EPA/600/R-07/033F | February 2008 | www.epa.gov/ncea
United States
Environmental Protection
Agency
A Screening Assessment of the
Potential Impacts of Climate Change
on Combined Sewer Overflow (CSO)
Mitigation in the Great Lakes and
New England Regions
National Center for Environmental Assessment
Office of Research and Development, Washington, DC 20460
-------
EPA/600/R-07/033F
February 2008
A Screening Assessment of the Potential Impacts of Climate
Change on Combined Sewer Overflow Mitigation in the
Great Lakes and New England Regions
Global Change Research Program
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection
Agency policy and approved for publication. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
Preferred Citation:
U.S. Environmental Protection Agency (EPA). (2008) A screening assessment of the potential impacts of climate
change on combined sewer overflow (CSO) mitigation in the Great Lakes and New England regions. Global
Change Research Program, National Center for Environmental Assessment, Washington, DC; EPA/600/R-07/033F.
Available from the National Technical Information Service, Springfield, VA, and online at
http://www.epa.gov/ncea.
ii
-------
TABLE OF CONTENTS
LIST OF TABLES iv
LIST OF FIGURES v
LIST OF ABBREVIATIONS AND ACRONYMS vii
PREFACE viii
AUTHORS ix
1. EXECUTIVE SUMMARY 1
2. INTRODUCTION 5
2.1. COMBINED SEWER SYSTEMS AND COMBINED SEWER OVERFLOWS 6
2.1.1. History 7
2.1.2. Effects on Water Quality and Public Health 8
2.2. CSO CONTROLS 9
2.2.1. Nine Minimum Controls and Long-Term Control Plans 10
2.2.2. CSO Control Policy Mitigation Requirements 10
2.3. STUDY GOALS 11
3. METHODS 13
3.1. CSS SELECTION 13
3.1.1. Great Lakes Region 13
3.1.2. New England Region 14
3.2. PRECIPITATION BENCHMARKING APPROACH 15
4. RESULTS AND DISCUSSION 19
4.1. CHANGES IN CSO EVENT FREQUENCY 19
4.1.1. Great Lakes Region 19
4.1.2. New England Region 20
4.2. POTENTIAL MITIGATION REQUIREMENTS 24
4.2.1. POTENTIAL CHANGES IN BENCHMARK DAILY PRECIPITATION.... 24
4.2.1.1. Great Lakes Region 25
4.2.1.2. New England Region 27
4.2.2. POTENTIAL CHANGES IN SYSTEM CAPACITY 30
4.3. LIMITATIONS AND FUTURE RESEARCH 31
4.3.1. Limitations 32
4.3.2. Future Research 34
5. CONCLUSIONS 36
REFERENCES 38
in
-------
LIST OF TABLES
1. Great Lakes region CSS communities by state 14
2. New England region CSS communities by state 14
3. Regional average percent change in CSO frequency in the Great Lakes region,
2060-2099 19
4. Regional average percent change in CSO frequency in the New England region,
2025-2050 22
5. Regional average percent change in the benchmark daily total precipitation in the
Great Lakes region during the future period 2060-2099 25
6. Regional average percent change in the benchmark daily total precipitation in the
New England region during the future period 2025-2050 27
7. Estimated regional average percent change in runoff volume for the Great Lakes
region for the future period 2060-2099 31
8. Estimated regional average percent change in runoff volume for the New England
region for the future period 2025-2050 31
IV
-------
LIST OF FIGURES
1. Distribution of combined sewer systems in the United States 7
2. Percent change in frequency of CSO events in the Great Lakes region relative to
historical values based on Canadian (CCCM) and Hadley (HADCM2) Model
climate projections for the future period from 2060-2099 (1-day) 20
3. Percent change in frequency of CSO events in the Great Lakes region relative to
historical values based on Canadian (CCCM) and Hadley (HADCM2) Model
climate projections for the future period from 2060-2099 (4-day) 21
4. Cumulative distribution of percent change in CSO frequency in the Great Lakes
region relative to historical values based on future climate projections for the period
2060-2099 21
5. Percent change in frequency of CSO events in the New England region relative to
historical values based on Canadian (CCCM) and Hadley (HADCM2) Model
climate projections for the future period 2025-2050 (1 -day averaging period) 22
6. Percent change in frequency of CSO events in the New England region relative to
historical values based on Canadian (CCCM) and Hadley (HADCM2) Model
climate proj ections for the future period 2025-2050 (4-day averaging period) 23
7. Cumulative distribution of percent change in CSO event frequency in the New
England region relative to historical values based on future climate projections for
the period 2025-2050 23
8. Percent change in the four-event benchmark daily total precipitation in the Great
Lakes region relative to historical values based on Canadian (CCCM) and Hadley
(HADCM2) Model climate projections for the future period 2060-2099 (1-day
averaging period) 26
9. Percent change in the four-event benchmark daily total precipitation in the Great
Lakes region relative to historical values based on Canadian (CCCM) and Hadley
(HADCM2) Model climate projections for the future period 2060-2099 (4-day
averaging period) 26
10. Cumulative distribution of percent change in four-event benchmark daily total
precipitation in the Great Lakes region relative to historical values based on future
climate projections for the period 2060-2099 27
11. Percent change in the four-event benchmark daily total precipitation in the New
England region relative to historical values based on Canadian (CCCM) and Hadley
(HADCM2) Model climate projections for the future period 2025-2050 (1-day
averaging period) 28
-------
LIST OF FIGURES (continued)
12. Percent change in the four-event benchmark daily total precipitation in the New
England region relative to historical values based on Canadian (CCCM) and Hadley
(HADCM2) model climate projections for the future period 2025-2050 (4-day
averaging period) 29
13. Cumulative distribution of percent change in four-event benchmark daily total
precipitation in the New England region relative to historical values based on future
climate projections for the period 2025-2050 29
VI
-------
LIST OF ABBREVIATIONS AND ACRONYMS
AOGCM Atmosphere-Ocean General Circulation Model
CSO Combined Sewer Overflow
CSS Combined Sewer System
IPCC Intergovernmental Panel on Climate Change
LTCP Long-Term Control Plan
NMC Nine Minimum Controls
TMDL Total Maximum Daily Load
VEMAP Vegetation/Ecosystem Modeling and Analysis Project
vn
-------
PREFACE
The U.S. Environmental Protection Agency's Global Change Research Program (GCRP)
is an assessment-oriented program within the Office of Research and Development that focuses
on assessing how potential changes in climate and other global environmental stressors may
impact water quality, air quality, aquatic ecosystems, and human health in the United States.
