United States          Office of Research and       EPA/600/R-00/020
Environmental Protection   Development                January 2000
Agency              Washington, DC 20460
Retrofitting Control Facilities For
Wet-Weather Flow Treatment

Research Report

-------
                                         EPA#/600/R-00/020
                                               January 2000
       Retrofitting Control Facilities For
        Wet-Weather Flow Treatment
                    by
      Peter E. Moffa, Howard M. Goebel,
     Daniel P. Davis and John J. LaGorga
   Moffa & Associates Consulting Engineers
         Syracuse, New York 13214

             In Association with

              Earth Tech, Inc.
        South Portland, Maine 04106

                Robert Pitt
          Environmental  Engineer
           Birmingham, AL 35226

        Contract No. 8C-R056NTSX
               Project Officer
            Thomas P. O'Connor
             Technical Advisors
       Richard Field & Mary K. Stinson
    Urban Watershed Management Branch
   Water Supply & Water Resources Division
National Risk Management Research Laboratory
             Edison, NJ 08837
National Risk Management Research Laboratory
     Office of Research and Development
    U.S. Environmental Protection Agency
           Cincinnati, Ohio 45268

-------
                                    Notice
The information in this document has been funded by the United States Environmental
Protection Agency under Contract No. 8C-R056-NTSX.  Although it has been subjected
to the Agency's peer and administrative review and has been approved for publication as
an EPA document, it does not necessarily reflect the views of the Agency and no official
endorsement should be inferred.  Also, the mention of trade names  or commercial
products does not imply endorsement by the United States government.

-------
                                   Foreword
      The U.S. Environmental Protection Agency is charged by Congress with protecting
the Nation's land, air, and water resources.  Under a mandate of national environmental
laws,  the Agency strives to formulate and implement actions leading to a compatible
balance between human activities and the ability  of natural  systems to  support  and
nurture  life.   To meet this mandate,  EPA's research program  is providing data  and
technical support for  solving environmental problems  today and building  a  science
knowledge base necessary to manage our ecological resources wisely, understand  how
pollutants affect our health, and prevent or reduce environmental risks in the future.

      The National Risk Management Research  Laboratory is the Agency's center for
investigation of technological and management approaches for reducing risk from threats
to human health and the environment. The focus of the Laboratory's research program  is
on methods  for the prevention and control of pollution to air, land, water and subsurface
resources;   protection  of  water  quality  in  public  water  systems;  remediation  of
contaminated sites and ground water; and prevention and control of indoor air pollution.
The goal of this research effort is to catalyze  development and implementation of
innovative, cost-effective, environmental technologies; develop scientific and engineering
information   needed by  EPA to support regulatory and policy decisions;  and provide
technical support  and  information transfer  to  ensure  effective  implementation  of
environmental regulations and strategies.

      This publication has been produced as part of the Laboratory's strategic long-term
research plan.  It is published and  made  available by EPA's Office of Research  and
Development to assist the user community and to link researchers with their clients.
                                    E. Timothy Oppelt, Director
                                    National Risk Management Research Laboratory

-------
                                     Abstract

Available technologies were evaluated to demonstrate the technical feasibility and cost
effectiveness of retrofitting existing facilities to handle wet-weather flow.  Cost/benefit
relationships  were  also  compared to construction  of new conventional  control and
treatment facilities.

Desktop analyses of 13 separate retrofit examples were  performed for 1) converting or
retrofitting  primary  settling tanks  with  dissolved  air flotation  and  lamellae  and/or
microsand-enhanced plate or tube settling units, 2) retrofitting existing wet-weather flow
storage tanks to provide enhanced settling/ treatment and post-storm solids removal, 3)
converting  dry ponds to wet ponds for enhanced treatment, 4)  retrofitting wet-weather
flow storage tanks for dry-weather flow augmentation, 5) using storage for sanitary sewer
overflow control , 6) retrofitting for industrial wastewater control in a combined sewer
system, and 7) bringing outdated/abandoned treatment plants back online as wet-weather
flow treatment facilities.

This analysis demonstrated that  retrofitting existing  wet-weather flow facilities can  be
technically  feasible  in most cases and may be more  cost effective than  construction of
new conventional control and treatment facilities. The feasibility and cost effectiveness of
retrofitting  was  found to be  a  function of  site-specific  conditions  and  treatment
requirements.  Retrofitting processes will better enable communities  to meet EPA's
National CSO Policy and stormwater permitting program requirements.
                                        IV

-------
                                   Contents
Notice	ii
Forward	iii
Abstract	iv
Tables	vii
Figures	ix
Acronyms and Abbreviations	x
Acknowledgement	xiii
Section 1    Introduction	1
Section 2    Conclusions	7
Section 3    Recommendations 	10
Section 4    Methods and Material	11
            4.1    Literature Search	11
            4.2    UseofSWMM	19
            4.3    Case Studies and Hypothetical Retrofit Examples	21
Section 5    Desktop Evaluation	22
            5.1    Conversion of Primary Treatment Tanks	22
                  5.1.a.  Augusta WWTF, Augusta, ME	22
                  5.1.b.  Hypothetical Retrofit of Primary Treatment Facilities
                        Using the ACTIFLO System	37
            5.2    Retrofitting Existing WWF Storage Tanks to Provide Enhanced
                  Settling/Treatment and Post-Storm Solids Removal;
                  Spring Creek AWPCP,  New York, NY.	53
                  5.2.a.  Cross-Flow Plate Settlers	56
                  5.2.b.  Chemical Addition to AWPCP	63
                  5.2.c.  Post-Storm Solids Removal	69
            5.3    Converting Dry Ponds for Enhanced Treatment	107
                  5.3.a.  Sunnyvale Detention Basin, Santa Clara County, CA	108
                  5.3.b.  Monroe St. Detention Pond, Madison, Wl	121
                  5.3.c.  Birmingham, AL Dry-Ponds	129
            5.4    Retrofitting WWF Storage Tanks for DWF Augmentation	139
                  5.4.a.  Erie Boulevard Storage System, Syracuse, NY.	139
                  5.4.b.  Spring Creek AWPCP,  New York, NY.	145
            5.5    SSO Control Using Storage	149
                  5.5.a.  Hypothetical Overflow Retention Facility Retrofit to an
                        Existing Sanitary Sewer System	149

-------
                             Contents, Continued

            5.6    Retrofitting for Industrial Wastewater Control in a
                  Combined Sewer System	156
                  5.6.a. Rockland WWTF, Rockland, ME	156
            5.7    Bringing Outdated/Abandoned POTWs Back Online as
                  Wet-Weather Flow Treatment Facilities	174
                  5.7.a. Auburn WWTF, Auburn,  NY.	174
Section 6    References    	185

Appendices
            Appendix A: Bibliographic Databases	191
            Appendix B: Selected Tables with English  Equivalent Units	195
                                      VI

-------
                                    Tables

2-1.      Summary of Retrofitting Desktop Analysis Projects	7
5.1 .a-1   Phase 1 CSO Abatement Program Wet-Weather Treatment Design
         Criteria	30
5.1.a-2   Estimated O&M Costs, CSO Abatement Program-All Phases	35
5.1.a-3   Phase 1 CSO Abatement Program Project Costs	36
5.1.b-1   ACTIFLO Design Summary	37
5.1.b-2   Comparison of ACTIFLO and Existing Primary Treatment Systems	39
5.1 .b-3   Average Annual Overflow Volume From the Pre- and Post-Retrofit
         Facilities (1962-1991)	44
5.1.b-4   Estimated O&M Costs	46
5.1.b-5   Estimated ACTIFLO Retrofit Capital Costs	47
5.1.b-6   Major-Operating Parameters of ACTIFLO Pilot Plant at Nominal Flow Rate
         of150M3/H	47
5.1.b-7   Removal Efficiency From 20 Daily Samples in Primary Treatment
         in Mexico City, Mexico	48
5.1.b-8   Effect of Coagulant Dosage on Removal Efficiencies in Galveston, Texas....50
5.1.b-9   Effect of Overflow Rate on Removal Efficiencies, Cincinnati, Ohio	51
5.2-1     Existing Dimensions of the AWPCP and Design  Flow Rate	55
5.2.a-1   Design Parameters for the Cross-Flow Piste Separator	56
5.2.a-2   Cross-Flow Plate Settler Design Data	63
5.2.b-1   Coagulant Dosages and Performance Data for Six Treatment Plants
         That Employ Chemical Coagulation for Advanced Primary Performance
         In Sedimentation Tanks	66
5.2.b-2   Estimated Capital Costs Chemical Addition Retrofit	67
5.2.b-3   Estimated Chemical Addition O&M Costs	68
5.2.C-1   Work Plan Prototype Test Matrix	76
5.2.c-2   Summary of Prototype Testing Operation	82
5.2.c-3   Operating Data for Existing Traveling Bridge	89
5.2.c-4   Tipping Bucket Cleaning System Test Results at Representative
         Solids Conditions	90
5.2.c-5   Spray Nozzle Cleaning System Test Results	93
5.2.c-6   Original Orifice Header Configuration Test Results	95
5.2.c-7   Modified Orifice Header Configuration Test Results at Representative
         Solids Concentrations	96
5.2.c-8   Projected Cleaning Requirements for a Single Basin Cleaning Event	102
5.2.c-9   Cost Comparison of Full Scale Tipping Bucket Versus Orifice
         Header Cleaning Systems	104
5.3.a-1   Comparison of Median Metal Concentrations at Inlet to Retrofitted Basin
         to Other Santa Clara Valley Stormwater Monitoring Station Data	115
5.3.a-2   Inlet and Outlet Observed Concentrations and Pollutant Removals	116
5.3.a-3   Sediment Observations (mg/kg)	117
                                      VII

-------
                              Tables, Continued

5.3.a-4   Comparison of Average Sediment Concentrations from Detention
         Basins and Swales (mg/kg)	118
5.3.a-5   Estimated Mean Annual Load Reduction and Cost Effectiveness	119
5.3.a-6   Estimated Annualized Costs for Capital Expenditures and Operation	120
5.3.D-1   Summary of Observed Influent and Effluent Pollutant Concentrations
         at Monroe St. Pond	124
5.3.D-2   Summary Table of Pollutant Control	127
5.3.b-3   Average Suspended Solids Removal for Variable n Values for the
         Monroe St. Ponds	128
5.3.C-1   Outlet Device Descriptions	133
5.3.C-2   Subcatchment Summaries for Design Storms	133
5.3.C-3   Pond Results of HydroCAD Simulations	134
5.3.C-4   DETPOND Summary for Design Storms	135
5.3.C-5   Water Quality Output Summary for 1976 Rain File	135
5.3.C-6   Water Quality Output Summary for 1952-1989 Rain File	138
5.4.a-1   Average Annual Capture Volume From the EBSS Under Pre- and
         Post-Retrofit Conditions (1962-19991)	143
5.4.a-2   Estimated O&M Costs	144
5.4.a-3   Estimated EBSS Retrofit Capital Costs	144
5.4.b-1   Wet-Weather Discharge Volume From the Spring Creek AWPCP
         Under Pre-and Post-Retrofit Conditions for 1985	146
5.4.b-2   Estimated O&M Costs	147
5.4.b-3   Estimated Spring Creek DWF Retrofit Capital Costs	148
5.5.a-1   Average Annusl Overflow Volume from the Pre- and Post-Retrofit
         Facilities (1971-1985)	151
5.5.a-2   ORF O&M Costs	153
5.5.a-3   Estimated ORF Construction Costs	154
5.6.a-1   SWMM Modeling Summary-Annusl Overflow Volumes	164
5.6.3-2   SWMM Modeling Summsry- Projected Peak Flows	164
5.6.3-3   SWMM Modeling Summsry- Projected Pesk Flows	165
5.6.3-4   Wet-Westher Trestment Design Criteris	167
5.6.3-5   Estimsted O&M Costs Phsse II CSO Improvements	172
5.6.3-6   Project Costs - Phsse II CSO Improvements	173
5.7.3-1   Bssis of Design of the Overflow Retention Fscilities	178
5.7.3-2   Average Annusl Overflow Volume From the Pre- snd Post-Retrofit
         Fscilities (1971-1991)	181
5.7.3-3   ORF O&M Costs	183
5.7.3-4   ORF Chlorinstion/Dechlorinstion Cspitsl Costs	183
                                     VIM

-------
                                     Figures

5.1.a-1   Existing Site Layout Prior to Retrofit	24
5.1.a-2   Process Flow Schematic Prior to Retrofit	25
5.1.a-3   Site Layout Following Retrofit	28
5.1.a-4   Process Flow Schematic Following Retrofit	29
5.1 .b-1   Layout of Hypothetical WWTF Prior to Construction of ACTIFLO System	38
5.1.b-2   Layout of Proposed ACTIFLO System in Hypothetical Setting	40
5.1.b-3   ACTIFLO Process Flow Diagram	42
5.1.D-4   Helminth Eggs Removal in Primary Wastewater Treatment in Mexico
         City, Mexico	49
5.1.D-5   SS Removal in Primary Wastewater Treatment in Mexico City, Mexico	49
5.1.b-6   Start-Up Efficiency of ACTIFLO Process	50
5.1.b-7   Twenty-Four Hour Continuous Demonstration Run,  Cincinnati, Ohio	51
5.2-1     Spring Creek AWPCP  Layout	54
5.2.a-1   Layout of Shell Co. Cross-Flow Plate Settler, Montreal Canada	57
5.2.a-2   Horizontal Projections  of Inclined Cross-Flow Lamella Plates	60
5.2.a-3   Plan View of Aquarius  Cross-Flow Plate Settler	61
5.2.a-4   Profile of Aquarius Cross-Flow Plate Settler	62
5.2.c-1   Wash Water Piping System Schematic	72
5.2.c-2   Tipping Buckets-Enlarged Plan	74
5.2.c-3   Tipping Buckets - Section A	75
5.2.c-4   Modified Plywood Training  Wall Schematic	78
5.2.C-5   Flushing Gate-Section B	79
5.2.c-6   Spray Nozzle Header- Section	81
5.2.c-7   Original Orifice Header Configuration - Section A	85
5.2.c-8   Modified Orifice Header Configuration - Section A	86
5.3.a-1   Sunnyvale Pump Ststion No. 2 Showing Retrofit Structures	111
5.3.C-1   Stage v. Surface Area  Curve	132
5.3.C-2   Pond Stage v. Particle Residue Control	136
5.3.C-3   Rain Depth v. Particle  Residue Control	137
5.3.C-4   Rain Intensity v. Particle Residue Control	137
5.4.a-1   EBSS Layout	140
5.6.3-1   Existing Site Layout Prior to Retrofit	158
5.6.3-2   Process Flow schemstic Prior to Retrofit	159
5.6.3-3   Site Lsyout Following Retrofit	162
5.6.3-4   Process Flow Schemstic Following Retrofit	163
5.7.3-1   Existing Wsstewster Trestment Flow Disgram	176
5.7.3-2   Conceptusl Wsstewster snd Sludge Trestment Flow Disgram	180
                                        IX

-------
                        Acronyms and Abbreviations
ac-ft
ASD
AWPCP
BATS
BMP
BOD
CCS
COM
CEAM
cfs
cm
CN
COD
COV
CPE
CSO
CSS
cyd
CWA
D
DAF
DEP
DWF
EA
EBSS
EPA
EMC
FDS
FeCI3
fps
ft
gal
gpd
gpm
ha
HP
h
IDF
In.
I/I
JSRS
kg
Acre-feet
Augusta Sanitary District
Auxiliary Water Pollution Control Plant
Burnet Avenue Trunk Sewer
Best Management Practice
Biochemical Oxygen Demand
Combined Collection System
Camp Dresser & McKee
Center for Exposure Assessment Modeling
Cubic Feet per Second
Centimeter
Curve Number
Chemical Oxygen Demand
Coefficient of Variation
Comprehensive Performance Evaluation
Combined Sewer Overflow
Combined Sewer System
Cubic Yards
Clean Water Act
Day
Dissolved Air Flotation
Department of Environmental Protection
Dry Weather Flow
Environmental Assessment
Erie Boulevard Storage System
U.S. Environmental Protection Agency
Event Mean Concentration
Flow Distribution Structure
Ferric Chloride
Feet per Second
Feet
Gallon
Gallons per day
Gallons per minute
Hectare
Horse power
Hour
Intensity Distribution Frequency
Inch
Infiltration and/or Inflow
James Street Relief Sewer
Kilogram

-------
                    Acronyms and Abbreviations, Continued
km
kW
I
Ibs
m
mA
mm
urn
mi
Metro
METRO
MG
MGD
mg
mg/L
jig
min
MIS
ml
msl
mt
MTP
NASA
NOAA
NPDES
NTIS
NURP
NWS
NYCDEP
NYSDEC
O&M
ORF
POTWs
ppm
PPS
PRF
psi
RCP
rpm
s
SCADA
scfm
Kilometer
Kilowatt
Liter
Pounds
Meter
Milliampere
Millimeter
Micrometer
Mile
Municipality of Toronto Metropolitan
Syracuse Metropolitan Wastewater Treatment Facility
Million gallons
Million gallons per day
Milligram
Milligrams per Liter
Microgram
Minute
Main Interceptor Sewer
Milliliter
Mean Sea Level
Metric ton
Main Treatment Plant
National Aeronautics and Space Administration
National Oceanic and Atmospheric Administration
National Pollution Discharge Elimination System
National Technical Information Services
Nationwide Urban Runoff Program
National Weather Service
New York City Department of Environmental Protection
New York State Department of Environmental Conservation
Operation and Maintenance
Overflow  Retention Facility
Publicly Owned Treatment Plants
Parts per Million
Particulate Pollutant Strength
Peak Reduction Factors
Pounds Per Square  Inch
Reinforced Concrete Pipe
Revolutions per Minute
Seconds
Supervisory Control  and Data Acquisition
Standard  Cubic Feet Per Minute
                                      XI

-------
scs
SESD
SOR
STLC
SS
SSES
SSO
SW
SWMM
SYNOP
TB
TKN
TOO
TTLC
USGS
WET
WHO
WPCP
WQO
WWF
WWTF
  Acronyms and Abbreviations, Continued

Soil Conservation Service
South Essex Sewerage District
Surface Overflow Rate
Soluble Threshold Limit Concentrations
Suspended Solids
Sewer System Evaluation Survey
Sanitary Sewer Overflow
Storm water
Stormwater Management Model
Synoptic Rainfall Analysis Program
Tipping Buckets
Total Kjeldahl Nitrogen
Total Organic Carbon
Total Threshold Limit Concentration
U.S. Geological Survey
Whole Effluent Toxicity
World Heath Organization
Water Pollution Control Plant
Water Quality Objective
Wet Weather Flow
Wastewater Treatment Facility
                                     XII

-------
                              Acknowledgements

Moffa & Associates gratefully acknowledges the support of the project by the National
Risk Management Research Laboratory, Office of Research and Development,  U.S.
Environmental Protection Agency, under the direction of Mr. Richard Field, Wet-Weather
Flow Research Program leader.

The cooperation of Thomas P. O'Connor, Project Officer, Urban Watershed Management
Branch, Water  Supply and  Water  Resources Division, National  Risk Management
Research  Laboratory  (NRMRL),  Office of Research  and Development (ORD),  U.S.
Environmental Protection Agency (EPA), is acknowledged as are Dr. David Fischer and
Dr. Ronald Rovanesk,  ORISE post graduates with EPA's  UWMB, in Edison, New Jersey
and Dan Murray, Technology Transfer Division, NRMRL, ORD, EPA, Cincinnati, Ohio for
providing reviews for the report.

In addition, special thanks are given to the following consultants for their contributions to
this report:

  Steve D. Freedman,  P.E., Earth Tech  Inc.

  Sorin L. Goldstein, P.E., Camp Dresser & McKee, Inc., Woodbury, NY  11797

  Dwight A. MacArthur, P.E., O'Brien & Gere Engineers, Inc., Syracuse, NY 13221

  Robert Pitt, Environmental Engineer, Consultant

  Robert B. Stallings, P.E., Earth Tech Inc.

The assistance of Gregory J. Hotaling with preparation of graphical materials presented is
hereby recognized, as is the  assistance  of Renee Davis for assistance  with word
processing.

This report was prepared by  Moffa  & Associates Consulting Engineers, by Howard M.
Goebel, Daniel  P. Davis,  and John J.  LaGorga under the direction of Peter E. Moffa,
Principal.
                                      XIII

-------
                                   Section 1
                                 Introduction
Background
Wet-weather  flow  (WWF) is currently one  of  the  leading causes of water-quality
impairment in the United States and improvement of controls is one of the two priority
water focus-areas cited by the U.S. Environmental Protection Agency's (EPA) Office of
Water in its National Agenda for the Future.   Wet-weather  pollution problems  are
primarily manifested in three different ways: separate sewer overflows (SSO), combined
sewer overflows (CSO), and stormwater  (SW) discharges.  SSO are the result of the
unplanned relief of a sewer system intended only for sanitary sewage but also carrying
infiltration and inflow (I/I), illegal stormwater connections, and stormwater. CSO are the
result of the "designed" relief of a sewer system that intentionally carries stormwater and
sanitary sewage.  Stormwater runoff discharges include flows that have  been collected
in a separate storm sewer system or a surface drainage system.

WWF pollution  is extensive throughout the country.  There are approximately  15,000
CSO  discharge points within  1,100 municipalities that have combined  sewers.  SSO
occur  in more  than  1,000 municipalities.   SW discharges  occur  in  over 6,500
municipalities with  populations greater  than 50,000 and  at  1.2  million industrial,
commercial, institutional, and retail sources.

It was not until data from several comprehensive municipal CSO and SSO studies were
collected that the types of pollutants discharged by overflow systems and their relative
impacts on receiving  waters were initially understood.  The  primary  parameters of
importance are  visible floatables and gross  solids,  bacteria  and  other  pathogens,
suspended solids  (SS),  biochemical  oxygen demand (BOD),  nutrients,  and toxic
organics and metals.

The projected costs for national CSO, SSO,  and SW abatement are in  the tens of
billions of  dollars.  Consequently,  inexpensive abatement alternatives are  required to
alleviate the high costs associated with  conventional treatment.  High-rate treatment
technologies can increase sewage treatment system  capacity  to allow treatment of a
significant portion of the WWF in sanitary and combined sewers. The primary benefits
of  high-rate treatment technologies are their ability to handle higher flow rates and
compactness  or small footprint compared to conventional treatment processes.  High-
rate treatment technologies require less tankage and space and in urbanized areas are
often  more cost effective than conventional treatment  processes since storm flows are
significantly greater than dry-weather flows.

Wet-weather discharges are reflections of the watersheds or sewersheds from which
they are generated and the climatic patterns those areas experience.  The impacts of
wet-weather discharges are determined by a variety of complex relationships between
the temporal  and  spatial  distributions of discharges.  These  include  the types and
magnitudes of pollutant loads being transported, the ecological  and hydraulic nature of
the receiving  water, as well  as the designated beneficial  uses, and the  desires and

-------
expectations of stakeholders. The result is a wide-scale problem that requires very site-
specific solutions.

Solutions to WWF problems require investigation of a wide variety of management
options and the relationships between management costs and water quality benefits.
WWF must be controlled by storage-treatment systems since WWF are intermittent and
highly variable in  both pollutant concentrations and flow rates. An optimized system
maximizes  the use of  the  existing  infrastructure  prior  to  construction  of  new
treatment/storage  devices.

Retrofitting  existing sewerage systems to handle increased WWF that  result in SSO,
CSO, and SW discharges may be a cost-effective alternative to the construction of new
treatment systems.   Retrofitting  existing sewerage systems and treatment plants to
convey and treat additional WWF can be accomplished by:

1.     Increasing the hydraulic loadings at the control facilities (CSO and  SSO);
2.     Increasing the amount of storage throughout the conveyance system (All WWF);
      and
3.     Providing the additional function of pollution removal by settling at flood-control-
      detention basins (SW).

Summary of Analyses

This report demonstrates  real  and hypothetical case studies.  Retrofits of available
technologies for conventional sewer systems are presented in the form  of desktop
analyses.  The desktop analyses were conducted to determine the technical feasibility
and  cost effectiveness of retrofitting existing sewage and WWF  systems in lieu of
constructing new  sewerage and conveyance systems.  Site-specific  analyses were
completed for  WWF  storage-treatment systems and their associated cost due to the
variability of wet-weather volume andtreatability (quantity and quality).

The desktop analyses including conceptual layouts and designs of systems to increase
treatment  and storage of WWF through  retrofits were performed for the following
scenarios:

1.     Converting  or retrofitting primary settling tanks at the publicly owned treatment
      works  (POTW) to  alternative sedimentation or sedimentation   enhancement
      methods or methods of treatment that  provide equivalent or higher levels of
      treatment (SSO and CSO);
2.     Retrofitting  existing CSO storage tanks to provide enhanced settling/treatment
      and post-storm solids removal;
3.     Converting  existing  "dry-ponds"  and/or  other flood control devices  to SW
      detention   ponds   for separate   SW  systems  to  allow  treatment  through
      sedimentation;
4.     Retrofitting  existing WWF storage tanks and ponds to serve a second function as
      DWF equalization tanks;

-------
5.     Retrofitting for industrial wastewater control in a combined sewer system;
6.     SSO control using storage; and
7.     Bringing  outdated  and/or abandoned POTWs back  online as satellite WWF
      treatment facilities (SSO and CSO).

The  analyses were  performed over a variety of orders of magnitude of flow  and/or
volume for each scenario.  The  desktop analyses of the retrofitted systems addressed
the following design considerations, where appropriate:

1.     Operations, maintenance  and level of automation
2.     Sludge handling requirements and disposal
3.     Continuous operation versus storm flow event operations
4.     Chemical storage and feed rate
5.     Bypass requirements

Treatment values, pollutant and  hydraulic loadings, and seasonal and diurnal variation
are also provided, where appropriate.

Simplified simulations were completed  using  historical  rainfall data  for an one-year
period using the EPA Stormwater  Management Model (SWMM) with hourly time steps
or an appropriate alternative.  The yearly rainfall data is provided including seasonal
variation, average antecedent duration, and storm intensity.

When feasible,  the  cost effectiveness of retrofitting  versus  construction of additional
conventional treatment works and/or storage  basins was  completed as part of each
desktop analysis. The effectiveness of each system, guidance on design and operation,
cost  breakdowns (new  construction  and modification of existing structure,  system
installation, and operation and maintenance [O&M]), and case studies are provided.


Regulatory Background

Since the 1970s, the U.S. Environmental Protection Agency's Construction Grant
Program financed the development and construction of hundreds of wastewater
treatment facilities (WWTF) around the nation.  During this time the main focus was
regional WWTF designed to treat dry-weather sewage flows. Little or no money was
invested to abate wet-weather pollution sources. Since then, many of the POTWs have
exceeded their design flows, discharge standards have become  more stringent, and
federal regulations and policies have begun to address WWF pollution.  All of these
factors contribute to a need for more and better WWF treatment  facilities.  Today,
federal funding for new environmental facilities is scarce and taxpayers  carry the
burden. Based on this climate, it becomes much more critical that any compliance
needs should first consider optimizing existing infrastructure through retrofitting.

-------
This report is intended to help municipalities who may be impacted by current policies
and regulations (i.e., National Pollutant Discharge Elimination System (NPDES) Phase I
permitting and the CSO Control Policy) and the newer regulations and policies that
address WWF (i.e., the upcoming NPDES (Phase II)  permitting regulations and SSO
regulations).  Also, this report can assist municipalities and others (e.g., industry and
commercial property owners) who are subject to local regulations.

Stormwater Regulations and Permits

Congress added  Section  402(p) to the  Clean Water Act  in 1987 to require the
implementation of a comprehensive approach for addressing stormwater discharges.
Section 402(p)(4) required  EPA to develop permit  application regulations under the
NPDES,  submission of NPDES  permit  applications,  issuance of NPDES permits, and
compliance with NPDES permit conditions. Section 402(p)(6) requires EPA to designate
stormwater discharges to be regulated (within  the statutory  definitions  provided  in
section  402(p)(2))  and establish  a  comprehensive regulatory program, which  may
include performance standards,  guidelines, guidance, and management practices and
treatment requirements (EPA, 1999).

The  first phase of  NPDES  stormwater permit application regulations  ("Phase I") was
promulgated on November 16, 1990 (EPA, 1990). The provisions addressing municipal
separate-storm-sewer systems (MS4s)  cover systems serving a population  of 100,000
or more. This includes 173 cities, 47 counties  and  additional systems  designated by
EPA or states.  A total  of 260  permits, covering approximately  880  operators  (local
governments, state highway departments,  etc.) were  identified as subject to Phase I
permit application  requirements.  The CWA requires that  MS4  permits effectively
prohibit non-stormwater discharges  into  the storm sewers as  well as  reduce the
discharge of pollutants to the  maximum extent  practicable (including management
practices, control techniques and system,  design and engineering methods, and  other
provisions appropriate for the control of such pollutants).

Phase I MS4 permittees were required to submit an application that included source-
identification information, precipitation data, existing  data on the volume and quality of
stormwater discharges, a  list of receiving water bodies  and existing information on
impacts on receiving waters, a field screening analysis for illicit  connections and illegal
dumping, and other information. The  MS4 permittees were to gather and provide
additional information including:

• discharge characterization data based on quantitative data from 5 to 10  representative
locations in approved sampling plans; estimates  of the annual pollutant load and event-
mean concentration of system discharges for selected conventional pollutants and
heavy metals; a proposed schedule to provide  estimates of seasonal pollutant  loads;
and the mean concentration for certain detected constituents in a representative storm
event; and

-------
• a proposed management program including descriptions of:  structural and source
control measures that  are to  be implemented to reduce pollutants in runoff from
commercial and residential areas; a program to detect and remove illicit discharges; and
a program to control pollutants in construction site runoff.

The  NPDES stormwater regulations for the second  phase of  stormwater discharge
control ("Phase II") was proposed on January 9,  1998 (US  EPA, 1998c). EPA was
required to promulgate the Phase II rule in 1999 under a separate consent decree. The
proposal  designated two  classes  of  facilities to be  automatically  covered  on  a
nationwide basis under the  NPDES program:
(1) small  municipal separate  storm sewer systems located in urbanized  areas (about
   3,500 municipalities would be included in the program); and
(2) construction activities (pollutants include sediments and erosion from these  sites)
   that disturb equal to or greater than one and  less than five acres of land (about
   110,000 sites per year will be included in the program).
Those facilities designated  above would need  to apply for NPDES stormwater permits
by 2002. EPA is  anticipating that most permittees would be covered under general
permits.

Municipalities and others may be able to use retrofitting to comply with the new Phase II
rules by converting existing dry ponds (or a portion  thereof), used typically for flood and
drainage  control, to wet ponds.  This is consistent the Phase I management program to
use structural  and  source control measures to reduce  pollutants  in  runoff  from
commercial and residential areas.  Wet ponds, because of the permanent pool and
longer detention times are thought to remove more pollutants than dry ponds (e.g., SS
and BOD and COD removals) and reduce or eliminate discharges.

Some facilities that EPA is  proposing to cover under the new Phase II rule are already
subject to state and/or local stormwater management requirements.  The Rockland,
Maine case study  demonstrates how retrofitting was incorporated into the  Maine's CSO
Abatement Program permitting requirements.   Among other  planned  retrofits a new
force main was commissioned to separate an industrial loading from the municipal
sewer and the  wastewater treatment facility  is being upgraded to handle additional
WWF.

CSO Control Policy
The need for high-rate  treatment systems comes  from  the National CSO Policy that
requires treatment of additional WWF in existing POTW during high flow periods caused
by rainstorms.   The recent EPA National CSO Control Policy (59 Federal  Register
18688) resulted in the publication of several guidance documents. These two guidance
documents are 'Combined Sewer Overflow - Guidance for Nine Minimum Controls"
(EPA 832-B-95-003) and "Combined Sewer Overflow - Guidance for Long Term Control
Plan" (EPA 832-B-95-002).   Required CSO abatement technologies that are addressed
in this report are:

-------
*  Maxim ization of flow to the POTW for treatment;
*  Evaluation of alternatives that enable the permittee, in consultation with the NPDES
   permitting  authority, Water Quality Standard (WQS) authority, and the public,  to
   select CSO controls that will meet CWA requirements;
*  Cost/performance   considerations  to  demonstrate the  relationships  among  a
   comprehensive set of reasonable control alternatives; and
*  Maximization of WWF treatment at existing POTW treatment plants


Consensus Recommendation of SSO Federal Advisory Subcommittee

The EPA/State SSO work group distributed a set of draft NPDES permit regulations and
related guidance for municipal sanitary sewer collection systems and SSO to a small
municipal outreach group, the SSO Federal Advisory Subcommittee and other members
of the   public.   The   SSO   Federal  Advisory   Subcommittee  made  consensus
recommendations to EPA on:
•  capacity, management, operation and maintenance (CMOM) program requirements;

•  a permit prohibition for SSO;

•  permit record keeping, reporting and public notification requirements; and

•  peak excess flow treatment facilities.

The SSO subcommittee recommended to EPA the substance of the CMOM, Prohibition,
Record  Keeping,  Reporting  and  Public  Notification, Remote  Treatment  Facilities
documents, and  Satellite Collection Systems and watershed  management agreed
principals on October 20, 1999. In addition,  the Subcommittee recommended principles
for satellite collection systems and watershed management.

The protocols for the Federal Advisory Subcommittee provide that EPA is committed to
reflect the consensus  reached  by the SSO Subcommittee in an upcoming notice of
proposed  rulemaking.    While  the Advisory Subcommittee  made recommendation
addressing core  issues,  some  outstanding  issues remain.   The  consensus
recommendation  of the SSO Federal  Advisory Subcommittee clears the way for the
next step in the development of a proposed  rule.

Retrofitting an existing treatment plant is often a  more cost-effective way to control
additional  flow than building new facilities.   With the upcoming  SSO regulations,
investigating the  economics of retrofitting  existing  systems to treat more flow, may
become the first step in a compliance exercise.

-------
                                   Section 2
                                 Conclusions
Desktop analyses were performed for  13 separate retrofit examples.   The analyses
demonstrated  that  in  certain circumstances  retrofitting existing  WWF facilities  is
technically  feasible  and  may  be  more  cost effective than  construction of  new
conventional control and treatment facilities.  A summary of each retrofit is provided in
Table 2-1 and is discussed below.

Table 2-1. Summary of Retrofitting Desktop Analysis Projects
Project
Augusta WWTF
Hypothetical WWTF
Spring Creek AWPCP
Spring Creek AWPCP
Spring Creek AWPCP
Sunnyvale Detention Basin
Monroe St. Detention Pond
Birmingham Dry-Ponds
Erie Boulevard Storage System
Spring Creek AWPCP
Hypothetical Sanitary Sewer
System
Rockland WWTF
Retrofit
Headworks Upgrade - Enhanced
Primary Treatment
ACTIFLO System -Enhanced Primary
Treatment
Plate Settlers for Enhanced
Treatment in WWF Storage Facilities
Chemical Addition for Enhanced
Treatment in WWF Storage Facilities
Post-Storm Solids Removal for WWF
Storage Facilities
Retrofit for Enhanced Treatment
Retrofit for Enhanced Treatment
Retrofit for Enhanced Treatment
DWF Augmentation
DWF Augmentation
SSO Control Using Storage
Retrofitting for Industrial Wastewater
Control in a CSS
Treatment
Benefit
Yes
Yes
No
Marginal
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Cost
Benefit
Yes
No
No
Marginal
Not
Determined
Not
Determined
Not
Determined
Not
Determined
Marginal
Marginal
Yes
Yes
The  headworks of the Augusta,  ME WWTF were retrofitted to maximize flow to the
existing primary and secondary tanks and to minimize wet-weather bypass. This was a
very cost-effective alternative by  maximizing existing plant capacity.  This  retrofit
allowed for minimizing the construction of new high-rate treatment facilities.  However,
there are only limited experiences of such retrofitting documented in the literature.

Enhanced sedimentation methods employ flocculation with lamellar settling to speed up
the clarification process. One such system, the ACTIFLO, was hypothetically retrofitted
to a  primary treatment facility.  This technology provides enhanced SS removal with
short detention  times and  quick  start-up  as  compared  to  conventional  primary
clarification. One of the benefits of the ACTIFLO system is the extremely small footprint

-------
of the process.   The downside is  the  high cost of chemicals and  increased  O&M
requirements.

Retrofitting existing WWF storage tanks  for flow equalization and enhanced treatment
was not cost effective.  The Spring Creek Auxiliary Wastewater Pollution Control Plant
(AWPCP) in New York, NY was used to evaluate this retrofit option.  This is largely due
to the high variability of WWF and limitations of the technologies evaluated. Retrofitting
existing WWF storage tanks with cross-flow plate settlers was found to be technically
limiting due to the hydraulic loading rates of cross-flow plate settlers.  This technology
may be more appropriate for  primary clarifiers where hydraulic load rates are  more
consistent than in WWF storage tanks.

Retrofitting existing WWF storage tanks  with chemical addition to provide chemically
enhanced sedimentation of WWF  is a feasible option.  This technology  could result in
60—90%  SS  removal,  40—70%  BODS  removal,  and  30—60% chemical  oxygen
demand (COD) removal.  This technology does not affect the operation  of the storage
tanks to serve their primary role of equalization.

The AWPCP was also used to evaluate several retrofitting options for post-storm solids
removal from WWF storage tanks including tipping buckets, flushing gates, and orifice
headers.  The results of pilot testing and design considerations including capital and
O&M costs are provided for each technology.

Retrofitting existing dry-ponds to provide  treatment through sedimentation was found to
be  a relatively simple and viable treatment option.  Three dry ponds were  evaluated
from different  regions of the country: 1) Santa  Clara, CA,  2) Madison, Wl,  and  3)
Birmingham, AL.  The retrofits significantly reduced effluent flows and downstream
channel erosion  and  also provided a water quality benefit.  This  was done largely by
developing features to better control influent and effluent flows and to reduce short-
circuiting within the ponds.

The Erie Boulevard Storage System (EBSS) in Syracuse, NY and the  AWPCP were
evaluated for  the purpose of serving a dual  role  as  wet-weather and  dry-weather
equalization facilities.  This analysis showed that  this could be done without drastically
reducing the ability of the  storage tanks  to capture WWF.  The primary  benefit of
equalization provided by the  DWF storage is the dampening of  peak flow rates that
would  otherwise result at the  treatment plant, allowing  the primary and secondary
treatment facilities to operate at a more  uniform  rate thereby provide more consistent
treatment efficiency throughout a typical day. However, this retrofit is only economically
viable for a treatment plant that is operating at or near design  capacity where increased
O&M costs can be balanced against cost  to increase capacity at the plant.

The retrofit of an unused aeration tank  at  the Rockland WWTF in Rockland, ME and the
construction of a separate force main from an industrial facility to the WWTF provided a
cost-effective  alternative for  the  elimination  of  high-strength  wastes from  CSO
discharges.  This retrofit conveyed high-strength waste from the industrial facility to the
                                       8

-------
dry-weather  headworks thereby  eliminating the  potential  for  CSO  discharges  of
industrial wastewater and consequent impact on the receiving water.

Retrofitting SSOs to  use storage from  overflow  retention facilities provided a  cost-
effective alternative to expansion of the sanitary collection system to convey the volume
of overflow to the treatment plant for primary and secondary treatment.  However, this
alternative requires the availability of a sufficient amount of land to construct the storage
facility.

Retrofitting abandoned POTWs for use as WWF treatment facilities was shown to be an
effective treatment alternative.  The Auburn WWTF in Auburn,  NY was retrofitted using
the abandoned primary treatment  tanks  as WWF  storage tanks to reduce the impact
associated with WWF in excess of the capacity of the existing treatment plant.  This
retrofit was an extremely cost-effective alternative to other abatement options due to the
availability and condition of the existing facilities. The abandoned primary tanks were in
good condition and required only minor  modifications.  This retrofit  is a cost-effective
alternative in cases where abandoned primary tanks are available in good condition for
retrofitting. This type of retrofit may be a less cost-effective solution for cases where the
existing tanks are in poor condition requiring a substantial cost to upgrade the tanks to
serve a functional role.

The feasibility and  cost effectiveness of retrofitting is in large part controlled by site-
specific conditions  and treatment requirements.  The physical condition  of existing
facilities, age, and level of maintenance, size, layout, hydraulic factors, water quality and
other site-specific factors all affect the suitability for retrofitting.  As such,  the opportunity
for retrofitting existing  WWF facilities  must be evaluated on a case by case basis.

-------
                                    Section 3
                               Recommendations

The findings of this study indicate that retrofitting existing control facilities can be a cost-
effective alternative  to the construction  of conventional  treatment facilities.    The
appropriateness of  retrofitting  is  however,  a  function of  site-specific  needs  and
conditions.  Therefore, it is recommended that retrofitting existing control facilities be
identified as an alternative for wet-weather abatement. Retrofitting should be evaluated
early on in the decision making process for wet-weather abatement.

The advanced sedimentation methods such as the ACTIFLO merit further study given
the results  for high-rate  treatment  applications in  Europe  and  due to the limited
experience  in  North  America.    This could  include  controlled,  side-by-side  pilot
demonstrations to assess the benefits and cost effectiveness of alternative advanced
sedimentation methods.
                                       10

-------
                                   Section 4
                             Methods and Materials

4.1    Literature Search

Introduction

Typically, urban infrastructure collects, conveys, and treats DWF.  In the case of a CSO
community, a component of WWF is collected and conveyed in the same infrastructure
as the DWF, but points  of relief are purposely provided before treatment because of
hydraulic constraints of  the conveyance system and the capacity of the dry-weather
treatment facilities.  In the case of an SSO community, the DWF (sanitary sewage) and
the stormwater are collected and conveyed in different systems.   However,  these
separated sanitary systems inadvertently  receive  I/I which  result  in overflows and
consequent pollutant concerns.  Regardless of how WWF are manifested, the essential
infrastructure is made up of the collection, conveyance, and treatment systems. In each
of these systems,  retrofitting specific structures can  provide improved  control and
treatment for WWF.  Often surprisingly simple and inexpensive retrofits  can provide
expanded capacity,  enhanced treatment,  extended facility  lifespan, and reduced
operating costs.

This literature review focuses on retrofits made to collection, conveyance, and treatment
systems in order to better control and treat wet-weather pollution. A majority of the cited
literature was found through queries in such databases as,

•  Environmental Sciences & Pollution Management Database
•  Compendex*Plus
•  Wilson Applied Science & Technology Abstracts
•  ProCite Database,
•  National Center for Environmental Publication and Information

Other  literature  was found through requests  for information  made to  consulting
engineering firms.  An emphasis was placed on national conference proceedings.  The
referenced databases are described in Appendix A.

General Retrofitting Information

Martin (1988) developed  a method to help  government  watershed planners and
managers to:  (1)  assess   the  pollutant  potential of  runoff from  existing  urban
development, (2) screen for feasible control measures, and  (3) develop and  implement
retrofit management  strategies.  This method strongly prompts the  implementation of
watershed retrofits as remedial actions for stormwater pollution.

Walesh (1992) discusses the potential for retrofitting stormwater facilities.  The  paper
summarizes  the history of urban stormwater detention facilities and suggests that many
of these existing facilities can be retrofitted to improve quantity control, include quality
                                      11

-------
control,  improve O&M, reduce safety hazards, enhance aesthetic attributes and have
recreational features.   Beale (1992) and England  (1995) also discuss retrofitting
stormwater facilities.   Both authors address design considerations  for stormwater
retrofitting techniques.  Such techniques include detention ponds, retention ponds,
underground  vaults,  exfiltration  pipes,  and  baffle  boxes.   Palhegyi,  Driscoll,  and
Mangarella (1991) evaluated the feasibility of  retrofitting stormwater control facilities to
improve pollutant-removal performance.  This  paper presents the results of an analysis
that was performed to predict the removal efficiency of suspended sediments in storm
runoff under existing conditions and under several practical retrofit alternatives.  A flood
risk analysis  was  performed for  each  condition.   It was concluded that stormwater
quality could  be improved by modifying operating strategies of existing facilities,  but a
flood risk analysis must be considered.

Field and O'Connor  (1997) discuss a strategy to optimize CSO control systems.  The
optimization strategy maximizes the use of the existing system and sizes the  storage
facility in accordance with the WWTF capacity to obtain the lowest-cost storage and
treatment system.  Roesner and Burgess (1992) discuss CSO  rehabilitation strategies
for urban areas. Roesner and Burgess address and cite examples for eliminating dry-
weather overflows, rehabilitating regulators, and optimizing existing interceptor capacity.

Collection System Retrofits

The practice  of utilizing inlet control concepts to reduce the impacts of  flooding from
urban  runoff  and  pollution  through catchbasin  adjustments  is  common.    Most
installations involve restriction of catchbasin outflow to cause surface inflow to be re-
directed  by overland routes to  more  suitable discharge points  or to  underground
storage.  The majority of installations involve retrofitting  existing systems to mitigate
collection-system  surcharging,  reduce  exfiltration  and  infiltration  exchange within
adjacent sewer systems, and to cause reduction of downstream CSOs (Pisano, 1990).

The area known as the "Triangle"  in Parma, Ohio, is a  12.1 ha (30 acre) low-relief area
in a 117.4 ha (290 acre) drainage system.  In  the "Triangle" storm  drains are generally
inadequate and severe street flooding  and  basement  flooding  are common.   Inlet
controls such as overland flow berms, vortex  throttles, and shallow drains to de-water
street areas were implemented throughout the  entire 117.4 ha (290 acre) drainage area.
The project mitigated surface-water ponding and provided basement-flooding protection
throughout the entire 117.4 ha (290 acre) area.   Construction  costs  in  1984  were
$875,000 (Pisano,  1982).

Lake  Quinsigamond, Worchester,  Massachusetts,  was  becoming eutrophic  from
stormwater discharges.  Studies  indicated that the  dominant nutrient  source  was
contaminated  DWF from the storm sewer.   It was  believed  that this contamination
resulted  from the mixing  of the sanitary  sewer  with the storm sewer due to  the
"over/under" sewer configuration.  In this configuration the sanitary sewer was placed
over the stormwater sewer in the same trench.  A system wide  pipe replacement or
rehabilitation was rejected based on cost.  Therefore, a number of system  controls were
                                       12

-------
installed; these were found to immediately eliminate much of the nutrient loading from
the storm sewer base flow.  A vortex flow throttle was installed in the storm sewer to
divert the polluted base flow and the first flush to a nearby combined  sewer.   Inlet
controls were then installed in the upper reaches of the combined system to increase its
capacity to convey the polluted base flow and the first flush.  Approximately 0.34  m3/s
(12 cfs) of stormwater (base flow and first-flush) were diverted to the combined system.
Total construction cost were estimated at $160,000 (Pisano, 1989).

Conveyance System Retrofits

The conveyance system is the infrastructure  component designed  to move water from
the site of origin to an acceptable receiving  water body.  In the case of dry-weather
sewage flow, the  conveyance system carries wastewater to a treatment plant.  In the
case of WWF, this water may be  discharged into a water body without treatment. In
many stormwater instances, detention  ponds are an integral part of the conveyance
system.  In the past, these detention ponds were used primarily for the purpose of urban
flood control. In some  instances, ponds are retrofitted for the capture and treatment of
stormwater, thereby improving the quality of the receiving water body.

Flow Equalization

Fairport  Harbor,  Connecticut,  developed a  unique  solution to eliminate  CSOs by
converting  an abandoned industrial fuel  oil tank into a  cost-effective wet-weather
retention tank. The fuel oil tank was utilized for flow equalization after the original  plan
to construct a new tank and pump station near the outfall met local resistance.  The
conversion of the fuel oil tank has  removed an environmental liability, saved the district
money, and reduced CSOs.  An initial examination of the 36.6 m (120 ft)  diameter fuel
oil tank indicated it had sufficient capacity 12,112 m3 (3.2 MG) for the intended capture
of the five-year intensity storm.  However, of primary concern were structural integrity
and the presence of industrial sludge and asbestos.  Rehabilitation of the tank began
with the removal of the  crude  oil sludge, interior piping,  asbestos-covered exterior
piping, and lead-based paint.   Welds  that  did not meet American Welding  Society
standards were replaced.  Utilization of the fuel  oil tank also required building a pump
station and force main designed for a peak flow rate of 19,870 m3/d (5.25 million gallons
per day or MGD).  The project total cost was $2.6 million (Shrout, 1994).

