Controlling Combined Sewer Overflows with Rainwater Harvesting in Olympia, Washington

Item

Title

Eng Controlling Combined Sewer Overflows with Rainwater Harvesting in Olympia, Washington

Date

2011

Creator

Eng Boyd, Lara

Subject

Eng Environmental Studies

extracted text

Controlling Combined Sewer Overflows with Rainwater
Harvesting in Olympia, Washington

by
Lara Boyd

A Thesis
Submitted in partial fulfillment
of the requirements for the degree of
Master in Environmental Studies
The Evergreen State College
May, 2011
i

© 2011 by Lara A. Boyd. All rights Reserved.

ii

This Thesis for the Master of Environmental Studies Degree
by
Lara A. Boyd

has been approved for
The Evergreen State College
by

Robert H. Knapp, Jr., Ph.D.
Member of the Faculty

Ralph Murphy, Ph. D.
Member of the Faculty

Tyle Zuchowski
Capitol Planning Manager, LOTT Clean Water Alliance

Date

iii

ABSTRACT
Controlling Combined Sewer Overflows with Rainwater Harvesting
in Olympia, Washington
Lara Boyd

Urban development creates impervious surfaces, such as roads, parking lots and rooftops
which have significantly altered the movement of water through the environment. Each
year, the precipitation that falls on urban areas in the United States results in billions of
gallons of stormwater runoff that collects various nonpoint source pollutants from
impervious surfaces. Combined sewer systems are designed to collect and convey
domestic, commercial, and industrial wastewater as well as stormwater runoff in the same
pipes. During heavy precipitation events, stormwater volume has the potential to exceed
a wastewater treatment facility’s capacity. When this occurs, wastewater and stormwater
are diverted from the facility and discharged directly into designated receiving surface
waters. This event is called a combined sewer overflow (CSO). CSOs are a major threat
to water quality as they are comprised of both raw sewage and stormwater runoff.
Rainwater harvesting has the potential to mitigate stormwater runoff and this thesis
examines its potential for controlling combined sewer overflows in Olympia,
Washington. Rainwater harvesting is defined as the collection, storage and reuse of
rainwater. On-site rainwater harvesting systems use cisterns to collect and store volumes
of rooftop storm runoff for later use. Approximately six hundred acres of downtown
Olympia is served by a combined sewer system. Under normal conditions, treated
wastewater is released into Budd Inlet from the Lacey, Olympia, Tumwater, Thurston
County (LOTT) Clean Water Alliance owned and operated Budd Inlet Treatment Plant.
However, for the first time in over fifteen years, a CSO event occurred from the Plant in
the early morning hours of December 3, 2007, releasing approximately 9 million gallons
of untreated wastewater into Budd Inlet. Rainwater harvesting systems were modeled
onto large-scale buildings (>10,000 ft2 roof area) served by the combined sewer lines.
Daily cistern levels were modeled based on actual precipitation that occurred on and
around the 2007 CSO event to determine the volume of precipitation captured in the
cisterns and thus prevented from entering the combined sewer lines. Results indicate that
approximately 1.22 million gallons and 275,000 gallons would have been captured on
December 2 and December 3, 2007, respectively, from the 102 buildings analyzed,
indicating a substantial volume of runoff would have been prevented from entering the
combined sewer lines, easing pressure on the Budd Inlet Treatment Plant.

TABLE OF CONTENTS
1. INTRODUCTION ............................................................................................................... 1
2. STORMWATER RUNOFF ................................................................................................... 3
2.1 NONPOINT SOURCE WATER POLLUTION ................................................................. 3
2.2 SEWER SYSTEMS ....................................................................................................... 5
2.2.1 SEPARATE SEWER SYSTEMS (SSSS) AND OVERFLOWS (SSOS) ...................... 8
2.2.2 COMBINED SEWER SYSTEMS (CSSS) AND OVERFLOWS (CSOS) .................. 10
3. STORMWATER MANAGEMENT ...................................................................................... 15
3.1 CONVENTIONAL S TORMWATER MANAGEMENT .................................................... 16
3.1.1 STORMWATER BMP OVERVIEW ..................................................................... 17
3.1.2 CSO CONTROL ................................................................................................ 19
3.1.3 CONVENTIONAL STORMWATER MANAGEMENT CONCLUSIONS ...................... 22
3.2 GREEN S TORMWATER INFRASTRUCTURE (GSI) AND LOW IMPACT DEVELOPMENT
(LID) .............................................................................................................................. 23
3.2.1 LID TECHNIQUES ............................................................................................ 24
3.2.2 LID IMPLEMENTATION FOR S TORMWATER MANAGEMENT (CSO CONTROL)
................................................................................................................................... 28
4. RAINWATER HARVESTING ............................................................................................ 31
4.1 INTRODUCTION/BACKGROUND .............................................................................. 32
4.2 RAINWATER HARVESTING SYSTEM COMPONENTS ............................................... 32
4.2.1 COLLECTION SURFACE .................................................................................... 33
4.2.2 CONVEYANCE SYSTEM .................................................................................... 34
4.2.3 PRE-TANK TREATMENT COMPONENTS ........................................................... 34
4.2.4 STORAGE .......................................................................................................... 36
4.2.5 DISTRIBUTION ................................................................................................. 38
4.2.6 ADDITIONAL TREATMENT FOR POTABLE WATER USE ................................... 39
4.3 REGULATIONS ......................................................................................................... 40
4.4 RAINWATER HARVESTING BENEFITS ..................................................................... 42
4.4.1 WATER CONSERVATION .................................................................................. 43
4.4.2 ENERGY CONSERVATION ................................................................................ 47
4.4.3 FINANCIAL ....................................................................................................... 50
4.4.4 STORMWATER MANAGEMENT ........................................................................ 53
5. CASE S TUDY SELECTION .............................................................................................. 53
5.1 PUGET SOUND ......................................................................................................... 55
5.2 OLYMPIA DEVELOPMENT AND IMPERVIOUS SURFACES ........................................ 57
5.3 OLYMPIA S TORM AND SURFACE WATER ............................................................... 61
5.4 OLYMPIA SEWER AND STORM SYSTEM ................................................................. 62
5.5 COMBINED SEWER OVERFLOWS IN OLYMPIA ....................................................... 65
6. METHODS AND ANALYSIS ............................................................................................ 69
6.1 GIS ANALYSIS ........................................................................................................ 69
6.2 SPREADSHEET S IMULATION MODEL ...................................................................... 78
7. RESULTS ........................................................................................................................ 84
8. CONCLUSIONS ............................................................................................................... 87
8.1 FUTURE P LANS FOR GREATER OLYMPIA ............................................................... 88
8.2 RAINWATER HARVESTING IN OLYMPIA ................................................................. 89
iv

8.3 RESEARCH APPLICATIONS ...................................................................................... 89
BIBLIOGRAPHY ................................................................................................................. 91
APPENDIX A-TREATMENT CONTROL METHOD DEFINITIONS (FROM WADOE, 2005) .. 96
APPENDIX B-ADDITIONAL DESCRIPTIONS AND/OR ILLUSTRATIONS OF SPECIFIC
RAINWATER HARVESTING COMPONENTS ........................................................................ 99
APPENDIX C-BUILDING CHALLENGE EXAMPLES .......................................................... 104
APPENDIX D- ENLARGED OLYMPIA PRECIPITATION CHART ........................................ 110

v

LIST OF FIGURES
Figure 2.1 Distribution of Sewer System Types ............................................................................. 7
Figure 2.2 Typical Separate Sewer System (SSS) .......................................................................... 9
Figure 2.3 Combined Sewer System Locations ............................................................................ 11
Figure 2.4 Typical Combined Sewer System ............................................................................... 12
Figure 4.1 Generalized Illustration of a Rainwater Harvesting System ....................................... 33
Figure 4.2 Percent of Total Water Use that is Non-Potable ......................................................... 44
Figure 4.3 U.S. Mean Daily Per Capita Domestic Water Use ...................................................... 45
Figure 4.4 20-Year Monetary Need for Infrastructure Updates by Water Sector (in 2007
dollars) .......................................................................................................................................... 49
Figure 4.5 Example Storm Runoff Fees ....................................................................................... 51
Figure 4.6 Example Water Rate Increases .................................................................................... 52
Figure 5.1 Olympia Location ........................................................................................................ 54
Figure 5.2 Puget Sound ................................................................................................................. 55
Figure 5.3 Olympia Drainage Basins ............................................................................................ 60
Figure 5.4 Olympia Rainfall 1955 through 2007 .......................................................................... 61
Figure 5.5 Olympia Sewer Lines .................................................................................................. 64
Figure 5.6 Olympia Drainage Basins and Combined sewer lines ................................................ 65
Figure 5.7 Budd Inlet Treatment Plant Outfall Locations ............................................................ 67
Figure 5.8 2007 CSO Event .......................................................................................................... 68
Figure 6.1 Olympia Combined Sewer Lines................................................................................. 70
Figure 6.2 Combined Sewer Lines and All Buildings .................................................................. 71
Figure 6.3 Selected Large-Scale Buildings ................................................................................... 72
Figure 6.4 Olympia Water Use Parcels......................................................................................... 73
Figure 6.5 Selecting Buildings and Parcels Served by the Combined Sewer Lines ..................... 75
Figure 6.6 One Building per One Parcel ....................................................................................... 76
Figure 6.7 Analyzed Building Sizes ............................................................................................. 77
Figure 6.8 Analyzed Building Uses .............................................................................................. 79
Figure 6.9 Daily Total Runoff Calculations ................................................................................. 80
Figure 6.10 Cistern Sizing Analysis ............................................................................................. 81
Figure 6.11 Analyzed Building Modeled Cistern Sizes................................................................ 81
Figure 6.12 Daily Cistern Volume Analysis ................................................................................. 82
Figure 6.13 Day to Day Volume Stored ....................................................................................... 83
Figure 7.1 Modeled Cistern Storage 12/2/2007-12/3/2007 .......................................................... 85
Figure 7.2 Modeled Potential Cistern Storage 12/2/2007-12/3/2007 ........................................... 86
Figure 7.3 Modeled Cistern Storage October 31, 2007 through January 30, 2008 ...................... 87

vi

ACKNOWLEDGEMENTS
I would like to thank Professor Rob Knapp for inspiring my interest in this subject as well
as performing the extensive duties as my primary thesis reader. He provided me with
invaluable guidance, direction and assistance throughout the process of preparing this
thesis. I would like to thank Tyle Zuchowski for supplying the data needed to perform
my analysis as well as his availability to answer my numerous questions. I would like to
thank Professor Ralph Murphy for believing in this thesis and guiding me through the
steps needed to begin my research. I would also like to thank Greg Stewart for assisting
me in GIS analysis and thanks are also due to Agnes Przeszlowska, Shannon Claeson,
and Steve Wondzell for letting me present my defense to them and providing me with
excellent feedback.

vii

LIST OF ABBREVIATIONS/ACRONYMS
BMP- Best Management Practice
BOD- Biological Oxygen Demand
CSO- Combined Sewer Overflow
CSS- Combined Sewer System
EPA- Environmental Protection Agency
GSI- Green Stormwater Infrastructure
LID- Low Impact Development
MGD- Million Gallons per Day
SSS- Separate Sewer System
SSO- Separate Sewer Overflow

viii

1. INTRODUCTION
Over one hundred million acres of land have been developed in the United States, and
development and sprawl are increasing at a faster rate than population growth (Kloss and
Calarusse, 2006). Development increases roads, parking lots, and rooftops, collectively
known as impervious surfaces. Urban landscapes, with large areas of impervious
surfaces, have significantly altered the movement of water through the environment. Not
only that, but urban landscapes allow for stormwater to collect a variety of pollution that
greatly affects the health of our nation’s waters. This nonpoint source pollution is a
major source of contamination of our nation’s waters (NRDC, 1999).
Each year, the precipitation that falls on urban areas in the United States results in
billions of gallons of stormwater runoff and combined sewer overflows (CSOs). CSOs
are the result of combined sewer systems (CSSs), which collect and convey storm and
wastewater in the same pipes. During heavy precipitation events, stormwater volume has
the potential to exceed a wastewater treatment facility's capacity and thus produces a
CSO. CSOs are composed of industrial and commercial wastewater, raw sewage and
urban stormwater runoff.
Scientific research has determined that CSOs represent a serious threat to water
quality (U.S. EPA, 2004; Kloss and Calarusse, 2006; Rochfort, 2000). Mitigation and
prevention of increased storm flows and pollutants as a result of urbanization is one of
the most challenging areas for water resource managers currently. Reducing runoff
decreases the amount of pollution introduced into waterways and relieves the strain on
stormwater and wastewater infrastructure.
Conventional stormwater management techniques rely on stormwater Best

1

Management Practices (BMPs) and often include costly projects constructed to achieve
increased storage capacities during high volume precipitation events. However, as
population growth and development continues, water authorities are looking for efficient
and low-cost stormwater management solutions for managing runoff.
One area of stormwater runoff management currently gaining momentum is low
impact development (LID). LID uses engineered, small-scale hydrologic controls to
replicate the pre-development hydrologic regime of watersheds through infiltrating,
filtering, storing, evaporating and detaining runoff close to its source (PGDER, 1999).
One LID method is rainwater harvesting, also known as rainwater catchment or
collection. Rainwater harvesting systems use cisterns to collect and store volumes of
precipitation, generally from a home or building rooftop. Rainwater harvesting is
extensively promoted as an alternative water supply source, however, harvesting systems
are seldom solely examined for their ability to control stormwater runoff.
Olympia, Washington was selected as the case study location for 4 main reasons.
First, the City borders Budd Inlet, located at the southernmost point of Puget Sound (a
unique, valuable, and threatened ecosystem). Second, Olympia is partially served by a
combined sewer system. Third, the combined sewer lines are located in the older
downtown area which contains concentrated areas of impervious surfaces. Finally,
Olympia is subject to high intensity precipitation events, thus elevating the risk for CSO
events.
In December 2007, Olympia experienced its first precipitation triggered CSO
event in over fifteen years, releasing over nine million gallons of untreated overflow into
Budd Inlet. This thesis examines the potential for controlling CSOs with rainwater

2

harvesting systems in Olympia by modeling rainwater harvesting systems onto largescale buildings served by the combined sewer lines, using the December 2007 CSO event
precipitation and then examining the volume of roof runoff potentially prevented from
entering the combined sewer lines to determine if the modeled rainwater harvesting
systems would have reduced the volume of stormwater runoff that triggered the 2007
CSO event.

2. STORMWATER RUNOFF
Urbanization has greatly affected stormwater runoff. The continued expansion
and growing population concentrations in urban areas has put stress not only on our water
supply and distribution infrastructure, but has directly and indirectly affected our water
supply sources themselves. Stormwater runoff carries with it a myriad of problems,
including nonpoint source water pollution and sewer system overflow events. Section 2.1
will review nonpoint source pollution and section 2.2 provides a background on sewer
systems, including an in-depth discussion of the two types of sewer systems and their
overflow risks, combined sewer systems (CSSs) in 2.2.1 and separate sewer systems
(SSSs) in 2.2.2.

2.1 NONPOINT SOURCE WATER POLLUTION
Development rates in the last twenty years have been double the population
growth here in the U.S. (Kloss and Calarusse, 2006). Of the 100 million acres of
development, impervious surfaces cover over 27 million acres, or 27% (Frazer, 2005).
3

This means massive displacement and conversion of areas formerly able to capture and
assimilate precipitation into impenetrable surfaces. The expected population increase of
25% between the years 2000 and 2025 will add another 68 million acres of development
(Beach, 2002).
The landscape vegetation of undeveloped land captures precipitation, allowing it
to mostly infiltrate where it falls. In urban and suburban areas, much of the landscape is
covered by impervious surfaces, such as rooftops and paved areas. Impervious surfaces
alone do not generate pollution. However, impervious surfaces are a critical contributor
to the hydrologic changes that degrade waterways, an acute contributing factor of the
intensive land uses that do generate pollution, prevent natural pollutant processing
through infiltration of precipitation into soil, and provide an effective conveyance system
to transport pollutants into storm drains or waterways (Arnold and Gibbons, 1996).
Impervious surfaces intensify stormwater runoff, enhance stream channel erosion and
diminish groundwater recharge (Stone Jr., 2004).
The U.S. EPA identifies urban stormwater runoff from impervious surfaces a
leading threat to water quality (U.S. EPA, 1996). This is because toxic and pathogenic
pollutants accumulate on impervious surfaces and are picked up and carried by
stormwater, generating what is known as nonpoint source water pollution. These
pollutants include sediment; oil, grease, and toxic chemicals from motor vehicles;
pesticides and nutrients from lawns, gardens and golf courses; viruses, bacteria and
nutrients from pet waste and failing septic systems; heavy metals from roof shingles and
motor vehicles; garbage and more (U.S. EPA, 2003). Nonpoint pollution resulting from
stormwater runoff has been identified as one of the major causes of the deterioration of

4

the quality of receiving waters (Lee and Bang, 2000).
Nonpoint source water pollution harms fish and wildlife populations, kills native
vegetation, pollutes drinking water sources and threatens safety of recreational areas.
There is a direct correlation between impervious surface proportion within a watershed
and water quality. A number of scientific studies have determined that when watershed
imperviousness exceeds 10%, aquatic ecosystem health tends to decline. At 30%
impervious surface coverage, the watershed will become critically impaired (Arnold and
Gibbons, 1996; Booth and Jackson, 1997; Wang, 2001).
The amount of impervious surfaces in urbanized areas ranges. In residential
areas, coverage increases from approximately 10% in low-density subdivisions to over
50% in multi-family communities. In industrial and commercial areas, impervious
surface coverage rises above 70% and in regional shopping centers and dense urban
areas, impervious coverage is over 90% (Schueler, 2000). In these densely impervious
locations, stormwater is often directed into storm drains to be transported to the local
sewer system to be treated and discharged into receiving waterways. In fact, most
communities rely on a municipal sewage system to deal with stormwater runoff and such
systems will be presented next.