The Program's focus on water quality is consistent with the Research Strategy of the U.S.
Climate Change Research Program—the federal umbrella organization for climate change
science in the U.S. government—and is responsive to U.S. EPA's mission and responsibilities as
defined by the Clean Water Act and the Safe Drinking Water Act. The GCRP's water quality
assessments also address an important research gap. In the 2001 National Assessment of the
Potential Consequences of Climate Change in the United States (Gleick, 2000), water quality
was addressed only in the context of the health risks associated with contaminated drinking
water. A comprehensive assessment of the potential impacts of global change on water quality
was not included.
This report is a screening-level assessment of the potential implications of future climate
change on combined sewer overflows (CSOs) in the New England and the Great Lakes Regions.
It is not a detailed analysis of individual systems. Rather, the purpose is to determine whether
the potential implications of climate change on CSOs in these regions warrant further
consideration and study. Wastewater treatment infrastructure was identified as a priority concern
because of the essential service provided by these systems to protect public health and
ecosystems. Investments in wastewater treatment infrastructure are also long-term, capital-
intensive, and, in many cases, irreversible in the short- to medium-term. Thus, today's decisions
could influence the ability of treatment facilities to accommodate changes in climate for decades
into the future.
This final report reflects a consideration of comments received on an External Review
Draft dated September 2006 (EPA/600/R-07/033A) provided by an external letter review, and
comments received during a 30-day public review period (March 29, 2007 through April 28,
2007).
Peter Preuss, Ph.D.
Director
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Research Program
Vlll
-------
AUTHORS
The National Center for Environmental Assessment (NCEA), Office of Research and
Development, was responsible for preparing this final report. Analysis and preparation of the
draft report was conducted by ICF International, Inc. under U.S. EPA Contract No. GS-10F-
0124J. The authors are very grateful for the many excellent comments and suggestions provided
through an external letter review and comments received during a 30-day public comment
period.
AUTHORS
U.S. Environmental Protection Agency, National Center for Environmental Assessment,
Global Change Research Program, Washington, DC
John Furlow
Thomas Johnson
Britta Bierwagen
ICF International, Fairfax, VA
J. Randall Freed
Jeremy Sharfenberg
Sarah Shapiro
IX
-------
1. EXECUTIVE SUMMARY
Combined sewer systems (CSSs) collect and co-treat storm water and municipal
wastewater. During high intensity rainfall events, the capacity of CSSs can be exceeded
resulting in the discharge of untreated storm water and wastewater directly into receiving waters.
These combined sewer overflow events (CSOs) can introduce high concentrations of microbial
pathogens and other pollutants into receiving waters.
The frequency and severity of CSO events is strongly influenced by climatic factors
governing the occurrence of urban stormwater runoff, particularly the form (i.e., rain or snow),
the amount, and the intensity of precipitation. Under U.S. EPA's CSO Control Policy, CSS
communities are required to implement mitigation measures as a component of the National
Pollutant Discharge Elimination System (NPDES) permitting process. CSO mitigation measures
include infrastructure upgrades to increase system capacity (e.g., storage) and stormwater
management to reduce the volume of runoff entering CSSs. Such practices are typically
engineered to handle precipitation or runoff events of a given intensity, duration, or frequency,
and most often there is an implicit assumption that precipitation and hydrology are statistically
stationary (e.g., constant mean, variance, autocorrelation structure) overtime.
During the last century, much of the United States experienced increased ambient air
temperatures and altered precipitation patterns (NAST, 2000). Projections of future climate
suggest these trends are likely to continue and potentially accelerate during the next century
(IPCC, 2007). If realized, these changes could present a significant risk to the future
performance of CSS infrastructure—including CSO mitigation. Little is known, however, about
the extent of this risk.
This screening-level report assesses the potential order of magnitude of climate change
impacts on CSO mitigation in the New England and Great Lakes Regions. The purpose is to
determine whether the potential implications of climate change on CSOs in these regions warrant
further consideration and study, and secondly, to evaluate the need for decision support tools and
information enabling CSS managers to better incorporate consideration of climate change into
their decision making processes. As such, this assessment and report is only a first step towards
understanding a complex issue, the implications of which will vary significantly in different
locations and for different systems. Results are thus not intended, nor the methods appropriate,
to provide site-specific information on the potential impacts or mitigation requirements for any
individual system or facility.