The Rohnert Park Demonstration  Project  included the design, construction, operation,
testing and evaluation of a surge facility designed to provide flow equalization and some
degree  of treatment  to all WWF  and to provide rate control of wet-weather and  dry-
weather wastewater flows to interceptor  sewers.  The  principal feature  of the surge
facility was a 2840 m3 (0.75 million gallon or  MG) sedimentation-equalization basin.  It
was tested under field conditions  with influent flows varying from 2,270—20,800  rrfVd
(0.6—5.5 MGD).  The ability of the surge facility to provide adequate hydraulic control to
diurnal flow variation was documented.  Under WWF conditions, the surge facility was
able to remove approximately 45% of the influent  SS.   BOD removal  under WWF
                                       13

-------
conditions was not significant.  The total construction cost for the demonstration facility
was $384,000 (Welbon, 1974).

Detention Basins

Santa Clara Valley, California  is protected from flooding by a series of pump stations
and levees.  Based on a preliminary evaluation of the feasibility of retrofitting detention
basins, a pilot study was implemented to retrofit a pump station and to conduct testing
to measure water  quality benefits  and costs.    In the  pilot study,  structural  and
operational retrofits were  implemented  at a  pump station consisting of four  primary
pumps rated at 1.1 m3/s (39 cfs) and one auxiliary pump rated at 0.25 m3/s (9 cfs).  The
detention basin area is 1.8 ha (4.4 ac.) with a 37,005 m3 (30 ac-ft) capacity and receives
water from a 187.4 ha (463 acre) urbanized watershed.  The structural retrofits included
installing a weir at the outlet of the detention pond, filling a channel that connected the
inlet and the outlet of the detention pond,  and blocking off a drainage pipe that ran
under the channel.  The  purpose of these  structural  retrofits was to  reduce short-
circuiting of the pond and provide better distribution of the flow into the outlet.   The
operational retrofit  consisted  of modifying  the pump schedule  to create  a two-foot
permanent  pool  in  the pond.  These retrofits provided greater  retention time in the
detention pond and therefore  enhanced gravitation settling of  SS.  Post-retrofit water
quality surveys showed removals of 29% for chromium, 42% for total copper, 53% for
lead, 51% for total  nickel,  44% for total zinc,  and 50% for SS. The  metals removals
correlated well with the SS removal.  The amortized cost over  20 years for  retrofitting,
operation, and maintenance was $8,200/year (Woodward-Clyde, 1996).

Dye-tracing experiments performed in a stormwater pond in Kingston Township,  Ontario
revealed short retention times,  mainly due to the  small size of the pond.  Mathematical
modeling of this pond  indicated irregular circulation patterns, resulting in short-circuiting
and dead zones.   A study was conducted  to  retrofit the pond for  the  purpose of
improving pollutant-removal  characteristics.   A series of three baffles were installed
along the length of the  pond,  roughly  perpendicular  to the  direction of flow,  and
extending an average of two-thirds of the width across the pond. The installation of the
baffles increased the length-to-width ratio of the flow path in the  pond from 1.5:1 to 4.5:1
and corrected the flow regime problems.  In addition, an increase in the hydraulic
efficiency of the pond (defined as the ratio of measured to volumetric retention times)
from 0.65—0.86 was observed. An increase in pollutant removal through sedimentation
processes was inferred from a  comparison of  retention time measured before and after
baffle installation (Matthews et al., 1997).

Florida's Indian River Lagoon suffers from drainage problems such as increased volume
of freshwater runoff  to  the  estuarine  receiving water and  deposition  of  organic
sediments,   reduced  water  clarity  because  of  increased discharge  of  SS,   and
eutrophication caused by  nutrient  loadings.   A  project was initiated  to create  a
watershed control system for the Indian River  Lagoon, develop  management strategies
to relieve stresses  resulting from runoff  to the lagoon,  and address the feasibility of
watershed  retrofitting  to reduce pollutant loads.   The partially developed,  768.9 ha
                                       14

-------
(1,900 acre), industrial area of NASA's Kennedy Space Center was selected as  a
representative drainage catchment.  After screening the retrofitting alternatives, it was
determined that constructing a weir in the main  drainage channel, with discharge of
diverted flow to a wetland area, would provide the best system performance. The cause
and effect relationships  between the catchment  hydrology, channel  hydraulics, and
pollutant loads were documented using a calibrated SWMM  model.  Model  results
concluded that the retrofit could  satisfactorily achieve flood control, increase  SS
removal, and maximize water depth, but  could have difficulty meeting the groundwater
discharge and water level-fluctuation criteria.  Without periodic drawdown  from  the
wetland,  water levels in the system would be nearly  static due to the nearly  flat relief.
Fixed cost of the proposed retrofit would be $28,000.  The annual O&M  costs were
estimated to be $17,100 (Bennett and Heaney, 1989).

Chemical Enhanced Treatment

As evidenced  by the above-referenced literature, many structural  retrofits have been
implemented to improve the treatment characteristics of wet-weather detention basins.
In addition to structural retrofits, some municipalities have experimented with chemically
enhanced wet-weather detention  basins to  improve the treatment characteristics of
these basins during wet-weather  events.  This  low-structural  retrofit  requires small
amounts of  chemicals  and polymer to  promote the coagulation  and flocculation of
smaller, lighter particles into larger, heavier particles.

The Washington State Department of Transportation designed and constructed a scale
model detention  basin to investigate contaminant-removal capabilities.  Using typical
contaminants  and  concentrations,  a  simulated highway stormwater  runoff  was
formulated and applied to the model detention basin over a  range  of flow rates. Four
coagulants were evaluated for their ability to enhance removal of sediment and metals.
Coagulant addition resulted in significant  increases in metals removal over the range of
stormwater flow rates studied.  The greatest  improvement was observed at the higher
flow rates. Further improvements  in contaminant removal were observed following  the
addition of an influent baffle.  This baffle increased the hydraulic detention time by
reducing short-circuiting with an associated improvement in contaminant removal (Price
and Yonge, 1995).

A restoration project began in 1987 in an effort to improve water quality in Lake Ella,
Tallahassee,  Florida.  This effort constituted the first  use  of alum for treatment of
stormwater inputs into a  receiving water body. Lake Ella is a shallow, 5.3 ha  (13 acre)
lake that received stormwater runoff through 18 separate storm sewers from 64.8 ha
(160 acre) of mostly impervious urban watershed.  With a volume of 113,550 rr?  (30
MG), Lake Ella receives 518,550 m3 (137  MG) of runoff each  year.  The lake was highly
eutrophic and fish  kills were common  before installation  of the alum  stormwater
treatment system.   Initially,  common  stormwater treatment  technologies  such as
retention basins and  exfiltration  trenches were considered.   Alum treatment was
considered when it was  determined that there was no available land surrounding  the
lake that could be used for retention and  the cost of purchasing homes to acquire land
                                      15

-------
for construction of retention basins was cost prohibitive.  The stormwater treatment
system was implemented using  flow meters  and variable speed injection pumps to
automatically inject liquid alum  into the  storm  sewers.   Mixing  of  the  alum  and
stormwater occurs as  a result of  turbulence in  the storm sewer  line, and  the floe
produced settles on the lake bottom providing an added benefit of nutrient inactivation in
the sediments.  In addition to the non-structural alum retrofit, bar type trash traps were
installed in underground structures  upstream of the lake; the primary outfall lines were
extended  into  the  lake  about  30.5  m (100 ft).   The total cost of  the alum  injection
system, excluding  the  extension of the storm sewer  lines into the lake, was about
$200,400.  The capital cost of this system was $3,140/ha  ($1,270/acre) of watershed
area treated.  The capital costs involved in constructing an alum treatment system are
relatively independent of watershed size and dependent primarily upon the number of
outfall locations treated (Harper and Herr, 1992).

Polymer injection has also been  used for the  control of sewer overflows in the city of
Dallas, Texas.  An EPA-sponsored research  program  studied a non-structural retrofit
system wherein the capacity of a  sewer was increased by the injection of certain water-
soluble chemicals to reduce turbulent friction. A polymer injection rate of 0.54 kg/min
(1.2  Ibs/min)  into a 38 cm (15  in.), surcharged  pipe  conveying  18.9 m3/min  (5,000
gallons per minute  [gpm]) reduced head from 2.5 m (8 ft) above the pipe to 1.8 m (6 ft)
above the pipe. The capacity of a sewage  lift station was also increased by injection of
a slurry of friction-reducing polymer into the pump intake.  Polymer doses of  120 and
350 ppm  were sufficient at increasing flow  in a 15.2 cm (6 in.) force main by 35% and
64%, respectively (Chandler and Lewis, 1977).

Real-Time Control

Real-time control systems are gaining a renewed interest for the control of wet-weather
overflows.  In 1967, the U.S.  federal government funded demonstration projects to
investigate available technologies for control  of  CSOs.  Demonstration  grants were
awarded  to sewer agencies  in  Cleveland, Ohio; Seattle, Washington; Minneapolis,
Minnesota; and Detroit, Michigan (Chantrill, 1990). These projects used computers for
the coordination of storage at different sites in a sewer collection system  to optimize use
of storage on a system wide basis.  All of the projects experienced  problems getting the
central computers operational largely due to the limitation of computer technology at the
time.  Today,  municipalities have  many more options for implementing computer-based
real-time control of wet-weather overflows.

Delattre (1990) outlined real-time control of combined  sewer systems from the user's
perspective. Delattre's discussion included design objectives and operating functions,
implementation of supervision and control  schemes,  and evaluation of operation
performance thorough case studies. Chantrill  (1990) discussed the history of real-time
control of combined sewers in the  United  States.  Chantrill's discussion  included the
control software  developed   in  Seattle,  Washington  and  Lima, Ohio,  and  the
development efforts by Cello Vitasovic and Wolfgang Schilling.
                                       16

-------
Treatment Facility Retrofits

Increasing Capacity/Efficiency of Existing WWTF

Retrofitting dry-weather treatment plants to handle increased flows from  wet-weather
events has proven to have merit in some communities.  Typically it  involves increasing
the capacity of  the  existing treatment  plant  by installing  newer,  more  efficient
equipment.

The town of Dexter, New York, solved an infiltration problem by upgrading a treatment
plant with new aeration units rather than rehabilitating the collection system. Peak flows
of more than 2,271  m3/d (0.6  MGD) from  wet-weather events  and  groundwater
infiltration  were washing bacteria out of the 379  m3/d (0.1  MGD) activated sludge
WWTF. The town of Dexter  decided to treat the excess infiltration, but did not have
enough space at the plant to build a larger mixed-liquor system to handle excess flows.
Installing  prefabricated fixed-film  treatment  units that offered improved  treatment
efficiency and quick bacterial growth (Myers and Wickham, 1994) solved this problem.

The city of Clatskanie, Oregon, conducted an infiltration/inflow (I/I) study and found that
the collection system was subjected to severe I/I.  Additionally, deficiencies existed in
the trickling filter plant. A cost-effectiveness analysis indicated  that  the development of
a joint dry/wet-weather facility without major collection  system improvements  was the
best solution. The principal features of this alternative included a single primary clarifier
followed by an activated sludge secondary treatment process with  a DWF capacity of
1,893  m3/d  (0.5  MGD).  The  secondary process can be  operated in a  contact-
stabilization mode, and the primary  clarifier can be operated in a dissolved air flotation
(DAF) mode.  With these process modifications, flows up to 4,731 rrfVd  (1.25 MGD)
could be effectively treated. SS and BOD removals during DWF were both 94%. WWF
removal efficiencies for flows ranging from 1,893—8,706  rrfVd (0.5—2.3  MGD) were
71% and  73%  for SS and BOD  respectively.  The  capital cost of the  DAF-contact
stabilization capability was estimated to be 14% more than the cost of a standard dry-
weather plant (Whitney-Jacobsen and Associates, 1981).

Modifications to the clarification system at the main WWTF in New  Orleans, Louisiana
have  resolved  a  recurring hydraulic  overloading  problem  caused  by inflow  and
infiltration.  The modifications also have increased average waste activated sludge
concentrations  by 50%.   Sluice gates  were installed  to direct flows  equally to the
clarifiers during normal flows.  These gates are adjusted during high flows to allow for
optimal flows through  the clarifiers, thereby eliminating sludge-blanket scouring that
occurred before the gates were installed. A polymer injection system was also added to
the thickener-clarifierto further enhance settling (Berry, 1994).

The Acheres WWTF in Paris, France was designed to treat the carbonaceous portion of
dry-weather sewage.   The city is currently undergoing  a major  rehabilitation  and
retrofitting plan to  upgrade the plant to include  nutrient  removal and  wet-weather
treatment.  By  the year 2001, the  plant will be retrofitted with tertiary treatment for
                                       17

-------
nutrient removal and physical-chemical treatment for wet-weather treatment.  The wet-
weather treatment facility was designed to treat 0.76 rrfVmin (200  gpm),  almost twice
the 0.42 m3/min (110 gpm) DWF (Gousailles, 1995).

Physical-Chemical Treatment

The physical-chemical treatment system at the Acheres wastewater plant will use the
ACTIFLO  process.    This  process   incorporates  lamellar  settling  and  weighted
flocculation by microsand.  Guibelin, Delsalle and Binot (1994)  presented operational
and performance results from the ACTIFLO process.  They established that the process
allows an upflow velocity of 131 m/h (0.12 fps) over the lamella. Under this flow regime,
80% SS and 60% BOD removal is achieved using 60g/m3 of ferric chloride and 0.8g/m3
of polyelectrolyte.  The process is also noted for its quick start  up time,  an important
criteria for treating WWF.

Another physical-chemical treatment system  being studied in France is the Densadeg
unit (Le  Poder and  Binot,  circa  1995;  Rovel  and Mitchell circa 1995).   This  unit
combines three  principles:  lamellar  settling,  coagulation-flocculation,  and  built-in
thickening.  The Densadeg pilot unit has been studied  at WWTF and at representative
sites along  a combined sewer system. The  researchers documented SS removals of
80% at a velocity of 50 m/h (0.5  fps) and noted that coagulant  dose (FeCb) affected
treatment results more than velocity.   Densadeg units are to be installed at the new
Colombes sewage treatment plant (240,000 m3/d [63.4 MGD]) in Paris, France.  A key
feature of the Colombes is how it will be operated during wet-weather events.  Rather
than sequential operation of the biological stage as during dry-weather, the stages will
switch to operate in parallel during periods of high flow (Hayward,  1996).

Nenov (1995) compared  the effectiveness and costs of physical-chemical  treatment
versus the  activated  sludge process.  Laboratory experiments confirmed the high
efficiency of physical-chemical treatment for SS and  BOD/COD removal.   The data
obtained show that chemical treatment over a large range of surface flow rate provides
a reduction of SS and BOD/COD greater than 60% and 50%, respectively.   Based on
capital and O&M costs from the  Ravad WWTF  it was shown that physical-chemical
treatment retrofitting could  be done at a  much  lower cost than biological  treatment
expansion.


Chemical Enhanced Treatment

As  an  alternative  to   physical-chemical  treatment,   other  municipalities  have
experimented with  chemically enhanced  treatment to increase  plant capacity during
wet-weather events.

Sarnia, Ontario, studied the effectiveness of chemically enhanced primary treatment to
increase  and sustain  removal efficiencies during high plant flows for more than ten
years at  an existing  primary plant.  Very little  decrease in  treatment efficiency was
observed at overflow rates as high  as  114.1  m3/d/m2 (2,800 gpd/ft2).  The study
                                      18

-------
concluded that, with the addition of 14 ppm of ferric chloride and 0.3 ppm polymer, the
SS and BOD removals averaged 84 and 60%, respectively (Heinke,  Tay, and Qazi,
1980).

In March 1990, a 22 day test of chemically enhanced treatment was conducted at the
South  Essex Sewerage   District   (SESD)   primary  treatment  plant  in  Salem,
Massachusetts.  The testing conditions were unique in that the flow was split and the
plant was operated as two distinct and parallel treatment lines: half of the plant received
chemical additives while the other half operated as a conventional primary plant.  BOD
removals on  the  conventional primary side  averaged 17%  but were 51%  on the
chemically  treated side  (a 200% increase).    SS, fats,  oils and  grease removals
increased by 60%. The test also demonstrated chemical addition can enable settling
tanks to handle  maximum stormwater conditions  without deterioration  in removal
efficiency.   Ferric chloride and polymer concentrations totaled  approximately 30 ppm.
Average overflow  rates of 32.6 m3/d/m2 (800 gpd/ft2) reached  a peak of 77.4 m3/d/m2
(1,900  gpd/ft2) during wet-weather events. The SESD plant's average  DWF of 94,625
m3/d (25 MGD) increased to over 340,650 m3/d (90 MGD) during a storm on the ninth
day of the test. SS and BOD removal rates did not decrease during or after this event
(Harleman, Morrissey, Murcott, 1991).

Converting  Dry-Weather WWTF to a Wet-Weather Facility

Conversion of existing or abandoned dry-weather treatment plants into wet-weather
holding or treatment facilities  is an option when communities build new dry-weather
treatment plants.

The  Municipality  of Toronto Metropolitan  (Metro)  has recently completed the Main
Treatment Plant (MTP) Environmental Assessment (EA). One of the goals of the  EA
was  to establish a preferred  alternative for meeting CSO treatment needs.   Metro
placed  a high priority on source control as a means of reducing CSO's and associated
pollutant loadings.  Sewer  separation was not considered as the only solution to the
CSO problem because of concerns relating to increased stormwater discharges, the
legality of separation on private properties, and high costs.   The preferred strategy
included, among  other solutions, conversion  of a WWTF to a CSO treatment plant
(CG&S, 1997).
4.2   UseofSWMM

SWMM is a large, complex model capable of simulating the movement of precipitation
and  pollutants  from  the  ground  surface  through  pipe  and  channel  networks,
storage/treatment  networks, and finally to receiving waters.  The model  is capable of
simulating a single event or on a continuous basis for extended periods of time.

SWMM has been  released under several different official versions (Metcalf and Eddy,
Inc.,  et al. 1971; Huber et  al. 1975, 1984; Roesner et al. 1984; Huber and Dickinson
                                      19

-------
1988; Roesner et al. 1988).  The official versions were primarily designed for mainframe
computer use; however, later  versions  have been  modified  for use  on  personal
computers.  The current "official" version, version 4, is available on disks or through the
Internet from the EPA at the following addressees:

      Center for Exposure Assessment Modeling (CEAM)
      U.S. Environmental Protection Agency
      960 College Station Road
      Athens, Georgia 30613
      Phone:      706/355-8400
      E-mail:      ceam@epamail.epa.gov
      WWW:      ftp://ftp.epa. gov/epa_ceam/wwwhtml/ceam home, htm

The complete  SWMM model documentation (Huber and Dickinson 1988; Roesner et al.
1988) can be obtained from  the National Technical  Information Services (NTIS) or
Oregon State University:

      National Technical Information Service
      5285 Port Royal Road
      Springfield, Virginia 22161
      Phone:      703/487-4650
      NTIS no. PB88 236 641 - Huber and Dickinson (1988)
      NTIS no. PB88 236 658 - Roesner et al. (1988)

      Dr. Wayne C. Huber
      Department of Civil Engineering
      Oregon State University
      Corvallis, Oregon 97331
      Phone:      503/737-4934
      E-mail:      huberw@ccmail.orst.edu

The SWMM consists of five basic blocks (or components) that can be used together or
separately.   The  blocks  are RUNOFF,  TRANSPORT,  EXTENDED TRANSPORT
(EXTRAN), STORAGE TREATMENT (S/T) and RECEIVING WATER (RECEIVE). The
RUNOFF  Block  simulates  rainfall,  the  resulting  surface  runoff  quantities,  and
hydrographs for each  drainage basin  of the combined sewer collection system.  The
EXTRAN Block routes the surface runoff hydrographs developed in the RUNOFF Block
through the collection system to the point of interception and overflow.

SWMM was used  to perform simplified simulations using historical rainfall data for  a
period of one  year or the average of  multiple years of hourly rainfall  data with hourly
time  steps or an  appropriate equivalent.   This was  done using the RUNOFF and
TRANSPORT Blocks of SWMM.  The  RUNOFF Block generates surface runoff
hydrographs from  a drainage area in  response to precipitation based on watershed
characteristics including size,  shape, imperviousness,  and slope. The  RUNOFF block
                                     20

-------
uses the continuity equation that tracks the volume and depth of water on the ground
surface and Manning's equation that governs the rate of surface runoff.

These  hydrographs were then used as input into the TRANSPORT Block that routes
flow through a sewer system that is depicted as a series of manholes, conduits,  pump
stations,  storage  tanks,  and  overflow structures.   TRANSPORT uses a simplified
version of the momentum  equation where  flow is a function of depth to  route flow
through the collection system.

Unless otherwise noted, simplified simulations conducted for these analyses were done
using existing calibrated and validated RUNOFF and TRANSPORT models, modified to
estimate the  pre- and  post-retrofit  conditions  to   assess the  treatment/capture
effectiveness of the retrofit.

4.3 Case Studies and Hypothetical Retrofit Examples

The desktop analyses provided in this document are a mixture of case studies of actual
retrofit  examples and hypothetical retrofit examples where actual retrofit examples were
not readily available.  The case studies  include site  specific details of each retrofit
including description of facilities, operational parameters, flow rates,  and costs.  The
hypothetical examples are used  to  provide  a  conceptual  overview  of  retrofitting
opportunities and  allow comparison to  other retrofitting examples. In some instances,
actual  case studies were labeled  as  "hypothetical" to allow the use of information
without revealing  the source  to  protect confidential information.   In these  instances,
actual information was used to the greatest extent possible to  increase the accuracy of
the desktop analysis.
                                      21

-------
                                    Section 5
                               Desktop Evaluation

5.1    Conversion of Primary Treatment Tanks

Conversion or retrofitting primary settling tanks at POTWs to provide equivalent or higher
levels of treatment may be a technically feasible and economically attractive alternative to
the  construction of new conventional primary  settling tanks.   The principal downside of
conventional  primary treatment facilities is the large footprint required due  to  the  low
surface overflow rate (SOR).  Alternative  sedimentation methods including micro carrier
enhancement, chemical precipitation, dissolved air floatation, and plate/tube settling may
provide an alternative to conventional primary treatment.

Two desktop analyses are presented to demonstrate the retrofit of primary treatment tanks
to provide equivalent or higher levels of treatment:

   1.  Retrofit of the Augusta Wastewater Treatment Facility, Augusta, ME

   2.  Hypothetical retrofit of the POTW using the ACTIFLO process

5.1.a. Augusta Wastewater Treatment Facility, Augusta, Maine

The Augusta, Maine Sanitary District (ASD) owns and operates a secondary wastewater
treatment facility (WWTF) that  discharges to the  Kennebec  River, which is  a Class C
waterbody for the state of Maine.  The WWTF was originally constructed in 1962 with
upgrades in 1966, 1982 and 1997. The ASD WWTF provides secondary treatment using
pure-oxygen activated sludge. The WWTF was designed to handle a monthly-average flow
of 30,280 m3/d (8 MGD) and an instantaneous-peak flow of 60,560 m3/d (16 MGD).  The
WWTF serves a partially combined collection  system (CCS). ASD's existing collection
system  services approximately 90%  of  the  residential population, nearly  all  of  the
commercial properties, and a portion of the industrial properties within the limits of the city
of Augusta. The system also serves four neighboring communities, the city of Hallowell and
the towns of Manchester, Wnthrop and Monmouth.  The system contains three types of
sewers:

    •  combined sanitary and storms
    •  separate sanitary
    •  storm

The majority of the collection system within the limits of Augusta flows  by gravity to the
WWTF.   Ten pumping stations exist serving the lowlands and outlying areas.   Currently,
there are over 167 km  (104 miles) of gravity sewer ranging from 15—107 cm (6—42 in.) in
diameter.  Pipe materials include stone and brick (dating back to the Civil War era) as well
as more  recent installations of vitrified clay, asbestos-cement, concrete, cast iron, and
PVC pipe.  According to the 1993 CSO Facilities Plan there are presently 38 diversion
                                       22

-------
manholes and 29 permitted CSO located within the system.  In an average year 219,530
m3 (58 MG) are discharged.

A majority of the wastewater currently entering the ASD sewer system is domestic sewage
from residential and commercial  sources.  The original WWTF design  included an
allowance for a significant organic loading from  local industry; however, many of the local
industries closed or relocated soon after Secondary Treatment System upgrade became
operational  in the early 1980's. Consequently, the WWTF is operating with less than one-
half of its original design flow and organic loading.  Currently, annual average wastewater
flows are 15,140 m3/d (4 MGD).

Prior to implementation of the CSO Abatement Program, the WWTF used the following unit
processes to provide preliminary, primary and secondary treatment:

   •  One  mechanically-cleaned bar screen with 3.8 cm (11/4n.) clear spacing located in a
      1.5 m (57 in.) wide channel

   •  One  detritor-type grit chamber 4.3 m (14 ft) in diameter

   •  Two  91.5 cm (36 in.) comminutors with a manual bypass screen

   •  Three primary clarifiers,  of which two are  16.8 m (55 ft) in diameter and one is 24.4
      m  (80 ft) in diameter, for a combined surface area of 910 m2 (9,800 ft2)

   •  Four main primary effluent lift pumps

   •  Two  pure-oxygen activated sludge aeration tanks with a total volume of 2,725 m3
      (0.72 MG)

   •  Three secondary clarifiers, each at 24.4 m (80 ft) diameter with a total surface area
      of approximately 1400 m2 (15,100ft2.)

   •  One  chlorine contact tank with a volume of 613 m3 (162,000 gal)

Figure 5.1.a-1 shows the  existing site layout prior to  implementing the CSO Abatement
Program.  Figure 5.1.a-2  shows a schematic of the existing WWTF prior to retrofitting.
When influent peak flow approaches 34,065 m3/d (9 MGD),  excess flow is bypassed to the
chlorine contact tank following degritting in the detritor.   The bypass was automatically
activated by a side overflow weir located along the grit chamber outlet channel. The water
level was regulated by a 30.5 cm (12 in.) Parshall Flume located downstream of the weir.

The CSO Abatement  Program approved by the Maine  Department of Environmental
Protection recommended a four-phase plan to address CSO discharges.  Phase I of the
CSO Abatement Program  included improvements to provide treatment for WWTF bypass
that is the third largest CSO in the CCS based on projected peak flows. The following is a
summary of Phase I Improvements,  which  is a  retrofit to the  existing  facilities:
                                       23

-------
Figure 5.1 .a-1.                    Prior to
                                    •*f
                      .' /'-t|  \ /CT\  .--"
                   11;  ___  xA-^./?"i|V
                                      24

-------
Figure 5.1 .a-2.           Flow             Prior to
                            111  5;     M
                              '-*i '•* •• '•*••  I" -•  ""*:•  1
                              fc '•!.' -1  '-V  ;l.:;s
                        II.  i:i  ''  !
                        •-, (4 •-   «r /    i,




                             *,
                                 .?   ?  >ii  IS
                                       --  - • -j j>|

                                  4   *   A
                                   * i   > *     _"~" ~" r~—- ^* * *
                                                25

-------
    •  A new headworks  containing two mechanical screens with 2.5 cm (1  in.) clear
       spacing located in a 1.5 m (60 in.) wide channels, two 167 m3 (44,000 gal) aerated
       grit chambers, and one 0.9 m (36 in.) parshall flume. The mechanically cleaned bar
       screens have  screenings compactors  to  remove excess water.   The grit is
       removed from  the  aerated  grit chambers  using recessed impeller type pumps
       which discharge to cyclone classifiers for washing and dewatering.

    •  A new flow distribution structure (FDS No. 1) for the primary clarifier influent.  The
       distribution structure utilizes  motorized weir gates that will allow off-line tanks to be
       activated during a wet weather event.

    •  A new primary effluent distribution structure (FDS No. 2) with motorized weir gate
       to regulate flow receiving secondary treatment and flow to the CSO bypass.

    •  A new 250 m3  (66,000 gal) chlorine contact tank with mechanical mixers  using
       sodium  hypochlorite for high-rate  disinfection.  A 24 m3 (6,400 gal) bulk storage
       tank  is  provided for sodium hypochlorite storage.   An  11  m3  (3,000 gal)
       dechlorination contact chamber with mechanical flash mixer is provided.  A 16 m3
       (4,150 gal) bulk storage tank is provided for sodium bisulfite. Chemical  metering
       pumps are provided with flow pacing from an ultrasonic flow meter.

These improvements will allow the  full 109,800 nf/d (29 MGD) of connected interceptor
capacity to receive preliminary and  primary treatment.  Daily-maximum flows of 45,400
m3/d (12 MGD) would receive secondary treatment while 64,300 m3/d (17 MGD) would be
diverted to the high-rate disinfection system after receiving primary treatment.

Later phases of the CSO abatement program (Phases II, III and IV) will allow additional
WWF from the Augusta CCS to be treated at the WWTF.   This will be accomplished
through the construction of:

    •  Consolidation conduits to intercept CSO discharges throughout the system and
       convey the flow to the WWTF;

    •  Three new vortex separators, each 10.7 m  (35 ft) diameter, and 2.4 m (8 ft) side
       water depth to provide the equivalent of primary treatment;
       Additional high-rate disinfection and  dechlorination tank capacity to increase the
       tank volume to
       dechlorination;
tank volume to 1,250 m3 (330,000 gal) for disinfection, and 62.5 m3 (16,500 gal) for
    •  A new wet-weather outfall from the WWTF to the Kennebec River.

After all phases are completed, the total influent peak flow capacity to the WWTF would be
increased from 109,800 m3/d (29 MGD) to 405,000 m3/d (107 MGD). Up to 295,200 m3/d
(78 MGD) will be treated in the vortex separators.  The remaining 109,800 m3/d (29 MGD)
would be conveyed to the upgraded headworks. Approximately 268,700 m3/d (71 MGD) of
vortex separator  effluent will  be disinfected and dechlorinated in the expanded high-rate
                                       26

-------
disinfection tank.  Approximately 26,495 m3/d (7 MGD) of vortex separator underflow would
                                                           3/
be conveyed to the upgraded headworks. A total flow of 136,260 m /d (36 MGD) would be
                                                       3/
treated within the headworks and primary clarifiers (109,765 m /d or 29 MGD influent plus
26,495 m3/d or 7 MGD vortex underflow).  Primary clarifier effluent flows of approximately
45,420 m3/d (12 MGD) would receive secondary treatment.  The remaining 90,840 nf/d
(24 MGD) of primary clarifier effluent will receive high-rate disinfection and dechlorination.
The basis of design of these facilities is based upon SWMM results using a one-year, two-
hour storm event.

Figure 5.1.a-3 shows a site  layout following completion of the CSO Abatement Program.
Figure 5.1.a-4 shows a schematic of the ASD WWTF following completion of the CSO
Abatement Program.

It should be noted that the  1993 CSO Facilities Plan is currently  being updated.  The
updated  plan will  re-assess the earlier recommendations for the later phases of the
program.

Stormwater Management Model Results

The EPA SWMM Version 4 was used to assess the net benefit of the retrofit by comparing
annual WWF volume  treated at the WWTF under pre- and post-retrofit  conditions.  A
simplified model was developed to simulate hourly rainfall, based upon the previously
calibrated model developed for individual rainfall events.

The data used for these projections consists of 21 years of hourly  rainfall data from the
National Weather  Service (NWS) station in Portland,  Maine for the years from  1971—
1991.  These data were used as input to the  simplified RUNOFF  model  that projected
flows  that were used  to simulate  long-term  overflow using the  TRANSPORT block of
SWMM.  The TRANSPORT model was constructed as a simplified  network consisting of
overflows and regulator pipes. The simplified model was used to project the annual volume
of treated WWF at the WWTF.

Based on this analysis, the  average  annual WWF conveyed to the treatment plant was
found to be 582,890  rrf (154 MG).  The layout of the pre-retrofit headworks allowed
484,480 m3 (128 MG) to receive treatment consisting of primary and secondary treatment
followed by disinfection while 98,410 m3  (26 MG) of WWF annually bypassed the treatment
plant and only received disinfection prior to discharge.  The retrofitted headworks allowed
the entire 582,890 m3 (154 MG) to receive primary treatment and 484,480 m3 (128 MG)
received secondary treatment and disinfection.  The remaining 98,410 m3 (26 MG) that did
not receive secondary treatment was bypassed to the high-rate disinfection system  prior to
discharge to the Kennebec River.
                                      27

-------
Figure 5.1 .a-3.
                                                                                             pv-.jt
                                                                                             a1 - ^
                                                                                             n
                                    i   '. IS1*  ^  ^=*  ***''  .     "'\
                                    :   ^, f a  f;:':l  |l% r.';0   **      •
                                                      28

-------
Figure 5.1 .a-4.
                                             J|	Ji
                                £i*.»  g   :  4



                                     A     .»





                                 |* li. 't1 ''i-'  -i'*"


                                                                     
-------
Design Considerations

The retrofitting of the ASD WWTF to handle increased WWF required consideration of
several factors:

      •  Pollutant and Hydraulic Loadings
      •  Treatment Objectives
      •  Operational Requirements
      •  Chemical Storage and Feed Requirements
      •  Sludge Handling Requirements
      •  Operations and Maintenance Requirements
      •  Construction  Sequencing and Site Constraints

Each of these considerations is described in this section.

Pollutant And Hydraulic Loadings - Pollutant and hydraulic loadings were developed for
the wet-weather  treatment units at  the ASD WWTF during the Facilities Planning stage.
These values were refined and updated as  the project progressed through the preliminary
engineering  and detailed design  stages.   The  primary  objective of the  wet-weather
treatment  units  is to  remove  floatable  and  settleable  solids  to  promote  effective
disinfection.

Table 5.1.a-1 shows  the design criteria for the wet-weather treatment processes at the
ASD WWTF. The primary focus of the Phase I  improvements are to remove hydraulic
restrictions at the WWTF preventing flows  of 109,800 nf/d (29 MGD) from  reaching the
existing primary clarifiers. As noted previously, the completion of Phases II,  III and IV will
provide  equivalent primary treatment and disinfection for additional  peak  WWF of
approximately 295,200  m3/d (78 MGD).

Table 5.1 .a-1. Phase 1  CSO Abatement Program Wet-Weather Treatment Design
      Criteria
Description and Criteria
Vortex Treatment
Number of Units (each)
Design Flow
Diameter Each
Loading Rate
Overflow to
High-Rate Disinfection
Underflow to Headworks
Phase I Phases II, I

0 3
0.30 m3/d
10.67m
1098m3/d/m2
268,735m3/d
26,495 m3/d
III & IV

78.0 MGD
35.0ft
27,000 gpd/ft2
71.0 MGD
7.0 MGD
Mechanical Screening
No. of Bar Screens (ea.)                  2                              2
Type                          Reciprocating Rake                 Reciprocating Rake
BarSize                   0.95x5.1cm         3/8x2 in       0.95x5.1 cm         3/8x2 in
                                       30

-------
Bar Spacing
  2.54 cm
  1 in
  2.54cm
  1 in
                                    Table 5.1.a-1. Continued
Description and Criteria
Peak Influent Flow
Vortex Overflow
Total Peak Flow
Approach Velocity
Half Clogged
Screen Velocity
Half Clogged
Differential Head
Estimated Wet Screening
Quantity
Average De-watered
Screenings Volume
Screenings Container Size
Screening Storage
Time (days)
Phase I
109,765m3/d
0.0
109,765m3/d
0.76 m/s
2.07 m/s
27.43cm
3.7x1 0'5m3/m3
0.28 m3/d
.57 m3/d
29.0 MGD
29.0 MGD
2.5 fps
6.8fps
0.9ft
(5ft3/MG)
0.37cyd/d
0.75 cyd
Phases II, II
109,765m3/d
26,495 m3/d
1 36,260 m3/d
0.84 m/s
2.29 m/s
33.22 cm
3.7x1 0'5m3/m3
0.57 m3/d
0.57 m3/d
II&IV
29.0 MGD
7.0 MGD
36.0 MGD
2.75 fps
7.5 fps
1.09ft
(5ft3/MG)
0.74cyd/d
0.75 cyd
Grit Removal
No. of Tanks
Type
Average Flow
Peak Flow
Volume Each Tank
Detention Time -
Avg. Flow (min)
Detention Time -
Peak Flow (min)
Air Supply

No. of Air Blowers
Air Flow Each

Pressure
Blower Motor Hp

Grit De-watering & Storage
Type
No. of Classifiers
Flow
No. of Grit Pumps
Pump Head
Pump Speed (rpm)
Pump power
Grit Washer Diameter
Grit Washer Speed (rpm)
Estimated Grit Quantity
Grit Container Size
Grit Storage Time (days)
Description and Criteria
2
Aerated
I 4.0 MGD
d 29.0 MGD
44,000 gal

Aerated
30,280 m3/d
1 36,260 m3/d
167m3
2

8.0 MGD
36.0 MGD
44,000 gal
 15,140m7d
109,765m3/d
   167m3
            31.7

             4.4

   0.27 to          3 to 8 cfm/ft
0.72 m3/min/m
         2+1 standby
   2.4 to           84 to 224 cfm
 6.3 m3/min
 3867 kg/m2         5.5 psig
       5/10, two speed
           Cyclone
             2
  1 m3/min           260 gpm
         2+1 standby
   8.5m              28ft
             900
        7.5kW(10HP)
  30.5cm             12 in
             18
 45 cm3/m3          6.0 ft3/MG
   4.6 m3             6 cyd
             7
           Phase I
                              16.8
                              3.5
                   0.27 to
               0.72 m3/min/m
                   3 to 8 cfm/ft
                          2+1 standby
                  2.4 to           84 to 224 cfm
                6.3 m /min
                3867 kg/ m2
                    5.5 psig
                        5/10, two speed
                            Cyclone
                               2
                 1 m3/min            260 gpm
                          2+1 standby
                  8.5m               28ft
                              900
                         7.5 kW (10 HP)
                 30.5cm              12 in
                              18
                45 cm3/m3          6.0 ft3/MG
                  4.6 rrr
                      6 cyd
                        Phases II, III & IV
Primary Clarifiers
Average Flow
 15,140m3/d
4.0 MGD
30,280 m3/d
8.0 MGD
                                                31

-------
Peak Flow
BOD Loading 1
SS Loading 1
No. Of Tanks (ea.)
                             109,765m3/d
                               3220 kg/d
                               5443 kg/d
                   29.0 MGD
                   7,100lbs/d
                   12,000 Ibs/d
                                136,260 m3/d
                                 7258 kg/d
                                  7348 kg/d
                    36.0 MGD
                   16,000 Ibs/d
                   16,200 Ibs/d
                                    Table 5.1 .a-1 Continued
Description and Criteria
Diameter Each (ft)

Sidewater Depth
Total Volume (MG)
Average Hydraulic Loading
Peak Hydraulic Loading
Average Weir Loading
Peak Weir Loading
Detention Time-
Average Flow(h)
Detention Time-
PeakFlow(h)
Estimated BOD5 Removal1
Estimated SS Removal1
Estimated Sludge
Concentration (mg/L)
Estimated Sludge
Quantity @4.0%1
Primary Sludge/Scum
Pumps
Pump Type
Pump Flow Each
Total Head
Pump Speed (RPM)
Pump power
No. of Grinders
           Phase I
                                         Phases II, III & IV
 2 @ 16.7m
 1 @24.4 m
   2.1 m
  1938m3
16.7m3/d/m2
 121 m3/d/m2
 83.2 m3/d/m
603.4 m3/d/m
                 2 @ 55 ft
                 1 @ 80 ft
                  7.0ft
                  0.512
                409 gpd/ft2
               2,966 gpd/ft2
               6,701 gpd/ft
               48,584 gpd/ft
             3.1

             0.4

  1352 kg/d          2,980 Ibs/d
  2994 kg/d          6,600 Ibs/d
           40,000

  75m3/d         0.0198 MGD

             5
            Disc
 0.3 m /min
                 85 gpm
4.6m              15ft
         300
     3.7 kW (5 HP)
          5
 2@16.7m
 1 @24.4 m
   2.1 m
  1938m3
33.3 m3/d/ m2
150m3/d/m2
166.5m3/d/m
749.0m3/d/m
                                                                                     2 @ 55 ft
                                                                                     1 @ 80 ft
                                                                                       7.0ft
                                                                                       0.512
                                                                                     818 gpd/ft2
                                                                                    3,682 gpd/ft2
                                                                                    13,403 gpd/ft
                                                                                    60,311 gpd/ft
               1.5

               0.3

  2205 kg/d          4,860 Ibs/d
  3674 kg/d          8,100 Ibs/d
             40,000

   92 m3/d          0.0243 MGD

               5

              Disc
  0.3 m3/min           85 gpm
   4.6m                15ft
              300
          3.7 kW (5 HP)
               5
CSO Disinfection
Chemical Type
Peak Flow
Dose Rate at Peak Flow
(mg/L)
Chemical Concentration
     Sodium Hypochlorite
64,345m3/d          17.0 MGD
            15.0

            15.0
                                       Sodium Hypochlorite
                                359,575m3/d          95.0 MGD
                                              15.0

                                              15.0
CI2 Feed Rate
No. of Pumps
Pump Capacity each
Contact Time (min)
Contact Chamber Volume
Length to Width Ratio
Description and Criteria
149.8 kg/m3
               1.25lbs/gal
  149.8 kg/m -
                                                                                     1.25lbs/gal
        1 + 1 standby
  0.27 m3/h           70.0 gph
             5.5
   250m3          66,000 gal
             7:1
           Phase I
                                          2 + 1 standby
                                 0.76 m3/h           200.0 gph
                                               5.0
                                   1250m3          330,000 gal
                                              40:1
                                	Phases II, III & IV	
CSO Dechlorination
Chemical Type
Peak Flow
Chlorine Residual (mg/L)
NaHSO3Dose (mg/L)
       Sodium Bisulfite
64,345m3/d          17.0 MGD
             5.0
            8.06
                                         Sodium Bisulfite
                                359,575m3/d          95.0 MGD
                                               5.0
                                               8.0
                                               32

-------
NaHSO3, Concentration (%)
                                   38.0                            38.0
Feed Rate NaHSO3           504.4 kg/m3      4.21 Ibs/gal      504.4 kg/m3       4.21 Ibs/gal
No. of Pumps                      1 + 1 standby                      2+1 standby
Pump Capacity each           0.05 m3/h         12.0gph       0.12m3/h          32.0 gph
Contact Time (s.)                       15.0                            15.0
Contact Chamber Volume        11.4m3         3,000 gal         62.5m3         16,500gal

Treatment Objectives. The state of  Maine  requires an  Escherichia coliform (E-coli)
concentration of 949 colonies per 100 ml to comply with the maximum daily E-coli limit in a
Class C river.   Sampling and monitoring performed during the development of the CSO
Abatement Program indicated that the  Kennebec River does not  currently meet water
quality criteria  for bacteria during  wet weather.   The  CSO  Abatement  Program
recommended a plan to treat CSO discharges to meet the current maximum-daily-water-
quality limit.  Primary treatment and high-rate disinfection were selected as  the  most
economical methods to meet the treatment objectives for this project.

Operational Requirements.  The Phase I Improvements to the Augusta WWTF designed
to operate on a  continuous basis include:

      •   Mechanical Screening
      •   Grit Removal
      •   P ri m ary Treatm ent

The high-rate disinfection and dechlorination systems are intended to operate in  an event-
based mode during periods  of excessive WWF.  Under current  DWF conditions, the
WWTF operates with only one primary clarifier in service. The other two primary clarifiers
can be kept out of  service and used as off-line storage during wet-weather events.  If one
24.4 m diameter (80 ft) primary clarifier is in service, the  remaining two 16.8  m (55 ft)
diameter primary clarifiers provide approximately 943 m3 (249,000 gal)  of off-line storage.
The WWTF staff can activate the off-line tanks using the new Supervisory Control and Data
Acquisition (SCADA) System Controls.  Motor-operated weir gates regulate the flow to the
primary clarifiers.

After the WWF subsides the primary  clarifiers can be drained  using  a  new pump  with
variable speed controls. The primary clarifier tank drain pump discharge is returned to the
new Primary Clarifier Flow Distribution  Structure (FDS No. 1).

The future improvements of Phases II,  III and  IV will use an event-based mode of operation
during wet weather periods.  The existing overflow pipes located within the CSS will be
intercepted by new consolidation conduits that will convey the combined sewerage to the
WWTF. A series  of vortex separators will be constructed to provide equivalent primary
treatment (removing floatable and  settleable solids) to facilitate effective disinfection.  The
high-rate disinfection and dechlorination tank will  be expanded with a new parallel  outfall
pipe provided to the Kennebec River from the WWTF.  At the end of the  wet-weather event
the vortex separators and high-rate disinfection tank will be  drained using a pump  with
variable speed control to the WWTF headworks.
                                        33

-------
Chemical Storage and Feed Requirements. The CSO Abatement facilities will utilize two
chemicals:

      •   Liquid Sodium Hypochlorite for disinfection
      •   Liquid Sodium Bisulfite for dechlorination

The liquid sodium hypochlorite system design will use on commercial-grade liquid sodium
hypochlorite (15% solution) which contains the equivalent of 150 kg of chlorine/m3 (1.25 Ibs
of chlorine/gal of solution).  The metering pumps are designed to deliver a dosage of 15
milligrams per liter (mg/L) at  the peak design flow with one unit out  of service.  A bulk
storage tank with an effective working volume of 240.2 nf (6,400 gal) is  provided.  The
tank is  used for both  dry-weather, secondary  effluent disinfection,  and  high-rate wet-
weather disinfection to minimize problems associated with product shelf life. The available
chlorine  in sodium hypochlorite solution declines with age.  The  anticipated sodium
hypochlorite storage time will be 51  days  initially, decreasing to 25 days at the future
design flow.

The storage system design is based on commercial grade liquid sodium bisulfite (38%
solution) which contains approximately 500 kg  of sodium  bisulfite per cubic meter of
solution (4.2 Ibs of sodium bisulfite/gal of solution).  The metering pumps are designed to
deliver sodium bisulfite dosages necessary to dechlorinate effluent with a chlorine residual
of 5.0 mg/L.  Approximately  1.6 kg of sodium  bisulfite will be used per kg of chlorine
residual. A  bulk storage tank with an effective working volume of 15.7 m3 (4,150 gal) is
provided.

The  sodium  hypochlorite  and sodium bisulfite metering pumps are flow paced.   An
ultrasonic level sensor is used for effluent flow measurement. A broad-crested rectangular
weir 3.4 m (11 ft) in length is the primary measuring device.

Sludge  Handling  Requirements.   Sludge  handling requirements will increase during
periods  of WWF.  The most pronounced increase is expected to occur during the initial
period  of WWF  (first flush).  The  configuration of FDS No. 1 will capture a significant
portion of the first flush solids in the off-line primary clarifiers.

A  solids balance was  performed to  estimate the quantity of sludge produced  during a
storm with a one-year recurrence interval. The daily sludge quantities increase as follows:

      •   Current Estimated Sludge Quantity (kg/d)                 3,290 (7,250 Ibs/d)
      •   One-Year Storm Estimated Sludge Quantity (kg/d)         7,550 (16,650 Ibs/d)

The  sludge  quantity is estimated to increase  230%  during a  one-year storm event.
However, over the course of a week, the overall increase  can be assimilated if sludge
pumping and processing schedules are adjusted.  The following is a summary of weekly
sludge quantities during a period with a one-year storm:

      •   Initial Estimated Sludge Quantity (kg/wk)              23,010 (50,730 Ibs/wk)
                                        34

-------
      •  One-Year Storm Projected Sludge Quantity (kg/wk)     27,270 (60,140 Ibs/wk)


The estimated sludge quantity would only increase by 19% during a week with a one-year
storm event.

Operation and Maintenance Requirements. The implementation of the CSO Abatement
Program will result in increased O&M costs.  The following list summarizes the areas of
expected O&M cost increases at the WWTF:

   •  Labor to perform sampling, laboratory  analyses, reporting, wet-weather process
      adjustments, and clean-up following an event.
   •  Power for in-plant pumping, sludge processing, and chemical feed systems.
   •  Chemicals  for disinfection, dechlorination,  and polymer for additional  sludge
      quantities.
   •  Sludge disposal tipping and transport fees.