2.2 SEWER SYSTEMS
Municipal sewer systems are a considerable and important sector of our nation's
infrastructure. Our daily activities generate liquid wastes that without proper treatment
and disposal would generate numerous environmental and health issues. Wastewater is
any water that has been adversely affected in quality by human influence. Sewage is
5

specifically the subset of wastewater that is contaminated with feces and urine, however,
the terms wastewater and sewage are often used interchangeably. To understand the
significance of sewer systems, it is important to provide a brief discussion concerning the
history of sewer systems and wastewater treatment here in the U.S.
Construction of municipal sewer systems did not start in the U.S. until the 1880s
(U.S. EPA, 2004). Before sewer systems, human waste was dumped into privy vaults
and cesspools, and stormwater ran into the streets or into surface drains. Rapid
urbanization between 1840 and 1880 resulted in increased quantities of wastewater that
could not be handled by privy vaults and cesspools. Sewer systems were constructed to
protect public health, to address flooding issues as well as to increase community
aesthetics.
Municipalities installed sewer systems using two prevailing design options,
combined sewer systems (CSSs), where wastewater and stormwater runoff are collected
and conveyed in a single pipe system or separate sanitary and storm sewer systems
(SSSs), where wastewater and stormwater runoff are collected and conveyed using two
separate systems of pipe. Figure 2.1 displays the distribution of the two sewer types and
shows that combined sewer systems occur less frequently than separate sewer systems
and are more concentrated in the Eastern U.S. However, there are a few located on the
West Coast, including one in Olympia, Washington. Both combined and separate sewer
systems are designed to overflow storm and/or wastewater. The discharge from both
types of sewer systems is called point source water pollution as it is a localized,
identifiable source of pollution. These systems will be discussed in detail next in 2.2.1
and 2.2.2.

6

Figure 2.1 Distribution of Sewer System Types

http://www.epa.gov/npdes/pubs/csossoRTC2004_executive_summary.pdf

As sewer systems were first being constructed in the 1880s, no model sewer systems
existed for guiding construction, and engineers were reluctant to experiment with
expensive capital works (U.S. EPA, 2004). For municipalities that needed both sanitary
and storm sewers, it was less costly to construct CSSs. For municipalities that needed
only a wastewater collection system, constructing a SSS was less expensive. Generally,
large cities opted to construct CSSs because of the flood prevention offered by combined
systems while smaller cities sought separate sanitary and storm sewers (U.S. EPA, 2004).
Sewer systems can be publicly or privately owned, meaning owned and operated
by a state or municipality, or a facility whose operator is not the operator of the treatment
works, respectively. Municipal sewer systems serve approximately 208 million people in
the U.S. and the EPA estimates that publicly-owned sewer systems account for 724,000
miles of sewer pipe while privately-owned sewer systems account for about 500,000
miles of piping to deliver wastewater into these systems (U.S. EPA, 2004). Pipe
corrosion occurs in virtually all types of piping systems and over time, susceptibility to
7

corrosion is increased. Leaching occurs, leaks develop and water quality can be
degraded. Over two-thirds of the U.S. population depends on hundreds of thousands of
miles of piping that is already ageing or will over time.
While the generalized average age of sewer system components is 33 years,
components of some systems date back over a century (U.S. EPA, 2004). Not only are
these systems aging, but marked variability exists in the current condition of sewer
infrastructure, as municipalities have used an array of materials, design and installation
practices and maintenance and repair procedures (U.S. EPA, 2004). Population growth
will likely contribute additional wastewater to be processed by these variable and ageing
municipal sewer systems, adding stress to the already burdened wastewater infrastructure.
Sewer systems are necessary in maintaining the health of our communities and
greatly improve local sanitary conditions. However, the age and structural design of both
CSSs and SSSs present health concerns for both communities and the natural
environment surrounding them. This is because these systems are designed to overflow
storm and/or wastewater during specific situations, resulting in separate storm sewer
overflows (SSOs) and combined sewer overflows (CSOs). The next two sections will
present background and potential risks associated with both SSSs and CSSs.
2.2.1 SEPARATE SEWER S YSTEMS (SSSS) AND OVERFLOWS (SSOS)

Separate sewer systems (SSSs), also known as separate sanitary and storm sewer

systems are comprised of two sets of pipes that collect and convey domestic, commercial
and industrial wastewater mixed with limited amounts of infiltration and inflow 1 to a

1

Inflow is defined as surface water entering the sewer via means other than groundwater. Inflow
is usually the result of precipitation events. Infiltration is defined as groundwater that enters the

8

treatment plant in one set, while collecting and conveying stormwater runoff directly to
surface waters in the other (see Figure 2.2). This is problematic because polluted
stormwater runoff is commonly discharged untreated into local water bodies. While the
concentration of pollutants in stormwater is generally more dilute than in wastewater, it
can still contain significant amounts of pollutants as discussed earlier.

Figure 2.2 Typical Separate Sewer System (SSS) Function during Dry and Wet Weather

http://www.phillyriverinfo.org/CSOLTCPU/Home/images/Sewers%20in%20Dry%20Weather.png

A properly designed, operated and maintained SSS is meant to collect and
transport all the sewage that flows into it. However, SSSs are at risk of experiencing
sanitary sewer overflows (SSOs). SSOs are partially treated or untreated sewage
overflows from a sanitary sewer collection system. While SSOs occur, they occur
infrequently and properly maintained SSSs are designed to handle and treat all incoming

sewer, usually through leaky sewer pipes and joints, manholes and service connections (LOTT,
2009).

9

wastewater. Combined sewer systems (CSSs) share wastewater and stormwater runoff in
the same piping and are designed to overflow when incoming flow levels exceed the
ability of the treatment plant. The focus of this thesis is on combined sewer systems
(CSSs) and combined sewer overflows (CSOs), which will be discussed next.

2.2.2 COMBINED SEWER S YSTEMS (CSSS) AND OVERFLOWS (CSOS)

As sewer systems were first constructed, it was recognized early on that

wastewater collection improved local health conditions and often reduced illness.
However, by the 1890s, drinking water being drawn downstream from untreated
wastewater discharges resulted in major cholera and typhoid outbreaks (U.S. EPA, 2004),
and the need to provide wastewater treatment was established. It was clear that although
CSSs were efficient means of collecting and conveying storm and wastewater, they also
made treatment more difficult because large variation in flows existed between wet and
dry weather. Most state and local authorities have not allowed construction of new CSSs
since the 1950s (U.S. EPA, 2004).
Approximately 772 communities nationwide contain CSOs, mainly within the
Great Lakes, Northeast and Puget Sound regions (see Figure 2.3). Of the 772
communities, approximately 30% have populations greater than 75,000, and the other
70% are small with total service populations of less than 10,000 (U.S. EPA, 2001).
Increased stormwater flows generated in urban areas containing CSSs increase the
potential for CSO events with increased pollutant concentrations. Also, because most
CSSs were designed before 1950, population increases in the following decades have

10

Figure 2.3 Combined Sewer System (CSS) Locations in the U.S.

Available from http://cfpub.epa.gov/npdes/cso/demo.cfm?program_id=5

increased dry weather wastewater flows, leaving less capacity available for storm flows.
As mentioned earlier, CSSs are designed to collect and convey domestic,
commercial and industrial wastewater as well as stormwater runoff in a single pipe
system. Generally, CSSs were designed to carry three to five times the average dry
weather flow, giving the pipe system considerable capacity during dry weather.
However, during wet weather events, CSSs are designed to overflow, discharging directly
to surface waters (rivers, estuaries and coastal waters) when total flows exceed the
capacity of the CSS or treatment plant (see Figure 2.4). CSO duration and frequency
vary from system to system and from outfall to outfall within a single CSS (U.S. EPA,
2001).
CSOs are composed of commercial and industrial wastewater, raw sewage and
stormwater runoff and are significant sources of point source water pollution. The EPA
estimates that 850 billion gallons of CSO outfall are discharged into our surface waters
each year (Kloss and Calarusse, 2006). CSOs contain substantial amounts of microbial
pathogens, oxygen-depleting substances, suspended solids (small particles housing
11

pollutants and pathogens on the surface that remain in suspension in water), toxics
(mostly metals and pesticides), and nutrients and floatables (water-borne litter and debris)
from both wastewater and stormwater runoff. CSO events are cause for serious public
health and water quality concerns.

Figure 2.4 Typical CSS Function during Dry and Wet Weather

http://www.phillyriverinfo.org/CSOLTCPU/Home/images/Sewers%20in%20Wet%20Weather.png

Generally, CSOs are induced during wet weather and overflows during dry
weather events are rare and prohibited under federal regulation. Pollutant concentrations
of CSO events vary substantially (U.S. EPA, 2004). The relative amounts of domestic,
commercial, industrial wastewater and urban stormwater carried by a CSS during storm
events are the key operatives of pollutant concentrations associated with CSO discharge.
There are other factors that influence CSO pollutant concentrations. First,
pollutants increase on surfaces over long periods of time between wet weather events and
the longer duration of time, the higher the pollutant concentrations. Also, the duration of
12

the wet weather event influences pollutant loads, as the early stages of a CSO event
(often referred to as the first flush) contain the highest pollutant concentrations. Finally,
the intensity and duration of the wet weather event influences pollutant concentrations.
According to the EPA, CSO pollutants impact five general areas: aquatic life
support, shellfish harvesting, fish and shellfish consumption, drinking water supply, and
water recreation (U.S. EPA, 2004). The following paragraphs will describe these impacts
in more detail.
When water provides a suitable habitat for protection and propagation of desirable
fish, shellfish and other aquatic organisms, it is designated as aquatic life support (U.S.
EPA, 2004). CSOs (and SSOs) contain oxygen-demanding substances that contribute to
impaired aquatic life support impairment. CSO pollutants are generally not widespread
causes of aquatic life impairment, however several states have experienced diminished
aquatic life support capacity within receiving waters of CSO outfall locations (U.S. EPA,
2004).
The shellfish industry in the U.S. is responsible for supporting thousands of jobs
and generating hundreds of millions of dollars for the U.S. economy (NOAA, 1998).
Unfortunately, commercial and recreational shellfish harvesting in populated coastal
areas has been steadily declining since the 1900s, when outbreaks of typhoid were linked
to untreated wastewater (U.S. EPA, 2004). In 1995, more than 33,000 square miles of
marine and estuarine water in the contiguous U.S. were classified as shellfish growing
waters (NOAA, 1998). These waters were carefully surveyed and classified for harvest
to protect public health and ensure safe harvests. Results determined that the primary
basis for shellfish harvest restriction was due to concentrations of fecal coliform bacteria

13

associated with human sewage from sewer system outfalls and organic wastes from
livestock and wildlife deposited via stormwater runoff. An average of 20% of classified
waters are prohibited from harvest annually (NOAA, 1998).
The U.S. population consumes millions of pounds of shellfish annually (NOAA,
1998). Bioaccumulation of the microbial pathogens found in CSO outfall occurs with
shellfish grown in contaminated waters. Pathogens can be passed on to humans by eating
whole, partially cooked or raw contaminated shellfish (U.S. EPA, 2004). Consuming
shellfish contaminated with raw sewage can lead to gastroenteritis in humans, an
inflammation of the gastrointestinal tract resulting in acute diarrhea. If untreated,
gastroenteritis can be fatal.
Public water supply systems 2 provide drinking water to 90% of Americans (U.S.
EPA, 2009). About 65% of the population served by these systems receives water drawn
from surface waters (rivers, lakes and reservoirs) while the other 35% drink groundwater.
It is possible to contract waterborne diseases from contaminated municipal drinking
water and well water. Contamination occurs when sewer system outfalls are located near
drinking water intake sources. The EPA identified fifty-nine CSO outfalls located within
one mile upstream of a drinking water intake source (U.S. EPA, 2004). Drinking water
authorities drawing freshwater near a CSO outfall are aware and in communication with
the CSS authorities responsible for outfalls and as such, CSOs generally do not pose a
major risk of contamination to most public drinking water intakes.
Each year, millions of people in the U.S. use oceans, rivers, lakes and streams for
a variety of recreational activities (U.S. EPA, 2004). Water-based activities, such as
2

A provision to the public of drinking water with at least fifteen service connections or regularly
services at least twenty five individuals

14

swimming and boating are restricted where CSO discharge occurs. This is because
documented cases of gastroenteritis and gastrointestinal illness has been observed from
contact with and ingestion of water near wastewater or storm drain outfalls (U.S. EPA,
2004).
The cost for mitigating CSOs nationwide has been estimated at $54.8 billion
annually (U.S. EPA, 2008a). CSOs during dry weather are prohibited by law, so the
regular events that occur are triggered by stormwater runoff. Stormwater management
practices are used to handle and prevent SSOs and CSOs as well as stormwater runoff.
Section 3 presents an overview of stormwater management.

3. STORMWATER MANAGEMENT
Stormwater management is the practice of managing the quantity and quality of
stormwater. Stormwater runoff has been a source of great concern regarding the health
of our country's waterways for several decades. Stormwater management in general
centers on the use of Best Management Practices (BMPs). Conventional stormwater
management is presented in 3.1, including an overview of stormwater BMPs in 3.1.1,
CSO control methods in 3.1.2, and general conclusions regarding conventional
stormwater management in 3.1.3. An increasingly popular stormwater management
methodology is called green stormwater infrastructure (GSI) or low impact development
(LID). This methodology uses stormwater BMPs that center on maintaining or restoring
the pre-development hydrologic regime of urban and developing watersheds. GSI and
LID are presented in 3.2, with a discussion of techniques in 3.2.1 and implementation in

15

various cities in 3.2.2.

3.1 CONVENTIONAL STORMWATER MANAGEMENT
The Clean Water Act was passed in 1972 to protect our nation's waters from
pollution and amendments to the Act in 1987 required the EPA to address stormwater
runoff in two phases under Section 402, the National Pollution Discharge Elimination
System (NPDES) permit program. The NPDES program controls water pollution by
supervising point sources that discharge pollutants into waters of the United States, and
regulatory monitoring of SSSs and CSSs was implemented in two phases.
Phase I of the NPDES program was issued in 1990 and requires medium and large
cities or certain counties with populations of 100,000 or more to obtain NPDES permit
coverage for their stormwater discharges. Phase II of the NPDES program was issued in
1999 and requires regulated small sewer systems in urbanized areas, meaning a
population density greater than 1,000 people per square mile, to obtain a NPDES permit.
In order to follow the NPDES program requirements, municipalities across the
U.S. have developed Best Management Practices (BMPs) for stormwater management.
BMPs are commonly used to describe structural or engineered control devices and
systems as well as operational or procedural practices designed to retain and/or treat
storm runoff. Stormwater management centers around the use of BMPs, and the most
commonly applied conventional stormwater management techniques display a
reactionary rather than preventative methodology for managing storm runoff. An
overview of the three common types of BMPs applied in conventional stormwater
management are presented in 3.1 as well as a brief discussion regarding detention ponds,
16

the most commonly applied management practice.
Municipalities containing SSSs are required to develop and implement a
stormwater management program to reduce the contamination of stormwater runoff while
municipalities containing a CSS are required to implement EPA's determined nine
minimum controls and develop long-term CSO control plans, both of which will be
discussed in 3.1.2.
Conventional stormwater management has focused on removing stormwater from
a site as quickly as possible to reduce on-site flooding. This means predominate
implementation of two types of management techniques. First are curb and gutter and
piping systems that discharge runoff to the nearest receiving water. Second is
implementation of detention type BMPs to reduce peak runoff discharge rates, such as
detention ponds (NHDOES, 2008).

3.1.1 STORMWATER BMP OVERVIEW

The three common types of BMPs are source control, treatment, and flow control.