A simple, precipitation "benchmarking" approach was used to examine the extent to
which CSO long-term control plans may be under-designed if planners assume that past
precipitation conditions are representative of future conditions. We assumed that each CSO
1
-------
community in the New England and Great Lakes Regions will design their system to achieve an
average of four CSO events per year (i.e., the Presumption approach threshold), and they will
base their design on historical precipitation data. The benchmark daily precipitation event (daily
total) equaled or exceeded, on average, four times per year was determined based on historical
and projected future precipitation data. The extent to which CSO mitigation may be under-
designed, if based on historical precipitation, was then determined by estimating changes in the
frequency of the historical benchmark event under future climate conditions. The additional
system capacity required to meet the mitigation target in the future was determined based on
estimated future changes in the magnitude of the benchmark daily event and an assumed range of
runoff coefficients describing proportional changes in stormwater runoff resulting from changes
in precipitation.
Daily precipitation data representative of historical and projected future climate
conditions were obtained from the Vegetation/Ecosystem Modeling and Analysis Project - Phase
2 (VEMAP Phase 2). VEMAP data representative of historical climate conditions were
developed using monthly average station data from Historical Climate Network and Cooperative
Network weather stations. Daily precipitation totals representative of historical climate were
generated from monthly average station values using a modified version of the stochastic
weather generator WGEN (Richardson and Wright, 1984). VEMAP data representative of future
climate conditions were developed based on transient climate modeling experiments using two
coupled atmosphere-ocean general circulation models (AOGCM): the Hadley Centre Model
(HADCM2) and the Canadian Climate Centre Model (CCCM). Daily precipitation totals
representative of future climate were generated based on monthly average AOGCM outputs
using WGEN (Richardson and Wright, 1984).
Regional average estimates of potential impacts on CSOs were determined by mapping
the 317 active CSS communities located in the Great Lakes and New England Regions to the
nearest grid location where VEMAP data was available, analyzing the VEMAP data at each of
these locations, and weighting the results from each VEMAP grid location within each region
according to the number of CSS communities represented by that grid location
Results suggest that if future climate change includes increased precipitation and
stormwater runoff volumes, the efficacy of CSO mitigation efforts may be diminished.
Specifically, in the Great Lakes Region, projected long-term (2060-2099) changes in
precipitation suggest that if CSO mitigation efforts are designed based on historical precipitation,
many systems could experience increases in the frequency of CSO events beyond their design
capacity resulting in increases in overflow volume discharged to receiving waters.
In the New England Region, projected near-term (2025-2050) changes in precipitation
are inconsistent, with projections based on the Hadley and Canadian AOGCM models
-------
disagreeing on the direction of change. This difference in direction complicates interpretation of
the results and highlights uncertainties associated with the AOGCM climate projections. The
near-term results for the New England Region are best considered inconclusive; neither result
confirms nor refutes the likelihood of future climate change impacts on CSOs. This analysis did
not include long-term (2060-2099) projections of change in the New England Region—a
limitation that precludes any direct comparison of impacts in the two study regions. Future study
is required to address this issue in the New England Region.
It must be noted that the methodology used in this study involves a number of
simplifying assumptions and limitations. Key limitations include the inherent uncertainty in
AOGCM projections, methods for downscaling to daily data, the lack of consideration of snow
and snowmelt, and the use of simple scaling factors to estimate stormwater runoff rather than
detailed modeling of individual CSSs.
Investments in water infrastructure tend to be long-term, capital-intensive, and, in many
cases, irreversible in the short- to medium-term. It is thus prudent to consider that today's
decisions could influence the ability of treatment facilities to accommodate changes in climate
for decades into the future. Faced with the prospect of future climate change, opportunities may
exist where current CSO mitigation efforts can be upgraded at little additional cost to provide an
added margin of safety to account both for near-term extreme events and the potential future
effects of climate change. No-regrets opportunities may also exist where actions taken today to
address current, other water quality concerns can provide additional benefits in the context of
adapting to climate change.
Finally, it is important to recognize that each CSS and CSS community has a unique set
of attributes, existing challenges, constraints, and other factors that must be considered in
determining what reasonable and appropriate actions should be taken to manage any increase in
risk associated with climate change. The focus of this report, CSOs, is not meant to imply that
CSOs are the single or even the greatest source of water quality impairment in these areas. Other
sources of impairment including non-point loading from agriculture, urban development, and
other sources also occur in the study regions. Accordingly, responding to climate change will
require a holistic approach that considers climate change in the context of other impacts on CSSs
and regional water quality to determine what reasonable and appropriate actions can be taken.
Although limited in scope, this screening-level analysis provides a first step towards a
better understanding of climate change impacts on CSOs in the New England and Great Lakes
Regions. Results suggest that certain systems may be vulnerable to future climate change and
that there is a need for more detailed, site-specific analyses including the development of
decision support tools and information. Regardless of whether or not CSS managers choose to
-------
include climate change in their long-term planning, it is preferable that the decision be
intentional and not due to lack of awareness of the problem.
-------
2. INTRODUCTION
Combined sewer systems (CSSs) collect and co-treat storm water and municipal
wastewater. During high intensity rainfall events, the capacity of CSSs can be exceeded
resulting in the discharge of untreated storm water and wastewater directly into receiving waters.
These combined sewer overflow events (CSOs) can result in high concentrations of microbial
pathogens, biochemical oxygen demand, suspended solids, and other pollutants in receiving
waters.
The frequency and severity of CSO events is strongly influenced by climatic factors
governing the occurrence of urban stormwater runoff, particularly the form (i.e., rain or snow),
amount, and intensity of precipitation. Under U.S. Environmental Protection Agency (U.S.
EPA)'s CSO Control Policy, CSS communities are required to implement mitigation measures as
a component of the National Pollutant Discharge Elimination System (NPDES) permitting
process. CSO mitigation measures include infrastructure upgrades to increase system capacity
(e.g., storage) and stormwater management to reduce the volume of runoff entering CSSs. Such
practices are typically engineered to handle precipitation or runoff events of a given intensity,
duration, or frequency, and most often there is an implicit assumption that precipitation and
hydrology are statistically stationary (e.g., constant mean, variance, autocorrelation structure)
over time. The rules guiding mitigation requirements are also based in part on an understanding
of how different characteristics of precipitation events affect sewer performance.
