately, acidification also redissolves
phosphate, thus recycling it to the effluent. Hence, the
only phosphorus removal is accomplished by blowdown of the
regenerated carbon-alum stream. This amounts to a phos-
phorus removal of about 31 percent when 5 percent blowdown
occurs.
322
-------
San Francisco. California [3]
The San Francisco City engineering staff has recommended
physical-chemical treatment as the most suitable treatment
process to manage its total wastewater system. The concep-
tualized plant would be built according to a building block
type construction method, allowing split flow integration
of dry- and wet-weather facilities. The split flow option
will permit variable degrees of treatment depending on the
volume of inflow. The building block method permits con-
struction to proceed on a smaller increment basis to keep
pace with the development and construction of the automated
combined sewer collection/storage system (see Section XIV).
Three levels of treatment would be provided. Level 1 would
handle up to 43,808 I/sec (1,000 mgd) while sequential
Levels 2 and 3 could handle up to 25 percent of this maxi-
mum inflow.
As conceived, the Level 1 physical-chemical treatment pro-
cess includes bar screens, grit chambers, chemical sedimen-
tation with ferric chloride and polymers, and chlorination
and dechlorination. Add-on Level 2 treatment includes high
dose lime chemical sedimentation, subsequent recarbonation,
and lime recalcination. Finally, add-on Level 3 includes
superchlorination or, alternatively, ammonia stripping,
dual-media filtration, and carbon adsorption.
Estimated removal efficiencies for the overall system,
taking into account flow splitting and overflow, are 90 per-
cent COD, 98 percent SS, 85 percent total nitrogen, 93 per-
cent total phosphorus, 90 percent hexane extractable
material, and 95 percent floatables.
As San Francisco is sewered on the combined plan, this
dual-use treatment prospect offers excellent potential for
maximizing both dry- and wet-weather performance.
Kingman Lake, Washington, D. C. [6]
This conceptual engineering study involved the investigation
of the reclamation of combined sewer overflows and utiliza-
tion of the reclaimed waters in a major water-oriented rec-
reation facility. The treatment process selected to follow
a large combined sewer overflow storage reservoir was a
total physical-chemical system including chemical clarifica-
tion, filtration, carbon adsorption, disinfection, and carbon
regeneration. To increase the beneficial use of the facili-
ties, lake water would be directly processed and recycled in
323
-------
the periods between storms. Lake quality suitable for boat
ing and fishing is expected in one segment and suitable for
swimming in a second. This project is described further in
Section XIV.
COSTS OF PHYSICAL-CHEMICAL TREATMENT
Representative capital and operation and maintenance costs
projected from data developed within each project are sum-
marized in Table 72, along with a description of what is
believed to be included. Note that the operation and main-
tenance costs presume continuous year-around operation.
Much higher unit costs would therefore be expected for
intermittent storm operations. The wide variations in both
systems and projected costs emphasize the requirement for
independent estimates for each particular installation.
Table 72. ESTIMATED CAPITAL AND OPERATION AND
MAINTENANCE COSTS FOR TYPICAL PHYSICAL-CHEMICAL
TREATMENT PLANT
Operation and
maintenance costs,
Capital costs, $ f/l.OOO gal.
Location 10 mgd 25 mgd 100 mgd 10 mgd
Hypothetical
CAST3 [11] 4,822,000 9,680,800 28,330,500 9.7
Hypothetical
PCTb [8] 6,656,000 13,409,000 42,379,000 16.3
South Lake Tahoe,
Calif. [9] 4,870,300 9,907,400 29,010,600 13.0
Ewing- Lawrence
Sewage Authority,
Trenton, N.J.
[12]
Packed bed 3,548,700 7,218,900 21,138,300 12.8
Expanded bed 3,411,900 6,940,600 20,323,400 12.3
Washington, D.C.
[2] 27,065,700 55,060,700 161,227,200 31.4
Rocky River,
Ohio [23] 2,416,000 4,914,700 14,391,200 3.0
Pomona, Calif.
[19] 2,942,700 5,986,200 17,528,600 4.2
Albany, N.Y. [22] 1,791,300 3,643,900 10,670,100 18.8
25 mgd 100 mgd costs
7.1 5.3 See definition
13.4 .10.3 See definition
10.8 8.6 See definition
10.6 8.0 See definition
10.2 8.0 See definition
26.0 19.6 See definition
2.5 1.9 See definition
3.5 2.6 See definition
15.6 11.7 See definition
1
2
3
4
5
6
7
8
9
a. CAST = conventional activated sludge treatment (for comparison only).
b. PCT » physical-chemical treatment
Note: mgd x 43.808 = I/sec
-------
Table 72 (Continued)
Definition of costs
Operation and maintenance costs include all
legal, administrative; and engineering
materials, supplies, and labor.
Capital costs include two-stage lime clarification, dual-media filtration, ammonia stripping,
granular carbon adsorption, and lime recalcination. Unknown whether pretreatment and chlori-
nation are included.
Capital costs include coagulation chemicals addition systems, flash mixing, flocculation,
phosphorus removal, ammonia stripping, recarbonation, multi-media filtration, granular acti-
vated carbon adsorption, chlorination, recalcination, and carbon regeneration. Operation
and maintenance costs include all materials and labor. The PCT is preceded by a complete
activated sludge treatment system. Costs for general facilities are not clearly serparated
into CAST and PCT components.
Capital costs include aerated grit chambers, flocculation, clarifier, sludge thickener, drum
filter, two-stage packed bed activated carbon contactors, combined duty multi-hearth furnace
for recalcination and incineration of clarifier sludge, carbon storage, multi-hearth carbon
regeneration furnace, all piping, foundations, roads, fences, instrumentation, pumps, build-
ings, utilities, and engineering. Operation and maintenance costs include all materials and
labor. The cost analysis is for the treatment of raw sewage; chlorination and nitrogen re-
moval systems are excluded.
Capital and operation and maintenance costs include the above except that the carbon con-
tactors are the expanded bed type.
Capital costs include bar screens, cyclone degritting, two-stage high-pH lime treatment,
dual-media filtration, ion exchange and regeneration, carbon adsorption, recalcination, re-
carbonation, sludge disposal, and complete automation of the plant. Operation and maintenance
costs include fuel, electricity, materials, and labor. Costs are scaled from a 13,140-1/sec
(300-mgd) PCT plant cost estimate. At first glance the estimated costs seem extremely high
until it is considered that (1) this is a complete PCT plant designed for treating raw sewage;
(2) application of PCT to a plant of this scale has no precedent; (3) ion exchange regenera-
tion by air stripping requires expensive heating during the winter months; and (4) other in-
dependent studies generated similar estimated costs [17, 16].
Capital costs include only an add-on granular activated carbon column process and the build-
ing for it without carbon regeneration. Operation and maintenance costs include only the
add-on activated carbon process materials and labor. Existing facilities to be utilized for
pretreatment consist of chemical coagulation, settling, and chlorination.
Capital costs include only gravity packed bed granular activated carbon adsorption columns
and a carbon regeneration system. Operation and maintenance costs include only labor and
make-up carbon. Process is designed to handle secondary process effluent, not raw sewage.
Capital costs include screens, grit chambers, overflow facilities, pipe reactor vessels, pumps,
chemical storage, carbon slurry tanks, sludge storage, agitators, flocculators, tube settlers,
filtration, chlorination, carbon regeneration/sludge incineration, fluidized bed furnace,
chemical make-up system, 10 percent contingencies, and land. Operation and maintenance costs
include all materials, power, and labor. Plant is designed for raw stormwater treatment.
325
-------
Section XIII
DISINFECTION
Three categories of human enteric organisms of the greatest
consequence in producing disease are bacteria, viruses, and
amoebic cysts. Typical waterborne bacterial diseases are
typhoid, cholera, paratyphoid, and bacillary dysentery.
Diseases caused by waterborne viruses include poliomyelitis
and infectious hepatitis.
Disinfection refers to the selective destruction of patho-
genic microorganisms. Generally, not all of the organisms
are destroyed during the process. Viruses, cysts, and
bacterial spores are the most hardy.
AGENTS AND MEANS
In the field of wastewater treatment, the most common meth-
ods for accomplishing disinfection are through the use of:
1. Chemical oxidizing agents
2. Physical agents
3. Mechanical means
4. Radiation
In applying the disinfection agents or means, the following
factors must be considered: (1) contact time, (2) concen-
tration and type of chemical agent, (3) intensity and nature
of physical agent, (4) temperature, (5) numbers of organisms,
(6) types of organisms, and (7) nature of the suspending
liquid. Typical removal efficiencies for various treatment
processes are reported in Table 73.
Chemical Agents
The requirements for an ideal chemical disinfectant and the
characteristics of the most commonly used chemical
326
-------
Table 73. REMOVAL OR DESTRUCTION OF BACTERIA BY
DIFFERENT WASTEWATER TREATMENT PROCESSES
Removal,
Process %
Physical
Coarse screens 0-5
Fine screens 10-20
Grit chambers 10-25
Plain sedimentation 25-75
Chemical precipitation 40-80
Mechanical
Trickling filters 90-95
Activated sludge 90-98
Chemical oxidation
Chlorination of treated sewage 98-99
disinfectants are listed in Table 74. Although no such
ideal compound has been found, these requirements should be
considered in evaluating proposed or recommended
disinfectants.
Chemical agents used for disinfection include chlorine gas
or liquid, sodium or calcium hypochlorite, ozone, bromine,
and iodine. Chlorine dioxide is being evaluated as a disin-
fectant in two current EPA-sponsored demonstration projects.
Chlorine — The common sources of chlorine for use in disin-
fection are (1) liquified chlorine gas, (2) sodium or calcium
hypochlorite, and (3) electrolytic generation.
Chlorine gas — Liquified chlorine gas is available in 45- to
68-kg (100- to 150-lb) cylindrical containers, 907-kg (1-ton)
containers, and 1,450- to 4,980-kg (16- to 55-ton) railroad
tank cars. In most cases, liquified chlorine gas is deliv-
ered to the plant site by truck or railroad. It is extremely
327
-------
Table 74. COMPARISON OF IDEAL AND ACTUAL CHEMICAL
DISINFECTANT CHARACTERISTICSa
Characteristic
Toxicity to
microorganism
Solubility
Stability
Nontoxic to
higher forms
of life
Homogeneity
Interaction
with extraneous
material
Toxicity at
ambient
temperatures
Ideal disinfectant
Should be highly toxic at
high dilutions
Must be soluble in water or
cell tissue
Loss of germicidal action on
standing should be low
organisms and nontoxic to man
and other animals
Solution must be uniform in
composition
Should not be absorbed by
organic matter other than
bacterial cells
Should be effective in
ambient temperature range
Chlorine
High
Slight
Stable
Highly toxic
to higher
life forms
Homogeneous
Oxidizes
organic
matter
High
Sodium
hypochlorite
High
High
Slightly
unstable
Toxic
Homogeneous
Active
High
Calcium
hypochlorite
High
High
Relatively
stable
Toxic
Homogeneous
Active
High
Chlorine
dioxide
High
High
Unstable, must
be generated
as used
Toxic
Homogeneous
High
High
Ozone
High
High
Unstable, must
be generated
as used
Toxic
Homogeneous
Oxidizes
organic matter
Very high
Penetration Should have the capacity to High High High High High
penetrate through surfaces
Noncorrosive Should not disfigure metals Highly Corrosive Corrosive Highly Highly
and nonstaining or stain clothing corrosive corrosive corrosive
Deodorizing Should deodorize while High Moderate Moderate High High
ability disinfecting
Detergent Should have cleansing action
capacity to improve effectiveness of
disinfectant
Availability Should be available in large Low cost Moderately Moderately Moderate High cost
quantities and reasonably low cost low cost cost
priced
a. Adapted from Reference 11.
toxic and must be handled with due care and adequate
safeguards. It may be introduced either directly into water
through diffusers or dissolved in a small flow of water and
carried as a solution to the point of application. Because
the gas may escape through the water in a direct feed appli-
cation, solution feed is preferable.
Hypochlorite — Hypochlorite is being used increasingly be-
cause it is safer than chlorine gas. It is available as
either calcium hypochlorite or sodium hypochlorite.
Calcium hypochlorite is available in either dry or wet form.
High-test calcium hypochlorite contains 70 percent available
chlorine. The dry form (powder or granules) is more stable,
but feeding the dry form is usually more costly and more
difficult to control.
328
-------
Sodium hypochlorite solution is the form of hypochlorite
used most often. It is available in concentrations from 5
to 15 percent; 30 percent is maximum. The solutions decom-
pose more rapidly at high concentrations and are affected
by exposure to light and heat. They must therefore be
stored in a cool location in corrosion-resistant tanks (such
as PVC-lined steel or fiber glass).
When large quantities of hypochlorite are needed, attention
has been given to on-site generation. Two methods are cur-
rently in use for on-site hypochlorite generation. In the
first, sodium hydroxide and chlorine are reacted to form
sodium hypochlorite. Such a plant, discussed in detail
later in this section, was constructed in New Orleans,
Louisiana. The second method of on-site hypochlorite gener-
ation is electrolysis of a sodium chloride brine solution.
One currently available process, developed by constructors
John Brown Limited (C.J.B.) of Great Britain, involves the
electrolysis of a combined flow of sewage and sea water at
a ratio of 20:1 for in situ hypochlorite generation. The
system will produce Equivalent hypochlorite concentration
of 80 ppm." In the United States a prototype plant for on-
site electrolytic hypochlorite generation from brine was
constructed for the Somerville Marginal Conduit Pretreatment
Facility described later in this section.
Chlorine dioxide — Chlorine dioxide (C102) is a good bacter-
icide at high pH values, but it decomposes rapidly (1 to
2 days) and is usually manufactured where it is to be used.
This is normally accomplished by mixing solutions of sodium
chlorite and chlorine in a reaction chamber. The resulting
chlorine dioxide has an oxidizing ability nearly as high as
hypochlorous acid, but does not react with ammonia to form
chloramines, and will oxidize phenols without the formation
of chlorophenols which affect taste and odor. Until re-
cently, the lack of a satisfactory test for residual chlo-
rine dioxide in the presence of residual chlorine has
hindered its application in water and wastewater treatment.
Its chief disadvantage, however, is that it costs more than
other chlorine'compounds. It also is explosive and picks up
metal contaminants if held in contact with metal parts.
Ozone — Ozone is an allotropic form of oxygen with powerful
oxidizing and disinfecting properties. It is about twice as
potent as chlorine and has much shorter reaction times. It
is very unstable, however, and requires on-site production
because its half-life before breaking down to normal oxygen
is measured in minutes. In water, it is difficult to obtain
residuals after longer than 5 to 10 minutes. Of the many
methods for producing ozone, the most practical for large-
scale use is the electrical corona-discharge method.
329
-------
As a disinfectant, ozone has the following advantages com-
pared with chlorine:
1. Ozone is a more rapid and stronger oxidant than
chlorine. Ozone also is much more effective
against viruses.
2. Ozone is more effective in eliminating odor-causing
compounds in water, such as phenols and amines.
Chlorination merely converts these to compounds
that are more amenable to oxidation, and the trend
toward recycling and reusing water will intensify
this problem.
3. The ozonation process introduces considerable
amounts of air or oxygen into the waste, thus in-
creasing the dissolved oxygen content of the
receiving stream.
4. Ozone acts almost instantaneously (600 to 3,000
times more rapidly than chlorine) so shorter con-
tact times are needed. It is only slightly af-
fected by changes in temperature and pH.
5. No chloramine or chlorinated hydrocarbons, which
can have toxic effects on aquatic life, are
produced.
6. Ozone also oxidizes soluble organics.
The disadvantages associated with ozone are as follows:
1. Capital costs for ozone production equipment are
relatively high.
2. Ozone is generally thought of as a dangerous
substance.
3. The lack of a measurable residual makes it diffi-
cult to monitor the application rate.
4. To disinfect with ozone effectively, as with chlo-
rine, SS reduction is of prime importance.
Ozone may be produced from either pure oxygen or air. The
theoretical limit for ozone generation from pure oxygen is
about 8 percent; the practical limit is about 6 percent.
From air, the concentrations are approximately 2 percent
and 1 percent, respectively. It takes about 50 kg (110 Ib)
of oxygen to produce 1 kg (2.2 Ib) of ozone. For this rea-
son, pure oxygen used to generate the ozone is recycled to
330
-------
the generator. If air is used, there is no need for
recycling.
Other Chemical Agents — Other chemical agents used in disin-
fection include bromine, iodine, fluorine, and potassium
permanganate.
When bromine is used as a disinfectant, it is usually used
for swimming pool disinfection. McKee [10] has shown that
to achieve a 99.995-percent kill of coliforms in sewage at
Pasadena, California, required about 45 mg/1 of bromine
(or iodine) but only 8 mg/1 of chlorine. Bromine chloride,
another chemical agent, acts like bromine but does not
form persistent amines.
Iodine has high bactericidal effectiveness, a more stable
residual than chlorine, and is not affected as much by pH.
On a weight basis, it kills more viruses in a wide range of
natural water than either chlorine or bromine, because in-
hibitory amines are not formed with iodine. At the present
time, however, elemental iodine is not competitively priced
at $2.86/kg ($1.30/lb) .
Fluorine has been found too reactive to store and apply
easily. Potassium permanganate is expensive and imparts a
color that must be removed after using. Also, the pH for
its most effective use is higher than the pH in most storm-
waters and combined sewer overflows.
Physical Agents
Physical disinfectants that can be used are solids separation,
heat, and radiation. Ultraviolet light has been used success-
fully for the sterilization of small quantities of water.
Ultraviolet Light — To ensure disinfection by ultraviolet
light, the water must be relatively free of suspended matter
that might shade the organisms against the light and the
water depth must be shallow for adequate light penetration.
Time and exposure intensity are also important. One hundred
microwatts per square centimeter of 2,537 Angstrom light will
produce high bacteria kills at a contact time of less than
1 minute [14] . Use of ultraviolet light for disinfection
has found limited application probably because other methods
are more economical. An example of an ultraviolet light
arrangement for disinfection is shown in Figure 67.
Heat — Heating water to the boiling point will destroy the
major disease-producing, nonspore forming bacteria. Heat is
commonly used in the beverage and dairy industry, but it is
331
-------
ULTRAVIOLET LAMPS
INLET
OUTLET
PHOTOELECTRIC
SENSING DEVICE
^
Figure 67. Ultraviolet I ight
used as a method of disinfection
not a feasible means of disinfecting large quantities of com
bined sewer overflows or storm sewer discharges because of
the high cost.
Mechanical Means
Bacteria are also removed by mechanical means during sewage
treatment, by such processes as trickling filters and acti-
vated sludge. These processes, when applied to combined
sewer overflows, also will remove bacteria. Additional
mechanical means for removing bacteria from combined sewer
overflow include microscreening; sand, dual-media and
multi-media filtration; and carbon adsorption.
Gamma Radiation
The major types of radiation are electromagnetic, acoustic,
and particle. Because of their penetrating power, gamma
rays have been used to disinfect wastewater.
Conventional municipal sewage disinfection generally in-
volves the use of chlorine gas or sodium hypochlorite as the
disinfectant. To be effective for disinfection purposes, a
contact time of not less than 15 minutes at peak flow rate
and a chlorine residual of 0.2 to 2.0 mg/1 are commonly
recommended.
Where storm sewer discharge or combined sewer overflow dis-
infection is concerned, a different approach is required
332
-------
mainly because such flows have characteristics of inter-
mittent, high flow rate and high SS content, wide range of
temperature variation, and variable bacterial quality.
Current research on storm sewer discharge and combined sewer
overflow disinfection techniques include such aspects as:
1. Application of high-rate (rapid-oxidizing)
disinfectants.
2. Application of increased concentrations of
disinfectants.
3. Use of various alternative disinfectants.
4. On-site generation of disinfectants to reduce the
unit cost and counteract concentration decay during
storage.
High-rate disinfection refers to achieving either a given
percent or a given bacterial count reduction through the use
of (1) decreased disinfectant contact time, (2) increased
mixing intensity, (3) increased disinfectant concentration,
(4) chemicals having higher oxidizing rates, or (5) various
combinations of these.
Chlorination
Chlorination is the most commonly used and usually the most
economical method of disinfection. Although chlorine gas is
inexpensive, it is extremely hazardous. Moreover, it is
available in a limited range of container sizes and these
containers must be moved to the point of use. For this
reason, it is undesirable to use chlorine gas for disinfec-
tion at locations that are not staffed continuously or that
are accessible to the public.
Factors Affecting Disinfection by Chlorination — The extent
of disinfection by Chlorination isa function of (1) chlo-
rine residual, (2) contact time, (3) mixing intensity,
(4) chlorine demand, (5) nature and concentration of pro-
tective suspended solids, (6) pH, and (7) temperature. The
first and last two parameters pertain to the chemical form
and activity at which the disinfecting agent exists.
Contact time and mixing intensity — Contact time deals with
the opportunity for destruction. Until recently, there has
been little information on the effect of mixing intensities.
Collins, Selleck, and White [3] have shown clearly the magni
tude and influence of mixing intensity by the comparison of
bacterial kills for two different mixing intensities, one
from a tubular reactor and one from a stirred reactor.
333
-------
Velocity gradient was first proposed by Camp and Stein to
explain flocculation rates [2]. Glover further suggested
the use of velocity gradient as a measurement of the mixing
intensity [8].
G =
where G = velocity gradient, sec"
y = viscosity, centipoise
V = velocity, ft/sec
S = slope, ft/ft
From the Manning's Equation for flow in open channels
S = V
where S = slope, ft/ft
V = velocity, ft/sec
n = roughness factor
R = hydraulic radius, ft
Since the velocity gradient increases proportionally with
n, Glover suggested and designed a contact basin incorpor-
ating parallel corrugated baffles. These baffles act to
increase n value, causing flow turbulence. Limited test
results indicated that by using corrugated baffles within
the contact tank, disinfection can be achieved within 2 min
utes to obtain a desired kill to less than 100 MPN total
coliforms per 100 ml. Within practical limits, the same
results were obtained for chlorine dosages of 15, 10, and
5 mg/1 [8] . No minimum dosage was established. A compari-
son of data on the corrugated baffle contact tank, the
tubular and stirred batch reactors in Collin's experiment,
and a conventional contact basin is presented in Table 75.
These results tend to support the following arguments:
1. Effective disinfection is achievable at low chlo-
rine residual and reduced contact times when there
334
-------
Table 75. COMPARISON OF DATA ON
DISINFECTION DEVICES [8]
Type of chamber
Tube [3]
Stirred batch [3]
Corrugated baffles
Corrugated baffles
Conventional
Chlorine Contact
residual, time, t,
mg/1 sec
5 60
5 60
3 240
3 120
1,800
Velocity
gradient, G,
_ i
sec 1
6,800
400
40
40
6
Fraction of
organisms
' Gt, surviving,
dimensionless dimensionless
4 x 105 6
2.4 x 104 1
1 x 104 3
4.8 x 10^ 1
1.1 x 104 1
x 10"
x 10"
x 10"
x 10"
x 10"
4
1
4a
3a
2est.
a. Based on limited tests.
is sufficient mixing intensity, although further
verification is essential.
2. By providing good mixing intensity and promoting
plug flow, good bacterial kill performance can be
achieved in corrugated baffle contact tanks.
3. The tubular reactor offers another possible appli-
cation, but requires a long outfall conduit, and
the conduit size and storm overflow rate must guar-
antee turbulent flow conditions.
One advantage of the high velocity gradient, short resident
time contact chamber is that it is less sensitive to flow
rate than the conventional contact chamber, and considerably
less likely to be impaired by short circuiting. Neverthe-
less, the use of a Sutro weir was recommended [8] for the
design, where sufficient head is available, to produce a
rather constant Gt product at the peak design flow condition
The primary concern in the corrugated baffle contact tank is
to promote plug flow while maintaining good mixing intensity
Since the objective is plug flow, it is necessary to dis-
perse the disinfectant throughout the flow before it enters
the contact tank. One method for dispersing the disinfect-
ant, which is becoming more commonly used, is rapid flash
mixing using a mechanical mixer. This method has proved
satisfactory in enhancing bacterial kill by ensuring com-
plete dispersion of the disinfectant.
335
-------
Temperature — It has been shown that temperature can be an
important variable in disinfection. Specifically, with pH,
initial bacterial count, and chlorine dioxide dosage held
constant, it required 5 times as much contact time to obtain
a 99 percent kill at 5 degrees C as it did to obtain the
same kill at 30 degrees C [1]. Thus, temperature may be im-
portant in addition to the usual disinfection control param-
eters because the temperature range for combined sewer
overflows is usually greater than that for municipal sewage.
As a result, it might be necessary to vary disinfectant
dosage seasonally or as affected by ambient temperature [4].
In results from recent tests conducted at Syracuse [12] , it
was reported that there appeared to be no difference in the
effectiveness of either sodium hypochlorite or chlorine
dioxide at temperatures ranging from 2 to 30 degrees C.
This somewhat surprising result was attributed to the com-
plex nature of the reactions between chlorine and chlorine
dioxide and the constituents of sewage.
Chlorination control system — Flow fluctuation is usually
greater for combined sewer overflows than for treatment -
plant effluents. Hence, it is difficult to maintain a de-
sired chlorine residual. The most suitable control system
should provide automatic control on both inflow and effluent
as compared to either one separately or manual control.
Shown on Figure 68 is a schematic diagram of an automatic
hypochlorite control system in New York City 116] . It in-
cludes the measurement of the upstream influent flow rate
and the downstream chlorine residual. Information concern-
ing the inflow rate and the residual are fed into a con-
troller which directs the hypochlorite feed.
Alternatives to Chlorine Gas
Alternatives to chlorine gas for disinfection of combined
sewer overflows and storm sewer discharges are being sought.