Source control BMPs reduce the exposure of materials to stormwater, therefore reducing
the amount of pollutants picked up by storm runoff. Source control BMPs target the
activities that produce contaminants and can be divided into two broad categories. The
first category includes BMPs dictating planning, design and construction of
developments and re-developments to minimize or eliminate adverse impacts. The
second category includes education and training to promote awareness of the potential
problems associated with stormwater runoff and of specific BMPs to help solve
stormwater runoff problems.
17

Treatment BMPs include facilities that remove pollutants by gravity settling of
particulate pollutants, filtration, biological uptake, and soil adsorption (WADOE, 2005).
Methods used for runoff treatment facilities include wetpools, biofiltration, oil/water
separation, pretreatment, infiltration, filtration, emerging technologies, on-line systems,
and design flow 3. Treatment BMPs can accomplish significant levels of pollutant load
reductions if designed and maintained properly.
Flow control BMPs typically control the rate, frequency, and flow duration of
stormwater runoff. The need to provide flow control BMPs is dependent on whether a
development directly or indirectly discharges to a stream system or a wetland (WADOE,
2005). The purpose of flow control is to store volumes of water that can later be slowly
released into the collection system. The volume of storage needed for flow control
depends on several factors: the size of the drainage area, the extent of disturbance of the
natural vegetation, topography and soils, the proportion of impervious surfaces, and the
target release rates, i.e. how rapidly the water is allowed to leave (WADOE, 2005).
Source control BMPs are preferred over treatment BMPs because it is more cost
effective to prevent pollutants from entering runoff rather than treating runoff to reduce
pollutants. However, detention ponds, a method of flow control, are the most commonly
applied stormwater BMP (WADOE, 2005). The function of detention ponds is to manage
flows by capturing and detaining stormwater runoff from developed areas and slowly
releasing outflow. Unfortunately, the design of some detention ponds prevents
groundwater recharge by restricting stormwater to large cement basins, serving as a
reactive rather than proactive method for handling stormwater runoff.

3

See Appendix A for treatment control method definitions

18

Both CSSs and SSSs implement control methods to prevent overflow events from
occurring. However, CSSs are designed to overflow when the treatment system is
overwhelmed whereas SSSs are not. This thesis examines the control of CSOs, therefore,
I only present methods for CSO control next in 3.1.2.

3.1.2 CSO CONTROL

Historically, CSO prevention was difficult due to an inability to fully quantify

CSO impacts on receiving water quality as well as the site-specific variability in the
volume, frequency and characteristics of CSO events (U.S. EPA, 1995). The EPA
published the CSO Control Policy on April 19, 1994 and sought to reduce negative
impacts from CSOs on water quality, aquatic biota, and human health (U.S. EPA, 1994).
As mentioned above, CSOs are subject to NPDES permit requirements. Permits
authorizing discharges from CSO outfalls include very strict water quality-based
requirements to meet water quality standards (U.S. EPA, 2004). The CSO Control Policy
directs implementation and enforcement responsibilities to NPDES authorities.
Currently, the 772 CSO communities have a total of 9,471 outfall locations that
are identified and regulated by 828 NPDES permits. CSSs are diverse, varying in
configuration, size, age, number and location of outfalls. The CSO Control Policy
establishes objectives for CSS communities, including implementation of prescribed nine
minimum controls and requiring development of a long-term CSO control plan. The nine
minimum controls are as follows:
1. Proper operation and regular maintenance programs for the sewer system;
2. Maximum use of the collection system for storage;
3. Review and modification of pretreatment requirements to assure CSO
impacts are minimized;
19

4. Maximizing flow to the sewage treatment facility for treatment;
5. Prohibition of CSOs during dry weather;
6. Control of solids and floatable materials in CSOs;
7. Pollution prevention;
8. Public notification to ensure that the public receives adequate
notification of CSO occurrences and CSO impacts;
9. Monitoring to effectively characterize CSO impacts and the efficacy of
CSO controls (U.S. EPA, 2004).

After passing the Control Policy, the EPA expected municipalities to implement
the nine minimum controls and to submit appropriate documentation to NPDES
authorities as soon as reasonably possible. Of the 828 active CSO permits identified by
EPA in July 2004, 94% (777 permits) required implementation of the nine minimum
controls (U.S. EPA, 2004).
In addition to implementing the nine minimum controls, EPA and NPDES
authorities expect CSO communities to develop and implement a long-term CSO control
plan that includes measures for achieving water quality standards set by the Clean Water
Act. The plan must evaluate a range of control options, including costs and benefits (the
EPA recommend a total nine elements they consider essential for a long-term CSO
control plan 4 to direct the selection of an alternative that would achieve water quality
objectives compliant with the Clean Water Act (U.S. EPA, 2001). Almost 90% of
permits (708 of 828) required development and implementation of a long-term CSO
control plan (U.S. EPA, 2004).
For communities developing long-term CSO control plans, municipalities are
required to consider significant structural controls. Common CSO control measures
4

1) characterization, monitoring and modeling of the CSS; 2) public participation; 3)
consideration of sensitive areas; 4) evaluation of alternatives to meet Clean Water Act
requirements; 5) evaluate cost and performance considerations; 6) an operational plan; 7)
maximization of treatment at the sewer treatment facility; 8) create an implementation schedule;
9) a post-construction compliance monitoring program (U.S. EPA, 2004).

20

identified by long-term control plans include off-line storage facilities, plant
modifications, sewer rehabilitation, disinfection facilities and sewer separation (U.S.
EPA, 2004). Off-line storage facilities store wet weather flows in near-surface storage
basins, such as tanks and basins or deep tunnels located adjacent to the sewer system
(U.S. EPA, 2004). Simple plant modifications, such as changing the physical treatment
processes and plant operations during wet weather can result in increased ability to
handle wet weather flows. Rehabilitating sewer systems refers to the replacement of
structural components that have deteriorated with time. The gradual breakdown of these
components allows more groundwater and stormwater to infiltrate into the sewer system
(U.S. EPA, 2004). Disinfection is necessary to protect public health from wastewater
discharges and facilities hold and treat wastewater before discharging it, although
discharge often includes high levels of toxic residual chlorine (U.S. EPA, 2004).
Sewer separation is the most commonly implemented long-term control plan
method for CSO control (U.S. EPA, 2004), and is the act of separating combined, single
pipe systems into separate sewers for sanitary and storm flows. Separating a CSS
contributes to improvements of water quality by reducing or eliminating sanitary
discharges to receiving waters (U.S. EPA, 2004). This in turn prevents contact risk with
pathogens and impacts to aquatic species. Separating a CSS also relieves regulations
associated with CSOs.
However, there are negative impacts associated with sewer separation. These
include impacts related to extensive construction, disturbances to residents and
businesses, potential disruptions in sewer service, the need for new stormwater discharge
controls to maintain positive water quality results, and high costs. Commonly

21

implemented CSO control measures are generally structural and can be very expensive.
Sewer separation can cost as much as $600 per foot of separation (Kloss and
Calarusse, 2006). However, costs are highly variable due to the location and layout of
existing sewers, the location of other utilities that will have to be avoided during
construction, other infrastructure work that may be required, land uses and costs, and the
construction method used (U.S. EPA, 1999). Sewer separation costs are increasing over
time and many cities have only been able to separate portions of their combined systems
due to high costs and physical limitations, such as location within heavy development.
The EPA estimates that $63.6 billion (an increase of $13 billion from 2000) is needed to
fund established water quality problems associated with CSOs existing as of 2008 or
expected to occur in the following 20 years (U.S. EPA, 2008a).

3.1.3 CONVENTIONAL STORMWATER MANAGEMENT CONCLUSIONS

Conventional stormwater management methods, such as the highly implemented

detention pond method, have proven to be problematic to downstream waters by altering
stream channel morphology5, reducing groundwater recharge, and increasing frequency
and magnitude of floods (NHDOES, 2008). These issues make less water available for
drinking water withdrawals and stream base flows.
The downfalls associated with conventional stormwater management are largely
due to methods that rely on conveyance efficiency and end-of-pipe treatment. The key to
efficient and effective management of stormwater runoff is to simply reduce the volume
of runoff generated at the source and maintaining as much of the original site hydrology.
5

Stream channel morphology includes stream alignment, cross-section geometry and streambed
composition

22

The same applies for CSSs. If CSSs are functioning properly and not
experiencing CSO events during dry weather periods, reducing or eliminating inflow
from stormwater runoff is the most efficient means of controlling CSO events. Reducing
or eliminating inflow also prevents costly infrastructure updates, such as CSS separation.
Reducing inflow from storm events is also cost effective by being preventative in nature,
therefore reducing treatment costs associated with wastewater flows. Many communities,
especially areas containing CSSs have recently shifted focus toward a new methodology
for stormwater management. This is called green infrastructure for stormwater
management and includes low impact development (LID) technologies for managing
storm runoff and will be presented next in 3.2.

3.2 GREEN STORMWATER INFRASTRUCTURE (GSI) AND LOW IMPACT
DEVELOPMENT (LID)
Green infrastructure is a method of stormwater management that is
environmentally sensitive, sustainable, and generally cost-effective (U.S. EPA, 2004).
The term green infrastructure refers to a class of stormwater BMPs or management
practices that slow, capture, treat, infiltrate and/or store runoff at its source, and include
both structural (stormwater capture and treatment) and non-structural (preservation of
open space) approaches (LaBadie, 2010). Green infrastructure techniques infiltrate,
evapotranspire, capture and reuse stormwater to maintain or restore natural hydrology
and can be applied at the site, neighborhood, or regional scale.
The term low impact development (LID) generally refers to development
approaches and principles that utilize green infrastructure techniques to create functional
23

drainage systems. Often these two terms are used interchangeably to define a new,
comprehensive land planning and engineering design approach with a goal of
maintaining and enhancing the pre-development hydrologic regime of urban and
developing watersheds (Low Impact Development Center, 2010). I will be referring to
this methodology as LID for the purpose of this thesis. Next, LID techniques are
presented in 3.2.1 and implementation of LID techniques for stormwater management is
discussed in 3.2.2.

3.2.1 LID TECHNIQUES

Integrated stormwater management involves the development and implementation

of a range of LID and conventional BMPs to improve the quality of urban stormwater
runoff before its discharge into the receiving environment. LID techniques are designed
to control the timing and volume of storm discharges from impervious surfaces as well as
the volume of wastewater (residential, commercial, and industrial) generated within a
community (U.S. EPA, 2004). LID principles and applications represent a compelling
conceptual shift from a purely structural approach to stormwater management, as LID
methods are runoff reduction stormwater BMPs (Puget Sound Action Team, 2005).
On a large scale, protecting and restoring natural landscape features (such as
forests, floodplains and wetlands) are significant elements of LID. Preserving these areas
simultaneously provides wildlife habitat and recreation opportunities while improving
water quality. On a smaller scale, LID techniques can be divided into broad categories
such as infiltration practices, filtration practices, and runoff storage practices.
LID infiltration practices are engineered structures or landscape features designed
24

to capture and infiltrate runoff. They are used to reduce both storm runoff volume
discharge from a site and to mitigate infrastructure needed to convey, treat, or control
runoff (U.S. EPA, 2007). Infiltration practices are also used for groundwater recharge,
which is particularly valuable in communities where maintaining drinking water supplies
and stream baseflow 6 is of special concern due to limited precipitation or high ratios of
water withdrawal to groundwater recharge rates. LID infiltration practices include
porous pavement, disconnected downspouts, infiltration planters, and rain gardens, which
will be presented briefly below.
LID filtration practices are similar to infiltration practices with the added
advantage of providing increased pollutant removal benefits. Rain gardens and
infiltration planters can also be categorized as filtration practices because they can
provide pollutant removal. Filtration practices treat runoff by filtering it through
elements designed to capture pollutants through the process of physically filtering
dissolved pollutants. Filtration practices offer many of the same benefits as infiltration,
such as reduced storm runoff volume, groundwater recharge, and increased stream
baseflow (U.S. EPA, 2007). Rain gardens, infiltration planters, vegetated swales,
sidewalk trees and tree planters are examples of filtration LID techniques.
Runoff storage LID practices are beneficial because storm runoff from impervious
surfaces can be captured and stored for reuse or gradually infiltrated, evaporated, or used
to irrigate plants (U.S. EPA, 2007). Runoff storage practices can reduce the volume of
runoff discharged to surface waters, reduce the erosive forces of high storm runoff flows
and irrigate landscaping. Runoff storage LID practices include green roofs and rainwater
6

Stream baseflow is defined as the sustained flow in a stream channel because of subsurface
runoff

25

harvesting systems. All of the example techniques will be discussed briefly below.
Porous pavement, otherwise known as permeable pavement, offers one solution to
managing increased stormwater runoff and decreased water quality associated with
transportation related surfaces, such as roads and parking lots. Porous pavement is
commonly made up of a matrix of interlocking concrete blocks constructed with voids
that allow stormwater to infiltrate through to the underlying soil, which provides
groundwater recharge and reduced urban storm flows (Brattebo and Booth, 2003).
Permeable pavement is very effective at mitigating storm runoff, however, are not
recommended for heavily trafficked areas, such as interstates and highways, as surface
durability can be degraded.
Disconnecting downspouts is a simple LID method to prevent storm runoff from
entering the sewer system. Downspouts on many homes are connected directly to the
local sewer system. Disconnecting downspouts directs roof runoff to drain to lawns and
gardens or any form of bioretention, where it can infiltrate into the soil, rather than being
transported to the sewer system.
Infiltration planters are containers with open bottoms to allow stormwater to
slowly infiltrate into the ground and are only recommended for areas where soil is well
drained. They contain a layer of gravel, soil and vegetation, where storm runoff
temporarily pools on the topsoil and slowly infiltrates through the planter into the ground.
Infiltration planters are variable and can contain a variety of vegetation as well as
constructed of many materials, such as wood, stone, brick, concrete and plastic. Planters
are commonly located in urbanized areas where space is limited, such as a downtown
(LaBadie, 2010).

26

Rain gardens are natural or dug shallow depressions planted with deep-rooted
native plants and grasses designed to capture and soak up stormwater runoff from
impervious areas such as sidewalks, walkways and compacted lawns. Stormwater is held
in the garden for a short period of time and is allowed to naturally infiltrate into the
ground. Although small scale and often implemented at residential homes, rain gardens
provide multiple benefits. They absorb, filter and assimilate nonpoint source pollutants
before they reach the storm drain, provide habitat for beneficial insects and birds, and
increased groundwater recharge (Dussaillant et al., 2005). Rain gardens are also easy to
implement as they require less technical expertise to install and maintain than other LID
methods.
Vegetated swales, also known as bioswales, are similar to rain gardens in that
they are a form of bioretention used to partially treat water quality, control flooding and
convey stormwater away from critical infrastructure (UF, 2008). These systems are linear
and applied as residential roadside swales, highway medians, and parking lot islands and
medians, parallel to roadways. These open-channel drainageways are designed to convey
storm runoff and the vegetation treats a portion of the stormwater by absorbing, filtering
and assimilating pollutants while well-drained soils enhance site infiltration, recharging
groundwater. Vegetated swales are often used as an enhancement of or alternative to
traditional stormwater piping (UF, 2008).
Sidewalk trees intercept precipitation and are planted to reduce stormwater runoff
and the urban heat island effect as well as improve the urban aesthetic and air quality.
Tree boxes are purchased to address spatial issues associated with sidewalk trees, which
are often compacted into restrictive spaces. To obtain the full range of potential for street

27

trees, a healthy soil volume is needed and tree boxes are constructed to provide adequate
soil space.
Green roofs are roofs with a vegetated surface and substrate and provide multiple
ecosystem services in urban areas (Oberndorfer, 2007). Green roofs improve stormwater
management by capturing and assimilating precipitation that would otherwise become
roof runoff. Green roofs also help regulate building temperatures by providing additional
insulation. Green roofs reduce urban heat-island effects 7 by increasing reflection of
incoming radiation away from a surface. They also increase urban wildlife habitat by
providing a stop for insects as well as resident and migrating birds. Green roofs generally
represent a higher monetary investment than other LID techniques, but energy savings
and potential credit for reducing greenhouse gas emissions and for stormwater
management result in significant savings.
Rainwater harvesting is a dynamic LID technique that not only controls storm
runoff but also promotes water conservation by supplying an alternative source of water
for non-potable uses such as toilet flushing and landscape irrigation. Described simply,
rainwater harvesting uses storage tanks, such as cisterns, to capture runoff, generally
from rooftops, which can be used later for non-potable water applications. This thesis
examines the ability to control CSO events with rainwater harvesting, therefore,
rainwater harvesting is discussed in-depth later in Section 4.
3.2.2 LID IMPLEMENTATION FOR STORMWATER MANAGEMENT (CSO CONTROL)

Many cities nationwide are using LID techniques as advanced components for

7

An urban heat island is a metropolitan area that is significantly warmer than the surrounding
rural areas due to less vegetation, high heat retention from impervious surfaces that absorb
sunlight, and increased levels of air pollution