The Fourth Assessment Report of the Intergovernmental Panel on Climate Change
(IPCC) states that warming of the climate system is now unequivocal, as is evident from
observations of increases in global average air and ocean temperatures, widespread melting of
snow and ice, and rising global average sea level (IPCC, 2007). The IPCC also reports that if
greenhouse gas emissions continue at or above current rates, changes in the global climate
system during the 21st century will very likely be larger than those observed during the 20th
century. In the United States, observed climate change during the last century varied regionally
but generally included warming temperatures and an increased frequency of heavy precipitation
events (IPCC, 2001). Anticipated future changes also vary regionally, but throughout most of
the United States changes include continued warming temperatures and increases in heavy
precipitation events (IPCC, 2007). If realized, these changes could present a significant risk to
the performance of CSS infrastructure including efforts to mitigate CSOs. Specifically, regions
experiencing an increased frequency of high intensity rainfall events may also experience an
increased risk of CSO events and associated water quality impairment. In the United Kingdom,
an assessment of climate change projections for the year 2080 suggests that future climate
-------
change could result in increased flooding and CSO frequency (Wilkinson and Balmforth, 2004).
Generally, however, little is known about the extent of this risk.
This screening-level report assesses the potential order of magnitude of climate change
impacts on CSO mitigation in the New England and Great Lakes Regions. The purpose is to
determine whether the potential implications of climate change on CSOs in these regions warrant
further consideration and study, and to evaluate the need for decision support tools and
information enabling CSS managers to better incorporate consideration of climate change into
their decision making processes. As such, this assessment is only a first step towards
understanding a complex issue, the implications of which will vary significantly in different
locations and for different systems. A simple, precipitation "benchmarking" approach was used
to examine the extent to which CSO long-term control plans (LTCPs) may be under-designed if
it is assumed that past precipitation conditions are representative of future conditions. The study
is not intended, nor are the methods appropriate, to provide detailed, site-specific information on
the potential impacts or mitigation requirements for any individual system or facility. The New
England and Great Lakes Regions were selected for study because CSSs in these two regions
account for nearly half of the total 746 CSS communities in the United States (U.S. EPA, 2004).
2.1. COMBINED SEWER SYSTEMS AND COMBINED SEWER OVERFLOWS
A CSS collects storm water and sanitary wastewater in a common conveyance system
and routes them to a treatment plant (U.S. EPA, 2004). The storm water component fluctuates
with the weather: during rainfall events, the collection system and treatment plant must
accommodate more volume due to runoff entering the system directly through street catch basins
and gutter downspouts. By design, when the volume of water entering a CSS exceeds the
system's capacity, excess water is discharged at different points in the system into receiving
waters through CSO outfalls.
The water that is discharged to receiving waters during a CSO event is typically a
mixture of raw or partially treated (screened for solids) sewage, other industrial wastewaters, and
storm water. The sewage component is typically of greatest concern due to bacterial and/or viral
contamination. These discharges usually occur in response to wet weather and are known as
CSOs. The term CSO refers to any discharge from a CSS prior to the treatment plant (U.S. EPA,
2004). The U.S. EPA (2004) estimates that 850 billion gallons of overflow are discharged into
the nation's waters each year.
According to U.S. EPA's 2004 Report to Congress on the Impacts and Control of CSOs
and Sanitary Sewer Overflows (SSOs), there are 746 communities with combined systems and a
total of 9,348 CSO outfalls identified and regulated by the NPDES permits (U.S. EPA, 2004).
-------
Combined sewer systems are found in 31 states and the District of Columbia, with the majority
located in older cities found in the Great Lakes and New England Regions (see Figure 1).
c>
Figure 1. Distribution of combined sewer systems in
the United States.
Source: U.S. EPA, 2004.
2.1.1. History
Construction of municipal sewer systems began in the 1880s. Prior to then, sewage was
collected in cesspools and privy vaults (Burian et al., 2000). Populations were sparse enough
that aesthetic concerns were not great, and the health risks were not well understood. As
population densities grew during the 19th century, the need to remove waste became more
critical. Cities had two initial goals in constructing sewers: (1) as populations grew larger and
more concentrated, privies and cesspools were no longer sanitary or aesthetically acceptable.
Sewers were constructed to remove wastes from population areas, (2) heavy rains could render
unpaved streets impassable, so storm waters needed to be quickly conveyed to rivers or lakes.
The two designs common to this time period were dedicated sanitary sewers and combined
sewers to convey sewage and storm water. A third option, separate sanitary and storm sewers,
was viewed as too costly for most communities (Burian et al., 2000).
At this time, waste waters were not treated; the purpose of sewer systems was simply to
convey the water away from the population and into some receiving waterbody. The design
choice depended on the community's needs. In general, smaller communities opted for sanitary
-------
systems only. Larger cities used combined systems because they efficiently and effectively
removed both storm and wastewater (U.S. EPA, 2004). In some cases, a combined system was
chosen because storm water would flush the sewage out of the system into receiving waters
(Burian et al., 2000; Schladweiler, 2005). At the turn of the 20th century, these systems were
linked to significant reductions in waterborne disease outbreaks within the cities they served.
There were, however, consequences with respect to water quality in receiving waters.