Table 5.1.a-2 summarizes  the expected increase  in annual O&M  costs at the WWTF
resulting from the  initial Phase I and future Phases II, III and  IV of the CSO Abatement
Program.

Table 5.1 .a-2. Estimated O&M Costs, CSO Abatement Program - All Phases	
  Description	Estimated Annual Cost
  Sampling                                                        $800
  Laboratory Analyses                                             $3,200
  Report Preparation                                               $1,500
  Process Control Adjustments                                      $1,200
  Clean-Up of Wet-Weather Treatment Structures                      $6,000
  Power                                                          $1,000
  Sodium Hypochlorite                                             $4,100
  Sodium Bisulfite                                                  $2,000
  Polymer for Sludge Dewatering                                     $1,000
  Sludge Disposal Transport and Tipping Fees	$11,500	
	Total Estimated Cost	$32,300	
Construction  Sequencing  and Site Constraints.   The existing ASD WWTF  site is
crowded, with little room for expansion.  Therefore,  it was necessary to construct some of
the CSO Abatement Facilities  in locations previously used for other purposes.  The
following is a listing of special  sequencing and site considerations related to the CSO
Abatement Program at the WWTF:

   •  Portions of the existing screen structure and detritor grit chamber were demolished
      to facilitate the construction of the new headworks.
   •  A temporary junction structure and CSO bypass was needed.
                                      35

-------
   •  The existing primary clarifiers were taken out of service in sequence,  to facilitate
      new interconnecting piping and mechanical refurbishment.
   •  The existing comminutor structure required temporary piping connections to keep it
      in operation during the new headworks construction.
   •  After the new headworks are built and placed into service, the existing comminutor
      structure will  require demolition  and  removal to facilitate installation  of new
      underground piping.

A completion time of 540 calendar days was specified for the Phase I  CSO Abatement
Program.

Project Costs

The bids for the Phase  I CSO Abatement Program were opened on January 23, 1997.
Costs for the CSO Abatement components are listed in Table 5.1.a-3.  Costs included
construction  requirements  and  project  services.   Project  services  included  design
engineering,  construction engineering, legal,  and fiscal  expenses  but did not  include
contingencies, right-of-way expenses, or other miscellaneous items.

Table 5.1 .a-3. Phase 1 CSO Abatement Program Project Costs	
      Description	Cost1	
      General Conditions                                     $200,000
      Sitework and Yard Piping                                $344,000
      Headworks Building and Chemical Feed Systems        $1,688,000
      CSO Disinfection and Dechlorination Structure             $536,000
      Primary Pump Room Renovations                         $80,000
      Primary Clarifier Renovations                             $40,000
      Flow Distribution Structure No. 1                          $117,000
      Flow Distribution Structure No. 2                          $122,000
      Operations Building Pump Room  Renovations               $60,000
      Yard Electrical Ductbanks and Lighting                     $88,000
      Instrumentation and SCADA System                      $167,000
                               Total Construction Cost      $3,422,000
    	Project Services2	$761,000	
                                    Total Project Cost      $4,183,000
      1 Based on January 23, 1997 Bid Prices
      2Includes Design Engineering, Construction Engineering, Legal, and Fiscal
       Expenses; does not include contingencies and right-of-way costs.

Conclusions
This retrofit will utilize the entire capacity of the existing primary clarifiers to provide primary
treatment for the maximum rate of influent flow, thereby increasing the portion of the WWF
that will receive primary treatment from 60,560 m3/d (16 MGD) to 109,765 m3/d (29 MGD).
This will result in net 20% increase: 484,480—582,890 m3 (128—154 MG) in the annual
                                       36

-------
WWF that receives primary treatment.  Only a few examples of retrofitting WWTF  have
been published.  Often these examples are strongly influenced by site-specific constraints
and therefore do not lend themselves  to comparison with other retrofit situations.  The
retrofit solution described here has been selected based on the cost effectiveness.  For
comparison purposes, the alternative treatment scheme of off-line storage at the WWTF for
the 13 MGD of WWF was estimated to be $8,336,000 versus the $4,183,000 for the
selected retrofit.
5.1.b. Hypothetical Retrofit of Primary Treatment Facilities  Using the  ACTIFLO
      System

ACTIFLO System

This  desktop analysis involves  the retrofit of hypothetical  primary treatment tanks  for
enhanced treatment of WWF in excess of the 908,400 m3/d (240 MGD) peak wet-weather
capacity of the existing WWTF. The treatment method selected for this retrofit scenario is
the ACTIFLO process that combines microsand-enhanced flocculation with lamellar plate
settling.  The process has been  shown to  be effective in treating  CSO/SSO and primary
wastewater flows in installations  in Paris,  France (Le Poder and Binot 1994). Typical
performance results are 85-95%  reduction  in SS, 50—80% reduction in Total BOD, 85—
95% reduction of total phosphorus and 10—20% reduction in total kjeldahl nitrogen (TKN).

The hypothetical WWTF has eight primary  clarifiers that are 41.2 m (135 ft) in diameter
with a 3 m (10 ft) side water depth. The surface-settling rate of the existing units  at the
design peak  flow of 908,400 rrf/d (240 MGD) is 85.6  rrf/d/m2 (2,100 gpd/ft2) with a
hydraulic detention time of 48  min.   Figure 5.1.b-1  shows the existing layout of the
hypothetical WWTF prior to construction of the Phase I.

Due to the limited side water depth, retrofit of the existing primary clarifiers is not practical.
A  more economical approach is the demolition of  one  of the  existing  clarifiers and
replacing it with  a 908,400 m3/d (240 MGD) ACTIFLO system which has a depth of 6.7 m
(22 ft).  The seven remaining clarifiers would be retained for storage of WWF during storm
events or for future plant modifications.

The proposed 908,400 m3/d (240 MGD)  ACTIFLO system design  consists of six 151,400
m3/d (40 MGD) process trains.  Each of the process trains is designed to operate at a
nominal overflow rate of 1.2 m3/min/m2 (30 gpm/ft2) with a  total hydraulic retention time of
5.6 min.

Table5.1.b-1. ACTIFLO Design Summary
       Design Parameter	Quantity	
       Number of Process Trains                        6
       Nominal Train Capacity           151,400 m3/d         40 MGD
       Total System Capacity            908,400 m3/d         240 MGD
                                       37

-------
 Hydraulic Retention Time
                               5.6 min

        I 35'tT(°E.!*

   ' ', •  css^yv^ it r "•
..'f"8*:.. ,.i ; •  typi ^  .. j *.,—j ;      .
»•*-!,.!»• «"- J "\      .•'{'•_.'      5
                     »TKi55ifn:.

[  *^tf*?r  ;  \        }  :  "w^  .  •*«
\ :-*- M'T^S .'  \ "^=Jei^4''1!!5 .'  3   I      **
; ••"" - f -•'   ', ,   .-\  i      '•


        5.1 .b-1.          of               WWTF       to                of ACTIFLO

System
                                          38

-------
The high overflow rate and short retention time offered by the ACTIFLO process result in an
extremely compact system. Overall, the proposed ACTILFO design will meet the treatment
objectives of the project in approximately 5% of the space currently occupied by the
existing primary clarifiers. A brief comparison of the existing facilities and the ACTIFLO
design are shown in Table 5.1.b-2.

Table 5.1 .b-2. Comparison of ACTIFLO and Existing Primary Treatment Systems

                                                                   Net Savings with
	Parameter	Existing System	ACTIFLO	ACTIFLO	
 Total Capacity              908,400 nf/d            908,400 nf/d
 Number of Units                 8                     6
 Unit Capacity              113,550 nf/d            151,400 nf/d
 Area Required           1,329.1 m2 (per unit)      87.8 m2 (per train) 526.7       94% per unit
                        10,632.7m2 (total)            m2 (total)              95% total
 Process Volume          3,987 m3 (per unit)        3,533 m3 (per unit)         85% per unit
                        31,898 m3 (total)         3,588.3 m3 (total)           89% total
 Hydraulic Retention             51 min                  5.6 min                 89%
 Time

A layout drawing of the hypothetical WWTF showing the location of the proposed ACTIFLO
system is provided in Figure 5.1 .b-2.

ACTIFLO Process Theory

Fundamentally, the  ACTIFLO process  for  wastewater treatment  is  very similar to
conventional water treatment technology of coagulation, flocculation  and sedimentation.
Although  influent characteristics vary for different wastewater applications and  influent
streams, the removal of suspended materials is accomplished by the same mechanisms in
each application.   Therefore,  for the  purposes of this discussion, the term wastewater
(WW)  collectively refers to any domestic wastewater stream  including primary sewage,
CSO, SSO and/or inflow and infiltration (I/I).

As in conventional water treatment technology, a coagulant is used for the destabilization of
suspended  materials entering the process and a  flocculent aid  polymer is  added to
aggregate the  destabilized solids into larger masses.  The resulting  floes are then
subsequently removed  by settling for  disposal.   The primary  advance made in the
ACTIFLO process is the addition of very fine sand as a "seed" for the development of high-
density floe.  The resulting floes, ballasted by the microsand, are more  easily removed by
settling.  It  is in this step that ACTIFLO differs significantly from conventional treatment
processes.  A  brief overview of the physiochemical  principals  involved  in conventional
coagulation  and flocculation is presented in "Section 5.2.b. Chemical Addition to AWPCP"
to help understand the advantages of ACTIFLO over chemical coagulation alone.
                                        39

-------
                                                      'Cn,.

           -^  ..            ;       r

            %- ',' v *5 : /      -•' 'M v.vrr^ 3 pi  ,
          Cfl3'  •.• Xw^.,.y  _„ c™  r *: \ •*? < :  ™  !
            -x^ st-vawt
                        H ; S ?*

                          •_' \
5.1 .b-2.        of          ACTIFLO         in
                                  40

-------
The primary advance made in the ACTIFLO process is the addition of a very fine (60—140
jam mean diameter) clean silica sand or microsand in the flocculation process step.  With
its addition, the microsand serves several key roles in the ACTIFLO process:

  •  The high specific surface area to volume ratio of the microsand particles serves as a
     seed for the development of floe.

  •  Together, the microsand  and  polymer seed promote  inter-particle contact and
     enmeshment of suspended materials resulting in development of large stable floe.

  •  The high microsand concentration within the ACTIFLO process effectively dampens
     the effects of fluctuating  influent quality  and  provides for stable  operation and
     treatment throughout changes in the raw water quality.

  •  The relatively high specific gravity of the microsand (~2.65) serves as a  ballasting
     agent creating heavier floe that settle very quickly.

  •  The chemically inert microsand  does not react with  the process chemistry and allows
     the microsand to  be efficiently separated from the sludge via a hydrocyclone and
     reused in the process.

The advantages of microsand ballasted flocculation allow ACTIFLO to offer surface loading
or overflow rates many times higher than competing processes.  Overflow rates as high as
3.3 m3/min/m2 (80 gpm/ft2) are  claimed in typical wastewater applications of ACTIFLO.
The high  loading  rates  result  in a process that is extremely  compact compared  to
conventional or competing processes of similar capacity. The high overflow rates and size
savings provided by ACTIFLO translate directly into significant savings in project cost.

The use of microsand for ballasted flocculation provides ACTIFLO with several operational
advantages  in wastewater applications.  The high solids  content within  the  ACTIFLO
process provides for effective treatment of influent with varying SS concentrations. In such
applications, the high microsand concentration within the process allows for SS removal by
particle enmeshment  in the  microsand-polymer  complex.  This  performance is also
extremely beneficial in  chemical precipitation  reactions  such as  tertiary phosphorus
removal/effluent polishing or the  treatment of dilute waste streams.

Microsand ballasted flocculation is  also effective in the  treatment of extremely dirty
wastewater  (high  turbidity and SS).   Here,  the high  microsand concentration within the
process allows it to provide  efficient removal  of high-suspended solids  concentrations
without difficulty.   For  similar  reasons, the process is virtually unaffected  by sudden
fluctuations in  influent quality or volume. Minor adjustments to chemical dosage are all that
are needed to provide effective treatment under such conditions.

Overall, the use of microsand results  in the development of chemical floe that is significantly
denser and  more  durable than floe from  conventional  clarification  processes.   The
                                        41

-------
ACTIFLO floes have considerably higher settling velocities than conventional floe allowing
much higher clarifier loading rates.  Ballasted flocculation also  provides for a process that
is rapidly started-up and optimized, as well as stable in treating with variations in raw water
quality.  These characteristics make the ACITLFO process ideal for treating  the difficult
conditions posed by wastewater/CSO treatment.

ACTIFLO Process Description

The ACTIFLO® process is a  compact high performance water clarification system that
combines the advantages  of microsand  enhanced flocculation with lamellar settling.
Wastewater treatment is accomplished through a series of consecutive process steps that
consist  of  influent screening,  coagulant  addition, microsand  and  polymer injection
(injection), floe maturation (maturation), settling and sand recirculation.  Each  step  in the
process  is  discussed  in  greater detail  in  the  paragraphs  that  follow.   A flow
diagram/process schematic of the ACTIFLO system is provided in Figure 5.1 .b-3.



                                                   TO




Prior to entering the ACTIFLO process, raw wastewater is directed through fine screens to
remove large debris and prevent equipment fouling in the process. Debris removed by the
fine screening process is typically sent directly to a sanitary landfill for disposal. Chemical
coagulant is  then  added to the screened raw water in the influent line via an in-line
mechanical or jet mixing system  to provide for thorough and instantaneous dispersion of
the coagulant into the influent stream. The coagulant, typically ferric
                                        42

-------
chloride (FeCI3) in wastewater applications, serves to destabilize suspended materials as
they enter the process.

The coagulated water enters the ACTIFLO system in the injection tank. Here, flocculent aid
polymer (polymer) and microsand are added to  initiate floe formation.  Microsand and
polymer are incorporated into the coagulated water via mixing over 1  min of retention time.
Together, the microsand and polymer form ballasted  pin-floe that is allowed  to develop
further in the next process step.

Treatment continues as water  passes through the underflow passage  from the injection
tank into the maturation tank.  Although chemical floe formation actually begins with the
addition of polymer and  microsand in  the injection tank, the majority  of  ballasted floe
formation  occurs during the  maturation  process step.   Gentle mixing  and  increased
hydraulic retention time of approximately 3 min provide ideal conditions for the formation of
polymer bridges  between the microsand and  the destabilized SS. This  process is further
augmented by the  large specific surface area of  the microsand that provides enhanced
opportunity for polymer bridging and enmeshment  of suspended materials.

Fully formed ballasted floe leave the maturation tank and enter the settling tank  where they
rapidly settle and are removed from the treated water via lamellar settling. Laminar up flow
through the lamellar settling zone provides effective removal of smaller floe and  suspended
materials in the treated water.  Clarified water exits the ACTIFLO system  by  a series  of
collection troughs for subsequent treatment.

Sludge Handling. The ballasted floe sand-sludge  mixture is collected at the bottom of the
settling tank and  withdrawn using a rubber-lined centrifugal slurry pump.  The sand-sludge
mixture  is then pumped to the hydrocyclone for separation.   Energy  from  pumping  is
effectively converted to  centrifugal forces within  the body of the hydrocyclone causing
chemical sludge to be separated from the higher density microsand.  Once separated, the
microsand is concentrated and discharged from the bottom of the hydrocyclone  and re-
injected into the process for re-use.  The lighter density sludge is discharged from the top
of the hydrocyclone and sent for thickening and/or final disposal.

Coagulant Process Data.  As previously discussed, chemical coagulant is added to the
raw water to destabilize suspended materials  as they enter  the ACTIFLO system. To
provide effective treatment, coagulant should be injected in the  influent line upstream of the
system.   In-line  mechanical or jet injection  mixing,  depending  on  flow rate  should be
provided to ensure  nearly instantaneous  incorporation of the coagulant into the influent
stream.

Although any number of different coagulants could be  used,  FeCb is typically used  in
wastewater and  CSO treatment applications since it  is less  expensive, and  thus  more
economical in high capacity treatment applications than other available coagulants.

FeCI3 also offers the advantages of being less sensitive to pH and more aggressive  in the
removal of organics.  The only disadvantage to FeCI3  use is the need for corrosion
                                       43

-------
resistant coagulant dosing and mixing equipment in the influent line. Once incorporated
into the raw water, FeCI3 does not pose a serious corrosion threat to process equipment
within the ACTIFLO system.

Stormwater Management Model (SWMM) Results

The EPA SWMM Version 4 was used to assess the benefit of the  retrofit by comparing
annual WWF volume treated at the WWTF under pre- and post-retrofit conditions for the
hypothetical retrofit.  A simplified model was developed to simulate  hourly rainfall for the
hypothetical collection system.

The data used for these projections consists of 30 years of hourly rainfall data for the years
from 1962—1991.  These data were used as input to the simplified  RUNOFF model that
projected flows that were used to simulate long-term overflow using the TRANSPORT
block of SWMM.   The TRANSPORT model was constructed as  a simplified network
consisting of hypothetical overflows and regulator pipes.  The simplified model was used to
project the annual volume of treated WWF at the WWTF.

Table 5.1.b-3  demonstrates that the hypothetical ACTIFLO  system  would provide
considerable annual capture of WWF that is presently discharged without treatment.  This
is due to the use of the remaining seven primary treatment tanks as wet-weather storage
devices  with the ACTIFLO system.  These tanks have a combined storage capacity of
28,388 m3 (7.5 MG).

Table 5.1 .b-3. Average Annual Overflow Volume from the Pre- and Post-Retrofit Facilities
      (1962-1991)
                                       Average Annual Overflow Volume
Pre-Retrofit WWTF Overflow
Post-Retrofit
439,060 m3/d
34,065 m3/d
116MG
9MG
   Percent Capture                                   92%
The proposed ACTIFLO system in  combination with the remaining 7 primary clarifiers
modified for storage of WWF would provide a system to treat 92% of the WWF that is
presently discharged at the WWTF.

Project Costs

Operational Costs.  The primary costs associated with the operation of the ACTIFLO
system are electrical power and process chemicals.  The major power components in the
system are the mixers, scraper drive and microsand recirculation pumps.
                                      44

-------
Additionally, a small  amount of electrical power is required for the PLC control panels,
polymer mix and feed system, and coagulant feed system.  Wherever  possible, energy
efficient equipment is selected for use in the ACTIFLO system to help reduce overall
operating  costs.  In  wastewater/CSO treatment applications, the ACTIFLO system  is
typically operated at less than full capacity, further reducing electrical costs.

Chemical  costs are the other major expense associated with operation  of the ACTIFLO
process.  The chemical costs consist of primarily of coagulant and flocculent aid  polymer.
Depending on the raw water characteristics, pH adjustment chemical costs might also be
required in certain applications.  As with any water treatment process,  chemical  dosages
are difficult to predict without pilot or full-scale operating data from the  site. Furthermore,
chemical dosages are subject to vary with changes in the influent conditions. Therefore, a
range of typical  or anticipated chemical dosages is used to calculate operating costs.

Although chemicals are a major component in the overall operating costs of the ACTIFLO
process, the chemicals are used  very efficiently due to the mixing provided within the
system.  Overall,  ACTIFLO is  typically capable  of producing similar or  better treatment
results with 20—50% reductions in chemical usage as compared to conventional physical-
chemical treatment systems of similar capacity. The estimated  annual  O&M costs are
provided in Table 5.1 .b-4.

Capital Costs.   Capital costs estimates for retrofitting the proposed ACTIFLO  system into
the existing WWTF that were provided by the ACTIFLO manufacturer  (Kruger  Inc.) are
listed in Table 5.1.b-5.  The costs include construction  requirements and project services
including design engineering, construction management, legal and fiscal expenses but do
not include contingencies or other miscellaneous items.

The conventional abatement alternative for the hypothetical retrofit would be  to construct
wet-weather storage facilities  adjacent  to the  existing primary treatment  facilities to
equalize WWF for ultimate treatment at the WWTF. This would require the construction of
a 28,388 m3 (7.5 MG) overflow retention facility (ORF). The estimated cost to construct this
facility  based upon EPA cost guidelines for CSO control technologies (EPA 1992) is $18.4
million  derived from the following equation:

            Cost = 3.577V0812  (1)
            where,
            Cost = Average construction cost, millions of dollars
            V    = Storage volume, MG
                                       45

-------
Table 5. 1. b-4. Estimated O&M Costs
DESCRIPTION
Estimated Annual Cost
SAND COMSUMPTION
Est. Sand Loss (g/m 3 water
produced)
Design Flow Rate (m 3/d)
Sand Loss (kg/day)
Sand Loss (mt/y)
Sand Cost ($/ton)
Typical Operation Cost ($/y)
1
908,400 (240 MOD)
793(1, 748 Ibs/d)
263 (290 tons/y)
$88 ($80/ton)
$9,253
POLYMER COMSUMPTION
Est. Polymer Dosage (ppm)
Polymer Consumption (kg/d)
Polymer Consumption (mt/y)
Polymer Cost ($/mt)
Typical Operation Cost ($/y)
0.8
205 (452 Ibs/d)
68 (75 tons/y)
$3,858 ($3,500/ton)
$104,694
COAGULANT COMSUMPTION
Coagulant Type
Est. Coagulant Dosage (ppm)
Coagulant Consumption (kg/d)
Coagulant Consumption (mt/y)
Coagulant Cost ($/mt)
Typical Operation Cost ($/y)
Ferric Chloride
65
26,008 (57,501 Ibs/d)
8,650 (8,540 tons/y)
$220 ($200/ton)
$760,978
ENERGY COMSUMPTION
Injection Mixer kW per Train
Maturation Mixer kW per Train
Scraper kW per Train
Sand Pump kW per Train
kW per Train
Total kW
kW/1,OOOm3
KW/d
Typical Operation Cost ($/y)
Total Typical Operation Cost
($/y)
11 (15 HP)
15 (20 HP)
3.7 (5.0 HP)
37 (50 HP)
67 (90 HP)
403 (540 HP)
0.44 (2.25 HP/MGD)
7,725 (9,688 HP/d)
$107,090
$982,015
(1) Typical annual O&M costs are based on operation at 33% capacity 302,800 m3/d (80 MOD)
for 365 days per year with the provision for 75 wet-weather events lasting 12 h  each requiring
operation at 100% capacity.  75  wet-weather  events  per year  are  based upon long-term
continuous simulation of 30  years of data through a hypothetical collection system using the
TRANSPORT block of SWMM.
                                        46

-------
Table 5.1.b-5.  Estimated ACTIFLO Retrofit Capital Costs	
           Description	Estimated Cost(1)
           Equipment Cost                           $12,000,000
           Installation Cost                               $960,000
           Concrete Cost	$700,000

          	Total Project Cost   $13,660,000	
             Includes Design Engineering, Construction Engineering, Legal, and
            Fiscal Expenses based on information provided by ACTIFLO
            manufacturer, Kruger, Inc. Does not include contingencies and right-of-
            way costs
In comparison,  the  capital cost for the ACTIFLO  process is  less  costly  than  the
conventional storage facility, $14.5 million versus $18.4 million; however, the  projected
O&M cost of the ACTIFLO facility at almost $1 million per year make it a less cost-effective
alternative to storage.  This is a result of high chemical costs associated the ACTIFLO
process.

Typical ACTIFLO Performance in WW/CSO Treatment

ACTIFLO Pilot Testing Results.  A pilot plant with a nominal capacity of 150 nf/h and a
maximum capacity of 200  rrf/h was used by Kruger  to conduct the pilot testing  of the
ACTIFLO system.  The pilot plant consists of three mixing tanks with a total volume of 12.5
m3 and  two parallel hopper settling tanks  each  having a surface  area of 1.2 rrf.  Each
mixing tank is equipped with an axial flow mixer designed to allow for optimal coagulation
and flocculation. A rubber lined centrifugal pump extracts the sand/sludge slurry from the
bottom of the hopper settling tanks.   The major operating parameters of the pilot plant at
the nominal flow rate of 150 m3/h are summarized  in Table 5.1 .b-6.

Table 5.1 .b-6.  Major-operating parameters of ACTIFLO pilot plant at nominal flow rate of
      150m3/h
Parameters
Coagulation Tank Detention Time
Injection Tank Detention Time
Maturation Tank Detention Time
Settler Overflow Rate
Microsand Recirculation Rate
Microsand Concentration
Velocity Gradient (G) in Coagulation Tank
Velocity Gradient (G) in Injection Tank
Velocity Gradient (G) in Maturation Tank
Values
1
1
1
150
8
3—6
200—300
200—300
150—200
Units
min
min
min
m/h
%
kg/m3
1/s
1/s
1/s
                                        47

-------
Case Studies

Primary Wastewater Treatment Application - Wastewater Reuse for Agricultural Irrigation
in Mexico City, Mexico. Wastewater reuse for agricultural irrigation must be treated for the
removal of Helminth eggs and protozoa cysts, and elimination of pathogenic bacteria and
inactivation of enteric viruses.  Of great importance is the removal of Helminth eggs. They
are extremely  persistent,  surviving in harsh environmental conditions, and due to their
latency period, are transmitted primarily through uncooked food.  The Helminth disease
transmission has been identified as the top-priority heath problem of the wastewater reuse
in developing country.  The World Heath Organization (WHO) has recommended that the
Helminth egg concentration be under 1 Helminth egg per liter for all kinds of agriculture
(WHO, 1989).

After several studies,  the Mexican authorities  had determined that the best and most
economical way to reuse  the Mexico City wastewater for agriculture irrigation was to use
advanced primary treatment.  Due to the existing situation and the extremely large flow of
74.5 m3/s (1,700 MGD), the Mexican authorities  decided  to evaluate new treatment
technologies and organized pilot studies to see their performances. A six-month pilot test
was conducted to investigate the performance of the ACTIFLO process in 1997 (Le Poder
1997).

The results are summarized in Table 5.1.b-7 and in Figure 5.1.b-4 and 5.1.b-5 and show
that it was possible to obtain very low SS (25 mg/l in average) and Helminth  eggs residual
concentration (1 egg per test in average) even at very high overflow (up to  200 m/h) and
low chemical dosage (<60 ppm  of alum).  It confirms that the ACTIFLO process, even
without the downstream filtration process, is able to produce an effluent water with less than
1 egg/liter in daily average samples the majority of the time, and less than 5 eggs/liter in all
cases, which is the maximum amount permitted for certain kinds of agriculture.

Table 5.1 .b-7. Removal efficiency from 20 daily samples in primary treatment in Mexico
      City, Mexico

Influent
Effluent
Removal (%)
SS
mg/l
302
24.7
91.7
Helminth
Eggs
Eggs/I
24.3
0.99
96
Fecal
Coliform
/100ml
8x1 08
1.4x108
82.8
COD
mg/l
442
172
61.1
Total N
mg/l
18.4
15.5
15.8
Total P
mg/l
10.3
1.9
81.6
Sulfurs
mg/l
7.1
1.3
81.7
                                       48

-------
1 nn -f 	 	 orrt

.2
S -in -
ti
in
D)
D)
LU
CO
-^ 1 -
c
E
0)
X
.1 n
/
s | i i Rise Rate = Influent TSS S Effluent TSS
^**"" "'- . ™ ™
.«_







—
•t-







••-






—
•*


/\



A



\
\


—

V

n
^ —




•&-







	



-^






—*»«„


	










„— — -~








-^.




--_.--




""---*»,


	




^K





*-"-"*"


	



-*





/



3F







^ /////////// /
Date

•200
£~
• 150 £
QJ

• 100 a
cc
•50


Figure 5.1.b-4.
Mexico
               Helminth eggs removal in primary wastewater treatment in Mexico City,
700

600 -

500 -

400 -

300 -

200 -

100 -
        0
                               Rate
                           Influent TSS
                           Effluent TSS

                                         \
                                             X
                                                              X
                                                                       250
• 200


• 150


 100


 50
                                                                           (0
                                                                           OL
                                                                           0)

                                                                           2
                                      Date
Figure 5.1 .b-5.  SS removal in primary wastewater treatment in Mexico City, Mexico

CSO  Treatment Application in Galveston,  Texas, Pilot Study.  A pilot  study  was
conducted by COM (1998) in Galveston, TX to evaluate the effectiveness of the ACTIFLO
process under simulated  CSO conditions.  In order to  simulate CSO conditions raw
wastewater was pumped through a screen into a holding basin where it was subsequently
blended with the secondary effluent from the existing WWTF prior to entering the pilot unit.

The effect of coagulant dosage on SS, COD and BOD5 removal  efficiencies is shown in
Table 5.1 .b-8.  The pilot unit was operated at an overflow of 75 m/h,  a polymer dosage of
1.0 mg/l and a pH of coagulation of 6.0 while the coagulant  dosage (FeCb) was varied
from  75—125  mg/l.  As  shown in  the figure, the SS  and COD  removal  efficiencies
improved as the FeCI3 was increased from 75 mg/l to 100 mg/l.  The removal  efficiencies
of SS and BOD5 remained constant as the FeCb was increased from 100 mg/l to 125
mg/l. Therefore, the optimum coagulant dosage was approximately 100 mg/l.
                                      49

-------
Table 5.1 .b-8.  Effect of Coagulant Dosage on Removal Efficiencies in Galveston, TX
   FeCI3                 SS                       COD               BODS
   Dose    Raw   Settled     %     Raw  Settled   %    Raw   Settled    %
   Mg/l     Mg/l     Mg/l    Rem.   Mg/l    Mg/l    Rem.   Mg/l    Mg/l    Rem.
75
100
125
123
147
135
16.9
13.4
10.9
86%
91%
92%
500
600
300
112
105
105
78%
92%
65%
105
123
195
39
35
52
63%
72%
73%
In the event of CSO conditions a process that is primarily used during overflow periods will
have to start up and achieve steady state very quickly. Figure 5.1.D-6 illustrates the rapid
start up capabilities of ACTIFLO process.  The first SS and COD samples were taken after
two hydraulic retention times (18 min), the time needed to achieve steady state. After the
process reached steady state the SS and COD  removal efficiencies were 88% and 75%
respectively, and the average SS and COD removal efficiencies over the testing period
were 91% and 73%, respectively. The short overall retention time of the process allows the
rapid start up capabilities of ACTIFLO process.  During this experiment the pilot unit was
operated at an overflow rate of 75 m/h, a FeCI3 dosage of 100 mg/l and a polymer dosage
of 1.0mg/l.
    100
     90--
  o
  0)
  ce
  0)
  0.
  0)
  D)
  ro
     80
70-
     60
           The process was started up at 15:00
     50-
      15:00
         15:10
15:20
15:30
15:40     15:50
    Time
16:00
16:10
16:20
16:30
Figure 5.1 .b-6. Start-Up Efficiency of ACTIFLO Process

CSO Treatment Application in Cincinnati, Ohio Pilot Study. Kruger Inc conducted a pilot
study  using the ACITFLO® process for the Metropolitan Sewer District at the Mill Creek
WWTF in Cincinnati, OH (Kruger 1998).  The testing period was from April 14,  1998
through April 29,  1998.  This pilot study was conducted to demonstrate the effect of
overflow rate on the percent removal of turbidity, BOD5, and total phosphorous during CSO
conditions.  On April 17, 1998 the pilot unit was operated at a polymer dosage of
                                       50

-------
1.0 mg/l, a FeCI3 dosage of 35 mg/l while the overflow was varied from 75 m/h to 150 m/h.
Table 5.1 .b-9 depicts the effect of overflow rate on the removal efficiencies.
Table 5.1 .b-9. Effect of overflow rate on removal efficiencies, Cincinnati, OH
   Rise          Turbidity                BOD5             Total Phosphorous
   Rate    Raw   Settled    %    Raw  Settled     %     Raw  Settled     %
   m/h    NTU    NTU    Rem.   mg/l    mg/l    Rem.   mg/l   mg/l     Rem.
75
100
150
64
72
80
2.06
2.35
4.65
96.6%
96.8%
94.2%
125
139
71
23
20
20
81 .6%
85.6%
71.8%
0.60
1.03
2.99
ND1
ND
ND
>99%
>99%
>99%
  ND = None Detected

ACTIFLO process achieved greater than 94% removal of turbidity and > 70% removal of
BOD5 and   >99% removal  of total phosphorous.  The process performance was not
compromised at the higher overflow rates in regards to total phosphorous removal. By
comparing the settled  water results,  higher overflow rates do not  adversely effect the
process performance.

An overnight run was performed to demonstrate the stability and the performance of the
process during CSO conditions.  The results of this overnight run are shown  in Figure
5.1 .b-7.  During the night of April 22 and the morning of April 23' the process was operated
at an overflow rate of  100 m/h, a polymer dosage  of 1.0  mg/l to 1.3 mg/l  and a  FeCb
dosage  of 45 mg/l to  100 mg/l.   The SS  removal  varied from 60% to 97.2% with an
average percent removal of 78.8% and a standard deviation of 7%.  The BOD5 percent
removal varied from 17.1% to 71.6% with an average  percent removal of 45.4% and  a
standard deviation of 18.2%.
    120
                                  10
                                        Hour
15
20
25
Figure 5.1 .b-7.  Twenty-four Hour Continuous Demonstration Run, Cincinnati, Ohio
                                      51

-------
Conclusions

As  a primary clarification process, the ACTIFLO process can  achieve high removal
efficiencies of pollutants including SS and Total Phosphorous. Additionally, the quick start-
up and  short detention time of ACTIFLO process make it a viable alternative for CSO
treatment.

This desktop analysis demonstrates that while the ACTIFLO  process  may be a viable
treatment process, the capital and O&M costs are high versus conventional alternatives.
One of the  primary benefits of the ACTIFLO system is the extremely small footprint of the
process. This can make the ACTIFLO system a more cost-effective alternative for sites
where space for large storage facilities is limited.

This process can essentially  be adapted and retrofitted into existing primary  treatment
facilities and for CSO control.
                                       52

-------
5.2 Retrofitting Existing WWF Storage Tanks to Provide Enhanced Settling/
Treatment and Post-Storm Solids Removal. Spring Creek AWPCP, New York, NY

Background

The Spring Creek Auxiliary Water Pollution Control Plant (AWPCP) is a wet-weather
combined sewage detention facility with an existing  maximum storage volume  of
approximately 49,200 m3 (13 MG): 37,900 m3 (10 MG) in basin storage and 11,400 m3
(3 MG) in influent barrel storage. The AWPCP facility  is located on Spring  Creek in
Queens,  New York, approximately 1 mile east of the 26 Ward Water Pollution Control
Plant (WPCP), near the Brooklyn-Queens border.  The function of  the AWPCP is  to
capture  combined sewage, above elevation -7.00, from tributary drainage  areas  in
Brooklyn and Queens.  Flow is conveyed to the plant by the four overflow barrels from
the Autumn Avenue regulator located in Brooklyn, and by two overflow barrels from the
157th Avenue  regulator  located  in Queens.  The plant has six basins that fill with
combined sewage to a depth of 4.6 m (15 ft) and up to a maximum elevation  of +1.00,
depending on Jamaica Bay tide elevations. Once the basins fill to elevations of -0.50 or
greater depending on tides, CSO is released into the bay through the flap gates at the
south end of the basins.  As flow recedes  in the sewers, combined sewage retained
within the basins and influent barrels above elevation -7.00 flow by gravity back into the
sewer system for treatment at the 26 Ward WPCP.  Retained  combined sewage below
elevation -7.00 is screened and pumped back into the  sewer system  through the 0.6 m
(24 in.) diameter plant effluent sewer. The layout of AWPCP is shown in Figure 5.2-1.

Overview

This  desktop analysis  evaluates the  concept  of retrofitting  the AWPCP to provide
enhanced treatment of stored WWF and removal of solids accumulated  in the basin
following storm events.  WWF  is presently retained in the AWPCP to minimize wet-
weather overflow through equalization, however, no provision for treatment is provided.

The residual solids and debris deposited within the storage basins are currently cleaned
by a system of traveling bridges equipped with a horizontal spray header and a water
cannon.   Each bridge,  one per  basin,  is operated in multiple passes over the basin
length and uses spray water (brackish bay water) to clean the basin walls and floor.

Two  alternatives,  cross-flow plate settlers and chemical addition, were  evaluated  to
determine the  dual effectiveness of retrofitting  AWPCP to serve as an  equalization
system and to enhance the removal of solids prior to discharge to Jamaica Bay.

The criteria for retrofitting the AWPCP  for enhanced treatment were selected to allow
conceptual design of each retrofit. The primary design factor of these alternatives is the
peak flow rate  that must be conveyed and treated.  The one-year frequency,  two-hour
duration design storm  was selected for this analysis.  The peak flow rate associated
with the one-year design storm is 66 m3/d  (2,330 cfs) based on SWMM modeling of the
combined sewer system tributary to Spring Creek and the 26 Ward WPCP.
                                      53

-------


i-«-^=- r^»*vi'  ^ si -'
                     V
    _^ jj_» ^^^.sp,^ . y~OT__»| rnTi'jTrTri in S
    JiJi^Ju,; ...... ^ J™L;i-i31i ! ,U        '
                                n- ! ' -,
                                "I :. '
                       • > ,    r i  i >
                        \.\  lilM<4

                         ''•\\ \ \


                           ViiU
            54

-------
The existing dimensions  of the AWPCP and the design  flow rate for the  proposed
treatment facilities are shown in Table 5.2-1.
Table 5.2-1. Existing Dimensions of the AWPCP and Design Flow Rate
          Design Data
Quantity
          Design Storm Flow (one-year storm)
          Number of Retention Basins
          Flow / Basin
          Length of Basin
          Width of Basin
          Depth of Basin
          Basin Volume
          Total Volume
          Basin Surface Area
          Total Surface Area
          Cross Sectional Area / Basin
          Surface Loading Rate
          Hydraulic Detention Time
63.2 m3/s (2,230 cfs)
6
11.0 m/s (388 cfs)
152.4m (500 ft)
15.2m (50 ft)
3.7m  (12.2 ft)
8,670  m3 (306,145 ft3)
52,000m3 (1,836,158 ft3)
2,322m2 (25,000 ft2)
13,935 m2(150,000 ft2)
56.7m2 (61 On2)
410 m3/d/m2 (10,067 gpd/ft2)
15 min
5.2.a. Cross-Flow Plate Settlers
Cross-flow plate  settlers use lamellar clarification  to remove solids from wastewater.
This process takes advantage of the decantation principle for SS that are denser than
water for clarification. Cross-flow plate settlers have performed as well as conventional
clarifiers, with a smaller footprint to achieve a similar level of settling.
Cross-flow lamellar clarification are principally used for the removal of oils and grease
present in residual  liquids  emanating from industrial  activities  in  the  petrochemical,
chemical,  mechanical, metallurgical, paper-processing and food-processing  sectors.
The cross-flow  plate  settler technology  has  also  been  successfully  utilized  in
conventional primary clarifiers and for gravity settling of surface water runoff.
Case History: Shell Montreal Canada
The Shell Company in Montreal, Canada retrofitted their existing  storage basin  with
cross-flow plate settlers to improve the water quality of the effluent prior to discharge.
This concrete  storage basin collects  surface runoff  from the  refinery area.   The
                                        55

-------
parameters that were used for design of the cross-flow plate settlers are shown in Table
5.2.a-1.

Table 5.2.a-1.  Design Parameters for the Cross-Flow Plate Separator

             Parameter                     Quantity
             Treatment Capacity (peak flow)     7,570 L/min (2,000 gpm)
             Influent TSS Concentration        200 mg/L
             Average TSS Density             1,159 mg/L
             Temperature                    20 deg. C
             pH                             8 to 9
             Overflow Rate                   404 m3/d/m2 (9,930 gpd/ft2)
The concrete tanks were modified to include cross-flow plate settlers  that provided a 27
m2 (290 ft2) surface area for  a  total  projected surface  of 223 rr2 (2,400 ft2).  The
installation included:

   •   Inlet deflectors
   •   Inlet screening
   •   Two cross-flow plate cells
   •   A V-notch weir
   •   An adjustable  skimmer
   •   An effluent weir

The layout of the retrofitted settling tanks is shown on Figure 5.2.a-1.

Limited evaluation of this retrofit has shown that  SS concentrations have been reduced
to between 100—125 mg/L throughout a range of influent flow rates and water qualities.

This retrofit  would improve the treatment capacity of the  existing  settling tank without
the requirement for costly expansion or the construction of new settling tanks to achieve
comparable  levels of SS removal.  Cost information for this retrofit is unavailable.

Proposed Retrofit of AWPCP with  Cross Flow Settlers

This retrofit consists  of converting the existing Spring Creek basins to include cross-flow
lamellar plate settlers to provide treatment through gravity settling of dense particles.
The use of inclined plates significantly increases the ratio of settling surface to footprint
that  would be  required for conventional primary clarifiers.   The  capacity of primary
clarifiers is primarily  a function of the surface area rather than depth.  The settling area
is a function of the surface area. Theoretically, the settling area of a  given spatial area
could be increased by stacking horizontal plates  to achieve a settling area equal to the
surface area times the number of plates.  However, this arrangement  does not allow for
removal of the accumulated sludge.
                                       56

-------
5.2.a-1           of       Co,
                     -,-.-.,,,-B* ;'  '  • *

                                                                       :

                                                57

-------
Inclining the plates would make sludge removal possible.  Each plate would have an
effective settling  area equal to  its horizontal projection.   The projections of inclined
plates  packed  close  together (e.g.  4 in.  apart) overlap.   The  total  settling area is
determined by adding the horizontal projections.  This can  increase  the theoretical
settling area by 10 times the settling area compared to a conventional clarifier  (Huebner
1979).

There  are two major force  vectors acting on SS contained within wastewater  in the
cross-flow separator.  The force of gravity acts to pull the SS down while the velocity
vector  pushes the SS up along the length  of the  plates.  The resultant of these  forces
directs the SS to the  plate where particles impinges on the plate and come  in contact
with other particles sliding downward along the plate.  The flow between the plates is
maintained in the laminar range to promote settling.

The treatment process begins with the screened influent entering the lamellar zone that
consists of cross-flow cells.  Particle settling occurs by gravity according to Stoke's Law
that describes the velocity of a particle falling through a fluid. Stoke's Law states that
the velocity of a particle is equal to the square of the particle radius.

In the design of sedimentation basins, the usual  procedure is to select a particle with a
terminal velocity Vc and to design the basin so particles that have a terminal velocity
equal to or greater than Vc will be removed.

The lamellar cell design is a practical  application of Hazen's theory on settling that
provides a large equivalent surface of separation (as noted by the projected  horizontal
area for each plate). Hazen's theory is stated by Huebner (1979) where the velocity is:

            V =  average linear velocity   =——    (2)
                                         d -w

            q = volumetric flow rate
            cf = distance between plates
            w =  width of plates
            0 =  angle of elevation to the horizontal
            Vs = particle settling velocity
            ds = maximum settling distance
            L =  length of plates

Particles are removed from the flow stream when its settling time is less than its flow
through time.
            t,   *  *q  (3)

            t,   =   —     (4)
                    Vs


                                        58

-------
                    L        L       L-d-w     .._,
             t   =   —  =  —,	  =  	     (5)
             1      rr       I 1                  V  '
                    V     q/d-w        q

The  maximum settling distance, cfs can  be calculated as  a function of the distance
between the plates, d and the angle of inclination of the plate, 0.

                       ^
                                 (6)
              ,
                    Cos®
Which equates to:

             d/Cos®      L-d-w
               Vs     "     q
and further simplifies to:
                   <  VsCos®    (8)
             L-w
Hazen's theory provides the surface loading rate (Y) for a conventional clarifier, as:

             7  =  -JL         (9)
                   L-w
which is equivalent to the simplified  inclined plate formula derived above in that q is the
total flow divided by the number of plates. The conventional surface loading rate occurs
when the angle 0 goes to zero, resulting in a Cos 0 of 1 .

The principal design parameters of cross-flow plate settlers are:

Wastewater flow rate - Settling systems are designed to accommodate the peak flow
rate, but the maximum removal efficiency is normally achieved at a lower design flow
rate.

Settling rate of solids - The settling rate of solids is measured in laboratory tests.  A
design velocity is then  selected  according to  the required removal  efficiency.  The
required  settling  area  is calculated by dividing  the design  flow rate by  the  design
velocity.

Angle/spacing of the plate -  The  angle  of  the  plates typically  varies  from 55 — 65
degrees  and the spacing varies  from 2.5 — 10.2 cm (1 — 4 in.) depending upon the
nature and  the concentration  of  solids and  space limitations.  The  10.2  cm  (4  in.)
spacing is commonly used in wastewater applications.

A depiction of the cross-flow plate  is shown on Figure 5.2.a-2.
                                       59

-------
Figure 5.2.a-2.  Horizontal Projections of Inclined Cross-Flow Lamellar Plates

Cross Flow Plate Settler Equipment

The cross-flow plate settlers  selected for this retrofit are from Aquarius  Services &
Technologies Inc.,  Quebec.   The major  equipment of the cross-flow lamellar plate
settlers is shown in Figures 5.2.a-3 and 5.2.a-4, and include the following:

1.  Flow Deflectors  - Flow deflectors at the inlet of each basin dissipate  the  energy
generated by the influent flow.  This encourages flocculation  and distributes the flow
evenly along the width of the basin and wet perimeter of the lamellar cell.

2.  Inlet Bar Screen - Floating material that may clog the lamellar cells is removed by a
inlet bar screen with 5.1 cm (2 in.) square openings

3.  and 4. Cross Flow Cells - Cross-flow type lamellar cells consist of inclined plates at a
60° angle and are constructed of either aluminum, stainless steel or synthetic materials.
Spacing  between plates may vary according  to the design  requirements.  For the
purpose of this  retrofit, stainless steel plates with 10.2 cm (4 in.) spacing were selected
based  on the manufacturer's suggestions for this installation.

5.  Skimming Device - Oils and hydrocarbons are skimmed from the surface through an
adjustable skimmer.

6.  Scum Baffle - A scum baffle is installed in back of the skimmer to retain the floatables
for collection and disposal through the skimmer.

 7. Weir-A V-notch weir is installed  at the discharge end of the basin to regulate the
liquid level in the basin.

8.  Sludge Vacuum Collector- Each sludge silo is equipped with  a sludge vacuum
collector to transfer and dispose of collected sludge.
                                       60

-------
5.2.a-3.  Plan View of          Cross-Flow
                      -Sit"
                       Si
A
wl
"1:
MS"'

I

:S
»/
0
                                  61

-------
Figure 5.2.a-4.  Profile of Aquarius Cross-Flow
^)r£^^ctb^^"^iSn^
*S™"™"™T™""""'****T! "T"* "™""'""™"
 ,.?: 	....-—„,,„	,  .

                                                                                   0,
                                                                                    jf
                                                                                  a) ?  i;
                                                                                  D ^  •'•
                                                                                     s
                                                                                   t
                                                62

-------
Following the completion of the storm, the basins would be automatically washed and
the sludge will be transferred for disposal.

The  hydraulic design criteria for the cross-flow  plate settler technology is shown on
Table 5.2.a-2.

Table 5.2.a-2.  Cross-Flow Plate Settler Design Data


       	Design Data	Quantity	
           Hydraulic Loading Rate*      0.15 m3/s/100 m2 (0.5 cfs/100 sq. ft)
           Overflow Rate                 132 m3/day/m2 (3,250 gpd/sq. ft.)
           * Loading Rate furnished by Aquarius Services & Technologies Inc. Quebec


Under the above loading rates and  available basin surface area, the one-year design
flow rate of 63 m3/s (2,230 cfs) would not be accommodated by a cross flow plate settler
system.  The proposed system is capable of treating a maximum flow rate of 1,835,725
m3/d (485 MGD) or 32% of the design flow rate based  on the physical size of the Spring
Creek basins and the limited hydraulic loading rates of the cross-flow plate settlers. The
proposed facilities  would provide  treatment under a small percentage of wet-weather
events observed at Spring  Creek and would create a hydraulic  limitation  to  WWF
entering  the facility.  Therefore, retrofit with cross-flow plate settlers is not considered
viable for this site.

5.2.b.  Chemical Addition to A WPCP

This  desktop analysis involves retrofitting the AWPCP by chemical  addition to provide
enhanced settling/treatment. Chemical addition describes a method  of treatment where
chemicals are added to enhance settling of particles present in the wastewater.  Treated
water is then decanted and conveyed for further treatment or discharged. The resultant
sludge is collected for disposal and can be de-watered to reduce the volume of solids.
Chemical precipitation can be used to remove metals, fats, oils and greases (FOG), SS
and some organics. It can  also to be used to remove phosphorus, fluoride, ferrocyanide
and other inorganics.