Three major alternatives to chlorine gas are sodium hypo-
chlorite, chlorine dioxide, and ozone, all of which can
be generated on-site. Typical examples are (1) sodium hypo-
chlorite generated either from sodium hydroxide and chlorine
or electrolytically from brine, (2) chlorine dioxide gener-
ated from sodium chlorite and chlorine, and (3) ozone gen-
erated from air. The disinfectants used by various cities
in treating combined sewer overflows and storm sewer dis-
charges and their sources are listed in Table 76. In the
case of generation involving chlorine as a reagent, the
disinfectant is generated at treatment plants either for use
at the treatment plant or for transfer to large holding
tanks at the application sites. In either situation, the
-------
FLOW
TRANSMITTER
HYPOCHLORITE
FLOW RECORDER I
HYPOCHLORITE
FLOW CONTROLLER
PARSHALL
FLUME
RESIDUAL
ANALYZER
HYPOCHLORITE
STORAGE TANK
MAGNETIC
FLOW METER
HYPOCHLORITE
FLOW CONTROL
VALVE TO
OIFFUSER
Figure 68. Chlorination control system,
Oakwood Beach sewage treatment plant, New York City [16]
volume of chlorine used is considerably less when it is used
as a reagent than when it is used alone.
Optimization of storage volume and generating capacity are
important for on-site generation projects. This allows
generation of the disinfectant to replenish supplies during
periods when it is not being used. The critical parameters
are the stability of the chemical produced and the frequency
and magnitude of the overflow event in the drainage area.
Sodium Hypochlorite — Sodium hypochlorite or liquid bleach,
manufactured by many companies in the United States, usually
contains 12 to 15 percent available hypochlorite at the time
of manufacture. It is available only in liquid form. For
combined sewer overflow and storm sewer discharge applica-
tions, it is produced on-site either by reacting chlorine
337
-------
Table 76. CHEMICAL DISINFECTION AGENTS AND SOURCES USED
BY VARIOUS CITIES FOR COMBINED SEWER OVERFLOWS
AND STORM SEWER DISCHARGES
Location and site
Chemical agent
Source
Cambridge, Massachusetts,
Cottage Farm Storm
Detention and Chlorination
Facility
Dallas, Texas,
Bachman Storm Water Plant
Kenosha, Wisconsin,
contact stabilization
Milwaukee, Wisconsin,
Humbolt Avenue Pollution
Abatement Plant
Mount Clemens, Michigan,
Combined Sewer Overflow
Collection and Treatment
Facility
New Orleans, Louisiana,
4 different storm sewer
discharge sites
New Providence, New Jersey,
Sanitary and Storm Water
Pollution Control Plant
New York City, New York,
Spring Creek Auxiliary
Pollution Control Plant
Philadelphia, Pennsylvania,
Callowhill Microstrainer
Facility
San Francisco, California,
Baker Street Dissolved
Air Flotation Facility
Syracuse, New York
Sodium hypochlorite Purchased. On-site generation from
brine (experimental).
Chlorine
Chlorine
Chlorine
Purchased.
Purchased.
Purchased.
Chlorine dioxide On-site generation from sodium
chlorite and chlorine.
Sodium hypochlorite Central generation from sodium
hydroxide and chlorine.
Transported to and stored at
application sites.
Chlorine
Purchased.
Sodium hypochlorite Purchased.
Sodium hypochlorite Purchased.
Ozone On-site generation (experimental)
Sodium hypochlorite Purchased.
Chlorine dioxide
Chlorine
On-site nitrosyl chloride
generation system.
Purchased.
with sodium hydroxide or by electrolysis from brine. Because
sodium hypochlorite solutions are to some degree unstable,
storage temperatures should not exceed 29 degrees C (85
degrees F) and the maximum shelf life is about 90 days. The
stability of various sodium hypochlorite solutions is shown
on Figure 69. As a result of the decay in strength indi-
cated, it is necessary to adjust the hypochlorite feed rate
at the time of application to maintain the desired concen-
tration in the combined sewer overflow or storm sewer
discharge.
338
-------
160
140 -
120
100
40 60 80 100 120 . 140
DAYS STORAGE AT NORMAL ROOM TEMPERATURE
160
180
Figure 69,, Stabi I i ty of
sodium hypochlorite solution [13]
Hypochlorite is attractive because it is safer to handle
than chlorine gas. It can be generated between storms and
stored until it is used. This reduces the capital costs by
allowing the use of a small generating plant that can oper-
ate almost continuously, thus avoiding the problems of
frequent shutdown and startup and having the plant idle for
long periods. An example of a sodium hydroxide-chlorine
gas hypochlorite generating plant for storm sewer discharge
disinfection is shown on Figure 70. A recently developed
electrolytic hypochlorite generator is shown on Figure 71.
Chlorine Dioxide — Chlorine dioxide is being used currently
to disinfect storm flows at Mount Clemens, Michigan, and
Syracuse, New York. It has a distinct advantage over chlo-
rine for treating water having a high ammonia content be-
cause it reacts readily with oxidizable materials without
339
-------
JHMPI P" mi
(a)
(c)
Figure 70. Hypochlorite generating plant
for stormwater disinfection, New Orleans
(a) Detail of solution pumps and transfer piping with sodium hydroxide storage tanks
in background (b) Overview of sodium hydroxide-chlorine gas hypochlorite generating
plant. Chlorine gas is delivered in tank car quantities on the railroad siding in
foreground (c) View of reaction chamber with manufactured hypochlorite storage
tanks in background (d) Typical hypochlorite holding tanks at application site
340
-------
Figure 71. Electrolytic hypochlorite generator
Small 12-cell electrolytic generator capable of producing 1,515 to 1,890 liters
per day (400 to 500 gpd) of 7-percent sodium hypochlorite solution
341
-------
combining with ammonia to form chloramines. Chloramines
have been reported to be toxic to some forms of aquatic life
at concentrations as low as 10 yg/1 [15].
Ozonation
Ozone is usually produqed by passing high-voltage elec-
tricity through dry atmospheric air or oxygen between sta-
tionary electrodes. A small percentage of the oxygen is
converted into ozone in the process. Ozone not only has a
more rapid disinfecting rate than chlorine but also the
further advantage of supplying additional oxygen to the
wastewater. The increased disinfecting rate for ozone re-
quires shorter contact times and thus effects a reduction
in capital costs compared to those for a chlorine contact
tank. Ozone does not produce chloramines or a long-lasting
residual as chlorine does, but it is unstable and must be
generated on-site just prior to application. Thus, unlike
chlorine, no storage is required. In recent tests on com-
bined sewer overflows in Philadelphia, equivalent disinfec-
tion was obtained using either 3.8 mg/1 of ozone or 5 mg/1
of chlorine [7].
DEMONSTRATION PROJECTS
New Orleans, Louisiana [5]
Storm sewer discharges being pumped from the east bank of
New Orleans into Lake Pontchartrain, as shown by bacterio-
logical testing during 1961 to 1966, contained excessive
coliform bacteria concentrations. It was felt that disin-
fection on a large scale could restore the quality of the
water to a level adequate for body contact sports. To accom-
plish this, the EPA sponsored a demonstration project in
which disinfectant was added to the storm sewer discharges
pumped by four drainage pumping stations located on three
outfall canals on the east bank of New Orleans. The four
pumping stations had a combined capacity of 313 cu m/sec
(7,140 mgd) and each normally pumps in excess of 566,000
cu m (20,000,000 cu ft) per day of stormwater on rainy days.
The project included the design and construction of a sodium
hypochlorite manufacturing plant to prepare the disinfectant.
The manufacturing plant was located at the water purifica-
tion plant of the Sewerage and Water Board of New Orleans
because of the availability of personnel experienced in
handling large quantities of chlorine. The disinfectant was
manufactured at a continuous, automatically controlled plant
with a capacity of 3,785 1/hr (1,000 gal./hr) of 120 g/1
sodium hypochlorite. Also included at the plant was storage
342
-------
for 151,400 1 (40,000 gal.) of finished sodium hypochlorite.
Photographs of the plant and storage tanks are shown on
Figure 70. Disinfectant prepared at the manufacturing plant
was stored at the feeding facilities located adjacent to
each of the pumping stations. The disinfectant was trans-
ported to the feeding facilities in two 11,400 1 (3,000 gal.)
tank trucks.
Sodium hypochlorite was manufactured by reacting 50-percent
sodium hydroxide, water, and chlorine under atmospheric
conditions. The average manufacturing cost for the hypo-
chlorite was $0.05/kg ($0.12/lb) of available chlorine.
Total cost of the manufacturing facilities was $581,700
(ENR 2000) .
Sodium hypochlorite was added to storm sewer discharges dur-
ing 16 high-volume storms and more than 20 low-volume storms.
During the 16 high-volume storms, 3,936,400 cu m
(1,040,000,000 gal.) of stormwater were treated with more
than 132,475 1 (35,000 gal.) of sodium hypochlorite. The
largest single treatment episode was 257,400 cu m
(68,000,000 gal.) of stormwater with 30,660 1 (8,100 gal.)
of hypochlorite. Based upon results from the 16 high-volume
storms, chlorine residuals greater than 0.5 mg/1 resulted in
99.99-percent or greater reduction of coliform
concentrations. However, upon cessation of disinfection,
coliform levels in the outfall canals recovered within 24 to
30 hours. Total coliform concentrations recovered to those
levels normally found in the outfall canal. Fecal coliform
recovery levels were approximately two orders of magnitude
less than normal endogenous levels. This rapid recovery
changes the significance of the coliform levels. Their use
as indicators of possible pathogenicity of the stormwater is
obscured once disinfection has occurred.
Ionics Hypochlorite Generator [6]
An advanced electrolytic generator has been developed for
on-site production of sodium hypochlorite from sodium chlo-
ride (salt) brine. In this system, an electrochemical cell
electrolyzes high-purity sodium chloride brine (saturated at
approximately 275,000 mg/1) to chlorine gas and sodium hy-
droxide solution, which are then reacted immediately outside
the cell to produce a 5- to 10-percent sodium hypochlorite
solution. The process requires 1.6 kwh of electricity and
0.95 kg (2.1 lb) of salt per 0.45 kg (1 Ib) of sodium hypo-
chlorite generated. The first field unit, shown on
Figure 71, is scheduled for installation in the Somerville
Marginal Conduit Pretreatment Facility, Somerville,
Massachusetts.
343
-------
™i J? is.beun? constructed to divert combined sewer
overflow which is being dumped into the Mystic River basin
into a tidal portion of the river. All overflow is to be
screened and disinfected with hypochlorite . The conduit
t^« fCt +t ^ detention chamber with a 9.5-minute contact
time for the 5-year storm and progressively longer times for
less severe loads The Somerville interception station has
been designed on the basis of 240 hours per year of combined
3T40 ?/~fl??n ^ ^^P £1°W °f thi* overflow wiH be
3,940 I/sec (90 mgd) and the desired dosage level will be
4.3 mg/1 of sodium hypochlorite. The annual consumption of
hypochlorite will be 14,620 kg (32,200 Ib) of available
44 4SOna*l°VPSr?Ximatel? i34>6°° to 168,250 1 (35,560 to
44,450 gal.) of 7-percent hypochlorite solution.
Q teSt Unit was desiSne » the Astern will
run 2,400 hours per year or about 27 percent of the time
This results in a cost of $0.200/kg ($0.091/lb) of available
chlorine. At that location, the quoted price for delivered
hypochlorite is $0.385/kg ($0.175/lb) of available chorine.
Delays in construction of the Somerville Marginal Conduit
made initial testing at an alternate site necessary. The
.
a^er£?te-Si^ sejected was ^e Cottage Farm Storm Detention
and Chlonnation Facility in Cambridge, Massachusetts.
Cottage Farm Storm Detention and Chlorination Facility
Boston, Massachusetts ~~~ - ~~ - — — - *-*-
The storage/sedimentation aspects of this facility have been
discussed previously in Section IX, Storage. Its other
major function is to disinfect all combined sewer overflows
to prevent degradation of the water quality of the Charles
River Basin This facility was designed to use truck-
delivered sodium hypochlorite solution (10- to 15-percent
concentration) The concentration of the solution decays in
SrA°9a£e* /he hypochlorite solution is stored in two 15,900-1
(4,200-gal ) fiber glass tanks. The estimated annual demand
at this site is 22,700 kg (50,000 Ib) of available chlorine?
Average chlorine demand of the combined sewage varies from
1.3 to 4.5 mg/1 .
The small 12 -cell Ionics hypochlorite generator was operated
at the facility on a test basis in October to December 1972
and part of 1973. High-purity brine was found to be neces-
sary for generation of the hypochlorite. The maximum concen-
tration produced was 7-percent hypochlorite at 1 515 to
1,890 I/day (400 to SOOgpd) . This, p?us decay! 'routinely
344
-------
meant a feed concentration of approximately 4 percent, which
required more solution than the hypochlorite tanks could
store. Additional operating problems plagued the tests.
The facility has returned to operating with truck-delivered
hypochlorite.
Qnondaga County, Syracuse, New York [121
As a result of bench-scale studies investigating high-rate
screening and high-rate disinfection, a demonstration com-
bined sewer overflow treatment facility is being constructed
in Syracuse, New York. The project is sponsored by EPA.
Three screening processes will be tested, each with a maxi-
mum capacity of 220 I/sec (5 mgd). The screening equipment
will include a microstrainer with a mesh opening of 23
microns, a drum screen with a mesh opening of 71 microns,
and a rotary fine screen with a mesh opening of 105 microns.
Following each screening unit will be individual contact
tanks so that several methods of disinfection can be tested
also.
Disinfecting agents to be used are chlorine and chlorine
dioxide. Chlorine will be provided from 1-ton chlorine gas
cylinders. Chlorine dioxide will be generated by means of
a Nitrosyl Chloride generation system at the demonstration
site. The generation facilities and the feed systems were
designed to provide disinfection with a single agent or a
combination of the two agents at various quantities of
either. The point of application of the disinfecting agent
is variable also. The contact tanks were designed to accom-
modate flash mixing at the point of injection or flash
mixing at several locations throughout the tank. Corrugated
baffles will be added to one of the contact tanks to test
the effects of continuous high turbulence on disinfection.
A concentrated effort will be made to determine the effects
of two-stage disinfection.
From bench-scale tests, it was determined that for a 12-hour
overflow of 219 I/sec (5 mgd) (the maximum pumping capacity
for each treatment unit), 227 kg (500 Ib) of chlorine would
be required at a dosage level of 25 mg/1 to disinfect the
entire overflow with a contact time of 60 seconds. The chlo
rine dioxide generator was sized on the basis of a 10-mg/l
dosage level and a contact time of 60 seconds. In both
cases, the target total coliform concentration level in the
effluent is 1,000/100 ml.
It was reported also that effluent total coliform bacteria
requirements of 1,000/100 ml and fecal coliform require-
ments of 200/100 ml can be reached by using 25 mg/1 of
345
-------
chlorine or 12 mg/1 of chlorine dioxide in a single-stage
disinfection process with a contact time of 60 seconds.
Equivalent disinfection was accomplished in the same length
of time using 8 mg/1 of chlorine, followed 30 seconds later
by 2 mg/1 of chlorine dioxide. It is hoped that the results
of bench-scale tests will be verified by the prototype
facility when operation begins.
Philadelphia, Pennsylvania [7, 8]
This work consisted of the design, installation, operation,
and evaluation of a commercial microstrainer, and of ozona-
tion and chlorination at a typical combined sewer overflow.
The results of the microstrainer evaluation were discussed
previously in Section X. Also included in this project was
the corrugated baffle chlorine contact tank mentioned previ-
ously in this section.
Chlorine contact times of only 2 minutes, under relatively
high turbulence conditions with chlorine dosages as low as
5 mg/1, reduced total coliform concentrations from 1,000,000/
100 ml in the combined sewer overflow down to 5 to 10/100 ml.
Fecal coliform concentrations were similarly reduced, under
the same conditions, from 100,000/100 ml down to 5 to 10/
100 ml. These results were obtained in a specially designed
contact tank using parallel corrugated baffles to provide
high mixing intensities.
It was reported also that the use of ozone for disinfection
of treated combined sewer overflows is feasible [7]. Ozone
was generated by a commercially available ozonator with air
as the input for application to microstrainer effluent.
Figure 72 shows the ozone generation and its arrangement
with the microstrainer unit. Ozone concentrations of 3.8
mg/1 were required to match disinfection efficiencies of
5 mg/1 of chlorine.
The capital cost of ozone generation facilities, when oxygen
is used for the production, was considered to be lower than
that for hypochlorite facilities.
COSTS
Cost data on chlorine gas and hypochlorite disinfection are
presented in Table 77. Disinfectant costs for combined
sewer overflow and storm sewer discharge treatment are
higher than those for sewage treatment, as indicated in the
table. This is the result of smaller total annual disin-
fectant volume requirements, increased disinfectant concen-
tration requirements, and higher unit operation and mainte-
nance costs for combined sewer overflow treatment facilities.
346
-------
OZONIZED AIR
DRY AIR
Ut
AIR
SUPPLY
OZONATOR DESSICATOR
COMBINED SEWER
OVERFLOW
TREATED
EFFLUENT
Figure 72. Schematic of ozone generation
and injection for microstrainer facility
These costs could be reduced by using the facilities in
conjunction with dry-weather flow treatment plants, whenever
possible.
Generally speaking, chlorine gas has the lowest production
and operation costs for large, continuous operation
requirements. Disinfection using commercially purchased
hypochlorite has low initial capital costs, but supply costs
depend on the location, source, and quantity. Capital costs
for on-site electrolytic hypochlorite generation are high,
while operation costs are dependent upon the solution used
(sea water or brine) for generation. The operation costs
are lower for sea water, but technical difficulties associ-
ated with the use of sea water have been reported.
The operating costs for tertiary treatment of secondary
treatment plant effluent using ozone are listed in Table 78.
These unit costs would increase considerably when applied to
347
-------
Table 77. COST DATA ON CHLORINE GAS AND
HYPOCHLORITE DISINFECTION [13]a
Location, agent,
and source
Capital cost.
Operating
$ cost, $/yr
Cost/lb
available
chlorine, $
Akron, Ohio*3
Sodium hypochlorite
Purchased
Cambridge, Massachusetts
and Somerville,
Massachusetts^
Sodium hypochlorite
Purchased
On-site generation
New Orleans, Louisiana
Sodium hypochlorite
On-site generation
Saginaw, Michigan
Chlorine gas
Sodium hypochlorite
Purchased
On-site generation
South Essex Sewerage
District, Massachusetts6
Chlorine gas
Sodium hypochlorite
Purchased
On-site generation
441,500
23,300
581,700
161,000
19,550
95,450-161,000
872,460
421,800
290,000
2,300
6,325-11,500
4,715-5,175
233,100
364,080
0.152-0.264
0.385
0.200
0.120
0.35
0.18-0.31
0.28-0.40
0.035
0.046
Sea water
Brine
1,665,000
1,665,000
160,950
303,030
0.035
0.051
a. ENR = 2000.
b. Combined sewer overflow disinfection.
c. Storm sewer discharge disinfection.
d. Combined sewer overflow disinfection at use rate of 42,000 Ib/yr
of chlorine.
e. Sewage treatment plant effluent disinfection at use rate of
24,000 Ib/day of chlorine.
Note: $/lb x 2.2 = $/kg
348
-------
Table 78. ESTIMATED COSTS OF TERTIARY TREATMENT
PLANTS USING OZONE [9]a
Plant capacity, Plant capacity, Plant capacity,
Item 1 mgd 10 mgd 100 mgd
Capital cost, $
Operating cost,
$/mil gal.
Operating cost,
$/l,000 gal.
254,520
172.63
0.173
1,360,800
96.65
0.097
9,399,600
61.84
0.062
a. ENR = 20QO.
b. Oxygen recycle system would be used.
Note: mgd x 43.808 = I/sec
$/mil gal. x 0.264 = $/Ml
$/l,000 gal. x 0.264 = $/l,000 1
combined sewer overflow disinfection because the flow would
not be continuous for both. Additional ozone generation
capacity would be required to handle the wet-weather flows.
However, the dual use of such facilities for both dry- and
wet-weather operation would reduce the cost slightly.
The capital costs for different disinfection agents and
methods resulting from a study conducted at Philadelphia are
shown in Table 79. The capital costs for ozone generation
are usually the highest of the most commonly used processes.
Ozone operation costs are very dependent on the cost of
electricity and the source of the ozone (air or pure oxygen)
349
-------
Table 79. COMPARISON OF ESTIMATED CAPITAL COSTS
FOR 3 DIFFERENT DISINFECTION METHODS [7]a
Capital cost,
Disinfection method $/mgd
2-Minute ozone contact
(chamber with once-through
oxygen-fed ozone generator)b 13,013
2-Minute chlorine contact
(chamber with hypochlorite
feeder)c 1,521
5-10 Minute conventional
chlorine contact^ 1,690
a. ENR = 2000.
b. Unit cost of ozone at $5.20/lb from
oxygen <§ $0.19/lb; dosage of 3.8 ppm;
Otto plate type generator.
c. Unit cost of hypochlorite at $0.42/lb
available chlorine; dosage of 15 ppm.
d. Unit cost at $0.42/lb available chlorine;
dosage of 5 ppm.
Note: $/mgd x 0.0228 = $/l/sec
$/lb x 2.2 = $/kg
350
-------
Part IV
IMPLEMENTATION
-------
-------
Section XIV
INTEGRATED SYSTEMS
The treatment and abatement methods described in Part III,
Management Alternatives and Technology, have been discussed
essentially as unique and singular solutions. In fact, many
of these methods can and should be combined to optimize the
effectiveness of any overall abatement program by maximizing
the pollution reduction, enhancing the aesthetic and reuse
potential, and minimizing the cost of the program. For ex-
ample, reducing the source contaminants, solids and debris
in stormwater runoff (source control); using regulators for
maximizing storage capacity in sewers (system control and
storage) and for coarse quantity/quality separation (pre-
liminary treatment); arid providing objective level treatment
for the storm flow prior to discharge (treatment) could sub-
stantially reduce the pollution load on the receiving water.
Thus, the purpose in this section is to deal with integrated
systems for storm sewer discharge and combined sewer over-
flow treatment and abatement.
The interfacing of storm flow pollution abatement facilities
with existing or proposed sewage collection systems and dry-
weather flow treatment facilities is discussed first. Next,
the availability and use of mathematical models for
"planning," "predictive," and "decision-making" purposes
is described and several examples of sewerage master plans
using integrated approaches are presented. Finally, the im-
portance of integrating the required facilities aesthetically
into each local environment is emphasized.
INTERFACING WITH DRY-WEATHER FACILITIES
The principal purpose of the storm flow pollution abatement
device or process is to reduce the contaminants in the flows
discharged to receiving waters. Two methods available are:
(1) source control to remove the contaminants before they
are picked up by the stormwater, and (2) treatment of storm
and combined sewer flows after contamination to remove pollu-
tants before discharge. In this latter method, existing
353
-------
dry-weather facilities or newly constructed facilities can be
used. Storage devices are used to reduce the rate of flow,
which enables the reduction in size of the required treatment
facilities and leads to their more effective utilization.
This discussion describes how each abatement device or pro-
cess can fit into an existing system and how the devices
interrelate with each other. The interfacing of the storm
flow pollution abatement scheme with the existing sewerage
system is important in the development of a total integrated
waste flow system.
First, it is necessary to understand where individual units
can be used in relation to a typical sewer network and sewage
treatment plant. Basically, a storm flow treatment device
may be located (1) at overflow points or possibly at key up-
stream locations, (2) at one or more central locations away
from dry-weather facilities, or (3) adjacent to municipal
sewage treatment plants. A fourth possibility exists when
storm flows are treated at dry-weather plants after being
stored. The first two possibilities are essentially the same
except for size and number of stormwater treatment devices
used. The first consists of many locations with small facili
ties; the second consists of one or two larger facilities at
strategic points. Whenever several units are needed, it is
usually economical to use the same type of device, equip-
ment, and design to reduce operation and maintenance costs.
Larger centralized facilities generally are associated with
storage tunnels or other large storage devices because of
the costs involved in transmitting large quantities of flow
and providing peak flow rate capacities. Representative
means of interfacing with existing sewerage systems are
shown schematically on Figure 73. Although all facilities
are shown downstream of the collection systems for clarity,
upstream locations may offer benefits in particular appli-
cations as determined by limitations within the existing
system.
As previously noted, the physical, biological, and physical-
chemical processes used on storm flows each have limitations
as to where they can be used. Biological treatment devices
should be located at sewage treatment plants to provide a
continuous active biomass. Physical and physical-chemical
treatment devices may be placed at satellite locations.
Although a complete physical-chemical treatment system could
be placed at overflow points, its size and complexity gener-
ally limit it to a central location or next to dry-weather
facilities. Physical treatment devices, with or without
chemical additions, lend themselves more easily to remote
locations at overflow points.
354
-------
COMBINED OR SEPARA
SANITARY SEWER
(TYPICAL)
/
TE
\ 1 IN-LINE
{STORAGE
(A1TERNATI)
STORM SEWER'"
.DIVERSION STRUCTURE
/ SWIRL CONCENTRATOR X
(TYPICAL) /
fc.f^ DiF W STP
Y
1 WET WEATHER FLOW (WWF)
OFF-LINE
STORAGE
^-SEWAGE TREATMENT [
PLANT (STP) \
1
~~ \
* 1
\
1
STORM
W TREATMENT " \
(OPTIONAL)
(a) SATELLITE AND CENTRALIZED FACILITIES
/
/
V
["IN-LINE
' STORAGE
•(ALTERNATE)
STORM SEWER"
DWF
»O
T
IWWF
OFF-LINE
STORAGE
OWF PA
TREATM
r
1 BY
t
..^
TH FOR POLISHIN8 OR TERTIARY
ENT BY STORM FACILITIES
STP
PASS
BOUNDARIES
y
\
\
j
/
STORM
TREATMENT
COLLECTION SYSTEM
(b) FACILITIES IN CONJUNCTION WITH SEWAGE TREATMENT PLANT
WWF. TEMPORARY DWF DIVERSIONS
STORM SEWER
Figure 73. Methods of Interfacing stormwater
facilities with existing systems
355
-------
Some storm flow treatment facilities offer the added advan-
tage of acting as effluent polishers or tertiary treatment
when operated in conjunction with a secondary sewage treat-
ment plant. In addition, such facilities can become dual
purpose by improving the effluent quality of the dry-weather
flow during non-storm conditions and treating storm flows
during wet weather. Storage and transmission (conduit)
facilities may and should also provide dual-purpose relief
or standby functions. This total system approach, of course,
has the added benefit of reducing the overall cost of waste-
water facilities construction and operation.