As the first sewer systems were being built, CSSs were a dramatic improvement over
cesspools and open sewers. However, as populations grew, impacts of discharges into receiving
waters grew as well. CSOs result mainly from two different events associated with CSSs: (1)
insufficient conveyance capacity within a portion of the sewer system and resultant surcharging
and overflow through manholes or designed outfalls and (2) insufficient capacity at the
wastewater treatment facility. In the latter case, the excess combined sewage must bypass the
facility and be discharged at a specified outfall. This effluent is often screened to remove solids
(primary treatment) before discharging.
2.1.2. Effects on Water Quality and Public Health
CSOs present a threat to water quality and public health. The pollutants found in CSOs
include microbial pathogens, suspended solids, nutrients, toxics, and debris. Pollutant
concentrations vary, but they can be high enough to cause violations of water quality standards.
It is common for local rivers and streams to be considered dangerous to human health after heavy
rains due to CSO pollution. It is difficult to attribute the violation of water quality standards
exclusively to CSO discharges, because CSOs occur during storm events when some of the same
pollutants are washed directly into receiving waters by storm water runoff.
Despite the difficulty of attributing causality to particular sources, the U.S. EPA has
compared data on CSO locations with data on 305(b) assessed water segments and 303(d)
impaired waters in 19 states (U.S. EPA, 2004).1 The study found that of a total of 59,335
assessed segments, 25% were impaired. For 733 segments that were within a mile downstream
of a CSO outfall, 75% were impaired. Though it is difficult to determine how much of the
impairment is due to the CSO, the high percentage of impairment associated with CSOs suggests
some correlation (U.S. EPA, 2004). CSOs should be considered as a potential source of
pollution during Total Maximum Daily Load (TMDL) development, and in some communities
substantial load reductions have been assigned to CSOs as a result of the TMDL process (U.S.
EPA, 2004).
and 303(d) are references to sections of the Clean Water Act that mandate assessment of water bodies, and
identification of impaired waters, respectively.
-------
CSOs also present risks to public health and the natural aquatic ecosystems. Humans can
become sick by drinking contaminated water, eating contaminated shellfish, or coming in direct
contact with contaminated water. The most common symptoms of pathogenic illness are
diarrhea and nausea, but respiratory and other problems can occur as well. Toxics present in
CSO discharges include metals and synthetic organic chemicals. Less is known about the risks
of biologically active chemicals such as antibiotics, hormones, and steroids (U.S. EPA, 2004).
2.2. CSO CONTROLS
Efforts to manage the risks of CSOs have evolved over the last several decades.
Following the passage of the Clean Water Act in 1972, publicly owned wastewater treatment
works (POTWs) were required to incorporate secondary treatment into their wastewater
treatment processes. Effluent from the treatment plant had to be treated, but it was unclear how
CSOs—discharges from the collection system, not the treatment plant—would be treated by the
law. A 1980 court ruling declared that CSO outfalls did not have to be subjected to the
secondary treatment required of discharges from a POTW. However, the discharges do fall
under the National Pollutant Discharge Elimination System (NPDES) permit program (U.S.
EPA, 2004). Under NPDES, all facilities which discharge pollutants from any point source into
waters of the United States are required to obtain a permit. The permit holder must provide
treatment based on technology accessible to all permittees in a particular industrial category
(U.S. EPA, 2006).
In 1989, the U.S. EPA issued the National CSO Control Strategy, which encouraged
states to develop statewide permitting strategies to ensure that all CSSs were subject to a
discharge permit. The strategy also recommended six minimum measures for controlling CSOs.
As the control strategy was being implemented, environmental groups pushed for further action
against CSOs, while many municipalities also called for greater clarity and a national approach
(U.S. EPA, 2004). In 1994, the U.S. EPA published the CSO Control Policy to establish
objectives for CSS communities in order to reduce the environmental impacts of CSOs. Four
key elements of the CSO Control Policy are meant to enable communities to cost effectively
reduce overflows and meet the objectives of the Clean Water Act:
(1) Provide clear levels of control that would be presumed to meet appropriate health and
environmental objectives;
(2) Provide sufficient flexibility to municipalities, especially financially disadvantaged
communities, to consider the site-specific nature of CSOs and to determine the most
cost effective means of reducing pollutants and meeting Clean Water Act objectives
and requirements;
-------
(3) Allow a phased approach to implementation of CSO controls considering a
community's financial capability; and,
(4) Provide for review and revision, as appropriate, of water quality standards and their
implementation procedures when developing CSO control plans to reflect the site-
specific wet weather impacts of CSOs.
2.2.1. Nine Minimum Controls and Long-Term Control Plans
The national CSO Control Policy also requires communities to implement nine minimum
controls (referred to as NMC) and to develop a LTCP to reduce the frequency and adverse
impact of CSOs. The NMC are expected to maximize the effectiveness of existing systems.
Among the controls are properly operating and maintaining the system; maximizing the flow to
the POTW from the collection system; eliminating overflows during dry weather; and notifying
the public of the occurrence and impacts of overflows. In addition to implementing the NMC,
communities are expected to develop LTCPs that will ultimately result in compliance with the
requirements of the Clean Water Act.
The development and implementation of LTCPs are in various stages of completion, but
all of the 746 communities that have CSSs must develop plans to comply with the CSO Control
Policy. Permit holders designing modifications to their systems generally base their plans on
historical weather data. The infrastructure investments made to implement LTCPs are expected
to have life expectancies of several decades, and the costs will be considerable. There is no
comprehensive source of individual municipal expenditures for CSO control because there are
multiple funding sources for CSO projects. However, the U.S. EPA has compiled expenditures,
to date, for 48 communities, roughly 6% of the nation's total. Those expenditures totaled $6
billion and ranged from $134,000 to $2.2 billion per community. The U.S. EPA estimates that
the capital costs of future CSO control over the next 20 years will exceed $50 billion (U.S. EPA,
2004).