Conventional wastewater clarification processes primarily involve the destabilization and
subsequent removal of colloidal SS materials that are not readily removed by gravitation
alone.   These suspended materials can  be natural or synthetic organic or  inorganic
compounds, microorganisms, and/or viruses that typically range in size from 10~4//m to
1,000//m.  In most natural systems, the stability of  colloidal suspended  materials is
attributed to a net negative surface charge that causes individual particles to repel each
other and remain in suspension. To counteract these  repulsive forces a highly charged
ionic  chemical,  such as  alum  (Ab (804)3),  FeCb,  ferric  sulfate  (Fe2(S04)s), poly-
aluminum chloride (PACI), is added to bring about a net reduction in the repulsive force
between the suspended materials.  This  process, termed coagulation, results in  the
                                       63

-------
destabilization and/or attraction of the SS to form chemical floe.  Although destabilized,
the chemical floe may remain in solution due to its extremely small size and  low mass.
These floe, often called "pin floe," are difficult to remove from the treated water without
first aggregating the smaller particles together into larger, heavier, more settleable floe
by flocculation.

Flocculation is the term used to describe the aggregation of pin floe into larger, heavier,
more settleable floe.  This is most readily accomplished with the addition of a chemical
(flocculant aid) polymer.  Such polymers are available from many sources in  both liquid
and dry form  with varying molecular weights and net surface charges.  In most cases
selection of the best polymer for a specific application and set of conditions is largely a
matter of  trial and  error.   Larger, more  settleable  floes  form  by inter-particle
polyelectrolyte bridges between floe and destabilized solids.  Although the formation of
inter-particle bridges can take many forms, the end result is the enmeshment of large
volumes  of particles to form  large  dense floe that are more  readily removed  by
gravitational settling.

The  use of chemical  addition  would enhance the settling  of discrete  particles and
suspended matter thereby increasing the overall removal of BODs and SS.  This would
be accomplished  by retrofitting the existing  basins with chemical  feed, mixing and
contact  facilities.   The retrofit  would also include dedicated sludge  collection and
removal  facilities.   Chemical coagulation of raw  wastewater before  sedimentation
promotes flocculation of finely divided solids into more  readily settleable floes, thereby
increasing  SS,  BODs,  and phosphorous  removal  efficiencies.   Sedimentation with
coagulation may remove 60—90% of the  SS,  40—70% of the BOD5, 70—90% of the
phosphorous, and  80—90% of the bacterial  loadings.  In comparison,  sedimentation
without coagulation may remove only 40—70% of the SS, 25—40% of the BOD5, 5—
10% of the phosphorous loadings and 50—60% of the bacteria loading.

Chemical Coagulants

Advantages of coagulation include greater removal efficiencies, the ability to treat higher
flow rates, and more consistent performance.  Disadvantages of coagulation include  an
increased mass of primary sludge that is often difficult to thicken and  de-water, and
increased in operational cost and operator attention (WEF 1992).  Iron salts, particularly
FeCb, are the most common coagulant used for primary treatment. WWTF  using lime
for primary treatment have found that it produces more  sludge than do metal salts and
is  more  difficult to store, handle and introduce into the waste  stream.  Coagulants
should be  selected for performance, reliability and cost and evaluations made  to
determine dosages and effectiveness.
                                      64

-------
Case Histories

Table 5.2.b-1 lists coagulant dosages and performance data for six (6) treatment plants
that employ chemical coagulation for advanced primary performance in sedimentation
tanks.  The Joint Water Pollution Control Plant (JWPCP) at Carson, California evaluated
30 different polymers  and found anionic polymers to be  the most effective type for
enhancing primary sedimentation.  The Hyperion plant in Los Angeles, California uses
to great  success FeCb and  an anionic polymer at the  headworks before primary
sedimentation.   This  process  increased  primary sludge  production 45%  through
improved SS and  colloidal matter  removal.  This increase was partially offset  by a
decrease in waste activated sludge production.

The Los Angeles and San Diego sedimentation tanks with chemical coagulation operate
with  overflow rates between 69.3—81.5  m3/d/m2 (I700—2000 gpd/ft2) as compared to
conventional overflow rates of 32.6 m3/d/m2 (800 gpd/ft2) for primary settling. Research
in Sarnia and Windsor, Ontario indicated that overflow rates  up to 97.8 m3/d/m2 (2,400
gpd/ft2) did not significantly affect effluent  quality.


Design Considerations
The design flow rate for this retrofit is 66 m3/s (2,330 cfs) as shown in Table 5.2-1 with a
corresponding overflow rate of 410.1 m3/d/m2 (10,067 gpd/ft2) based on the dimensions
of AWPCP.  This high surface SOR  is greater than what would typically be  applied for a
chemical coagulation system;  however, for the purposes of this  retrofit it is assumed
that during the higher SOR's, a reduced efficiency will result.  It is  important to note that
the frequency at which these high SOR's  are likely to occur  are expected to be relatively
infrequent (e.g.,  once per year).

The proposed retrofit consists  of equipping the existing influent barrels with a chemical
feed and mixing system.  These facilities would  provide the necessary mixing and
contact for the rapid mix and flocculation  steps, which include the addition  of FeCb and
an anionic  polymer.   An important element in this  analysis is  an evaluation of  the
hydraulic and physical parameters dictated by the existing basins  as depicted on Table
5.2-1.

This retrofit would include the specific elements outlined below.  Based on a ratio of 452
kW/ m3/d (0.16  HP/MGD) at a peak flow  of 953,820 m3/d (252 MGD) per barrel, a total
of 30 kW (40 HP) is necessary to achieve the mixing required for the FeCb per barrel.
There will also be 2 mixers per basin for mixing the polymer. Therefore, 24—15 kW (20
HP) mixers (4 /barrel)  are necessary for  the 6 basins. Chemical storage requirements
are based on the need to treat approximately 2,500,000 m3 (660 MG) of CSO per year.
Storage of FeCb (37% solution) and polymer (0.25  mg/l) to treat a month (on average)
of CSO would require approximately 16 m3 (4,200 gal) of chemical storage.
                                      65

-------
Table 5.2.b-1.  Coagulant Dosages and Performance Data for Six Treatment Plants That Employ Chemical Coagulation
           for Advanced Primary Performance in Sedimentation Tanks
                                      Advanced Primary Performance

Location
Point Loma
City of San Diego
Orange County
Plant No. 1
Orange County
Plant No. 2
JWPCP
Los Angeles County
Hyperion
Los Angeles City
Sarnia
Ontario, Canada

Flow
(m3/day)

722,935

227,100

696,440

1,438,300

1,400,450

37,850

(MGD)

191

60

184

380

370

10

Influent
mg/L

276

263

248

365

300

98
BOD
Effluent
mg/L

119

162

134

210

145

49

Rem.
%

57

38

46

42

52

50

Influent
mg/L

305

229

232

475

270

124
TSS
Effluent
mg/L

60

81

71

105

45

25
Chemical Addition
Rem.
%

80.3

64.6

69.4

77.9

83.3

79.8
Type
FeCI3
Anionic Poly
FeCI3
Anionic Poly
FeCI3
Anionic Poly

Anionic Poly
FeCI3
Anionic Poly
FeCI3
Anionic Poly
Cone.
mg/L
35
0.26
20
0.25
30
0.14

0.15
20
0.25
17
0.30
Duration

Continuous
8 hours
Peak Flow
12 hours
Peak Flow

Continuous

Continuous

Continuous
                                                  66

-------
Chemical feed pumps will be necessary to deliver the correct dosing of chemical during
wet-weather events.  The FeCb solution will require a feed rate of approximately 0.05
m3/min (13 gpm) per barrel at peak flow; resulting in 2 metering pumps per barrel for a
total of 24. The pumps would be hose (or peristaltic type) pumps.  Half as many feed
pumps would be required to deliver the polymer to the influent barrels.

Operators could control  the chemical feed system using a new  control panel and
SCADA system housed in the existing control  room.  The  capital costs for this retrofit
are shown on Table 5.2.b-2.

Table 5.2.b-2.  Estimated Capital Costs Chemical Addition Retrofit
             Description	Cost	
             General Conditions                                    $25,000
             Sitework                                          $350,000
             Chemical Storage                                    $25,000
             Chemical Feed System                                 $75,000
             Chemical Mixers (4 in each of 6 basins)                    $720,000
             Instrumentation and SCADA System	$25,000
                                    Total Construction Cost       $1,220,000
           	Project Services1	$240,000
                                       Total Project Cost       $1,460,000

            1 Includes Design & Construction Engineering, Legal, and Fiscal Expenses
             Does not include contingencies and right-of-way costs
Operational Requirements

The existing AWPCP operates intermittently based on wet-weather events.  Following
CSO events, the facility is de-watered by gravity and pumping. These operations would
not change under the proposed retrofit.  The following list summarizes the areas of
expected O&M cost increases at the facility:

•  Increased labor to perform wet-weather process adjustments and clean-up following
   a wet-weather event.

•  Power for chemical feed and mixing systems , and

•  Chemicals

The costs are based on  the facility treating approximately 2,498,100 m3 (660 MG) of
CSO per year and maintaining a chemical storage of approximately one month.  The
annual volume of CSO is based on modeling performed for  a typical year using the
Storm Water Management Model (SWMM).  Hourly rainfall data were used from a local
NWS Weather Station as  input to the model.  It is estimated that approximately 100 wet-
weather events occur per year and that each event is approximately 6 h in duration.
The O&M costs are shown in Table 5.2.b-3.
                                       67

-------
Table 5.2.b-3.  Estimated Chemical Addition O&M Costs

              Description	Estimated Annual Cost
              Coagulant Consumption	
               Coagulant Type                                   Ferric Chloride
               Coagulant Dosage (ppm)                                      20
               Design Flow Rate (MGD)                                   1,510
               Coagulant Consumption (Ibs./day) at Peak Flow               252,000
               Coagulant Consumption (Ibs/month)                          9,200
               Coagulant Consumption (tons/year)                              44
               Coagulant Cost ($/ton)                                      200
               Coagulant Cost ($/vear)	$8.800
              Polymer Consumption	
               Est. Polymer Dosage (ppm)                                  0.25
               Design Flow Rate (MGD) at peak flow                         1,510
               Polymer Consumption (Ibs./day) at Peak Flow                   3,100
               Polymer Consumption (Ibs./month)                            110
               Polymer Consumption (tons/year)                              0.53
               Polymer Cost ($/ton)                                       3,500
               Polymer Cost @ 100% 724/365 ($/vear)	$1.848
              Labor	
               Cleanup/Operations	$21,000
              Energy Consumption	
               Rapid Mixer HP per Basin                                      30
               Chemical Feed HP per Basin                                    2
               Total HP (for 6 basin compartments)                           192
               kWh/year                                             115,200
               Total Annual Energy Cost ($/vear)	$6,912
              Total Annual Operation Cost ($/year)	$38.600

              6 hours per event
Conclusions

This  retrofit  utilizes the existing storage  facilities  to provide  chemically enhanced
sedimentation of WWF.  This could result in the removal of 60—90% of the SS, 40—
70% of the BOD5, and 30—60% of the COD from the effluent to the receiving  water.
This is an expected 58% increase over current performance without chemical addition.
At the one-year design flow, the facility will exceed  typical SORs by a factor of 4—5,
however, these would occur relatively infrequently.
                                         68

-------
5.2.c.  Retrofitting Existing WWF Storage Tanks For Post-Storm Solids Removal,
      Spring Creek AWPCP, New York, NY

In 1997, COM conducted a basin-cleaning prototype  testing  program at the Spring
Creek AWPCP in New York City (COM 1997).  The prototype testing was performed in
response to the New  York City Department of Environmental  Protection's (NYCDEP)
desire to investigate alternatives to the existing traveling bridge-cleaning system.  This
analysis was performed to evaluate the results of the prototype testing and provide
recommendations for basin cleaning system design.

Background

The residual solids and debris deposited within the storage basins are actively cleaned.
Currently, for cleaning of solids and debris within the basins, the facility uses a system
of traveling bridges equipped with a horizontal spray header and a water cannon. Each
bridge, one per basin,  is operated in multiple  passes over the basin length and uses
spray water (brackish bay water) to clean the basin walls and floor.

The basin cleaning alternatives evaluated through prototype installation and testing at
the facility were:
   •  Tipping buckets
   •  Fixed spray nozzles
   •  Orifice headers
   •  Flushing gate (Hydroself Fluid Flap)
   •  Re-suspension/mixing (used to enhance any of the above)

Objectives and Scope

The effectiveness of the cleaning systems at the Spring Creek AWPCP were evaluated
to establish design criteria  for feasible alternatives for full  implementation and to
develop operational requirements and operational and capital costs  associated  with
each of the systems. The data collected from the prototype study was used to either
determine or indicate the following:
   •  Effectiveness of floor cleaning
   •  Effectiveness of wall cleaning
   •  Water usage
   •  Power usage
   •  Overall cleaning time required
   •  Special operating requirements
   •  Safety concerns (if any) associated with each system
   •  Cost
Each of the systems  listed above was installed  in test areas within  three of the six
basins  at Spring Creek. As discussed below, each system was arranged in  different
                                      69

-------
configurations  and operated during actual  basin  cleaning events for a total of 16
cleaning test events, over the period from July 1 to December 16, 1997.  The cleaning
systems were tested under conditions where solids accumulation varied widely.  Each
system was operated under a variety of cleaning water  volumes,  flow rates,  and
pressures.  As the testing proceeded, it became apparent which technologies were not
providing effective cleaning. As a result, the original test protocol was modified to focus
on, and optimize the performance of the most effective systems.

The actual test conditions and the equipment used in the study are summarized below.

Basin Cleaning Prototype Equipment and Operation. The prototype testing consisted of
16 test runs, from July 1, 1997 through December 17, 1997.  The prototype fieldwork
was  performed in conjunction  with  NYCDEP's routine  basin  cleaning operations  to
benchmark the existing cleaning operations at the facility.

The following parameters were monitored and recorded during the  prototype testing:

   •  Solids depth within basins
   •  Equipment test parameters (flow rate, total wash-water volume, pressure, etc.)
   •  System cleaning effectiveness
   •  Cleaning duration

The overall cleaning effectiveness was based on a qualitative assessment of the degree
of cleaning and the approximate  percentage of the floor area cleaned.  The test area
was judged to be clean  when approximately 99% of the test area was without solids.
Water consumption and operating pressures were measured  using flow meters and
pressure  gauges set up  throughout the system.  Cleaning time was measured for the
prototype  systems and  then estimated regarding full-scale  implementation of each
system.

An evaluation of the existing basin cleaning operation was  performed and  provided a
benchmark for the prototype  cleaning systems.  This benchmarking was  performed  by
observing the cleaning operations, documenting the cleaning time and using the rated
pump capacity  to calculate  the flow rate  and water used by the existing  traveling
bridges.

Using the traveling bridges,  the  basins were generally cleaned one side  at a time.
Typically, the traveling bridge requires a minimum of two passes per side.   The first
pass loosens the solids and provides some cleaning, while the latter passes provide for
a more thorough cleaning.  A manually controlled water cannon breaks up large heavy
solids.  In addition to cleaning the basin floor, the water cannon cleans the sidewalls of
the basin and the walkways of the basin area.
                                      70

-------
General  Prototype  Systems Installation  and  Operation.   Five prototype cleaning
systems were tested: tipping buckets, flushing gate, fixed spray headers with nozzles,
fixed spray headers with orifices, and a submersible mixer.  The prototype systems
were located  in basins 1, 2 and 3.

Bay water was  used  as  the  source  of cleaning water for the prototype systems.
Cleaning water was supplied  from  a single, 37.3 kW (50  HP),  three stage vertical
turbine wash-water pump,  installed in the existing spray water channel between basins
1 and 2. The pump was sized for a flow rate of 3 m3/min (800 gpm) against a head of
50.3 m (165  ft).  A  PVC piping system distributed wash-water to the various cleaning
systems.  A  flow schematic of the  wash-water equipment and  piping is provided in
Figure 5.2.C-1.  Flow  meters,  pressure gages, and butterfly  valves  were installed
throughout the cleaning-water piping-system to provide a positive means of regulating
the system flow and pressure.  Pressure control valves were installed within the piping
system to regulate the  delivery pressure of the cleaning water to the required level. A
pressure sustaining  valve was installed on the main piping header to relieve excess flow
and to maintain system pressure  during times the full 3 rrfVmin (800 gpm) flowed but
was not required.

Prototype testing occurred  in  conjunction  with  NYCDEP's normal  basin cleaning
operations.  The basins were  pumped and drained by plant staff prior to the  normal
cleaning operations.  Plant staff cleaned the length of the basins, except for the 6.1 m
(20 ft) test sections located in  basins  1, 2,  and 3 using the existing traveling bridge
system.  On  occasion,  some test  sections were not available for testing either due to
cleaning  of the test section by plant staff or by inflow of the tide through leaking tide
gates.  Generally after  the plant staff completed cleaning, the prototype system testing
began.   Sediment depth  in the prototype system test  sections  was  measured and
recorded by entering the basins.  The distribution of solids within each test section was
documented.  The tipping buckets, spray nozzles and orifice headers at  the influent end
of the basins were  generally tested first.  After the testing at the influent end of the
basins was  finished,  the  tipping buckets, flushing  gate, spray  nozzles  and orifice
headers at the effluent end of the basins were then tested. During several test runs, the
prototype testing was halted early due to overloading of solids in the AWPCP's cross
collector channel and the chamber screen channels. As discussed later, sudden slugs
of cleaning water, carrying large quantities of solids, overloaded the channels causing
the testing to  be ended  prematurely.
                                       71

-------
5.2.c-1.
                                        '"••—-\«,Arf-^-*--_-\.,A,A	A,,A,,
                              5 »
                                      IJ
                                                         ,

                                                       »! ft. --
                                                            "'
                                                   t    -, f  •             .if

                                                       •a Si. *4  •           (*   "


                                                    ">"1 •.<•!"' j,  'i x     i  "i  7
                                                     72

-------
As the results of the  prototype testing were analyzed, the operating conditions were
adjusted and the actual prototype system was  modified or the entire prototype system
was removed from testing when deemed ineffective.

Tipping Buckets.  Grande, Novae & Associates, Inc., of Montreal, Canada supplied the
units tipping buckets.  Four tipping  buckets were installed in basin 1. Two buckets, one
each on opposite sidewalls, were installed at elevation +1.5 m  (+ 5.00 ft) at both the
influent (TB-1A and TB-2A) and overflow (TB-1B and TB-2B) test sections of the basin,
as shown in Figures 5.2.C-2 and 5.2.C-3.  The buckets were installed at either end of the
basin to test the cleaning effectiveness of the highest  and lowest solids loading in the
basin.  The overflow end buckets were used to determine the ability of the buckets to
convey the solids down the full length of the center velocity channel.

Each bucket was 6.1 m (20 ft) long and had a liquid capacity of 0.15 m3 per linear meter
(11 gal per linear foot [gal/ft]) for a total volume of 0.8  m3 (220 gal) per bucket.  During
testing,  a constant supply of cleaning  water  was fed into the buckets.  When  the
maximum water  level  was reached in  the buckets, the buckets would automatically tip
due to the change in  the bucket's  center of gravity, a 0.8  m3 (220 gal) wave  of water
down the basin wall and along the floor, to the velocity channel.

A training wall, stretching the width of the basin, from  the velocity channel to the basin
sidewall, was originally installed  along the edge of the test area.  The  purpose of the
training wall was to prevent the wave of cleaning water from dispersing laterally, thereby
maintaining the full  force of the wave within the test section. In addition, curved wall
fillets were installed at the bottom of the sidewall  to  allow for  a smooth transition of
water flow from  the wall to the floor.  Figure 5.2.C-3 shows the approximate size and
location of the training walls and wall fillets.

Once the NYCDEP's cleaning operations were completed  and the  sediment in  the
basins measured, filling of the tipping buckets commenced.  The two  buckets at the
influent end of the basins were filled simultaneously and then discharged automatically.
The  process was repeated as necessary until  the test section was considered clean.
Table 5.2.C-1 summarizes the test conditions for the tipping buckets along with the other
prototype systems tested.

After the  installation  of the  equipment it was found that bucket TB-2B did not  tip
automatically.  A counter weight was  added to correct bucket mounting/misalignment
and to permit the bucket to tip.  Periodically,  bucket TB-1B also  experienced hesitation
in tipping.

Following the initial test runs,  modifications were made  to the training walls and fillets to
improve their effectiveness at eliminating dead zones within the basins.  The original
training walls,  made of steel angle approximately 15—30 cm (0.5—1 ft)  in height, were
bolted to the basin floor.  The training  walls extended from the walls at the column  line
to the velocity channel. During testing  it was observed that the solids along the training
                                       73

-------
5.2,c-2.  Tipping            -
                              *"!?"""'" *'"*!*' 	IsSKSi !^nS|KJit VMl'^SW^W™' ^^rf% ,
           /  I'"' >  	»-- !




          <*
                     11
                            X,


                                                                       >
                                             74

-------
5.2.c-3. Tipping         -        A
           w	

                                   75

-------
5.2.01.
     ft.
                                76

-------
wall were the most difficult to clean.  It was believed that this was due to a dead zone
between the edge of the bucket and the basin wall pilaster. A modified plywood training
wall was installed along the width of the pilaster to eliminate this dead zone as shown in
Figure 5.2.c-4.

In separate subsequent tests, the training  wall from bucket TB-2A and the wall  fillets
from buckets TB-1B and TB-2B were removed.  The training wall was removed  to
determine what effect its removal would have on the solids located in the dead zones.
The wall fillets were removed for use in the modified orifice header configuration.

Flushing Gate.   The flushing gate system was tested for possible  installation  in the
facility's influent barrels,  and/or cross collector channels.  Currently, accumulated solids
in the influent barrels must be removed  manually.  The wave produced  by a flushing
gate can produce a high velocity flow of water over a relatively long distance.  The basin
velocity channel offered an  opportunity to test the effectiveness of the  flushing gate
without  actually  entering the influent barrels.  Grande, Novae & Associates, Inc.  of
Montreal, Canada supplied the flushing gate.  A single  flushing gate,  1.22 m (4 ft) wide
with two 2.74 m  (9 ft) long steel subwalls was installed at the overflow end of basin 1.
The flushing gate was centered at the head of the existing velocity channel, while the
two subwalls were located on either side of the  flushing  gate.   The purpose  of the
subwalls was to accommodate the installation of the  gate and to provide a storage
volume of approximately 15.1 m3 (4,000 gal) of flushing  water.  This system utilized both
CSO  and bay water  as its  source  of flushing water.   Cross sectional  views  of the
flushing gate are shown in Figure 5.2.C-5.

When the desired volume of water  behind the flushing gate was  obtained, a remote
hydraulic actuator opened the flushing gate and sent a wave of cleaning water down the
length of the velocity channel.  The time  of travel for the leading edge of the wave
produced by the  gate was measured at two points in the center velocity channel, 30.5 m
(100 ft)  and 76.2 m (250 ft)  downstream of the gate.  The velocity versus time of the
flow  following the wave was determined by  measuring the time of travel of objects
floating in the channel.

There was no active means of closure for the flushing gate.  After the stored wash-water
volume  had discharged, the  flap portion of the gate would close under its own weight.
The flap was required to sit flush against the gate frame in order for the hydraulic  unit to
lock the flap in the closed position.  As a  result, any debris or leaking tide water flow
would interfere with the gate closure.  When the gate closed, debris caught between the
flap and frame would often cause leakage  of the stored volume.  Several times during
the testing, a person was required to enter the basin in order  to clear debris from the
gate and allow it to be closed.  On one occasion, an automobile tire became wedged in
the gate opening.
                                       77

-------
5.2.c-4.                       Training
                                      i.	


                                    r ffft, ~v^ f ^* at   p
                                    .kllO-^  ^^
                  Fioure S.2.C-*  Mo-difi'SO Pfywocxs T-'aifiirifl- Waji Sc
                                         78

-------
5.2.C-5.                 -          B

                                                                      a
                                                                      6s
                                     79

-------
Fixed  Header with  Spray  Nozzles.  Two  separate  spray  header  arrangements
(Alternative-1 and  Alternative-2) were installed in basin 2.  Spraying Systems Inc. of
Wheaton,  Illinois provided spray nozzles installed on these headers. Alternative-1 used
the  model  HH30100  nozzle  that  delivered  a  full  cone  spray pattern  of  0.05
m3/min/nozzle (12.3 gpm/nozzle) at approximately 42,190 kg/m2 (60 psi).  Alternative-2
used the  model  PSS35200 nozzle that  delivered  a  flat spray pattern  of  0.1
m3/min/nozzle (24  gpm/nozzle)  at  approximately  42,190  kg/m2  (60  psi).    Both
arrangements  also   utilized  a  flat-spray-pattern-nozzle-model   50200   [delivering
approximately 0.1 m3/min/nozzle (24 gpm/nozzle) at 42,190 kg/m2 (60 psi) to clean the
sidewalls.  Each spray manifold was installed with 3 separate 10.2 cm  (4  in.) header
pipes.

Two pipes were dedicated to clean the  floor, and one pipe was dedicated  for wall
cleaning  as shown in Figure 5.2.C-6.  The spray water manifolds were  located on
opposite walls at both the influent and overflow ends of the basins at elevation +1.5 m
(+5.00 ft).  At this elevation the nozzles were approximately 4.3 m (14 ft) from the basin
floor. This maintained the nozzles above  the basin flood line  and reduced the  potential
for nozzle damage  and clogging.

The  flow rate and  pressure of the cleaning water supplied to the nozzles  were held
constant throughout each test run.  Provisions to adjust the  flow and pressure,  using
throttling valves located in each header, were made so that these parameters could be
varied from run to run.  The cleaning system was allowed to run for up to 15 min.  The
flow  rate  and operating pressure were recorded for both nozzle  arrangements.  Table
5.2.c-2 provides the test conditions for the  fixed spray headers with nozzles.

Even at the highest attainable pressures and flows, the nozzles proved ineffective at
cleaning the basin floor and were removed from service. The  nozzles were also subject
to clogging  from debris found  in the cleaning water  supplied from the spray water
channel.

Fixed Spray Header with  Orifices.   Initially,  four  headers were installed  in  two
arrangements in the influent  and effluent ends of basin 3. Two orifice  headers were
located on opposite walls at the influent end of the basin, and two orifice headers were
located on opposite walls at the  overflow end of the basin.  Each orifice header  was 6.1
m (20 ft) long and equipped with 1.6 cm (5/8 in.) orifices spaced on approximately 30.5
cm (1 ft)  centers.  The spacing and size of the holes were selected to provide  a water
flow  rate  of 0.5 m3/min/m (40 gpm/ft) at a discharge pressure of 7,030 kg/m2  (10 psi).
One pair of headers was installed at a centerline elevation of +1.5 m (+5.00 ft)  (influent
header 1A and overflow header 2B) with the orifices aimed directly at the sidewalk
These headers were designed to send cleaning water down the wall in  a  continuous
sheet, thereby washing the basin floor and carrying solids to the velocity channel.  The
second pair of headers (influent header 2A and overflow header 2B) was installed  at a
centerline of elevation -8.50 with the header aimed directly at bottom of the wall.
                                       80

-------
5.2.C-6.                         -
                                     81

-------
Jable 5.2.C-2.  Summary of Prototype Testing Operation
                 '".nf.i« Ke«J«rs

                                                               «JTJ£ '^JlJ 'lljitli ifta.lM.-ffed iuftE S£! '"kT- OiJS
                                                               .9 ALT inyiJfiTj tltar v?r?t nj-fi .tffi-ifitxr? ?
                                              ••9~
                                              • :s;
                                                    82

-------
Table 5.2.C-2  page 2
                                                                                                  cmt  i.t u^
                                                                        !;'-£t£.S/MS thfi "'I'' -ITS*


                                                          k s''*5 ! B-." *"• "'" .=  '-t-'-i&s i-(**t:-i'** r-^cr
                                                     ft.£. «l
                                                    .pr •
                                                             83

-------
This system was also designed to provide a sheet flow of cleaning water along the floor.
Since  no nozzles were utilized  under this  alternative,  there  was  less  concern for
clogging while the headers were submerged.   Cross sectional views of this original
orifice header configuration are shown in Figure 5.2.C-7.

Training walls stretching the width of the basin from the velocity channel  to the basin
sidewalls were installed  at the outside end of the  basin test section.   Similar to the
tipping  buckets,  the  purpose  of the training walls  was  to prevent dispersion of the
cleaning water and maintain the flow within the basin test section.

After the initial six test runs, the original orifice header configuration was determined to
be ineffective at cleaning.  Subsequently, modifications to the header configuration were
made including the following:

•  Removal of the overflow end orifice headers (for reuse in the modified configuration
   at the basin influent)
•  Rotation of the influent orifice header 1A (elevation +1.5 m [+ 5.00 ft]) to orient the
   water jets nearly tangential  to the sidewalk
•  Rotation of the influent orifice header 2A to orient the spray downward and towards
   the velocity channel at a spray angle of approximately 30°s with the floor.
•  Installation of two new influent orifice headers (2B and 3) on  the eastern side of the
   test section.  Header 2B was installed on the  basin floor at elevation -2.6 m [-8.5 ft]
   and header 3 was installed on the sidewall at elevation +1.5  m (+5.00 ft). The spray
   on header 2B was oriented similar to header 2A while header 3 was oriented similar
   to header 1A.
•  Installation of wall fillets at the wall-floor joints beneath influent headers 1A and 3.

Figure 5.2.C-8 provides a section showing the modified header configuration.

Flow rate,  operating  time and pressure was  recorded for each orifice header. Table
5.2.c-2 provides the operating conditions and performance for this system.

Few operational  and mechanical problems  were  encountered while  operating this
system.  Occasionally rags and other debris became entangled  around the supports of
the orifice headers; however, this debris did not effect the cleaning operation.

Submersible Mixer. A single submersible  18.6  kW  (25 HP) jet  mixer was supplied by
Davis EW  of Thomasville,  Georgia (now a division of US  Filter).  The jet mixer was
installed at the influent end of Basin No.  2  using steel  guide  rails and lifting cables
similar to those typically associated with submersible pumping equipment.  The mixer
was  provided with an integral diffuser and air supply  piping system to achieve a roll
mixing effect within the basin.
                                       84

-------
Figure 5.2.C-7.  Original Orifice Header Configuration - Section A
                                        85

-------
5.2.c-8.                                        -        A

                                    86

-------
The jet-mixer was placed  into continuous service after test run no. 3. The mixer was
controlled using automatic level  control floats to turn  on while the basins were filling
during a CSO event and  off when the basins were almost completely drained.  The
mixer remained in operation sometimes  for more than one day until the basin was
drained.

The purpose of the mixer  was to provide re-suspension of the solids, such that solids
would be removed with the  wastewater during  the de-watering operation, minimizing
solids deposition within the basins.  During testing the mixer proved  ineffective at re-
suspending the solids in the test area and reducing solids deposition.  The mixer was
removed from service after  test run no.  7.  Following removal of the jet mixer from
service, it was found  that  the diffuser and discharge piping  of the mixer were almost
completely clogged with leaves, rags and other debris.

Prototype Test Results

Solids Composition and Distribution.  As would be expected for a  CSO facility,  the
composition and distribution of residual solids in the detention basins varied significantly
between events. The measured  depth of solids ranged from  a trace coating (less than
1.5 cm) up to 45.7 cm  (1.50 ft) at  individual piles. The solids ranged in consistency from
a thin watery mixture, to a viscous "clay like" consistency, to heavy  grit.  Test events
with the greatest depth of  solids  and heavier grit generally followed rainstorms of high
intensity.  Test events with fewer solids and solids with a thinner consistency generally
followed rainstorms with low intensity or low total precipitation.

In general, the influent ends of the basins contained the heavier solids, with the overflow
ends  characterized primarily  by  floatables,  leaves, and a thin coating  of solids. The
solids depth was greater at the influent end than at the overflow end. Dead zones at the
influent headwall of each basin contributed to the formation of piles or long mounds of
solids in the influent test sections. The largest piles were observed on the basin side of
the influent  headwall  columns and along the sidewalk  Solids appeared to build up
behind these obstructions, as scouring  of  solids during CSO inflow was prevented.
These and other piles of  solids often posed the most resistance to cleaning.  Within
basins no. 1 and 3, a ridge  of deeper solids was observed several times along the
western wall of the basin  influent.  At the influent end of basin No. 1, the solids depth
was generally greater  on the western side of the test section than the eastern side. This
likely contributed to the formation of piles along the western sidewalk  The difference in
solids depth between  the two sides of the basin  No. 1 influent may have been due to a
partial blockage of the influent barrel  feeding the eastern side of this basin, resulting in
unequal flow distribution.

Based on subjective observations, it appeared that the consistency of the residual solids
was generally thicker  and  more cohesive when more days had transpired between the
inflow event and the cleaning test  as compared to  when  cleaning was performed
immediately following  an inflow. The nature of the solids consistency affected cleaning
system performance, with the thicker/ cohesive solids being more resistant to cleaning.
                                       87

-------
Benchmark of Existing  Cleaning  Operations.  As discussed previously,  the existing
traveling bridge cleaning system operations were  observed  to  document general
operations,  cleaning  time,  and  water  consumption.   The  cleaning  time,  water
consumption and estimated electrical cost for this system are presented in Table 5.2.c-
3.

The  bridges provide very effective cleaning of the basin walls  and floor.  Generally,
several  complete passes  along the basin length  were required to clean each basin.
With the bridge mounted water  cannon, solids  or  objects which  might have  been
unaffected  by the  spray nozzles were readily moved into the velocity channel and
cleaned from the basin floor.  The bridges readily achieved complete basin cleaning.

The  primary advantage of the traveling  bridge system is  that it is very effective at
cleaning the basins.  The degree of cleaning is only  limited by the amount  of time  an
operator expends in cleaning. Disadvantages observed with the bridge system included
the following:

•  Cleaning with the bridges  was  labor intensive, requiring staffing  of  each bridge
   during operation
•  Because there  is  no separation  from the basin  airspace,  the  operator had the
   potential to be exposed to H2S levels exceeding standards
•  The bridges appear to require frequent maintenance. Often during the test runs, one
   of the six  bridges was observed to be temporarily out of service, due to problems
   with the bridge controls or with the electrical bus bar supplying each bridge

Tipping  Buckets. The test results for the tipping bucket prototype cleaning system are
presented  in Table  5.2.c-4.   This  table  presents  tabulated  results for  what are
considered representative solids conditions.   For several test events, the  preceding
inflow did not result in any significant deposition of solids,  or the test section may have
been cleaned by tidal inflow or by plant personnel. The data for such events was not
considered representative and is not included in Table 5.2.c-4.  Data for the influent and
overflow test sections are presented separately, with averages for the number of bucket
tips per  test section and projected water usage per basin calculated over each test run,
each test section (influent, overflow), and over the entire basin.  The average over the
basin length  assumes a linear distribution of solids (and  associated cleaning water
consumption) between the solids depth measured at  the influent test section, and that
measured at the overflow test section.

The tipping buckets provided very effective cleaning of the sidewalls and the basin floor.
The  high velocity wave  produced by each bucket tip easily cleaned solids and debris
from the basin floor and conveyed them to the center velocity channel.  With greater
depths of solids or for heavier solids or grit, more tips were required as compared to test
conditions with fewer solids or with looser solids.
                                       88

-------
Jable 5.2.C-3.  Operating  Data for Existing Traveling Bridge
             Tali!? 5.2.C-3 Operating       fur fxisling Travelling         Clean ing System
                '* ; i
-------
Table 5.2.C-4. Tipping Bucket Cleaning System Test Results at Representative Solids
              Conditions
            tliHwi'h.r
                   •»f*	

                                        90

-------
Averaged over the basin length, 7 tips are estimated to be required per bucket, resulting
in a projected water usage of 291.4 m3 (77,000 gal) per basin.  The range in projected
water usage is from a low of 41.6  m3 (11,000 gal) per basin, to greater than 1249 m3
(330,000 gal) per basin.  The  highest projected water consumption was  required to
clean a pile of matted leaves mixed with solids in the influent test section during test run
no.  16.  In this case, the leaves  seemed to bind the solids together, resulting in a single
large  mass that was difficult to clean.  This pile was slowly eroded with  each tip. After
30 tips the test was terminated while the pile still remained.  Difficulty cleaning leaves
was also observed in the overflow test section of run no. 10.  There was a pile of matted
leaves in the test section  for bucket TB-1B.   With each tip, the entire mass of leaves
was lifted, carried a short  distance, and redeposited.  Ultimately,  12 tips were required
to wash this pile into the center velocity channel.

The solids that had collected along the training wall proved difficult to clean and were
often  the limiting factor in  determining when the basin was 99%  clean.  Elsewhere on
the  test section floor, the wave of wash-water was able to mix and resuspend the solids
in the wash-water.  However, along the training wall  the solids would generally erode
slowly with each tip.  In the prototype installation, the wall-floor fillet on the eastern tank
wall was installed flush against the  sidewall in the recess between the column pilasters.
As a result, over the pilaster width, there was no smooth transition for the wave between
the  wall and floor, and the wave would lose energy in this dead zone when it impacted
the  floor.   The modified  plywood  training-wall improved  the  situation somewhat by
eliminating the dead zone created  by the sidewall pilaster.  The plywood training-wall
worked by preventing the deposition of solids in the area blocked by the pilaster. Solids
along the modified training wall still eroded slowly with each tip. However,  the amount
of solids collected along the  training wall was less  than for the original vertical training-
wall installation.

When the training-wall was removed from the test area for bucket TB-2A, the cleaning
effectiveness was the same if not more effective than with the training-wall.  Without the
training-wall the wave could penetrate laterally into the solids, resulting in more effective
resuspension of the solids  at this location.

Operationally, the buckets  were very simple to operate, requiring only that the water be
turned on;  however,  difficulty in tipping first one bucket (TB-2B) then  another bucket
(TB-1B) was observed.   Due to possible misalignment of TB-2B during  installation,
counterweights were added to the bucket so  that it would tip automatically during filling.
This worked for a time.  However,  later in the test  program the bucket had  to again be
tipped manually.  Periodically, the same problem was observed with bucket TB-1B.

Flushing Gate.  The flushing gate  produced a high velocity flow in the  center velocity
channel that extended at least 76.2 m (250 ft) down the length of the tank. The velocity
of the leading edge of the  wave is a function of the slope of the channel and the steep
hydraulic grade  created by  the wave surge.  The center velocity channel  slopes
approximately 0.63  percent towards  the cross  channel.   The  flow velocity  was
                                       91

-------
calculated using Mannings Equation ranged  between 1.3—1.9 m/s (4.4—6.2 ft/s) for
uniform flow at typical wash-water depths of 0.15 and 0.30 m (6 and 12 in).  During
operation of the flushing gate, the leading wave velocity was measured at 3.4—4.3 m/s.
(11-14 ft/s).  The water velocity  in the channel later dissipated to 1.2 m/s (4 ft/s) after
approximately 40 s and 100 s at the 30.5 m and 76.2 m (100 ft and 250 ft) measuring
stations, respectively.  Although there were  generally  no significant solids within the
velocity channel,  this high  initial flow rate would be expected to scour most if not all
solids and objects from the channel.

Based on the size and configuration of the gate supplied, which was primarily  designed
to flush the velocity channel, the lateral extent of the high velocity flow was limited to the
velocity channel.  In  the immediate vicinity of the gate,  the high velocity flow extended
laterally approximately 61—152.4 cm (2—5 ft) from the center channel. However, the
high rate of discharge produced backwater within the basin effluent test area, which did
not result in any effective cleaning of the test area outside of the velocity channel.

As  indicated previously, significant  operating  problems were encountered  with the
flushing  gate.   These problems were primarily related to floatables and debris that
caused problems with the closure and sealing of the gate.

Fixed Header with Spray Nozzles.  The test data for the spray nozzle prototype system
are presented  in Table 5.2.C-5.  The spray nozzles did not provide effective  cleaning.
The nozzles only cleaned the immediate area impacted by the central portion of the
spray jet. Outside the central portion or cone of the spray jet, the nozzles produced a
fine mist, which was  ineffective at cleaning.  Away from this immediate area the wash-
water would form channels  in the deposited  solids and would not provide any further
cleaning.  Overall, approximately only 25% of the floor test-area was cleaned  by either
nozzle configuration.  This degree of cleaning was achieved within the first minute of
operation.  The  system was generally allowed to operate for 15 min.  However, the
added operating time did not result in any further cleaning.  Since the spray nozzles did
not provide  complete cleaning,  the cleaning times and volumes presented  in  Table
5.2.c-5 have been qualified to indicate that greater cleaning time and volumes  would be
required.  The flat spray nozzles directed at  the sidewall provided an adequate job of
cleaning the wall only.

Based upon field observations, it appears that in  order to provide adequate cleaning
with this spray nozzle configuration, the headers would need to be swept back and forth
across the floor during operation, similar to the existing traveling bridge operation.

Fixed Header  with Orifices.  The test data for the original and  modified orifice header
configurations  are presented in  Tables 5.2.C-6 and 5.2.c-7, respectively. The original
header configurations  provided  only partial  cleaning of the floor test  area.  An area
extending approximately 1.5—3 m (5—10 ft) from the sidewall was completely cleaned
within the first minute of operation.   These solids were usually cleaned with  the  initial
surge of wash-water.
                                       92

-------
Table 5.2.C-5. Spray Nozzle Cleaning System Test Results (2 pg.)
          £ •  ; 5 • 1l i " 	~ ?

          "^ '*  '',   '- \ *| .?" "

          L i  :    i ~  S I
           •- !  = b   i £  « :   «
           ~ i   ^'  i ?  *i   -
                                        93

-------
Table 5.2.C-5. Spray Nozzle Cleaning System Test Results (2nd pg)
                 I!
                                      94

-------
Table 5.2.C-6.  Original Orifice Header Configuration Test Results
                  Q«fr3c-».;tfc:nJsr St'i. ! i
              &«.».! (hrrftu-*, itittiin Xo. 23-* Ei -S.
                                    ;.;-,;	
                                               95

-------
Table 5.2.C-7.  Modified Orifice Header Configuration Test Results at Representative
             Solids Concentrations (3pg.)
                   •%» """-
                   1C i
                  tl
               •5 iS
               a i,
          l:iil€ :
                                       96

-------
Table 5.2.C-7.  Modified Orifice Header Configuration Test Results at Representative
             Solids Concentrations 2nd page
                                       97

-------
Table 5.2.C-7.  Modified Orifice Header Configuration Test Results at Representative
             Solids Concentrations (3rd page)
                                             '1 I'.
                                       98

-------
Further away from the wall, the wash-water would form channels and the remaining
floor area would be  cleaned only very slowly, as the flow eroded the edges of the
channel.  Based upon field observations, it appeared that a higher velocity of wash-
water was required to completely resuspend the solids over the full width  of the test
section and resist the formation of channels. The headers installed at elevation +1.5 m
(+5.00 ft) provided adequate cleaning of the sidewalk   The  system was generally
allowed to operate for 15 min, by which time only 30—75% of the floor test area was
effectively cleaned. Since the cleaning was incomplete, the cleaning times and volumes
shown in  the table are qualified to indicate that actual values are greater than  those
shown.

Overall, the modified  orifice header configuration showed significant improvement over
the original orifice  header configuration.  The  modified orifice header no. 1A provided
the same degree of  cleaning, approximately  30—75% of the floor test area, as the
original configuration.  However, the area appeared to  be cleaned sooner than the
original configuration. Additionally, the wall-floor  fillet minimized the quantity of solids
deposited along the sidewalk  While no significant solids/grease buildup was observed
on the basin sidewalls, the spray from modified header no. 1A provided  better cleaning
action  than original configuration.  Because of the tangential flow  spray of the wash-
water onto the wall, all of the water flow washed down the wall in a sheet of water. In
the original configuration in which the orifices were directed more  at the wall, a  large
portion of the flow sprayed off the wall and rained down on the basin floor.

The combination of modified headers no. 2A, 2B, and 3 provided complete and effective
cleaning of  the  test  area  floor  and  sidewall within a  short  period  of time.   The
combination of the two lower headers (2A  and 2B) spraying towards the velocity
channel provided a high scouring velocity and prevented the formation of channels in
the flow path that would otherwise limit cleaning effectiveness.  With all three headers
operating, the test area was generally 95% clean within the first minute of operation.  As
shown in Table 5.2.C-7, the average required cleaning time for the test section was 2.68
min with a projected basin wash-water  usage of 530 rrrVbasin  (140,000  gal/basin).
During the prototype testing, the header operating sequence and flow rates were varied
to determine the optimum operating conditions. Simultaneous operation of headers 2A
and 3, followed by header 2B alone,  appeared to minimize the consumption of wash-
water.  Simultaneous operation of all  three headers was also tested, as this was  most
representative of the future  full-scale operating  scenario.   This  provided cleaning
effectiveness equal to sequential operation in a shorter time period,  but at higher
projected  water consumption.  A wash-water flow rate of 2.3 rrr/min (600  gpm) per
header (30 gpm/lf /header) was sufficient to clean the majority of the solids.   However,
in test run no. 13, a partially full trash bag was present in the test area and a flow rate of
2.8 m3/min (750 gpm) per header (37.5 gpm/lf /header) was required to move the trash
bag into the velocity  channel. The purpose of header no. 3 was to clean the sidewall
and the area behind  header no.  2A.   The flow rate on header number 3 was varied
between 0.8—1.5 m3/min (200—400 gpm or 10—20 gpm/ft).  A flow rate of between
                                      99

-------
0.8—1.2 rrrVmin (200—300  gpm) appeared to be sufficient to meet these cleaning
objectives.
Operation of the orifice headers was simple.  The biggest difficulty was modulating the
flow at startup to minimize the occurrence of an initial surge of water.  The steps taken
to minimize this effect  included slowly opening the valves at startup and tapping the
feed pipe to provide air relief points.  During operation, some rags and debris became
entangled on the orifice header pipe supports.  The  amount of material caught on the
supports was minor and did not affect cleaning performance. It is important to note that
after a total of 6 months of testing  the  lower headers, no  clogging of orifices  was
observed. This was originally identified as a concern for submerged orifices.

Submersible Mixer.  There is no specific test data on the submersible mixer other  than
solids depth measurements and qualitative observations.  From observations of the
residual  solids  distribution following basin de-watering,  the mixer  did not appear to
effectively minimize the deposition of solids.  Typically, solids were scoured from a small
area approximately 1.5—3.0 m  (5—10 ft) in diameter immediately at the discharge of
each of the mixer nozzles.  However,  the solids  depth immediately surrounding the
scoured areas was deeper than for other basins, indicating that the scoured solids were
immediately redeposited away from the nozzle discharge.  Often there was a deep pile
or ridge of solids formed around the scoured area. Elsewhere on the test area floor, the
solids were unaffected by the mixer.

The mixer did not have any effect on floatables and leaves, which remained  in the test
area following basin de-watering.

The discharge nozzles would periodically become clogged as indicated by decreased
mixing turbulence  in  the tank  during  mixer operation.   Following removal  of the
submersible mixer prototype, the flow distribution box and  several of the nozzles and
feed lines were found to  be  clogged with  leaves and debris.  This material apparently
became so  compacted in the distribution box that the top weld of the box cracked open.

Comparison of Technologies and Recommendations for Design

Summary of Cleaning Effectiveness.   The results of CDM's field  testing program
demonstrated  that either the tipping buckets or the modified orifice header system
would provide satisfactory cleaning of the basin floor, equivalent in performance to the
existing traveling bridge system.  Each of these  two  systems achieved  complete
cleaning of  the test areas.

The spray headers equipped with nozzles and  the original  orifice header configuration
provided only partial cleaning of the test areas. This configuration was considered to be
unsuitable for application at Spring Creek given the existing basin geometry.  The spray
headers  only provided  cleaning of the immediate area impacted by the nozzle spray.
This resulted  in cleaning only 25% of the test  area.   The  original orifice header
configuration was designed to provide sheet flow of water over the basin floor.  With this
                                      100

-------
system, channeling of the flow occurred rapidly, which limited the ability of the system to
clean the entire test area.  Under the best of conditions, this system cleaned only up to
75% of the test area.  These two systems are, therefore,  not considered further.  The
submersible mixer did not minimize the deposition of solids in the test area and created
deep piles and ridges of solids adjacent to scoured areas.  Also, the  mixer became
clogged with leaves and debris.   Therefore,  the submersible  mixer is not considered
further.