28

controlling stormwater runoff in urban areas (Hyland and Zuravnsky, 2008). While the
consideration of utilizing LID for stormwater management is commonly recognized, the
application of LID techniques for CSO control has been limited (U.S. EPA, 2004).
Although LID techniques are not currently used as a major combined sewer overflow
(CSO) control method in most urban areas, it is being used to complement engineered
solutions and has shown capability as part of larger stormwater management programs to
reduce the need and sizes of structural controls, such as storage (Hyland and Zuravnsky,
2008; U.S. EPA, 2004). Cities currently using LID to assist in preventing CSOs via
stormwater runoff prevention are San Francisco, Philadelphia, Chicago, Portland
(Oregon), and Seattle.
San Francisco's total land area is approximately 45 square miles of which almost
80% is composed of impervious surfaces, such as streets, sidewalks, rooftops and parking
lots. San Francisco is also one of only two cities in California containing a CSS.
Although their CSS currently meets regulatory requirements, wet weather events produce
occasional CSOs (Kennedy et al, 2008). As of 2008, the San Francisco Public Utilities
Commission (PUC) was developing a long-term sewer system master plan and under that
plan requested an analysis of LID techniques for reducing wet weather flows into their
CSS. The project team responsible for this task analyzed LID techniques for urban areas
and then performed GIS 8 spatial analyses and LID modeling for drawing conclusions.
The team examined several LID practices, including green roofs, porous pavement, street

8

GIS is a geographic information system that captures, stores, analyzes, manages, and presents
data that are linked to location

29

trees and urban forested areas, rainwater harvesting and bioretention methods 9. The LID
model results demonstrated reductions in storm runoff peak flow rates (responsible for
triggering CSOs) for all LID practices modeled (Kennedy et al, 2008).
Philadelphia is 135 square miles and is comprised of 56% impervious surface area
and is also home to CSS that experiences occasional CSO events (Neukrug et al, 2004).
The Philadelphia Water Department (PWD) has standardized LID implementation
throughout the city and LID techniques have been utilized since 2006 to control storm
runoff. Techniques used by the city include green roofs, rain gardens, vegetated swales,
porous pavement, downspout disconnection and rainwater harvesting. The new city
policies promoting LID and green infrastructure have drastically reduced CSO inputs and
have saved the city millions of dollars (U.S. EPA, 2010a).
Chicago covers approximately 228 square miles and contains a CSS as well as
considerable impervious surface areas. The city has adopted a suite of municipal policies
that promote decentralized stormwater management and promote the implementation of
LID techniques in new construction projects (U.S. EPA, 2010a). The LID techniques
utilized in Chicago include green roofs, rain gardens, vegetated swales, porous pavement,
downspout disconnection and rainwater harvesting.
Portland, Oregon spans 125 square miles and as home to a CSS has been a leader
in implementing sustainable stormwater infrastructure in the form of LID techniques for
several decades. The drive to implement LID techniques was generated by the need to
control CSO events and the city has pursued several strategies that promote decentralized
stormwater management approaches (U.S. EPA, 2010a). Portland building codes require
9

Bioretention refers to dispersed, small-scale landscape features designed to attenuate and treat
stormwater runoff, and includes techniques described in 3.2.1 as filtration and infiltration LID
techniques, such as rain gardens, vegetated swales and tree boxes (Kennedy et al, 2008).

30

on-site stormwater management for all new construction projects and LID techniques
used include green roofs, rain gardens, porous pavement, vegetated swales, downspout
disconnection and rainwater harvesting. In the future, about 40% of Portland's total CSO
control will be managed via LID techniques (Hyland and Zuravnsky, 2008).
The city of Seattle, Washington area is 84 square miles of which approximately
one-third is served by a combined sewer system. A majority of Seattle's LID projects are
focused on the concept of a natural drainage system that reduces storm flows through
retention and infiltration (Hyland and Zuravnsky, 2008). Although Seattle is
predominately implementing bioretention and bioinfiltration LID methods, other LID
techniques used by Seattle include green roofs, rain gardens, vegetated swales,
downspout disconnection and rainwater harvesting.

4. RAINWATER HARVESTING
The LID technique that is the focus of this thesis is rainwater harvesting.
Rainwater harvesting is simply the collection, storage and reuse of rainwater. Although
rainwater harvesting literature and research most often focuses on its ability to provide an
alternative source of water, rainwater harvesting is a dynamic tool for stormwater
management. Unlike the other LID techniques, it simultaneously promotes water and
energy conservation while storing stormwater runoff. The focus of this thesis is to
examine the capacity of rainwater harvesting to prevent volumes of stormwater runoff
from entering a sewer system, specifically CSSs. It is important to discuss rainwater
harvesting in depth and the following sections will review several aspects of rainwater

31

harvesting, including a brief introduction in 4.1 and a review of system components in
4.2. Section 4.3 reviews the multiple benefits associated with rainwater harvesting.
Finally, 4.4 briefly presents regulations for rainwater harvesting.

4.1 INTRODUCTION/BACKGROUND
Rainwater harvesting is not a new technique and has been practiced for over 4000
years in cultures throughout the world (Kinkade-Levario, 2007). The development of
centralized water treatment facilities in the early 1900s to address health concerns led to
the discontinuation of rainwater as a primary water source. Rainwater harvesting is
currently gaining momentum as an alternative source of water for many areas, especially
areas such as the Southwest U.S., where water supplies are diminishing. It is important
to provide background information regarding the makeup of a rainwater harvesting
system to fully understand how a system operates, which is presented below.

4.2 RAINWATER HARVESTING SYSTEM COMPONENTS
There are five fundamental components of a rainwater harvesting systems; 1) a
collection surface, 2) a conveyance system, 3) pre-tank treatment 4) water storage, and 5)
distribution (Lawson et al, 2009). Please refer to Figure 4.1 for a generalized illustration
of a rainwater harvesting system. If harvested rainwater is intended for human
consumption, an additional treatment/purification system component would be necessary
(TWDB, 2005). Consideration of component interactions when designing a rainwater
harvesting system can raise system efficiency and reduce economic costs.

The

following sections will describe these categories in more detail.
32

Figure 4.1 Generalized Illustration of a Rainwater Harvesting System

Catchment
Surf ace

Gutter
Downspout

First-f lush diverter
Cistern
Filter & pump shed
Figure available from The Texas Manual on Rainwater Harvesting, TWDB 2005

4.2.1 COLLECTION SURFACE

The collection surface is an instrumental component of a harvesting system. The

most common collection surface for RWH is a rooftop. Generally, desired roofing for
rainwater harvesting is smooth and non-porous, allowing for as much rain collection as
possible. It is also important to note the roof pitch, as steep slopes are most
desirable for rainwater harvesting as transfer of rainwater to conveyance guttering and
downspouts is made simpler. Roofing can consist of several materials, including asphalt
shingles, wood shingles, cement tile, terra cotta tile, metal and membrane (Lawson et al,
2009). However, copper roofs or any roofing with lead components should not be used
for rainwater harvesting if collected water is intended for human consumption (Lawson et
33

al, 2009). Copper and lead roofing often leach into rainwater falling on them and both
lead to serious health risks if consumed over long periods. Popular roofing for RWH are
membrane roofs and coated metal roofs due to high associated runoff coefficients 10
during precipitation events (Lawson et al, 2009).

4.2.2 CONVEYANCE SYSTEM

Gutters and downspouts, also known as the conveyance system, are integral

components of a rainwater harvesting system because they capture rainwater from the
collection system (rooftop) and convey it to the storage tank or cistern. Popular gutter
and downspout materials are galvanized steel, vinyl, half-round PVC, piping, and
seamless aluminum (TWDB, 2005). It is important to properly size gutters and
downspouts for rooftops with multiple pitches and/or connections, as under-sizing can
lead to overflow during high volume precipitation events.

4.2.3 PRE-TANK TREATMENT COMPONENTS

Directly correlated with the conveyance system are pre-tank treatment

components and include leaf screens or leaf guards, first-flush diverters, and pre-tank
filters.
Leaf screens or leaf guards are designed to remove debris from the rainwater after
it is collected from the rooftop and before it enters the storage tank. Leaf screens or leaf
guards cover gutters, generally with fine mesh, and they effectively keep large debris
10

A runoff coefficient is defined as the ratio of runoff to precipitation, meaning the ratio of the
amount of water that is not absorbed by the surface to the total amount of precipitation that falls
during a storm event

34

from entering gutters during storm events. This reduces excess burden on the filtration
system.
During dry periods, rooftops accumulate debris and contaminants. The initial
runoff from a roof surface, called the first-flush, generally rinses the roof, leading to
cleaner water as the rainfall continues. Sediment, metals and bacteria are reduced as a
rain event continues and steady state is generally reached after the first 1 mm of rainfall
(Lawson et al, 2009). Diverting about 20 gallons (approximately 1 mm) of rainwater per
1,000 square feet of collection surface is a general rule of thumb for first-flush diversion
to ensure harvest of only the cleanest rainwater. The first-flush diverter is a device
placed between the gutter and storage tank that discards this initial quantity of rainfall
and prevents accumulated rooftop debris and contaminants from entering the storage
tanks 11. The discarded water should then be diverted to a nearby pervious area, ideally a
form of bioinfiltration, such as a rain garden or vegetated swale. By discarding the firstflush of water, diverters improve water quality, reduce tank maintenance and protect
pumps. Fitting an appropriately sized first-flush diverter is critical to achieving good
water quality.
Organic debris that enters the cistern and is stored results in nutrient buildup and
low oxygen levels due to decomposition (Lawson et al, 2009). Anaerobic conditions lead
to bacterial growth in the tank as well as the potential to develop odors. Pre-tank filters
are designed to filter harvested rainwater through a straining action, usually with screen
or mesh to eliminate contaminants and small debris. A high quality filter also supplies
oxygen to the water during the filtration process (Lawson et al, 2009). The best filter
11

See Appendix B for additional descriptions and/or illustrations of specific rainwater harvesting
components presented in 4.2.3 Pre-tank Treatment Components and 4.2.5 Distribution

35

material is considered stainless steel because it can withstand all weather conditions, does
not rust, keeps its shape, and is self cleaning and drying (Lawson et al, 2009).
Roof washers, a type of pre-tank filter, are placed just ahead of the storage tank
and are designed to filter small debris (TWDB, 2005). Roof washers are comprised of a
30 to 50 gallon capacity tank with leaf strainers and a filter that holds water to be filtered
before being released into the tank 12. Roof washers use filters as fine as 30-microns, a
filter with pores approximately one-third the diameter of a human hair (TWDB, 2005).
There are also filters on the market that perform both the first-flush and filter
straining mechanisms. These modern filters are low-maintenance, needing to be cleaned
only twice annually, are designed to last the lifetime of a building, and are efficient at
collecting more than 90% of filtered water (Lawson et al, 2009). First flush fine filters
are designed to allow the first 5% of rainwater to flow through to be discarded. After
discarding the 5% first flush, the fine filter becomes wet and begins filtering fine
contaminants before diverting water to a storage tank. There are many filter styles and
models to select from, 13 but to ensure filter performance and efficiency, the filter should
be sized according to roof area as many are designed to filter variable square footage.

4.2.4 STORAGE

Cisterns, also referred to as storage tanks, are a central component of rainwater

harvesting systems. They are often the most costly feature of a rainwater harvesting
system and proper sizing is essential to ensure least cost with maximum storage potential,
which will be discussed below. Cisterns can be installed within structures and also above
12
13

See Appendix B for additional description and/or illustration of a roof-washer
See Appendix B for additional rainwater harvesting filter illustrations

36

or below ground, although below ground tanks are generally more expensive to install
due to costs associated with excavation and heavier reinforcement (TWDB, 2005).
Tanks or cisterns can be composed of several materials, however, the four most
common materials used are plastic, metal, concrete and wood and each of these has
advantages and disadvantages. Concrete tanks are very durable while plastic tanks are
lightweight, easy to move and clean, durable, and inexpensive. Metal tanks are easy to
relocate and both metal and wood tanks are aesthetically pleasing. Cistern material
choice will depend on several factors, including geographic location, cistern location
(above ground, below ground, inside building or home), local weather patterns, size, and
budget.
Cistern costs generally increase with increasing volume and like cistern material,
the size is directed by several variables, including local precipitation and precipitation
patterns (i.e. dry periods), desired water use from the system, the collection surface area,
budget and personal preference. It is important to properly size cistern because oversizing a cistern results in underutilized cistern volume, translating to a monetary loss.
Under-sizing a cistern results in decreased storage capacity, frequent tank overflow
(decreased storm runoff control capability), and underutilized water supply from
harvesting rainwater. However, properly sized cisterns reduce capital investment while
ensuring that water demands are met, making rainwater harvesting an environmentally
beneficial and economically viable LID technique.
All cisterns must be equipped with an overflow outlet to prevent backup into
gutters and downspouts when cistern volume has reached capacity. The overflow
diameter should be at least the same diameter as the inlet pipe and a properly designed

37

overflow can also benefit water quality in the cistern (Lawson et al, 2009). Exactly as
with first-flush, cistern overflow should be directed a nearby pervious area, ideally with
infiltration capacity. Cisterns should be placed as close to supply and demand points as
practical to reduce the distance the water is conveyed.

4.2.5 DISTRIBUTION

The distribution (or delivery) system is responsible for transporting harvested

rainwater from the cistern to the point of end-use in the home or building. Components
of the distribution system include calming inlets, pressurized storage tanks, pumps, and
float-filters.
Functional rainwater harvesting cisterns collect small amounts of sediments that
settle to the bottom of the tank, where higher contaminant concentrations are likely to be
found (Gromaire-Mertz et al, 1999). A calming inlet is a device placed at the bottom of
the cistern that directs entering water upward to prevent agitation of the settled sediment
layer at the bottom of the tank 14. The calming inlet allows rainwater to enter at the
bottom of the tank versus the top. The motion of the calming inlet motion also aerates
stored rainwater.
The laws of physics generally require that stored rainwater be distributed with the
assistance of a pump and pressure tank to transport water to its intended end use.
Standard municipal water pressure is between 40 and 60 pounds per square inch (psi)
(TWDB, 2005). Pump systems draw water from the cisterns, pressurize it, and then store
the water in a pressure tank until it is needed (TWDB, 2005). Pump systems are designed

14

See Appendix B for an illustration of a calming inlet

38

to push water rather than pull it, so it is wise to place the pumps at the same level and as
close to the storage tanks as possible. New, on-demand pumps terminate the need for a
pressure tank as well as the associated cost and space because they combine the pump
and pressure tank function into one device that activates in response to demand.
A floating filter is a device that allows harvested rainwater to enter the pumping
system through an elevated uptake point and is attached to the end of the pump's suction
hose to draw water from the tank 15. As mentioned earlier, collected sediment at the
bottom of the cistern has elevated concentrations of pollutants, therefore water should not
be drawn from the bottom. The cistern surface water also contains increased
concentrations of bacteria and also should not be where water is directly drawn from. A
floating filter floats on the surface of the cistern water and allows the pump to draw water
from the calm, clean water that is in the middle of the tank, generally 10 to 16 inches
below the surface.

4.2.6 ADDITIONAL TREATMENT FOR POTABLE WATER USE

Rainwater harvested solely for outdoor uses does not require additional treatment.

However, both potable and non-potable indoor water uses need additional treatment in
the form of sediment filtration and disinfection.
A considerable percentage of contaminants found in rainwater is found bound to
small sediment particles (Gromaire-Mertz et al, 1999). Sediment filters are designed to
remove those remaining small particles, and EPA guidelines for non-potable indoor uses
require a sediment filter of 5 microns or less (Kloss, 2008).

15

See Appendix B for an illustration of a floating filter

39

There are several forms of disinfection for potable rainwater use, including
chlorination, ozonation, reverse osmosis, and ultraviolet light (Lawson et al, 2009).
Disinfection methods are best placed after the pressure tank or on-demand pump
(TWDB, 2005). Chlorination is an inexpensive disinfection method, however is not
always effective at killing all pathogenic contaminants potentially present in harvested
rainwater and alters the taste of the water (TWDB, 2005). Ozonation uses ozone to
destroy bacteria and viruses, however is less effective at eliminating viruses, but does not
leave behind a residual taste. Reverse osmosis mechanically filters bacteria and viruses
rather than destroying them, however reverse osmosis units waste large quantities of
water and is not the most desirable filtration method for a rainwater harvesting system.
The most popular form of filtration is ultraviolet (UV) light. UV light destroys
bacteria and viruses in a non-chemical process, it does not alter the taste of the water and
is inexpensive and low-maintenance (Lawson et al, 2009). UV light is only productive
on clear water, so it is very important to include a sediment filter if utilizing a UV
filtration device. An activated charcoal filter is also recommended for rainwater intended
for potable uses. Proper rainwater harvesting system design ensures harvested rainwater
is a safe water source. Next, 4.3 examines rainwater harvesting regulations.