2.2.2. CSO Control Policy Mitigation Requirements
The U.S. EPA's CSO Control Policy allows for three basic approaches to be taken in
order to meet CSO mitigation requirements: First, a system may allow no more than four
overflow events per year (though the permitting authority may allow an additional two). Second,
a system may eliminate or capture at least 85%, by volume, of the combined sewage collected in
the system during a precipitation event. These first two approaches are considered to be
"presumptive" in nature. Finally, the system may eliminate or remove no less than the mass of
the pollutants identified as causing the water quality impairment for the volume that would be
eliminated or captured by the 85% approach (U.S. EPA, 1994). This final, or "Demonstration"
approach, allows communities to demonstrate that their system, though not meeting the criteria
10
-------
of the Presumption approach, is adequate to enable receiving waters to meet water quality
standards and protect designated uses (U.S. EPA, 2004).
Investments in CSO control tend to be long-term, capital-intensive, and, in many cases,
irreversible in the short- to medium-term. Given the influence of climate on stormwater runoff
and the occurrence of CSO events, it is thus prudent to consider the potential impacts of climate
change on the effectiveness of efforts to mitigate CSOs over the next several decades. To the
extent that climate change may result in increased precipitation, if CSO mitigation is designed
based on current climate and/or hydrology (e.g., calculations of required system storage
capacity), it is possible that mitigation actions taken as part of CSO long term planning may not
be sufficient to meet the desired objective.
2.3. STUDY GOALS
The goal of this screening-level analysis is to assess the potential order of magnitude of
climate change impacts on CSO mitigation in the New England and Great Lakes Regions. The
purpose is to determine whether the potential implications of climate change on CSOs in these
regions warrant further consideration and study, and to evaluate the need for decision support
tools and information enabling CSS managers to better incorporate consideration of climate
change into their decision making processes. As such, this assessment and report is only a first
step towards understanding a complex issue, the implications of which will vary significantly in
different locations and for different systems. An improved understanding of the potential
impacts of climate change on CSOs is important because the occurrence and mitigation of CSO
events is highly sensitive to climate, is one of the highest-priority programs at the state and
federal level, and will involve significant investment in wastewater collection, storage, and
treatment infrastructure. CSOs are also a timely subject of strategic discussion due to the large
gap in funds available versus funds needed for treatment system improvements.
A simple, screening-level approach is used to examine the extent to which CSO long-
term control plans may be under-designed if planners assume that past precipitation conditions
are representative of future conditions. An informal survey of U.S. EPA staff indicated that the
most common approach to LTCPs in the New England and Great Lakes Regions was the
Presumption approach: controlling and providing a minimum level of treatment to all but four
overflow events per year. This study considers the following two questions:
(1) If CSSs currently meet the U.S. EPA's CSO Control Policy Presumption approach of
four events per year based on historical precipitation, what is the potential change in
CSO event frequency in the future as a result of climate change?
11
-------
(2) What is the potential required change in the design capacity of mitigation measures
needed to meet the Presumption approach of four CSO events per year in the future as a
result of climate change?
Focusing on the four-events-per-year Presumption approach threshold provides a useful
benchmark for assessing impacts. This requirement alone, however, is not indicative of
compliance with the Clean Water Act. The analysis thus takes a relatively simple approach to a
complex problem. The study is not intended, nor are the methods appropriate, to provide
detailed, site-specific information on the potential impacts or mitigation requirements for any
individual system or facility. Analysis of the spatial variability of potential changes within each
region is also not addressed in this study. The New England and Great Lakes Regions were
selected as the focus of this study because of the large number of CSSs in these areas. It should
be noted, however, that results in the New England and Great Lakes Regions cannot be directly
compared to one another because work in the different regions was conducted as part of two
separate projects, and there are methodological differences in the future time periods considered.
It should also be noted that CSOs are just one of many potential sources of water quality
impairment in the New England and Great Lakes Regions. The focus of this report, CSOs, is not
meant to imply that CSOs are the single or even the greatest source of water quality impairment
in these areas. Other sources of impairment including non-point loading from agriculture, urban
development, and other sources may also be highly sensitive to precipitation changes associated
with climate change. Accordingly, responding to climate change will require a holistic approach
that considers climate change in the context of other impacts on local or regional water quality to
determine what reasonable and appropriate actions can be taken.
12
-------
3. METHODS
This study uses a simple, precipitation benchmarking approach to assess how future
changes in precipitation could impact the frequency and volume of CSO events in the New
England and Great Lakes Regions. The desired mitigation target for CSSs in each region was
assumed to be the LTCP Presumption standard of reducing CSO frequency to no more than four
events per year. Using this mitigation target, the historical benchmark daily event (daily total
precipitation) that is equaled or exceeded, on average, four times per year was determined. The
extent to which CSO mitigation may be under-designed if based on historical precipitation was
then determined by estimating changes in the frequency of the historical benchmark event under
future climate conditions. The additional system capacity required to meet the mitigation target
in the future was estimated based on future changes in the magnitude of the benchmark daily
precipitation event and an assumed range of runoff scaling factors describing the proportional
changes in stormwater runoff resulting from changes precipitation.
The precipitation benchmarking approach provides a simple and straightforward method
for assessing order-of-magnitude changes in each of the study regions. Detailed modeling of
individual CSSs to account for specific characteristics of each sewershed affecting the generation
and routing of stormwater runoff was not conducted due to the size (large) of geographic area
considered. Results are thus not intended to provide site-specific information on mitigation
requirements for any individual system or facility. The subsequent sections provide a more
detailed discussion of the methodology.