For example, storm flow facilities used in conjunction with
sewage treatment plants, particularly the biological treat-
ment devices, provide benefits by increasing dry-weather
capacity, possibly postponing planned expansions, and pro-
viding versatility in handling breakdowns, maintenance, and
other factors because of the ability to treat dry-weather
flow in emergencies. Two advantages of interfacing storm
flow treatment facilities with existing sewerage systems
are that (1) solids removed from the storm flow can be dis-
charged to the interceptor, and (2) most stormwater treat-
ment processes can accommodate higher flow rate variations
or provide flow equalization and control. A summary of
potential interfacing actions is presented in Table 80.
INTERFACING UNIT PROCESSES
The interrelationships between storm flow devices and unit
processes is second in importance only to interfacing with
the dry-weather facilities. As indicated previously, stor-
age, either in-line or off-line, is used to remove the peaks
and valleys in the storm hydrograph. As such, its use with
treatment devices is set. The relationships between the
different treatment devices, on the other hand, are not so
well defined. Classifying the units into main and comple-
mentary treatment units is an advantage, as it delineates
the main blocks from which to build an abatement scheme.
These are listed in Table 80.
Main treatment units (i.e., primary, secondary, and terti-
ary treatment) are considered as devices that remove a
significant portion of the SS and/or dissolved organics.
The remainder are pretreatment units. They can be used to
assist the main process by protecting against damage (bar
screens), or reducing coarse solids loading on the main
process (fine screens ahead of dual-media filters, dissolved
air flotation units, etc.). These interrelationships are
shown in Table 81. The reader is referred to the discus-
sions of the individual processes for details on their abil-
ity to interface with other processes. Of noteworthy
356
-------
importance, however, is the ability of filters and micro-
strainers to polish sewage treatment plant effluent and of
physical-chemical processes to act as a tertiary treatment
plant. They therefore can act in a dual capacity.
Sedimentation is easily incorporated into a storage facility,
thus reducing the total cost of clarification and storage.
One process not mentioned yet in this discussion is
disinfection. It is a process that will be used at almost
all storm flow treatment facilities to reduce bacterial and
virus counts. In some cases, disinfection has been applied
within storage/sedimentation or other tankage to eliminate
the need for a separate contact tank; in others, a contact
tank is required.
The objective of interfacing is to use processes and/or
equipment that are compatible with other processes and/or
equipment and also with the existing system. At the same
time, the devised storm flow facility must meet the pollu-
tion removal criteria necessary to protect the receiving
waters. The objective pollution abatement system is one
that makes a total wastewater system for the least cost.
The planning fundamentals were outlined in Section IV. A
key planning tool, perhaps the most important aid in storm
flow evaluations, is mathematical modeling. The available
programs and application techniques are discussed next.
MATHEMATICAL MODELING TECHNIQUES
The complexity of modern sewerage systems, whether combined
or separate sanitary and storm, makes it extremely difficult
to evaluate the impact of additions, proposed modifications,
or new systems rapidly and accurately. The time and man-
power required to make such an evaluation by traditional
methods can be tremendous. The application of computerized
mathematical simulation modeling techniques to evaluate
additions or changes to drainage areas and sewerage systems,
to interpret storm phenomena, and to assess receiving water
response, not only greatly reduces this time and manpower
requirement but also allows the comparison of alternative
courses of action. The use of computerized modeling tech-
niques facilitates the necessary regional planning and di-
rection of pollution abatement.
Under a concurrent EPA contract, existing mathematical
models for the engineering assessment, control, planning,
and design of storm and combined sewerage systems are
being evaluated [1]. The purpose of the study is to eval-
uate comprehensive models and to provide guidelines for the
357
-------
Table 80. INTERFACING STORM FLOW FACILITIES
WITH EXISTING SEWERAGE SYSTEM
Storm flow facility
STORAGE
In-line
Off-line
Basins
Tanks
Underground
silos
Underwater
Deep tunnel
Mined
labyrinths
TREATMENT
Sedimentation
Dissolved air
flotation
Bar screens
Rotary fine
screens
Fine screens
Microstrainers
Filtration
Swirl
concentrator
Contact
stabilization
Trickling filters
Rotating biologi-
cal contactors
Treatment lagoons
Physical -chemical
treatment
Disinfection
Type of unit
--
--
--
--
--
--
--
..
Primary
Primary
Pretreatment
Pretreatment
Pretreatment
Primary
Primary
Pretreatment
Primary i
Secondary
Secondary
Secondary
Secondary
Tertiary
Post
treatment
Primary purpose
Flow attenuation
Flow attenuation
Flow attenuation
Flow attenuation
Flow attenuation
Flow attenuation
Flow attenuation
Flow attenuation
Removes SS
Removes SS
Protects down-
stream equipment
Roughing filter
Roughing filter
Removes SS
Removes SS
Quantity and
quality regulator
Removes SS and
dissolved organics
Removes SS and
dissolved organics
Removes SS and
dissolved organics
Removes SS and
dissolved organics
Removes suspended
and dissolved
material
Reduces bacterial
contamination
Facility is
dependent
on DWF
plants
a
a
a
a
a
a
a
a
No
No
No
No
No
No
No
No
Yes
Yes
Yes
Not
necessarily
No
No
Facility is
good for
satellite
locations
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
Yes
Yes
Yes
Facility
increases
DWF plant
versatility
Yes
Can
Can
Can
Can
Can
Can
Can
No
No
No
No
No
Yes
Yes
Can
Yes
Yes
Yes
Can
Yes
Can
358
-------
Table 80. (Continued)
Storm flow facility
STORAGE
In- line
Generally
acceptable
max flow
variation
(times base
flow) Advantages Limitations
>10 Uses existing facilities. Limited to excess sewer
Off-line
Basins
Tanks
Underground
silos
Underwater
Deep tunnel
Mined
labyrinths
TREATMENT
Sedimentation
Dissolved air
flotation
Bar screens
2-4"
2-4c
capacity.
Location versatility. Usually expensive.
Can be combined with
sedimentation.
Inexpensive.
Minimum land requirements.
Large land requirements.
Large land requirements.
Minimum land requirements. Solids removal problems.
Minimum land requirements; Expensive.
flow transmission.
Minimum land requirements. Expensive.
Can be combined with
storage.
Good for satellite
locations.
Rugged.
Low removals .
Somewhat complicated
equipment .
1-2
5-10
5-10
2-4b
Rotary fine
screens
Fine screens
Microstrainers
Filtration
Swirl >10
concentrator
Contact 1-2
stabilization
Trickling filters 5-10
Rotating biologi- 5-10
cal contactors
Treatment lagoons >10
Physical-chemical 2-4
treatment
Low flow variation.
Versatile; low land
requirements.
Versatile; low land
requirements.
Good SS removal.
Solids separation.
Easily combined with
existing activated
sludge plants.
Easily combined with
existing trickling
filter plants.
Easily combined with
existing rotating
biological contactors.
Can be used in con-
junction with
recreation facilities.
Produces a reusable
effluent.
Must be combined with
DWF plant.
Must be combined with
DWF plant.
Must be combined with
DWF plant.
Large land requirements.
High sludge (dry weight)
volume.
Disinfection
Protects public health.
Expensive.
a. Generally yes for dewatering and solids disposal. May also apply to primary and
pretreatment devices.
b. For short periods of time.
c. Can be made to handle higher variations by using multiple units in parallel.
359
-------
Table 81. GENERAL INTERFACING BETWEEN TYPES OF
STORAGE AND TREATMENT DEVICES3
Proposed Complementary/Supplementary Process
V
00 «
I-1 O O
O *-» -H
« M
CJ +J
CJ c C
C -H u
Proposed Process ,5 o £ c
c
D
u
I/I
in (U C
ecu
4) -rt OJ
U X U)
n
r-l
•H
00
C
•H
O
'^
C
o
e
I O
00
* .
,_,
c
^* j
rt
s
HI
\.
s
c
o c
•H O
CO '»J
.-i a
3 h
00 <-J
O .H
», u u.
c
o
'l-t
D.
o
•a
«
c
Carbo
M
B
a
Ammon
c
o
tJ
c
•H
In-line storage
Off-line storage
Sedimentation
Dissolved air flotation
Bar screens
Rotary fine screens
Fine screens
Filtration
Swirl concentrators
Contact stabilization*"
Trickling filters0
Rotating biological contactors0
Treatment lagoons
Physical-chemical treatment
Coagulation w/sedimentation
Filtration
Carbon adsorption
Ammonia removal
\
O1
9
9
•
0
«
9
•
•
•
9
•
9
s «^
3 \^
aH
9
•
9
9
9
•
•
•
9
•
9
O
U
N.
9
•
9
9
9
9
U
O
\
U
O
9
•
•
\
•
•
•
•
•
•
•
•
•
•
O
o
9
9
\
•
9
9
9
9
9
9
O
0
9
9
\
•
9
9
9
9
9
O
O
\
O
O
0
9
O
O
O
O
O
O
O
\
0
O
O
O
9
9
9
9
9
9
\
9
9
9
9
9
9
O
O
0
O
O
0
\
O
O
O
0
O
O
\
O
O
O
O
o
o
\
0
o
0
o
o
o
o
o
o
\
0
o
o
o
o
o
o
o
o
o
\
\
9
9
9e
O
\
9
-------
practicing engineer to aid him in selecting the model most
suited for his requirement. In light of this, only a gen-
eral overview of the available models is included herein.
Available Models
Several models are presently available, each having a dif-
ferent level of sophistication. These range all the way
from models that produce only the urban runoff hydrograph
without the associated runoff quality to models that pro-
duce the runoff quantity and quality, route both through
the sewer system along with the dry-weather flow, simulate
the effects of various treatment and control facilities,
and route the resulting quantity and quality through the
receiving waters.
There are a number of mathematical models that carry out
routine design functions or that simulate portions of com-
plex urban wastewater management systems (such as rainfall-
runoff computations and limited sewer flow routing) or the
operation of a single storage reservoir or treatment plant,
but only a few models have been developed that consider the
entire sewerage system. Six of the more useful models are
discussed briefly below.
Hydrograph-Volume Method - The Hydrograph-Volume Method
(Hitter, 1971) was developed in Germany [18] . This model
calculates the dry-weather flow and storm runoff, and routes
the combined flows through a complex sewerage network. Its
benefits appear to be in conduit sizing and design. The
model does not simulate flow quality or control regulators.
Cost computations appear to be external.
Road Research Laboratory Hydrograph Method - The British
Road Research Laboratory (RRL) Method uses storm rainfall
to provide a stormwater runoff hydrograph for the purpose of
storm drainage design [23] . Rainfall is applied to the
paved area of the drainage basin which is directly connected
to the storm drainage system. Travel time to the nearest
storm drainage inlet is computed for various increments of
the total paved area that is directly connected. From this
time-area information, the surface hydrograph arriving at
the inlet is computed. The surface hydrograph is then
routed through the storage available in a particular sec-
tion of pipe. The surface hydrograph at the next downstream
inlet is added, and the combined hydrograph is routed on
downstream. Thus, the successive addition and routing of
surface hydrographs produces an outflow hydrograph at the
downstream discharge point. In the RRL method, the quality
associated with the runoff is not computed. The applica-
tion and use of the method is described in a recent EPA
report [16].
361
-------
Stanford Watershed Model (Hydrocomp Simulation Program! -
The Stanford Watershed Model (Crawford and Linsley, 1966 [6])
along with its commercial successor, the Hydrocomp Simulation
Program, is a comprehensive mathematical model that simu-
lates watershed hydrology and flow routing. This model has
been used extensively to simulate existing and planned sur-
face water systems. Recently, it has been expanded to
include water quality computations. It does not perform
cost calculations, however, and has not been adapted to
sewerage systems.
Urban Runoff Model - The Urban Runoff Model PDF-9 (UROM-9)
was developed at the University of Minnesota for the Metro-
politan Sewer Board, St. Paul, Minnesota [7]. The purpose
of this model is to predict discharges in the Minneapolis-
St. Paul interceptor sewers, given rainfall readings at
various points around the Twin Cities. The model computes
storm runoff from major catchments, combines the runoff
with estimates of dry-weather flow, routes the combined
flows through the interceptor system to the treatment plant,
and computes overflows to the receiving water at control
regulators. It uses monitored rainfall and flow level data
from various points for real-time control of the overflows.
The model has not been adapted to consider water quality
aspects of the overflows. It is not intended for use for
design purposes.
Urban Wastewater Management Model - The Urban Wastewater
Management Model (Battelle-Northwest and Watermation, Inc.,
1972) is a comprehensive mathematical model developed to
continuously simulate time-varying wastewater flows and
qualities in complex metropolitan combined sewerage systems
[2, 17]. The model simulates major sewer system components,
such as trunk and interceptor sewers, regulators, storage
facilities, and treatment plants. It provides a means
of evaluating the time-varying performance of a planned or
existing sewerage system under a variety of rainfall condi-
tions (considering both time and spatial rainfall variations)
without simulating every pipe or manhole. The model simu-
lates seven wastewater quality parameters: SS, BODs, COD,
phosphate, nitrate, ammonia, and Kjeldahl nitrogen. The re-
quired operation of control regulators during real-time
rainstorm events to minimize overflows is modeled. The
model can also be used for design and planning-studies. It
computes sizes and costs of structural sewer system modifi-
cations, such as sewers, regulators, and storage and treat-
ment facilities, that will result in the least-cost combina-
tion of alternatives for improving system performance.
Storm Water Management Model - The Storm Water Management
Model (SWMM) (Metcalf § Eddy, Inc., University of Florida,
362
-------
and Water Resources Engineers, 1971) was developed under the
sponsorship of the EPA [19, 20, 21, 22]. It is a comprehen-
sive mathematical model capable of representing urban storm-
water runoff, storm sewer discharge, and combined sewer
overflow phenomena. The SWMM has been demonstrated at more
than 20 sites throughout the country ranging from 76 to
8,100 ha (187 to 20,000 acres). During demonstration, the
SWMM has been verified to be capable of representing the
gamut of urban stormwater runoff phenomena for various
catchment systems [20] . This includes both quantity and
quality from the onset of precipitation on the basin,
through collection, conveyance, storage, and treatment sys-
tems , to points downstream from outfalls that are signifi-
cantly affected by storm discharges. The SWMM is intended
for use by municipalities, governmental agencies, and con-
sultants as a tool for evaluating the pollution potential of
existing systems, present and future, and for comparing
alternative courses of remedial action. The use of correc-
tional devices in the catchment, along with evaluation of
their cost effectiveness, has also been demonstrated.
Application of Mathematical Models
The first step in model selection and application should be
verification using local data. To this end, it is necessary
to have data from one or more actual storm(s) for comparison
with the computed values. For even a simple program that
computes only the stormwater runoff hydrograph and routes it
through the sewer system ' (without the associated quality),
it is necessary to have data on the rainfall, catchment
characteristics, sewer network, and measured sewer flows.
As the programs become more and more sophisticated, addi-
tional input data are required. An example of the general
data requirements for a comprehensive mathematical model is
listed in Table 82.
Although the gathering of the mass of input data required
for a very comprehensive model can be a formidable task,
the use of such a model for evaluation of additions or pro-
posed modifications to a sewer system can contribute to
large cost savings.
The utility of model applications is illustrated in the fol-
lowing two examples representing a comprehensive approach
and a "first cut" or roughing approach, respectively.
Example of Comprehensive Approach
The SWMM was applied in the Development of a Flood and
Pollution Control Plan for the Chicagoland Area in 1972 [9].
The application spanned a two-year period.
363
-------
Table 82. GENERAL DATA REQUIREMENTS, STORMWATER
MANAGEMENT MODEL [19]
Item 1. Define the Study Area (Catchment)
Land use, topography, population distribution, cen-
sus tract data, aerial photos, area boundaries.
Item 2. Define the System
Furnish plans of the collection system to define
branching, sizes, and slopes. Types and general
locations of inlet structures.
Item 3. Define System Specialties
Flow diversions, regulators, storage basins.
Item 4. Define System Maintenance
Street sweeping (description and frequency).
Catchbasin cleaning. Trouble spots (flooding).
Item 5. Define the Receiving Waters
General description (estuary, river, or lake).
Measured data (flow, tides, topography, water'
quality).
Item 6. Define the Base Flow (DWF)
Measured directly or through sewerage facility oper-
ating data. Hourly variation and weekday vs.
weekend. DWF characteristics (composited BOD and SS
results). Industrial flows (locations, average
quantities, quality).
Item 7. Define the Storm Flow
Daily rainfall totals over an extended period (6
months or longer) encompassing the study events.
Continuous rainfall hyetographs, continuous runoff
hydrographs, and combined flow quality measurements
(BOD and SS) for the study events. Discrete or com-
posited samples as available (describe fully when
and how taken).
364
-------
Study Area — The study area encompassed a 970-sq km (375-
sq mi) combined sewer area, of which 56 percent was within
the limits of the City of Chicago. All major sewers and
drains in the study area discharge to the Chicago River sys-
tem (a series of natural and artificial waterways connected
at the turn of the century to divert flows away from Lake
Michigan and into the Illinois River). Altogether, there
are more than 300 discharge points to the river system in
its 120 km (75 miles) of waterway above the control works at
Lockport, Illinois.
Step One - The consultant (Metcalf $ Eddy, Inc.) applied the
model to three large test areas selected by the city. One
of these, Roscoe Avenue, an area of approximately 2,430 ha
(6,000 acres), was selected because extensive monitoring
data were available from an earlier Public Health Service
study. The modeled results for several storms were compared
with measured data on the basis of flow, BOD^, and SS.
The results were close, considering the difficulties in
measuring flows and obtaining representative samples in very
large sewers under storm conditions.
Step Two - The results were analyzed and adjustments recom-
mended to improve correlation.
Step Three - The study area was broken down into approxi-
mately 60 subareas by combining adjacent catchment areas
which could be assumed to overflow to the receiving water at
a single point. These subareas were modeled by city person-
nel, trained and advised in the use of the model by the
consultant.
Step Four — The waterways were modeled and appropriate con-
trols and boundary conditions established. Dry-weather flow
inputs were computed, including inflows diverted from Lake
Michigan for flushing purposes. A comparison and correla-
tion was made with extensive in-stream monitoring data which
had been collected by the USPHS and a close fit established
for flow, BODs, and DO.
Step Five - The Runoff-Transport results, prepared by city
personnel and stored in tape files, were applied to the
waterway model.
Step Six - Alternate storage-treatment schemes, developed by
the city and its consultants, were tested, using the same
input tape files, and the resulting modified output files
were then applied to the waterway model and results compared
365
-------
Altogether, 18 plans were compared under each of four system
constraints. The following conditions were fundamental to
all plans:
1. Prevent backflow to Lake Michigan for all storms
of record.
2. Satisfy applicable waterway standards (DO,.
coliforms).
The input hydrology spanned 21 years of rainfall record, in-
cluding the largest storms in recorded history. The pres-
ently favored plans call for quarry (surface) storage in 2
or 3 units with treatment in conjunction with the existing
dry-weather flow facilities.
Example of "First Cut" Approach
In a second application for a reconnaissance-level study for
the District of Columbia [15], a simplified approach was
successfully applied for determining the required storage
volume for stormwater storage tanks coupled with alternative
treatment rates. The procedure, similar to the mass-diagram
or Rippl method for determining storage required in impound-
ing reservoirs, lends itself to computer application.
Assuming that the tank is empty at the beginning of the wet
season, the maximum amount of stormwater that must be stored
is equal to the difference between the runoff entering the
tank and the amount drawn off to the treatment plant subse-
quent to the next wet period. By using daily or hourly
rainfall records spanning several months or years, it was
possible to determine both the maximum storage volume re-
quired and the presumed optimum storage volume/treatment
rate to provide the most cost effective solution.
Increased use of simulation models to predict storm dis-
charge and combined sewer overflow quantity and quality will
allow many municipalities to evaluate the impact of changes
or additions to their present sewer system more easily.
The ability to evaluate a large number of alternatives in a
relatively short period of time and for much less money
(once the initial system has been computerized) is very
attractive. The significant point is that, with comprehen-
sive models, data can be developed on quantity (and quality
if desired) of storm discharges and combined sewer overflows
that can be used in the decision-making process on problems
concerning the handling of such discharges or overflows.
This capability alone is a major landmark, because in the
366
-------
past decisions on the amount of treatment or storage have,
for lack of data, been somewhat arbitrary and without ade-
quate consideration of the actual conditions in an existing
combined sewer system.
MASTER PLAN EXAMPLES
Recognition of the pollution potential of combined sewer
overflows and stormwater discharges, and concern about it,
has emerged only during the last decade. While several
large cities embarked on long-range sewer separation pro-
grams in the 1950s and earlier, these programs, in light of
altered water quality objectives, are now subject to ques-
tion and reevaluation.
Selected master plan examples demonstrating integrated
approaches are summarized in Table 83. All represent com-
bined systems, but the represented depths of investigations
are widely varied.
The size and complexity of urban runoff management programs
are so great that in-depth analyses, such as those of San
Francisco and the city and metropolitan areas of Chicago,
are rare. Other cities and metropolitan areas have contrib-
uted, frequently with EPA assistance, to limited-objective
demonstration projects that explore alternative, and some-
times innovative, control schemes upon which to build master
plan programs. Examples of both the in-depth approaches and
limited-objective studies and projects are discussed herein
for purposes of comparison.
An important consideration with respect to master plans is
that regulatory constraints and public attitudes on pollu-
tion and environmental objectives are subject to change.
The results tend to alter the ground rules for engineering
assumptions so frequently that plans lacking flexibility may
be or have become grossly outdated before implementation can
be effected.
San Francisco, California
The San Francisco Master Plan for wastewater management has
been under intensive development by the Department of Public
Works and its consultants since 1969 [3, 8, 10]. Under the
existing system, which is totally combined, overflows may
occur whenever the rainfall exceeds 0.05 cm (0.02 inches)
per hour. This happens approximately 80 times per year (on
46 days), through 41 bypass locations on the City's periph-
ery, representing a total annual duration of approximately
200 hours. The estimated annual bypassed (overflowed) vol-
ume is 22,700 Ml (6 billion gallons).
367
-------
Table 83. COMPARISON OF MASTER PLANS AND
PROJECTS IN VARIOUS CITIES [8]
Item San Francisco
Year study completed 1972
Average annual rainfall, 18.7
in.a
Average summer rainfall 0.8
(May-Sept), in.
Total combined sewered 24,000
area, acres
Proposed treatment Alt. A Alt. D
of dry weather flowb 8.0 8.0
Total storage capacity, 0.10 0.63
in. runoff
Chicago
1972
33.2
17.1
240,000
1.5
3.14
Boston
1967
42.8
16.9
12,000
NA
1.84
Seattle
1971
38.9
7.2
23,400
(equiv.)
3.0
0.05
New York
1968
42.4
19.0
3,256
1.5
0.26
Washington, D.C.
1970
40.8
20.3
10,240 (DT)
4,066 (K)
DT K
5.0 2.0
4.31 1.58
°Verfl°WS 8
p
Available level of
treatment for captured
flows
per'aSc
unknown 45
$17'18° $36'94° $11'945 *61 .905 $9,523 $6,523 $34,505 $16,840
a. At nearest major airport, 30 years of record.
b. Includes dry weather capacity where joint use planned.
c. ENR = 2000
Legend: P = primary
S = secondary
T » tertiary
C = chlorination only
DT = deep tunnel
K = Kingman Lake
NA = not available
Note: in. x 2.54 = cm
acre x 0.405 = ha
$/acre x 2.47 = $/ha
Ihe conceived master plan will consolidate overflows to only
15 locations and reduce their probability of occurrence to
eight times per year under a minimum program and to one time
in five years under the maximum program. Control is to be
achieved through a series of inland and shoreline storage
basins (up to 30 to 45 total units) and conveyance/storage
tunnels operated under extensive computer monitoring and
control. Schematics of the basic components and their
functional relationships are shown on Figure 74.
368
-------
ALTERNATE
lull FLO! AHO STOII DIIEC1
TO O.I !»./«• «•«•• CONNECTION
STORAGE BASIN
.PRETREATED STORN FLOI
IN EXCESS OF BYPASS
CAPACITY GOES INTO
STORAGE
CONTROL CtTES
rat FLOI
REGULATION
IHEN
IASIN
IS FULL
I 10 TRANSPORT
SfSTEV
EIISTINC TRUN«
I SEIER 10
S TRANSPORI
I SYSTEN VIA
I SHORELINE
IASIN
CONTIOl SUES FO
FLOI HEGUIUION
EIISTING TRUN«
SEIER CONTINUtTIO
TYPICAL UPSTREAM BASIN
10 SEWER DURING
AND AFTER STORII
PREIREHTEO STOM
FIOI IN EXCESS OF
ITPASS CAPACITY
GOES INTO STORAGE
JIT IEATKER
FLOI ANt ITOIK
FIOI UP TO O.I
IN./HI MI.*
NOTE: FLOI IN10 EXISTING
INTERCEPTOR LI»ITEO TO
APPROX. 0.03 IN. 'HR
TYPICAL SHORELINE BASIN
TYPICAL TUNNEL STORAGE MODULE
HOOIFIEO SYSTE» IILL
ALLOW UP TO 0,3 IN.
COMPONENT SCHEMATICS
SAH FRANCISCO BAY
REMOTE FROM TRANSPORT
LINE RETURNS ALL FLO*S
OVERFLOW I J—LJ__
I ^eunDEI IUF
.PSTREAM BASIN
CLOSE TO TRANSPORT
LINE »ITH DIRECT
CONNECTION. ONLY BASIN
OVERFLOWS CONTINUE TO
SHORELINE BASIN.
LAKE MERCED WET WEATHER
TREATMENT PLANT
CONCEPTUAL ARRANGEMENT
Figure 74. San Francisco master plan elements [8]
369
-------
Dunng the major portion of the year, wastes will receive
secondary treatment at one of two dry-weather treatment
tnnn^', I ?* ^eated e«l«ents will be transmitted through
tunnel and pipeline systems for ocean discharge approxi-
mately 6.4 km C4 miles) offshore. During storm conditions
flows exceeding the capacity of the secondary treatment '
plants will be transported to a 43.8-cu m/sec (1,000-mgd)
§J^h f°r Primary higher-level treatment and the effluent
discharged approximately 3 . 2 km (2 miles) offshore. Thus
the bypassing of untreated waste will be virtually elimi-'
nated and a secondary benefit of flooding control (by the
achie6™"1118 effects of uPstream basins) will be
within the plan, in addition to its advanced
concepts of automated systems control and monitoring, are
the relative wet-weather treatment capacity of 8 times the
average dry-weather flow, the proposed use of physical-
™6?nC?n treatment> and the approximate storage volumes of
*it* „ 2-PerCent of the runoff from a 1-year storm in the
alternatives considered most feasible. The total program
™S£-" es^lmated to be in the range of $450 to $900 million
menmiSn if tn*^ *?' ^ wet-weather programs and imple-'
mentation is to be phased over a period of 30 years.