The flushing gate was effective at producing a high velocity wave with initial  velocity of
3.4—4.3 m (11—14ft) per second in the center velocity channel at 30.5—76.2 m (100—
250 ft) downstream of the gate.  This velocity would be more than sufficient to  scour
solids from the velocity channel or the influent barrels.  It was not intended to evaluate
the flushing gate for cleaning of the basin floor.  The  gate was demonstrated on the
center velocity channel as a surrogate for possible  application to the influent barrels.
Operational problems were encountered with floatables and debris interfering with gate
closure and sealing.  For full-scale application, some means of positive closure would
be recommended. Also, if an alternate  method of sealing without significant leakage
could not be developed, an alternate water supply would be required to fill the storage
volume prior to a flushing event.

The remainder of the discussion in this section will address the two successful prototype
systems for basin cleaning, tipping buckets and orifice headers.

The average cleaning requirements determined from the prototype testing for the tipping
bucket and modified orifice header systems are summarized in Table 5.2.C-8 and are
compared against the existing  traveling bridge system. The cleaning  requirements for
the tipping buckets and  orifice headers are based upon cleaning a single 6.1 m (2  ft)
length of the basin at one time, which was the length  of the test areas. The cleaning
time  indicated for both the buckets and  orifice headers  could be reduced from that
shown in  the  table by increasing  the size of the  cleaning zone, and consequently,
increasing wash-water pumping and conveying requirements.  However, for the purpose
of comparing the prototype test data, the cleaning requirements for the two systems are
presented on the same 6.1 m (20 ft) basis.

As shown in the Table 5.2.C-8, the average basin cleaning time for the tipping  buckets is
more than twice that for either the orifice headers or traveling bridges.  This is primarily
due to the multiple tips required  to clean each zone.  For full-scale implementation of
either the tipping buckets or orifice headers, the actual time to clean all six basins would
not be six times the  average basin cleaning time, but  rather some fraction of this
quantity.   The total  cleaning  time  would be  dependent  upon a number  of factors
including:

•  Length and number of cleaning zones in each basin
•  Number of basins to be cleaned simultaneously
•  The maximum allowable or cost-effective wash-water flow rate
                                      101

-------
•  The maximum  allowable  solids loading rate to the cross  channel and climber
   screens
The  last two factors would limit the number of cleaning zones in operation  at any one
time. The flow and solids loading rates selected would impact the sizing and cost of the
cleaning and de-watering screening equipment. Therefore, while it is difficult to assess
the minimum  full-scale cleaning  time without  considering all the  above factors, it  is
believed that  the use of the tipping buckets will result  in a  significantly greater total
cleaning time than either orifice headers or traveling bridges.

Table 5.2.C-8.  Projected Cleaning Requirements for a Single Basin Cleaning Event
       (Based  on 20-ft. Cleaning Zone).
Cleaning Parameter
Average Time (min/basin)
No. of Cycles/Zone (averaged
Tipping
Buckets
175
7 tips
Mod. Orifice
Headers (2 A, 2B,
and 3)
\
67
1
Traveling
Bridge
72
2-6 passes
   over basin length)

   Water Consumption

     - Influent Test Area   (mVbasin)      457.99            530.77             NA
                       (gal/basin)      121,000           140,229            NA
     - Average Entire Basin (m3/basin)      291.45             NA             272.05
                       (gal/basin)      77,000             NA             71,875

   Estimated Electrical Consumption
   (kW-hr/basin)                         40               40               40

   Level of Staffing Required for          Automated         Automated       Fully Manned
   Operation

   Note: NA denotes data not available.
The performance of the tipping buckets and the modified orifice headers compare
favorably with each other and with the traveling bridge system in that all three systems
generally are capable of achieving complete cleaning of the basin floor and walls.  In
one instance, run no. 16, the tipping buckets could not completely clean a pile of matted
leaves mixed with solids. After 30 tips the test was terminated as it was determined that
no further cleaning could be achieved. Similarly, difficulty with the tipping buckets
moving leaves was also noted in the overflow test area of run no. 10, where 12 bucket
tips were required to move a pile of leaves. The orifice headers exhibited similar
difficulty moving a trash bag.  However, once the lower header flow rate was increased
to 2.8 m3/min (750 gpm) per header, the bag was swept into the velocity channel.  With
                                       102

-------
the proper design flow rate, the orifice headers should be capable of providing complete
cleaning.

The average wash-water consumption for the tipping buckets and the traveling bridges
is similar at 270—290 m3 (71,875—77,000 gal) per basin. There is no average over the
entire basin  length for the modified orifice headers as this system was only tested at the
basin influent end.   This represents the worst case solids cleaning condition.  For the
influent test areas,  projected basin water consumption values are presented for the
tipping buckets and  modified orifice headers.  Although,  based on this test area the
tipping buckets use  approximately 14% less water than the orifice headers. The wash-
water volumes ranged from 100 m3 to greater than 1,250  m3 (27,500 to >330,000 gal)
per basin for the tipping buckets and 175—880 m3 (46,050—232,900 gal) per basin for
the modified orifice headers.  This data variability would need to be accounted for in full-
scale design by assuming  conservative design values.   Therefore, the difference in
wash-water consumption between these two systems is not considered significant.

Estimates of power consumption for the three systems are presented  primarily for
informational purposes.  Because the cleaning systems are operated only intermittently,
on the order of 8 h/week, the differences in electrical costs are insignificant.  As shown
in the table,  the tipping buckets are estimated to use the least amount of electricity while
the orifice headers use the most.  This is because the orifice headers are a high flow
rate,  moderate head system, while the buckets are a low flow and low head system.

Both the tipping buckets and orifice header system would be designed for automatic
operation in full-scale application.  This is an advantage over the traveling bridges that
are continually manned during operation.

Cost Comparison

The  estimated  cost  of implementing either tipping buckets or orifice  headers at the
Spring Creek facility are presented  in Table 5.2.c-9.  These estimates are based upon
average  solids conditions and water consumption.  For the  tipping  buckets, water
consumption was averaged over the length of each basin  based upon the influent and
overflow test area results.  The modified orifice header system was only tested on the
basin influent  end.   Similar to the tipping  buckets,  the water consumption  for the
overflow end of the basin would be expected to be less than the influent end, due to the
distribution of solids. For this analysis, this was not taken into account. It is likely that
the operating time for the orifice headers,  and associated electrical cost, could  be than
the calculated electrical cost.
                                      103

-------
Table 5.2.C-9.  Cost  Comparison  of Full  Scale Tipping  Bucket  Vs.  Orifice Header
Cleaning Systems
Item
Direct Capital Costs
Piping and Valves
Mechanical Equipment
Instrumentation and Controls
Electrical Equipment
Indirect Capital Costs
Contractor OH&P (1)
Design / Contingency (2)
SUBTOTAL CAPITAL COSTS
Annual O&M Costs
Electrical Costs
SUBTOTAL O&M COSTS
O&M PRESENT WORTH (3)
TOTAL
Tipping
Buckets
$ 781,000
$ 4,956,000
$ 10,000
$ 362,000
$ 1,283,000
$ 916,000
$ 8,308,000
$ 1,500
$ 1,500
$ 17,205
$ 8,325,205
Orifice Headers
Piping Alt 1
Pipe Gallery
$ 10,255,107
$ 120,000
$ 16,000
$ 300,000
$ 2,245,000
$ 1,069,000
$ 14,005,107
$ 3,000
$ 3,000
$ 34,410
$ 14,039,517
Piping Alt 2
Pipe on Roof
Deck
$ 5,840,000
$ 120,000
$ 16,000
$ 300,000
$ 1,318,000
$ 941,000
$ 8,535,000
$ 3,000
$ 3,000
$ 34,410
$ 8,569,410
  Notes:
    1.  Contractor's overhead and profit are assumed to be 21 % of the direct capital cost.
    2.  Contingency is assumed to be 15% of the direct capital cost for the Tipping Buckets
       and Orifice Headers Alt 2, and 10% for Orifice Headers Alt 1.
    3.  Present worth analysis assumes a 20-year design life and an interest rate of 6.0%.

The design assumptions used in this cost analysis were as follows:
      Tipping Bucket Design Assumptions
             Bucket unit volume
             Bucket length
             Number of buckets/basin
             Average water consumption
             Material of construction

      Orifice Header Design Assumptions
             Number of headers/zone
             Length of cleaning zone
             Orifice header diameter
             Orifice pressure drop
2.7 m3/m (220 gal/ft)
6.1 m(20ft)
50
290 m3 (77,000 gal)/basin
316L Stainless steel
24.4 m (80 ft)
25.4 cm (10 in.)
7030kg/m2(10psi)
                                      104

-------
            Cleaning flow rate (for each side of basin)
            Wall header                    0.2 m3/min/m (15 gpm/lin ft)
            Floor headers                  0.5 m3/min/m (40 gpm/lin ft) /header
            Total cleaning flow rate          1.2 m3/min/m (95 gpm/lin ft)
            Average water consumption     530 m3 (140,229 gal) /basin

Conceptual  level capital  and electrical operating costs are presented.  The total  net
present worth cost is calculated based upon design life  of 20 years and an interest rate
of 6.0%. Costs  are presented for two different feed pipe  layouts for the  orifice header
system. Alternative-l is based upon installation of the main wash-water feed headers in
pipe galleries located in  the former spray water channels. This alternative has been
included in the 30% design phase.  Due to the space constraints  in the gallery and
obstructions by support columns, this alternative involved  significantly more fittings and
pipe flanges than the second  alternative.  Alternative-2,  the preferred alternative for the
orifice headers, is based  on installation of the main wash-water feed headers on top of
the basin cover system.  Without the obstructions of the pipe gallery, the feed headers
could be installed in straight pipe runs with pre-fabricated sections and  with branches
running 90°  off the main  header.  This significantly reduced the number  of fittings and
pipe flanges. For comparison to the tipping  buckets, only the cost for alternative-2 will
be considered.

As shown in Table 5.2.c-9, the estimated present worth costs to implement either  the
tipping bucket cleaning system or the orifice header cleaning system are  equivalent.
There is only approximately a 3% difference  in these estimates, with the  tipping bucket
cost being slightly less.  The  operating electrical costs  are not significant compared to
the total capital cost. The cost for regular maintenance of  the equipment  is not included
in the estimate. The primary maintenance items could include the following:

•  Maintenance and periodic replacement of  pumps and valves
•  Maintenance and periodic replacement of  tipping bucket bearings
•  Repair/replacement of orifice headers
•  Entry into basins to remove large debris items
Currently long-term (e.g., 10 or 20 year) maintenance  data is not available on tipping
bucket or orifice header  installations in CSO facilities.   It would be  difficult to assign
maintenance/  replacement costs for  servicing  of the  bucket bearings  or the orifice
header  pipes.   Also, the cost for staffing during cleaning operations  has not  been
included in the cost comparison.  Both the tipping bucket and orifice header systems
would be automated for full-scale application.  Some minimal staffing would be required
to monitor and troubleshoot the cleaning operations.  However, this  could  largely be
combined with other facility staffing responsibilities.  A staffing labor estimate for either
of these two systems would be dependent upon operating philosophy and configuration
(e.g., sequential vs. concurrent basin cleaning,  staffing level). It is expected that  the
staffing  cost for these two alternatives would be similar and would be significantly less
than the costs for the existing traveling bridges.  Assumptions on the extent of O&M
costs required could significantly affect the total costs.   Given this uncertainty on  the
                                       105

-------
O&M  costs, only the electrical costs which are well defined have been included in the
total present worth cost.

Implementability and Operability

Both  cleaning technologies, tipping  buckets  and  orifice  headers,  are  readily
implementable.   There are  no  known  constructability  issues  that would limit the
applicability of one technology over the other.   Both technologies provide effective
cleaning.  However, matted leaves posed a cleaning problem to the tipping buckets
while  this was not observed to be an issue with the  orifice headers.  Because of the
higher velocity imparted by the orifice headers, this  technology  appears to be better
suited at cleaning matted leaves, piles, and large objects.

There were two operability issues noted with  the tipping buckets,  including potential
bucket  misalignment and  transient  overloading  of  the  cross channel  and  climber
screens by waves carrying  high solids loads.  Misalignment of one of the test buckets
(TB-2B) caused it not to tip automatically.  This was  retrofitted in the field  by  adding
counterweights.   Later a  second  bucket (TB-1B) developed  a hesitation  in  tipping
automatically.  With  up to 300 buckets planned to be installed in a full-scale facility, this
would present a significant concern.  The second issue was a concern whenever the
facility received a high inflow of grit. During testing, high quantities of grit were washed
into the cross channel with the initial few tips  of the buckets.  Because the buckets
produced discontinuous flow  surges separated  by zero flow conditions, the  grit would
often  settle out in the  channel and immediately before the climber screen when the
water velocity decreased.   This  caused the channel and climber screen to become
clogged with grit, which then had to be removed manually.

With the orifice headers there are two operability concerns, entanglement of debris on
the pipe supports and potential clogging of the orifices.  Regarding the first issue, some
rags/debris became entangled on the pipe supports  of the prototype system during
testing.   However,  the spray from the orifices  was sufficient to prevent a  buildup of
solids with the rags.  With  proper design, the number  of pipe  supports  and their
configuration can  be selected so as to minimize  this effect.  Periodic entry of  the basins
will likely  be required to remove entangled material.   As  for potential clogging of the
orifices, no clogging of the prototype  system was observed during the 6 months the
headers were  in the basins.  If clogging does occur,  the  orifices in the  PVC headers
may be easily cleaned by plant staff.
                                      106

-------
5.3 Converting Dry Ponds for Enhanced Treatment

Dry ponds are extensively used throughout the U.S. and other countries (EPA 1983).
These  ponds  have been  constructed to  reduce peak runoff rates (peak  shaving),
typically with little consideration to runoff quality improvement.  Dry ponds have been
widely  used in flood control projects to detain runoff and thus reduce peak flows and
water elevations  in the  receiving waters. These flow reductions can also improve the
aquatic habitat by reducing flushing of fish and other organisms from urban creeks (Pitt
and Bissonnette 1984).  Flow reductions also reduce downstream channel bank erosion
and bottom scour. The  use of many dry ponds in a watershed, without regard to their
cumulative effect, can actually increase downstream flooding or channel scour  problems
(McCuen, et a/. 1984). The delayed discharge of a mass of water from a dry pond may
be superimposed on a more critical portion of the receiving water hydrograph.

Because these ponds  are normally dry and  only contain  water for relatively short
periods of time, they can be constructed as part of parking  lots, athletic fields, tennis
courts and other multi-use areas. The outlets are designed to transmit all flows up to a
specific design flow rate, after which excess flows are temporarily backed-up. In many
cases, the ponds only contain water during a few rains each year.

Several dry detention ponds were  examined  as part of  the National  Urban Runoff
Program (NURP), with monitored pollutant removals ranging from insignificant to quite
poor (EPA 1983). Sedimentation may occur in dry ponds, but only during the major
storms when flows are retained in the pond. The deposited material must be  removed
after each treated rain, or it can easily be re-suspended by later rains and washed into
the receiving waters. Adler (1981) found that new sediment deposits have little cohesion
and without  removal as part  of a  maintenance program, or without several feet  of
overlaying water,  bottom  scour is  probable.  Because  of  the  poorly documented
stormwater pollutant control  effectiveness  of  dry detention  ponds,  they  cannot,  by
themselves, be recommended as viable water quality control measures. However, they
can be very effective when used in conjunction with other  stormwater control  practices
(e.g., between  a wet detention pond and an infiltration or grass filter area).

Stanley (1996) examined the pollution removal  performance at a dry detention pond in
Greenville, NC, during eight storms. The pond was 0.7 ha in size and  the watershed
was  81 ha of mostly  medium  density single  family  residential homes,  with some
multifamily units,  and a short commercial strip.  The observed reductions were  low  to
moderate  for  SS  (42—83%),  phosphate  (-5—36%),  nitrate nitrogen (-52—21%),
ammonia nitrogen (-66—43%),  copper (11—54%), lead (2—79%), and zinc (6—38%).
Stanley (1996) also summarized the median concentration reductions at dry  detention
ponds studied  by others.   In all cases, the removal of the stormwater  pollutants was
substantially less than would occur at well designed and operated wet detention ponds.
The  re-suspension of previously deposited sediment during subsequent rains was
typically noted  as  the likely cause of these low removals. The conditions at one of the
ponds  in Greenville, NC were  observed three years after its construction. The most
notable change was that the pond bottom and interior banks of the perimeter dike were
                                     107

-------
covered with weeds and many sapling trees (mostly willows), which indicated that the
interior areas were too wet to be mowed.  The perforated  riser was also partially
clogged and some pooling was occurring near the pond outlet.  It appeared that the dry
pond was evolving into a wetland. The monitoring activity had been  conducted a few
months after the pond was constructed and was not affected by these  changes. Stanley
felt that the wetlands environment, with the woody vegetation,  if allowed  to spread,
could actually increase the pollutant trapping performance of the facility. With continued
lack of maintenance,  the dry pond will eventually turn into a wet pond, with a significant
permanent pool. The pollutant retention capability would increase, at the expense of
decreased hydraulic benefits and less flood protection than originally planned.

Nix and Durrans (1996) examined the benefits of off-line stormwater  detention ponds.
Off-line ponds (side-stream ponds) are designed  so  that only the peak portion  of a
stream flow is  diverted to the pond  (by  an in-stream diversion structure). They are
designed to reduce the peak flows from developed areas, with no direct water quality
benefits,  and  are typically  dry ponds. Off-line ponds are smaller (by as much as 20—
50%) than on-line ponds (where the complete storm flow passes through the pond) for
the same peak flow reductions. However, the outflow hydrographs from the two types of
ponds are substantially different. The off-line ponds produce peak outflows earlier  and
the peak flows do not occur for as long a period of time as on-line ponds. If located in
the upper portion of a watershed, off-line ponds  may worsen flooding problems further
downstream,  whereas  downstream on-line ponds tend  to worsen basin outlet area
flooding. Off-line dry ponds can be used in conjunction with on-line wet ponds to provide
both water quality and flood prevention benefits. Off-line ponds have an advantage in
that they do not interfere with the passage of fish and other  wildlife and they do not
dramatically affect the  physical character  of the by-passed stream  itself.  On-line dry
ponds may substantially degrade the  steam habitat by removing cover and  radically
changing the channel dimensions. The peak flow rate created by online  dry ponds
reductions can  also have significant bank  erosion benefits in the vicinity of the pond,
although these benefits would be decreased further downstream.

Case Studies

The retrofitting  detention  pond case studies presented in this section cover a broad
range of conditions, from underground  facilities for very small paved drainage areas, to
large detention  facilities for major drainage areas.  The problems encountered in each
case history  are  described, along with the costs and benefits,  to the extent  that
information was available.

5.3.a. Sunnyvale Detention Basin, Santa Clara County, California

South San Francisco Bay  has a serious heavy metal pollution problem, especially for
copper,  and numerous methods are being investigated to  reduce the  discharges of
metals to the Bay.  Woodward Clyde Consultants (1994) conducted a  retrofit project for
the Santa Clara Valley  Nonpoint Source Pollution Control Program to demonstrate the
                                      108

-------
benefits of modifying an existing dry detention pond for enhanced water quality benefits.
The discussion in this section is mostly taken from that report.

According to a survey  conducted in 1990 by Woodward-Clyde (1990),  there  are
seventeen municipally owned and operated pump stations in Santa Clara Valley. These
pump stations generally consist of pumps, storage units such as a sump or a detention
basin, and inlet and outlet works.  The sumps and detention basins reduce the required
pump capacity needed to pass the peak flood flows. The pump stations provide flood
protection to low lying areas that have historically subsided and are now protected by
levees.  The pump stations have generally operated as single-purpose flood control
facilities. The pumps are designed to go on as soon as water begins to fill the basin
with the goal of emptying the basin as soon as possible after the event. Therefore, no
net pollutant removals are assumed to be occurring.

One retrofitting option to achieve water quality benefits is to change the pump-operating
schedule in order to increase detention time and  to provide for a seasonal wet pond. A
preliminary evaluation of retrofitting detention basins was encouraging and a pilot study
to retrofit a facility and perform testing to measure water quality benefits and costs was
August 1990 through July 1993.

The following tasks were conducted as part of this study:

      • Retrofit  the pump station and modify pump schedules to increase water
        retention and thus improve SW pollutant removal,

      • Conduct water quality sampling to estimate the pollutant removal effectiveness

        of the retrofitted detention basin, and
      • Measure  sediment concentrations in the basin in order to determine whether
        sediments are classified as hazardous waste.

The detention basin has a channel  between the  inlet and  outlet that,  prior to  the
modifications, encouraged short-circuiting. A gabion baffle was installed at the outlet to
reduce short-circuiting and  to provide better distribution of flow into the outlet.  Rock
was dumped into the channel  leading  from the inlet,  and a drainage pipe that ran below
the channel was blocked off.   Operational changes consisted of  modifying the pump
schedule to create a 0.61 m (2 ft) permanent pool at the outlet and  to provide temporary
storage and slow release of water over the depth  range of 0.61—0.73 m (2—2.4 ft).

Site Description.  The northern portion of Santa Clara Valley has a history of subsidence
caused by groundwater pumping.  In order to  protect subsiding areas from  flooding, a
system of levees and pump stations was built. The pump stations collect and pump  SW
runoff from these low-lying areas through the levees into the flood control channels. In
order to accommodate large flows  and reduce the number and pumping capacity of the
pumps required, some pump stations include relatively large sumps or detention basins.
An inventory of the pump stations indicated that there are nine such facilities in  the
Valley with relatively large detention basins (Woodward Clyde Consultants 1990). The
design and operating philosophy of these systems is to: 1) attenuate the peak flow to
                                      109

-------
reduce required pump size, and 2) drain the basins as soon as possible following the
storm so that flood capacity is available for subsequent storms.

An example of one of these pump stations is Sunnyvale Pump Station No. 2, located at
the junction of the Milpitas/Alviso Road (Route 237) and Calabazas Creek. The Pump
Station consists of four primary pumps, each rated at 1.1 m3/s (39 cfs) capacity, and
one auxiliary electric pump with a capacity 0.25 m3/s (9 cfs). The detention basin area is
approximately 1.8 ha (4.4  ac.)  in area and has a capacity of approximately 37,005 m3
(30 ac-ft)  as shown in Figure  5.3.a-1.  It receives runoff from a 187.4 ha (463 acre)
catchment consisting  of the following  land uses: industrial park  (30%),  commercial
(10%), and residential (60%). There is a 2.13 m (7 ft) diameter concrete  reinforced pipe
that drains into the basin.  A second 0.91 m (3 ft) diameter line drains a 101.2 ha (250
acre) catchment (primarily  open space) and bypasses the basin to the north and directly
enters the pump house.

Treatment Concepts and Retrofitting Objectives.  The pump stations could provide an
opportunity to reduce nonpoint  source loads entering the South Bay if they can be cost-
effectively retrofitted  and  maintained. The  primary treatment is settling of particles.
Settling  can  be  an effective  treatment for some pollutants that are mostly in  the
particulate fraction in stormwater. The particulate fraction of locally collected stormwater
exhibited a range of 36—94%  (mean of 69%) for copper and 24—97%  (mean of 66%)
for lead.  These high particulate fractions allow sedimentation to be an effective control
practice.

The retrofitting scheme would increase the detention time to allow more particulates to
settle out  into the basin  while  not significantly increasing the flood risk. A goal of this
retrofit was to prevent high flows from  re-suspending previously settled sediments in the
detention  pond. Scour protection was  provided  by  having  at  least a two-foot deep
permanent wet pool during the wet weather season.

A 24—40 h hydraulic detention time for a pool several  feet deep  was found to  be
necessary to  effectively settle out most of the suspended sediment in  the local
stormwater.

In  all  cases, the  basin  must  maintain a  relatively  large flood control capacity and
associated outlet  works and pumps  in order to provide  the necessary flood control
benefits.

Description of Sunnyvale Retrofit Activities.

Change Pump Operational Rules to Create a Permanent Pool and Temporary Storage.
In order to create  the permanent pool in the pond, the pumps were  set to turn off when
water levels in the basin (as measured at the outlet) dropped below 0.61 m (2 ft)  In
order to create temporary  storage, pump settings were adjusted to phase in (and out)
very slowly for depths between 0.61—0.73 m (2—2.4 ft). These operational conditions
created  a  temporary storage depth above the permanent pool of 0.12 m (0.4 ft) with a
capacity of 2158.6 m3 (1.75 ac-ft).  Because this is an existing flood control facility, the
                                      110

-------
5.3.a-1                             No, 2
                         i |
!L
                             ;J4!
                            ~~ ;t"-\-.
                                      111

-------
temporary  storage  depth was  determined primarily based on flood  control  and
secondarily on  water quality considerations.  The  temporary storage  depth  was the
maximum depth that would still allow the basin to pass the 100-year flood.

Prevent Short Circuiting. The pond contains a trapezoidal open pilot channel (2.44 m (8
ft) bottom width, 5.2 m (17 ft) top width, and 1.4 m (4.5 ft) depth) between the inlet and
outlet. In addition to this open channel, a 0.76 m (2.5 ft) reinforced concrete pipe (RCP)
is situated below the channel to convey low flows between  the inlet and outlet.  These
conveyances effectively  "short circuited" flows between the  inlet and outlet, a condition
that is highly unsuitable for water quality control.

In order to  limit this short-circuiting, three modifications were made. At the outlet weir
near the pump house, a  gabion baffle was constructed around the  original  outlet weir to
prevent short-circuiting of flows along the channel and to promote a better distribution of
flow  from the basin  into the outlet weir. A second modification involved placing rocks
into the channel near the inlet. The third modification involved covering the entrance of
the 0.76  m (2.5 ft) RCP with a steel plate  and  vertical riser that reduced the rate at
which flow would enter the drain below the trapezoidal channel.

Plug Storm Drain that Directly Entered Pump House. A 0.91 m  (3 ft) RCP drained a
101.2 ha (250 acre) undeveloped area west of the detention basin directly to the pump
house sump.  This pipe was plugged with sandbags  in one of the manholes upstream of
the sump to prevent the runoff from this drainage area from mixing  with outflow from the
detention basin  in the sample collection area.

Problems Encountered.  No significant problems were encountered during  the structural
retrofitting of the detention basin.  However, the pump control system needed major
repairs in order to operate the basin within  the water level  tolerances required for the
study. Specific problems were encountered with the liquid level sensors and transmitter
(inaccurate flow monitoring because of the very  low flow  rates) voltage instabilities
caused  when certain pumps came on  line, and  fluctuations in the  power supply.
Therefore, an important aspect in evaluating the feasibility of retrofitting pump stations is
the design and  condition of the pump control system and the possible need for repairs
and upgrading.

Monitoring Program.  The goal of this study was  to measure the total runoff and collect
flow-weighted composite water samples at both the inlet and outlet of the detention
basin during  and after  storms  in  order to  estimate pollutant removal  performance.
Sediment samples were  also taken to characterize basin sediments.

Station Design  and Equipment.  Automated  flow and water quality monitoring stations
were located  at the inlet and the outlet of the basin. The inlet pipe was a 2.13 m (7 ft)
diameter RCP that was quite low, placed at a level below the basin floor, and tended to
be full of water  during most of the wet weather season. The inlet sampling station was
located  10.7  m (35  ft)  upstream of the  end of the pipe  and consisted of  a  Druck
pressure transducer,  velocity meter, ISCO Model 3700 automatic water  quality sampler
and  Campbell Scientific CR-10  data logger/controller.  Flow volumes were estimated
                                      112

-------
using the measured velocity times the wetted area of the pipe and the flow-weighted
composite samples were to be collected at the inlet based on these estimates. Initially,
a Montadero-Whitney electromagnetic velocity meter was  used. However, the velocities
in the pipe were too low to measure with this instrument and it was replaced  in March
1992  by a Detectonics  I.S.  Surveylogger, which relies on  the  Doppler effect and
suspended sediment  passing  the instrument.  When  compared  with  estimates  of
anticipated runoff volumes, neither instrument appears to have measured flow velocities
in a consistent and accurate  manner. The primary cause appears  to be the  relatively
low velocities in the large pipe.

The  outlet  sampling was conducted  in  the pump  house where  a Druck  pressure
transducer, an ISCO Model 2700 automatic sampler, and CR-10 datalogger/controller
were  installed.  Sampling based on estimated flow  through the  pumps was planned.
These estimates were based on the pump run times and pump  characteristic curves
(which show the relationship between flow and head for the design rpm of  700). To
achieve  this,  the  data logger was connected to the  pump  house  control  panel  to
determine pump run times and calculate  discharge from  the sump.  Field  visits during
the 1991—92  season revealed that the  pumps  did not operate at the  design rpm,
especially during the warm-up period, resulting in inaccurate flow estimates much of the
time.

Sampling Methods. At the inlet, a pressure sensor was used to estimate the water level
in the detention basin. During each sampling event, flow was calculated as a product of
velocity and area by the CR-10 microprocessor.  Based on the flow estimate (which was
generally poor), the  CR-10 initiated and continued water quality sampling at pre-
specified flow intervals. During a sampling event, instantaneous velocity and  pressure
were  recorded each time a water quality sample was  taken. Based on anticipated
rainfall, the sampling algorithm in the CR-10 was designed to  instruct the water quality
sampler to collect twenty  500 ml sub-samples in  a 10 liter borosilicate bottle over the
duration of the storm event. Following the sampling event, the pressure sensor was also
used to measure water level drops in the pond.

At the outlet, average hourly flowrate was estimated based on the pump run times and
the pump characteristic curves (also inaccurate), and was recorded over the duration of
the wet-weather  season.  To begin an  event,  field  crews  manually   initiated the
automated samplers based on anticipated flow volumes for that storm. As with the inlet,
the automated samplers  recorded  instantaneous  flow measurements  when each
sample was collected.

Stations were visited prior to, during, and  after monitored  events to  ice  samples,
exchange sample bottles, and ensure proper equipment operation.  Conductivity, pH,
and temperature were measured during the site visits.

Data  Collected.   Eight storm events  were sampled. For six of these events, flow-
weighted composite water quality samples and hydrologic measurements were taken at
both inlet and outlet stations. In most cases, only partial flow measurements were made
                                     113

-------
because equipment malfunctioned, threshold  velocities were  not achieved, or there
were problems with the pump control system.

Due to the uncertainty in flow volume measurements, pollutant loads were not used to
estimate treatment effectiveness. Instead, effectiveness was estimated based on the
flow composite water quality concentration data,  assuming that  the  inlet and outlet
volumes for an event are equal.

Sediment samples were taken at three locations: in the center of the  basin, near the
inlet, and near the outlet. Three sets of sediment samples were  collected  during dry
periods when the basin was empty, or nearly empty, of water (June 15, 1990, May 14,
1992,  and on July 12, 1993). The first samples were obtained using a 10.2 cm (4 in)
stainless steel hand auger, while the other samples were collected by scraping the top
1.27 cm (0.5 in) of sediment with a Teflon™-lined scraper.

Flooding Analysis, Storm Hydrology, Water Quality and Sediment Monitoring Results

Flooding Analysis. Woodward Clyde used a reservoir routing model to estimate water
levels in the basin for the 100-year inflow event and for two pump operating scenarios.
The first pumping scenario corresponded to the  original pumping schedule used for
flood control. The second scenario corresponded to the revised pumping  schedule
appropriate for a multipurpose flood control and water quality control facility. Based on
the results of the model, the maximum water level in the basin for the 100-year flood did
not change by modifying the pump operation schedule.

Precipitation. Rainfall was measured with a tipping bucket, which registered the time
when  the bucket  collected 0.25 cm  (0.1 in) of rainfall. The ranges of storm volumes
during  the sampling period were from 1—5.6 cm (0.4—2.2 in) and the storm durations
ranged from 6—60 h.  The Synoptic Rainfall Analysis Program (SYNOP) examined the
long period characteristics of the local  historical rainfall data that was collected by the
NWS at the San Jose Airport  (Gage No. 7821).  The median event rainfall volume for the
San Jose Airport gage for the period from 1948—1989 was 1.27 cm (0.5 in).

Runoff.  Flow measurements  collected  at the inlet and outlet for various events were
compared  with  rainfall  to  calculate  the  volumetric  runoff  coefficients.  The flow
measurements at both the inlet and outlet stations were not considered very reliable, as
the measured runoff coefficients ranged from  0.1—1.89. Woodward Clyde estimated
that the actual  values  would  be  about  0.5—0.8 for these  rains  and watershed
characteristics.

Comparison of Inlet Water Quality to Other Santa Clara SW Monitoring Station Data.
Laboratory chemical analyses were conducted on the water samples collected at the
basin  inlet and outlet stations during the six storm events. The median flow-weighted
composite concentrations of total metals (cadmium, chromium, copper, lead, nickel, and
zinc) from the inlet station are  summarized in  Table 5.3.a-1. The table shows median
concentrations  obtained  from  other  Santa  Clara Valley  SW  monitoring  stations
representing  residential-commercial,   industrial,   and  open   land  uses.   The  inlet
concentrations of copper, lead, nickel, and zinc are higher  than  concentrations from
                                      114

-------
open land use, but lower than concentrations at residential-commercial  and industrial
land use  stations. The cadmium concentration  appears to be  very  similar to  the
residential-commercial land use, while the chromium concentration is closer to the open
land use.

Table 5.3.a-1.  Comparison of Median Metal Concentrations at Inlet to Retrofitted Basin
      to other Santa Clara Valley Stormwater Monitoring Station Data
            Inlet to Retrofit  Residential/Commercia   Industrial Land    Open Space Land
 Metal          Basin        I Land  Use Station      Use Station       Use Station
               (n=6)            (n=21)             (n=25)            (n=4)
               (mg/l)	(mg/l)	(mg/l)	(mg/l)
Cadmium
Chromium
Copper
Lead
Nickel
Zinc
1.1
12
24
38
21
180
1.0
16
33
45
30
240
3.9
24
51
91
46
1,150
0.3
11
11
2.0
5.0
5.0
 *n= number of storm events

Pollutant  Removal  Effectiveness.  Table  5.3.a-2   summarizes   inlet  and  outlet
concentrations for total and dissolved metals (cadmium, chromium, copper, lead, nickel,
and zinc), SS, hardness and total oil and grease for  several storm events. Based on
these data, pollutant  reductions were estimated as the outlet minus inlet concentration
divided by the inlet concentration. The average pollutant removal effectiveness for the
metals ranged  from  about 30—50%. The metals removal  data  indicated that  the
removal of total chromium, copper,  lead, nickel and zinc were well correlated with SS
removal.

Comparison to Water Quality Objectives (WQOs). Of these metals,  total and dissolved
chromium,  lead  and nickel did not exceed the acute WQOs.  Total  and dissolved
cadmium exceeded the WQO in only one  storm  out of six  monitored storms. Total
copper at the inlet station exceeded WQOs in four out of six storm events. However,
concentrations  at the outlet  station never  exceeded  WQOs   (though  the  outlet
concentration was essentially equal to the WQO for one event). None of the dissolved
copper concentrations exceeded the acute WQOs. Total zinc concentrations at the inlet
and  outlet stations  exceeded  the  acute  WQOs  for all   six  storms.  Dissolved
concentrations of zinc  at the outlet station exceeded the  WQOs  in three of the  six
events.

Sediment Quality. The  objectives  of the  sediment  sampling was to  characterize
sediment quality in the detention basin and to compare the sediment concentrations to
hazardous waste criteria. Results of these sediment samples are summarized in Table
5.3.a-3.
                                      115

-------
Table 5.3.a-2. Inlet and Outlet Observed Concentrations and Pollutant Removals



SE17
Inlet
Outlet

Reduction
SE20
Inlet
Outlet
Reduction
SE21b
Inlet
Outlet
Reduction
SE23
Inlet

Outlet

Reduction
SE24
Inlet

Outlet

Reduction
SE27
Inlet

Outlet

Reduction
Average
Reduction
Cadmium
(l-ig/L)
nf | f
0.4 <0.2
0.2 <0.2

--

6.6 1.3
4.8 2.5
--

1.1 0.2
1.5 <0.2
--

1 0.2

0.6 <0.2

--

1.6 <0.2

1.3 0.2

--

1 0.5

0.6 0.4

--

--
Chromium
(l-ig/L)
nf | f
3.6 1.8
2.7 1.1

25% --

12 1
6 1
50% --

18 1
14 1
22% --

11 <1

8.3 1.4

25% --

21 1.1

15 8.6

29% --

6.3 1.4

4.9 1.7

22% --

29% --
Copper
(l-ig/L)
Nf
8.7
6.8

22%

24
9
63%

24
16
33%

27

12

56%

40

24

40%

14

8.9

36%
f
5.
4
4.
7
--

3
3
--

2
2
--

5

4.
7
--

2.
1
5

--

5.
4
4.
5
--
42% --
Lead
(^g/L)
nf | f
6.4 2.
2
3.4 1

47%

45 1
10 1
78%

53 <1
35 <1
34%

30 1

12 <1

60% --

76 <1

40 1.
4
47% --

13 <1

6.6 <1

49%

53%
Nickel
(l-ig/L)
nf | f
1.7 <2
1.7 <2

0%

16 1
4 1
75% --

25 <1
19 <1
24% --

13 3.
9
5.8 2.
2
55% --

42 9.
6
29 15

31% --

83 63

25 20

70% --

51% --
Zinc
(l-ig/L)
nf f
46 28
26 19

43% --

180 19
73 22
59% --

180 5
120 7
33% --

190 41

82 45

57% --

270 22

160 31

41% --

70 35

47 26

33% --

44% --
ss
(mg/L)

12
73

39%

90
24
73%

140
93
34%

74

31

58%

180

96

47%

30

15

50%

50%
TH
(mg/L)

97
120

--

110
63
--

--
--
--

100

90

--

140

140

--

110

220

--

--
O&G


1.5
1.4

7%

0.2
<0.2
--

--
--
--

0.7

0.5

--

0.6

3.5

--

1.6

1.3

--

--
nf: non-filtered (total)
f: filtered ("dissolved")
Removals are only given
SS: suspended solids
TH: total hardness, as CaCOs
if most observations were >PQL
O&G: oil and grease
                                          116

-------
Table 5.3.a-3.  Sediment Observations (mg/kg)

          %TOC   Cadmium   Chromium   Copper  Iron    Lead   Manganese  Nickel   Zinc
6/15/90     --       2.2        --        92      --      36       --        61    320
core
5/14/92
SWface     3.8       23       200       150   49,000    280      610       94    750
Inlet
Middle      5.5       17       220       140   38,600    350      640       87    570
Outlet      1.9       35       140       47    47,700    18      680       76    260
7/12/93
SWface     2.4       1.0       170       110   34,000    260      560       96    220
Inlet
Middle      0.65      0.2       120       37    36,000    12      700       75     85
Outlet      0.93      0.3       110       43    30,000    24      570       73     63
TILC                100       2,500     2,500    --     1,000      --       2,000   5,00
                                                                             0
      TTLC: Total Threshold Limit Concentration

In the second and  third  rounds of sampling, the  highest concentrations for copper,
nickel and zinc were found at the inlet station. Cadmium, chromium and lead were also
highest at the inlet station for the July 12, 1993 sampling round. The high concentration
of the majority of the metals near the inlet is consistent with other studies.

Average  sediment concentrations  observed in Pump Station No. 2 are  compared  in
Table 5.3.a-4 with sediment data collected from other detention basins in the Valley and
elsewhere. Results  from the various  basins  differ  substantially  and  indicate that
sediment quality is highly site  specific and varies depending on soils, catchment land
use, and other factors, especially time when the samples were analyzed (for lead).

To evaluate whether the sediments were hazardous, concentrations were  compared to
standards established in  the California Administrative Code, Title 22. Under Title 22,
there are two criteria for designating solids as hazardous  waste. The first criterion is that
the sediment  concentrations  not  exceed the Total Threshold  Limit  Concentrations
(TTLC). The second criterion is that the extract obtained  from the whole effluent toxicity
(WET) extraction method  not exceed  the  Soluble Threshold  Limit Concentrations
(STLC). For this pilot scale screening level of analysis, it was considered adequate to
compare with the TTLC only. In situations where disposal is being considered, the WET
extraction test should also be conducted.

None of the sediment sample  concentrations collected in the Sunnyvale Pump Station
basin exceeded the TTLC. The highest concentrations of cadmium, lead, and zinc were
4, 3, and  7  times   lower than the  TTLC,  respectively. The  highest  concentrations
reported for chromium, copper, and nickel were 11, 17 and 21  times  lower than the
TTLC, respectively.  Based  on these concentrations, sediments are not  considered
hazardous.
                                       117

-------
Table 5.3.a-4. Comparison of Average Sediment Concentrations from Detention Basins
      and Swales (mg/kg)
Detention Basin
This Retrofit Basin
Other Santa Clara County
Eastside Basin A
Easts! de Basin B
Eastside Basin C
River Oaks
Fresno NURP
Recharge F
Recharge G
Recharge M
Recharge EE
Recharge MM
Wigington (1983)
Bulk Mail Basin
Kmart Basin
Nightingale (1975)
Detention Basin
Special Pit
Wigington (1986)
Fairidge Swales
Stratton Woods Swales
Rte. 234 Rd. Swales
Cadmium
11.2

0.37
0.37
1
Nd

--
--
--
--
--

2.8
0.8

--
--

0.26
0.18
0.82
Copper
88

32
36
71
24

37
25
55
25
9.5

19
13

20
23

4.2
10
23
Nickel
80

36
40
100
72

32
37
53
22
11

--
--

--
--

--
--
-.
Lead
140

17
6
11
14

713
487
1333
297
93

112
368

224
801

42
18
936
Zinc
324

68
73
330
84

--
--
--
--
--

224
114

107
236

102
70
106
Cost Effectiveness Evaluation

The mean annual runoff volume of 433,000 m3 (351 ac-ft) was estimated based on
mean annual rainfall of 33 cm (13 in.) in the vicinity of the basin, an assumed runoff
coefficient of 0.7, and the 187.4 ha (463 ac.) catchment area. Mean concentrations and
removal efficiencies are averages of observed data.  For the metals, annual load
reductions ranged from 272 grams (0.6 Ibs) for cadmium to 29.5 kg (65 Ibs) for zinc. For
copper, the annual load reduction is estimated at 4.1 kg  (9 Ibs), which represents
approximately 40% of the total copper that enters the basin. Table 5.3.a-5 summarizes
the estimated cost effectiveness for the removal of heavy metals from the pond.
                                     118

-------
Table 5.3.a-5.  Estimated Mean Annual Load reduction and Cost Effectiveness*
METAL
Cadmium
Chromium
Copper
Lead
Nickel
Zinc
SS
Mean Inlet
Concentration
(mg/L)
0.002
0.012
0.023
0.037
0.038
0.156
87
Average Annual
Removal
Efficiency (%)
35
29
42
53
51
0.44
0.50
Load
Reduction
(kg/yr.)
0.3
1.5
4.2
8.2
8.2
29.5
18,600
Cost
Effectiveness
(kg/$1,000)
0.03
0.18
0.5
1.0
1.0
3.6
2,270
       *Assuming an annual runoff volume of 433,000 m3 (351 ac-ft)


Solids Accumulation and Removal.  About  18,600  kg  (41,000 Ibs) of SS would be
collected annually in the retrofitted detention basin, which represents about one-half of
the annual input of solids.  Assuming a  specific gravity  of about  1.5  for saturated
sediment collected in  the basin,  this would correspond to about  12.2 m3 (16 cyd) of
material annually. If uniformly distributed over the 1.78 ha (4.4 acre) basin, the mean
annual accumulation rate would be 0.08 cm per year (0.03  in. per year). Sediments are
expected to accumulate near the  inlet and, in this specific case, in the pilot channel. In
ten years, this accumulation rate would equal about  123.3 m3 (0.1 ac-ft) compared with
the capacity of the basin that is 37,000 m3 (30 ac-ft).  Therefore,  this  accumulation of
sediments does  not pose  a risk to  reducing  the flood  control capacity of  the basin.
Accumulation of at least 15.24 cm  (6 in.) of sediment  is required before removal is
practical. This amount of sediment may take as long as 10 or 20 years to accumulate.

Capital, Operation and Maintenance  Costs. Capital and O&M costs were estimated for
the retrofitted pump station and are  shown in Table 5.3.a-6. Costs were classified as
capital expenditures, O&M, and disposal. Capital costs for the structural retrofitting were
based on actual costs; whereas  the costs  for repair of the pump electronic control
systems were estimated. Operations and maintenance  assumes  100 hours per year
labor  in addition to that already being conducted to operate and maintain the facility for
flood control.  Disposal costs assume disposal is conducted every 10 years and include
estimated future costs for landfill fees, trucking, and excavation. The total annualized
cost is therefore estimated to be $8,200 for the 187.4 ha (463 acre) watershed, or about
$7.3 per ha ($18 per acre) of watershed per year.  The removal costs for copper were
estimated to  be about 0.5 kg per $1,000  (1.1 Ibs per $1,000)  that compares very
favorably with other  stormwater  control alternatives, e.g. copper removal via street
cleaning is approximately 0.7 kg per $1,000 (1.5 Ibs per $1,000) (Pitt  1979,  1985, and
1987).
                                      119

-------
Table 5.3.a-6.  Estimated Annualized Costs for Capital Expenditures and Operation

           	CATEGORY	COST
            1. Capital Expenditures
               Structural retrofitting = $15,000
               Amortized over 20 years at 8%                  $1,500
            2. Operations and Maintenance
               Inspection and repair (100 h @ $50/h)            $5,000
            3. Disposal (every 10 years)
               Landfill (122 m3 @ $38/m3                      $8,000
               Trucking (16 trips @$75/h x 2 h/trip              $2,400
               Excavation (122 cyd @ $7.7/m3                 $1,600
               Amortized over 10 years at 8%                  $1,700
            4. Total Cost per Year                           $20,200


Conclusions

Implications for Other Facilities. According  to an  inventory conducted by Woodward-
Clyde in 1990, nine of the existing 17 municipal pump stations in Santa Clara Valley are
designed with detention basins (rather than  sumps) and are, therefore, suitable  for
comparison with the pilot project. The detention basins range from 0.6—5.7 ha (1.5—14
ac.), with capacities of 5,550—182,556 m3 (4.5—148 ac-ft).  The watershed area  for
each pump station ranges from 10.1—405 ha (25—1,000 ac.) and the total watershed
area served by all nine stations is 1,725 ha (4,260 ac.) or 17 km2 (6.6 mi2). This is about
2% of the 900 km2 (350 mi2)  area of the Santa Clara Valley below the upland reservoirs.
If we assume that other similar facilities could  be retrofitted to achieve a performance
comparable to that measured at Pump Station No. 2, the net reduction in copper load to
the Bay would be about 45  kg (100 Ibs). This is only about 1% of the estimated  mean
annual load of 6,350 kg (14,000 Ibs) of copper entering San Francisco Bay.

   •  A 100-year flood analysis indicated that modification of the pump schedule to
     achieve water quality benefits did not increase the maximum 100-year elevation in
     the pond.

   •  Based on measured inlet and outlet flow composite concentrations from 6 storm
     events, the average pollutant removal efficiencies were: total chromium, 29%; total
     copper, 42%; total lead, 53%; total nickel, 51%; total zinc, 44%;  and total SS,  50%.

   •  The removal efficiencies for  chromium, copper, lead,  nickel and zinc  correlated
     well with SS removal, indicating that SS may be used as a surrogate parameter to
     monitor effectiveness of metals removal in detention basins.

   •  Metal  concentrations of basin sediments were generally  highest  at the inlet
     location.
                                      120

-------
  •  None of the sediment concentrations exceeded the TTLC standard, indicating that
     the sediments are not hazardous.

  •  The estimated mean annual load reduction of  metals ranged between 2.7—29.5
     kg (0.6—65  Ibs), depending on the metal. The mean annual load reduction for
     copper was 4.1 kg (9 Ibs)

  •  The amortized annual capital and O&M cost for retrofitting the Sunnyvale Pump
     Station No.  2  is estimated at  $8,200. The cost-effectiveness removal rate for
     copper is 0.5 kg/$1,000 (1.1 lb/$1,000).