4.3 REGULATIONS
There are several barriers to promoting the widespread implementation of
rainwater harvesting systems in the U.S. These barriers include lack of regulations and
codes, improper pricing of water, and water right doctrines. Each of these will be
discussed below.
40

Rainwater harvesting is largely excluded within regulations and codes here in the
U.S. and is not addressed in either the Uniform Plumbing Code (UPC) or the
International Plumbing Code (IPC) (Kloss, 2008) 16. When actually incorporated in the
UPC or IPC, guidance for the use of rainwater will be similar to reclaimed water (sewage
that is treated to remove solids and other impurities and is purified to a level suitable for
controlled non-potable applications) and graywater (wastewater generated from domestic
activities such as sinks, showers, and washers that is purified and reused for non-potable
water applications), both of which are included in the IPC and UPC. This lack of public
policy 17 is restrictive to developing rainwater harvesting. Changing our current codes
will be the first step towards widespread rainwater harvesting implementation.
The current price of water here in the U.S. creates little incentive for people to
look at alternatives, with the cost of water ranging from $0.70 to $4.00 per thousand
gallons and the average being $2.00 per thousand gallons (Kloss, 2008). Water demand
is relatively inelastic, so increasing the price of water will not result in an extremely
reduced demand. Implementing full pricing of water (which includes external costs such
as water treatment and distribution) in the price of water for customers will assist in
promoting rainwater harvesting as well as water conservation.
Water rights in the U.S. are dictated by state water laws and consist of two basic
doctrines, the riparian doctrine and prior appropriation doctrine (U.S. GAO, 2003). The
riparian doctrine is common in the Eastern U.S. while the prior appropriation doctrine is
16

The UPC is designated as the American National Standard and is a model code developed to
govern the installation and inspection of plumbing systems as a means of promoting public
health. The IPC is a building code and standard which sets out minimum requirements for
plumbing systems in their design and function, and which sets out rules for the acceptances of
new plumbing-related technologies.
17
Public policies are rules, regulations, laws and codes that are developed to ensure public safety
and economic activity.

41

common in the Western U.S. The riparian doctrine links water rights to land ownership
and states that if water is on someone's property, it is theirs to use and all other
landowners bordering the water will be given equal rights. The prior appropriation
doctrine (often referred to as western water rights) instead links water rights to priority
and beneficial use. Prior appropriation makes harvesting rainwater difficult as it is
capturing rainwater previously appropriated for groundwater recharge and/or surface
water recharge. However, many western states, including Washington, allow rainwater
harvesting because it is a water conservation practice that can reduce the withdrawal and
use of potable water drawn from groundwater and/or surface water sources, therefore
making a greater quantity of water available for all. Furthermore, rainwater reused for
irrigation purposes would simply mimic rainfall. Next, rainwater harvesting benefits are
discussed in depth in 4.4.

4.4 RAINWATER HARVESTING BENEFITS
Rainwater harvesting is a powerful tool that can be used to collectively address
issues related to water and energy conservation as well as stormwater management.
Water shortages are occurring globally and rainwater is being used in many areas as a
clean, reliable water source. Rainwater harvesting not only offsets peak water demand
periods (generally summer months when irrigation demands are highest), but also reduces
high water consumption rates by supplying a safe water source for non-potable water
uses. Rainwater harvesting systems collect and store volumes of rainwater, thereby
providing stormwater runoff control function as well.
Rainwater harvesting has both upstream and downstream benefits. Upstream (at
42

source) benefits include water conservation, energy conservation, and financial savings.
Rainwater harvesting for stormwater management is the downstream (mitigating impact)
benefit. This next section will describe the benefits associated with rainwater harvesting
in more detail.

4.4.1 WATER CONSERVATION

Hoekstra and Hung introduced the water footprint concept in 2002. The water

footprint of a nation is defined as the total volume of freshwater that is used to produce
the goods and services consumed by the population of that nation (Hoekstra and
Chapagain, 2007). Most water use data typically divides consumption information into
three categories: domestic water withdrawals, agricultural sector water withdrawals and
industrial water withdrawals. However, the water footprint concept incorporates internal
and external water footprint analysis within these three categories. An internal water
footprint is the volume of water used from domestic water resources while the external
water footprint is the volume of water used in other countries to produce goods and
services imported and consumed by the people of the country (Hoekstra and Chapagain,
2007).
The size of a water footprint is largely influenced by consumption of food and
other agricultural products. The global water footprint is 1240 m3/capita/year (327,450
gallons/capita/year) average. The U.S. has the largest water footprint, double the global
water footprint at 2480 m3/capita/year (654,900 gallons/capita/year) average, while
China's average water footprint is 702 m3/capita/year (185,380 gallons/capita/year)
(Hoekstra and Chapagain, 2007).
43

Rainwater harvesting is distinct because it is effectually the only stormwater BMP
that can also provide an alternative water supply. As population growth continues, U.S.
regional water authorities and municipalities must anticipate and secure sufficient water
supplies for their customers. All publicly supplied water is treated to potable water
(drinking water) standards. In fact, our water delivery infrastructure requires that we use
water treated to potable standards for all water applications. However, a significant
portion of our per capita water use is applied for non-potable purposes (see Figure 4.2).

Figure 4.2 Percent of Total Water Use that is Non-Potable

Figure adapted from Foraste and Hirschman, 2009

Non-potable water uses include toilet flushing, vehicle washing, laundry washing,
household cleaning and landscape irrigation. Almost 80% of U.S. domestic water
demand does not require potable water (Kloss, 2008). When reviewing U.S. household
water use, examining indoor and outdoor end uses of water is revealing.
A 1999 study of the 1188 homes in 12 diverse study sites to determine residential
(domestic) end uses of water in the U.S. found indoor water use comprises 40.3%
44

and outdoor water use comprises 58.7% of the total determined average 171.8
gallons per day per home water use (see Figure 4.3) (AWWA, 1999). A large portion of
indoor water use is for non-potable applications. A clothes washer and toilet alone
account for approximately 50% of indoor home water use and the bathroom is
consistently the most water consumptive part of a household (see Figure 4.3).

Figure 4.3 U.S. Mean Daily Per Capita Domestic Water Use

Fixture/EndUse
Toilet

Avg. gallons
per capita per
day

Avg. liters
per capita
per day

Indoor
use%

Total
use%

18.5

70.0

30.9%

10.8%

15

56.8

25.1%

8.7%

Shower

11.6

43.9

19.4%

6.8%

Faucet

10.9

41.3

18.2%

6.3%

Other domestic

1.6

6.1

2.7%

0.9%

Bath

1.2

4.5

2.0%

0.7%

1

3.8

1.7%

0.6%

59.8

226.3

100.0%

34.8%

Leak

9.5

36.0

NA

5.5%

Unknown

1.7

6.4

NA

1.0%

Outdoor

100.8

381.5

NA

58.7%

TOTAL

171.8

650.3

NA

100.0%

Clothes washer

Dishwasher
Indoor Total

Figure available from AWWA, 1999

As mentioned above, outdoor water use comprises 58.7% of total average home
water use. However, outdoor water use is more variable as regional locations and
variable property sizes lead to large differences in need for outdoor applications. There is
a direct correlation of lot square footage and the percentage of irrigable landscape to the
quantity of water used outdoors (AWWA, 1999). Outdoor water use is also dictated by
seasonal variability as dominant outdoor water uses are for irrigation, car washing and
45

pool filling. These activities most often occur during warmer and/or drier months, which
are highly variable due to climatic variation within the U.S. These periods are commonly
referred to as peak demand periods and are problematic when partnered with warmer
temperatures and diminishing precipitation events associated with climate change.
Commercial buildings consume a significant portion of our potable water here in
the U.S, estimated at 15 trillion gallons of water (USGBC, 2010). Commercial water use
includes water for motels, hotels, restaurants, office buildings and other commercial
facilities as well as government and public institutions (USGS, 1998). Water use in
commercial buildings varies with plumbing fixture type, equipment installed and building
function. Regardless of these factors, the restroom plumbing fixtures (toilets, urinals,
faucets and showers) account for the majority of building water use (Gilmer and Hughel,
2008). Approximately 60% of total water use in office and administrative buildings is
linked to restroom and plumbing fixtures while heating and cooling systems are
responsible for the remaining 40% (Gilmer and Hughel, 2008).
The majority of water uses, both domestic and commercial, are for applications
that do not require potable water. Rainwater harvesting can supply an alternative nonpotable water source that can relieve demand on potable water sources and ease peak
demand periods. Rainwater harvesting is an alternative non-potable (and potentially
potable) water source that has the ability to relieve demand on potable water supplies,
ease peak demand periods, increase water use efficiency and promote water conservation.
Next, energy conservation as related to rainwater harvesting will be discussed.

46

4.4.2 ENERGY CONSERVATION

Water use and energy use are closely linked. In providing and using water we

consume large quantities of energy. The EPA estimates drinking water and wastewater
services account for an estimated 3% (56 billion kWh) of total national energy
consumption (U.S. EPA, 2010b). Energy is used to withdraw raw water from its source,
treat that water and then distribute it (Foraste and Hirschman, 2009). The end uses of this
water (further treatment, circulation, heating and cooling) also requires additional energy,
however is not included in the drinking and waste water energy consumption estimate
(NRDC, 2004).
Estimates of energy required to perform various functions in the water use cycle
are wide ranging (Griffiths-Sattenspiel and Wilson, 2009). The existing water-energy
nexus research and corresponding literature is focused on California, which transports
water over extremely long distances and will realize immediate energy savings through
water conservation practices. Therefore, energy consumption per unit of publicly
supplied water varies from system to system and region to region as it is dependent on
community size and water use priorities (Foraste and Hirschman, 2009).
The water-energy nexus depends on several factors. First, the drinking water
source and location (energy used for water withdrawals), where more energy is required
as distance from source to end use increases. Second, the drinking water utility
treatment, meaning treatment level and energy used to achieve that level. Third,
community water use (per capita water use and estimated population increases to
determine energy used to treat water volume over time scales). Fourth, drinking water
utility location (centrally located facility or is water distributed over long distances.)

47

Increased transportation to and from drinking water utility equates to increased energy
used. Fifth and sixth respectively are wastewater utility treatment (increasing standards
of treatment require increasing quantities of energy) and wastewater utility location
(centrally located facility or is wastewater transported over long distances). Finally,
energy used in the water sector depends on whether the community energy source is
derived from fossil fuels or renewable energy.
Water supply, use, and disposal are easy to interpret in 3 stages: supply and
conveyance, water treatment, and distribution. Most water used in the U.S. is drawn
from groundwater aquifers or diverted from surface sources, such as rivers, streams or
lakes (NRDC, 2004). Considerable quantities of energy may be needed to create a source
of water and transport it to where it will be treated or consumed. From the total energy
consumed by public water systems, water distribution accounts for 83%, while supply
and conveyance and water treatment account for 10% and 7% respectively (Foraste and
Hirschman, 2009), meaning the majority of energy used by public water systems is used
to distribute water.
Not only is water distribution the most energy consumptive stage of water supply,
use and disposal, but many of the public water supply systems were constructed either
around the turn of the 20th century or shortly after the passage of the Clean Water Act in
1972. These systems are now considered outdated and will need costly updates. Water
supply and treatment infrastructure includes reservoirs, pump stations, storage tanks,
water treatment plants and piping distribution systems (Foraste and Hirschman, 2009)
and each of these components represent considerable investments. A 2009 EPA report
estimated a $334 billion investment will be needed for repairs and maintenance of

48

municipal water systems for the 20-year period from January 2007 through December
2026 and priority need originates in the transmission and distribution infrastructure (see
Figure 4.4) (U.S. EPA, 2009). As rainwater harvesting is a decentralized source of water
supply (available on-site), energy consumption associated with water treatment and
distribution for municipally supplied water is eliminated for the duration stored rainwater
is used.

Figure 4.4 20-Year Monetary Need for Infrastructure Updates by Water Sector (in 2007
dollars)

Available from EPA Report 816-R-09-001, 2009

A large, centralized water system that fails can have a greater negative impact on
a community than the failure of small, decentralized systems (Villarreal and Dixon,
2004), such as rainwater harvesting systems. As the water transportation sector
represents the category containing the largest need for update and investment, on-site
rainwater harvesting also mitigates the need to perform costly updates to our water
supply and treatment infrastructure if widespread implementation occurs.
The 56 billion kWh of energy accounted for in the water and wastewater utilities
49

produces 45 million tons of greenhouse gases (U.S. EPA, 2010b). A highly generalized
estimate using the average mix of energy sources results in 1 kWh of energy consumption
relative to water infrastructure producing 1.6 pounds of greenhouse gas emissions (U.S.
EPA, 2010b). This generalized estimate leads to the conclusion that each U.S. resident is
responsible for consuming 182 kWh annually for drinking and wastewater needs,
generating approximately 293 pounds of GHG annually. The EPA estimates that $400
million savings nationwide can be realized annually if the water and wastewater utility
sector reduces energy use by 10% (U.S. EPA, 2008b). Decreasing potable water demand
by 1 million gallons can result in reducing electricity use by nearly 1500 kWh (Kloss,
2008), preventing approximately 2400 pounds of GHG emissions.
A rainwater harvesting system installed at the King Street Center in Seattle,
Washington collects rainwater harvested from the rooftop into three 5400 gallon cisterns.
The collected rainwater in this building saves approximately 1.6 million gallons of
potable water annually by using collected rainwater for flushing toilets and irrigation
purposes (Kinkade-Levario, 2007). One building can make a considerable impact in
reducing potable water need and the associated embedded energy in the water
transportation sector. Not only do rainwater harvesting systems conserve water and
energy, they also represent long-term financial investments, which will be discussed next.

4.4.3 FINANCIAL

There is a variety of different rainwater harvesting systems and installation costs

for a system are variable. Despite this variety, systems are most economically viable and
efficient when incorporated into new construction rather than retrofitting onto an existing
50

structure (Kinkade-Levario, 2007).
Installing a rainwater harvesting system is a single investment that pays back over
its lifetime. Many communities charge increasing (and often substantial) storm runoff
fees associated with higher proportions of impervious surfaces on land parcels (see
Figure 4.5) (SPU, 2010; City of Tacoma, 2010). These fees can often be discounted by
incorporating an approved stormwater BMP, such as a rainwater harvesting system.

Figure 4.5 Example storm runoff fees

Not only are storm runoff fees substantial, but water rates across the nation are
rising and are projected to increase exponentially over time with population growth (see
Figure 4.6) (SPU, 2010; City of Tacoma, 2010). Owners of rainwater harvesting systems
that utilize stored water will realize immediate and direct financial benefits through
51

decreased purchases of municipal water. The payback period for a harvesting system is
dependent on system cost as well as discounted site storm runoff fees and decreased
municipal water use.

Figure 4.6 Example Water Rate Increases 18

Seattle Municipal Water Rate
$5.00
$4.50
$4.00
$3.50
$3.00
$2.50
$2.00
$1.50
$1.00
$0.50
$0.00

Municipal Water Rate
Linear (Municipal
Water Rate)

1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
2015
2017
2019

per ccf (748 gallons)

An example is a rainwater harvesting system installed on Eggleston Laundry in
Virginia in 2001. The 34,000 square foot facility paid $30,000 to retrofit for 20,000
gallons storage capacity. The initial predicted return-on-investment was 3 years but after
replacing the need for a substantial quantity of municipal water, the payback period was
reduced to 12 months (Kinkade-Levario, 2007).

18

Line breaks due to water rates for all prior years unavailable

52

4.4.4 STORMWATER MANAGEMENT

By collecting and storing volumes of stormwater, a RWH system inherently

reduces the volume of stormwater runoff entering a community's storm system. RWH
systems designed for large-scale commercial buildings that are highly occupied, such as
schools, hospitals and office buildings, have the ability to effectively store large volumes
of stormwater that will be utilized on a regular basis, maintaining available storage. The
goal of this thesis is to clarify areas of uncertainty regarding the potential of RWH to
assist in SWM goals by modeling the performance of RWH systems during high volume
precipitation events in Olympia, Washington. It is important to discuss below why
Olympia was selected as the case study location.

5. CASE STUDY SELECTION
The city of Olympia is Washington’s state capitol and is located within Thurston
County. Olympia is located on the southernmost point of Puget Sound and borders Budd
Inlet (see Figure 5.1). Although smaller than its Puget Sound counterparts Seattle and
Tacoma, Olympia comprises 16.71 square miles and is home to just over 40,000
residents. Because Olympia is the capitol of Washington, economic activity is primarily
derived from state government activity, which provides a stable work force, an engaged
and educated community, and a well-supported school system (City of Olympia, 2010a).
Olympia was selected as the case study location for five reasons that will be
discussed in the upcoming sections. 5.1 will discuss Olympia’s location in that it borders
the diverse Puget Sound ecosystem. 5.2 will discuss Olympia land development and

53

Figure 5.1 Olympia Location

impervious surface coverage. 5.3 will discuss Olympia’s precipitation, as high intensity
precipitation events are likely while 5.4 will discuss Olympia’s sewerage as the city is
home to a CSS. Finally, 5.5 will briefly discuss Olympia’s role as the state capitol, as
there are several large scale government and commercial buildings within the CSS area
that remain highly occupied during the week, making RWH system implementation a
54

very attractive option. I will discuss each of the above reasons in detail below.