Chicago, Illinois
The Chicagoland area studies date back to the mid-1960s and
«h?n relt WuS. comP1leted in 1972 ™der the joint sponsor-
ship of the Metropolitan Sanitary District of Greater
Chicago; the Institute for Environmental Quality, State of
Illinois; and the Department of Public Works, City of
S^3!° [9]: Th? "iteria established for the comparison
Sfrhia»n f P <* WeTe *° : W P^vent backflow to Lake
for al? Jn™f %°f P0t2ble !?a^er and hiSh recreational use)
w«v «JiB5 5 f°f rec?Jd5 and W ">eet the applicable water-
way standards Cprescribe tertiary levels of treatment under
dry-weather flow conditions and a high degree of Meatmen?
fhlt^in* Tges) \ Backflow actions are emergency flood
abatement measures wherein locks are opened permitting re-
verse flow in the canal system with resulting releases to
and contamination of Lake Michigan.
In the final analysis, up to 4 modifications of each of 18
alternative plans were considered. The features of the
recommended plan, incorporating conveyance tunnels, quarry
storage, and low-rate feedback to dry-weather facilities
SectionairntAnWere di?cussed earlier in this section and in
on FiSSre 7-; An °Peratlona:l schematic of the system is shown
on Figure 75. Of special interest are the provisions for
selective bypassing should the storm exceed the system
370
-------
RAINFALL
I
LEGEND
NORMAL FLOW PATHS
DEWATERING AND UNUSUAL
EX'CESS FLOW PATHS
EXISTING AND
PROPOSED
SEWERS
SLUDGE FLOW
AFTER PROLONGED
f" AERATION
| ^
NOTE: ACRE-FT x 1,233.4 - cu M
Figure 75. Operational schematic of the
recommended pollution control plan for Chicago [9]
371
-------
=nS »i X> ? staging capabilities for filling the basin*
and the provisions for alternative levels of treatment ™
general operation is described as follow!: treatment. The
...Rainfall runoff and snow melt enters the sewer
wL'tes mThing wit*. h°»sehold and industrial '
wastes. This combined flow travels through the
neIrrthe°w^COntr01 °? <4versi°n chamber located
near the waterways. In dry weather or very minor
storm periods all of the flow is diverted to the
-
In storm runoff periods exceeding the intercentor
Das£rftent E'T "Pacity. stofm overflow P
passes through the drop shafts to the large con-
veyance tunnels under the waterways. Flow is con-
veyed to the storage reservoir, first in the pri-
acrLfeaItnStheSc l ^ 2 ' f " how exceeds
• '
pal reservoir site to pump the water in the
tunnels... to the reservoir ---- The pumps have caoac
ity to perform this operation in three dayl? P
in from the waterway
anctunnels
In the post storm period, the dewatering pumps
SiH c6 °Perated t° Pun>P the stored water to the
a? »7r«? thw"t treatment plant. Pumping will be
at a rate which when added to the plant, raw sew-
flow " W111 eqUal 1
-------
The total program cost, if phased over a 10-year construc-
tion period, is estimated to exceed $3 billion, of which
approximately 45 percent is solely for wet-weather
facilities.
Boston, Massachusetts
A wet-weather flow master plan, based largely on preliminary
Chicago deep tunnel studies, was presented to the City ot
Boston in 1967 [12]. The four alternatives studied were:
m complete separation, (2) chlorine detention tanks,
(3) surface holding tanks, and (4) deep tunnels. Conclu-
sions reached in the study were:
At the present time [1967] no projects have been
constructed or planned which will significantly
improve water quality in the Boston area. To
date, overflows of mixed sewage and storm water
have been treated as minor problems, and solutions
attempted have been and are totally inadequate.
Because of the extremely high counts of coliform
bacteria, indicative of pathogenic organisms in
sewage, no relatively small amount of reduction in
either the frequency, quantity or duration of
overflows will significantly reduce the pollution
hazards.
The only solution worth the major effort required
is one that would completely eliminate overflows.
The proposed Deep Tunnel Plan for the Boston
regional area appears to offer the best and the
only feasible method for the elimination of
overflows.
Simply stated, the proposed Deep Tunnel Plan in-
volves the construction, under the city, of large
tunnels in rock, into which all of the overflows
can be discharged through vertical shafts. The
tunnels will store and conduct the overflows to a
point where they can be screened, chlorinated and
pumped through a long ocean outfall and disposed
of into deep waters of Massachusetts Bay. [11]
The system was described as 4,860 ha (12,000 acres) of com-
bined sewered area plus 4,050 ha (10,000 acres) of sepa-
rately sewered area connected to the system for sanitary
sewage flow transport.
The design basis was to capture completely the 15-year fre-
quency 24-hour duration storm, assuming 90 percent runoff.
A storage capacity of 4.64 cm (1.84 inches) of runoff was
373
-------
indicated, coupled with continuous pumping to the
ocean at
ss
* f5-S mlles) from the nearest land area
Mm tr?a^ent Planned was chlorination (30-mg/l dose'
90-minute minimum retention time) with deep water ocean
discharge and dilution ratios of 200 to 1 or belter A
maximum of 2 days was allowed for dewatering the tunnels
^JXt5,thi! c?nstraint> father treatment at the Try-
unfeasTbfe™ T^-1*7 (De?r ,Island) was Considered Y
unfeasible. To dispose of the runoff over a longer neriod
6d t0 U a mUCh greater ^crease if co?? o?
"" COmPensated *°* ™ reduced pumping
In May 1971, a demonstration surface detention and chlorina
tion facility was placed into operation in Cambridge
tnnn3? Uiettr' indicating a viable alternative to the deep
Hon .P CS6e Secjlon IX) • A new comprehensive evalua?
°m
Seattle, Washington
Combined sewer overflow abatement activities in the Seattle
area are described in a recent EPA report [14]? The p?oiect
objectives are to: project
1. Continuously monitor water depths and other factors
needed to compute flows and capacities in sewers
and treatment works.
2. Receive and process meterological data and predict
runoff intensity and volume on the basis of his-
torical records.
3. Reduce flow and store sewage in portions of the
pipeline system to permit increased flow intercep-
tion from areas experiencing high runoff rates.
4. Eliminate or reduce overflows to a level that
would meet receiving water quality standards.
Toward this end, a series of remotely controlled and moni-
tored regulator stations and a central control system were
constructed (discussed earlier in Sections VIII and I xT
Interception capacity is generally 3 times the estimated
/^^dry-weather flow; thus, for many areas, excess cauac
tiLWof th6 available for ^ny years. Ve optimal ulili^a-
°Vhls excess capacity, both for storage and acceler-
con r^Tsm^ ^^^ tO ^atment is
374
-------
To supplement these capabilities, a partial separation proj-
ect has been underway for several years. Approximately
7,290 ha (18,000 acres) of a total 14,580 ha (36,000 acres)
of combined sewer area are presently programmed for partial
separation, which is expected to result in a 70-percent re-
duction of flows in the combined system. All intercepted
flows are presently given primary treatment prior to
discharge. No special wet-weather treatment facility is
contemplated presently.
This program first became operational in 1971. Highest pri-
ority is given to abating pollution associated with summer
storm runoff. A 1-year frequency summer design storm having
a peak runoff rate of 30 to 60 times dry-weather flow was
cited in the report. The partial separation program,
coupled with in-system storage, is expected to reduce the
release of contaminants (total wet-weather) to the receiving
waters by an estimated 30 to 50 percent. In the case of
Lake Washington, a further reduction up to a total of 50 to
60 percent is anticipated by redirecting three-fourths of
the new storm drainage out of the inland basin to the Sound.
Early operations have been quite successful and programs are
being developed and tested for complete automated control.
However, population growth in the Seattle area will pro-
gressively reduce the available safe storage over a period
of years and may largely offset improvements in system
management.
Washington, D.C.
The District of Columbia started on a 50-year sewer separa-
tion program in the late 1950s. Costs for fringe areas
separated to date have averaged between $51,110 and $76,660
per ha ($20,700 and $31,050 per acre). Recognizing the limi-
tations of separation as a total solution, the program has
not been funded since 1970, pending an assessment of
alternatives.
A conceptual evaluation, completed under an EPA grant in
August 1970, followed the earlier approaches in Chicago and
Boston and recommended a system of deep rock tunnels and
mined storage with dual treatment at Blue Plains (the
regional dry-weather flow treatment plant) and a new wet-
weather reclamation plant at Kingman Lake [4]. The storage
would fully contain the combined area runoff (4,542,000 cu m
[1,200 mil gal.] or 10.95 cm [4.31 inches] of overflow) from
a 15-year frequency 24-hour duration storm. The combined
treatment capacity of the two plants would be 19,275 I/sec
(440 mgd); thus, they would be capable of dewatering storage
from the 15-year storm in less than 3 days. The estimated
375
-------
project cost of $353.3 million was approximately half th*
duceTn6?!0??* f°r t0tal ^ration aSd the blne£i?s in re
duced pollution were considerably greater. lc"ts ln re
EPA^was »adfy;nC?mple^d concurrently and also funded by
EPA, was made to investigate the reclamation asoects of
combined sewer overflow abatement [S]. In Uiifltudy
¥ € -J ! i:i!rr°^F""s "'^™
i/ rrn 4-T 8 lnches) of runoff. In addition a 2 IQft
Sss:
«d ^strict recently engaged a consultant to review these
and other alternatives and to recommend a Master Plan
ln°lT97^ ^/!;COnnaiSrnC? level reP°rt was completed early
pt7 d -
CeiT^^ waters. A total storage capacity of 2 271 000
cu » (600 mil gal.), coupled with an average excess flow
s^L-^ajy^.-^'jg); r0 sa-j 3?..
r?.;±/ -,r'f s
™uld
ENVIRONMENTAL COMPATIBILITY
sewer^^w1^
^iSTtg^S^^ Kcat^ 2T(r "
™ivinTwa^ ±??!!n! ^Lf-!-1 -quiremen?f f^om'a^e-
376
-------
Characteristically, the location of storm flow pollution
abatement facilities, based on need and cost constraints,
will coincide with lands of high potential use and public
access. It becomes imperative, therefore, that facilities
not only perform effectively but that they do so without
nuisance and that they enhance, not blight, the landscape.
Many EPA Research,. Development, and Demonstration projects
have accomplished this task with distinction. A few of the
more noteworthy are highlighted in this section.
Satellite Facilities
Satellite storage and treatment facilities, generally lo-
cated near the intersection of a main trunk sewer and an
interceptor, should receive priority consideration for aes-
thetic architectural treatment. These facilities usually
are constructed in highly developed urban areas. In many
cases, they are near, or within, public recreation areas.
Examples include the Baker Street Dissolved Air Flotation
Facility in San Francisco, the Cottage Farm Detention and
Chlorination Facility in Boston, the Humboldt Avenue Reten-
tion Basin in Milwaukee, and the remotely operated regulator
stations in Seattle. The addition in time and money re-
quired to enhance such facilities architecturally has
resulted in increased public acceptance and, in some cases,
increased real estate values.
The Baker Street Dissolved Air Flotation Facility, Figure 76,
was built fronting on San Francisco Bay adjacent to a large
yacht harbor and a popular public park. It is questionable
whether most of the people who use the area even know there
is a combined sewage overflow treatment facility there.
Planter boxes have been incorporated into the design along
with access hatch covers that double as benches. During the
summer, there are usually a number of bathers sunning them-
selves on the seawall. The architectural treatment and low
profile of the exterior of the building make it blend well
with the surrounding area.
The Cottage Farm Detention and Chlorination Facility in
Boston, Figure 39 (Section IX), is located on the bank of
the Charles River across from Boston University. Because
this reach of the river is a major boating and rowing recre-
ation area, the storm flow detention tanks have been buried,
and the surface has been planted with grass. The control
and operations building blends well with the surrounding
area.
The Humboldt Avenue Retention Basin in Milwaukee is located
on the bank of the Milwaukee River, Figure 37 (Section IX).
Although the facility is located in a present industrial
377
-------
(a)
(c)
(d)
Figure 76. Baker Street dissolved air flotation facility
San Francisco
378
-------
Figure 77.
Remote-controlled regulator station
Seattle, Washington
area, aesthetics were a major consideration during design.
The tank was buried and the surface was planted with grass,
as was done at the Cottage Farm facility in Boston. The
architectural treatment given to the control and operations
building has produced a striking effect, while still pro-
viding vandal-resistant security. There has been no need to
fence the tank area so it is open to public use.
The environmental compatibility of storm flow facilities is
further demonstrated in the regulator stations in Seattle,
Washington, as shown on Figure 77. One station in particu-
lar, the Dexter Avenue Regulator Station (not illustrated),
serves a dual purpose, providing in addition to its storm
flow control function both a minipark, complete with
planted areas and benches, and a bus stop. The only evi-
dence that the regulator facilities exist is a single access
door in the retaining wall at the rear of the park.
Multiple-Use Facilities
Multiple-use facilities on a grander scale are demonstrated
in the Kingman Lake conceptual plan and in the Mount Clemens
combined sewer overflow reclamation facility, both discussed
earlier in the text.
379
-------
J public interest in abating pollution and
th5°f add^onal environmental and recreational f«il
'11 *?easibillty °f a m«ltiPurpose project combining
™~ I*®0*?0!! and treatment °f combined sewer overflows and
recreational development in the Kingman Lake area was
investigated [S]. The Kingman Lakewea was ^cognized as
for'n8^6 P°tentlal to be developed as a major urtan center
for outdoor recreation. Yet in its present state the "lake"
sewfrUd? ^ ParVy combined sew« overflows and storm
?« ?h d"charf?s fTom the District of Columbia discharging
to the Anacostia River. The conceptual plot plan for the
project II the"? ?? FJ?Ure 7,8' The 2ene?al scheme of the
££i«Li% ?? collection and storage of the combined sewer
itv sii?.b£ £We2-by-treat?e!lt °f the over£1°ws to a qual-
^L™ f v j fishing and boating in the lower section of
Kingman Lake and for public bathing in the upper section of
been aJ*: ' 2 3 P?rt^On °£ the treatment facilities has
been envisioned as having multiple uses with the surface of
navt T3^ storage basin being used as a parking lot and
facility r°Ute trUCkS servicing the reclamation
A system similar in concept but smaller in scale has been
constructed and placed into operation in Mount Clemens?
Michigan. The lakelets (described previously in Section XI)
will provide a recreation area with boating/fishing
PROPOSED EAST LEG INNER LOOP FREEWAY
TRUNK SEWER
STRIBUTION HEADER
TREATMENT
EAST CAPITOl STREET
Figure 78. Conceptual plot plan
of Kingman Lake water reclamation facility [5]
380
-------
picnicking, etc., as well as a transitional buffer zone be-
tween residential developments and commercial industrial
complexes. A model of the facility is shown on Figure 79.
This project was the recipient of the Eminent Conceptor--
1971 Consulting Engineers Council of Michigan Engineering
Excellence Award, and was also the recipient of an Honor
Award--1971 of the Consulting Engineers Council/U.S.A.
Indicative of the project's success, the land around the
pond site is being built up with high-rise apartment
buildings.
Total Planned Communities
An appropriate final example is a new EPA project as a part
of a planned community being developed near Houston [13].
Entitled "Maximum Utilization of Water Resources In a
Planned Community," the study will focus on how a "natural
drainage system" can be integrated into a reuse scheme for
recreation and aesthetic purposes. Runoff will flow through
low vegetated swales and into a network of wet-weather ponds,
strategically located in areas of porous soils, as well as
into variable-volume lakes. This attenuation process will
allow some of the runoff to seep into the ground and retard
Figure 79. Combined sewer overflow
collection and treatment facility,
Mount Clemens, Michigan
381
-------
the flow of water downstream, thus preventing floods caused
T '^In addition> P0™us pavements will be
to further attenuate runoff, and the efficiency of
t
a ..
and disinfectant residuals in runoff and their effect in the
receiving lake system will be measured. Evaluation of a
multiprocess and flexible stormwater pilot plan" which will
treat runoff before it enters the lake, will also be part of
System "valuations wil? relul? in
Considering urban runoff as a benefit as opposed to a waste-
^e£'xal°!}g "lth the concePt of new community development
which blends into and enhances its environment rather ?han
thlfirs? tkeWllH b%em?loyed and thoroughly evaluated for
the ni?,,L? • H°Pefully- ll "ill be shown that man and
the natural environment can beneficially coexist.
Thus, as discussed in this section, integrated approaches
mathematical modeling, and environmental compatibility are
both desirable and cost-effective adjuncts to wastewater"
management systems planning, design, and program
implementation. Highlights pertaining to the operation and
maintenance of storm flow facilities Ire detailed in ?he
next section.
-------
Section XV
OPERATION AND MAINTENANCE
The term maintenance in an engineering sense may be defined
as the art of keeping plant equipment, structures, and
other related facilities in a suitable condition to perform
the services for which they are intended [12]. Operation
of a storm flow system requires not only the physical opera-
tion of the various components but also their operation in
unison and on-eall. The components include the wet-weather
treatment facilities, the dry-weather flow treatment facil-
ity, the collection and control network (whether multiple
systems or combined), and the sludge-handling and disposal
program.
A perspective of the problems, such as high turbulence, flow
momentum, vast areas of tankage and channels, and restric-
tive environment, encountered may be seen on Figure 80 which
illustrates a combined sewer overflow detention facility
before, during, and after a storm event. Note the relative
scale of the facilities from the figure in photo (e). One
of the most remarkable insights from the series of photo-
graphs is that all were taken within the space of a few
hours.
Discussed in this section are operating controls and op-
tions, sustaining (dry-weather) maintenance, support facili-
ties and supply, safety, and solids-handling and disposal.
The importance of each one of these aspects of operation
and maintenance cannot be overemphasized. If the facility
does not come on-line when needed, nothing will be
accomplished. If it is not kept in good repair, the re-
sults may be catastrophic. Finally, if provisions are in-
adequate for the removal and ultimate disposal of the
separated solids, a greater problem may have been generated
than solved.
383
-------
(a)
(f)
Figure 80, Stormwater detention facility (Boston)-
before, during, and after storm event
Residual solids in tanK and on effluent horizontal Mo. sheens at"! sio
384
-------
OPERATING CONTROLS AND OPTIONS
Operating controls can vary from simple orifices and other
static devices to complete computer-run automation. The
options range from flow routing and control to equipment
startup and shutdown, to chemical feed, to process adjust-
ments and equipment surveillance, and finally, to alarms
and emergency action.
Collection system control has been discussed in Section VII
A brief review of the control panel from a storm flow
detention-chlorination facility, shown on Figure 81, will
serve to illustrate typical off-line facilities control.
Equipment status and a simplified flow diagram are dis-
played across the upper half of the board, and actuating
Figure 81. Control panel at the Cottage Farm
detention and chlorination facility (Boston)
385
-------
f™;!,C?%and- alaiT are arran8ed below. For example, reading
from left to right are the status lights (open, throttled
66 *1 SlUie ate» »ith tL -- -
OFuTOMAr tir --
OFF-AUTOMATIC) control switches below; the sets of coarse
and fine screens similarly controlled; the wet well level-
the main sewage pumps (level -controlled) ; the pump discharge
theWrha *' the 1!Vel in the ^tention tanks; and finally^
^et^°"n%re!1ldual monitor. The important thing, again,
is that the facility come on-line when needed and that it
dana^V protect it?*" and/°r the public from major
danger (e.g., automatic closure of inlet gates coincident
with high wet well level alarm) . . ^"incident
Several examples of operating controls and options are pre-
sented in Table 84. Note that the various control devices
are fairly diverse, while the monitoring or status -
Because storm events occur at random intervals, storm flow
«E£5« systems must be capable of automatic startup and
r£ft?£T'* *1S reason' the instrument and equipment
reliability requirements may be much more demanding than
those for dry-weather flow facilities. Equipment reaction
time is very important. Large motor-operated sluice gates
typical of those used in combined sewer overflow and f?o?m'
sewer discharge applications, generally operate a? a stand-
per minuteaP?h^im\tlng 30'51™ J12 incheS) of gate movement
per minute, thus, changes will be relatively slow.
.
To ensure reliable startup and shutdown, all instrumenta-
tion must be checked and calibrated on a regular ba^is
All equipment must be exercised regularly to check and'en-
H^/eadlneSS> ?nd £acility cleanup, lubrication, and
dewatenng must be done following each storm. Degradation
of the retained flow and solids must be expected with in-
creased retention times, and dewatering rates must be
selected accordingly.
Components crucial to the operation, such as valves flow-
meters, chemical feeders, telemetry equipment; live! sen-
sors, Pressure sensors, and limit switches, should be
Careful planning and design of combined sewer overflow and
cle^.rn dlschar8e facilities permits minimization of
cleanup and maintenance via increased automation. It must
nointT^ered> however- that there will be a break-even
°
386
-------
Table 84.
Location,
[Ref.]
Operation
scale
Operation
controlled
Monitoring devices
Control devices
Seattle, Wash. [14]
Regulators Full
Boston, Mass. [7]
Cottage Farm Storm Full
Overflow Detention
and Chlorination
Facility
Dallas, Tex. [11]
Bachman Creek Full
Polymer Injection
Station
San Francisco,
Calif. [9]
Baker Street Full
Dissolved Air
Flotation Facility
Mount Clemens, Mich.
[13]
Combined Sewer Full
Overflow
Collection and
Treatment Facility
Milwaukee, Wis. [2]
Humboldt Avenue Full
Combined Sewer
Overflow Detention
Tank
Minneapolis, Minn.
[8]
Regulators Full
Overflow Level sensors, rain gages,
quantity wind gages, automatic sam-
plers, telemetry units,
computer, position sensors
Overflow Level sensors, automatic
quality and sampler, Dal! tube, re-
quantity sidual chlorine analyzer
Sewer Level sensor, magnetic
surcharge flowmeter, temperature
probe, computer
Overflow
quality
Overflow
quality
Magnetic flowmeters, level
probes, differential pres-
sure sensor
Magnetic flowmeters, level
probes, automatic samplers,
differential pressure
sensors
Overflow Level sensors, automatic
quality and samplers, magnetic
quantity flowmeters
Overflow
quantity
Rain gages, computer,
pressure sensors
Gate regulators, tide
gates, pumping stations
by remote centralized
operation
Pumping station, sodium
hypochlorite feeder, gate
regulators
Polymer injection feeder
Polyelectrolyte feeders,
solids pump, bypass
gates, dissolved air
flotation units
Pumping stations, aerated
lagoons, microstrainer
drive, pressure filters,
chlorine-chlorine dioxide
feeders
Mixers, chlorinators , bar
screen
Fabridam regulators, gate
regulators by remote
operation
SUSTAINING (DRY-WEATHER) MAINTENANCE
Sustaining, or preventive, maintenance is the one key to a
successful storm sewer discharge or combined sewer, overflow
pollution abatement and control system. The performance of
many existing systems could be greatly improved by strict
adherence to a well-planned sustaining maintenance program.
An example of what can result from a poor or nonexistent
maintenance program is illustrated in one of the conclusions
387
-------
drawn by investigators looking into the status of the ten
sewage collection systems contained in that portion of the
Hudson River Basin which lies within the waters of the
Interstate Sanitation District (Connecticut, New Jersey
New York): 7 »
In far too many cases, the personnel assigned to
regulator maintenance duty are not properly
equipped with either maintenance tools or the
necessary safety equipment to properly service
the regulators. A noticeable lack of under-
standing on the part of many of the maintenance
personnel as to the purpose and proper operation
of regulators was evident during field
inspection. Additionally, many of the regula-
tors were intentionally jammed or chained in the
open position to maximize wet weather flow to
the treatment plant. This can create a three-
fold problem of (a) an imbalance of sewage mix-
ture from each regulator drainage basin during
wet weather flow; (b) an unreasonably high wet
weather flow to the treatment plant which in
turn minimizes the effective treatment of this
sewage; and (c) a surcharge of the interceptor
system with associated local flooding of sewage
into streets and basements. [5]
There are two general categories of combined sewer overflow
and storm sewer discharge pollution abatement and control
« -[?-ties requiring sustaining maintenance: in-line and
ott-line. The in-line facilities include various types of
regulators tide gates, polymer injectors, flushing systems,
and flow and quality monitors. Off-line facilities include
the various physical, biological, and physical-chemical
treatment processes, as well as large storage facilities.
These are generally much more sophisticated than the
regulators.
In-Line Facilities
Generally, the sustaining maintenance required for regula-
tors and tide gates increases as the degree of collection
system control increases. For this reason the overall main-
tenance required increases as the complexity of the regu-
lators increases (from static to semiautomatic dynamic to
automatic dynamic). All regulators and tide gates are sub-
ject to the main problems of corrosion and clogging bv
debris Detailed maintenance problems of the most common
types of regulators are discussed elsewhere in the
literature [4]. Maintenance problems associated with the
388
-------
more recently developed regulators and other in-line facili-
ties are discussed here.
Regulators — Static regulators require periodic checks to
ensurethat weirs and orifices have not become clogged or
blocked. Floatables and deposited solids may need to be
flushed into the interceptor to prevent their accumulation
and the formation of nuisance problems. Included within
this category are vortex, horizontal and vertical orifice,
and spiral flow regulators; swirl concentrators; and side-
spill and high side-spill weirs.
Semiautomatic dynamic regulators (such as tipping gates, cylin^
drical gate, and float-operated regulators) also require peri-
odic inspection and maintenance. Regular lubrication of
pivot points, bearings, and mechanical linkages is necessary.
Floats and linkages must be free of debris to ensure smooth
operation. Metal parts should be checked for corrosion. Ori-
fices must be free of clogging and debris to ensure correct
operation of the regulator. Cylinder seals must be tight.
Broad-crested inflatable fabric dams are not usually sub-
ject to clogging or jamming since they can be deflated.