•  Solids accumulation rates are very low and are estimated to be approximately 123.3
   m3 (0.1 ac-ft) over 10 years. Given that the basin has a capacity of 37,000 m3 (30
   ac-ft), increased deposition caused by retrofitting does not increase flood risk.


Implications for Management.

•  The  total watershed area in Santa Clara Valley  served by the nine  pump stations
   with  retention basins  is approximately or 17 km2 (6.6 mi2) (only 2% of the total area
   of the Valley downstream  of the  reservoirs). Thus, even if an improved  treatment
   performance could be obtained from these basins, the total load reduction to the Bay
   would be minimal. For example, the load reduction of copper would only be 45.4 kg
   (100 Ibs), which is less than 1% of the estimated mean copper load to the Bay.

•  Since pump stations  are relatively easy to retrofit, water quality benefits could be
   achieved by simply changing the pumping schedule.

•  If a retrofitting  program  is to be pursued, it would be important to ensure that the
   pump control equipment is operational and well maintained, and that staff are well
   trained in its use.

5.3.b. Monroe Street Detention Pond, Madison, Wisconsin

Introduction and Site Description

This case study considers the benefits of retrofitting  the outlet works of an existing wet
detention  pond. The original  pond was  creating severe downstream erosion  in the
channels between the pond and the receiving water, and the pond storage volume was
not effectively being used for either flood control or water quality benefits. The outlets
were modified and the pond has undergone extensive monitoring to confirm  the water
quality benefits of the retrofit.

The  US Geological Survey  (USGS),  in cooperation  with the Wisconsin  Department of
Natural  Resources (WDNR) investigated the Monroe St. wet detention pond  located in
Madison, Wl (House, et a/. 1993). The University of Wisconsin originally constructed the
pond to protect the water quality and  ecology of Lake Wingra and surrounding wetlands
from stormwater.  The pond is located on the downstream side of Monroe Street at the
                                      121

-------
outlet of a storm sewer that drains a 0.96 km2 (237 acre) urbanized area. Land use in
the watershed area consists mostly of single-family residences and  commercial strip
development, with some  institutional uses (i.e.,  schools and churches). The average
basin slope is 2.2%.

The Monroe Street pond has a surface area of 5,670 m2 (1.42 acre), a maximum depth
of 2.3 m (7.5 ft) and an average depth of 1.1 m (3.6 ft)  at normal pool elevation. The
shape of the pond is  round to  oval with  a small island. The inlet side is nearest to
Monroe Street and the two outlets are on the far side away from Monroe Street.  The
pond  has a surcharge storage volume above  the normal pool elevation that is capable
of holding the 10-year, 24 h storm-runoff volume without overtopping the containment
berm  around the pond.  The pond has two outlets, each controlled by 90° V-notch weirs
that drain to  channels  leading to Lake Wingra. The weirs are located in 24 m (8 ft)
diameter concrete vaults, with 0.8 m (2.5 ft)  concrete pipes leading to the pond. The
initial  primary outlet configuration consisted of two 2.4 m (8 ft) long rectangular weirs
located in the vaults, made with concrete block walls. The original flow  capacity of these
two weirs was enormous,  being  about 1.4 m3/s (50 cfs) at 0.3 m (1 ft) head and 7 m3/s
(250 cfs) at 0.9 m (3 ft) of head.  The outlets in the pond are submerged and the bottom
of the pond consists of a clay layer that inhibits infiltration  of water from or into the pond.

As  noted above,  the discharges from the  pond were little attenuated from the  inflow
velocities and severe channel  erosion was  occurring in the wetlands, negating  the
sediment trapping benefits of the pond.  The emergency spillway appeared to never
have been used since construction,  even with  several massive storms. In fact, the pond
elevation barely  fluctuated.  The water quality benefits of the existing dry pond were
assumed to be negligible.

The outlets were therefore modified in  1995 to  reduce  the  downstream  erosion
problems by  removing  several courses of concrete blocks  and installing 90° V-notch
weirs made of plate steel in each  vault.  The  pond normal water level was dropped
about 15.3 cm (6 in.) with a lowered invert.  The new primary outlets have total flow
capacities of about 0.14 m3/s (5 cfs) at 0.3 m  (1  ft) head  and 2.3 m3/s (80 cfs) at 0.9 m
(3 ft)  head.  The  pond surface fluctuates more  now, and the emergency spillway has
been  active every few years. Most significantly, the downstream channels are now
stable.

The pond was designed for an expected 90%  event mean concentration (EMC) removal
for  SS  (particulate  residue).  The ratio  of  pond to drainage area is 0.6%.   This
percentage is close to the  value  (0.4—0.8%) required for  5  micron  control, which
generally corresponds to a 90% reduction of SS.

The Monroe Street pond may have more water quality monitoring data than any pond in
the country. A total of 64 events were extensively monitored between February 1987
and April 1988. The monitored  rains varied  from 2  to more than 82 mm during this
period. Periodic water quality and flow monitoring has also continued at this pond since
1988.
                                      122

-------
Method of Investigation

Water-quality data were collected by the U.S.  Geological Survey (House, et al.  1993)
using programmable automatic refrigerated water samplers installed at the  inflow and
outflow sites  of the pond.  The outflow data was collected at two locations, east and
west. The samplers were programmed to obtain flow-proportional  samples for each
storm. These samples represent the flow-averaged constituent concentrations during a
runoff event. These samples were removed from the samplers, preserved, and shipped
to the Denver USGS laboratory for analysis within 24 h of being collected. The samples
were analyzed for SS, volatile total solids, total and dissolved chemical oxygen demand
(COD), total chloride, total and dissolved phosphorus, phosphate total and dissolved
forms of TKN, nitrates, and total and dissolved forms of copper, zinc, and lead.

Precipitation data were also recorded at 5 min intervals during the storm events using a
recording rain gage located at the pond site. Storm runoff (pond inflow) was monitored
at the box culvert that was the terminus of the 0.96 km2 drainage area.  Discharge rates
and flow volumes passing through the culvert were determined by use of a flow velocity
sensor and water  level  indicator installed  inside the  culvert. The velocity  and depth
sensors were connected to a data logger that recorded the water level and velocity data
and computed discharge rates based on the culvert geometry.

Data Analysis and Observations

The  pond inlet and outlet  pollutant concentrations were analyzed to determine  the
pollutant reduction.  Various relationships between inlet and outlet concentrations were
investigated by statistical analysis,  including Particulate Pollutant Strength  (PPS) and
percent controls. Each statistical process is described in the following paragraphs. The
basic data are contained in the USGS report. (House, et al. 1993).

Hydrograph/Flow Calibration.  An  important part of the Monroe Street project was
validating  the DETPOND wet detention pond water quality model  that was used to
design the retrofit of the outlet structures (Pitt and Voorhees 1995). The first  step in the
validation  was  to  check  flow  volumes  and  peak  flow  rates, and the  complete
hydrographs.

Fifteen storm events were used to validate the flow portions of the DETPOND program.
The  program predicted  outflow flow values from the inflow  hydrographs  using  the
modified pulse  routing method. The outfall predictions (at five-minute intervals) were
compared to  the observed outfall flow values.  The predicted outflow hydrographs very
closely matched  the corresponding  observed  outflow  hydrographs.  In addition  to
comparing the general shape of the discharge  hydrographs, the outflow total discharge
volume, peak discharge flow rate,  SS  removal, and outflow particle size distribution
were also compared for validation. The predicted outflow volumes and peak discharges
very closely matched the observed outflow conditions.

Observed Influent and Effluent Pollutant Concentrations. Table 5.3.b-1  lists the influent
and  effluent conditions observed at the pond. The number of observations, and  the
mean, maximum, and average concentrations are shown. The standard deviation (Std.
                                      123

-------
Dev.) and coefficient of variation (COV) are given. The Mann Whitney a values, which
indicates whether the influent and effluent values are statistically significantly different,
are also summarized.

Table 5.3.D-1.  Summary of Observed Influent and Effluent Pollutant Concentrations at
       Monroe St. Pond
RESIDUE

Number of
Observations
Minimum
Max.
Mean
Std.
Dev.
COV
*
Mann
Whitney
Total Solid (mg/L)
IN
OUT
64
57
76
63
5810
1180
526
276
816
259
1.55
0.94
0.0003
Suspended Solid (mg/L)
IN
OUT
64
57
21
4
1330
140
230
32
257
28
1.11
0.87
0.0000
Filtered Residue (mg/L)
IN
OUT
64
57




303
244
789
244
2.6
1.0
0.0758
Volatile Residue (mg/L)
IN
OUT
62
55
28
0
376
232
92
48
63
30
0.68
0.63
0.0000
Total Chlorides

IN
OUT
Number of
Observations
65
57
Minimum
0.8
2.1
Max.
3100
570
Mean
120
90.9
Std.
Dev.
430
135.5
COV
3.58
1.49
Mann
Whitney
.0032
COD
Total COD, (mg/L)
IN
OUT
63
57
21
16
370
350
90
37
63.4
44
0.70
1.2
0.000
Particulate COD, (mg/L)
IN
OUT
56
52
8
-3
327
281
53
16
53
38.5
1.0
2.4
0.000
Filtered COD, (mg/L)
IN
OUT
56
52
10
9
160
69
38
21.4
32.3
11
0.85
0.51
0.0019
          Coefficient of Variation - standard deviation divided by the mean.
                                        124

-------
Table 5.3.b-1. Summary of Observed Influent and Effluent Pollutant Concentrations at
      Monroe St. Pond, Continued
RESIDUE

Number of
Observations
Minimum
Max.
Mean
Std.
Dev.
COV
Mann
Whitney
PHOSPHORUS
Total Phosphorus, (mg/L)
IN
OUT
65
59
0.16
0.13
1.7
1.6
0.56
0.26
0.39
0.19
0.69
0.73
0.000
Particulate Phosphorus, (mg/L)
IN
OUT
60
53
0.08
0.04
1.53
1.52
0.33
0.16
0.27
0.2
0.82
1.25
0.000
Filtered Phosphorus, (mg/L)
IN
OUT
60
53
0.06
0.02
1.33
0.27
0.23
0.1
0.28
0.06
1.2
0.6
0.0001
PHOSPHATES
Total Phosphates, (mg/L)
IN
OUT
56
51
0.026
0.004
1.22
0.25
0.2
0.08
0.27
0.06
1.35
0.75
0.0002
COPPER
Total Copper, (i-ig/L)
IN
OUT
64
57
10
3.5
130
60
50
46
14
12.43
0.28
0.27
0.592
Particulate Copper, (ug/L)
IN
OUT
60
52
7
.5
120
49
43
42.3
13.93
12.02
0.32
0.28
0.0652
Filtered Copper, (ug/L)
IN
OUT
60
53
3
3
16
13.5
6.35
4.5
3.32
2.17
0.52
0.48
0.0035
LEAD
Total Lead, (ug/L)
IN
OUT
64
57
18
4
420
100
85
68
52.4
39
0.61
0.57
0.100
Particulate Lead, (ug/L)
IN
OUT
59
52
15
1
417
97
83
66
54
39
0.65
0.59
0.1800
                                     125

-------
Table 5.3.b-1.  Summary of Observed Influent and Effluent Pollutant Concentrations at
      Monroe St. Pond, Continued
RESIDUE

Number of
Observations
Minimum
Max.
Mean
Std.
Dev.
COV
Mann
Whitney
Filtered Lead, (ug/L)
IN
OUT
59
53
3
3
12
4
3.5
3.0
1.7
0.17
0.49
0.06
0.54
ZINC
Total Zinc, (ug/L)
IN
OUT
64
57
35
30
1100
1400
152
191
136
265
0.89
1.39
0.0450
Particulate Zinc, (ug/L)
IN
OUT
58
47
-3.0
-15.5
980
630
103.5
60.3
131
101
1.26
1.68
0.0000
Filtered Zinc, (ug/L)
IN
OUT
57
48
9.0
20.0
160.0
1218
51
150
34
215
0.66
1.43
0.0020
The coefficient of variation (COV) is calculated by dividing the standard deviation by the
mean and  normalizes  the standard deviation.  A high  value  indicates  that the data
spread is wide, requiring many data observations to obtain a  precise estimate  of the
EMC.

COV varied from 3.58 for inlet chlorides, which may not be significant due to seasonal
variations, to 0.06 for outlet filtered lead. All filtered lead had lower COV values which
indicated small concentration variations for samples.

The Mann-Whitney test is a non-parametric analysis comparing two sets of data. The
null hypothesis used in the Mann-Whitney is inlet pollutant concentration minus outlet
pollutant concentration equals zero. Generally, an a value <  0.05  indicates  that the
sample  sets  are from  two  different  populations  (significantly different at the 95%
confidence level). The following constituents had significant a values indicating that the
pond significantly affected the concentrations:
      Total solids
      Suspended solids
      Volatile solids
      Chlorides
      COD (all forms)
      Phosphorus (all forms)
Phosphate
TKN (total and particulate)
Nitrate
Copper (filtered)
Zinc (all forms)
                                       126

-------
Particulate Pollutant Strength.  PPS is the ratio of a participate pollutant concentration
to the suspended solid concentration expressed in mg/kg. PPS was calculated for each
pollutant with a particulate form.  The small  particles in stormwater have greater PPS
values than the large particles. Therefore, all pollutants had higher outlet than inlet PPS
values due to preferential removals of large particles  in the detention  pond, leaving
relatively more small particles in the discharge water. A wide difference between influent
and effluent PPS values indicates that the main components of the contaminant (such
as for TKN and phosphorus) are associated with fine particles.

Control of Pollutants. The reduction of pollutants was calculated from the difference in
pollutant concentrations in the inlet and outlet water for each event, as shown on Table
5.3.b-2.   As expected, control was higher for all  particulate forms  of the constituents
than for filtered forms,  except  for copper. Filtered constituents (<0.45  |j,m)  have
characteristics like colloidal particles and will tend to be transported through the wet
detention pond  relatively unchanged. A well designed wet detention pond will remove
70—90% of SS, 70% of COD, 60—70%  of nutrients and 60—95% of the  particulate
forms of the heavy metals.

Table 5.3.b-2. Summary Table of Pollutant Control*
           Parameter	10%	50%	90%
           Suspended solids         35          87            97
          Total Residue             <0          52            86
          Volatile Residue           <0          41            76
           Filtered Residue           <0          <0            56

           Particulate COD           15          80            95
          Total COD               29          60            84
           Filtered COD              <0          24            80

           Particulate               -20          60            80
           Phosphorus
          Total Phosphorus         <0          47            81
           Filtered Phosphorus       <0          43            83

           Particulate TKN           -40          40            80
          Total TKN                <0          45            75
           Filtered TKN              <0          12            68

           Particulate Zinc          -117         70            95
          Total Zinc                <0          31            69
           Filtered Zinc              <0          <0            59
               * Copper and Lead observations were mostly below the detection limits
                and are therefore not shown.
                                      127

-------
Particle Size Distributions and Short-Circuiting

Seven events were studied to find the short-circuiting "n" factors using observed and
predicted  particle size distributions in  effluent water.  Particle  size distributions were
measured using the Sedigraph method at the USGS Denver laboratory. This technique
measures settling rates of different size suspended solid particulates down to 2 microns.
The value of n is calculated using the concentrations of large particles that are found in
the effluent. In ideal settling,  no particles greater than the theoretical critical size (about
5 |j,m for Monroe Street) should appear in the effluent. However, there is always a small
number of these larger particles in the effluent.  It is  generally  assumed that short-
circuiting is responsible for these large particles. The measured values for n were one,
or less,  indicating  a high  degree  of short-circuiting  in the pond.  However, these
observations  were possibly affected by scour of bottom deposits near the subsurface
effluent pipes. The maximum  effect of  short-circuiting on pond performance is shown in
Table 5.3.b-3, showing the average reduction in  SS removals for different n values,
compared to the best performance (n value equal to 8):

Table 5.3.b-3. Average Suspended Solids Removal for variable n values for the
      Monroe
              St. Ponds
               n value      % SS removal       Reduction in % SS
                             (average)	removal compared to n=8
8
3
1
0.5
0.2
85
84
80.7
78.5
59

1
4.3
6.5
26
The median value of n observed was about 0.35,  indicating a reduction in SS capture
efficiency of no more than about 10%.  The effects  of this short-circuiting, has a minimal
effect on the SS percentage removals. The Monroe Street pond provided an average
SS reduction of 87%, compared to the design goal of 90%.  The small difference (3%)
between actual  and  design  SS  reduction indicates that the short-circuiting  has a
negligible effect on actual performance.


Monroe St. Pond Retrofitting Conclusions

The retrofitting of the Monroe Street wet detention pond was very successful. Changing
the outlet structures from large rectangular weirs to V-notch weirs significantly reduced
effluent flows,  reduced downstream   channel  erosion  and  improved water  quality
benefits of the pond.

All effluent concentrations were lower than associated influent concentrations,  except
for chlorides. The measured  short-circuiting factor indicated a severe short-circuiting
problem, but that could be a false indication due to minor scour near the effluent works
                                      128

-------
in the pond. Data suggest that  short circuiting does not adversely  impact  treatment
efficiency.  The  Monroe  Street pond is meeting all reasonable expectations  in both
downstream channel protection and in contaminant capture.

5.3.c. Birmingham, Alabama Dry-Ponds

This retrofit case study examines a dry  detention pond at a new apartment complex in
Birmingham, Alabama. This is an example of a typical small pond installed to meet local
regulations for peak flow control. The pond is not even expected to operate satisfactory
even as a peak flow rate pond based on review of the current hydrological assessment.
This analysis makes suggestions for retrofitting this pond to a wet pond and to improve
the overall benefits of  the device.  It also  demonstrates  how a combination of water
quality and flow  analysis computer programs can be used  together to evaluate and
design effective multiple benefit SW controls.  The water quality benefits of the existing
dry pond were  assumed to be negligible for this analysis.

Redesign of Dry Pond for Water Quality Benefits

There are a  multitude of options for retrofitting the existing dry pond:  excavation below
the outlet, raising the outlet, or some combination of the two. However, since  relocating
and/or reconstructing  the existing  outlet structures  to gain additional water volume
below the outlet would  be quite expensive we will only consider excavating below the
existing outlet  bcated at 219.5 m (720 ft) mean sea level (msl).  This  redesign will only
require excavation within the existing space with a minimum of re-grading.

There are several factors to consider when designing a wet pond, among these are
depth,  safety,  and storage criteria.  Pitt (1998) recommends 0.9—1.8 m  (3—6 ft) of
storage below  the invert  of  the lowest outlet, 219.5 m  (720 ft).  In addition,  safety
requires a shallow ledge, submerged 0.3 m (1 ft) below  the permanent storage area.
The redesign presented here will include the preferred depth guideline of 1.8 m (6 ft),
requiring excavation from  219.5—217.6 m (720—714 ft), and a minimal three-foot ledge
at 219.2 m (719 ft) to  match existing grades. The side slope will be  1:1 from 219.5 m
(720 ft) to the  ledge at 219.2 m (719 ft), and also 1:1 from the  interior of the ledge to
217.6 m (714 ft). Assuming  a prismatic cross-section, the additional wet storage to be
constructed below the 219.5 m (720 ft) of elevation is approximately 250 m3 (0.204 ac-
ft).

Safety Criteria. Pitt (1998) recommends  a gradual slope, 1:1 to 1:4, on the approach to
the pond's normal high water mark, 221.6 m (727 ft) in this case. The existing pond
may be dangerous,  as the side slope is approximately 1:2 and quite long about 23 m
(75 ft). As an additional safeguard, a barrier should be constructed to prevent access to
the pond, preferably some type of barrier vegetation.  This apartment  complex is a new
development with the majority of the construction occurring in 1997. Therefore, some of
the landscaping is not complete as yet. There are some shrubs and small trees planted
around the perimeter, however these do not completely surround the  pond. In fact, the
western slope,  which is the longest and  steepest approach to the pond, is unfortunately
the area that  does  not have barrier vegetation.   Since the stormwater runoff is  not
                                      129

-------
planned for water contact activities because of high levels of pollutants, e.g. bacteria,
some proactive effort should be made to discourage the public from making contact with
the water.

Peak Reduction Factors (PRF).  The dry pond barely  reduces the  peak outflow rates
from the contributing area.  The  100-year storm's runoff rate is expected to be reduced
from 4.3 m3/s (153 cfs) to about 4.1  m3/s (145 cfs) by constriction at the pond's outlet
structures and by storage in the pond.

      PRF = 1-Qo/Qi. (10)

      Where: Q0 = effluent
              Qi = influent

The  PRF  for this event is only 0.05, corresponding to a 5% reduction of the inflow
hydrograph in the pond. For the 50-year and 25-year storms, the PRFs are 0.06 and
0.07,  respectively.   This pond is  likely inadequate  for any  measurable peak flow
reduction.  The pond also likely exceeds  the maximum stage of 233.4 m (733 ft).  This is
especially a problem because the water  coming through the spillway flows directly onto
a busy street with high speed traffic.

Upflow and Critical Settling  Velocities.  The most important  rain events to control for
water quality improvement are  the small events  that occur with great  frequency,  or
those that produce less than approximately 3.18 cm (1.25 in.) of runoff.  For the smaller
rain events,  the DETPOND simulations demonstrated that the  retrofitted pond would
satisfy the maximum up flow velocity (or critical settling velocity) maximum of  0.0004
mm/s (0.00013 fps) which is necessary for 5 micron  particle control.   Analyses also
indicated that removal of 90% of SS was  easily obtained on all but  the largest rain
events.  The analyses using  DETPOND and the 1976 Birmingham rain file (a typical
year) showed that the annual average suspended control was 86%.

Pond's Water Quality Storage. A reasonable goal is to  have the pond's water quality
storage be equal to the runoff associated with 3.18 cm (1.25 in.) of rain for the land use,
development,  and  cover of  the watershed served  by the pond (Pitt 1998).  The
composite CN for  the  watershed was estimated to  be  75.  This  corresponds  to
approximately 1 cm (0.40 in)  of runoff, from a storm of 3.18 cm (1.25  in.) (Pitt 1987).
Therefore  the minimum active pond storage (between the invert elevation of the lowest
outlet and the secondary outlet discharge devices) required should be a least 1727 m3
(1.4 ac-ft). Since the retrofitted pond's water quality storage  is about 826 m3 (0.67 ac-
ft), this requirement is not met.

Pond's Surface Area Requirements. A pond's surface area should be sized as a percent
of watershed's area based on  land use, development, cover, and the particle size
control desired. The apartment site has residential and woodland land uses. Pitt (1998)
recommends  0.8% of the watershed area for 5 micron particle control in a commercial
area (based  on  NURP, National Urban  Runoff Program,  studies).    Therefore the
                                      130

-------
minimum surface area of the pond should be at least 0.13 ha (0.33 ac.), which is close
to the normal water surface area of the retrofitted pond of .125 ha (0.31 ac.).

Possible Problems. Experience and observations of similar ponds in this region of the
country have shown that the shallow areas near the shore of the retrofitted pond might
become a breeding area for mosquitoes. The shallow shoreline  might allow too much
aquatic vegetation to grow  into the pond's interior area causing congestion and
periodically dying and subsequently increasing the BOD. In  addition,  the aquatic plants
located on the shallow shelf could grow in and around the outlet and reduce its output
capacity. Reptiles  (including snakes), waterfowl,  rodents (i.e., muskrats) might  become
too plentiful in the  pond area and become nuisances.  Decaying vegetation may result in
strong odors  in the  summer,  and these odors  might  have  a  negative impact on
residents.

Background Information Related to Site Evaluation

The peak inflow hydrograph values were  determined by HydroCAD's soil conservation
service (SCS) TR-20 methodology. For the site, a SCS Type III rainfall  intensity
distribution  frequency (IDF) curve was selected.  Rainfall depths for the 100-year, 50-
year, and 25-year  storms were approximately 22 cm (8.6 in.), 20 cm (7.8 in.), and 18 cm
(7.1 in.) respectively.  Time of concentration for the watershed was also calculated
using  HydroCAD's built-in TR-20  methods; Tc = 24.3 min  for the apartment  complex
area, and Tc = 33.8 for the Woodland area.

Land Use, Development, Cover, Soils Type, and CNs. SCS soil maps for the area were
examined, and it was determined that the site consisted of Nauvoo-Sunlight complex,
with  15—25%  slopes, and Townley silt  loam, with  12—18% slopes. The  SCS
Hydrologic  Soil Groups for these soils are type B and type C respectively. Research
conducted at the University of Alabama Birmingham has shown that development, due
to construction disturbances, compaction, and soil mixing, can significantly reduce the
actual  infiltration  rates from those assumed.   Therefore,  the  CN assigned to the
developed  area was  for the worst case,  type D  soil. However,  the  undeveloped
woodland area, mostly Nauvoo soil, was assigned a CN based  upon the type B soil.
Therefore for the  developed area of 10.5 ha (26.4  ac.), a composite CN of 87 was
assigned.  This composite CN of 87 is a weighted average at the sub-basin CNs.  The
basins include 6.5 ha (16 ac.) of residential land use with one-eighth-acre lots, SCS soil
type D, and 65%  impervious cover (CN = 92),  plus the remaining 4.2 ha (10.4 ac.) of
open  lawns with  good grass  cover (> 75%), and type  D  soil  (CN = 80).   For the
woodland area of  6.1 ha (15 ac.), a CN  of 55 was assigned, corresponding to woods
with good hydrologic condition and type B soils.

Analysis of  Design Storms

HydroCAD™. The  HydroCAD Stormwater  Modeling System  (version 4.53) was used to
analyze the pond for hydraulic  characteristics. This computer software allows for
calculating  hydrographs,  based upon design  storm and watershed characteristics, and
then  routing these through a drainage system composed of subcatchments, reaches,
                                     131

-------
and  ponds.  The program does not consider the dead storage below the first outlet,
assuming that this is always full of water, therefore the flow characteristics of both the
existing pond and the retrofitted pond are identical.

The subcatchment component is used to model a given drainage area or watershed.  In
this case, there are two subcatchments: subcatchment 1 refers to the 10.5 ha (26.4 ac.)
of the apartment  complex,  and subcatchment 2 consists of  the 6.1  ha (15 ac.)  of
woodland area that drains to the complex.  The program  uses built-in SCS TR-20
hydrology  methods  for  determining  the  hydrograph  characteristics.    Next,  the
hydrograph  is routed through a series of defined reaches and/or ponds. In  this case,
there are two hydrographs, one from each of the two subcatchments, which are routed
through a single pond.

The pond component of this model is described using a stage v. surface area curve,
shown in Figure 5.3.C-1 and Table 5.3.c-1. In addition, the model requires descriptions
of the outlet structures.
                              Elevation v. Surface Area
                           Stonecrest Dry Detention Pond
            0.32
            0.28
            0.08
            0.04
              720  721   722  723  724  725  726   727  728  729  730  731   732  733

                                  Water Surface Elevation (ft)
Figure 5.3.C-1. Stage v. Surface Area Curve.
                                       132

-------
Table 5.3.C-1. Outlet Device Descriptions
        #
 Route
     Invert
                                          Outlet Device
        1
Primary     219.5 m (720ft)
       2
       3
 to#1
 to#1
219.5m (720 ft)
221.6 m (727 ft)
             Secondary   223.1 m (732ft)
                                          61 cm (24 in) culvert
                                          n=0.013, length=56.4 m (185ft),
                                          slope=0.02 m/m, Ke=0.5, Cc=0.9
                                          30.5 cm (12 in) orifice
                                          sharp-crested  rectangular weir
                                          length = 3.7m (12 ft)
                                          height = 20.3 cm (8 in)
                                          (square concrete box with cap)
                                          3 m (10 ft) broad-crested
                                          rectangular weir emergency
                                          spillway
       Ke = 0.5 represents the entrance loss for a flush entrance
       Cc = 0.9 represents the coefficient of discharge for the culvert

The  HydroCAD  simulations were run for three, 24  h,  SCS Type  III  design  storm
frequencies: 25-year 18 cm (7.1  in.), 50-year 20 cm (7.8 in.), and 100-year 22 cm (8.6
in.).  Table 5.3.C-2 summarizes the model's output for these three storms, showing the
hydrographs peaks and volumes. As expected, an examination of the last two columns
shows that the most significant contribution to the hydrograph flowing into the pond
comes from the  apartment  complex area.  The surface area of the complex is  larger
than that of the woodlands and land use, soil  characteristics,  and  other  hydrologic
characteristics of the complex all contribute to generating much more and rapid  runoff
per unit area.
Table 5.3.C-2. Subcatchment Summaries for Design Storms
Subcatch- Description Design Storm Rainfall Peak Flow
ment No. Frequency (yr) (cm) (in) (m3/s) (cfs)
1
2
1
2
1
2
apartment
complex
woodland
apartment
complex
woodland
apartment
complex
woodland
25
25
50
50
100
100
18
18
20
20
22
22
7.1
7.1
7.8
7.8
8.6
8.6
2.91
0.54
3.24
0.67
3.62
0.82
102.6
19.13
114.4
23.59
127.9
28.94
Volume
(m3) (ac-ft)
13,310
3,035
14,850
3,690
16,615
4,490
10.79
2.46
12.04
2.99
13.47
3.64
Next the hydrographs were routed through the pond.  Table 5.3.C-3 summarizes  the
results. As previously mentioned, the peak reductions are quite low (5—7%),  and  the
peak discharge lag is only a few minutes. The predicted peak water surface elevation in
the pond is higher than the maximum elevation  in the pond, 223 m (733 ft) (column 3),
                                      133

-------
indicating flooding. This is a dangerous situation  because it means that the water is
flowing uncontrollably over the dam. This could damage the emergency spill way and
cause erosion of the dam itself.  These events occur even at the lowest storm frequency
modeled 25-year. It would appear that the pond is inadequate for the amount of runoff
generated by these storms.

Table 5.3.C-3. Pond Results of HydroCAD simulations

 Design  Peak Elev.** Peak Storage    Peak Q jn     Peak  Q out   Peak Q out|ett  Peak Q ^gmt Atten.  Lag
 Event*                                                                     Time
        (m)   (ft)   rn3   ac_ft.  m3/s   cfs    m 3/s   cfs   m3/s   cfs    m 3/s   cfs    %    (min)


  25-yr.  233.5   733.4  2,603  2.11   36.1  118.3   33.6  110.2  15.4   50.5   18.2   59.7    7    8.1
  50-yr.  233.7   733.8  2,726  2.21
                                 134.2
                                       38.4
                                            126.1
                                                       51
                                                                 75.2
  100-yr. 223.£
             734.1  2,850  2.31
     * Design storms are type III 24 h for Shelby County (SCS methods).
      ** Flood elevation is at 223.4 m (733 ft).
      f Peak flow through the first and second outlets to a 0.6 m (2 ft) diameter culvert.
      t Peak flow in the emergency spillway (flowing onto Bowling Drive).

DETPOND Modeling.  DETPOND generally uses a simplified triangular hydrograph for
individual  storms in   long-term  simulations  (several  decades)  for  water  quality
evaluations, so this model is not suitable for modeling synthetic rainfall distributions,
such as those appropriate for design storms. However,  HydroCAD is capable of these
types of simulations, and one can input the hydrograph generated  by  HydroCAD into
DETPOND using the "user defined hydrograph" option. Therefore, the 25-year, 50-year,
and  100-year hydrographs generated using HydroCAD's TR-20 methods were input,
and these results are presented in the Table 5.3.C-4.  A comparison of the hydraulic
results from  HydroCAD  (Table 5.3.C-4)  shows  that these results are  similar.  For
instance during the 25-year event, HydroCAD calculated a maximum flow rate into the
pond of 36.1 rrfVs (118.3 cfs),  while  DETPOND's calculations resulted in 36.0  m3/s
(118.0  cfs).  Even  under  these worst-case  conditions,   the  pond  is  removing
approximately 50% of the SS.
                                      134

-------
Table 5.3.C-4.  DETPOND Summary for Design Storms
Design
Event
25 yr
50 yr
100 yr
Max.
(m)
223.6
223.7
223.9
Stage
(ft)
733.5
734.0
734.5
Max. Inflow
(m3/s) (cfs)
36.0
40.9
46.5
118.0
134.2
152.6
Max. Outflow Max particle
(m3/s) (cfs) size discharged
(Mm)
30.5
36.0
41.6
100.1
118.0
136.6
95
95
95
Avg. Min Particle
Size Controlled
(Mm)
26.
28.
29.
,5
,1
,7
%SS
Removed
50,
48,
46,
.8
.6
.7
Analysis Using True Rainfall Characteristics

DETPOND evaluates long-term continuous  simulations using many years  of  actual
rainfall data.  Rain files are created by determining the rain depth, rain duration, and
interevent duration  for  multiple storms.  DETPOND then routes a simple, triangular
hydrograph for each event, and routes it through the pond to evaluate the SS removal.
DETPOND  simulations  were  conducted  using rain files created  from  the  1976
monitoring data,  and  also the 1952 through  1989  rain record. The 1976 file is
representative of a typical year, and was used to enable more model runs  because of
shorter run time. The larger file was then run to  continuously evaluate the pond based
on almost forty years of actual storm data. There are only 23  events, out of a total of
4,107 in the large 40 year rain file in  which  the pond stage rises to the level of the
second  outlet. All of the DETPOND simulations were conducted using the retrofitted
pond, as water quality benefits in the existing dry pond were assumed to be negligible.
The results of the simulations using the 1976 file are presented in Table 5.3.c-5.

Table 5.3.c-5. Water Quality Output Summary for 1976 Rain File
Inter- Max Pond Flow- Approx.
Rain Depth Rain nEvent Rain Intensity Stage wnei^d Pnart' Peak
Duration DuratlO ,m> lm PaitlCle ReS. D-Hnrtinn
Statistic (cm) (in) Duaton n (ds) (cm/h) (m) (ft> size Control* ^f
(h) (in/h) Factor
Mean 1.27 0.50 12.01 1.81 0.10 0.04 1.94 6.30 4.26 86.3% 0.07
Std.Dv. 1.91 0.75 10.77 2.36 0.15 0.06 0.16 0.51 4.23 13.5% 0.07
COV 3.84 1.51 0.90 1.30 3.76 1.48 0.03 0.08 0.99 0.2% 1.00
Minimu 0.03 0.01 1.00 0.00 0.00 0.00 1.84 6.00 0.00 57.3% 0.01
Event
Flushing
Ratio**
1.75
2.41
1.37
0.00
 Max.     9.75  3.84    45    11.68  0.79   0.31  2.73  8.84    15.70    100%     0.31      9.34

      *  Approximate Particle Residue Control (SS).
      * * Event flushing ratio - the storm runoff volume divided by the watervolume contained in the pond at the beginning
         of the event.


The control for the 5//m particles corresponds to approximately 90% SS control based
on the mean values of the columns of Flow  Weighted Particle Size  and Approximate
                                       135

-------
Particle Control. On average, in a typical year, the pond will collect particle sizes 4.26
microns and greater  in size.   The average  rain depth  is 1.27 cm  (0.5 in.),  and the
average duration is 12 h.  For  the smallest storms, the pond is achieving greater than
95% control, and for the largest storm in 1976, the SS control is still 57%.

Figure 5.3.C-2 shows the maximum pond stage. Axis labels denote the elevation above
the pond bottom (1.8 m [6 ft] corresponds to 219.5 m [720 ft] msl elevation,  the invert of
the first outlet device) versus the percent particle control. There  is an expected trend;
the control decreases with increasing stage.


                            Redesigned Pond, Bham76
       o
       1:
       o
       O
       a>
       u
uu m •• •



<
/b
1
/U


I

• V
•
1 £
1 £
•







•_










* •


















»








•
                     6.5
                                7         7.5         8

                               Maximum Pond Stage (ft above bottom)
                                                               8.5
Figure 5.3.C-2.  Pond Stage v. Particle Residue Control


Figure 5.3.C-3  shows the water  quality  performance of the retrofitted pond (percent
particulate control) versus the rain depth in inches. There is considerably more scatter
in this plot, but  one can still observe a trend. Generally, control decreases as rain depth
increases, as expected. The  results are similar  for Figure 5.3.C-4  depicting control
versus rain intensity.
                                       136

-------
                                  Redesigned Pond, Bham76
IUUW -
r
95
90 "
5= 85 •
"o
0
o <
0) _c .
o 75
1 .

65 "

• •
•• •
• ••
'.
, v"v
• •
•
•
•
•
•




• •
• •
•
f
• •
• '

•
•




, •
1
•
•
•







• •.
• •








• :



























• •

                         0.5
                                        1.5       2       2.5
                                           Rain Depth (in)
Figure 5.3.c-3. Rain Depth v. Particle Residue Control
                                  Redesigned Pond, Bham76
                                                                      3.5
1001
95 '

IS 85~
"o
o
0 <
«
ra «


60 '
1
>:
• *
: -
i,;
• •
•
. I
<
•
<






•
•
• 1
"l.
1







•
i

















•









•
•








•
                         0.05      0.1      0.15      0.2      0.25      0.3      0.35
                                         Rain Intensity (in/hr)
Figure 5.3.c-4. Rain Intensity v. Particle Residue Control
                                            137

-------
Long-term Simulation of 1952-1989 Rains

Table 5.3.C-6 shows the statistics for the 4,107 rain events that occurred in Birmingham
from 1952-1989. Notice that the minimum and maximum values are different than those
from the 1976 simulations, but the  mean values are quite similar. The most important
value in this  table  is the  mean particle  control, 80%. For  1976, the control  was
approximately 86%.  This is not significantly different. With close to forty years of rains,
the redesigned pond still averages 80% SS  removal.

Table 5.3.C-6.  Water Quality Output Summary for 1952—1989 Rain File
                           Inter-              Max Pond    Flow-    Approx.
Statistic
Mean
Std. Dv.
COV
Minimu
m
Max.
Rain Depth
(cm) (in)
1.27
1.91
3.84
0.03

34.4
9
0.50
0.75
1.50
0.01

13.5
8
Rain
Duration
(h)
6.31
6.88
1.09
1.00

93
tveni
Duration
(ds)
2.57
3.54
1.38
0.00

44.31
Rain Intensity
(cm/h)(in/h)
0.10
0.15
3.76
0.00

0.79
0.04
0.06
1.48
0.00

0.31
die
(m)
1.94
0.16
0.03
1.84

2.73
ige
(ft)
6.30
0.51
0.08
6.00

8.84
weigniea
Particle
Size
6.43
4.91
0.76
0.20

23.2
rare. Kes.
Control*
(%)
80.13
14.36
0.18
48.50

99.8
Peak
Reductio
n Factor
0.13
0.10
0.76
0.00

0.57
Event
Flushing
Ratio**
1.91
2.44
1.28
0.00

9.96
* Approximate Particle Residue Control (SS).
** Event flushing ratio-the storm runoff volume divided by the water volume contained in the pond atthe beginning of the
event.
Conclusions

As described above, retrofitting a dry pond to a wet pond can provide improvements to
the overall benefits of the device.  Average SS removal for particles of 5 |j,m or greater
is expected to approach 80% removal annually.  Additionally,  a combination  of water
quality and flow analysis computer programs can  be used together to evaluate and
design effective stormwater controls and SW control structures can provide both water
quality protection and flood control.
                                       138

-------
5.4   Retrofitting WWF Storage Tanks for DWF Augmentation

The operation of a WWTF requires the accommodation of wide fluctuations in flow due
to naturally occurring  variations  in  the generation of wastewater and the associated
effects of I/I. Diurnal variations in flow rates are typically observed with higher flow rates
occurring during the daytime when water use is high and lower flow rates at night when
water use is low.

Wastewater equalization can  dampen these diurnal  variations to achieve a  relatively
constant hydraulic loading to the WWTF.   Equalization  basins  are  normally located
within the  WWTF following  grit and  screening facilities;  however, equalization is
sometimes located in the collection system to relieve overloaded trunk sewers.

Equalization can relieve overloaded treatment facilities in existing treatment plants and
can improve the efficiency, reliability,  and operability of these facilities.  Equalization
may also be a more cost-effective solution to modifying existing facilities.  Retrofitting
existing WWF storage tanks to serve a second function as DWF equalization tanks may
prove to be a cost-effective alternative to the modifying existing treatment facilities or
the construction of new equalization  tanks at the treatment plant.   Two  desktop
analyses will be evaluated to demonstrate this:

•  Retrofit of the existing Erie  Boulevard Storage System (EBSS) in Syracuse, NY

•  Retrofit of the existing Spring Creek Auxiliary Wastewater Pollution Control Plant in
   New York, New York.

5.4.a. Erie Boulevard Storage System, Syracuse, New York

Facility Description

The  EBSS was  constructed as  part  of  the  Best  Management  Practice  (BMP)
improvements in 1982.  The existing large diameter storm sewer running underneath
Erie Boulevard 2x3 m (7.5x10.5 ft) was retrofitted with automated sluice gates  and level
sensors to entrap CSOs  that  discharge  to this  structure.   These discharges  are
temporarily stored in the EBSS until the Syracuse Metropolitan Wastewater Treatment
Facility (METRO) has  capacity to accept the WWF.  The  18,950 m3 (5 MG) volume of
the EBSS can adequately contain  the  discharge from the 90 percentile storm in its
tributary area. Operational problems with control equipment prevented the system from
functioning as intended. Reactivation of the EBSS as a WWF storage facility is targeted
for completion by July 2002.

As shown in Figure 5.4.a-1, the EBSS receives flow from three major sources:

   1.  Burnet Avenue Trunk Sewer (BATS) overflows, including the James Street Relief
      Sewer (JSRS).
   2.  Fayette Street Trunk Sewer (FSTS) overflows.
   3.  Storm Sewers at the upstream end of the system as part of the natural drainage
      basin.
                                      139

-------
Figure 5.4.a-1.
                           i: J|

                               T  Z.
               II'
                                 •

                               15
                               -
                               	   s
                             ,,.<--   I
                             il   fr-1
                               _ SI

         ^--
                 fa**,,/   11:   i I-
               •'---"    «f  ,:-
                  '	
;•**; ,;*fc



  .?*-;-
    3«i^
                                                                                   li I
                                             140

-------
The EBSS will be automatically controlled through the use of state-of-the-art SCADA
control system.  This system consists of aboveground sluice gate control structures for
each of the  four control gates that  operate based on level sensors in the influent
sewers. The SCADA system also controls operation of the inflatable dam that regulates
flow from the BATS, the major source of flow to the EBSS.

Three sluice  gates (Gates 1, 3, and 4) on the EBSS establish maximum water surface
elevations as shown on Figure 5.4.a-l  Gate Chamber 2 is located on the regulator
sewer that connects the EBSS to the main interceptor sewer (MIS) that conveys flow to
METRO.  This control gate operates based on the water level in the MIS.

Upon the initiation of WWF into the EBSS, it is filled sequentially through the closure of
Gate Chambers 1, 3,  and 4.  Flow fills each basin sequentially until the 18,950 m3 (5
MG) volume is exceeded, at which time, flow exits the EBSS  through the  high-water
overflow at Gate Chamber 1. Gate Chamber 2 is closed except for dewatering of the
EBSS following a wet-weather event as described below.

Stored volume behind Gate Chamber 1 is dewatered first by opening Gate Chamber 2
to convey flow to the MIS. Gate Chamber 3 is dewatered following dewatering of Gate
Chamber 1 by opening Gate Chamber 3, and then Gate Chamber 4 is opened allowing
the stored volume to flow into the  lower chambers  to the MIS for ultimate treatment at
METRO.   Gate Chamber 1  remains closed during this phase.  Gate Chamber 1 is
opened upon final dewatering of the system and EBSS is returned to standby mode.

This facility in its intended wet-weather operational mode will be used to evaluate the
hypothetical  potential  for retrofitting the EBSS to  serve a second  function for DWF
equalization.   Equalization of DWF  in the EBSS could  prove to  be beneficial to the
treatment plant by eliminating the diurnal flow variability, providing METRO a relatively
constant DWF component from the EBSS tributary area.

Proposed Retrofit

A portion of the DWFwill  be directed  from the BATS to the EBSS through the JSRS by
the  use of regulator structure located in the BATS immediately downstream of the
JSRS.  This structure  is proposed  as part of the EBSS reactivation to direct WWF into
the EBSS. This structure contains a bottom-outlet orifice device to allow the DWF to be
conveyed downstream  through the BATS. The proposed retrofit will modify this orifice
to allow only a portion of the dry-weather to pass downstream while regulating the
higher  DWF into the EBSS.  DWF in excess of the  orifice capacity will be directed into
the EBSS where it will be stored and regulated to the MIS using  a new vortex valve that
would be retrofitted into Gate Chamber 2, for ultimate treatment at METRO.

The storage volume  between Gate  Chambers  1  and  3  targeted  for  use for DWF
equalization  is 8,680  m3 (1.5 MG)  based on the storage volume  below the invert
elevation of the JSRS.  This will allow the remaining 13,250 m3  (3.5 MG) of the 18,950
m3 (5  MG) EBSS capacity (2,650  m3 (0.7  MG)  remaining capacity between Gate
                                     141

-------
Chambers  1  and 3 and  10,600 m3 (2.8 MG) upstream of Gate Chamber  3) to be
available for storage of WWF.

Monitoring of the BATS downstream of the JSRS conducted from July 29, 1998 through
August 6, 1998 found the DWF to vary between 0.16-0.69  m3/s (5.7-12.8 cfs) with an
average of 0.22 m3/s (7.6 cfs).  It is proposed to divert DWF in excess of 0.15  m3/s (5.2
cfs) to the EBSS for equalization to  provide, on average, 8,680 m3 (1.5 MG) of flow to
be equalized  per day utilizing the capacity of the previously described DWF volume.
Stored  DWF will be routed to METRO through a new vortex valve that will be retrofitted
into the existing Gate Chamber 2 with  a capacity of 0.065 rrrYs (2.3 cfs).  This low
release rate will allow effective  equalization of  DWF and  utilization of the proposed
8,680 m3 (1.5 MG) dry-weather storage volume.  Gate Chamber 2 is directly connected
to the MIS and will serve as the drain of the entire EBSS during wet-weather operation.

Stormwater Management Model Results

The EPA SWMM Version  4 assessed the impact of the retrofit by comparing the use of
the EBSS under pre- and post-retrofit conditions.  A simplified model was simulated
hourly rainfall, based upon a previously calibrated model.

The data consisted of 30 years of hourly rainfall data form the NWS station at  Hancock
Field in Syracuse,  NY for the years from 1962-1991 and were used as input to the
simplified RUNOFF model.  The projected flows were used to simulate long-term
overflow using  the TRANSPORT block  of SWMM.  The  TRANSPORT  model  was
constructed as a simplified network consisting of overflows and regulator pipes.  The
simplified model was used to  project the annual  wet-weather capture from the EBSS
tributary area.

Table 5.4.a-1  demonstrates that the EBSS operated in a wet-weather mode provides a
considerable  annual  capture of WWF.   The  retrofitting of the EBSS for  DWF
equalization reduces the ability to capture and retain WWF.  Table 5.4.a-1 shows that
this retrofit reduces the overall WWF capture from 80% to 75%.
                                     142

-------
Table 5.4.a-1.  Average Annual Capture Volume from the EBSS Under Pre- and Post-
        Retrofit Conditions (1962 - 1991)
                        Average Annual Total
                        WWF in the Combined      Average Annual      percent
                            Sewer System         WWF Captured       Volume

  EBSS CONDITION       (m3)        (MG)        (m3)     (MG)      CaPture

Existing Wet-Weather     1,071,465      283      859,450    227        80%
Operation

Dry-Weather Retrofit     1,071,465      283      802,650    212        75%
Operation*1}

 (1) Assumes dry-weather portion of storage is fully utilized during wet-weather events to
   represent worst case


Design Considerations

Retrofitting the EBSS to  provide equalization of a portion of the DWF from the BATS
required careful consideration of several key design factors, as follows:

   •  Hydraulic Loadings

   •  Flushing Requirements

   •  Operational and Maintenance Requirements

These design factors are described in further detail in this section.

Hydraulic Loadings.  It is proposed to divert DWF in excess of 0.15 m3/s (5.2 cfs) to  the
EBSS for equalization. This would result in an average of 8,680 m3 (1.5 MG) of flow to
be equalized per day.  DWF in excess of the available 8,680 rr? (1.5 MG) storage
volume will be directed to METRO through the existing BATS.