3.1. CSS SELECTION
A national list of CSS locations was obtained from a 2004 U.S. EPA Report to Congress
(U.S. EPA, 2004). Latitude and longitude were determined for each CSS by cross referencing
NPDES permit numbers with location information in the Permit Compliance System, or based on
city location. CSS communities within the Great Lakes and New England Regions were selected
as described in the following sections.
3.1.1. Great Lakes Region
The Great Lakes are part of the largest freshwater system in the world, and are bounded
by eight states: Minnesota, Wisconsin, Michigan, Illinois, Indiana, Ohio, Pennsylvania, and
New York (U.S. EPA, 2004; GLRA, 2000). A total of 182 CSS communities with active CSO
permits were identified within the Great Lakes Region (within the Great Lakes watershed);
Table 1 presents a breakdown of CSS communities within the Great Lakes Region by state.
13
-------
Table 1. Great Lakes region CSS communities by state
State Number of CSS Communities
Ohio 47
Michigan 46
Illinois 34
Indiana 24
New York
(in Great Lakes watershed)
Minnesota 3
Pennsylvania 3
Wisconsin 2
Total 182
3.1.2. New England Region
The New England Region was defined to include seven states: Maine, New Hampshire,
Vermont, Massachusetts, Connecticut, Rhode Island, and New York (excluding New York City).
A total of 135 active CSO permits were identified in the New England Region (U.S. EPA, 2004).
Table 2 shows the breakdown of CSS communities by state in the New England Region.
Table 2. New England region CSS communities by state
State
New York
(upstate)
Maine
Massachusetts
Vermont
New Hampshire
Connecticut
Rhode Island
Total
Number of CSS Communities
53
39
22
7
6
5
3
135
14
-------
3.2. PRECIPITATION BENCHMARKING APPROACH
There is considerable heterogeneity among CSSs in terms of baseline water quality
conditions, progress toward complying with the U.S. EPA's national CSO Control Policy, and
the site-specific approaches that will be used to reduce the frequency of CSO events. The U.S.
EPA requires, however, that all CSS communities develop a LTCP that includes an evaluation of
alternatives to meet CWA requirements by using either the Presumption approach or the
Demonstration approach. One of the most common design objectives of LTCPs is to achieve the
U.S. EPA's Presumption approach threshold of no more than an average of four CSO events per
year. Under this criterion, a CSO event is defined as any overflow from a CSS that does not
receive the minimum level of treatment defined in the CSO Control Policy.
In this study, it is assumed that each CSO community in the New England and Great
Lakes Regions will design their system to achieve an average of four CSO events per year (i.e.,
the Presumption approach threshold), and will base their design on historical precipitation data.
The benchmark daily total precipitation that is equaled or exceeded, on average, four times per
year was determined based on historical and future precipitation data. The extent to which CSO
mitigation may be under-designed if based on historical precipitation was then determined by
estimating changes in the frequency of the historical benchmark event under future climate
conditions. Potential increases in system capacity required to meet the mitigation target in the
future was estimated based on future changes in the magnitude of the benchmark daily
precipitation event and an assumed range of runoff scaling factors describing the proportional
changes in stormwater runoff resulting from changes precipitation.
Daily precipitation data representative of historical and projected future climate
conditions were obtained from the Vegetation/Ecosystem Modeling and Analysis Project—Phase
2 (hereafter referred to as VEMAP), administered by the National Center for Atmospheric
Research (NCAR) data group. VEMAP data representative of historical climate conditions were
developed using monthly average station data from Historical Climate Network and Cooperative
Network weather stations. Daily precipitation totals were generated from monthly average
station values using a modified version of the stochastic weather generator WGEN (Richardson
and Wright, 1984). Station data were spatially interpolated to a grid with intervals of one degree
latitude by one degree longitude (110 Km latitude and approximately 80 Km longitude in these
study regions) using PRISM (Parameter-elevation Regressions on Independent Slopes Model)
(Dalyetal., 1994).
VEMAP data representative of future climate conditions were developed for each
VEMAP grid location based on transient climate modeling experiments for a simulated period
from 1994 to 2100 using two coupled atmosphere-ocean general circulation models (AOGCM):
the Hadley Centre Model (HADCM2) and the Canadian Climate Centre Model (CCCM). Daily
15
-------
precipitation totals were developed based on monthly average AOGCM outputs using WGEN
(Richardson and Wright, 1984).2 Detailed documentation and discussion of the VEMAP Phase 2
data can be found in the VEMAP Phase 2 Users Guide (Rosenbloom et al., 2003).
It should be noted that the VEMAP data sets used in this study are subject to a number of
uncertainties and limitations related to the AOGCM projections, the daily precipitation totals
generated using WGEN, and other modeling methods and assumptions. The AOGCM modeling
experiments used to develop the VEMAP data sets are several years old and subject to inherent
modeling uncertainties. The future data sets used in this study should thus not be considered
predictions, but rather as representative, plausible futures. In addition, VEMAP daily data were
developed using a stochastic weather generator and thus may not represent extremes well.
Analysis by the VEMAP team determined that frequency distributions and extremes of daily data
from WGEN compare well to those of observed station data (Rosenbloom et al., 2003).
However, because daily data were developed using a weather generator, the values should only
be considered estimates. Finally, the daily precipitation totals available from VEMAP do not
allow consideration of sub-daily event characteristics known to influence the frequency and
magnitude of CSO events (e.g., system response to a thunderstorm yielding 2 inches of rain in
one hour versus a steady rain that accumulates 2 inches over a full 24-hour period). A more
detailed discussion of the VEMAP data can be found in the VEMAP Phase 2 Users Guide
(Rosenbloom et al., 2003).