However, even when deflated, some buildup of solids occurs
immediately upstream and requires periodic flushing. Fabric
dams must be inspected regularly for punctures. The sophis-
ticated control system, consisting of compressors, solenoid
valves, sealed potentiometers, pressure transducers, etc.,
requires attention from skilled maintenance personnel. In a
demonstration project in Minneapolis-St. Paul [8], about
50 percent of the Fabridams had to be replaced in the first
two years of service (one after only one month). Most fail-
ures were due to problems with anchorage, which is extremely
critical. Other failures were related to air leaks, punc-
tures, and autodeflation. Two of the failures were de-
scribed as "catastrophic," indicating the failures occurred
with considerable sewage in storage, thereby releasing
flood waves through the overflow pipes. Now, however,
after 4 years of operation, 13 of the original 15 regulator
stations are still in service. The weekly sustaining main-
tenance and inspection time for these stations involves
approximately 12 to 16 hours by electricians and 120 hours
by interceptor servicemen.
Fluidic regulators require periodic inspection to make cer-
tain that control system components are not clogged. The
control system piping should be checked for air leaks at
joints and seals. Occasional flushing of sediment and accu-
mulated solids may be necessary after long periods of dry-
weather flow.
389
-------
Tide Gates - Periodic inspections are necessary to ensure
correct operation. Lubrication of pivot points is a regular
requirement. Hinge arms and gate openings must be checked
regularly to be certain they are free of trash, timbers, or
other debris that might keep the gate in a partially open
position allowing inflow to the sewer system. The seating
surfaces, subject to corrosion and warping, should be in-
spected to ensure positive sealing. At locations where
shnn^h 6 gateS aT,e US6d' rem°te alarm Proximity switches
should be incorporated in each gate. Such alarms would notify
plant operators or maintenance personnel of tide gate openings
or malfunctions along with malfunctioning regulators? °penings
Other In-Line Facilities - Polymer injection facilities re-
qu*re checking of the storage tanks for leaks. Dehumidifi-
cation equipment must be operational to prevent solidifica-
tion of the polymer during storage. Injection pumps, mixers,
and polymer feeders should be checked for clogging. Sewer
liquid level sensors should be checked and cleaned
periodically.
Regular inspection and maintenance of flushing systems is
required to prevent corrosion and clogging. If sophisti-
cated control systems are employed, skilled maintenance of
valves, operators, etc., is necessary.
Capacitance probes, bubble tubes, and pressure transducers
used for flow depth measurement require constant attention
to prevent clogging or coating with grease. Any depth-
measuring device presenting an obstruction to flow must be
inspected regularly for structural damage. Instruments
must be checked and calibrated regularly and require highly
skilled maintenance. & y
A very graphic example of the types of problems encountered
in one study pertaining to monitoring combined sewer over-
flows follows:
During the monitoring program, several operating
problems were encountered, the most significant
of which were:
1. Difficulties in Pumping Wastewater Up from
the Sewer--The submersible pump anchored
to the bottom of the sewer was often
clogged by solid wastes (such as cans,
rags, wire, wood chips, tree stems, gravel,
sand, etc.) and stopped working. There
were also some pump stoppages during low-
intensity storms, probably because of in-
sufficient water depth in the sewer.
390
-------
2. Physical Damage to Equipment Installed in
the Sewer--During intense storms, heavy
solid wastes (such as tires, concrete
slabs, 55-gallon drums, mattresses, auto-
mobile radiators, chains, etc.) slammed
into the protective cages of the submersi-
ble pumps and caused extensive damage to
various equipment items. Bubbler lines
were broken and torn loose; the protective
cage was severely deformed and even disin-
tegrated; pump braces were sheared off;
pumps were washed away in the sewer; and
the electric conduit was pulled out of its
fastening studs.
3. Flooding of Lithium Chloride Release Sta-
tion at Rose Park Playground--This station
was sunk half way into the ground, at the
request of the local residents. Conse-
quently, the lower part of the structure
was inundated by excess storm water runoff.
This flooding caused minor damage to the
bubbler instruments, the lithium chloride
release system and the pressure-to-current
transmitter. This problem was overcome by
reinstalling the equipment above grade.
The equipment malfunctions and physical damage
described above prevented complete coverage of
all the storms that occurred during the monitor-
ing period. [3]
Off-Line Facilities
Usually off-line treatment or storage facilities are larger
and of more sophisticated design than in-line facilities
Several subtasks are integrated to accomplish the overall
treatment task. Effective control and operation of such
facilities are usually dependent upon varying degrees of
instrumentation. This may range from a simple liquid level
sensor all the way to highly instrumented, fully automated
operation.
Required maintenance common to most off-line facilities may
include lubricating of equipment; inspecting and cleaning of
chemical pumps, electrical and pneumatic sensing probes
flow measuring and recording devices, and automatic sam-
plers; checking and calibrating instruments; checking emer-
gency power generators and starting batteries; and inspect-
ing all pumps, valves, and piping. All equipment should be
391
-------
operated for short periods at least twice a month. Good
general housekeeping is essential for both successful oper-
ation and safety.
Maintenance problems peculiar to some of the more commonly
used off-line abatement and treatment processes are de-
scribed in the following discussion.
Physical Processes - Sedimentation basins and dissolved air
flotation sludge holding tanks and flotation chambers should
be dewatered and cleaned following each storm to prevent
nuisance odors and restore full treatment capacity. Solids
removal equipment, pumps, and other auxiliary equipment
should be cleaned, inspected, and lubricated on a regular
basis. Pretreatment devices, such as fine screens and cy-
clone grit removers, may require special attention.
Chemical metering, mixing, and injection equipment should
be serviced after each storm.
Screening devices, such as rotary fine screens, drum screens
and microstrainers, should be thoroughly inspected after
each storm. Special attention should be given to the condi-
tion of the screens because they are susceptible to damage
from sharp objects (causing rips, tears, or breaks), grease
blinding, and algae growths. Screens should be cleaned or
replaced as necessary. The equipment should be lubricated
and operated on a regular basis during dry-weather periods.
Filters should be visually inspected after backwashing to
verify their cleanliness. Pumps, valves, and other auxil-
iary equipment should be checked and serviced regularly.
Disinfection may be necessary to prevent biological growth
within the filter. Sludge holding tanks must be cleaned
after each storm. Chemicals and polymers must be replen-
ished, and metering and mixing equipment must be cleaned
and inspected. The strength of the stored disinfectant
should be monitored.
Biological Processes - Generally, biological processes do
n°t require the detailed after-storm maintenance needed by
physical processes. This is true because biological com-
bined sewer overflow treatment processes are usually oper-
ated as, or in conjunction with, dry-weather flow facilities
They therefore receive the same maintenance care as the dry-'
weather facilities. Generally, it is necessary only to wash
down the facility following a storm. During extended dry-
weather periods, such equipment as pumps and valves, used
only during combined sewer overflows, should be operated on
a regular basis.
392
-------
Lagoons require only minor maintenance, mostly of a cleanup
^e;-Unle^ ae?;at0rS are used- Aerators require periodic
reaufred°hva? lubricatlon« .The general ongoing maintenance
required by lagoons is aquatic weed removal, embankment and
access road inspection and repair, vegetation control
and periodic removal of accumulated sludge deposits.
Physical-Chemical Processes - Physical-chemical processes
are generally ratner complex in that considerable auxiliary
equipment is required for their operation. This includes
pumps, valves, instruments, chemical feeders and mixers, etc,
filled personnel are required to maintain and calibrate the
instruments and process controllers on a regular schedule.
bludge holding tanks must be cleaned following each storm
it recalcination and carbon regeneration equipment is used
special maintenance is usually required, especially for the
instrumentation. J
Practices for Improved Operation and Maintenance
Satisfactory operation of both in-line and off-line com-
bined sewer overflow abatement and treatment facilities
depends, to a large extent, on adequate regular inspection
and maintenance. The purpose of this is twofold: first to
locate and correct any operational problems or failures and
second, to prevent or reduce the probability of such prob-
lems or failures.
Inspection should be as frequent as necessary to keep such
facilities in good operating condition. Generally, this
means inspections both on a weekly schedule and following
each major storm. Regulators with small orifices or drop
inlets with grates may require more frequent inspection.
It is recommended, however, that no regulator or tide gate
be inspected less frequently than twice each month and after
each storm [6]. This also applies to off-line facilities.
Maintenance of regulators should be carried out by crews of
three to five men, depending on the type and complexity of
the regulators used. A minimum crew of three men is recom-
mended so that one man may remain at the surface while the
other two enter the chamber. Other appurtenances, such as
gate operators, depth probes, and telemetry equipment
may require additional highly skilled maintenance personnel.
Maintenance of off-line facilities should be carried out by
as many men as necessary to complete the job rapidly and
safely. This will, of course, depend on the size and com-
plexity of the facility. For example, in the Humboldt
Avenue Detention and Chlorination Facility demonstration
project [2], it was assumed that normal tank operation would
393
-------
require the attention of one man for approximately 2 hours
per day, seven days per week, plus special visits during and
immediately following each period of rainfall. Prior to
tank startup, it was further assumed that two men would be
required full-time to operate and maintain the tank equip-
ment through a 3- to 6-month shakedown period. Because of
equipment malfunctions and because it was necessary to ob-
tain maximum benefit from the facility, the two-man opera-
tion was extended a full year. The tendency to underesti-
mate actual staffing requirements appears common to most, if
not all, of the demonstration projects surveyed--largely
because of the unusual and demanding service requirements.
Complete records should be kept of all inspection and
maintenance. The time and date of each inspection should
be recorded, together with a description of the condition of
the equipment and the work performed. The number of man-
hours spent on each piece of equipment should be noted.
These data should be tabulated for each piece of equipment
and summarized for each facility, thus quickly revealing
any equipment requiring excessive maintenance or that is
out of service with unusual frequency. These records can
provide the data needed to compare the cost and efficiency
of different types of equipment for guidance in the design
and purchase of new equipment or the remodeling of existing
equipment. Such records also aid in the scheduling of pre-
ventive maintenance.
Maintenance Costs
The costs for maintaining sewer regulators, as reported in
a recent national survey, are presented in Section VIII.
Maintenance costs for off-line facilities are also pre-*
sented in the discussion of the various different processes
in Sections IX to XIII. Adjustments and the basis of costs
are noted with the presentation of the costs.
SUPPORT FACILITIES AND SUPPLIES
The importance of maintenance support in the operation of
both in-line and off-line control and treatment facilities
increases as the number and/or size of such facilities
increases. In view of the wide variety of control and
treatment processes, no attempt will be made to cover the
specific requirements of each individual process; only the
common general requirements will be discussed. The four
major requirements are (1) access to equipment, (2) ade-
quate tools and equipment, (3) a specialized work area,
and (4) spare parts stock.
394
-------
The layout of control and treatment facilities should meet
two criteria: (1) hydraulic requirements, and (2) operation
and maintenance requirements. Very often, insufficient at-
tention is given to operation and maintenance requirements.
As a result, on-site maintenance may be badly hampered by
lack of maneuvering space and room to use tools. The re-
sulting inadequate maintenance may cause improper operation
or failure of the equipment. Access means providing not
only adequate servicing room, but also room and a means for
removing the equipment from the structure. Invariably (in
accordance with one of Murphy's Laws that if anything can
go wrong, it will) every piece of equipment installed, at
some time for some reason, may have to be removed. It is
usually expensive to remove equipment by dismantling it
and even more expensive to have to demolish a portion of
the structure. Thus, attention should be paid to providing
access and working room when designing control and treat-
ment facilities.
Adequate tools should be readily available to perform the
necessary maintenance. The nature of the tools stocked at
any facility will be dependent on the particular type, size,
and equipment involved. At a small facility where little
or no mechanical equipment is used, only a few tools are
needed. At a large plant, on the other hand, necessary
tools and equipment may sometimes be extensive.
The following items afe considered necessary for maintain-
ing in-line control facilities [6]:
1. A half-ton panel truck with a two-way radio, winch,
and A-frame
2. A portable generator
3. Two submersible pumps
4. A blower unit
5. Various chains, ropes, hoses, ladders, pikepoles,
sewer hooks, sewer rods, chain jacks, tool kits,
etc.
6. Sewer atmosphere testing meters, safety equipment,
helmets, harnesses, first-aid kits, danger flags,
signs, barricades, life jackets, gas masks, air
packs, gas detector lamps, fire extinguishers, ex-
tension cords, protective clothing, etc.
7. Various spare parts
395
-------
Maintenance shops are usually required in support of off-
^^ treatment facilities. Occasionally, in-line control
and treatment facilities also include large pieces of
equipment. The importance of shops in the operation and
maintenance of treatment facilities increases with the size
of the plant. Shops may be classified as general repair
shops and machine shops. These shops should be in the same
building in adjoining rooms.
A single shop should be capable of supporting several in-
line facilities and, possibly, several off-line facilities
also. This is dependent on the number, size, and location
of the facilities. It may be possible to expand existing
dry-weather flow treatment plant shops to handle the addi-
tional maintenance. Cost savings can usually be expected
when using this centralized maintenance concept. Wet-
weather maintenance shops should be equipped and laid out
similarly to those for dry-weather treatment facilities be-
cause most off-line wet-weather treatment facilities have
maintenance requirements similar to those required for drv-
weather treatment processes.
Since storm events occur without regard for unfilled parts
orders or shipping dates, a spare parts inventory is essen-
tial as a backup to any maintenance program. A large stock
o± all parts is unnecessary, but those most subject to wear
and frequent breakage should be on hand. This involves a
storage room or compartments where these parts may be pro-
tected until used. This section of the maintenance facility
should be separate from the repair shop itself so that it
may be locked up and parts dispensed as necessary under an
inventory system.
Administration facilities should be centralized by making
use of existing dry-weather facilities, if possible. This
avoids unnecessary records duplication and affords easy
data assembly and analysis.
SAFETY
The Occupational Safety and Health Act (OSHA) of 1970, which
^Ca?! fi^t^ll ^Pril ?8? 1971' aPPlies to every employee
in the United States and its territories. The Act specifies
three areas of principal concern:
1. Record-keeping requirements related to employees.
2. Rules and regulations pertaining to design fea-
tures and safety requirements of facilities.
3. Rules and regulations pertaining to construction
safety practices and precautions.
396
-------
The hazards are a function of the working environment, oper
ating procedures and practice, and condition and design of
facilities. Sewage works deaths reported are considerably
higher on a man-hour basis than those occurring in machine
shops. Deaths in sewer manholes have been as many as 12 in
a 2-year period and, in one case, 2 men died and 2 others
were overcome, all in the same manhole.
In Westchester County, New York, 1956, two men
lost their lives by entering a manhole on a
large diameter sewer to rescue a third man who
had collapsed. The rescuers apparently were
overcome and were swept away through the large
sewer. Their bodies were recovered five hours
later, a half mile down stream. [12]
The intention here is not to give a full synopsis of safety
considerations, but rather to furnish a reminder that storm
tlow management applications expose personnel to very real
and very dangerous environmental conditions. For example,
in connection with sewer gas:
Sewer Gas is a misnomer since it is not a single
but a mixture of gases from the decomposition of
organic matter. It is actually sewage sludge
gas with a high content of carbon dioxide and
varying amounts of methane, hydrogen, hydrogen
sulfide and a small amount of oxygen. The
hazard is usually from an explosive mixture of
methane and oxygen or, more often, from an oxy-
gen deficiency. This definition does not in-
clude the extraneous gases or vapors which may
be present in sewers from gas main leaks or from
gasoline or other volatile solvents which fre-
quently find their way into sewers. [12]
The chemicals used or stored, such as chlorine, present
another problem because of their toxicity, corrosiveness,
etc. Plant features, such as railings, kickboards, safety
treads, multiple access/egress points, ventilation, light-
ing, auxiliary power sources, and detection and observation
points, must be fully incorporated into design and practice.
SOLIDS HANDLING AND DISPOSAL
Often, on process flow diagrams, the solids disposal problem
is solved simply and easily by showing an arrow pointing off
the diagram labelled "to solids disposal." In actuality,
solids disposal is one of the largest problems faced by
operators of treatment facilities. In stormwater and com-
bined sewage treatment facilities, the solids disposal
397
-------
problem may be orders of magnitude greater than that at
separate sanitary sewage treatment plants. For example
the deposited solids in a combined sewer available for
resuspension by storm flows may be similar to those shown
on Figure 82.
The removal of solids deposited in sewers used for storm-
water storage is a definite problem. One method for re-
moving deposited solids is that used in the Red Run Drain
Area near Detroit [1]. There, the detention structure,
19.8 meters by 6.1 meters by 3.35 km (65 by 20 by 11,000
feet) , is being constructed with a partially sloping bottom
leading to a central trough and a flushing system. Nozzles
spaced at 2.03-meter (6.67-foot) centers along the ceiling
at the outer walls discharge flushing water at rates up to
12.6 I/sec (200 gpm) to wash down the walls and the bottom
to the central trough, as shown on Figure 83. Other sewer
tlushing methods are described in Section VIII. The remain-
der of this section will deal with the handling and disposal
of solids from storm sewer discharge and combined sewer
overflow treatment facilities.
The problem of solids handling and disposal from off-line
combined sewer overflow facilities is threefold: (1) esti-
mating the quantities, (2) collection, and (3) ultimate
disposal.
Estimating Solids Removal
One of the greatest handicaps to the designer of combined
sewer overflow abatement facilities is the lack of viable
data on the amount of solids to be expected. This presently
inadequate data base will be bolstered as further operating
experience with the various demonstration projects is
gained. Sharp departures from the norm must be expected,
however, in individual applications. EPA is presently spon-
soring a project to determine the solids content of combined
sewer overflows.
Solids removal and disposal methods for several combined
sewer overflow treatment demonstration projects are listed
in Table 85. The figures listed in this table have been
calculated from reported data and are presented here only
to indicate orders of magnitude.
One item of major interest is the comparison of the solids
resulting from normal dry-weather operation of a treatment
facility and those resulting from wet-weather flows The
impact of imposing the additional solids from wet weather
upon existing dry-weather facilities is illustrated in
the following example.
398
-------
Figure 82. Sludge bank in Conner Street sewer
Detroit [15]
A 2.3 m (7.5-foot) deep sludge bank resulting from in-line storage; two barrels of a
three barrel sewer; each barrel, 4.8 m X 5.3 m (15.75 X 17.5 feet), contained sludge
deposits extending downstream about 2,150 m (7,000 feet) and ranging from 1 5 m to
1.8 m (5 to 6 feet) in depth for the first 1,220 m (4,000 feet)
399
-------
Figure 83. Flushing System, Red Run drain, Detroit [1]
The total sanitary sewage flow from a 1,000-ha (2,469-acrel
area will be 141 7 I/sec (3.23 mgd), assuming 32.3 person!/
ha (13.1 persons/acre) and 378 I/capita/day (100 gpcd) of
flow. If the sewage contains 200 mg/1 SS and 200 mg/1 BODc
and is treated in a conventional activated sludge plant
(assuming 65 percent SS and 35 percent BODr removal in pri-
mary clarifiers) the sludge production will be approxi?
mately 2,470 kg/day (5,500 Ib/day) of dry solids.
If a 1.27 cm/hr (0.5 in./hr) 2-hour duration rainstorm over
the same area will result in 1.27 cm (0.5 inch) of runoff
(assuming a runoff coefficient of 0.5), the total runoff
t£ir?hplU ^ 126'9°° CU m (1°2'9 acre-ft). Assuming
?na th Vlf * iS constant and al* runoff occur! dur-
ing the 2-hour storm duration, and that the storm flow can
be characterized as having 600 mg/1 SS during the first
30 minutes ("first flush" phenomenon) and 145 mg/1 SS during
f72 ^'TlM'V0 minutes>.the total solids load is 32,835 kg
(72 483 Ib) SS, or approximately 13.3 times the average
daily dry-weather flow solids production. Thus, the tradi-
tional oversight of not being concerned about storm flow
solids beyond sending them to existing dry-weather flow
facilities is somewhat frightening.
If the runoff is totally contained and sent to the dry-
weather flow facility over 3 days (assuming no further
storms occur during this period), the additional loading on
400
-------
Table 85. SLUDGE PRODUCTION AND SOLIDS DISPOSAL
METHODS FOR VARIOUS TREATMENT PROCESSES*
Treatment process
Sedimentation
Dissolved air
flotation
Sludge
concentration,
t solids
2.5-5.0b
1.0-2.0
SS removal
efficiency,
1
40-75
40-70
Wet sludge
volume
cf/mil gal.
260-1,000
670-2,300
Dry s.olids
volume,
cf/mil gal.
10-20
10-20
Sludge
disposal
method for
demonstration
projects
Return to
interceptor
Return to
interceptor
Comments
Ultimate disposal with
existing dry weather flow
sludge
Ultimate disposal with
existing dry weather flow
1-4S
sludge
Landfill For 3/4 in. to 1/2 in. bar
spacing
Rotary fine screens
Ultrafine screens and
microstrainers
Filtration 0.4-1.5*
Contact stabilization 0.5-1.5
Trickling filters and 3. 0-10. O8
rotating biological
contactors
Physical -chemical 2.0-5.0
27-34
25-90
50-90
80-95
60-90
80-100
5-10
S-25
1,100-7,500 10-25
1, 800-6, 300£ ' 20-2Sf
200-1,000 15-25
530-1, 700h 20-2Sh
Return to
interceptor
Return to
interceptor
Return to •>
interceptor
Return to
interceptor
Return to
interceptor
Incineration
Ultimate disposal with
existing dry weather flow
sludge
Ultimate disposal with
existing dry weather flow
sludge
Ultimate disposal with
existing dry weather flow
sludge
Ultimate disposal with
existing dry weather flow
sludge
Ultimate' disposal with
existing dry weather flow
sludge
Ultimate disposal by
landfill
a. Assuming 250 mg/1 SS in the stormwater and dry solids specific gravity - 1.30.
b. Assuming continuous sludge collection.
c. NA • not available.
d. Volumes shown for screenings only, not SS.
e. Low value for unsettled backwash water and high value for settled backwash water.
f. Does not include waste biological solids produced in aeration tanks.
g. Assuming sludge recycle.
h. Does not include added chemicals.
Note: cf/mil gal. x 0.0075 - cu m/Ml
in. x 2.54 - cm
the dry-weather facility will be 10,945 kg/day (24,161
Ib/day) SS. This amounts to 4.4 times the design SS
loading. Such SS loadings could easily overload the grit
chambers, primary clarifiers, and digesters, and could re-
sult in broken sludge collector mechanisms and in major
process upsets.
The factors controlling SS removal are the total volume of
wet-weather flow and the flow rate. In the case of the
foregoing example, an increase in primary sedimentation
capacity at existing dry-weather facilities may be required
to enable them to handle the additional SS loadings from
combined sewer overflows adequately or, alternatively, the
401
-------
y6 in°rganic solids may be removed at the
Collection
Equipment used in the collection and removal of solids from
storage/sedimentation basins include rakes, chain and
flights, water jets, and mechanical mixers. The sludge col-
lection rakes, a direct adaptation from circular dry-weather
flow sedimentation practice, are the most expensive^ and the
•? lnactivity °r nonsubmergence have resuUed
°ther
The high-pressure water jets appear to be a favored solution
incorporated in many designs. Location of the nozzles wiJh
respect to the surface to be cleaned (the force diminishes
considerably with distance), easy access to the nozzles fo
- "
mixerf. have been successfully applied in the
caUv ocatedeHenti°n 1*°^ deS^n W' The" stra-
=nft5 A7 l°cated devices have been used to resuspend the
Ultimate Disposal
is landlniT.f Crm°n rlh°d °f ultimate sludge disposal
is landfilling. Several treatment processes are available
for substantially reducing the sludge volume befor^ulUmate
disposal These include heat treating, aerobic and anaero
bic digestion and incineration. All of these processesLe
VOlUmetriC reduction of Budges containing a
c b
te?ianutf^dge^°f 3 biodegradable nature becaue th bac-
teria utilize the organics as an energy source for cell
synthesis. In storm sewer discharges and combined sewer
overflows, the majority of solids are inorganic in nature
"ally ve^y little3.1™ heaUng ValUe and Wifl degradeabior!ogi
To date, no storm flow treatment facility has been ni
specxfically for solids handling and disposal. Generally,
402
-------
as noted in Table 85, the solids removed by storm sewer dis-
charge and combined sewer overflow treatment facilities,
other than those operated as part of the dry-weather flow
treatment facilities, are discharged to the interceptor for
conveyance to the dry-weather plant for removal and ultimate
disposal.
It is apparent that storm flow solids handling and disposal
is indeed an important problem that no longer can be
ignored. The solution of this problem will be different
for each project, but each one will require careful storm
flow quantity and quality characterization studies. Toward
this end, the EPA has recently awarded a contract to study
the handling and disposal of sludges arising from combined
sewer overflow treatment [10]. An attempt is to be made to
define the amounts, constitution, and composition of the
sludges as a function of: (1) the nature of the combined
sewer overflows from which they are derived, and (2) the
treatment processes to which they are subjected. It is
hoped that this identification endeavor will provide the
background data necessary for the selection of appropriate
solids handling and disposal systems for various treatment
processes. The resulting recommendations for solids treat-
ment and disposal methods should be a useful reference for
designers of complete combined sewer overflow treatment and
control systems in the future.
403
-------
-------
Part V
REFERENCES, GLOSSARY,
AND
CONVERSION FACTORS
-------
-------
Section XVI
REFERENCES
Cited References
Section III
1. A Letter Opinion. State of Maryland, Attorney General, Annapolis, Md. April 6, 1971.
2. Combined Sewer Overflow Abatement Alternatives: Washington, D C. Roy F. Weston Inc
11024 EXF. Environmental Protection Agency. August 1970. '
3. Comptroller General of the United States. Need to Control Discharges From Sewers
Carrying Both Sewage and Storm Runoff. Report to the Congress. March 28, 1973.
4. Georgia Water Quality Control Act. Georgia Code Annotated (Supplemental 1966) Ch 17-5
and Ch. 88-26, as Amended by Public Acts of 1971 and Senate Bills 493 and 494 1972
Legislature, Sections 2, 3, 10, 21A. 1972. '
S. Illinois Water Pollution Control Rules. Illinois Pollution Control Board Rules and
Regulations, Chapter 3, Parts I and IV. March 7, 1972.
6. Orange County, Florida, Subdivision Regulations. Orange County Planning Department
(Ed.), Fla. October 1972.