Operational Requirements. The only adjustable control point necessary for this retrofit
is a pinch  valve located on the dry-weather orifice device.  The dry-weather orifice will
be automatically  adjusted when flow in the BATS exceeds the maximum DWF of 0.69
m3/s (12.8 cfs) or when the dry-weather storage  volume is fully utilized.  This will be
done by fully opening the pinch valve on the dry-weather orifice to allow the entire DWF
to  be conveyed  downstream through the BATS.   This will be controlled through  the
EBSS SCADA system. The pinch valve will be engaged in the normal operational mode
to only allow passage of the lower portion of the DWF.

Flushing Requirements. Accumulated solids deposition will increase during equalization
of  DWF, however, the layout of the EBSS serves to  provide considerable flushing
following wet-weather events.  This is accomplished by  staged release of  Gate
                                     143

-------
Chambers 3 and 4 to provide a wave of water to direct accumulated solids downstream
for conveyance to METRO through the MIS. This system will be an effective method for
the removal of accumulated solids from the DWF equalization.

Operation and Maintenance Requirements. The use of EBSS for DWF equalization will
result in increased  O&M costs.  The following  list summarizes the areas of expected
O&M cost increases:

   •  Labor to perform inspection of the facilities
   •  Power for the control system

Table 5.4.a-2 summarizes the expected increase in  annual O&M costs resulting from
the use of the EBSS for DWF equalization.

      Table 5.4.a-2. Estimated O&M Costs

                                                  Estimated Annual
          	Description	Cost	
           Process Control Adjustments                  $3,000
           Clean-Up of Wet-Weather  Treatment       $10,000
           Structures
           Power	$2,000	
                       Total Estimated O&M Cost       $15,000

Project Costs

The projected capital costs for retrofitting  the EBSS for DWF equalization are provided
in Table 5.4.a-3.

      Table 5.4.a-3. Estimated EBSS Retrofit Capital Costs

           Description	Estimated Cost(1)
           Flow Diversion Structure Retrofit               $10,000
           Flow Regulator Structure Retrofit               $15,000
           Electrical Controls                            $35,000
           Instrumentation and SCADA System	$55,000	
                                Total Project Cost      $140,000
      (1) Includes Design Engineering, Construction Engineering, Legal, and Fiscal
         Expenses, does not include contingencies and right-of-way costs
                                     144

-------
5.4.b. Spring Creek AWCP, New York, New York

The  Spring Creek Auxiliary Water Pollution Control Plant (AWPCP)  is a CSO wet-
weather detention facility with approximately 49,200 m3 (13 MG): 37,900 m3 (10 MG) in
basin storage and 11,400 m3 (3 MG) in influent barrel storage. The details of the Spring
Creek facility are provided in Section 5.2 of this document.


Proposed Retrofit

A portion of the DWF will be directed from the Autumn Avenue regulator in Brooklyn to
basin 6  to minimize the diurnal fluctuations  of DWF from Autumn Ave.  regulator
drainage area  that is tributary to the 26 Ward WPCP.  The DWF will be conveyed to
AWPCP to allow a specified level of DWF to be conveyed to the 26 Ward WPCP.

The storage volume in basin 6 targeted for use for DWF equalization is 3,790 m3 (1.0
MG) or approximately one-half of the storage  volume.  This will allow the remaining
34,000 m3 (9.0 MG)  of the 37,900 m3 (10 MG) AWPCP capacity to be available for
storage of WWF.

O'Brien & Gere Engineers previously performed an extensive evaluation of the 26 Ward
WPCP throughout the past 15 years.  Their work has shown the average DWF from the
Autumn Avenue regulator to vary between 1.06-2.07 m3/s (37-73 cfs) with an average of
1.5 mP/s  (53 cfs).  It is proposed to divert DWF in excess of 1.33 rrrVs  (47 cfs) to
AWPCP Basin No. 6 for equalization  to provide,  on average, 3,790 m3 (1.0 MG) of flow
to  be equalized per day.  Stored DWF will be routed to the 26 Ward WPCP through a
new dry-weather pump station with a capacity of 0.28 m3/s (10 cfs).  This low release
rate will allow  effective equalization  of DWF and utilization of the proposed  3,790 m3
(1.0 MG) dry-weather storage volume.

The  proposed  retrofit will consist of a  physical  modification to the  existing  diversion
structure by installation of sluice gate on the 1.5 m (60 in.) diameter dry-weather outlet
of  the regulator structure.  This sluice gate will be controlled by a flow meter located
downstream of the sluice gate in the 1.5 m (60  in.) diameter interceptor sewer to limit
flow to 1.33 m3/s (47 cfs). The regulator chamber will also be modified to include a weir
structure to convey DWF in excess of 47 cfs to Bay Number 6 for equalization. Level
sensors in the dry-weather equalization tank will also be used to control the dry-weather
sluice gate when the 1 MG volume is  utilized.

The dry-weather sluice gate will also  be  designed to open fully when sewage overflows
the Autumn Ave. regulator structure into  Spring Creek to maximize flow to the treatment
plant by the used of a level sensor located at the invert elevation of the overflow weir.
                                      145

-------
Stormwater Management Model (SWMM) Results

The SWMM Version 4 assessed the impact of the retrofit by comparing the use of the
Spring Creek AWPCP under pre- and post-retrofit conditions.  A simplified model was
use to simulate hourly rainfall, based upon a previously calibrated model.

The  data  consisted  of  hourly rainfall  data  from the  National  Oceanographic and
Atmospheric Administration rain gauge No. 305803 at J.K.F. International Airport in New
York, NY for the 1985. This data was used as input  to the simplified RUNOFF model.
The projected flows were used to simulate long-term continuous  simulation using the
TRANSPORT block  of  SWMM.   The  TRANSPORT  model was  constructed as  a
simplified network consisting of the overflow and regulator pipes.  The simplified model
was used  to project the  annual wet-weather capture from the Spring Creek AWPCP
under pre- and post-retrofit conditions.

Table 5.4.b-1 demonstrates  that  the AWPCP presently discharges a  considerable
amount of WWF to Spring Creek.  The retrofitting of the AWPCP for DWF equalization
reduces the ability to capture and  retain WWF only  to a small extent. Table  5.4.b-1
shows that this retrofit increases the annual WWF by 12% or approximately 50,000 m3
per year (13.2 MG per year).

  Table 5.4.b-1. Wet-Weather Discharge Volume from the Spring Creek AWPCP  Under
                Pre- and Post-Retrofit Conditions for  1985
                                                   Wet-Weather
                                                Discharge for 1985

            Spring Creek Storage Volume            (m3)       (MG)
Existing (37,850 m3 [10 MG])
Retrofitted (34,065 m3 [9 MG]) (1)
Increase In Wet- Weather Discharge
Percent Increase
433,005
482,967
49,962
12%
114.4
127.6
13.2
12%
        (1) Assumes dry-weather portion of storage is fully utilized during wet-weather
            events

Design Considerations

Retrofitting the AWPCP to  provide equalization of a  portion of the DWF from the
Autumn Avenue regulator required several design factors:

   •  Hydraulic Loadings
   •  Operational Requirements
   •  Flushing Requirements
   •  Operational and Maintenance Requirements
                                     146

-------
These design factors are described in further detail in this section.

Hydraulic Loadings - It is proposed  to divert DWF in excess of 1.33 rrrYs (47 cfs) to
AWPCP for equalization. This would result in an average of 8,680 m3 (1 MG) of flow to
be equalized per day. DWF in excess of the available 8,680 m3 (1 MG) storage volume
will be directed to the 26 Ward WPCP.

Operational Requirements - The two primary control points for this retrofit are the new
DWF sluice gate that will be  installed in the Autumn Ave. regulator and the new DWF
pump station that will be used to convey equalized flow to the 26 Ward WPCP.

The dry-weather sluice  gate will  be automatically adjusted when  flow in the 1.5 m (60
in.) diameter interceptor exceeds  1.33 rrfVs (47 cfs) or  when the 1 MG dry-weather
storage volume is fully utilized.

New level sensors will be used to control operation  of the DWF pump. These controls
will be set to energize the pump station when flow is sensed in the storage tank and to
turn the pump off when the tank is empty.

Flushing  Requirements -  Accumulated  solids  deposition will   increase   during
equalization  of DWF, however, a flushing  system will remove accumulated solids
following wet-weather events.  The design of new flushing system for the Spring Creek
AWPCP is presently being  evaluated to provide more efficient removal of accumulated
solids.  The selected system will be an effective method for the removal of accumulated
solids from the DWF equalization.

Operation and  Maintenance Requirements - An increased O&M burden will be created
for the  use  of Spring Creek for DWF equalization.  The  following list summarizes the
areas of expected O&M  cost increases:

   •  Electrical Power to operate the dry-weather pump station and the orifice control
   •  Labor to perform  inspection of the facilities
   •  Labor to perform flushing of accumulated solids

Table 5.4.b-2 summarizes the expected increase in annual O&M costs resulting  from
the use of Spring Creek for  DWF equalization.

      Table 5.4.b-2. Estimated O&M  Costs

          Description	Estimated Annual Cost
          Electrical Power                              $3,000
          Labor (Maintenance of Facilities)               $10,000
          Labor (Tank Flushing)	$25,000	
                  Total Estimated O&M Cost             $38,000
                                     147

-------
Project Costs

The  projected capital  costs for retrofitting Spring  Creek for  DWF  equalization  are
provided in Table 5.4.D-3.

      Table 5.4.b-3.  Estimated Spring Creek DWF Retrofit Capital Costs

        Description	Estimated Cost1
        Regulator Structure Retrofit                        $205,000
        DWF Pump Station and Force Main                 $170,000
        Process Control System                            $45,000
        Instrumentation and SCADA System	$65,000	
                                   Total Project Cost      $485,000
      11ncludes Design Engineering, Construction Engineering, Legal, and
       Fiscal Expenses,  Does not include contingencies and right-of-way costs

Conclusions

Retrofitting the EBSS and  Spring  Creek AWPCP were  evaluated  for the purpose of
serving  a dual role  as wet-weather and dry-weather  equalization  facilities without
drastically reducing the ability of the storage tank to capture WWF.  The primary benefit
of equalization provided by the DWF storage is the dampening of peak flow rates that
would otherwise result  at the treatment plant.  This  retrofit will increase the treatment
capacity for a treatment plant that is operating at or  near design capacity and will also
allow the primary and secondary treatment facilities to operate at a more uniform rate to
provide  consistent treatment efficiency throughout a typical day.   However,  there is
adequate hydraulic  and biological  capacity the Syracuse METRO and  26  Ward
treatment facilities.

This retrofit option may also provide a reduction  in operation of high flow pumps, where
appropriate, as a result of eliminating the  high diurnal  flows.   This  could result in a
reduction in O&M at the treatment plant; however, the O&M required at the equalization
facility may offset any benefit at the treatment plant.
                                      148

-------
5.5   SSO Control Using Storage

5.5.a. Hypothetical Overflow Retention Facility Retrofit to an Existing Sanitary
Sewer System

This desktop analysis involves a hypothetical retrofit to a sanitary sewer system to
eliminate  SSOs from an  existing regulator  structure.   This  desktop analysis was
developed from a real  life  project  from  a confidential  source.   Overflow from this
regulator occurs  during and following rainfall  events  due  to  a well-established  I/I
problem.  The I/I  problem was assessed as part of a sewer system evaluation survey
(SSES) that  identified and quantified various sources of I/I with the collection system.  A
rehabilitation program was implemented in accordance with the findings of the SSES to
provide removal of the primary I/I sources. This program also requires the elimination of
the existing regulator overflow.

Sewer system modeling was performed using the Environmental Protection Agency's
SWMM Version 4.  The previously calibrated model was expanded  to identify flow
conditions at the regulator following the implementation of the I/I rehabilitation program.
The model was used to identify peak overflow rates, volumes  and hydraulic grade lines
for various overflow abatement alternatives.
Project Approach

A 0.9 m (36 in.) diameter interceptor and an 0.5 m (18 in.) diameter interceptor enter the
regulator chamber.  A 0.9 m  (36 in.)  diameter changes to  a 1 m (42 in.) diameter
interceptor then  exits  the chamber  and continues to the  WWTF.   The existing
interceptor and other downstream piping configurations restrict the flow to the treatment
plant. An 0.5 m (18 in.) diameter overflow also exits the chamber that discharges to the
receiving water body when wet-weather events cause the chamber to overflow.

The regulator overflow was likely developed to relieve a  persistent flooding problem in
the vicinity of  the regulator.  The interceptor network in the vicinity  of the regulator
presently  surcharges  during the two-year design storm resulting in  approximately
12,100 m3 (3.2 MG) of surface flooding. The existing overflow relieves surcharge and
minimizes flooding.  The peak hydraulic grade line cannot increase beyond surcharge
with the overflow closed.  Closing the overflow would result  in pronounced  basement
and street flooding  in the  areas upstream of the existing  overflows  and would not
increase the peak flow to the treatment plant from the interceptor. The high head losses
are associated with the overflow and limitations of the interceptor network that includes
two double-barrel siphons downstream of the regulator overflow.  The alternative  of
constructing  a  new sewer parallel to the existing sewer to convey flow  to the treatment
plant was not considered feasible given long length of sewer,  deep rock excavation, to
long siphons, and a shoreline construction.

The two-year  frequency, two-hour duration design storm was selected  in order to
determine what control was necessary  in terms of treatment,  transportation or storage
of the regulator overflows.   Based on the SWMM model, during a two-year storm, a
                                      149

-------
peak rate of approximately 28,000 m3/d (7.4 MGD) overflows to the receiving water with
a total  overflow  volume  of 6,400  m3  (1.7  MG).   One recommended  abatement
alternative is to use a 6,400 m3 (1.7 MG) ORF to eliminate overflows up to the two-year
design storm and to treat overflows in excess of  the storage capacity with high-rate
disinfection.

The WWTF is presently permitted to treat 64,000 m3/d (17 MGD).  The monthly average
flow during a two-year monitoring period of the SSES was 76,000 m3/d (20 MGD).  As
the plant already exceeds its flow limit, recommendations to increase the peak WWF to
the plant were  not made, but to store the WWF during wet events with a slow release to
the WWTF following the storm.


Overflow Retention Facilities

The proposed retrofit consists of a 6,400  m3  (1.7 MG) ORF  located  between the
regulator chamber and the existing outfall to capture the flow associated  with the two-
year design storm.   Overflow events  producing  less  than  6,400 m3 (1.7 MG) are
retained and returned to the treatment  plant when influent flows subside.   Overflow
events producing in excess of  the available  storage  volume are disinfected using
chlorine and dechlorinated before being discharged to the receiving water.  The retained
volume of 6,400 m3  (1.7 MG) is returned to the treatment plant through the interceptor
sewer utilizing  an overflow return pump.

The overflow retention facilities include the following components:

ORF Tanks. The ORF tanks will consist of an in-ground  reinforced concrete tank with a
dimension of 30.5 m (100 ft) wide by 73.2 m (240 ft) long with a 3 m (10 ft) depth to
provide a useable retention volume of 6,400 m3 (1.7 MG).

Overflow Return.  A 10.2  cm (4 in.) overflow return pump will be used  to return the
captured volume to  the WWTF through the existing interceptor sewer within  48 h after
the occurrence of an overflow event.

Disinfection. The disinfection system will utilize sodium hypochlorite introduced into a
contact tank sized for 5 min detention time at peak flow  rate of 28,000 m3/d (7.4 MGD)
and requires the use of  mechanical  mixers to induce energy  to  provide  effective
disinfection. The mixing velocity gradient and chlorine dose were calculated on the
basis  of estimated  influent and effluent fecal coliform  levels for sanitary  sewer
overflows.    The  high-rate  disinfection system  was sized by estimating disinfectant
dosage and mixing energy required to  assure the inactivation of fecal coliform. Based
on the optimum velocity  gradient calculated, horsepower requirements for mixers were
determined. From this analysis, it was determined that an optimum  chlorine  dosage of
20 mg/l and approximately 1.1 kW(1.5 HP) of total  mixing energy are required to insure
proper disinfection.  A 5 cm (2 in.) disinfection  tank pump will be used  to return the
disinfection tank volume  to the WWTF through the existing interceptor sewer within  48 h
after the occurrence of an overflow event.
                                      150

-------
Dechlonnation.  Dechlorination, which reduces the chlorine residual prior to discharge,
will be accomplished by injecting sodium bisulfite in a  10 m3 (2,600 gal) tank with a
high-rate mixer to provide detention time of 30  s at a peak flow of 28,000 rrfVd (7.4
MGD). The dechlorination facilities are metered automatically as a function of overflow
flow rate to minimize chlorine residual.

Overflow  Metering.    Overflows  will  be  metered  following  discharged  from the
dechlorination tank through an open channel-type metering device located in the outfall
pipe.

Control Building. A 63.2 m2 (680 ft2) structure will be provided as the control building for
this facility. This structure will include process control equipment and be used to house
the chemical storage tanks.


Stormwater Management Model Results

The EPA SWMM Version 4 was used to assess the benefit of the retrofit by comparing
annual WWF volume presently discharged to the volume captured by the retention tank.
A  simplified model  was  developed to  simulate hourly  rainfall for  the  hypothetical
collection system

The data used for these projections consisted of 15 years of hourly rainfall data for the
years from  1961-1985. These data were used as input to the simplified RUNOFF model
that projected  flows  that were  used  to  simulate  long-term overflow using the
TRANSPORT  block of  SWMM.  The TRANSPORT  model was constructed  as a
simplified  network consisting  of hypothetical overflows  and  regulator pipes.   The
simplified model was used to project the annual volume presently discharged and the
percent of the annual volume captured by the proposed storage facility.

Table  5.5.a-1  demonstrates that the  6400 m3 (1.7  MG) storage facilities  provide
considerable  annual capture  of WWF compared to  that was  presently discharged
without treatment.

Table 5.5.a-1. Average Annual Overflow Volume from the Pre- and Post-Retrofit
       Facilities (1971 -1985)

                                      Average Annual Overflow Volume
                                          m3	(MG)
Pre-Retrofit WWTF Overflow
Post-Retrofit (1)
Percent Capture
70,400
18,000



75%
18.6
4.7

Although the ORF  was sized on  the  two-year  design  storm, the system annually
captures only 75% of the WWF that is presently discharged.  This occurs as a result of
the regulator chamber modification required to reduce the hydraulic grade line at the
                                      151

-------
regulator and  relieve surcharge and  street flooding in the vicinity of the regulator
chamber.  Therefore, the ORF will be active more frequently than overflow presently
occurs from the existing overflow.

Design Considerations

The retrofitting of a sanitary sewered collection system  to eliminate overflows through
storage and disinfection required the following design considerations:

      •  Hydraulic Loadings

      •  Treatment Objectives

      •  Operational Requirements

      •  Chemical Storage and Feed Requirements

      •  Sludge Handling Requirements

      •  Operations and Maintenance Requirements

Hydraulic  Loadings  - Hydraulic loadings  were developed for  the  ORF to provide
retention of WWF for subsequent treatment at the WWTF following a wet-weather event
and to provide disinfection of WWF beyond the 6,400 m3  (1.7 MG) ORF capacity.

Treatment Objectives - The objective of this retrofit is to eliminate  overflow from the
sanitary sewer system up to the two-year  design storm and to provide disinfection of
WWF beyond the two-year design storm level.

Operational Requirements - The ORF are  designed to operate on a continuous basis
include:

      •  WWF retention

      •  High-rate disinfection of flow in excess of ORF capacity

      •  Chemical Storage and Feed Requirements - The ORF utilize two chemicals:

      •  Liquid Sodium Hypochlorite for disinfection

      •  Liquid Sodium Bisulfite for dechlorination

The liquid sodium hypochlorite system is  based on commercial-grade  liquid sodium
hypochlorite (15%  solution) which contains the equivalent of 150 kg of chlorine/m3 of
solution (1.25 Ibs of Cb/gal of solution).  Two metering pumps are designed to deliver a
dosage of 20  mg/L at  the  peak design flow with one  unit as  a backup.  The liquid
sodium hypochlorite is stored in a bulk storage tank with an effective working volume of
11.4 m3 (3,000 gal).  The available chlorine in sodium  hypochlorite solution declines
with age. The anticipated sodium hypochlorite storage time will be 50 days.
                                      152

-------
The liquid sodium bisulfite storage system is based on commercial grade liquid sodium
bisulfite (38% solution) that contains appropriately 503.2 kg of sodium bisulfite/m3  of
solution (4.2 Ibs of sodium bisulfite/gal of solution).  Two metering pumps are designed
to deliver sodium  bisulfite dosages necessary to dechlorinate effluent with 1.5 mg/L of
sodium bisulfite per mg/L of chlorine residual. Approximately 1.6 kg of sodium bisulfite
will  be used per kg of chlorine residual. A bulk storage tank with an effective working
volume of 5.7 m3 (1,500 gal) will be used for sodium bisulfite storage.

The sodium hypochlorite and  sodium  bisulfite metering  pumps  are  flow paced.  An
ultrasonic type level sensor is used for effluent flow measurement.

Sludge Handling Requirements - Accumulated sludge will be returned to the WWTF
during the pump back of retained WWF and the associated tank flushing following tank
draining.  Tank flushing will be  accomplished manually using a 7.6 cm (3 in.) municipal
water service.

Operation and Maintenance Requirements - The implementation of the ORF will create
an O&M cost. The areas of expected O&M burden include:

      •  Chemicals for disinfection and dechlorination

      •  Power for tank draining and chemical feed systems

      •  Water for tank flushing.

The estimated annual O&M costs are provided in Table 5.5.a-2.

Table 5.5.a-2. ORF O&M Costs

                Annual Chemical Costs           Cost
             Sodium Hypochlorite           $ 8,200
             Sodium Bisulfite               $ 3,800
             O&M Labor                   $10,000

                           TOTAL COST  $22,000


Project Costs

The estimated construction costs for the ORF are provided in Table 5.5.a-3.
                                      153

-------
Table 5.5.a-3.  Estimated ORF Construction Costs

Item Description
Site Work
Common Excavation
Rock Excavation-Blasting Method
Select fill

Sheeting
Site Dewatering
New Pavement & Restoration
Fencing Restoration
(General Site & Bldg Restoration

Structural & Architectural

CIP Concrete-Base
CIP Concrete-Walls
CIP Concrete-Roof
Control Builriinp Brink & Block

Mechanical
Chemical Metering Pumps
Submersible Mixers-1.1 kW
Oder Control Blowers-1 1 .2 kW
ORF Tank Dewatering Pump-10.2 cm
Disinfection Tank Dewatering Pump-5.1 cm
Back Flow Preventer-7.6 cm
Ra^k\«rator Valvo-^T T r-m

Special Construction
Carbon Absorption Unit
Chemical Storage Tank (Chlorination)
Chemical Storage Tank (Dechlorination)
Flow Meter
SCADA Tansmitter
Jib Crane
Misc Safety Fquipment

Site Piping & Utilities
106.7 cm Interceptor Relocation
91 .4 cm Interceptor Relocation
61 cm ORF Inf/Effl. Sewer
45.7 cm Interceptor Relocation
182.9 cm Storm Sewer Relocation
Bypass Pumping Interceptor
Diversion Structure
Storm Sewer Junction Manhole
Special Manhloe for Sewer Crossing
10.2 cm Force Main
7.6 cm Facility Water Service
Wet Tap for Water Main Service
Temporary Overflow Bypass

Construction Cost Subtotal
Other Contracts
Electrical
Heating & Ventilating
Plnmhing
Subtotal Constuction Costs
Estimating Contingency
Total Construction Cost
Pnninoorinri & 1 Anal

Quantitv

1,529
574
459

743
1
785
61
1



214
191
199
63


2
5
1
1
1
1
1


1
1
1
1
1
2
1


18.3
55
9.1
4.6
67
1
1
1
1
6
67
1
9 1



15%
7%
9%

25%

9HOA

Unit

3
m
3
m
3
m
2
m
Ea
2
m
m
Fa


3
m
3
m
3
m
2
m


Ea
Ea
Ea
Ea
Ea
Ea
Fa


Ea
Ea
Ea
Ea
Ea
Ea
Fa


m
m
m
m
m
Ea
Ea
Ea
Ea
m
m
Ea
m



PCT
PCT
PP.T

PCT

PP.T

Unit Cost

$26
$131
$29

$377
$20,000
$24
$98
SIP nnn



$523
$1 ,046
$1,177
S1 507


$3,500
$3,700
$4,500
$3,500
$2,200
$2,000

-------
Conclusions

Retrofitting SSOs with ORF can provide a cost-effective alternative in lieu of expansion
of the sanitary collection system to convey the volume of overflow to the treatment plant
for  primary  and secondary  treatment.  This alternative requires  the availability of a
sufficient amount of land to build such a facility.
                                       155

-------
5.6   Retrofitting for Industrial Wastewater Control in a Combined Sewer System

EPA has identified maximizing flow to the POTW in combined sewer systems as one of
the nine minimum controls for CSO  abatement (EPA  1995).  The objective of this
control is to reduce the magnitude, frequency, and duration of CSOs that flow untreated
into receiving waters.   Industrial wastewater is often characterized by high-strength
pollutants that can  be detrimental to  receiving waters.    The control of industrial
wastewater in a combined sewer system can reduce the impact of the CSO discharges
on receiving waters.  A desktop analysis of the Rockland WWTF in Rockland,  ME, was
evaluated to demonstrate the technique.

5.6.a.  Rockland Wastewater Treatment Facility, Rockland, Maine

The city of Rockland, Maine owns and operates a secondary WWTF that discharges to
Rockland Harbor,  a  Class  SC water body.  The  WWTF was originally constructed in
1978  with upgrades in 1994  and  1998.  The Rockland WWTF provides secondary
treatment using conventional activated sludge.  The WWTF was designed to  handle a
monthly-average flow of 11,000 rrfVd (2.9 MGD) and an  instantaneous-peak flow of
22,000  mP/d (5.7 MGD).  The WWTF serves a CCS.   Rockland's existing collection
system services most of the residential population, commercial properties, and  industrial
properties within the limits of the  city  of  Rockland.   The system also serves the
neighboring community of  Rockport and the Samoset Resort.  The system  contains
three types of sewers:

      •  combined sanitary and storm
      •  separate sanitary
      •  storm

Nine pumping stations exist serving various areas of the collection system.  At present
there are approximately 21  km (13 miles) of gravity sewer ranging from  15 cm (6 in.) to
76 cm (30 in.) in diameter.   Pipe materials include stone and brick (dating back to the
Civil  War era) as  well as more recent installations of vitrified clay, asbestos-cement,
concrete, cast iron, and PVC pipe. According to the 1998 CSO Facilities Plan, there are
presently two permitted CSOs located within the system. In addition, five other CSOs
were located during the Facility Planning effort that began in 1993. In an average year,
187,000 m3 (49.3 MG) are discharged from these seven CSOs.

A majority of the wastewater currently entering the Rockland sewer system is domestic
sewage from  residential  and  commercial  sources.   The  WWTF also receives a
significant flow and loading  from the FMC Corporation Food Ingredients  Division facility.
This facility manufacturers carrageenan, a seaweed based food additive. The WWTF is
designed to receive approximately 67% of the organic loading and approximately 22%
of the DWF from the FMC Facility.  Annual average wastewater flows to the WWTF are
7,570 m3/d (2.0 MGD) and average daily BOD5 loadings are 1,565 kg/d (3,445 Ibs/d).

Prior to  implementation of the CSO Abatement Program, the WWTF used the following
unit processes to provide preliminary, primary and secondary treatment:
                                     156

-------
      •  Three raw sewage pumps with a capacity of 10 m3 (2,600 gal) per min each,
      •  One 92 cm (36 in.) comminutor with a manual bypass screen,
      •  Two primary  clarifiers (each at 14 m [45 ft] with a total surface area of
         approximately 295 m2 [3,180 ft2]).  Primary sludge is conveyed to two cyclone
         classifiers to remove grit prior to thickening,
      •  Six mechanically aerated activated sludge aeration tanks with a total volume
         of6,590m3(1.74MG),
      •  Two secondary clarifiers (each at 17 m [55 ft]  with a total surface area of
         approximately 440 m2 [4,750 ft2 ]),
      •  A chlorine contact tank with a volume of approximately 345 m3 (91,000 gal),
      •  A dechlorination chamber with a total volume of approximately 15 m3 (4,000
         gal).

Figure 5.6.a-1 shows the existing site layout prior to implementing the CSO Abatement
Program.  Figure 5.6.a-2 shows a schematic of the existing WWTF prior to  retrofitting
with a description of the numbered overflow points.  When influent peak flows would
reach 18,930 m3 (5 MGD), the WWTF influent pumps would be throttled and excess
flow  bypassed  to the outfall  downstream  of  the chlorine  contact tank resulting in
untreated, undisinfected discharges.

These overflows are discharged to Rockland Harbor via permitted CSO #002 (located in
the wetwell of the WWTF influent pump station Figure 5.6.a-2).  The wetwell receives
the FMC Industrial Wastewater plus WWTF sidestream flows from the primary sludge
gravity thickener overflow, waste activated sludge DAF thickener overflow, and sludge
dewatering system vacuum  filter filtrate.   The  industrial wastewater  and WWTF
sidestream flows are mixed with the combined  sewage upstream of the influent pump
station wetwell.  Consequently, these high-strength wastes are included in the overflow
discharged via CSO #002.

In March 1998,  a  CSO  Facilities Plan was completed. The CSO Facilities Plan was
approved  by the State  of Maine Department of Environmental Protection (DEP) and
EPA.   The approved  plan recommended  a three-phase program to address CSO
discharges.

Phase I of the CSO Abatement Program included rehabilitation of the Lindsey Brook
Sewer.  This sewer is located within the streambed for Lindsey Brook and has been
determined to be a considerable source of I/I during wet weather or high groundwater
periods.  The Lindsey Brook  Sewer is also suspected to be a source of exfiltration
during periods of dry weather.
                                     157

-------
5.6.a-1.                     Prior to
                                  158

-------
Fioure F> 6 a-?          Flow Schematic Prior to Retrofit
                                      tiff
                                      » S < '
                                    = i-  = - *
                                       159

-------
Phase II of the CSO Abatement Program  included upgrading the  WWTF to handle
additional WWF.   Phase II  also  includes new consolidation conduit  to intercept
overflows from the second and third largest CSO's (#004 and #005) that are located
approximately  92  m  (300 ft)  upstream of  the  WWTF on Main  Street (Route 1) in
downtown Rockland.  Phase II also includes a  new force main to separate industrial
wastewater produced at  the  FMC  Food Ingredients  Division  Facility from the City
interceptor sewer  and provide a  direct connection to  the Rockland WWTF.  Finally,
Phase II will redirect WWTF sidestream flows to the primary and secondary treatment
units downstream of the influent pumping station.

Phase III of the CSO Abatement Program  includes selected sewer separation in  the
south portion of the Rockland CCS to address the remaining CSO discharges.

This case study will focus on Phase II Improvements at the Rockland WWTF. Phase II
Improvements include:

      •  A new 76 cm (30 in)  diameter consolidation conduit from  CSO #004 and CSO
         #005 to the WWTF.
      •  A new flow diversion structure (SS#1) at the WWTF.
      •  Conversion of the existing headworks  into a  submersible pumping station to
         handle DWF and WWF.  Two dry weather pumps are  provided.  Each dry
         weather pump  has  a  capacity of 15 rr? (4,000 gal) per min.    Three wet
         weather pumps are provided.  Each wet weather pump has a capacity of 30
         m3 (7,800 gal) per min.
      •  Construction of a new headworks with  mechanically cleaned  bar screen,
         screenings  compactor,  manually  cleaned bar rack, grit chamber, recessed
         impeller grit pump, cyclone grit classifier, and 31 cm (12 in.) parshall flume.

      •  Construction of two new  upflow  style  flow distribution structures for the
         existing primary and  secondary clarifiers to improve hydraulic flow splitting.
      •  Construction of a new vortex separator to provide high-rate primary treatment
         for projected peak wet weather CSO flows of 127,560 m3/d (33.7  MGD).
      •  Construction of a new CSO  Disinfection and Dechlorination  Structure  to
         provide high-rate disinfection and  dechlorination for projected peak WWFs of
         114,690  nf/d (30.3  MGD); 127,560  m3/d (33.7 MGD) peak WWF minus
         12,870 m3/d (3.4 MGD) vortex underflow.
      •  Constructions of a  new 20 cm  (8 in.) diameter  force main for industrial
         wastewater.   This force  main  would  convey  industrial wastewater to  the
         existing  primary clarifiers,  existing aeration tanks, or a  new 1,100 m3 (0.29
         MG)  equalization tank.  The  equalization tank was created by refurbishing
         one (1) unused  existing aeration tank.  The equalization tank utilizes diffused
         aeration for mixing.  The equalization  tank contains two submersible pumps
         (1 primary and  1 standby). Each pump has a variable speed drive to allow
         output to be adjusted.
                                     160

-------
      •  Construction  of new plant  drain piping to eliminate discharge of WWTF
         sidestreams to the influent sewer. The gravity thickeners for primary sludge
         are eliminated by  replacing the recessed impeller primary sludge pumps with
         new positive displacement units designed to remove thickened sludge directly
         from the primary  clarifiers.   The waste activated  sludge filtrate drain  was
         redirected to the existing aeration tank inlet channel.  The sludge dewatering
         filtrate drain was redirected to the primary clarifier influent.

Figure 5.6.a-3 shows a  site  layout of the Rockland WWTF following construction of the
Phase II CSO Abatement Program. Figure 5.6.a-4 shows a process flow schematic of
the Rockland WWTF following Phase II of  the  CSO Abatement  program.   These
improvements will  allow  high-rate  primary treatment and disinfection for peak wet-
weather CSO flows at  128,000 rrfVd  (33.7 MGD).  This provides treatment  for CSO
#002, #004, and #005  that  represent approximately  90% of the total annual  overflow
volume from the Rockland CCS.

Stormwater Management Model Results

The EPA SWMM Version 4 was used  in the facility  plan to predict overflow rates and
volumes at all of Rockland's CSOs.  The SWMM  is  a relatively complex  mathematical
model. To model the various physical processes associated with stormwater runoff and
CSO discharge, a large  amount of input data is required. Drainage basin characteristics
(such as  size,  average  ground slope,  width, percent  imperviousness, roughness,
surface storage, soil infiltration and gutter or pipe flow) must be supplied to the model.

The RUNOFF and EXTRAN blocks of SWMM were used in this study.  The model was
constructed using records of the physical system: pipe sizes,  slopes, materials, lengths
and elevations.  Each CSO  diversion structure was modeled. The existing Park Street
Pumping Station was also modeled. Then drainage area characteristics were  modeled
for such factors as total acres, percent imperviousness, slope and dimensions.  Once
the SWMM model was mathematically constructed, it was calibrated using the extensive
rainfall and CSO quantity data  gathered during 1993,  1994 and 1995.


Unlike the calibration process, where actual "real-time" rainfall  data are used, SWMM
projections are made using two types of rainfall data:

      •  Synthetic rainfall hyetographs (or designs storms) for determining peak rates
         of overflow.
      •  Long-term rainfall records from  NOAA for determining statistically  based
         overflow volumes.

The former data are used in  the evaluation of pipes and conduits, pumps and CSO
treatment facilities where peak rates of discharge are  used as the basis of design.  The
latter data are used to design storage facilities and  to test the long-term effectiveness of
overall CSO abatement  programs.
                                      161

-------
5.6.a-3.
                     __i^
                                 »  :  i

                                 162

-------
5.6.a-4.          Flow
                                                 i
                                       l»

                  1 .--"--, P||      : JS^i;
                 *-* -* "     '.   t      1 J: 4 -, i
                                     163

-------
For the city  of Rockland CCS, the predictions were made using design storms with
return frequencies of 0.5, 1, 2 and 5 years.  A two-hour storm duration was selected for
the design to reflect an average of time of concentration of the Rockland  system.  The
SWMM  Modeling  identified several important hydraulic characteristics of Rockland's
CCS:

      •  CSO #005 has the  highest projected peak overflow  rate, CSO #002 (WWTF
         Bypass) has the second highest peak overflow rate.
      •  CSO #002 (WWTF  Bypass) discharges the greatest overflow volume with the
         North Subsystem overflows  (CSO #004 and #005) discharging the second
         greatest volume.

Table 5.6.a-1 shows the predicted overflow activity in terms of average annual volume
of overflow  per year.  These data were developed  using  the SWMM model  and
continuous simulation. Tables 5.6.a-2 and 5.6.a-3 show the  projected peak overflow
rates for each CSO using the design storms.

Table 5.6.a-1. SWMM Modeling Summary - Annual Overflow Volumes
Drainage Area
Description
WWTF Bypass
North/Lindsey/Downtown
South
CSO
ID Number
002
004 and 005
003, 006, 007
and 008
Estimated
Annual
CSO Discharge
(m3/yr.)(1)
110,900
54,000
21,200
Estimated
Annual
CSO Discharge
(MG/yr.)(1)
29.3
14.4
5.6
                                TOTAL         186,100           49.3
  (1)Using long-term simulation based upon 21 years of hourly rainfall data form the
    NWS station in Portland, ME for the years from 1971 to 1991.

Table 5.6.a-2. SWMM Modeling Summary - Projected Peak Flows
Estimated Peak Flow (m3/d)
CSO
ID Number
002
003
004
005
006
007
008
0.5 Year
Storm(1)
23,850
5,300
0.0
47,700
380
6,440
1,900
1 year
Storm(1)
31,800
5,300
10,100
53,800
6,440
6,440
3,400
2 year
Storm(1)
33,000
5,300
23,100
54,900
14,000
6,440
5,300
2 year
Storm(1)
34,100
5,300
37,100
56,000
21,200
6,440
7,600
           (1)
             Using 2 h duration.
                                     164

-------
Table 5.6.a-3. SWMM Modeling Summary - Projected Peak Flows
Estimated Peak Flow (MGD)
CSO
ID Number
002
003
004
005
006
007
008
0.5 Year
Storm(1)
6.3
1.4
0.0
12.6
0.1
1.7
0.5
1 year
Storm(1)
8.4
1.4
2.9
14.2
1.7
1.7
0.9
2 year
Storm(1)
8.7
1.4
6.1
14.5
3.7
1.7
1.4
2 year
Storm(1)
9.0
1.4
9.8
14.8
5.6
1.7
2.0
           (1)Using2 h duration.
Design Considerations
The following is a list of key design considerations for the Rockland WWTF retrofit:
      •  Impacts of High Strength Wastes
      •  Hydraulic Loadings
      •  Treatment Objectives
      •  Operational Requirements
      •  Chemical Storage and Feed Requirements
      •  Sludge Handling Requirements
      •  Operations and Maintenance Requirements
      •  Construction Sequencing and Site Constraints
Each of these considerations is described in this section.
Impacts of High-Strength Wastes. One of the nine minimum CSO controls is the review
and modification of pretreatment programs to assure that CSO impacts are minimized.
The elimination of the  industrial  wastewater connection from FMC  Food Ingredients
Division to the  city of Rockland CCS, and construction of a separate  force main to the
WWTF will satisfy this objective.
The industrial wastewater design loading to the Rockland WWTF is approximately 2,990
kg (6,583 Ibs) per day.  This is approximately 67% of the total organic loading of 4480
kg (9,871 Ibs) that the WWTF  receives daily. Following the implementation of Phase II
improvements, this industrial wastewater can be directly discharged to three locations at
the Rockland WWTF:
      •  Existing Primary Clarifier Influent at New Flow Distribution Structure No. 1
      •  Existing Aeration Tank Inlet Channel
      •  Equalization Tank
                                      165

-------
The industrial wastewater enters the WWTF via a 20 cm (8 in.) diameter force main
connected to a new pre-cast concrete valve vault (SS #2).  SS #2 contains valves to
direct flow  to the three discharge locations.   A 20 cm (8 in.)  magnetic flow meter
measures flow.  If the flow is directed to the equalization tank, then it is mixed using a
fine bubble diffused air system.  The design airflow is 30 scfm per thousand cubic feet.
The air blower has a variable frequency drive allowing adjustments to the airflow. The
WWTF staff is able to mix the equalization basin at a lower rate, if desired.  A total of
324 membrane diffusers each with a maximum airflow of 3.5 scfm are provided in the
equalization basin.

Two submersible pumps (1  primary and 1 standby) convey industrial wastewater from
the equalization basin to the WWTF.  Each pump has a  rated capacity of 2.3 m3 (600
gal) per min at a total dynamic head of 4.5 m (14 ft).  The pumps are driven with 3.7 kW
(5 HP) motors controlled  with variable frequency drives that allow the WWTF staff to
adjust the  flow of industrial wastewater to the process.   The  discharge from the
submersible pumps can be directed to two locations:

      •  Existing Primary Clarifiers via new Flow Distribution Structure No. 1
      •  Existing Aeration Tank Inlet Channel

A 15 cm (6 in.) diameter magnetic flow meter is provided on the submersible pump
discharge allowing  the rate of industrial wastewater flow to be monitored by the WWTF
staff.

Other high-strength wastes eliminated from the CSO  discharge  with  the Phase  II
Improvements are the sidestream flows from  the sludge processing  system.  The
sidestream  flows represent a relatively small portion of the daily load.  Approximately
127 kg  (279 Ibs) per day  are  discharged  (approximately 3% of the  overall design
organic  loading of 4480  kg (9,871  Ibs)  per  day).   However,  these  loadings are
concentrated within a few  hours during the day, which magnifies their effects.

The  sludge  thickener filtrate   is  produced in  a  3.5-4  h period and   represents
approximately 8% of the hourly  organic loading while the sludge de-watering filtrate is
produced in a 3.5-4 h period and represents approximately 12% of the hourly organic
loading.  Overall, the sidestreams represent approximately 20% of the hourly organic
loading at the WWTF when the units are in service.  The sludge thickening sidestream
was redirected to the existing Aeration Tank Inlet  Channel while the sludge dewatering
sidestream   was redirected  to  the existing Primary Clarifier  influent at  new  Flow
Distribution Structure No. 1 with the Phase II Improvements.

Hydraulic Loadings.  Pollutant and hydraulic loadings were developed for the wet-
weather treatment units at the  Rockland  WWTF during the Facilities Planning and
Design stages.  The primary objective of the wet-weather treatment  units is to remove
floatable and settleable solids to  promote effective disinfection.  The primary objective of
the industrial  wastewater  equalization basin  is to  remove high-strength industrial
wastewater from the CCS to allow full secondary treatment of these wastes at all times.
                                      166

-------
Table 5.6.a-4 shows the design criteria for the wet-weather treatment processes at the
Rockland WWTF.
Table 5.6.a-4. Wet-Weather Treatment Design Criteria
Description and Criteria
A. Flows and Loadings
WWTF Peak Monthly Avg. Flow
WWTF Peak Daily Flow
WWTF Instantaneous Maximum
Total Annual CSO Volume
CSO Annual Volume at WWTF
CSO North/West Subsystem Peak (5 yr., 2 h)
CSO North/West Subsystem Peak (2 yr., 2 h)
Phase II Design Value

10,100m3/d
1 2,500 m3/d
21,600m3/d
186,600m3
188,100m3
1 27,550 m3/d
11 0,900 m3/d

2.90 MOD
3.30 MOD
5.70 MOD
49.3 MG
43.7 MG
33.7 MGD
29.3 MGD
       CSO North/West Subsystem Peak (1 yr. 2 h)

       CSO North/West Subsystem Peak (1/2 yr. 2 h)
   B.  Influent Pumping - Dry Weather
       No. of Pumps (each)
       Type
       Flow
       TDH
       Power (each)
   C.  Screenings Handling System
       Type
       No. of units (each)
       Channel Width
       Bar Screen Clear Spacing
       Bar Size
       Headless, half clogged
   D.  Grit Removal
       Type
       No. of Units (each)
       Diameter
       Hopper Diameter
       Type of pump
       No. of pumps (ea.)
       Flow
       TDH
       Horsepower (each)
       Type of De-watering
       Pressure
       Screw Diameter
   E.  Wet Weather Pumping
       No. of pumps (each)

       Type
96,500 m3/d      25.5 MGD

71,500m3/d      18.9 MGD
   Non-clog Submersible
 15m3/min       4,000 gpm
   9m             SOn
  37.3 kW         50 HP
     Mechanical Screen
            1
                  36 in.
                  1.0 in.
               2-1/2x3/8 in.
                  9.0 in.
  91.0cm
  2.5cm
6.5x1.Ocm
  22.9 cm
           Pista
            1
   3.7m           12.0ft
   1.5m           5.0 ft
     Recessed Impeller
            1
 1 m3/min         260 gpm
   9m            30n
            10
         Cyclone
4,922 kg/m2       7.0 psig
 30.5cm          12.0 in.
   Non-Clog Submersible
                                      167

-------
Table 5.6.a-4.  Wet-Weather Treatment Design Criteria, continued
       Description and Criteria
   Phase II Design Value
Flow each
TDH
Power each
Maximum Speed (rpm)
Minimum Efficiency (%)
F. Vortex Separator
No. of Units (each)
Type
Diameter each
Sidewater Depth each
5 year Hydraulic Loading
2 year Hydraulic Loading
1 year Hydraulic Loading
V* year Hydraulic Loading
Influent Pipe Size
Effluent Pipe Size
Underflow Pipe Size
Underflow Rate, Minimum
Underflow Rate, Maximum
30 m3/min
8.5m
74.6 kW





10.7m
2.7m
1,430m3/d/m
1,240m3/d/m
1,080m3/d/m
780 m3/d/m2
107cm
107cm
35.5 cm
6435 m3/d
1 2,870 m3/d
7,800 gpm
28 ft
100 HP
720
76

1
EPA
35 ft
8.75 ft
2 35,000 gpd/ft2
2 30,400 gpd/ft2
2 26,500 gpd/ft2
1 9,600 gpd/ft2
42 in.
42 in.
14 in.
1.7 MOD
3.4 MOD
       No. of Underflow pumps
       Type of pump
       Pump Flow (each)
       TDH
       Motor power each
       Minimum Efficiency (%)
       Maximum Pump Speed (rpm)
   G.  CSO Disinfection And Dechlorination
       No. of Units (each)
       Volume
       Length to Width Ratio
       Chlorine Detention Time - 5 yr.
       Q = 114,685.5 m3/d (30.3 MOD) (min)
       Chlorine Detention Time - 2 yr.
       Q = 98,031.5 m3/d (25.9 MOD) (min)
       Chlorine Detention Time - 1 yr.
       Q = 83,648.5 m3/d (22.1 MOD) (min)
       Chlorine Detention Time - !4 yr.
       Q = 58,667.5 m3/d (15.5 MOD) (min)
       Dechlorination Tank Volume
       Dechlorination Detention Time - 5 yr.
       Q = 114,685.5 m3/d (30.3 MOD) (s)
       No. of NaHSO3 pumps (each)
       Capacity of NaHSO3 Pumps (each)
       No. of Sodium Bisulfite Pumps (each)
       Capacity of NaHSO3 Pump (each)
   Self Priming Non-Clog
4.5 m3/min
  4.3m
  7.5 kW
     1,170 gpm
        14ft
       10 HP
  400m3
 55
860

 1

40:1

 5.0

 5.8

 6.7

 9.2
105,300 gal
  19m'
      5,100 gal
           15
       2+1 standby
 0.4m3/h         105gph
       1+1 standby
 0.1 m3/h         25gph
                                       168

-------
Treatment Objectives. Sampling and monitoring performed during the development of
the CSO Abatement Program indicated that Rockland Harbor does not meet current
water quality criteria for enterococcus bacteria during wet weather. Currently, the State
of Maine allows a maximum geometric mean enterococcus concentration of 14 colonies
per 100 ml and  a daily maximum  enterococcus concentration  of 94 per 100  ml  to
comply with Standards  established for Class SC waters.  The  primary focus of the
phase II Improvements are:

      •  Increasing the quantity  of CSO  conveyed to the WWTF for treatment and
         disinfection.
      •  Segregating high-strength industrial wastewater and WWTF sidestreams from
         the overflow at the influent pump station.
      •  Providing high-rate  primary treatment to  remove floatable and  settleable
         solids from the CSO to facilitate disinfection.

The CSO Abatement Program recommended a plan to treat CSO discharges  to meet
the current maximum daily water quality limits.  The  high-rate primary treatment and
disinfection processes were  selected as the most economical  method to meet the
treatment objectives to meet water quality objectives for this project.