5.1 PUGET SOUND
Puget Sound is a unique and valuable ecosystem located in Western Washington
(see Figure 5.2). Retreating glaciers carved Puget Sound 11,000 to 15,000 years ago at
the end of the last ice age (Puget Sound Partnership, 2010). It is the second largest
estuary in the U.S., with over 3000 kilometers of shoreline and fjord-like geomorphology

Figure 5.2 Puget Sound

55

is unique in the U.S.
Like many coastal ecosystems worldwide, Puget Sound is showing evidence of
degradation (Puget Sound Partnership, 2010). Coastal ecosystems both globally and
locally are experiencing trends including increasing numbers of imperiled species,
disrupted food webs, degraded and/or loss of habitat for many species, and increasing
levels of toxic contaminants (Puget Sound Partnership, 2010).
In 2007, the Washington State Legislature enacted Engrossed Substitute House
Bill 5372 with the stated goal of restoring the health of Puget Sound by 2020. The
responsibility of overseeing and implementing this legislation was given to a newly
created state agency, the Puget Sound Partnership. The agency goal is to have Puget
Sound considered a "healthy" ecosystem, a self-sustaining system that supports human
societies by providing goods and services in the form of energy, food, building materials,
water purification, flood and erosion control, as well as providing spiritual enrichment,
recreation and aesthetic experiences (Puget Sound Partnership, 2010). Washington is
dependent upon Puget Sound to provide goods and services to maintain a high quality of
life for residents.
The Puget Sound Partnership is founded on four fundamental beliefs. First, Puget
Sound is a national treasure and the life-blood of Washington State. Second, the Puget
Sound ecosystem is in serious decline and will likely worsen through time. Third, current
activities to protect and restore the Puget Sound ecosystem are fragmented,
uncoordinated, and mostly ineffective at the ecosystem scale and fourth, Puget Sound is
worth protecting and restoring. The agency further emphasizes that the need for action is
urgent as population growth, climate change, and other forces are fundamentally altering

56

Puget Sound (Puget Sound Partnership, 2010).
The Puget Sound marine nearshore is at the nexus of the aquatic and terrestrial
environments and provides habitat for many species. Many of these species are
economically important, such as sea grasses, kelp, and forage fish (Puget Sound
Partnership, 2010). The nearshore habitat is extremely important in maintaining
ecosystem function in Puget Sound because it is created and maintained by processes
involving transfers of sediment, nutrients, water and other constituents. It is also the
location where substantial human development has occurred.
As mentioned above, Olympia is located on the southernmost point of Puget
Sound and borders Budd Inlet. Olympia was established in 1850 and expeditiously
developed around the waterfront in the mid-1850s and quickly became a central point for
maritime commerce (City of Olympia, 2010c). Budd Inlet was historically a favorite
shellfish gathering site for many coastal Native American Salish tribes before European
settlement (City of Olympia, 2010a). However, shellfish harvesting for human
consumption is currently not recommended because of degraded water quality 19.
Recently, Olympia's growth and subsequent urbanization has aggravated stressors
placed on the natural hydrologic system. The next section will discuss Olympia's
population growth and link to urban development and increased impervious surfaces.

5.2 OLYMPIA DEVELOPMENT AND IMPERVIOUS SURFACES
Olympia is a naturally aqueous environment that historically contained extensive
wetlands and forests. Over the last 150 years, human occupation and activity has
19

A primary reason behind degraded water quality of Budd Inlet is due to historical sewer system
outfall and will be discussed below in 5.4

57

disturbed the intricate hydrologic cycle that sustains Olympia's natural ecosystem
processes (City of Olympia, 2003).
Olympia contains 1.81 square miles of water and Olympia's surface water consists
of eight major streams totaling 16 miles in length, four major lakes comprising
approximately 430 acres within the city, and about 6 miles of Budd Inlet shoreline (City
of Olympia, 2003). These waters along with adjacent landscapes are home to a diverse
population of aquatic and terrestrial species that are impacted by anthropogenic activities.
The major trends affecting Olympia's aquatic ecosystems include forest removal,
wetland filling, covering land with impervious surfaces, routing creeks below-ground,
and allowing toxic substances to multiply, all of which are associated with human
activity. The results of these activities include water quality degradation, variable
seasonal high and low flows, flooding, streambed alteration and habitat loss, erosion,
stream-channel widening, and impediment to fish migration and spawning (City of
Olympia, 2003).
As discussed earlier in 2.1, impervious surfaces degrade water quality by
replacing natural land covers with new surfaces such as buildings, streets, parking lots,
driveways and sidewalks (City of Olympia, 1995). These surfaces do not allow rainwater
to infiltrate into the soil, contain nonpoint source pollutants that are collected by storm
runoff, and increases the amount of polluted storm runoff entering a watershed's 20 surface
water. As little as 10% impervious surface coverage within a watershed has been shown
to degrade water quality (City of Olympia, 2003).
The City of Olympia states that urban growth is surpassing science and
20

A watershed is a region of land within which water flows down to a specified receiving body,
such as a stream, river, lake or ocean

58

technology's ability to provide solutions for complicated water resource problems
characteristic within the urban locale (City of Olympia, 2003). Olympia has experienced
one of the highest growth rates in the nation between 1970 and 2000 (City of Olympia,
2003), growing from 23,111 residents in 1970 to 42,000 residents by 2000 (TRPC, 2009).
This large population growth period contributed to Olympia's impervious surface area.
Olympia is estimated to reach a population of over 79,000 by the year 2025, indicating
the importance of protecting water quality from polluted storm runoff generated from
impervious surfaces.
Reducing impervious surfaces and mitigating the impacts from impervious
surfaces for the purpose of maintaining water quality is a top priority for Olympia
because surface water provides 80% of the City's drinking water while the other 20% is
drawn from groundwater supplies (City of Olympia, 2010b). By not allowing rainfall to
infiltrate into the soil and generating storm runoff, impervious surfaces pollute surface
water and deplete groundwater supplies that Olympia depends on for drinking water (City
of Olympia, 1995).
Olympia is composed of several drainage basins (see Figure 5.3). As little as 10%
impervious surface cover in watersheds/drainage basins has been shown to degrade the
health of the aquatic ecosystem. Downtown Olympia houses the most urbanized
drainage basins (shown in red) and includes Moxlie Creek basin, containing 47.5%
impervious surface cover and Capitol Lake basin, which contains 41.1% impervious
cover (TRPC, 2003). Indian, Mission, and Percival Creek basins all contain
approximately 25% impervious surface cover, all of which are well above the threshold
of cover that degrades aquatic ecosystem health.

59

Although development and increased impervious surfaces are problematic for any
community's watershed health, they are most troublesome for communities that receive
high intensity precipitation events, therefore generating large volumes of storm runoff.
The next section will discuss Olympia's precipitation patterns to understand the City’s
storm and surface water patterns.

Figure 5.3 Olympia Drainage Basins 21

>40%
impervious
surface cover
= 25%
impervious
surface cover
<25%
=
impervious
surface cover

=

21

The drainage basins denoted LOTT could not be located within Thurston County data so the
presumption is that all water falling in the LOTT basin goes directly to Budd Inlet Treatment
Facility, which will be discussed in depth later in 5.4. City and county drainage basin maps
indicate that Capitol and Moxlie basins touch boundaries and absorb the mapped LOTT basin and
the same for Mission and East Bay basins

60

5.3 OLYMPIA STORM AND SURFACE WATER
Like most of western Washington, Olympia experiences mild climatic conditions,
including sunny summers and wet winters. Olympia receives an average of 51 inches
annual rainfall, and while less than other U.S. city's average annual rainfall (i.e. 67 inches
in New Orleans, 63 inches in Atlanta and 53 inches in Houston), the rainfall tends to be
dispersed over extended temporal periods unlike the other mentioned regions (TRPC,
2009). Olympia is subject to experience high intensity precipitation events (see Figure
5.4) (note the storms where greater than 4 inches of precipitation fell in a single day).
Although these events occur only occasionally, climate change is expected to alter
precipitation patterns for Olympia (City of Olympia, 2007e). Only one of these very
intense precipitation events took place between 1955 and 1990, but between 1990 and
2008, three have occurred.

Figure 5.4 Olympia Rainfall 1955 through 2007 22

>4” daily rainfall

22

This figure represents Olympia rainfall with a focus on events where over 4" of rain fell in a
single day (potential CSO trigger precipitation), thus the focal point is the occurrence of red
circles. Please see Appendix D for a larger, more readable precipitation graph.

61

Climate change is expected to alter Pacific Northwest precipitation by producing
a moderate increase from November through January (Salathe et al, 2008) and Olympia
specifically will experience increased precipitation over winter months (City of Olympia,
2007e). A statistically significant increase of extreme precipitation events is also
predicted to occur (Tebaldi et al, 2006). Warmer ocean temperatures will evaporate
greater quantities of moisture and warmer air temperatures are capable of holding more
moisture. When this moister air moves over land, more intense precipitation is produced
(Tebaldi et al, 2006).
Olympia is likely to experience increased potential for high intensity precipitation
events. These events are most problematic in urban areas containing high proportions of
impervious surfaces that are also served by a CSS. Olympia is partially served by a CSS
and the next section will discuss Olympia's storm and sewer system in detail.

5.4 OLYMPIA SEWER AND STORM SYSTEM
After Olympia's establishment in 1850, the first permanent sewers were installed
in 1892 (City of Olympia, 2007d). These first sewer lines were generally short pipes
flowing directly into Budd Inlet or Deschutes Waterway that with little planning were
extended as the City developed. The Budd Inlet Treatment Plant was originally built in
1949 to manage Olympia's wastewater needs and the plant's outfall was discharged
directly into Budd Inlet. Until the mid-1950's, Olympia's sewer lines were combined,
carrying both sanitary and storm flows in single pipes that discharged into Budd Inlet. In
1955, the City mandated that future sewer systems in Olympia be separate systems,
62

disconnecting sanitary from storm flows.
Plant discharge has gone through various levels of treatment over the years due to
impacts on Budd Inlet aquatic ecosystem health and is monitored by the Washington
State Department of Ecology under the NPDES Permit Program.
Primary sewage treatment removes larger floating objects through screening and
sedimentation and removes approximately half of suspended solids and significantly
reduces biological oxygen demand (BOD). In the early 1950s the Plant treated
wastewater at the primary level before releasing outfall into Budd Inlet (LOTT, 2010a).
Primary treatment alone is not considered adequate for the protection of the environment
or people's health.
Budd Inlet Treatment Facility was upgraded in 1985 to secondary treatment,
which relies on processes similar to natural biological decomposition and removes over
90% of suspended solids and BODs. However, this level of treatment does not remove
viruses, heavy metals, dissolved minerals or certain chemicals.
In 1994, upgrades to tertiary treatment processes (advanced level of treatment
removing approximately 99% suspended solids and BOD) were incorporated by
integrating nitrogen removal and ultraviolet disinfection. In 2004, the Budd Inlet
Treatment Plant was further upgraded with the addition of a new Class A reclaimed
water 23 sand filter system.
The Budd Inlet Treatment Plant handles storm and wastewater from Olympia as
well as from the nearby cities of Lacey and Tumwater as well as the rest of Thurston
County, and is owned and operated by the LOTT (Lacey, Olympia, Tumwater, Thurston)
23

Reclaimed water is former sewage that is treated to remove solids and impurities and purified
to a level suitable for further use, such as sustainable landscape irrigation, to recharge
groundwater aquifers, or discharge into surface waters.

63

Alliance. Currently Olympia is primarily served by a conventional sewer system 24 that
covers approximately 18 square miles and consists of over 698,000 feet of sewer pipe
(LOTT, 2005). Although Olympia's sewer system is primarily a separate sewer system,
approximately 600 acres of the downtown area is served by a combined sewer system
(see Figure 5.5) (LOTT, 2005).

Figure 5.5 Olympia Sewer Lines

24

A conventional sewer system is a system that collects municipal wastewater in gravity sewers
and conveys it to a central treatment facility before discharge into receiving waters

64

5.5 COMBINED SEWER OVERFLOWS IN OLYMPIA
As discussed above, the most concentrated impervious surfaces are within the
Moxlie Creek and Capitol Lake drainage basins, followed by Indian, Mission, and
Percival Creek basins. Olympia’s combined sewer lines predominately reside within the
Moxlie Creek and Capitol drainage basins, and overlap the East Bay and Indian Creek
basins as well (see Figure 5.6) 25.

Figure 5.6 Olympia Drainage Basins and Combined sewer lines

>40%
impervious
surface cover
= 25%
impervious
surface cover
<25%
=
impervious
surface cover
= Combined sewer
lines

=

25

As mentioned in footnote 21, the drainage basins denoted LOTT cannot be located within
Thurston County or City of Olympia data or maps so the presumption is that all water falling in
the LOTT basin goes directly to Budd Inlet Treatment Facility. Careful examination of maps
indicates majority of LOTT basin is essentially Capitol and Moxlie basins

65

This means that the area most at risk for generating large volumes of storm flows to be
sent to Budd Inlet Treatment Plant is also the area where there is the least amount of
pervious surface to capture and assimilate precipitation.
About 10 to 12 million gallons of wastewater flow through the Plant on an
average day, however, flows have averaged as high as 22.3 million gallons per day (mgd)
during the wettest months. Under normal conditions, treated wastewater is released into
Budd Inlet via the North Outfall from the Plant (see Figure 5.7). However, when
experiencing high intensity precipitation events, the Budd Inlet Treatment Plant is subject
to flow spikes due to the combined sewer area in older downtown Olympia.
Approximately 53% of all inflow and infiltration is estimated to come from the combined
sewer system area (LOTT, 2005).
A CSO event occurs at the plant when the facility is overwhelmed with excess
volume of storm and sewer water (specifically when the equalization basins are full and
influent pumps are at capacity 26) and wastewater is discharged to the emergency
Fiddlehead outfall location (see Figure 5.7) (LOTT, 2005). LOTT has experienced only
two CSO events between 1991 and 2009, which took place on December 3, 2007 and
January 7-8, 2009. Both CSO events occurred due to high intensity precipitation events
in which greater than 5 inches of rain fell in a 24-hr period (LOTT, 2009).
My analysis focused on the December 2007 CSO event as it was the first to occur
in 16 years. Between 1:00 pm December 2 and 1:00 pm December 3, 5.5 inches of rain
fell in Olympia and triggered the event that took place early morning, December 3. Of
the 57.5 million gallons of flow that entered the plant for the 24-hour period, 45.74
26

Equalization basins are used to store peak wastewater flows for later treatment and influent
pumps are required to move wastewater to higher elevations (when gravity alone cannot move
wastewater).

66

Figure 5.7 Budd Inlet Treatment Plant Outfall Locations

Regular
discharge
location

Emergency
discharge
location

million gallons were fully treated, 2.75 million gallons were partially treated, and 9.01
million gallons were released untreated over an 8-hour period (see Figure 5.8).
The Plant experienced another CSO event January 7-8, 2009 as a result of 5.06
inches of rain that fell between 12:00 am January 7 and 12:00 pm January 8 (LOTT,
2009). The Plant was was forced to divert 1.5 million gallons of screened effluent

67

Figure 5.8 2007 CSO Event

12/2/2007 - 12/3/2007 Precipitation
1
0.9

Inches per hour

0.8

December 3

December 2

CSO Event over 8-hr period

0.7
0.6

5.52" Precipitation over 24-hr period

0.5
0.4
0.3
0.2
0.1
0

around the secondary treatment process and blend it with fully treated final effluent.
Following the initial diversion, 6.3 million gallons of the blended, disinfected effluent
was discharged through the emergency Fiddlehead outfall location. Unlike the 2007
CSO event, the outfall released for this event was all at least partially treated.
LOTT only experiences CSO events during high intensity precipitation events,
therefore, reducing inflow into the combined sewer lines in downtown Olympia
generated during high intensity precipitation events will relieve pressure on the Budd
Inlet Treatment Plant and is a primary objective for preventing CSO events.
The goal of this thesis was to investigate if rainwater harvesting can reduce the
volume of stormwater runoff that triggers combined sewer overflow (CSO) events in
Olympia. By modeling rainwater harvesting systems onto large-scale buildings served by
the combined sewer lines and using the 2007 CSO event precipitation, the additional
storm flow storage provided by the harvesting systems during high intensity precipitation

68

events could be evaluated. The next section will present my methods for analysis.

6. METHODS AND ANALYSIS
I predominately used Geographic Information Systems (GIS) and Microsoft Excel
programs to perform my analysis. My methodology included using GIS for a visual
analysis and then creating an Excel simulation model to evaluate daily cistern levels on
large-scale buildings to determine storage capacity for the 2007 CSO event.