7. Pollution Effects of Stormwater and Overflows From Combined Sewer Systems- A
Preliminary Appraisal. U.S. Department of Health, Education, and Welfare] 1964.
8. Prevention and Correction of Excessive Infiltration and Inflow Into Sewer Systems- A
Manual of Practice. American Public Works Association. 11022 EFF. Environmental
Protection Agency. January 1971.
9. Problems of Combined Sewer Facilities and Overflows 1967. American Public Works
Association. 11020--- 12/67. Environmental Protection Agency. December 1967.
Tr6atment Dat6S ' Mississippi River. Illinois Pollution
11. Rules and Regulations, Sediment Control. State of Maryland, Department of Natural
Resources, 8.05.03.01. Annapolis, Md. April 4, 1972.
12' ^°rUtl?n N?-C7°"?3' Adding Re?olution No. 67-64. Special Time Schedule for the City
and County of San Francisco Relative to Regulation of Discharges From Combined Sewers
California Regional Water Quality Control Board- -San Francisco Bay Region
November 24, 1970.
13. Sediment Control Law. State of Maryland, Annotated Code, Chapter 245, Acts of 1970.
A.p n J. L L j J. y 7 (J •
14. WPC Technical Policy 20-24, Design Criteria--Waste Treatment Plants and Treatment of
Sewer Overflows. Illinois Environmental Protection Agency. Revised July 1971.
407
-------
8.
REFERENCES (Continued)
Section IV
2.
3. Giessner. W-^.^t al.^ Management of Wet Weather Flow. Presented at CWPCA Conference.
4. Metcalf § Eddy, Inc. Reconnaissance
5.
6. U. S
Area! ISJy1™" . "* COm"'erCe- "^ropolitan Climatological Summaries: National Capital
Section V
5- in D-c- R '
7. Combined Sewer Overflow Seminar Papers. Seminar at Hudson-Delaware Basins Office
Agency! Marchlg^. *'*' 1969' U°2°"- °3/70' Environmental Projection
8. Combined Sewer Overflow Study for the Hudson River Conference. Contract 68-01-0055
August ?972 n C°mmission> New York City. Environmental Protection Agency?
9' °th Massachusf ts- Special Report of the Department of Public Health
aratlref S™^* ^64 f" ^ D1SP°Sal °f ^^^ " thg
^ Hercules, inc. 11020-10/69.
11. Dispatching System for Control of Combined Sewer Losses. Metropolitan Sewer Board
St. Paul, Minnesota. 11020 FAQ. Environmental Protection Agency. March 1971? '
408
-------
REFERENCES (Continued)
12. Engineering Investigation of Sewer Overflow Problem: Roanoke, Virginia. Hayes, Seay,
Mattern § Mattern. 11024 DMS. Environmental Protection Agency. May 1970.
13. Environmental Impact of Highway Deicing. Edison Water Quality Laboratory. 11040 GKK.
Environmental Protection Agency. June 1971.
14. FY 1972 Annual Report on the Quality of Urban Storm Runoff Entering the San Francisco
Bay. The Hydrologic Engineering Center of Engineers, U.S. Army. June 1972.
15. J.B. Gilbert § Associates and Metcalf § Eddy, Inc. Evaluation San Francisco Wastewater
Master Plan. Draft report. March 2, 1973.
16. Hypochlorination of Polluted Storm Water Pumpage at New Orleans. U.R. Pontius et al.
11203 FAS. Draft report for Environmental Protection Agency. March 1973.
17. Kluesener, J.W. and Lee, G.F. Nutrient Loading From a Separate Storm Sewer in Madison,
Wisconsin. Bechtel Corporation and University of Wisconsin. Presented at 45th Annual
Conference, Water Pollution Control Federation. Atlanta, Georgia. October 12, 1972.
18. Letter from Robert W. Agnew of Rex Chainbelt, Inc. to Richard Field of Environmental
Protection Agency on CSO Data From Central States Meeting. June 19, 1972.
19. Los Angeles County Flood Control District, Memorandum on Storm Water Sampling Program
from J.K. Mitchell to E.J. Zielbauer. File No. 2-19.012. August 16, 1968.
20. McKee, J.E. Loss of Sanitary Sewage Through Storm Water Overflow. Jour. Boston Soc.
Civ. Engrs, 34, No. 2, p 55. 1947.
21. Metcalf $ Eddy, Inc. American Sewage Practice, Vol. I: Design of Sewers, 2nd Edition.
New York, McGraw-Hill Book Company. 1928.
22. Ministry of Housing and Local Government. Technical Committee on Storm Overflows and
the Disposal of Storm Sewage. London, Her Majesty's Stationery Office. 1970.
23. Onondaga Lake Study. Onondaga County, Syracuse, New York. 11060 FAE. Environmental
Protection Agency. April 1971.
24. Palmer, C.L. the Pollutional Effects of Storm-Water Overflows From Combined Sewers.
Sewage and Industrial Wastes, 22, No. 2, pp 154-165. 1950.
25. Personal communication from K.L. Zippier, Elson T. Killam Associates, Inc. to
W.G. Smith, Metcalf § Eddy, Inc., regarding 413TP - Treatment Plant, Borough of New
Providence. April 26, 1973.
26. Pollutional Effects of Stormwater and Overflows From Combined Sewer Systems: A
Preliminary Appraisal. U.S. Department of Health, Education, and Welfare. 1964.
27. Problems of Combined Sewer Facilities and Overflows 1967. American Public Works
Association. 11020 12/67. Environmental Protection Agency. December 1967.
28. Schmidt, 0. J. Pollution Control in Sewers. Jour. WPCF, 44: No 7 p 1384
July 1972. ' v
29. Sewer Bedding and Infiltration - Gulf Coast Area. Tulane University. 11022 DEI.
Environmental Protection Agency. May 1972.
30. Shifrin, W.G. and Homer, W.W. Effectiveness of the Interception of Sewage-Storm
Water Mixtures. Jour. WPCF, 33, No. 6, pp 650-720. 1961.
31. Storm and Combined Sewer Pollution Sources and Abatement: Atlanta, Georgia. Black,
Crow and Eidsness, Inc. 11024 ELB. Environmental Protection Agency. January 1971
32. Storm Water Management Model, Vol. I: Final Report. Metcalf § Eddy, Inc., University
of Florida, and Water Resources Engineers, Inc. 11024 DOC. Environmental Protection
Agency. July 1971.
33. Storm Water Pollution From Urban Land Activity. Economics Systems Corporation.
11034 FKL. Environmental Protection Agency. July 1970.
34. Storm Water Problems and Control in Sanitary Sewers: Oakland and Berkeley, California.
Metcalf § Eddy, Inc. 11024 EQG. Environmental Protection Agency. March 1971.
409
-------
REFERENCES (Continued)
3S-
' ?ISS oo'bS*?^!?'"" V01' 1: *»Wic.l Studi... University of Cincinnati
38'
39-
Section VI
1.
M
ftgaray, r.j. and Kiado, M.L. Effects of Refri apmteri c*
pps2!o:23?:ocMaydi;ii6of "" 2ut *"*» '"<-«"" ^ste",^^^ ass'K&X'f
2. Berthouex, P.M. Evaluating Economy of Scale. Journal WPCF, 44, No. 11, pp 2111-2119.
3.
4.
5.
6.
reporfo ' ' «»««••». «r- ,
report for Environmental Protection Agency. 1973
tLa^oPrOJeCt- «°* *' "•"«' '«• »»« «X-'
Sader"eL™rchCeCenterate?70f9or--eruoa6^ W""V«.r Treating Processes. Robert A.
1969. Center. 17090--- 06/69. Environmental Protection Agency.
June 1969.
7.
9.
^ , ,__ „.„ _ . D^^
11. uiessnpr w u &+ «i \i . ,. ,,
of Wet Weather Flow. Presented at CWPCA Conference
12. Hydrospace-Challenger, Inc. Sewer Flow Sampler. Preliminary rough draft. 1972.
ission of the City of
13.
14.
15.
16.
410
-------
REFERENCES (Continued)
17. McPherson, M.B. Feasibility of the Metropolitan Water Intelligence System Concept
(Integrated Automatic Operational Control). American Society of Civil Engineers.
December 1971.
18. Metcalf § Eddy, Inc. Reconnaissance Study of Combined Sewer Overflows and Storm
Sewer Discharges. Prepared for District of Columbia. 1973.
19. Metcalf § Eddy, Inc. Wastewater Engineering: Collection, Treatment, Disposal.
New York, McGraw-Hill Book Company. 1972.
20. National Pollution Discharge Elimination System Application for Permit to Discharge,
Appendix A - Standard Analytical Methods (Interim). Federal Laws 71:5598-71:5603.
21. A Portable Device for Measuring Wastewater Flow in Sewers. Hittman Associates.
Contract No. 14-12-909. Draft report for Environmental Protection Agency. May 1973.
22. Schontzler, J.G. Electronic Flow Measurement for Wastewater. Presented at the
WWEMA Industrial Water and Pollution Conference. March 15, 1973.
23. Smith, R. Cost of Conventional and Advanced Treatment of Wastewater. Journal
WPCF, 40, No. 9, pp 1546-1574. 1968.
24. Suspended Solids Monitor. American Standard, Inc. 11020 DZB. Environmental
Protection Agency. August 1972.
25. System Monitoring and Remote Control. Detroit Metropolitan Water Department.
11020 FAX. Draft report for Environmental Protection Agency. November 1972.
26. Underwater Storage of Combined Sewer Overflows. Karl R. Rohrer Associates, Inc.
11022 ECV. September 1971.
27. USEPA Clean Water Fact Sheet: Storm and Combined Sewer Projects. April 1969.
28. Visit to the EPA Regional Office in Chicago, Illinois, and Four of Their On-going
Projects. Memorandum for the Record, C.D. Tonkin, Metcalf § Eddy, Inc.
January 29-February 6, 1973.
Section VII
1. Environmental Impact of Highway Deicing. Edison Water Quality Laboratory. 11040 GKK.
Environmental Protection Agency. June 1971.
2. Everhart, R.C. New Town Planned Around Environmental Aspects. Civil Engineering-ASCE,
43, No. 9, pp 69-73. September 1973.
3. Guidelines for Erosion and Sediment Control Planning and Implementation. Maryland
Department of Water Resources and Hittman Associates, Inc. 15030 FMZ. Environmental
Protection Agency. August 1972.
4. Investigation of Porous Pavements for Urban Runoff Control. The Franklin Institute
Research Laboratories. 11034 DUY. Environmental Protection Agency. March 1972.
5. Mammel, F.A. We are Using Salt--Smarter. Department of Public Works, Ann Arbor,
Michigan, American City, LXXXVII, No. 1, pp 54-56. 1972.
6. Poertner, H.G. Better Storm Drainage Facilities at Lower Cost. Civil Engineering -
ASCE, 43, No. 10, pp 67-70. October 1973.
7. Sartor, J.D. and Boyd, G.B. Water Pollution Aspects of Street Surface Contaminants.
EPA-R2-72-081. Environmental Protection Agency. November 1972.
8. Sartor, J.D. , Boyd, G.B. , and Agardy, F.J. Water Pollution Aspects of Street Surface
Contaminants. Presented at the 45th Annual Conference of the Water Pollution Control
Federation. Atlanta, Georgia. October 1972.
9. Sediment Movement in an Area of Surburban Highway Construction, Scott Run Basin,
Fairfax County, Virginia, 1961-64. Geological Survey Water-Supply Paper 1591-E.
411
-------
REFERENCES (Continued)
or Fun and Profit. Water Spectrum, 2, No. 3, pp 29-34,
11.
12.
13. Fi«
A^cy7JMay'/i973?1SOn WSter QU3llty ReS6arch Laboratory, Enviro^mentaTpro^ction
Section VIII
1.
2.
ASCE
3.
and
:, 40,
»„* c*-4- « T Poilution From Combined Sewe Cincinnati, Ohio
1090S°UrCeS Englneerin« Meeting, Memphis,
, __.j *»«»x* *iv/v^iicm» i»i . r*. . .^an ^T»or»x»i*-*-rt \g ~ — *. •*-»•• <• _ _ " ~ - r -I • \^ • «J i •
ment: Preliminary Report.
7.
i, B.C. Roy F.
8.
^_ i4,puil. i(J1 environment Protection Agency 1973
9.
'
10.
11.
12.
13.
14.
IS.
Agency. September~1973".' ""^ *"*my Kesearcn Laboratory, Environmental Protection'
16.
and
17. EP
412
-------
REFERENCES (Continued)
18. Feasibility of a Stabilization-Retention Basin in Lake Erie at Cleveland, Ohio.
Havens and Emerson. 11020 05/68. Environmental Protection Agency. May 1968.
19. Field, R. Combined Sewer Overflows. Civil Engineering-ASCE, pp 57-60.
February 1973.
20. Flood Control Coordinating Committee. Development of a Flood and Pollution Control
Plan for the Chicagoland Area: Summary of Technical Reports. Metropolitan Sanitary
District of Greater Chicago, Institute of Environmental Quality, and Department of
Public Works. August 1972.
21. Flow Augmenting Effects of Additives on Open Channel Flows. Columbia Researcy
Corporation. EPA-R2-73-238. Environmental Protection Agency. February 1973.
22. A Flushing System for Combined Sewer Cleansing. FMC Corporation. 11020 DNO.
Environmental Protection Agency. March 1972.
23. Gibbs, C.V., et al. System for Regulation of Combined Sewage Flows. ASCE San.J., 98,
SA-6, pp 951-972. 1972.
24. Greeley, S.A. Marston, F.A., and Requardt, G.J. (Board of Engineers). Report to
District of Columbia, Department of Sanitary Engineering,on Improvements to Sewerage
System. February 28, 1957.
25. Ground Water Infiltration and Internal Sealing of Sanitary Sewers. Montgomery County
Sanitary Department, Montgomery County, Ohio. 11020 DHQ. Environmental Protection
Agency. June 1972.
26. Harlan, T.S. and Allman, W.B. Polyethylene Pipe Slipped Into Defective Sanitary
Sewer. Civil Engineering, 43, No. 6, pp 78-81. 1973.
27. Heat Shrinkable Tubing as Sewer Pipe Joints. The Western Company. 11024 FLY.
Environmental Protection Agency. June 1971.
28. Impregnation of Concrete Pipe. Southwest Research Institute. 11024 EQE.
Environmental Protection Agency. June 1971.
29. Improved Sealants for Infiltration Control. The Western Company. 11020 DIH.
Environmental Protection Agency. June 1969.
30. Maximizing Storage in Combined Sewer Systems. Municipality of Metropolitan Seattle.
11022 ELK. Environmental Protection Agency. December 1971.
31. Metcalf § Eddy, Inc. Reconnaissance Study of Combined Sewer Overflows and Storm
Sewer Discharges. Prepared for District of Columbia. 1973.
32. Metcalf § Eddy, Inc. Wastewater Engineering: Collection, Treatment, Disposal.
New York, McGraw-Hill Book Company. 1972.
33. Microstraining and Disinfection of Combined Sewer Overflows. Cochrane Division,
Crane Company. 11023 EVO. Environmental Protection Agency. June 1970.
34. Oakley, H.R. Practical Design of Storm Sewage Overflows. Symposium on Storm
Sewage Overflows. London, Institution of Civil Engineers. 1967.
35. Palmer, C.L. Feasibility of Combined Sewer Systems. Journal WPCF, 35, No. 2,
pp 162-167. 1963.
36. Personal Communication from Dallas Water Utilities to W.G. Smith, Metcalf § Eddy, Inc.,
regarding site plan and break-down of costs for the Bachman Polymer Injection Station.'
March 9, 1973.
37. Personal communication from M.L. Robins of Metropolitan Sewer Board, St. Paul,
Minnesota, to J.A. Lager, Metcalf $ Eddy, Inc., regarding Minneapolis-St. Paul's
combined sewer overflow control program. February 2, 1973.
38. Personal Communication from Paul L. Brunner, Fort Wayne City Utilities, to W.G. Smith,
Metcalf § Eddy, Inc. regarding cost data for stormwater demonstration. Grant EPA-
11020 GYU. May 8, 1973.
413
-------
REFERENCES (Continued)
»«•»«• Environmental
• A
Overflows 1967. American Public Works
"-1 Protection Agency. December 1967
43.
__.... ^..r, _,„ - . Motions Applied to Storm Sewage
" •> • «xwis j.41^ . n.ij. ft* r»i nn fl i Ttrnmnn -«y*n4--;MM -t-^urr' c» «^.»_ &« . «
to W.G. Smith, Metcalf 5 Eddy, Inc., regarding
46.
'. 11022 DEI.
May
47.
Ho* i-» i. v j, ate cinu iT~f»aTm*»r»T r^-h f*-*mK-i*.Aj o_. OVA -f 1
49.
50.
51.
ECV. Environmental Protection
Section IX
2-
:. Proceedings of A
6. Devenis, K. Cottage Farm Stormwater Treatment Station. Public Works, November 1971
7.
414
-------
REFERENCES (Continued)
8. Field, R. and Struzeski, J. Management and Control of Combined Sewer Overflows.
Presented at 36th Annual Sewage Works Operators' Conference. Springfield, 111.,
April 21, 1971. In WPCF Journal, 44, No. 7, pp 1393-1415.
9. Flood Control Coordinating Committee. Development of a Flood and Pollution Control
Plan for the Chicagoland Area: Summary of Technical Reports. Metropolitan Sanitary
District of Greater Chicago, Institute of Environmental Quality, and Department of
Public Works. August 1972.
10. Foerster, R.E. The Spring Creek Combined Sewer Overflow Project. Greeley and Hansen,
Engineers. Presented at the New York State Water Pollution Control Association Annual
Meeting, January 28, 1972.
11. Hydraulics of Long Vertical Conduits and Associated Cavitation. St. Anthony Falls
Hydraulic Laboratory. 11034 FLU. Environmental Protection Agency. June 1971.
12. Letter from Andrew Warren to Tony Tafuri on bid cost of Spring Creek Project.
September 1972.
13. Letter from D.R. Zwickey to Commissioner of Public Works, City of Milwaukee on bidder
costs of Humboldt Avenue Pollution Abatement Demonstration Project. July 10, 1969.
14. Maximizing Storage in Combined Sewer Systems. Municipality of Metropolitan Seattle.
11022 ELK. Environmental Protection Agency. December 1971.
15. Metropolitan Water Intelligence Systems Completion Report-Phase 1. Colorado State
University, Department of Civil Engineering. U.S. Department of the Interior, Office
of Water Resources Research, Grant No. 14-31-0001-3410. June 1972.
16. Poertner, H.G. Better Storm Drainage Facilities at Lower Cost. Civil Engineering-
ASCE, 43, No. 10, pp. 67-70. October 1973.
17. Rohrer, K.R. Assoc, Inc. Demonstration of Void Space Storage with Treatment and
Flow Regulation. EPA Project 11020 DXH.
18. Storage and Treatment of Combined Sewer Overflows. C. C. Oster. Lake Superior Basin
Office-EPA, Chippewa Falls, Wis. Environmental Protection Agency. October 1972.
19. Underwater Storage of Combined Sewer Overflows. Karl R. Rohrer Associates, Inc.
11022 ECV. Environmental Protection Agency. September 1971.
Section X
1. American Petroleum Institute. Manual on Disposal of Refinery Wastes, Volume on Liquid
Wastes, Chapter 5, Oil-Water Separator Process Design. 1969.
2. American Petroleum Institute. Manual on Disposal of Refinery Wastes, Volume and Liquid
Wastes, Chapter 9, Filtration, Flocculation, Flotation. 1969.
3. Babbit, H.E. Sewerage and Sewage Treatment, 8th Ed. New York, John Wiley § Sons.
1958.
4. Backman Stormwater Treatment Plant: Monthly Report. City of Dallas, Texas. 11020 FAW.
Environmental Protection Agency. January 1973.
5. Bachman Stormwater Treatment Plant: Monthly Report. City of Dallas, Texas. 11020 FAW.
Environmental Protection Agency. November 1972.
6. Bodien, D.G. and Stenberg, R.L. Microscreening Effectively Polishes Activated Sludge
Plant Effluent. Water § Wastes Engineering, 3, No. 9, pp 74-77. 1969.
7. Cohen, J.M. Suspended and Colloidal Solids Removal. PPB 1703, EPA Internal Communi-
cation. June 19/0.
8. Combined Sewer Overflow Seminar Papers. Edison Water Quality Research Laboratory.
EPA 670/2-73-077. Environmental Protection Agency. September 1973.
9. Conceptual Engineering Report - Kingman Lake Project. Roy F. Weston, Inc. 11023 FIX.
Environmental Protection Agency. August 1970.
415
-------
REFERENCES (Continued)
10.
11.
12.
13.
14.
16.
17.
18.
Hercules, Inc. 11020- 10/69.
«t, of Portland,
^
^fi;: ^Strainin* of C°"">-ed Sewer Overflows. Journal
1S- ^Kj-icSE"1?:; ^^inZr^1 ?issord Air Fiotati°- '«'»«•••
environmental Protection Agency. July 1971.
:0^^^^
^
19.
21.
Effluents. Cost
22' l?n5/FY? Fijtration of Combined Sewer Overflows. Hydrotechnic Corporation
11023 EYI. Environmental Protection Agency. April 1972. Corporation.
23. Hulme. All You See is the Stream. The American City, pp 77. 1970.
24' ppei317^i333.ali972iltering C°mbined Sewer Overflows. Journal WPCF, 44, No. 7,
25.
26.
27.
29.
3°' EewTaiiaSereandTEiC?lnvP Tegfarrdi^ desW equations for microstrainers with
t.«Li. Diaper and G.E. Glover of Cochrane Division of Crane Company. February 21,
31. Personal ohone rommnnirrat i nn va,,-,,-,!; „„ ^-_ ,
characteristics of microstrainers
of Crane Company. February 20,
416
-------
REFERENCES (Continued)
32. Planning Drawings for the Wastewater Collection System and Treatment Facility at
Mount Clemens. Spalding, DeDecker § Associates, Inc. January 1973.
33. Post'Construction Evaluation Plan, Mount Clemens, Michigan. Spalding, DeDecker §
Associates, Inc., 1973.
34. Rapid-Flow Filter for Sewer Overflows. The Rand Development Corporation. Contract
No. WA 67-2. Environmental Protection Agency. August 1969.
35. Reclamation of Water. Water Pollution Research, pp 120-123. 1966.
36. Rock Tunnel Summaries. Data for Chicago's Deep Tunnel. September 1970.
37. Roesler, J.F. Preliminary Design of Surface Filtration Units (Microscreening).
Environmental Protection Agency. June 1969.
38. Rotary Vibratory Fine Screening of Combined Sewer Overflows. Cornell, Howland,
Hayes and Merryfield. 11023 FDD. Environmental Protection Agency. March 1970.
39. Screening/Dissolved-Air Flotation Treatment of Combined Sewer Overflows. Envirex,
Inc., for Combined Wastewater Systems Training Course, sponsored by EPA. November 1972-
February 1973.
40. Screening/Flotation Treatment of Combined Sewer Overflows. Rex Chainbelt, Inc.
11020 FDC. Environment Protection Agency. January 1972.
41. Strainer/Filter Treatment of Combined Sewer Overflows. Fram Corporation. 11020 EXV
Environmental Protection Agency. July 1969.
42. Ultrasonic Filtration of Combined Sewer Overflows. American Process Equipment
Corporation. 11023 DZF. Environmental Protection Agency. June 1970.
43. Visit to the EPA Regional Office in Chicago, Illinois, and Four of Their On-Going
Projects. Memorandum for the Record, D.C. Tonkin, Metcalf § Eddy, Inc. January 29-
February 6, 1973.
Section XI
1. An Evaluation of Three Combined Sewer Overflow Treatment Alternatives. Clark, Dietz
and Associates-Engineers, Inc. 11020 FAM. Draft report for Environmental Protection
Agency. October 1972.
2. Antonie, R.L. Response of the Bio-Disc Process to Fluctuating Wastewater Flows
Presented at the 25th Purdue Industrial Waste Conference. May 5-7, 1970.
3. Antonie, R.L. and Van Aacken, K. Rotating Discs Fulfill Dual Wastewater Role Water
and Wastes Engineering, 8, No. 1, pp 37-38. January 1971.
4. Autotrol Corporation. Bio-Surf Design Manual. 1972.
5. Canter, L.W., Englande, A.J., Jr., and Mauldin, A.F., Jr. Loading Rates on Waste
Stabilization Ponds. ASCE San. Eng., 95, No. SA 6, pp 1117-1129. 1969.
6. Clark, J.W. and Viessman, W., Jr. Water Supply and Pollution Control. Scranton
International Textbook Company. 1965.
7. Combined Sewer Overflow Treatment by the Rotating Biological Contactor Process.
Autotrol Corporation. Environmental Protection Agency. September 1972.
8. Combined Wastewater Collection and Treatment Facility, Mount Clemens, Michigan.
V.W. Mahida. Spalding, DeDecker § Associates, Inc. Presented at 4th Annual
Conference, Water Pollution Control Federation. San Francisco, California
October 3-8, 1971.
9. Design, Construction, Operation and Evaluation of a Demonstration Waste Treatment
Device Termed the "Rotating Biological Contactor". Allis-Chalmers. Contract No.
14-12-24. Environmental Protection Agency. August 1969.
417
-------
REFERENCES (Continued)
10.
March 1967~
11.
, E.W.J. Oxidation Pond Effluent Improvement. Presented at'the 49th Texas
and^Sewage Works Association's Short School. College Station, Texas.
lia in Wastewater Oxidation Ponds.
• 1972.
12. Gloyna, E.F. Basis for Stabilization Ponds Designs. Water Quality Improvement-
Hater Resources Symposium No. 1, pp 397-408. Austin, University of Texas Press! 1968,
13. Gloyna, E.F. Waste Stabilization Ponds. Geneva, World Health Organization. 1971.
14. _
^ „„„. ^IL f _ and
15.
' SpaSy!"' lati. E"8in«ri"8 Management of Water Quality. New York, McGraw-Hill Book
' C""'"1™. *""««. Disposal.
1 y « i^i^uiidj. LUiiiiniiiiicjirinn -i- vrmi UQVTI r\ •*-,-! T« « *. ~ v i _ *.. . ,, ._ .
Utility regarding EPA Grant
— ->s Report No. 43.