Operational Requirements. The Phase II Improvements to the  Rockland WWTF which
are designed to operate  on a continuous basis include:

      •  Dry Weather Pumping Station
      •  Industrial Wastewater Equalization
      •  Mechanical Screening
      •  Grit Removal
      •  Primary Treatment

The wet-weather pumping system, high-rate primary treatment (vortex separator) high-
rate disinfection, and dechlorination systems are intended to operate in an event-based
mode during periods of excessive WWF.  Under normal dry weather conditions flow
from the north and south CCS enter a new diversion structure (SS #1) upstream of the
influent pumping  station.  SS #1 contains a side overflow weir separating the overflow
from the normal sewage flow through the structure. As sewage flows increase during a
wet-weather event, the new submersible dry  weather  pumps will  be throttled to control
flow to the secondary WWTF. As water levels rise in the dry-weather pump wetwell flow
will automatically be diverted to the overflow  chamber. These overflows are conveyed
to the wetwell  of the wet-weather pumping system.   When levels in the wet-weather
wetwell rise, the wet-weather pumps will automatically start and convey flow to the new
vortex separator via a 1.0 m (42 in.) force main.  Once flow through the vortex separator
has  been established,  a level switch located at  the  inlet  to the high-rate CSO
Disinfection and  Dechlorination Structure will  start  the vortex underflow pumping
system. The vortex underflow pumps are rated to deliver 4.50 m3 (1,170 gal) per min at
                                     169

-------
4.5 m (14ft) total dynamic head.  Each pump has a 7.5 kW(10 HP) electric motor with a
variable frequency drive allowing the WWTF staff to adjust the rate of vortex underflow.

The level switch located at the inlet of  the  CSO Disinfection and  Dechlorination
Structure  will also  start the  chemical  metering   pumps  (sodium  hypochlorite for
disinfection and sodium bisulfite for dechlorination) and the tank mixers.

An ultrasonic flow meter  is provided at the 160 cm  (64 in.) broad crested rectangular
weir at the inlet of the CSO Disinfection and Dechlorination Structure.  The flow meter
records CSO flows,  and provides a 4-20 mA signal to control the speed of the sodium
hypochlorite  and  sodium bisulfite chemical metering  pumps.   Flow pacing  of the
chemical feed rate will improve disinfection and dechlorination efficiency, and  reduce
overall chemical consumption.

At the end of the  wet weather event automatic level controls will stop the submersible
wet-weather pumps.  When the influent flow stops, the level in the vortex separator will
fall as it is drained with the vortex underflow pumping system.  The chemical metering
pumps and tank mixers automatically stop when the level in the CSO Disinfection and
Dechlorination Structure falls below the pre-selected  set point. The WWTF staff would
then drain the CSO Disinfection and  Dechlorination Structure and wash  down the
structures to prepare for the next event.

Chemical Storage and Feed Requirements.  The CSO  Abatement facilities  will utilize
two chemicals:

      •   Liquid Sodium Hypochlorite for disinfection
      •   Liquid Sodium Bisulfite for dechlorination

The liquid sodium hypochlorite system is designed for commercial-grade  liquid sodium
hypochlorite (15% solution) which contains the equivalent of 150 kg of chlorine/m3  (1.25
Ibs Cb/gal) of solution. The metering pumps can deliver a dosage of 25 mg/L at the
peak design flow with one unit out of service.  The bulk storage tank has an effective
working volume of 4 m3 (1,000 gal).  This tank is interconnected with the existing 15 m3
(4,000 gal)  sodium hypochlorite  bulk  storage tanks   used for secondary effluent
disinfection,  giving the WWTF a total of 19 m3 (5,000  gal) of hypochlorite storage.  The
available  chlorine in sodium  hypochlorite  solution declines with age,  the anticipated
sodium
MGD).
sodium hypochlorite storage time will be 52 days at the design flow of 10,980 m3/d (2.9
The  liquid  sodium bisulfite storage system  is designed  for commercial  grade liquid
sodium bisulfite (38% solution) which contains appropriately 500 kg sodium bisulfite/m3
(4.2  Ibs NaHSOs/gal)  of solution.  The  metering  pumps  can deliver sodium  bisulfite
dosages necessary to  dechlorinate effluent with a  chlorine residual  of  5.0 mg/L.
Approximately 1.6 kg of sodium bisulfite will  be used per  kg of chlorine residual.  The
bulk storage tank has an effective working volume of 4 m3 (1,000 gal).   This tank is
interconnected  with the  existing  8 m3 (2,000 gal) sodium bisulfite tank used for
                                      170

-------
secondary effluent dechlorination, giving the WWTF a total of 12 rr? (3,000 gal) of
sodium bisulfite storage.

Sludge Handling Requirements. Sludge handling requirements increase during periods
of WWF. The most pronounced increase occurs during the initial period of WWF (first
flush).

A solids balance was performed to estimate the quantity of sludge produced  during a
one yr frequency, two hr duration design storm.  The daily sludge quantities would
increase as follows:

      •  Estimated Sludge Quantity                  2,870 kg (6,326 Ibs) per d
      •  One-Year Storm Estimated Sludge Quantity   3,640 kg (8,029 Ibs) per d

The sludge quantity is estimated to increase by approximately 27% during a one-year
storm event. The overall increase can be assimilated if sludge pumping and processing
schedules are adjusted during the remainder of the week.  The following is a summary
of weekly  sludge  quantities  during  a  period with a one-year  duration, two-hour
frequency design storm:

      •  Initial Estimated Sludge Quantity (kg/week)                     20,090
      •  Projected Sludge with One-Year Design Storm (kg/week)        20,860

The estimated weekly sludge quantity would increase by approximately 4% during a
week with a one-year storm event.

Operation and Maintenance  Requirements.  The implementation of the Phase  II
Improvements will result in increased O&M costs.  The following list summarizes the
areas of expected O&M cost increases at the WWTF:

      •  Labor  to  perform  sampling,  laboratory analyses, reporting,  wet-weather
         process adjustments, and clean-up following an event.
      •  Power for in-plant pumping, sludge processing and chemical feed systems.
      •  Chemicals for disinfection, dechlorination,  and polymer for additional sludge
         quantities.
      •  Sludge disposal tipping and transport fees.

Table 5.6.a-5 summarizes the expected increase in O&M costs at the WWTF resulting
from the Phase II Improvements.
                                     171

-------
Table 5.6.a-5. Estimated O&M Costs Phase II CSO Improvements

                                                     Estimated Annual
      	Description	Cost	
      Sampling                                              $600
      Laboratory Analyses                                    2,500
      Report Preparation                                     1,200
      Process Control Adjustments                            1,000
      Clean-Up of Wet-Weather Treatment Structures            4,500
      Power                                                  700
      Sodium Hypochlorite                                    3,100
      Sodium Bisulfite                                        1,600
      Polymer for Sludge De-watering                            300
      Sludge Disposal Transport and Tipping Fees	4,000	
                               Total Estimated Cost         $19,500


Construction Sequencing and Site Constraints. The existing  Rockland WWTF site is
crowded with little room for new facilities in the area where the  influent sewers enter the
WWTF.  Therefore, it was necessary to construct some of the CSO Abatement Facilities
in  locations previously used for  other purposes.  The following is a listing of special
sequencing and  site  considerations related to the CSO Abatement Program  at the
WWTF:

      •   A storage garage was demolished to facilitate construction of the vortex
         separator.
      •   There was  limited space available for  wet-weather pumping.  The existing
         preliminary  treatment  building  (which  contained the influent pumps)  was
         converted into  a submersible pumping  station for DWF and WWF. The dry
         weather pumps are installed in the existing wetwell.  The wet-weather pumps
         are installed in  the existing pump room that is converted into a wetwell for the
         submersible wet-weather pumps.  The modifications will  require temporary
         bypass pumping of flow into the WWTF during construction.
      •   New junction structures are required on the existing outfall pipe (SS #4 and
         SS #5) to increase the hydraulic capacity.  An emergency relief discharging to
         Lermond Cove (adjacent to the WWTF site) will be preserved.

A substantial completion time of 510 calendar days was specified for the Phase II
Improvements.
                                     172

-------
Project Costs

The bids for the Phase II Improvements were opened on July 8, 1998.  Costs for the
CSO related components  of the project are listed  in  Table 5.6.a-6.   Costs include
construction  requirements  and project  services.   Project services  include design
engineering,  construction engineering, legal,  and fiscal expenses.   Project services
costs  do not include contingencies,  right-of-way expenses, or other miscellaneous
items.

Table 5.6.a-6. Project Costs -Phase II CSO Improvements

      Description                                               Cost1
      General Conditions                                     $175,000
      Site Work and Yard Piping                                375,000
      Consolidation Conduit                                   140,000
      Special Structure No.  1                                     90,000
      Dry Weather/Wet Weather Pumping Station Renovation     180,000
      Headworks/CSO  Disinfection and Dechlorination           1,080,000
      Structure
     Vortex Separator                                        240,000
      Special Structure No. 2                                     40,000
      Equalization Tank Renovation                               58,000
      Industrial  Wastewater Force Main                            22,000
      Flow Distribution Structure No. 1                             60,000
      Flow Distribution Structure No. 2                             65,000
      Special Structure No. 4                                     60,000
      Special Structure No. 5                                     55,000
     Yard Electrical Ductbanks and Lighting                       90,000
      Instrumentation and SCADA System	70,000
                                  Total Construction Cost    2,800,000
    	Project Services2     600,000
                                        Total Project Cost    3,400,000
     1 Based on July 8,  1998 Bid Prices
     Includes  Design Engineering, Construction Engineering, Legal, and Fiscal
      Expenses. Does  not include contingencies and right-of-way costs.

Conclusions

The retrofit of an unused aeration tank and the construction of  a separate force main
from the FMC Corporation  Food Ingredients Division facility to  the WWTF provides a
cost-effective alternative for the elimination of a  established portion of high-strength
wastes from CSO discharges at the Rockland WWTF. This retrofit, combined with other
WWTF improvements directed  at CSO abatement, will  direct high-strength waste from
the FMC  facility to  the dry-weather headworks,  eliminating the potential  for CSO
discharge of industrial wastewater. This retrofit was part of an overall CSO Abatement
Program necessary to address system-wide CSO discharges.
                                      173

-------
5.7   Bringing Outdated/Abandoned POTWs Back Online as Wet-Weather Flow
      Treatment Facilities

The wastewater treatment facilities remaining following upgrades or construction of new
POTWs are often  abandoned or demolished. These outdated or abandoned facilities
could possibly serve a useful function as part of the new treatment facilities depending
on their condition and  location with respect to the new  treatment facilities and the
combined  sewer system.  These facilities may be  utilized as wet-weather storage or
flow equalization facilities.


5.7.a. Auburn WWTF, Auburn, New York

Background

The  city of Auburn Water Pollution Control  Plant was initially constructed during the
1930s and provided primary treatment.  In 1970, the original primary treatment facilities
were upgraded to provide secondary treatment. Since that time, several improvements
have been completed.

The  WWTF  includes preliminary treatment (screening  and grit removal),  primary
sedimentation, trickling filtration, final sedimentation, and chlorination. Sludge treatment
facilities include gravity sludge thickeners, belt filter press for sludge dewatering, and
incinerators for sludge disposal.  The ash from the incineration process is landfilled.

The Auburn WWTF services the city of Auburn and the surrounding Towns of Aurelius,
Fleming,  Sennett, and Owasco.   Portions  of the existing sewer system consist of
combined  sewers  that carry both sanitary sewage flow and storm sewer flow in the
same pipe.

In December  1989, the city of Auburn entered into a consent order agreement with the
New York State Department of Environmental Conservation (NYSDEC).  The  consent
order required the city of Auburn to complete an  Infiltration/Inflow Analysis,  a CSO
Analysis, BMP Program, and a Comprehensive Performance Evaluation (CPE) for the
city of Auburn Water Pollution Control Plant.

Phase I  improvements to the  city  of Auburn  WWTF  were  performed  between
1993-1995, as summarized below:

     1.  New  preliminary  treatment facilities consisting  of mechanical bar screens
        housed in a new influent  building, aerated grit channels, and new influent flow
        metering.

     2.  New  primary treatment facilities,  including new  primary settling tanks and
        sludge  handling  pumps  and  rehabilitation of the existing- settled  sewage
        pumps and controls.
                                     174

-------
     3.  A new ORF utilizing the abandoned primary  settling tanks, with chlorination
        (using sodium hypochlorite) and dechlorination (using sodium bisulfite) facilities
        and flow metering equipment for the ORF overflow.

     4.  Trickling filter rehabilitation, including replacement of rotary distribution arms,
        rehabilitation of existing recirculation pumps and controls, and installation of
        recirculation flow metering equipment.

     5.  New septage  receiving facilities, including an enclosed  septage receiving
        station, mechanical septage screening, underground septage storage tank, and
        septage handling equipment.

     6.  Odor control facilities to treat ventilated  air from influent building,  screenings
        and grit loading area, septage receiving station and  septage storage tank, and
        belt press room of the existing administration building.

Construction costs for  the Phase I  improvements  were estimated to be $9,600,000.
Figure 5.7.a-1  shows the WWTF flow schematic prior to construction of the  Phase I
improvements

The following summarizes the retrofit of the abandoned primary treatment tanks for use
as overflow retention facilities for the treatment of WWF in excess of the treatment plant
capacity.

Previous Studies

As a  requirement of the consent order with New York State, the city of Auburn has
completed studies on I/I and CSO abatement.

Infiltration and Inflow Study.   In January 1991, the city of Auburn completed  an  I/I study
of the wastewater collection system.   The  purpose of this study was to develop  a
recommended plan of  corrective action  to  eliminate  excessive  I/I within the  city's
sanitary  sewer system.   The  study  recommended  the removal of sewer  system
conveyance restrictions and the elimination of direct inflow sources. The results of the
I/I study indicated that infiltration was not excessive and that it was more cost effective
to transport and treat, than to eliminate. The study also suggested that approximately
31,420 m3/d (8.3 MGD) of peak inflow could cost effectively be removed.

CSO  Abatement.   In July  1993, the city of Auburn  completed  a CSO  abatement
study/comprehensive water collection system plan. The CSO abatement study included
the evaluation of alternatives for the elimination, storage, and/or treatment of CSOs that
discharge to the Owasco Outlet within the city of Auburn. The CSO study identified the
alternative  of  centralized  high  rate  treatment  as  the  most   cost  effective  and
environmentally sound  alternative for  the city of  Auburn's collection system.  This
alternative includes the construction of a new interceptor to increase the capacity of the
existing North  Owasco  and  South Interceptors,  and the construction of a high  rate
treatment facility  and diversion structure at the existing  WWTF.  The  CSO study
recommended the implementation of the North Interceptor improvements as described
                                      175

-------
5.7.a-1
                                        176

-------
in the North Interceptor Replacement Sewer Basis of Design Report dated April 1993.
Also included in the CSO study was an estimate for the future (year 2010) total peak
flow to the WWTF of 191,000 m3/d (50.46 MGD). Of this total flow, 44,970 m3/d (11.88
MGD) was estimated for total base flow plus infiltration.

Comprehensive Performance Evaluation.  In June 1992, an interim  CPE  report for the
city of Auburn  was completed.  This interim report developed the  conceptual WWTF
improvements  required to meet the anticipated reclassification of Owasco Outlet and
revised effluent limits.

Project Approach

The CSO report  recommended the construction  of peak flow storage facility on the
North Interceptor and the construction of parallel  interceptors to convey peak flows to
the WWTF.  This  report had also recommended the construction of new overflow
facilities,  such  as a swirl concentrator, to treat flows above 78,400 nf/d (20.7  MGD).
These facilities were completed and operation began in July 1996.

In  lieu of the  swirl concentrators to  treat peak flows at the WWTF, an  ORF was
evaluated.  The  ORF would treat flows  above  the 95,000  m3/d  (25 MGD) WWTF
capacity and be constructed from  the abandoned  primary settling tanks, which became
available  after the treatment plant was upgraded. The reuse of this tankage represented
a significant cost saving to the city of Auburn


Overflow  Retention Facilities

The abandoned primary settling tanks were rehabilitated to function as an ORF.  The
basis of design for the ORF is summarized in Table 5.7.a-1 as described below.

Table 5.7.a-1.  Basis of Design of the Overflow Retention Facilities

 Overflow Retention Tanks	Peak hourly flow 96,500  m3/d (25.5 MGD)
 Number of Units                           5 (abandoned primary settling tanks)
 Total Surface Area                                         782 m2 (8,416 ft2)
 Maximum SOR                                    123 m3/d/m2 (3,025 gpd/ft2)
 Equipment	
 Overflow Return Pumps
 Number of Units                                                         4
 Type                                                           Centrifugal
 Return Duration                                                      24 h
 Overflow Influent Metering	
 Type                                                       Open Channel
 Flow Element                                                        Weir
 Flow Meter                                                       Ultrasonic
 Range                                          0—11,355 m3/d  ( 0—3 MGD)
                                     177

-------
Table 5.7.a-1. Basis of Design of the Overflow Retention Facilities, continued
  High-Rate Disinfection
  Number of Units
  Contact Tank Volume
  Detention Time
  Tank Dimensions (existing)
  (Width x Length x Side Water Depth)
  Mixing Energy
  Chlohnation Chemical Feed Equipment
1 (abandoned primary settling tank #6)
                 318m3 (11,220 ft3)
       4.7 min at peak hourly flowrate
              5.9m x 20.9m x 2.6 m
             (19.5 ft x 68.5 ft x 8.4 ft)
                   22.4 kW (30 HP)
  Type
  Dosage
  Dechlorination
          Liquid Sodium Hypochlorite
                            25 mg/l
  Detention Time
  Contact Tank Volume
  Tank Dimensions
  (Width x Length x Side Water Depth)
  Equipment
  Dechlorination Chemical Feed
  Type
  Dosage per mg/l of chlorine residual
  Mixers
  Number of Units
  Type
  Tank Flushing System	
             12 s at peak hourly flow
                     13m3 (458 ft3)
               1.5 mx 3.6 mx2.4m
              (5 ft x 11.85 ft x 7.74 ft)
              Liquid Sodium Bisulfite
                           1.5 mg/l

                         1 per tank
                    Constant Speed
  Source
  Overflow Effluent Metering
             Chlorinated Plant Effluent
  Type
  Flow Element
  Flow Element
  Flow Meter
  Range
                       Open Channel
                       Open Channel
                               Weir
                           Ultrasonic
           0—113,550 m3/d (30 MGD)
Influent flows in excess of the 95,000 m3/d (25 MGD) WWTF capacity are directed to
the retention facility  through  a  controlled  diversion structure.  The maximum  SOR
provided  by using five of the six abandoned primary  settling tanks as an overflow
retention  tank(s),  is 123 m3/d/m2 (3,025 gal/d/fr) at the estimated  peak hourly flow of
96,500 m3/d (25.5 MGD).  Overflow events producing less than approximately 4,100 m3
(1.083 MG), which includes volume disinfection contact chamber, are retained  and
returned  to  the  treatment process when  influent flows subside.   Overflow  events
producing in excess of the available storage volume are disinfected using chlorine and
                                     178

-------
dechlorinated  before  being  discharged  to  the  Owasco Lake Outlet.  The retained
volume is returned to the aerated grit chambers utilizing four overflow return pumps.
Figure 5.7.a-2 shows the flow schematic of the Auburn WWTF following construction of
the ORF.

The overflow retention facilities include the following:

Modifications to the Abandoned Primary Settling Tanks. The primary sludge longitudinal
and cross collectors were removed to use the abandoned primary settling tanks as an
ORF.  Deteriorated concrete surfaces were repaired and weirs were installed to allow
flow  between  tanks.   A high  volume tank flushing system was provided  that uses
chlorinated plant effluent to wash the tank walls and bottom following an overflow event.

Overflow Return. The old primary sludge pump was replaced with overflow return
pumps. The new pumps are suitable for pumping grit and primary sludge.  In order to
prevent odors, the entire overflow retention tank volume is  returned  to the treatment
process within 24 h after the occurrence of an overflow event provided influent flow
rates have subsided.

Disinfection.  Overflow events producing stormwater in excess of the available storage
volume are disinfected using a high-rate  disinfection system.  The system consists of  a
contact tank sized for a 4 min detention time at peak hourly flows of 96,520 m3/d (25.5
MGD) and requires the use of mechanical mixers to induce energy to  provide effective
disinfection.  The mixing velocity gradient and chlorine dosage were calculated  on  the
basis of estimated influent and effluent fecal coliform levels for CSOs, detention time,
concentration of TKN, BOD, and an estimated value for pH.  The high-rate disinfection
system was sized  by estimating disinfectant  dosage and  mixing energy required to
assure the  inactivation  of total  coliform.   Based  on the optimum  velocity gradient
calculated, horsepower requirements for  mixers were determined.  From this analysis,  it
was determined that an optimum chlorine dosage of 25 mg/l and approximately 22.4 kW
(30 HP) of total mixing energy are required to insure proper disinfection.

The  abandoned Primary Settling Tank #6 was modified  for use  as the  high-rate
disinfection  contact tank.  The chlorination facilities are housed in  a new  structure
constructed over a portion of the new overflow  retention tank.  This structure houses the
chlorination and dechlorination equipment. The chlorination equipment was designed to
be automatically flow-paced according to the overflow flow rate.

Dechlorination.  Dechlorination facilities  were installed to reduce the  chlorine residual
before discharge to  the Owasco Outlet.   The  dechlorination facilities  are metered
automatically as a function  of overflow flow  rate and will  be designed to  minimize
chlorine residual. The dechlorination tank  contains a high-rate mixer designed for  a
detention  time of 30 s at peak hourly flows of 96,500 m3/d (25.5 MGD). This  tank was
constructed within a portion of abandoned Primary Settling Tank #6.

Overflow Metering. Overflows discharged to the Owasco Lake Outlet are metered after
being discharged from the dechlorination tank through an open channel-type metering
device located in a separate concrete structure.
                                      179

-------
5.7.a-2.                       and
                                180

-------
Stormwater Management Model Results

The  EPA SWMM Version 4 was used to predict the annual capture of WWF  that
previously discharged at the WWTF and is now treated through the ORF.  A simplified
model of the WWTF drainage area was used to determine the impact of pre- and post-
retrofit.  This analysis is based on 21  years of hourly rainfall from  the NWS Station at
Hancock Field in Syracuse,  NY.  Table 5.7.a-2 demonstrates that  use of the ORF
resulted  in a considerable annual capture of WWF that was previously discharged
without treatment.

Table 5.7.a-2. Average Annual Overflow Volume from the Pre- and Post-Retrofit
      Facilities (1971 -1991)

                                   Average Annual      Average Annual
                                   Overflow Volume      Overflow Volume

     	Condition	(^)	(MG)
      Pre-Retrofit WWTF Overflow        29,000                7.65

      Post-Retrofit Overflow(1)             7,000                1.85
      Percent Capture                               76%
     (1)0verflows receive preliminary treatment and disinfection prior to discharge

The ORF annually captures 22,000 m3 (5.8 MG) of wet-weather overflow at the WWTF
that is returned to the WWTF influent for treatment following storm events.


Design Considerations

The retrofitting of the abandoned primary settling tank to provide enhanced wet-weather
treatment through storage and  disinfection required careful consideration of several
factors.  The following is a list of key design considerations for this retrofit:

      •  Pollutant and  Hydraulic Loadings

      •  Treatment Objectives

      •  Operational Requirements

      •  Chemical Storage and Feed  Requirements

      •  Sludge Handling Requirements

      •  Operations and Maintenance Requirements

Each of these considerations will be described in this section.

Pollutant and Hydraulic Loadings. Pollutant and hydraulic loadings were developed for
the ORF during the  Facilities Planning stage.  These values were refined and updated
as the project progressed through the preliminary engineering and detailed design
                                     181

-------
stages.   The primary objective of the ORF is to  provide preliminary treatment and
disinfection of WWF beyond the 95,000 m3/d (25 MGD) WWTF capacity.  The basis of
design for the ORF is shown on Table 5.7.a-1.

Treatment Objectives.  The rationale for this retrofit was meeting the revised effluent
limits from the WWTF in the Owasco Lake outlet.  Preliminary treatment and high-rate
disinfection were selected  as the most  economical method to meet the treatment
objectives for this project.

Operational Requirements.  The ORF are designed to operate on a continuous basis
include:

      •   Preliminary Treatment

      •   High-Rate Disinfection

Each of the five ORF tanks fill  uniformly during an overflow event.  After the  wet-
weather  flows  subside, the  flow  retained in  the  ORF  is  pumped  to  the  WWTF
headworks for treatment. WWTF staff manually controls this operation. The tanks are
flushed by a manually controlled system using chlorinated plant effluent.

Chemical Storage and Feed Requirements.  The ORF utilize two chemicals:

      •   Liquid Sodium Hypochlorite for disinfection

      •   Liquid Sodium Bisulfite for dechlorination

The  liquid sodium hypochlorite system design is  based  on  commercial-grade liquid
sodium  hypochlorite  (15%  solution)  which contains the  equivalent of  160 kg  of
chlorine/m3 (1.25 Ibs Cb/gal) of solution. The metering pumps are designed to deliver a
dosage  of 25  mg/L at the peak design flow with one unit out of service.  The liquid
sodium hypochlorite is stored in a bulk storage tank with an effective working volume of
10 m3 (2,500 gal).  The available chlorine in sodium hypochlorite solution declines with
age.   The anticipated  sodium hypochlorite storage  time  will be 51  days initially,
decreasing to 25 days at the future design flow.

The  liquid sodium bisulfite storage system design is based on commercial grade liquid
sodium bisulfite (38% solution) that contains appropriately 503 kg of sodium bisulfite/m3
(4.2  Ibs  NaHSOs/gal) of solution. The metering pumps are designed to deliver sodium
bisulfite dosages necessary to dechlorinate effluent with 1.5 mg/L of sodium bisulfite per
mg/L of chlorine residual. Approximately  1.6 kg of sodium bisulfite will be used  per kg
of chlorine residual.  A  bulk storage tank with an effective working volume of  3.8 rrr
(1,000 gal) is provided.

The  sodium hypochlorite and  sodium bisulfite  metering pumps are flow paced.  An
ultrasonic type  level  sensor is used for effluent flow measurement.  A broad-crested
rectangular weir 3 m (10 ft) in length is provided as the primary measuring device.
                                      182

-------
Sludge Handling Requirements. Accumulated sludge will be handled through the pump
back of retained WWF and the associated tank flushing following tank draining.
Operation and Maintenance Requirements.  The implementation of the ORF will result
in  increased O&M costs.  The following list summarizes the  areas of expected  O&M
cost increases at the WWTF:
   •  Chemicals for disinfection and dechlorination
   •  Power for tank draining and flushing, and chemical feed systems.
The estimated annual O&M costs are provided in Table 5.7.a-3.
Table 5.7.a-3. ORF O&M Costs
                  Annual Chemical Costs        Cost
                 Sodium Hypochlorite                 $7,000
                 Sodium Bisulfite                     $3,000
                 O&M                              $8,000
                 Total                             $18,000
Project Costs
Project costs for the capital costs for the ORF are provided in Table 5.7.a-4.
Table 5.7.a-4. ORF Chlorination/dechlorination Capital Costs
                            Chlorination                 Cost
               Chemical Feed Pumps and Controls         $20,000
               Storage Facilities                         $14,000
               Installation                               $21,000
                   Subtotal                             $55,000
               Dechlorination
               Chemical Feed Pumps and Controls         $20,000
               Storage Facilities                         $16,000
               Residual Analyzer System                 $10,000
               Installation                               $19,000
                  Subtotal                              $65,000
                  Total Cost                           $120,000
                                     183

-------
The conventional treatment alternative for this retrofit would be to construct wet-weather
storage facilities to equalize WWF for ultimate treatment at the WWTF.  This alternative
would require the construction of a 4,100 m3 (1.083 MG) ORF.   The estimated cost to
construct this facility based upon  EPA  cost guidelines for CSO control technologies
(EPA 1992) is $3.816 million derived from the following equation:


            Cost  = 3.577V0812   (1)
            where,
            Cost  = Average construction cost, millions of dollars
            V     = Storage volume, MG

The ORF retrofit is a simple solution, requiring  minimal O&M cost, resulting  in a total
cost that is  a fraction of the cost of constructing new wet-weather storage facilities. The
ORF retrofit provides an average annual  76% wet-weather volume capture. The cost for
this retrofit is $120,000 dollars compared  to  the  $3,816,000  cost for  conventional
storage facilities.

Conclusions

Retrofitting  the abandoned primary treatment tanks for use as ORF was a success, in
large part due to the availability and condition of the existing facilities. The abandoned
primary tanks were in  good condition and required only minor modifications for use as
an ORF.  The  ORF retrofit is a cost-effective alternative  in cases where abandoned
primary tanks are available in good condition for retrofitting.  This type of retrofit may be
a less cost-effective solution for cases in where the existing tanks are in poor condition
requiring a substantial cost to upgrade the tanks to serve a functional role as an ORF.
                                      184

-------
                           Section 6 - References

REFERENCES

Alder, J. (1981) "Automatic Device for Emptying a Detention Pond At a Constant Low
Flow Rate and Velocity," International Symposium on Urban Hydrology, Hydraulics, and
Sediment Control, University of Kentucky, Lexington, July 1981.

Beale, D.C. (1992) Recent Developments in the Control of Urban Runoff. J.IWEM, 6,
141-150.

Bennett, D.B and J.P.  Heaney (1991) Retrofitting for Watershed Drainage.   Water
Environments, Technology, 3, 9,63-68.

Berry, B.   (1994)  Clarifier  Changes  Minimize I/I  Effects.  Water Environment  &
Technology, 6, 8, 22-24.

Camp Dresser & McKee (COM) (1998) DRAFT ACTIFLO Evaluation Report.

Camp Dresser & McKee (COM) (1998)  DRAFT Basin Cleaning Test  Repot, Spring
Creek AWPCP Upgrade, Capital Project No. WP-225, Volume I and II.

Chandler,  R.W.  and W.R. Lewis (1977)  Control of Sewer Overflows  by Polymer
Injection.  EPA-600/2-77-189, U.S. Environmental Protection Agency, Cincinnati, OH.

Chantrill, S.G. (1990) Real Time Control of Combined Sewers: A U.S. View.  In Urban
Stormwater Quality Enhancement-Source Control, Retrofitting, and Combined Sewer
Technology. American Society of Civil Engineering, NY, NY, pp. 390-401.

CG&S-  CH2M  GORE  & STORRIE  LIMITED.    (1997)  Main  Treatment  Plant
Environmental Assessment, Draft Document. The Municipality of Toronto Metropolitan.
CH2M GORE & STORRIE  LIMITED, Waterloo, Ontario.

Delattre, J.M. (1990) Real-Time Control of Combined Sewer Systems form The User's
Perspective.  In  Urban  Stormwater Quality Enhancement-Source  Control, Retrofitting,
and Combined Sewer Technology.  American Society of Civil Engineering, NY, NY, pp.
266-289.

England,  G.  (1995) Stormwater retrofitting techniques for water quality benefits.  In
Proceedings of the 22 Annual Conference Integrated Water Resources Planning 21st
Century. American Society of Civil Engineering, NY, NY, pp. 25-28.

Field, R., and T.P. O'Connor (1997) Optimization of CSO Storage  and Treatment
Systems.  Journal of Environmental Engineering, 123, 3, 269-274.
                                    185

-------
Gousailles, M.  (1995) The Acheres Waste Water Plant A New Master Plan by the Year
2001."  In 7th International Association of Water Quality International Conference on
Design and Operation of Large Wastewater Treatment Plants.

Guibelin,  E., F.  Delsalle and P. Binot (1994) The ACTIFLO Process: A Highly Compact
and Efficient Process to Prevent Water Pollution by Stormwater Flows.

Harleman, D.R.F.,  S.P.  Morrissey  and  S.  Murcott   (1991)  The  Case  for Using
Chemically Enhanced Primary Treatment in a New Cleanup Plan for Boston Harbor.
Civil Engineering Practice, 6, 1, 69-84.

Harper, H.H.  and J.L. Herr  (1992)  Stormwater Treatment Using Alum.  Public Works,
123,10,47-49,89-90.

Hayward, K.  (1996) Capital Investment.  Water Quality International, pp. 22-25.

Heinke, G.W.,  A.J.  Tay and  M.A.  Qazi (1980)  Effects  of Chemical Addition  on the
Performance of Settling  Tanks.  Journal Water Pollution  Control Federation, 52,  9,
2946.

House, L.B.,  R.J. Waschbusch, and  P.E. Hughes. (1993)  Water Quality of an Urban
Wet Detention  Pond  in Madison, Wisconsin,  1987-1988. U.S. Geological  Survey,  in
cooperation with the Wisconsin Department of Natural Resources. USGS  Open File
Report, pp. 93-172.

Huber, W.C., and R.E. Dickinson. (1988) "Storm Water Management Model, Version 4:
User's  Manual," EPA-600/3-88-001a, U.S. Environmental Protection Agency,  Athens,
GA.

Huber, W.C., J.P. Heaney, M.A. Medina, W.A. Peitz, H. Sheikh, and G.F. Smith. (1975)
"Storm Water Management Model User's Manual-Version II," EPA-670/2-75-017,  U.S.
Environmental  Protection Agency, Cincinnati, OH.

Kruger Inc., (1998) ACTIFLO® Pilot Study for the Metropolitan Sewer District At the Mill
Creek Wastewater Treatment Plant, Cincinnati, OH.

Le Poder, N. and P. Binot (circa 1995) Treatment of Combined Sewer Overflow (CSO)
with High Speed Microsand  Settling.  O.T.V. -  "L'Aqarene" - 1 place  Montgolfier -
94417 Saint Maurice Cedex - France.

Le Poder, N. (1997)  High Rate Technology in Mexico. Presented at the £Sh SEED
Seminar,  Reuse of Water  and Solids: The  Squaring of the Water Cycle, Chicago,
October 19, 1997., O.T.V. - "L'Aqarene" - 1 place Montgolfier - 94417 Saint  Maurice
Cedex - France.
                                     186

-------
Martin, S.R.  (1988) An urban stormwater retrofit assessment and planning method.  In
Proceedings of  the  Symposium  on Coastal Water Resources,  American Water
Resource Association, pp. 369-382.

Matthews, R.R.,  E.  Watt,  J.  Marsalek, A.A. Crowder, and B.C. Anderson (1997)
Extending retention times in a stormwater pond with retrofitted baffles.  Water Quality
Research Journal Canada, 32, 1, 73-87.

MCCuen, R.H., S.G. Walesh, and W.J. Rawis. Control of Urban  Stormwater Runoff by
Detention and Retention,  Misc. Public. No. 1428, U.S. Dept. of Agric. 1984.

Metcalf and  Eddy,  Inc.,  University of Florida, and Water  Resources Engineers, Inc.
1971. "Storm Water Management Model," Vols. I, II, III, and  IV; EPA Reports 11024
DOC 07/71,  08/71, 09/71, 10/71; U.S. Environmental Protection  Agency, Washington,
D.C.

Myers, J., and J.  Wickham (1994) Infiltration  Problems Solved Through  Treatment
Upgrade. Water Environment & Technology, 6, 6,15.

Nenov, V.  (1995) SS/BOD  Removal Efficiency and Cost Comparison of Chemical and
Biological Wastewater Treatment.  Water Science Technology, 7, 2, 207-214.

Nix, S.J.  and S.R.  Durrans.  (1996) "Off-line Stormwater Detention  Systems." Water
Resources Bulletin, 32, 6, 1329-1340.

Palhegyi, G., G. Driscoll and P. Mangarella (1991) Evaluation of the feasibility to retrofit
stormwater control facilities  to  improve  pollutant  removal  performance. In Water
Resources Planning and  Management and Urban Water Resources. American Society
of Civil Engineering, NY, NY, pp. 787-791.

Pisano, W.C. (1990) Inlet Control Concepts on Catchbasins  U.S.  Experience. In Urban
Stormwater Quality Enhancement-Source Control,  Retrofitting, and  Combined Sewer
Technology.  American Society of Civil Engineering, NY, NY,  pp. 268-296.

Pisano, W.C.  (1989)  Feasibility  Study for  Stormwater Treatment at Belmont Street
Drain,  Worcester, Ma.   Environmental Design  & Planning,  Belmont MA.  City  of
Worcester, MA.

Pisano, W.C. (1982) An overview of Four Inlet Control Studies for Mitigating Basement
Flooding  in Cleveland and Chicago Areas.  American Society of Civil  Engineering,
Cleveland, OH.

Price, F.A., and D.R. Yonge  (1995) Enhancing Contaminant Removal  in stormwater
Detention Basins by Coagulation. Transportation Research Record, 1483, 105-111.
                                     187

-------
Pitt, R. (1979) Demonstration of Nonpoint Pollution Abatement Through Improved Street
Cleaning  Practices,  EPA-600/2-79-161,  U.S.   Environmental  Protection  Agency,
Cincinnati, OH, pp. 270.

Pitt, R. (1985) Characterizing and Controlling Urban Runoff Through Street and Sewage
Cleaning,  U.S. Environmental  Protection  Agency,  Storm and Combined  Sewer
Program, Risk Reduction  Engineering Laboratory, EPA/600/S2-85/038,  NTIS PB  85-
186500, Cincinnati, OH, pp. 467.

Pitt, R. (1987) Small Storm Urban Flow and Particulate Washoff Contributions to Outfall
Discharges.  Ph.D. dissertation. Dept. of Civil and  Environmental Engineering,  the
University of  Wisconsin,  Madison,  Wisconsin.  Listed  in   dissertation   Abstracts
International, University Microfilms International, Vol. 49, No. 1, 1988, November, 1987.

Pitt, R. (1998) Personnel Discussions.

Pitt, R. and P. Bissonnette. (1984) Bellevue  Urban Runoff Program, Summary Report,
U.S. Environmental  Protection Agency  and  the  Storm  and  Surface  Water  Utility,
Bellevue, Washington.

Pitt, R. and J. Voorhees. (1995) "Source Loading and Management Model (SLAMM),"
Seminar Publication: National  Conference on  Urban Runoff Management: Enhancing
Urban Watershed Management at the Local,  County, and State Levels. March 30 -April
2, 1993. Center for Environmental Research  Information, U.S. Environmental Protection
Agency, EPA/625/R-95/003, Cincinnati, OH.

Roesner, L.A., and E.H. Burgess (1992) CSO Rehabilitation Strategies for Urban Areas.
In Water Resource Planning and Management.  American Society of Civil Engineers,
NY, NY, pp. 654-660.

Roesner, L.A., J.A. Aldrich, and  R.E. Dickinson. 1988.  "Storm  Water Management
Model  User's Manual, Version IV: Addendum I, EXTRAN," EPA-600/3-88-001, U.S.
Environmental Protection Agency, Athens, GA.

Roesner, L.A., R.P. Shubinski,  and J.A. Aldrich. 1984. "Stormwater Management  Model
User's  Manual,  Version   III:  Addendum   I,  EXTRAN,"  EPA-600/2-84-109b,  U.S.
Environmental Protection Agency, Cincinnati, OH.

Rovel,  J., and A.  Mitchel (circa 1995)  Characteristics and Treatment of  Municipal
Wastewater During Rainy Weather.  Unknown Source.

Shaver, E.  (1993)  Beach  Community  Adds Sand Filter  to  Storm  Drains.   Water
Environment & Technology, 5, 5, 18.
                                     188

-------
Shrout,  P.W.  (1994)  Unique Solution to Flow  Equalization.  Water Environment  &
Technology, 6,6, pp. 16-18.

Stanley, D.W.  "Pollutant  Removal  by a Stormwater  Dry Detention  Pond."  Water
Environment Research, 68, 6, 1076-1083.

U.S. Environmental Protection Agency (EPA) (1983), Results of the Nationwide Urban
Runoff Program.  Water Planning  Division, NTIS# PB  84-185552,  Washington,  D.C.,
December 1983.

U.S. Environmental Protection Agency (EPA) (1990), "National Pollutant Discharge
Elimination System  Permit Application  Regulations  for  Storm Water Discharges."
(Phase I Final Rule.) Federal Register. 55 FR 47990, November 16, 1990.

U.S. Environmental  Protection Agency (EPA)  (1992),  Cost Estimates  for  Select
Combined  Sewer Overflow Control Technologies, Office of Municipal Pollution Control,
EPA Contract No. 68-08-0023, Washington, D.C., January 1992.

U.S. Environmental  Protection Agency  (EPA) (1995),  Combined Sewer  Overflows,
Guidance for Nine Minimum Controls, Office of Wastewater Management, EPA 832-B-
95-003, Washington, D.C., May 1995.

U.S. Environmental  Protection Agency  (EPA) (1995),  Combined Sewer  Overflows,
Guidance for Long Term Control Plan, Office of Wastewater Management, EPA 832-B-
95-002, Washington, D.C., May 1995.

U.S. Environmental Protection Agency (EPA) (1998), "National Pollutant Discharge
Elimination System-Proposed Regulations for Revision of the  Water  Pollution Control
Program Addressing Storm Water Discharges."  (Phase II Proposed Rule.)  Federal
Register. 63 FR 1536, January 8, 1998.

U.S. Environmental Protection Agency (EPA)  (1999),  Preliminary Data Summary of
Urban Storm Water Best Management Practices. Office of Water, EPA-821-R-99-012,
Washington, DC, August 1999.

Walesh, S.G.   (1992) Retrofitting Storm Water Facilities  for Quantity and Quality
Control.  In Water Resource Planning and  Management.  American Society of Civil
Engineers, NY, NY. 786-791

Welborn, H.L.  (1974)  Surge  Facility for Wet and Dry Weather Flow  Control.  EPA-
670/2-74-075, U.S. Environmental Protection  Agency, Cincinnati, OH.

Whitney-Jacobsen.  (1981) Joint Dry/Wet Weather Treatment of Municipal Wastewater
at Clatskanie, Oregon.  U.S. Environmental Protection Agency, Municipal Environmental
Research Lab., Cincinnati OH. Whitney-Jacobsen and Associates, Portland, OR.
                                    189

-------
Woodward-Clyde Consultants. (1990) Evaluation of Existing Stormwater Pump Stations
in Santa Clara County for Pollution Control Potential. Prepared for Santa Clara Valley
Water District, July, 1990.

Woodward-Clyde  Consultants.  (1994)  Sunnyvale Detention  Basin  Demonstration
Project. Prepared for Santa Clara Valley Water District, July, 1994.

Woodward-Clyde  Consultants.  (1996)  Sunnyvale Detention  Basin  Demonstration
Project. Santa Clara Valley Nonpoint Source Pollution Control  Program.  Woodward-
Clyde Consultants, Oakland, CA.

World  Health  Organization (WHO).  (1989)  Guidelines  for  Use  of  Wastewater  in
Agriculture and Aquaculture. WHO Technical Report Series 778, WHO, Geneva.7
                                     190

-------
                                APPENDIX A

                        BIBLIOGRAPHIC DATABASES
Wilson Applied Science & Technology Abstracts (WBA)
H.W. Wilson Company                      Phone: (718) 588-8400
950 University Ave.                         800-367-6770
Bronx, NY 10452                           Fax (718) 538-2716

Contact: Technical Support Department.
E-mail: techmailinformationo.hwwiison.com (Internet).

Type: Bibliographic,

Content:  Contains  abstracts to all articles cited in the Applied Science Technology
Index (described in a separate entry) from March 1993 to date.  Includes an index of
non-English-language periodicals English abstracts are  provided.  Sources include 400
English language scientific and technical publications.

Subject Coverage.   Applied  science and  technology, including  such  areas  as
aeronautics and space science, chemistry, computer  technology,  construction, data
processing, energy, engineering,  the  environment, fire  prevention,  food, geology,
machinery,  marine  technology  and  oceanography,   mathematical  metallurgy,
meteorology,  mineralogy, optics,  petroleum and gas, physics,  robotics, solid  state
technology, telecommunications, textiles, transportation,  and water and waste.

Language: English.

Geographic Coverage:  International.

Time Span: October 1983 to date (indexing coverage);  March 1993 to date (Abstracts).

Updating: Approximately 1,000 records per week.
                                     191

-------
Compendex*Plus
Engineering Information, Inc. (Ei)             Phone: (201) 216-8500,
Castle Point on the Hudson                  800-221-1044
Hoboken, NJ 07030                        Fax: (201) 216-8532

Type: Bibliographic.

Content:    Contains more than 3.63 million citations,  with abstracts, from more than
2,500 journals,  reports, books,  and  conference proceedings  covering  fields  of
engineering and technology.  Sources  include approximately 2500 journals,  reports,
books, and conference proceedings.  Conference review  records contain conference
code numbers that are links to complete coverage of all individual papers.  Corresponds
in part to The Engineering Index Monthly and Engineering Index Annual.

Subject Coverage: All fields of engineering,  including related areas of app science,
management,  and  energy.    Specific fields  covered  include  civil  engineering,
construction  and   materials,  transportation,  waterworks,  sanitary,  engineering,
bioengineering, ocean  engineering, mining, petroleum, fuel  technology, metallurgical
engineering, mechanical engineering,  aerospace, engineering, automotive engineering,
electronics  and  electrical engineering,  control engineering,  chemical engineering,
nuclear engineering, and computers and robotics engineering.

Language: English.

Geographic Coverage: International.

Time Span: 1970 to date.

Updating:  Weekly;  13,000 records a month.
                                      192

-------
Environmental Sciences & Pollution Management Database
Cambridge Scientific Abstracts (CSA)         Phone: (301 )961 -6750
7200 Wisconsin Ave., Ste. 601                800-843-7751
Bethesda, MD 20814-4823             Fax: (301 )961 -6720

Type: Bibliographic.

Content:  Contains more than 64,000 citations, with abstracts, to literature covering the
environmental  sciences.   Covers  some  1498  core journals, monographs, and
conference proceedings, as well as 5500 secondary sources, drawn from the primary
sources of abstracts journals such  as:  Ecology, Pollution  Digest  of  Environmental
Impact Statements, toxicology, Health & Safety Science, Bacteriology,  Environmental
Engineering, and others.

Subject  Coverage:   Environmental sciences, including agricultural  biotechnology,
aquatic    environments,   bacteriology,   ecology,   environmental   biotechnology,
environmental  engineering, environmental  impact  statements,  industrial  hygiene,
microbiology applied to industrial and environmental issues, pollution (land, air, water,
noise, and solid waste), risk assessment, safety science, toxicology, water quality, and
water resource issues.

Language: English.

Geographic Coverage: United States.

Time Span:  1983 to date.

Updating: Monthly. 76,000 records/year.


National Center for Environmental Publications and Information
United States Environmental Protection Agency
Washington, D.C.
http://www.epa.gov/ncepihom/index.html

Type: National Publications Catalog

Content: 5,500 titles in paper and/or electronic.

Subject Coverage: Repository for all EPA documents.

Language: English.

Time Span:

Updating: Yearly.
                                     193

-------
ProCite
Robert E. Pitt, P.E., Ph.D., DEE
University of Alabama at Birmingham
Hoehn Engineering Building
1075-13th St. So.
Birmingham. AL  35294
Phone: (205)934-8434
Fax: (205) 975-9042
e-mail: RPITT@ENG.UAB.EDU
Type: Bibliographic.

Content: 3,700 wet-weather references.

Subject Coverage:  The database was  prepared by Robert Pitt for the Wet Weather
Floe Design for the Future project under contract by the United States Environmental
Protection Agency.

Language:  English.

Time Span: 1899-1997

Updating: Database constructed in 1997.
                                    194

-------
                APPENDIX B




SELECTED TABLES WITH ENGLISH EQUIVALENT UNITS
                    195

-------
       Table 5.1 .b-2. Comparison of ACTIFLO and Existing Primary Treatment Systems

                                                                         Net Savings with
Parameter
Total Capacity
Number of Units
Unit Capacity
Area Required
Existing System
240 MG
8
30MGD
14,307 ft2 (per unit)
ACTIFLO
240 MG
6
40MGD
945 ft2 (per train)
ACTIFLO
94% per unit
                          114,453 IT (total)           5,670 fT (total)             95% total
Process Volume          143.066 ft3 (per unit)        20,790 ft3 (per unit)          85% per unit
                         1,114,530 ft3 (total)          124,740 ft3 (total)           89% per total
Hydraulic Retention              51 min                   5.6 min                  89%
Time
                                           196

-------