6.1 GIS ANALYSIS
I first mapped Olympia's combined sewer lines with data received by Whitney
Bowerman with the City of Olympia (see Figure 6.1), and added a building layer from
data available in Evergreen's GIS database to evaluate building vicinities to combined
sewer lines (see Figure 6.2).
I then wanted to select large-scale buildings for modeling rainwater harvesting
systems, as the larger the roof area, the more potential for runoff prevention. I used the
building layer and separated all buildings into four size classes by manually adjusting the
properties of the building layer (see Figure 6.3). All buildings smaller than 10,000 ft2
remained colored lavender.

69

Figure 6.1 Olympia Combined Sewer Lines

70

Figure 6.2 Combined Sewer Lines and All Buildings

71

Figure 6.3 Selected Large-Scale Buildings

72

I needed a way to estimate water uses for the buildings. I received a parcel
shapefile from Tyle Zuchowski at LOTT which contains water use information for each
parcel in Olympia (see Figure 6.4).

Figure 6.4 Olympia Water Use Parcels

73

The water use provided in the shapefile is the 2008-2009 average gallons per
month water use for the winter months November, December, January and February. I
made the assumption that parcel water use was generated by the building or buildings that
lie within those parcels. I could then use the water use provided in the parcel shapefile to
determine a building's water use. The higher a buildings daily water use, the more
storage is made available for capturing runoff each day.
In order to find the most effective buildings for rainwater harvesting system
placement, I only wanted to look at parcels that contained selected buildings that were
served by the combined sewer lines. I first selected buildings that were within 125 feet of
the combined sewer lines and then selected parcels that were within 125 feet of the
combined sewer lines, but that selected all parcels along the combined lines and I only
wanted parcels that contained selected buildings. I then reselected parcels that were
within 50 feet of the selected buildings to narrow the parcel selection and Figure 6.5
displays the end result.
I determined an estimated water use for each selected building served by the
combined sewer lines. For a simple analysis, I only needed to locate the identification
number (ID) of each selected building, find the ID of the parcel that contained the
building, and then look at the specific water use of that parcel. The easiest way to do this
was to label the parcel IDs and building IDs and reference the attribute tables of each to
find the building's roof area and the parcel's water use, which I would then record in an
Excel spreadsheet, see Figure 6.6 for an example.

74

Figure 6.5 Selecting Buildings and Parcels Served by the Combined Sewer Lines

75

Figure 6.6 One Building per One Parcel

76

However, I quickly realized that many of my selected buildings did not just reside
within one parcel (See Appendix C for building challenge example images). Some
buildings overlapped several parcels. When this occurred, I recorded all parcels the
building overlapped and summed each parcel’s water use to derive a total water use
estimate for the building. Others had multiple selected buildings within a single parcel.
In a case like this, all three buildings are contributing to parcel water use, so the roof area
of all buildings within the parcel were summed to derive a total roof area for analysis.
Other buildings were within a parcel that also contained several very small
parcels. In this case, the small parcel water uses were located in the parcel attribute table
and were added to total building water use. If my selected building slightly overlapped a
parcel that contained a significantly sized unselected building, I eliminated that parcel
from the selected building's water use, as I determined that the unselected building was
likely the only user of water for that parcel.
Other selected buildings equally overlapped more than one parcel where one of
those parcels included a substantially sized unselected building, but had less than 10,000
ft2 roof area. In this case, I labeled the building areas and if the building was larger than
1000 ft2, concluded that the smaller building contributed to water use on that shared
parcel. Therefore, I added the square footage of the unselected building to the square
footage of the selected building to account for the additional roof area during my
analysis. Another case was a selected building that overlapped a parcel that also
contained a smaller, unselected building. If the unselected building was smaller than
1000 ft2, the building was determined to not contribute to parcel water use and square
footage was not added to selected building square footage.

77

6.2 SPREADSHEET SIMULATION MODEL
After performing the above steps, I then had a spreadsheet containing all selected
buildings, the parcels they overlap, the parcel water uses, and the parcel use itself. If the
building water use was less than 75 gallons per day water use, the building was not
selected for analysis. I also had to physically visit some buildings to determine if they
used water and also to verify if some buildings still existed (because the layer was created
in 2008). If there were issues with certain buildings, they were not selected for analysis.
After those eliminations, a total of 102 buildings were selected for analysis. Most
buildings were sized between 10,000 and 25,000 ft2 (see Figure 6.7).

Figure 6.7 Analyzed Building Sizes

Analyzed Building Sizes
1
4

= 10,000-25,000 ft2

27

= 25,000-50,000 ft2
68

= 50,000-100,000 ft2
= 100,000-300,000 ft2

I examined parcel uses and determined the building use from that information. It
was important for me to analyze building uses to see what kind of proportion of non-

78

potable water use the building might be using as many commercial buildings have a very
high proportion of non-potable water use. General merchandise retail, followed by
government and mixed use accounted for almost half of all the analyzed buildings (see
Figure 6.8).

Figure 6.8 Analyzed Building Uses

Analyzed Building Uses
1%

3%

3%

1%

1%
Retail-General Merchandise
Government

3%

Mixed Use

21%

Professional Service

3%

Business Services

4%

Retail-Food
Public Assembly

6%

5+ Units
15%

7%

Recreational
Financial Service
Hotel-Motel
Education

9%
12%
11%

Repair Service
Retail-Auto
Transportation

I then examined the roof runoff produced from each building. I used daily
precipitation from October 31, 2007 through January 30, 2008. Approximately 0.62
gallons per square foot of collection surface per inch of rainfall can be collected for
79

rainwater harvesting. Therefore, all precipitation was converted to roof runoff using this
coefficient. The value then was runoff per square foot, so that was multiplied by the roof
area of the building. The result was the total daily roof runoff produced from that
building, and this was done for each selected building, see Figure 6.9 for the process.

Figure 6.9 Daily Total Runoff Calculations

Precipitation
(inches/day)

*

0.62 gallons/
inch rainfall/ft2

*

Roof Area
(ft2)

=

Daily total runoff

I then used the daily total roof runoff values to determine an appropriate cistern
size for each building. To do this, I generated bin values representing a range of cistern
volumes. The bin sizes used were in 5000 gallon increments. I created a histogram
based on the daily roof runoff values and the bin sizes. I then determined the cistern size
by looking at where the numbers significantly decreased within the bins, see Figure 6.10
on the next page for an example.
For this example, the roof runoff produced dropped off after 15,000 gallons so I
sized the cistern at 15,000 gallons. Although I could have sized the cistern much larger
for runoff storage purposes, the general rule is that cisterns cost approximately one dollar

80

Figure 6.10 Cistern Sizing Analysis

Building 7491
60

Frequency

50
40

15,000 gallon cistern
selected for this example

30
20
10
0
0

5000

10000

15000

20000

25000

30000

35000

40000

More

Cistern Volumes

per gallon of storage, so it was important for me to keep the costs accurate and as low as
possible while still providing storage for storm runoff. Most cisterns ended up providing
between 15,000 and 25,000 gallon storage capacities, although a few of the buildings
with larger roof areas were given 30,000 gallon cisterns (see Figure 6.11).

Figure 6.11 Analyzed Building Modeled Cistern Sizes

Cistern Sizes for All Buildings

= 15,000 gallon cistern
7
13
40

39

= 20,000 gallon cistern
= 25,000 gallon cistern
= 30,000 gallon cistern

81

I next determined daily cistern volumes (or levels). I used IF, THEN logic
statements in Excel to create an analysis of daily cistern levels to determine daily
capacity available for the time frame between October 31 and January 30 (see Figure
6.12).

Figure 6.12 Daily Cistern Volume Analysis
=

k

j
h
g

i

• g = gallons stored at
beginning of day
• h = daily water use
• i = inches precipitation
• j = gallons roof runoff
• k = cistern capacity
g-h+j = cistern volume
If (g-h+j)>k, then k
If (g-h+j)<0, then 0

As mentioned earlier, the daily precipitation was converted to gallons roof runoff for each
building. The equation to determine daily cistern volume levels is gallons stored at the
beginning of the day (g) minus the daily water use (h) plus the gallons roof runoff (j).
However, this equation as is would generate negative values and values greater than the
cistern storage capacity (k). Negative values occurred if there was no roof runoff and the
cistern volume was at a level less than the daily water use while values greater than
82

cistern storage capacity resulted due to high roof runoff (or j) values. Therefore I created
logic statements that would prevent these values from occurring. This logic equation was
used for each building analysis and generated daily cistern levels for the 3 months I
examined.
I then determined the volume of water prevented from becoming runoff each day
because it was water stored in the cistern (see Figure 6.13). To do this, another logic

Figure 6.13 Day to Day Volume Stored

j

h
m

• m = cistern volume current
day
• n = cistern volume day
before
• h = daily water use
• j = gallons roof runoff current
day

n

m-n+h = volume stored
If (m-n+h)>j, then j

statement was created. The cistern volume the current day (m) minus the cistern volume
the day before (n) plus the daily water use (h) determines the quantity of roof runoff
stored, thus prevented from becoming runoff sent to the combined sewer lines. However,
if the equation generated a value greater than the gallons roof runoff produced the current
day (j), than the only volume stored can be the gallons of roof runoff generated that day

83

(j). This logic equation was used for each building and generated daily quantities of
gallons roof runoff stored for the period between October 31 and January 30.

7. RESULTS
Approximately 150 acres or 25% of the 600 total acres served by the combined
sewer system are buildings. I analyzed and modeled rainwater harvesting systems onto
only buildings larger than 10,000 ft2 that are served by the combined sewer lines, and the
roof area of those 102 buildings totaled approximately 53 acres. Therefore, I analyzed
approximately 33% of the roof area served by the combined sewer lines and
approximately 33% of the buildings served by the combined sewer system are larger than
10,000 ft2.
A substantial quantity of roof runoff was stored in the modeled cisterns December
2 and 3, 2007 (see Figure 7.1). My model determined that on December 2nd, 1.22
million gallons roof runoff were stored in the modeled cisterns of a total 3.9 million
gallons actual estimated runoff produced from the analyzed buildings, meaning
approximately one-third or 33% of the total roof runoff that occurred on the 53 acres of
analyzed roof area would be stored in the modeled cisterns. On December 3, the model
determined 275,000 gallons of runoff were stored in the modeled cisterns of an actual
estimated total 6.6 million gallons runoff produced from the analyzed roof area. The
modeled quantity stored for December 3 was significantly lower due to most of the
cisterns being full from December 2 precipitation.

84

Figure 7.1 Modeled Cistern Storage 12/2/2007-12/3/2007

Stored vs. Total Roof Runoff
7,000,000
6,000,000

gallons

5,000,000
4,000,000

Stored Runoff

3,000,000

Total Roof Runoff

2,000,000
1,000,000
0
12/2/2007

12/3/2007

However, because the CSO event occurred in the early morning hours of
December 3, I feel that it was likely triggered by precipitation December 2 and
aggravated by continuous heavy precipitation December 3. Therefore, a substantial
quantity of roof runoff would be prevented from entering the combined sewer lines prior
to the CSO event if the modeled cisterns were actualized. Not only would a substantial
quantity of roof runoff be stored in the modeled cisterns, but the early runoff stored on
December 2 most likely contained a larger proportion of nonpoint source pollutants than
the runoff that occurred late December 2 and early December 3 due to little precipitation
occurring the days prior to the CSO event. Thus, by the time the CSO event occurred in
the early morning hours of December 3, most of the runoff would essentially be only
rainfall containing little (if any) nonpoint source pollutants.
85

Another important factor is the modeled cisterns provide a total of 1.9 million
gallons of storage capacity if they are all empty (see Figure 7.2). If cisterns were
implemented, a management option could be to slowly drain cisterns to empty or near
empty to greatly increase storage capacity for predicted high intensity precipitation
events.

Figure 7.2 Modeled Potential Cistern Storage 12/2/2007-12/3/2007

Stored vs. Total Roof Runoff
7,000,000

CSO occurred early morning 12/3

6.6

6,000,000

Total Roof Runoff

5,000,000

gallons

Predicted Stored Runoff

4,000,000

3.9

Stored
Runoff
Max
Cistern
Storage
Max
Cistern
Storage

3,000,000

Total Roof Runoff

2,000,000
1,000,000

1.22
0.275

0
12/2/2007

12/3/2007

My primary focus days were December 2 and December 3, 2007. However, the
spreadsheet also allowed easy calculation of the total quantity of runoff stored in the
modeled cisterns for precipitation that fell on the analyzed buildings October 31, 2007
through January 30, 2008 by summing all analyzed buildings daily volume stored and
then summing all buildings daily total roof runoff (see Figure 7.3). An actual estimated
86

total 34 million gallons of roof runoff would have been generated between October 31
and January 30 for the 53 acres of roof area analyzed. With the inclusion of modeled
cisterns on the buildings, 15.2 million gallons would be stored for later use in the
building. That means almost half of roof runoff from the 102 buildings would be stored
and presumably utilized on a regular basis for the time examined.

Figure 7.3 Modeled Cistern Storage October 31, 2007 through January 30, 2008

Stored vs. Total Roof Runoff
40
35

34

million gallons

30

Total Roof Runoff

25
Predicted Stored Runoff

20
15

Stored Runoff

15.2

Total Roof Runoff

10
5
0
November
January
October
31, 2007Through
– January
30, 2008

8. CONCLUSIONS
The modeled rainwater harvesting systems on the 102 buildings comprising 53
acres of land cover would reduce the volume of stormwater runoff that triggered the CSO
event December 3, thus showing that rainwater harvesting is an effective low impact
87

development solution for providing additional storm runoff storage. The predicted 1.5
million gallons of total runoff stored over December 2 and 3, 2007 was a small
proportion compared to the extreme flows received at the Budd Inlet Treatment Plant, so
I have to conclude the modeled rainwater harvesting systems would not have prevented
the CSO event from occurring. However, the analyzed roof area is only 17% of the total
area served by the combined sewer system, indicating additional storage potential.

8.1 FUTURE PLANS FOR GREATER OLYMPIA
As discussed earlier, mitigating CSOs can require costly infrastructure updates.
Separating combined sewer lines and redirecting stormwater can cost as much as $600
per foot of piping (Kloss and Calarusse, 2006), resulting in multi-million dollar
investments.
In the late 1990s as a result of a major inflow and infiltration study as well as a
newly created wastewater resource management plan, LOTT and the City of Olympia
investigated separation of the downtown combined sewer lines into a separate storm
system, and determined the process as not cost-effective.
To handle increased volumes of wastewater, LOTT is planning "just-in-time"
construction of several satellite reclaimed water 27 facilities based on population and
employment projections to meet future wastewater treatment capacity needs (LOTT,
2010b). Each satellite facility will have the ability to treat one million gallons per day
(MGD) and will be expandable for up to 5 MGD (LOTT, 2003). LOTT's second

27

Reclaimed water is former sewage that is treated to remove solids and impurities and purified
to a level suitable for further use, such as sustainable landscape irrigation, to recharge
groundwater aquifers, or discharge into surface waters

88

reclaimed water facility, the Hawks Prairie Reclaimed Water Satellite, cost
approximately $35 million and provides 2 MGD of treatment capacity.

8.2 RAINWATER HARVESTING IN OLYMPIA
The modeled rainwater harvesting systems could cost as little as $2.5 million and
provide a potential 1.9 million gallons of storage capacity daily (assuming they are all
empty) while also promoting water and energy conservation. As mentioned above,
LOTTs reclaimed water satellite facilities are designed to treat one million gallons per
day at the lowest level. The modeled rainwater harvesting systems with only partial
coverage of the CSS area could handle almost 2 million gallons per day at only a fraction
of the cost. Also, significant financial savings would be observed by LOTT due to
decreased flows to be treated and buildings would save by decreasing municipal water
purchases for non-potable water applications.
There are also still 97 acres of smaller-scale buildings with less than 10,000 ft2
roof area served by the combined sewer lines that were not analyzed. If modeled the
same as the large-scale buildings, they could potentially triple roof runoff storage.

8.3 RESEARCH APPLICATIONS
Most rainwater harvesting models that I have encountered use average monthly
precipitation to determine the proper sizing of cisterns as they are looking at rainwater
harvesting for an alternative water supply. My model is the first that looks at daily
cistern levels for stormwater runoff storage, thus having the ability to analyze cistern
storage capacity for single storm events. The cistern sizes are also adjustable and can be
89

easily increased or decreased for different storage scenarios. My research is also
applicable for any community served by a combined sewer system, and only precipitation
quantities, building areas and water uses would need adjustments.
If I had additional time for further research, I would like to model rainwater
harvesting systems onto all buildings served by the combined sewer system in Olympia
to evaluate total storage potential. In the future, I would like to apply my model to other
communities served by a combined sewer system.