20.
-L* ZiPPler' Elson T- Killam Associates, Inc. to W G
' Borough'f "£ "'
^ J- • * wj i, \-.vjioLiULLJ_un r. VHiiiarinn \J \ ar» M/-tn^i +> r"i«.«« »« • _ i • ,-, - , .
Spalding, DeDecker
22.
23. Rex Chainbelt, Inc. Kenosha Biosorption Project. January 1973.
24. Siddiqi, R.H. et al. The Role of Enzymes in the Contact Stabilization Process.
Advances in Water Pollution Research, Vol. 2, pp 353-366. Water Pollution Control
Federation. 1967.
25. Thirumurthi, D. Design Principles of Waste Stabilization Ponds. ASCE San. Ene 95
No. SA 2, pp 311-330. 1969. '
26. Torpey, W.N. et al. Rotating Discs with Biological Growths Prepare Wastewater for
Disposal or Reuse. Journal WPCF, 43, No. 11, pp 2181-2188. November 1971.
27. Visit to the EPA Regional Office in Chacago, Illinois, and Four of Their On-Going
Projects. Memorandum for the Record, C.D. Tonkin, Metcalf § Eddy, Inc. January 29-
February 6, 1973.
28. Walters, C.F. et al. Microbial Substrate Storage in Activated Sludge. ASCE San
Eng., 94, No. SA 2, pp 257-269. 1968.
29. Waste Treatment Lagoons--State of the Art. Missouri Basin Engineering Health Council.
17090 EHX. Environmental Protection Agency. July 1971.
418
-------
REFERENCES (Continued)
Section XII
1. Anooshian, A. A., Jenks , J., and Agardy, F.J. Fluidized Bed Sludge Incineration
System- Review and Analysis, Phase 1. Estero Municipal Improvement District, Foster
City, California. October 1970.
2. Bishop, D.F. et al. Physical-Chemical Treatment of Municipal Wastewater. Journal
WPCF, 44, No. 3, pp 361-371. 1972.
3. City and County of San Francisco. Geissner, W.R., Cockburn, R.T. , Coffee, H.C. Jr.,
Moss, F.H. Jr., and Noonan, M.E. San Francisco Master Plan for Waste Water Manage-
ment: Preliminary Report. Department of Public Works. September 1971.
4. Cohen, J.M. Dissolved Refractory Organics. PPB 1702, EPA Internal Communication.
June 1970.
5. Cohen, J.M. Suspended and Colloidal Solids Removal. PPB 1703, EPA Internal Communi-
cation. June 1970.
6. Conceptual Engineering Report - Kingman Lake Project. Roy F. Weston, Inc. 11023 FIX.
Environmental Protection Agency. August 1970.
7. Conley, W.R. and Hsiung, K. Design and Application of Multimedia Filters. Journal
AWWA, 61, p 97. February 1969.
8. Cost and Performance Estimates for Tertiary Wastewater Treating Processes. Robert A
Taft Water Research Center. 17090--- 06/69. Environmental Protection Agency.
7
June 1969.
Gulp, R.L. and Gulp, G.L. Advanced Wastewater Treatment. New York, Van Nostrand-
Reinhold Company. 1971
£o!P10' sJ-sT"^?8'"1611' °f ^ Sewage-2' Water and Wastes Engineering, 4,
11. Estimating. Costs and Manpower Requirements for Conventional Wastewater Treatment
* 17°9° DAN- Envi™nmental Protection Agency
FMC CorPo-tion- 17050 DAL. Environmental
' H'J< Dallas WateT Utilities. Journal AWWA, 64, No. 10, pp No. 638-641.
14. Kugelman I.J. and Cohen, J.M. Chemical-Physical Processes. Presented at the
Advanced Waste Treatment and Water Reuse Symposium, Cleveland, Ohio. March 1971.
15' iafcaJfJ*Apda;d fSth'y!;S; ^em?Sa"dum f°r the Record-Dallas Inspection Trip.
Inc. March 16, 1973.
16. Metcalf S Eddy, Inc. Report to District of Columbia, Department of Sanitary
Engineering Washington. D.C., on Comparative Evaluation of Advanced Waste Treatment
17. Metcalf 5 Eddy, Inc. Report of District of Columbia, Department of Sanitarv
Engineering, Washington, D.C., on Comparative Evaluations of Secondary and Advanced
Wastewater Treatment for the Water Pollution Control Plant. July 12 -" AdVanced
Disposal.
20. Process Design Manual for Suspended Solids Removal. Burns and Roe, Inc. 17030 GNO
Environmental Protection Agency Technology Transfer. October 1971.
21. Process Design Manual for Carbon Adsorption. Swindell-Dressier Co. 17020 GNR
Environmental Protection Agency. October 1971. A/UZU I>NK.
419
-------
REFERENCES (Continued)
22.
23.
24.
25.
2-
10.
11.
12.
13.
14.
15.
16.
S«.g.. B.ttell. Northwest
Techni
-------
REFERENCES (Continued)
Section XIV
1. Battelle Memorial Institute. Evaluation of Mathematical Models for Engineering
Assessment, Control, Planning, and Design of Storm and Combined Sewerage Systems.
Project No. CI-73-0070. Environmental Protection Agency.
2. Battelle Pacific Northwest Laboratories. A Mathematical Model for Optimum Design
and Control of Metropolitan Wastewater Management Systems. April 1973.
3. City and County of San Francisco. Geissner, W.R., Cockburn, R.I., Coffee, H.G. Jr.,
Moss, F.H., Jr., and Noonan, M.E. San Francisco Master Plan for Waste Water Manage-
ment: Preliminary Comprehensive Report. Department of Public Works. September 1971.
4. Combined Sewer Overflow Abatement Alternatives: Washington, D.C. Roy F. Weston,
Inc. 11024 EXF. Environmental Protection Agency. August 1970.
5. Conceptual Engineering Report - Kingman Lake Project. Roy F. Weston, Inc.
11023 FIX. Environmental Protection Agency. August 1970.
6. Crawford, N.H., and Linsley, R.K. Digital Simulation in Hydrology: Stanford
Watershed Model IV. Stanford University, Dept. of Civil Engineering Technolocv
Report 39. 1966. b/
7. Dispatching System for Control of Combined Sewer Losses. Metropolitan Sewer Board,
St. Paul, Minnesota. 11020 FAQ. Environmental Protection Agency. March 1971.
8. Evaluation San Francisco Wastewater Master Plan. J. B. Gilbert § Associates.
Sacramento, California. May 1973.
9. Flood Control Coordinating Committee. Development of a Flood and Pollution Control
Plan for the Chicagoland Area: Summary of Technical Reports. Metropolitan Sanitary
District of Greater Chicago, Institute of Environmental Quality, and Department of
Public Works. August 1972.
10. Giessner, W.R., et al. Management of Wet Weather Flow. Presented at CWPCA
Conference, San Diego. May 10, 1973.
11. Horsefield, D.R. The Deep Tunnel Plan for the Boston Area. Journal of the Boston
Society of Civil Engineers, 55, No. 4. October 1968.
12. Improvements to the Boston Main Drainage System. Camp, Dresser, 5 McKee. Report to
the City of Boston. September 1967.
13. Lager, J.A. and Field, R. Counter Measures for Pollution From Overflows -- The State
of the Art. Presented at Water Pollution Control Federation 46th Annual Conference
Cleveland, Ohio. September 30-October 5, 1973.
14. Maximizing Storage in Combined Sewer Systems. Municipality of Metropolitan Seattle.
11022 ELK. Environmental Protection Agency. December 1971.
15. Metcalf § Eddy, Inc. Reconnaissance Study of Combined Sewer Overflows 2nd Storm
Sewer Discharges. Prepared for District of Columbia. 1973.
16. Paved Area Runoff. Illinois State Water Survey. 11030 FLN. Environmental
Protection Agency. July 1972.
17. Pianno, F.R., Anderson, D.J., and Derbin, S. The Use of Mathematical Modeling to
Evaluate Sewer Design and Operation. City of Cleveland exhibit handout, WPCF 46th
Annual Conference. Cleveland, Ohio. September 30-October 5, 1973.
18. Ritter, F.G., and Warg, C. Upgrading City Sewer Installations. Engineering Digest.
April 1971.
19. Storm Water Management Model, Vol. I, Final Report. Metcalf § Eddy, Inc., University
of Florida, and Water Resources Engineers, Inc. 11024 DOC. Environmental Protection
Agency. July 1971.
20. Storm Water Management Model, Vol. II, Verification and Testing. Metcalf § Eddy, Inc.,
University of Florida, and Water Resources Engineers, Inc. 11024 DOC. Environmental
Protection Agency. August 1971.
421
-------
REFERENCES (Continued)
'
Section XV
1.
Stores
2.
L, B.C.
4.
...„- ^..^. .,_,-,...,-, ^r,,™ _ _.._ ..... ... Conference.
-/ ~.. ..I* viimwiii.a.1. nutettion Agency. January 1973.
6.
ic Filtration In Treating the Overflows
Associates, Inc. 11020--- 09/67.
, nneso a. 020 FAQ. Environmental Protection Agency. March 1971. '
9' Dlssolved Air "~"X^^:ra^^^ra «;-*«*
10.
11.
Inspection Trip.
12.
13. . _ „...„.„
:ility, Mount Clemens,
Pollution Control
14. Maxim: „ ^.w^,.™^^ ... . ....,.,,„,.„ ^ _ ^
of Metropolitan Seattle.
15. IK oiiu RemnTP ,.nnrm,
_.^ ^ 11020
422
-------
REFERENCES (Continued)
Uncited References
Abstracts
1. Storm Water Runoff from Urban Areas: Selected Abstracts of Related Topics. Robert A.
Taft Sanitary Engineering Center. Environmental Protection Agency. April 1966.
2. Selected Urban Storm Water Runoff Abstracts. The Franklin Institute Research Laboratories
11020 DES 06/69. Environmental Protection Agency. June 1969.
3. Selected Urban Storm Water Runoff Abstracts. The Franklin Institute Research Laboratories
Contract No. 14-12-467. Environmental Protection Agency. January 1969.
4. Selected Urban Storm Water Runoff Abstracts: July 1968-June 1970. The Franklin Institute
Research Laboratories. 11024 EJC. Environmental Protection Agency. July 1970.
5. Selected Urban Storm Water Runoff Abstracts: First Quarterly Issue. The Franklin
Institute Research Laboratories. 11024 EJC. Environmental Protection Agency. October
1970.
6. Selected Urban Storm Water Runoff Abstracts: Second Quarterly Issue. The Franklin
Institute Research Laboratories. 11024 EJC. Environmental Protection Agency. January
7. Selected Urban Storm Water Runoff Abstracts: Third Quarterly Issue. The Franklin
Institute Research Laboratories. 11024 FJE. Environmental Protection Agency. April 1971.
8. Selected Urban Storm Water Runoff Abstracts: June 1970-June 1971. The Franklin Institute
Research Laboratories. 11024 FJE. Environmental Protection Agency. July 1971.
9. Selected Urban Storm Water Runoff Abstracts: July 1971-June 1972. The Franklin Institute
Research Laboratories. 11020 HMM. Environmental Protection Agency. September 1972.
10. Selected Urban Storm Water Runoff Abstracts: July 1971-June 1972. The Franklin Institute
Research Laboratories. Environmental Protection Agency Municipal Technology Branch.
11020 HMM. Environmental Protection Agency. December 1972.
11. Storm and Combined Sewer Pollution Control Program: Progress Report. Edison, New Jersey
Water Quality Laboratory. Environmental Protection Agency. September 1970. '
12. Storm and Combined Sewer Pollution Control Program Reports: Research, Development, and
Demonstration Grant, Contract, and In-House Project Reports. Edison, New Jersey Water
Quality Laboratory. Environmental Protection Agency. June 1972.
13. Storm and Combined Sewer Pollution Control Program Reports: Research, Development, and
Demonstration Grant, Contract, and In-House Project Reports. Edison, New Jersey, Water
Quality Laboratory. Environmental Protection Agency. January 1972.
14. Storm and Combined Sewer Demonstration Projects. Environmental Protection Agency
August 1969. '
IS. Storm and Combined Sewer Demonstration Projects. Environmental Protection Agency
January 1970. ' '
16. Field, R. and Weigel, P. Urban Runoff and Combined Sewer Overflow. Journal WPCF, 45.
No. 6, pp 1108-1115. 1973.
Biological Treatment
1. A Literature Search and Critical Analysis of Biological Trickling Filter Studies-Vol. I.
The Dow Chemical Company. 17050 DDY. Environmental Protection Agency. December 1971.
2. A Literature Search and Critical Analysis of Biological Trickling Filter Studies-Vol. II.
The Dow Chemical Company. 17050 DDY. Environmental Protection Agency. December 1971.
3. Feasibility Studies of Applications of Catalytic Oxidation in Wastewater. Southern
Illinois University. 17020 ECI. Environmental Protection Agency. November 1971.
423
-------
REFERENCES (Continued)
4. Design Guides for Biological Wastewater Treatment Processes. The City of Austin Texas
Augus^^l*01" ReS6arCh ln Water Resources' "010 ESQ. Environment Protection IgeJcy
?^hnj EK S65!!11! ?f ExPerimen*s by EWAG with the Rotating Biological Filter Bids
Technische Hochschule, Zurich-Forthildungskurs der EWAG. 1964. 'liter, tidg.
5' E
••
Disinfection
An Investigation of Light-Catalyzed Chlorine Oxidation for Treatment of Wastewater
DeSe 3 Institute« 1702° (14-12-72). Environmental Protect?on Agency? '
"'ewater Disinfection in
4* ppeR162-R170.' mf1"6111 °f Urban Stormwater Runo£f- Journal WPCF, 40, No. 5', Part 2,
5. Glover, G.E. High Rate Disinfection of Combined Sewer Overflow. Presented at
Combined Wastewater Systems Training Course sponsored jointly by the N Y State
Department of Environmental Conservation and Environmental Protection Aeency
Rochester. November 29, 1972. 7"
6. Glover, G.E. Problems in Obtaining Adequate Sewage Disinfection. ASCE
Engineering, 98, No. 4, pp 671-673. 1972.
StUdi" °f Chl°-e Contact
8'
Sons 1958
awT; G*n".^yer? J'C'i and Okun' D'A- Water and Wastewater Engineering: Volume
ons 1958 1Ca" Wastewater Treatment and Disposal. New York, John Wiley §
cfflca^orjoration? P[??"ti6S' TransPortation Equipment, and Safe Handling. Allied
12' pp°893-894.f°1970rtiary Treatment' Environmental Science § Technology, 4, No. 11,
13. Pearce, L. (Chairman) Chlorination in Sewage Disposal. Report of APHA Committee
on Sewage Disposal. Printed by Wallace $ Tiernan Co., Inc. 1933.
14. Proceedings of the National Specialty Conference on Disinfection. American Society
of Civil Engineers. University of Massachusetts. Amherst. July 8-10, 1970.
1S* wlctnn5 ?eSigni?nonaJMn°r "PSradin8 Existing Wastewater Treatment Slants. Roy F.
Weston, Inc. 17090 GNQ. Environmental Protection Agency. October 1971.
16' ?IiSi,J:E' Treatment of Combined Sewer and Storm Water Overflows. Progress report
to O'Brien $ Gere Engineers, Inc., Syracuse, N.Y.; July 10, 1973.
424
-------
REFERENCES (Continued)
17. Struzeski, E. J. , Jr. Current Status on Control and Treatment of Storm and Combined
Sewer Overflows. Presented at the EPA Technology transfer Program, Design Seminar
on Waste Water Treatment Facilities. Boston. May 26-27, 1971?
Treat"ent *«*"*, Saginaw, Michigan.
19. pzonation in Sewage Treatment. University Extension, University of Wisconsin.
November 9-10, 1971.
20. Pavia, E.H. and Powell, C.J. Stormwater Disinfection at New Orleans. Journal WPCF. 41.
No. 4, pp 591-606. 1969.
21. Post Construction Report: Storm Detention and Chlorination Station. The Commonwealth
of Massachusetts, Metropolitan District Commission. 11020 FAT (7-Mass-l). Draft
report for Environmental Protection Agency. March 1972.
22. Spring Creek Marinac Pollution Control Project- -Report on Use of Hypochlorite for
Treatment of Storm Overflows. Greeley ft Hansen, Engineers. New York, N.Y. March 1964.
Dissolved Air Flotation
1. Development of a Flocculation- Flotation Concept for Solids Separation in Storm Sewer
Agency!' April"!"!. "^ U023 EYC' Interim rePort *»r Envi?on»entSl Protection
2' Sif«MyedDAiirJ1°tu*ion'*A?pe^ixJAi Phase !- 'Pro-Construction Studies on Quality and
SlSni J ReJatl°nshlPs ?f Combined Sewage Flows and Receiving Water Studies^ Oute?
r *"** N°' ^~2^^' Environmental
3. Gupta, M.K. Feasibility Studies for the Upgradation of Flotation Treated Combined
March ?973 P°rt the DePartment of Natu"l Resources. Madison, Wisconsin.
4< Si«i'«.V; aSd LSCyi Y'P- Air Flotatio" Pilot Plant Operation at Niagara Falls.
Presented to New York Sewage and Industrial Wastes Association. 1960.
5* l~ rK!-°K Scre«»ing/D"solved-Air Flotatipn for Treating Combined Sewer Overflows.
?hica£o June IT^ f5?nntep " Sy»POsium on Storm and Combined Sewer Overflows?
Lfticago. June 22-23, 1970. Report for Environmental Protection Agency.
Drainage Area Control
1. A Search: New Technology for Pavement Snow and Ice. Control. Edison Water Quality
Research Laboratory. Z-800615. Environmental Protection Agency. December 1972.
2. Chiang, S.L. A Crazy Idea on Urban Water Management. Water Resources Bulletin. 7
No. 1, pp 171-174. 1971. * '
3. Cleveland, J.G. , Reid, G.W. , and Walters, P.R. Storm Water Pollution From Urban Land
Activity. Presented at ASCE Annual and Environmental Meeting. Chicago. October 15-
17, 1969. Preprint 1033.
4. Dawdy, D.R. Knowledge of Sedimentation in Urban Environments. ASCE Journal Hydraulic
Division, 93, No. HY 6, pp 235-245. 1967.
5. Guy, H.P. Research Needs Regarding Sediment and Urbanization. ASCE Journal Hydraulic
Division, 93, No. HY 6, pp 247-254.
6. Peters, G.L. and Troemper, A. P. Reduction of Hydraulic Sewer Loading by Downspout
Removal. Journal WPCF, 41, No.. 1, pp 63-81. 1969.
7. Reduction of Ground-Water Infiltration Into Sewers by Zone Pumping at Meridian, Idaho.
Hoffman 5 Fiske Consulting Engineers. Grant 29-IDA-2. Environmental Protection Agency.
June 1969. * '
425
-------
REFERENCES (Continued)
Remus, G.J. Keep Street Debris Out of the Sewers. The American City. D 52
December 1968. /»*»•"••
Request for Proposal Related to the Environmental Impact of Highway Deicing. Edison
Water Quality Research Division, Environmental Protection Agency. February 24, 1972.
93*eNo?'sA'5* En^™nmenJ{[J7Ef£ects of Highways. ASCE Sanitary Engineering Division
11. Urban Soil Erosion and Sediment Control. National Association of Counties Research
Foundation. 15030 DTL. Environmental Protection Agency. May 1970.
12. Wolman, M.G., and Schick, A.P. Effects of Construction on Fluvial Sediment, Urban
and Suburban Areas of Maryland. Water Resources Research, 3, No. 2, pp 451-464. 1967
Filtration
10
§'
'• srss-
*' "' li?o. Ftlter'Mllt)' «»d Ml«05cr..n.r Desi«n. Journ.! WPCF. 42. No.U, pp 1944
" Tri"e"* """• *SCB S.nl«ry Journ.1. 94. SA:l,
Mathematical Modeling
JJSVit .
Sewers. University of Minnesota, Memorandin No. M-118. 1968
of urban
sr.
426
-------
REFERENCES (Continued)
10' il!LEfi7CCtSv!!£ Urba"izfion °n Unjt Hydrographs for Small Watersheds, Houston, Texas,
R««Jcii. ^JIib.J'im?1 N°' "-O^0001'158'' «•«. Office of Water Resources
lo!UEfi7eCI™0f,,yrbaninatio2 on.Unit Hydr°g"Phs for Small Watersheds, Houston, Texas,
i?5J ii JPPn2*-XeS:*Da.ta Coi"Pilation, Volume 2. Tracor. Contract No. 14-01-0001-
1580. U.S. Office of Water Resources Research. September 1968.
12. iucKer. L.S. Availability of Rainfall-Runoff Data for Sewered Drainage Catchments.
Society of Civil Engineers. Technical Memorandum No.8. 1969.
and Cost
Physical -Chemical Treatment
l' ?™?npari???,co0f ExPanded"Bea and Packed-Bed Adsorption Systems. FMC Corporation.
17020 — 12/68. Environmental Protection Agency. December 1968.
2. Advanced Wastewater Treatment Using Powdered Activated Carbon in Recirculatine Slurry
'"01"1^'"* InfilC0' 17020
-------
REFERENCES (Continued)
?*£' 1970.' 6t a1' Physiocheraical Treatment of Wastewater. Journal WPCF, 42, No. 1,
4-
Dow
—..-..w.,,..^ni.e»i i-iuueciion Agency. September 1970.
20
- —— — j ..*««« ^ WVSAIII ttMu isuraDinea sewer
H3w»»« ..^.__ J 4M . . ^*»|t 9 1*111 X 1 _ _
1971.
Sampling
^^
Screenin
Microstraining. W.t.r 8 Sewage Korks. 116. No. 6,
Se«g,
SeB" Overflow-
?< m"' 19?6. Filterabllity and Microscreener Design. Journal WPCF, 42, No. 11,.pp 1944^
. 11,,,
428
-------
REFERENCES (Continued)
Sewer Design, Material, and Maintenance
1. Ahrens, J.F., et al. Chemical Control of Tree Roots in Sewer Lines. Journal WPCF, 42,
No. 0, pp 1643-1655. 1970.
2. Evaluation of External Sealing Methods to Reduce Storm Flow Effect in Sewerage Systems.
Final Progress Report, County of Sonoma Sanitation Department, WPD 111-01-66.
Environmental Protection Agency. 1966.
3. FRP Pipe Design Manual. The Ceilcote Company. 1971,
4. Godfrey. What's New in Water and Sewer Pipe? 'Civil Engineering, 40, No. 10, pp 46-52.
1970.
5. Graeser, H.J. Dallas Has Modern Approach to Sewer Installation. Water § Sewage Works,
116, pp 326-331. 1969.
6. Lysne, O.K. Hydraulic Design of Self-Cleaning Sewage Tunnels. ASCE Sanitary
Engineering, 95, No. SA 1, pp 17-36. 1969.
7. Peters, G.L. Report of A Sewer Sealing Process Using Infilcheck 110, A Product of
Sealite, Inc. Springfield Sanitary District. October 9, 1968.
8. Polyethylene Manholes Reduce Pipeline Leaks. Chemical Engineering. April 3, 1972.
9. Pomeroy, R.D. Sanitary Sewer Design and Hydrogen Sulfide Control. Presented at the
1970 Concrete Pipe Sewer and Culvert Seminar, Sponsored by the Portland Cement
Association and the California Precast Concrete Pipe Association: San Diego, April 14;
Downey, April 15; Sacramento, April 16; Oakland, April 17, 1970.
10. Report on Springfield Illinois Treatment Project, Infilcheck 10. Springfield Sanitary
District. October 15, 1968.
11. Rooting out a Sewer Problem. Water Q Wastes Engineering, 9, No. 11, p. 26. 1972.
12. Sewer Renewal with Aldyl D Polyethylene Pipe. E. I. DuPont DeNeraours § Co. Brochure
and miscellaneous articles. 1973.
13. Stall, J.B. and Terstriep, M.L. Storm Sewer Design—An Evaluation of the RRL Method.
11030 FLN. Environmental Protection Agency. October 1972.
14. Supplemental Report on Sewer Sealing Using Infilcheck flO. Springfield Sanitary
District. October 29, 1969.
15. Techite Technical Information. United Technology Center. 1968.
16. The Construction and Technical Evaluation of the Various Aspects of an Aluminum
Storm Sewer System. Schedule of Prices and Tabulation of Bids: Chamlin $ Associates,
Inc., Peru, Illinois. 11032 DTI. June 1971.
17. Trenchless Sewer Construction and Sewer Design Innovations. Sussex County Council,
Georgetown, Delaware. Demonstration Project No. S-800690. Environmental Protection
Agency. June 1972.
18. Willhoff, T.L. ABS Truss Pipe for Sewers. Water ft Wastes Engineering, 6, pp 40-42.
1969.
19. Lysne, O.K. Hydraulic Design of Self-Cleaning Sewage Tunnels. ASCE Sanitary
Engineering, 95, SA 1, pp 17-36. 1969.
Sewer Separation
1. A Pressure Sewer System Demonstration. EPA Region I. 11022 DQI. Environmental
Protection Agency. November 1972.
2. ASCE and Greeley and Hansen. Report on Milwaukee Study Area. American Society of Civil
Engineers Combined Sewer Separation Project. December 1968.
429
-------
REFERENCES (Continued)
Natlon's
C.lu.bia, D.part.,»t. of
6. Sewer Within a Sewer. Water Works « Wastes Engineering, 1, No. 1, pp 36-37. 1964.
Sewer System Operation and Control
; ppJ60-ol.ali97"?1< " C°ntr01 Se"er °*«»°»- »»« « Wastes Engineering,
contr°i
- ASCE
'
'» M.K. Seattle-Portland inspection trip.
11. Leiser, C.P. Samplers Used in Seattle, Washington. December 8, 1972.
Milwaiiir*fc« 11 A *? 4 cirr\ rt *. \ * ~ * * * wCwcTftffc Coiiunission of the p i t~v f\-f
• •"••-iwouiv^rc* J. .A V/ A *|< r V U • OC tlOnPT* 1 fi 1Q77 '••»'**'••*'•»• vim^v/iujrvi
Octobe?°1969.' WaSt?Water Measuring and Sampling Techniques. Tri-Aid Sciences, Inc.
14. McPherson. M.R Caoeiv.^ i i«... _^r ^i_ ., . , .
-Sii:??011!"?!!!'.^!.1.1??^???-*- Concept
) . _ — -~**»B».J»»,C VJ/3I.CU v>oncep
. American Society of Civil Engineers.,
15. Magnetic Flowmeters. Fischer $ Porter Catalog C10D.
No?r7?epp'7n-720OVei961WS fr°m C°mbined Sewers in ^shington, D.C. Journal WPCF, 33,
17.
18. Proposal for the Development of Fluidic Sampling Equipment for rnmK^ AC
t. Freeman Associates, Inc. Mav 1972 B
-------
REFERENCES (Continued)
19. Report on Operations. Municipality of Metropolitan Seattle. 1971.
20. Shelley, P.E. and Kirkpatrick, G.A. An Assessment of Automatic Sewer Flow Samplers.
Hydrospace-Challenger, Inc. Contract No. 68-03-0155. Environmental Protection Agency.
June 1973.
21. Swirl Concentrator Promises to Reduce Pollution. APWA Reporter, p 24. May 1972.
22. Summary of Experimental Tests Results of Effects of Polymer Additives on Open Channel
Flows. Columbia Research Corporation. Contract No. 68-01-0168. Environmental
Protection Agency. December 1972.
23. Suspended Solids Monitor. American Standard, Inc. 11020 DZB. Environmental Protection
Agency. August 1972.
24. Tarazi, D.S., et al. Comparison of Wastewater Sampling Techniques. Journal WPCF, 42,
No. 5, Part 1, p 708. 1970.
25. Thomas, H.A. The Hydraulics of Flood Wave Movements in Rivers. Carnegie Institute of
Technology Engineering Bulletin. 1934.
26. Tucker, L.S. Raingage Networks in the Largest Cities. American Society of Civil
Engineers. Technical Memorandum No. 9. 1969.
27. Tucker, L.S. Oakdale Gaging Installation, Chicago - Instrumentation and Data.
American Society of Civil Engineers. Technical Memorandum No. 2. 1968.
28. Weidner, R.B., Wiebel, S.R., and Robeck. G.G. An Automatic Mobile Sampling and Gaging
Unit. Public Works, 99, No. 1, pp 78-80. 1968.
29. Eiffert, W.T and Fleming, P.J. Pollution Abatement Through Sewer System Control.
Journal WPCF, 41, No. 2, pp 285-291. 1969.
30. Riddle, W.G Infiltration in Separate Sanitary Sewers. Journal WP.CF, 42, No. 9, p 1676.
1970.
1. City of Milwaukee. Humboldt Avenue Pollution Abatement Demonstration Project.
December 1971.
2. City of Milwaukee. Humboldt Avenue Pollution Abatement Demonstration Project -
General Project Description. December 1971.
3. Combined Sewer Temporary Underwater Storage Facility. Melpar: An American Standard
Company. 11022 DPP. Environmental Protection Agency. October 1970.
4. Combined Underflow-Storage Plan for Pollution and Flood Control in the Chicago
Metropolitan Area. City of Chicago Department of Public Works, Bureau of
Engineering. September 1969.
5. Composite Drainage Plan for the Chicago Area. City of Chicago, Department of Public
Works, Bureau of Engineering. September 1968.
6. Demonstration of Void Space Storage With Treatment and Flow Regulation. Karl R. Rohrer
Associates, Inc. 11020 DXH. Progress report for Environmental Protection Agency.
January 1973. .
7. Drainage Criteria Manual. Chapter on Storage. December 1968.
8. Escritt, L.B. Re-Examination of the Storm Tank Problem. Water 5 Waste Treatment,
pp 298-300. September/October 1969.
9. Feasibility of a Stabilization-Retention Basin in Lake Erie at Cleveland, Ohio. Havens
and Emerson. 11020 05/68. Environmental Protection Agency. May 1968.
10. Gregory, J.H., et al. Intercepting Sewers and Storm Stand-By Tanks at Columbus, Ohio.
ASCE Proceedings, October 1933, Paper No. 1887, pp 1295-1331.
11. Knudsen, Jorgen. Operation of Detention Tank for Humboldt Avenue Project.
431
-------
REFERENCES (Continued)
12'
• Svlronmen^i
^i Abatin, Pollution
15. Silo in the Sewer Stores Storm Water Overflow. Engineering News-Record. May 25, 1972
for^hicafo^^DeerTunnel." 197™.** IntcrcePtin« Sewer 18E, Extension-A System. Data
17. Table of Significant Data - Southwest 13A. Data for Chicago's Deep Tunnel. July 1970
19. Preload Prestressed Concrete. Preload Company, Inc. December 1972.
^^
22'
Stormwater Problem
^
contro1
.
Protection Agency. June 1970.
aiifc™Di:-Hs%iss.T4e?: s
6
432
-------
REFERENCES (Continued)
8. DeFilippi, J.A. and Shih, C.S. Characteristics of Separated Storm and Combined
Sewer Flows. Journal WPCF, 43, No. 10, pp 2033-2058. 1971.
9. Dobbins, W.E. Quantity and Composition of Storm Sewage Overflows. Presented at
ASCE Sanitary Engineering Division Symposium on Treatment of Storm Sewage Overflows.
April 17, 1962.
10. Engineering Investigation of Sewer Overflow Problem: Roanoke, Virginia. Hayes,
Seay, Mattern 5 Mattern. 11024 DMS. Environmental Protection Agency. May 1970.
11. Espey, W.H., Morgan, C.W., and Masch, F.D. A Study of Some Effects of Urbanization
on Storm Runoff From a Small Watershed. Technical Report HYO 07-6501. Texas Water
Commission. July 1965.
12. Gibbs, C.F., Henry, C.J., and Kersnar, F.J. How Seattle Beat Pollution. Water §
Wastes Engineering, 9, No. 2, pp 30-40. 1972.
13. Field, R. Annual Literature Review. Journal WPCF. June 1973.
14. Field, R. Management and Control of Combined Sewer Overflows: Program Overview.
Presented at 44th Annual Meeting of the New York Water Pollution Control Association.
New York. January 26-28, 1972.
15. Johnson, D.F. Nation's Capital Enlarges its Sewerage System. Civil Engineering,
pp 428-431 and 502-505. 1958.
16. Johnson, R.E., et al. Dustfall as a Source of Water Quality Impairment. Journal of
the Sanitary Engineering Division, SA 1, pp 245-267. February 1966.
17. McPherson, M.B. Hydrologic Effects of Urbanization in the United States. Technical
Memorandum No. 17. ^American Society of Civil Engineers. June 1972.
18. McPherson, M.B. Urban Runoff. Technical Memorandum No. 18. American Society of
Civil Engineers. August 1972.
19. Nash, N., Jerome, D., and Epstein, R. Characteristics of Combined Overflows. Paper
received during visit to Jamaica Bay (Spring Creek) Project.
20. Nutrient Removal Using Existing Combined Sewer Overflow Treatment Facilities (Supple-
mental Work Proposal). Onondaga County, Syracuse, N.Y. 11020 HFR. Environmental
Protection Agency. February 2, 1973.
21. Nicoli, E.H. Work in Scotland. Surveyor, pp 37-39. September 23, 1972.
22. Pollutional Effects of Stormwater and Overflows from Combined Sewer Systems: A
Preliminary Appraisal. U.S. Department of Health, Education, and Welfare. 1964.
23. Romer, H. and Klashman, L.M. How Combined Sewers Affect Water Pollution. Parts 152.
Public Works, 94, Nos. 3 5 4, pp 100 and 88. 1963.
24. Romer, H., Lagerre, G., and Gallagher, T. Effect of Combined Sewage Overflows on
Waters Around New York City. Presented at the ASCE Sanitary Engineering Division
Symposium on Treatment.of Storm Sewage Overflows. April 17, 1962.
25. Rice, I.M. New Approach to Applied Research. Water § Wastes Engineering, 8, No. 2,
pp 20-22. 1971.
26. Rosenkranz, W.A. Developments in Storm and Combined Sewer Pollution Control.
Presented at Meeting of the New England Water Pollution Control Association.
27. Rosenkranz, W.A., Wright, D.R., and Cywin, A. Improving the Efficiency of Sewerage
Systems. Presented at the Public Works Congress and Equipment Show. Miami Beach.
October 19-24, 1968.
28. Soderlund, G., Lehtinen, H., and Friberg, S. Physicochemical and Microbiological
Properties of Urban Storm-Water Run-off. Presented at the 5th International Water
Pollution Research Conference. July-August, 1970.
29. Special Report of the Department of Public Health Relative to the Preparation of Plans
and Maps for the Disposal of Sewage in the Merrimack River Valley. House No. 3733.
The Commonwealth of Massachusetts. 1964.
433
-------
REFERENCES (Continued)
3U« Storm and Combined Sewer
31. Systematic Study and
Research (First Year
U.S. Office of Water
32' SalaryYwater5^ Presented at the Fourth Annual
resources Engineering Conference. Vanderbilt University. June 4 1965
"' Ej'1972: RedUCing Polluti^ From Combined Sewer Overflows. APWA Reporter, pp 10-14.
34' ppe89-92.' W1969?SeSSing ^ Quality of Urban Drainage. Public Works, 100, No. 10,
35. JJeather Bureau Rainfall Frequency Atlas of the United States. Technical Paner No A
U.S. Department of Commerce. Washington, D.C. May 1961? 'ecimical Paper No. 4.
4 i. Conta"ln"""
"•
ASCB
Kir5»«.Dr"10P"" a"d »— »"«1.. Projects. Bnviro^ent.1 Protection
Agency.
Move"nts in
Textbooks
! vo1- '•
pU: a?f6?°dle- E'B- •«•' Supply and Waste Dispos.L Scranton, mt.rn.tion.l
434
-------
REFERENCES (Continued)
4. Imhoff, K. and Fair, G.M. Sewage Treatment. New York, John Wiley ft Sons. 1956. 2nd Ed.
5. Klein, L. River Pollution --2: Causes and Effects. Washington, D.C., Butterworth ft Co.,
Ltd. 1967.
6. Klein, L. River Pollution --3: Control. Washington, D.C., Butterworth ft Co., Ltd. 1966.
7. Lund, H.F., ed. Industrial Pollution Control Handbook. New York, McGraw-Hill Book Company.
1971.
8. Metcalf ft Eddy. American Sewerage Practice, Vol. II: Construction of Sewers. New York,
McGraw-Hill Book Company. 1915.
9. Statistical Summary: 1968 Inventory--Municipal Waste Facilities in the United States.
No. CWT-6. Environmental Protection Agency. 1970.
10. Thomann, R.V. Systems Analysis ft Water Quality Management. New York, Environmental Research
ft Applications, Inc. 1972.
Urban Runoff Characteristics
1. Urban Runoff Characteristics--Vol. 1: Analytical Studies. University of Cincinnati.
11024 DQU. Environmental Protection Agency. October 1972.
2. Urban Runoff Characteristics--Vol. II: Field Data. University of Cincinnati. 11024 DQU.
Environmental Protection Agency. October 1972.
3. Urban Runoff Characteristics--Vol. II (Cont'd): Field Data. University of Cincinnati.
11024 DQU. Environmental Protection Agency. October 1972.
435
-------
Section XVII
GLOSSARY, ABBREVIATIONS, AND CONVERSION FACTORS
GLOSSARY
. treatment processes--Means of treatment in which
or biochemical action is intensified to stabilize
nd. nitTlrv 1~n^ iir»ef-oKl« ~~~ • ,. , >
T • i i • ' " •*-*-A A j-/ me uiibLaDie organic nia.tter Dreseni"
areexampleSUterS' SCtivated slud8e Processes, and lagoons
both domestic sewage and
-rface
Combined sewer overflow- -Flow from a combined
sewer in
436
-------
Disinfection--The art of killing the larger portion of micro-
organisms in or on a substance with the probability that all
pathogenic bacteria are killed by the agent used.
Domestic sewage--Sewage derived principally from dwellings,
business buildings, institutions, and the like. It may or
may not contain groundwater.
Dual treatment--Those processes or facilities designed for
operating on both dry- and wet-weather flows.
Dynamic regulator--A semiautomatic or automatic regulator
device which may or may not have movable parts that are
sensitive to hydraulic conditions at their points of instal-
lation and are capable of adjusting themselves to variations
in such conditions or of being adjusted by remote control to
meet hydraulic conditions at points of installation or at
other points in the total combined sewer system.
Equali z at ion--The averaging (or method for averaging) of
variations in flow and composition of a liquid.
First flush--The condition, often occurring in storm sewer
discharges and combined sewer overflows, in which a dispro-
portionately high pollutional load is carried in the first
portion of the discharge or overflow.
Industrial wastewaters--The liquid wastes from industrial
processes, as distinct from domestic sewage.
Infiltrated municipal sewage --That flow in a sanitary sewer
resulting from a combination of municipal sewage and exces-
sive volumes of infiltration/inflow resulting from
precipitation.
Infiltration--The water entering a sewer system and service
connections from the ground, through such means as, but not
limited to, defective pipes, pipe joints, connections, or
manhole walls. Infiltration does not include, and is dis-
tinguished from, inflow.
Infiltration ratio--The ratio of rainfall volume entering
the sewers to the total rainfall volume.
Inflow--The water discharged into a sewer system and service
connections from such sources as, but not limited to, roof
leaders, cellar, yard, and area drains, foundation drains,
cooling water discharges, drains from springs and swampy
areas, manhole covers, cross connections from storm sewers
and combined sewers, catch basins, stormwaters, surface
437
-------
runoff, street wash waters, or drainaa^ TT^I~ A
include, and is distinguished ^infiltration. ^ ^
- Within the physical confines of the sewer pipe
S '
Physical-chemical treatment prnr.«.,...u..,- of
438
-------
Pollutant--Any harmful or objectionable material in or
change in physical characteristic of water or sewage.
Pretreatment--The removal of material such as gross solids,
grit, grease, and scum from sewage flows prior to physical,
biological, or physical-chemical treatment processes to im-
prove treatability. Pretreatment may include screening,
grit removal, skimming, preaeration, and flocculation.
Regulator--A structure which controls the amount of sewage
entering an interceptor by storing in a trunk line or di-
verting some portion of the flow to an outfall.
Retention--The prevention of runoff from entering the sewer
system by storing on a surface area or in a storage basin.
Sampling train--The intake, method for gathering and trans-
porting" tubing, and storage container of a sewer flow
sampler.
Sanitary sewer--A sewer that carries liquid and water-
carried wastes from residences, commercial buildings, indus-
trial plants, and institutions, together with relatively low
quantities of ground, storm, and surface waters that are not
admitted intentionally.
Sewer--A pipe or conduit generally closed, but normally not
flowing full, for carrying sewage or other waste liquids.
Sewerage--System of piping, with appurtenances, for collect -
ing and conveying wastewaters from source to discharge.
Static regulator—A regulator device which has no moving
parts or has movable parts which are insensitive to hy-
draulic conditions at the point of installation and which
are not capable of adjusting themselves to meet varying flow
or level conditions in the regulator-overflow structure.
Storm flow--Overland flow, sewer flow, or receiving stream
flow caused totally or partially by surface runoff or
snowmelt.
Storm sewer--A sewer that carries intercepted surface run-
off, street wash and other wash waters, or drainage, but
excludes domestic sewage and industrial wastes.
Storm sewer discharge--Flow from a storm sewer that is dis-
charged into a receiving water.
439
-------
440
-------
ABBREVIATIONS
Organizations
APWA
ASCE
DMWS
EPA
MSDGC
NPDES
NSF
OWRR
RD§D
USPHS
Symbols
%
$
°C
American Public Works Association
American Society of Civil Engineers
Detroit Metropolitan Water Service
U.S. Environmental Protection Agency
Metropolitan Sanitary District of
Greater Chicago
National Pollution Discharge
Elimination System
National Science Foundation
Office of Water Resources Research
Research, Development, and Demonstration
Program (EPA)
U.S. Public Health Service
percent
dollar
degree(s) Celsius (Centigrade)
approximately
greater than
less than
greater than or equal to
equal to or less than
cents
per
441
-------
Abbreviations
ABS
avg
BOD5
cf of cu ft
cfs
cf/lb
cm
cm/hr
cm/yr
COD
CPVC
cu m
cu in/day
cu m/min/day
cu m/min/ha
cu m/min/100 1
cu m/sec
cy
degree C
degree F
DO
DWF
eff.
ENR
F/M
fpm
fps
FRP
ft
g
gal.
gal./inch diameter/
mile/day
acrylonitrile butadiene styrene
average
biochemical oxygen demand (5-day)
cubic foot (feet)
cubic feet per second
cubic feet per pound
centimeter(s)
centimeter(s) per hour
centimeter(s) per year
chemical oxygen demand
chlorinated polyvinylchloride
cubic meter(s)
cubic meter(s) per day
cubic meter(s) per minute per day
cubic meter(s) per minute per hectare
cubic meter(s) per minute per 100 liters
cubic meter(s) per second
cubic yard
degree (s) Celsius (Centigrade)
degree (s) Fahrenheit
dissolved oxygen
dry-weather flow
effluent
Engineering News-Record (index)
food-to-microorganism ratio
feet per minute
feet per second
glass fiber-reinforced plastic
foot (feet)
gram(s)
gallon(s)
gallons per inch diameter per mile
per day
442
-------
gpd/sq ft
gpm/sq ft
ha
hr/yr
HST
in.
in./hr
in./yr
JTU
kg
kg/sq cm
kl
km
kwh
Ib
Ib/day
Ib/mil gal.
Ib/min
Ib/yr
1/day/ha
I/kg
1/min/sq m
I/sec
1/sec/sq m
m
max
mg
mgad
mgd
mg/1
mil gal.
min
min
Ml
gallon(s) per day per square foot
gallon(s) per minute per square foot
hectare
hour(s) per year
heat shrinkable tubing
inch(es)
inch(es) per hour
inch(es) per year
Jackson turbidity unit
kilogram(s)
kilograms per square centimeter
kiloliter(s)
kilometer(s)
kilowatt hour(s)
pound(s)
pounds per day
pounds per million gallons
pound(s) per minute
pounds per year
liter(s) per day per hectare
liter(s) per kilogram
liter(s) per minute per square meter
liter(s) per second
liter(s) per second per square meter
meter (s)
maximum
milligram(s)
million gallons per acre per day
million gallons per day
milligrams per liter
million gallons
minute(s)
minimum
megaliter(s)
443
-------
ml
MLSS
ml/1
mm
MPN
m/sec
N
NA
N/A
P
pcf
ppm
psf
psi
PVC
PVDC
rpm
RRL
scfm
sq cm
sq ft
sq ft/mgd
sq in.
sq km
sq mi
sq m/cu m/sec
SS
STP
SWMM
TOG
TS
TVS
VS
VSS
yr
milliliter(s)
mixed liquor suspended solids
milliliter(s) per liter
millimeter(s)
most probable number
meter(s) per second
nitrogen
not available
not applicable
phosphorus
pounds per cubic foot
parts per million
pounds per square foot
pounds per square inch
polyvinylchloride
polyvinyldichloride
revolutions per minute
Road Research Laboratory
standard cubic feet per minute
square centimeter(s)
square foot (feet)
square feet per million gallons per day
square inch(es)
square kilometer(s)
square mile(s)
square meter per cubic meter per second
suspended solids
sewage treatment plant
Storm Water Management Model
total organic carbon
total solids
total volatile solids
volatile solids
volatile suspended solids
y.ear(s)
444
-------
CONVERSION FACTORS
English to Metric
English unit
acre
•ere -foot
cubic foot
cubic feet per Minute
cubic feet per second
cubic inch
cubic yard
degree Fahrenheit
feet per Minute
feet per second
foot (feet)
gal Ion (s)
gallons per acre per day
gallons per capita per day
gallons per day
gallons. per day per
square foot
gallons per minute
gallons per minute per
square foot
gallons per square foot
horsepower
inch(es)
inches per hour
million gallons
million gallons per
acre per day
million gallons per day
mile
parts per billion
parts per million
pound (s)
pounds per acre per day
pounds per day per acre
pounds per 1,000 cubic feet
pounds per million gallons
pounds per cubic foot
pounds per square foot
pounds per square inch
square foot
square inch
square mile
square yard
standard cubic feet
per minute
ton (short)
yard
Abbr.
acre
acre-ft
cf
cfm
cfs
cu in.
cy
deg F
fpm
fps
ft
gal.
gad
gcd
gpd
gpd/sq ft
gpm
gpm/sq ft
gpsf
hp
in.
in./hr
mil gal.
mgad
mgd
mi
pph
ppm
Ib
Ib/acre/day
Ib/day/acre
lb/1.000 cf
Ib/mil gal.
pcf
psf
psi
sq ft
sq in.
sq mi
sq yd
scfm
ton
yd
Multiplier
0.40S
1.23S.S
28.32
0.0283
28.32
16.39
0.0164
0.765
764.6
O.S5S ('F-32)
0.00508
0.305
0.305
3.785
9.353
3.785
4.381 x 10"S
1.698 x 10"1
0.283
0.0631
2.445
0.679
40.743
0.746
2.54
2.54
3.785
3.785.0
0.039
43.808
0.0438
1.609
0.001
1.0
0.454
453.6
0.112
1.121
16.077
0.120
16.02
4.882 x 10"4
0.0703
0.0929
6.452
2.590
0.836
1.699
907.2
0.907
0.914
Abbr.
ha
cu •
1
cu m/min
I/ sec
cu cm
1
cu m
1
deg C
•/sec
m/sec
m
1
1/day/ha
I/capita/day
I/sec
cu m/hr/sq m
cu m/min/ha
I/sec
cu/m/hr/sq m
1/sq m
kw
cm
cm/hr
Ml
cu m
cu m/hr/sq m
I/sec
cu m/sec
km
mg/1
mg/1
kg
8
g/day/sq m
kg/day/ha
g/cu m
mg/1
kg/cu m
kg/sq cm
kg/sq cm
sq m
sq cm
sq km
sq m
cu m/hr
kg
metric ton
m
Metric unit
hectare
cubic meter
liter
cubic meters per minute
liters per second
cubic centimeter
liter
cubic meter
liter
degree Celsius
meters per second
meters per second
meter (s)
liter(s)
liters per day per hectare
liters per capita per day
liters per second.
cubic meters per hour
per square meter
cubic meters per minute
per hectare
liters per second
cubic meters per hour
per square meter
square meter
liters per square meter
kilowatts
centimeter
centimeters per hour
megaliters (liters x 10 )
cubic meters
cubic meters per hour
per square meter
liters per second
cubic meters per second
kilometer
milligrams per liter
milligrams per liter
kilogram
grams
grams per day per square
meter
kilograms per day per
hectare
grams per cubic meter
milligrams per liter
kilograms per cubic meter
kilograms per square
centimeter
kilograms per square
centimeter
square meter
square centimeter
square kilometer
square meter
cubic meters per hour
kilograms
metric ton
meter
445
-------
Section XVIII
LIST OF PUBLICATIONS
p«??r;-J'Ai and Field> R- Co"nter Measures for
Pollution from Overflows - The State of the Art
Presented at Water Pollution Control Federation '46th
Ohio- Septet 30-"
2. Lager, J. A. Stormwater Treatment: Four Case
Histories. Prepared for presentation at ASCE National
Meeting on Water Resources Engineering. Los Angeles
California. January 21-25, 1974. "iigeies ,
3.
Smith, W.G. Application of a Simulation Model for
Routing Storm Flows through Sewer Systems. Presented
ral
. esene
ral1S^at!s Water Pollution Control Association
June lT-ll, 19°73?renCe' R°ckton' Illinois.
446
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
EPORT NO. 2.
EPA-670/2-74-040
TLE AND SUBTITLE
RBAN STORMWATER MANAGEMENT AND TECHNOLOGY:
i Assessment
JTHOR(S)
Dhn A. Lager and William G. Smith
RFORMING ORGANIZATION NAME AND ADDRESS
stcalf $ Eddy, Inc.
alo Alto, California 94303
PONSORING AGENCY NAME AND ADDRESS
ational Environmental Research Center
ffice of Research and Development
. S. Environmental Protection Agency
Lncinnati, Ohio 45268
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
December 1974; Issuing Date
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT
10. PROGRAM ELEMENT NO.
1BB034/ROAP 21ASZ/Task
NO.
1
11. CONTRACT/GrajQXNO.
68-03-0179
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
UPPLEMENTARY NOTES
comprehensive investigation and assessment of promising, completed,
id ongoing urban stormwater projects, representative of the state-of-
le-art in abatement theory and technology, has been accomplished. The
?sults, presented in textbook format, provide a compendium of project
iformation on management and technology alternatives within a frame-
Drk of problem identification, evaluation procedures, and program
>sessment and selection. Essentially every metropolitan area of the
lited States has a stormwater problem, whether served by a combined
jwer system (approximately 29 percent of the total sewered population)
r a separate sewer system. However, the tools for reducing stormwater
Dilution, in the form of demonstrated processes and devices, do exist
id provide many-faceted approach techniques to individual situations.
lese tools are constantly being increased in number and improved upon
> a part of a continuing nationwide research and development effort.
le most promising approaches to date involve the integrated use of
mtrol and treatment systems with an areawide, multidisciplinary
jrspective.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
infection, Drainage, *Water pollu-
n, *Waste treatment, *Sewage treat-
t, *Surface water runoff, *Runoff,
ste water, *Sewage, Contaminants,
ter quality, Cost analysis, *Cost
ectiveness, *Storage tanks, *Storm
ers, *0verf lows- -sewers , *Combined
ers, Hydrology, Hydraulics, *Mathe-
ical models, Remote control
b.lDENTIFIERS/OPEN ENDED TERMS
Drainage systems, Water pollution
control, Biological treatment,
Pollution abatement, *Storm runoff,
*Water pollution sources, Water
pollution effects, *Wastewater
treatment, *Urban hydrology, *Com-
bined sewer overflows, Physical
processes, Physical-chemical
treatment
ISTRIBUTION STATEMENT
RELEASE TO PUBLIC
rorm 2220-1 (9-73) 4^
19. SECURITY CLASS (This Report)
UNCLASSIFIED
20. SECURITY CLASS (This page)
UNCLASSIFIED
c. cos ATI Field/Group
8H
13B
21. NO. OF PAGES
471
22. PRICE
17
GOVERNMENT PRINTING OFFICE: 1975-657-590/5331 Region No. 5-11
-------
------- |