90

BIBLIOGRAPHY

American Waterworks Association (AWWA). 1999. Residential End Uses of Water,
AWWA Research Foundation, Denver, Colorado
Arnold, Chester and C. James Gibbons. 1996. "Impervious surface Coverage-The
Emergence of a Key Environmental Indicator", Journal of the American Planning
Association 62, p. 243-258.
Beach, Dana. 2002. The effects of Urban Design on Aquatic Ecosystems in the United
States, Pew Ocean Commission.
Booth, Derek B. and C. Rhett Jackson. 1997. "Urbanization of Aquatic Systems:
Degradation Thresholds, Stormwater Detection, and the Limits of Mitigation",
Journal of the American Water Resources Association 33, no. 5, p. 1077-1090.
Brattebo, Benjamin O. and Derek B. Booth. 2003. Long-Term Stormwater Quantity and
Quality Performance of Permeable Pavement Systems, Center for Water and
Watershed Studies, University of Washington.
City of Olympia (a). "About Olympia", available from
http://olympiawa.gov/en/community/about-olympia.aspx, accessed 4/15/2010.
City of Olympia (b). "Drinking Water Facts", available from
http://olympiawa.gov/city-utilities/drinking-water/water-quality/, accessed
4/15/2010.
City of Olympia (c). "History of Olympia, Washington", available from
http://olympiawa.gov/community/about-olympia/history-of-olympiawashington.aspx, accessed 4/15/2010.
City of Olympia. 1995. Impervious Surface Reduction Study, City of Olympia, Public
Works, Water Resources Program, Olympia, Washington.
City of Olympia. 2003. Storm and Surface Water Plan-Draft, City of Olympia, Storm
and Surface Water Utility, Olympia, Washington.
City of Olympia (d). 2007. Wastewater Management Plan 2007-2012, City of Olympia,
Public Works, Olympia, Washington.
City of Olympia (e). 2007. Olympia's Response to The Challenge of Climate Change,
City of Olympia, Public Works, Olympia, Washington.
City of Tacoma. "1818-Stormwater Rates", available from
http://www.cityoftacoma.org/Page.aspx?hid=3625, accessed 4/25/2010.

91

Dussaillant, A. R. et al. "Raingardens for stormwater infiltration and focused
groundwater recharge", Water Science and Technology 5, no. 3 (2005): 173-179.
Frazer, Lance. 2005. "Paving Paradise: The Peril of Impervious Surfaces", Environmental
Health Perspectives 113, no.7, p. A457-A462.
Gilmer, Laurie and Gregory Hughel. 2008. "Improving Water Efficiency in Your
Building", Facility Engineering Associates, Volume 16, No. 4.
Griffiths-Sattenspiel, Bevan and Wendy Wilson. 2009. The Carbon Footprint of Water,
River Network.
Gromaire-Mertz, M. C. et al. 1999. "Characterization of urban runoff pollution in Paris",
Water Science and Technology 39, no. 2, p. 1-8.
Hoekstra, A. Y. and A. K. Chapagain. 2007. "Water footprints of nations: Water use by
people as a function of their consumption pattern", Water Resources Management
21, p. 35-48.
Hoekstra, A. Y. and P. Q. Hung. 2002. "A Quantification of Virtual Water Flows
Between Nations in Relation to International Crop Trade", Value of Water
Research Report Series No. 11, UNESCO-IHE Institute for Water Education,
Delft, The Netherlands, http://www.waterfootprint.org/Reports/Report1.pdf
Hyland, Ray and Lauren Zuravnsky. 2008. "Natural Assistance-Green infrastructure
approaches for CSO control in urban areas", Water Environment and Technology
20, no. 4, p. 63-69.
Kennedy, Lori et al. 2008. "Low-Impact Development: San Francisco's green approach to
stormwater management promises to reduce energy use, increase natural habitat,
and enhance the quality of urban life", Water Environment and Technology 20,
no. 4, p. 34-43.
Kinkade-Levario, Heather. 2007. Design for Water: Rainwater Harvesting, Stormwater
Catchment, and Alternative Water Reuse. Gabriola Island, B.C., New Society
Publishers.
Kloss, Christopher. 2008. Managing Wet Weather with Green Infrastructure Municipal
Handbook, Rainwater Harvesting Policies, EPA-833-F-08-010, Low Impact
Development Center. [Cited under U.S. EPA as well]
Kloss, Christopher and Crystal Calarusse. 2006. Rooftops to Rivers: Green Strategies for
Controlling Stormwater and Combined Sewer Overflows, Natural Resource
Defense Council.
LaBadie, Katherine. 2010. Identifying Barriers to Low Impact Development and Green
92

Infrastructure in the Albuquerque Area, Masters Degree Professional Project, The
University of New Mexico.
LaBranche, Adrienne et al. 2007. Virginia Rainwater Harvesting Manual, The Cabell
Brand Center, Salem, VA.
Lawson, Sarah et al. 2009. Virginia Rainwater Harvesting Manual, The Cabell Brand
Center, Salem, Virginia.
Lee, Jun Ho and Ki Woong Bang. 2000. "Characterization of Urban Stormwater Runoff",
Water Resources 34, no. 6, p. 1773-1780.
LOTT Alliance. 2003. Wastewater Resource Management Plan, LOTT Alliance.
LOTT Alliance. 2005. Fact Sheet for NPDES Permit WA0037061-Budd Inlet Wastewater
Treatment Plant, LOTT Alliance.
LOTT Alliance. 2009. 2009 Annual Capacity Report, prepared by Adam Klein and Tyle
Zuchowski.
LOTT Alliance. (a) 2010. "A Brief History", available from
http://www.lottonline.org/history.htm, accessed 5/25/2010.
LOTT Alliance. (b) 2010. "Reclaimed Water Facilities", available from
http://www.lottcleanwater.org/reclaimFacilities.htm, accessed 5/25/2010.
Low Impact Development Center, available from
http://www.lowimpactdevelopment.org, accessed 3/10/2010.
National Oceanic and Atmospheric Administration (NOAA). 1998. Classified Shellfish
Growing Waters, by C.E. Alexandar, NOAAs State of the Coast Report.
Natural Resources Defense Council. Stormwater Strategies-Community Responses to
Runoff Pollution, Chapter 12 Low Impact Development, 2001.
Oberndorfer, Erica et al. 2007. "Green Roofs as Urban Ecosystems: Ecological
Structures, Functions, and Services", BioScience 57, no. 10, p. 823-833.
Prince George's County, Maryland. Department of Environmental Resources (PGDER).
Larry Coffman et al. 1999. Low Impact Development Design Strategies, An
Integrated Design Approach. Published by the U.S. EPA, Washington D.C. EPA841-B-00-003.
Puget Sound Action Team. 2005. Low Impact Development-Technical Guidance Manual
for Puget Sound, Washington State University, Pierce County Extension.

93

Puget Sound Partnership. 2010. Strategic Science Plan-January 2010 Review Draft,
prepared by the Puget Sound Partnership Science Panel, Olympia, Washington.
Rochfort, Q. et al. 2000. "Using Benthic Assessment Techniques to Determine Combined
Sewer Overflow and Stormwater Impacts in the Aquatic Ecosystem", Water
Quality Research Journal of Canada 35, p. 365-397.
Salathe, Eric P. et al. 2008. "A high-resolution climate model for the United States
Pacific Northwest-Mesoscale feedbacks and local responses to climate change
Draft", Journal of Climate, in press.
Seattle Public Utilities (SPU). 2010. Drainage Rates, available from
http://www.cityofseattle.net/util/Services/Drainage_&_Sewer/Rates/DrainageRate
s/index.htm, accessed 4/28/2010.
Schueler, Thomas R. 2000. "The Importance of Imperviousness", Watershed Protection
Techniques 1, no. 3, p. 100-111.
Stone Jr., Brian. 2004. "Paving over paradise: how land use regulations promote
residential imperviousness", Landscape and Urban Planning 69, p. 101-113.
Tebaldi, Claudia et al. 2006. "Going to the Extremes-An Intercomparison of ModelSimulated Historical and Future Changes in Extreme Events", Climate Change
79, p. 185-211.
Texas Water Development Board (TWDB). 2005. The Texas Manual on Rainwater
Harvesting-3rd Edition, Texas Water Development Board, Austin, Texas.
Thurston Regional Planning Council (TRPC). 2009. The Profile, prepared by the
Thurston Regional Planning Council.
University of Florida (UF). Florida Field Guide to Low Impact DevelopmentBioswales/Vegetated Swales, IFAS Extension, 2008.
U.S. EPA. 1994. Combined Sewer Overflow (CSO) Control Policy, EPA-830-94-001,
Office of Water, April 1994.
U.S. EPA. 1995. Combined Sewer Overflows-Guidance for Nine Minimum Controls,
EPA-832-B-95-003, Office of Water.
U.S. EPA. 1996. Nonpoint Source Pollution: The Nation's Largest Water Quality
Problem, EPA-841-F-96-004A, Office of Water.
U.S. EPA. 1999. Combined Sewer Overflow Management Fact Sheet: Sewer Separation,
EPA-832-F-99-041, Office of Water.

94

U.S. EPA. 2001. Report to Congress-Implementation and Enforcement of the Combined
Sewer Overflow Control Policy, EPA-833-R-01-003, Office of Water.
U.S. EPA. 2003. Protecting Water Quality from Urban Runoff, EPA 841-F-03-003.
U.S. EPA. 2004. Report to Congress-Impacts and Controls of CSOs and SSOs,
EPA-833-R-04-001, Office of Water.
U.S. EPA. 2007. Reducing Stormwater Costs through Low Impact Development (LID)
Strategies and Practices, EPA-841-F-07-006, Nonpoint Source Control Branch.
U.S. EPA. 2008a. Clean Watersheds Needs Survey 2004: Report to Congress, Office of
Wastewater Management.
U.S. EPA. 2008b. Ensuring a Sustainable Future-An Energy Management Guidebook
for Water and Wastewater Utilities, Office of Water.
U.S. EPA. 2009. 2006 Community Water System Survey Volume I Overview, EPA815R-09-001, Office of Water.
U.S. EPA. 2010a. “Green Infrastructure Case Studies”, available from
http://cfpub.epa.gov/npdes/greeninfrastructure/gicasestudies.cfm#Municipal,
accessed 2/25/2010.
U.S. EPA. 2010b. "Energy and Water", available from
http://www.epa.gov/waterinfrastructure/bettermanagement_energy.html, accessed
4/20/2010.
U.S. General Accounting Office. 2003. Freshwater Supply- States' Views of How Federal
Agencies Could Help Them Meet the Challenges of Expected Shortages, Report
GAO-03-514.
U.S. Geological Survey. 1998. Estimated Use of Water in the United States in 1995, U.S.
Geological Survey Circular 1200.
Washington State Department of Ecology (WADOE). 2005. Stormwater Management
Manual for Western Washington, Publication Numbers 05-10-029 through 05-10033, Water Quality Program.
Wang, Lizhu et al. 2001. "Impacts of Urbanization on Stream Habitat and Fish Across
Multiple Scales", Environmental Management 28, no. 2, p. 255-266.

95

APPENDIX A-TREATMENT CONTROL METHOD DEFINITIONS (FROM WADOE,
2005)
•

Wetpools: Wetpools provide runoff treatment by allowing settling of particulates
during inactive periods by biological uptake and by vegetative filtration.
Wetpools can be single-purpose facilities that only provide runoff treatment or
they can also be combined with a detention pond or vault to also provide flow
control.

•

Biofiltration: Biofiltration uses vegetation in coordination with slow and shallowdepth flow for runoff treatment. As runoff passes through the vegetation,
pollutants are removed through the combined effects of filtration, infiltration and
settling. These effects are aided by the reduction of the velocity of stormwater as
it passes through the biofilter. Biofiltration facilities include swales designed to
convey and treat concentrated runoff at shallow depths and slow velocities, and
filter strips that are broad areas of vegetation for treating sheet flow runoff.

•

Oil/Water Separation: Oil/water separators remove oil floating on the surface of
water by using gravity to remove surface and dispersed oil. The two general
types of separators include the American Petroleum Institute (API) separators and
coalescing plate (CP) separators.

•

Pretreatment: There are several methods for pretreatment, however, presettling
basins are often used to remove sediment from runoff prior to discharge into other
treatment facilities. Pretreatment often must be provided for filtration and
infiltration facilities to help protect groundwater or to prevent clogging. Devices
can include a pre-settling basin, a wetpond/vault, biofilter, constructed wetland or
96

an oil/water separator.
•

Infiltration: Infiltration refers to the use of the filtration, adsorption, and
biological decomposition properties of soils to remove pollutants. Infiltration can
provide multiple benefits including pollutant removal, peak flow control,
groundwater recharge, and flood control. However, one condition that can limit
the use of infiltration is the potential adverse impact of groundwater quality. To
adequately address the protection of groundwater when evaluating infiltration it is
important to understand the difference between soils that are suitable for runoff
treatment and soils only suitable for flow control. Sufficient organic content and
sorption capacity to remove pollutants must be present for soils to provide runoff
treatment. The use of coarser soils to provide flow control for runoff from
pollutant generating surfaces must always be preceded by treatment to protect
groundwater quality. Thus, there will be instances when soils are suitable for
treatment but not flow control, and vice versa.

•

Filtration: A relatively new application of pollutant removal system for
stormwater is the use of various media such as sand, perlite, zeolite, and carbon,
to remove low levels of total suspended solids (TSS). Specific media such as
activated carbon or zeolite can remove hydrocarbons and soluble metals. Filter
systems can be configured as basins, trenches or the novel cartridges.

•

Emerging Technologies: Emerging technologies are new technologies that have
not been evaluated using approved protocols, but for which preliminary data
indicate that they may provide a desirable level of stormwater pollutant removal.
They have not been evaluated in sufficient detail to be acceptable as stand alone

97

BMPs for general usage in new development or redevelopment situations
requiring Basic Treatment. A few emerging technologies are allowed to help
remove metals, hydrocarbons, and nutrients. Otherwise, their use is restricted in
accordance with their level of development [as explained in Chapter 12]. The
recommendations for these emerging technologies will change as we collect more
data on their performance. Updated recommendations on their use will be posted
on the Ecology website. Meanwhile, emerging technologies can also be used for
retrofit situations.
•

On-line Systems: Most treatment facilities can be designed as "On-line" systems
with flows above the water quality design flow or volume simply passing through
the facility with lesser or no pollutant removal efficiency. However, it is
desirable to restrict flows to treatment facilities and bypass the remaining higher
flows around them. These are called "Off-line" systems. An example of an online system is a wetpool that maintains a permanent pool of water for runoff
treatment purposes.

•

Design Flow: Design criteria for treatment facilities are assigned to achieve the
applicable performance goal at the water quality design flow rate (e.g., 80% TSS
removal)

98

APPENDIX B-ADDITIONAL DESCRIPTIONS AND/OR ILLUSTRATIONS OF SPECIFIC
RAINWATER HARVESTING COMPONENTS
•

First-flush diverter

First-flush diverters come in various designs but all perform the same general
function: to discard the first-flush of water from the rooftop as it is considered the
most contaminated. Below are a series of illustrations depicting the basic funtion
of a first-flush diverter.

Images courtesy of www.rainharvest.com

99

•

Roof washer
Roof washers are designed to filter small debris from the rooftop. Below left is an
illustration of a box style roof washer. They are effective, however, modern
filters are replacing the need for roof washers by combining first-flush and filter
straining mechanisms into one device. Below right compares older, roof washer
style filters with a modern filter.

Image courtesy of TWDB, 2005

•

Image courtesy of LaBranche et al, 2007

Rainwater Harvesting Filters
There are several designs and styles of rainwater harvesting system filters with a
range of purposes (roof area capture abilities) and prices. Below are two images
depicting various filter styles 28. Below those are three illustrations of an example
modern rainwater harvesting filter and what it would look like installed.

28

For above left figure, numbers provide additional information. 1: Incoming rainwater; 2:
Larger debris is removed; 3: Pre-filtered water flows over a secondary fine filter sieve; 4: Cleaned
water flows to the cistern; 5: Dirty water is discarded

100

Image courtesy of www.starkenvironmental.com

Image courtesy of LaBranche et al, 2007

All above images courtesy of www.rainwaterharvestingstore.com/filters

101

•

Calming Inlet
A calming inlet allows incoming rainwater to enter from the bottom of the tank
and directs water upward so as not to disturb fine particles at the cistern bottom.
Below is an illustration of an example calming inlet, however, again there are
many different styles on the market.

Image courtesy of Lawson et al, 2009

102

•

Floating filter

Floating filters are important for drawing the cleanest water from the cistern (generally
from10 to 16 inches below the surface). An air-filled ball suspends the floating inlet
filter and allows for connection of the floating inlet to a pump or section line. Below are
illustrations.

Image courtesy of http://www.beingwater.com/rainwater

Images courtesy of www.starkenvironmental.com

103

APPENDIX C-BUILDING CHALLENGE EXAMPLES
a) One Building-Multiple Parcels

104

b) Multiple Buildings-One Parcel

105

c) Multiple small parcels within one large parcel

106

d) Selected building small parcel overlap

107

e) Shared parcel-significantly sized unselected building

108

f) Shared parcel-insignificantly sized unselected building

109

APPENDIX D- ENLARGED OLYMPIA PRECIPITATION CHART

>4” daily rainfall
110

Media

Boyd_LMESthesis2011.pdf