Although subject to uncertainty, VEMAP data was used in this screening-level
assessment because it was readily available, and because the data set is well known and
documented. Use of VEMAP data set also simplified the analysis by providing historical and
projected future climate data on the same geographic footing (a one-degree grid).
Regional-average estimates of potential CSO impacts were conducted by first mapping
each of the 317 active CSS communities in the Great Lakes and New England Regions to the
nearest VEMAP grid location. Analyses of precipitation data were then conducted for each
VEMAP grid location with at least one associated CSS. Estimates of the regional average
changes for all CSSs within each study region were determined by weighting results from each
VEMAP grid location within each region according to the number of CSS communities
represented by that point.
WGEN is a weather simulation model developed at the USDA-ARS Grassland, Soil and Water Research
Laboratory that is used to scale down monthly AOGCM outputs to a daily time-step. The model uses a probability
function (first-order Markov chain) where the chance of precipitation is conditioned on the wet or dry status of the
previous day, and the intensity is based on a gamma distribution where small intensity events occur more frequently
than large intensity events.
16
-------
The analysis of precipitation data at each VEMAP grid location was conducted in two
ways: using the daily precipitation totals and using the 4-day moving average of daily totals.
Evaluation of the 1- and 4-day averages accounts for potential differences in the temporal
characteristics of storage in CSO mitigation measures (i.e., the time lag between onset of
precipitation and peak flows within a CSS). The choice of 1- and 4-day timeframes provided
approximate lower and upper bounds on both (a) time of travel within the sewershed or area
draining to the treatment plant (from "upstream" boundaries to the treatment plant) and (b) the
effects of multiple rain events in quick succession. In addition, the analysis was simplified by
treating all precipitation as rainfall; making no accommodation for the occurrence of snowfall or
snowmelt.
In the Great Lakes Region, 40 years of historical precipitation data representative of the
period from 1954-1993 were compared to 40 years of projected future precipitation data
representative of the period from 2060-2099. In the New England Region, 25 years of historical
precipitation data representative of the period from 1968-1993 were compared to 25 years of
projected future precipitation data representative of the period from 2025-2050. As indicated
earlier, it is important to note that the different future time periods considered in the two study
regions preclude direct comparison of results. Work in the two regions was done at different
times as two independent projects. Work in the New England Region was done subsequent to
that in the Great Lakes Region, and the focus on the period from 2025-2050 in this region was
intended to provide information more relevant to near-term decision making.
As mentioned previously, it was assumed that CSSs in each study region will design their
systems to meet the four-event per year standard based on historical daily precipitation totals.
This served as the historical benchmark daily event for each system. In theory there will be only
four events (in an average year) that exceed this benchmark if a CSS community is meeting the
objectives of the CSO Control Policy. In the case of the Great Lakes Region, the benchmark
event was identified as the 160th largest daily precipitation event (daily total) in each of the
40-year, aggregated 1- and 4-day moving average precipitation data sets (four events per year *
40 years of data). The magnitude of benchmark events for the Great Lakes Region were
determined by ranking the VEMAP daily precipitation totals, and selecting the 160th largest
event. For the New England Region, the benchmark event was identified as the 100th largest
daily precipitation event in each of the 25-year, aggregated 1- and 4-day precipitation data sets
(four events per year * 25 years of data). Benchmark events for the New England Region were
determined by ranking the VEMAP daily precipitation totals and selecting the 100th largest
event. Note that in each case this methodology provides a simple estimate of the daily
precipitation event (1- or 4-day) with a recurrence interval of 3 months. Formal frequency
analysis based on an assumed statistical distribution of daily precipitation was not considered
17
-------
necessary for this screening analysis given the other uncertainties inherent in the approach.
Using this methodology, the benchmark 1- and 4-day daily precipitation events needing to be
captured to meet the LTCP Presumption standard of no more than four CSO events per year were
estimated at each VEMAP grid location under conditions representative of historical and future
climate (based on CCCM and HADCCM2 AOGCM projections).
The extent to which CSO mitigation may be under-designed if based on historical
precipitation was estimated by determining changes in the frequency of the historical benchmark
daily precipitation event under future climate conditions. Potential future changes in system
storage capacity required to meet the mitigation target in the future were then estimated based on
changes in the magnitude of the four-event-per-year benchmark event, and estimates of changes
in stormwater runoff generated by these changes in precipitation. Detailed modeling of
stormwater for individual systems was beyond the scope of this study. Rather, for each VEMAP
location and projected change in precipitation, percent changes in stormwater runoff (relative to
historical conditions) were estimated by applying a high- and low-range multiplier (scaling
factors) to projected percent changes in benchmark daily precipitation events. Potential future
changes in system capacity (expressed as a percent relative to that necessary to meet the
four-events-per-year Presumption standard under historical climate conditions) to meet the
four-events-per-year mitigation target in the future were then assumed equal to percent changes
in stormwater runoff volume.
It is important to note that the benchmarking approach used in this study does not
consider the unique attributes of any individual CSS, and it is not intended to provide site-
specific information for any individual CSS community. Rather, the approach is intended as a
simple, screening-level assessment of the potential order of magnitude of climate change impacts
in each study region. More detailed study and modeling are required to determine specific
mitigation design specifications for individual CSS communities. The estimates of potential
increases in system capacity also assume that infrastructure upgrades are the predominant
approach used in CSO mitigation. In practice, stormwater management can be implemented to
reduce the volume of runoff that enters a CSS, thus reducing the need for changes in CSS
infrastructure. Although not explicitly addressed, however, the range of scaling factors used to
estimate stormwater runoff does capture a range of potential changes in hydrologic response that,
in part, could result from changes in future stormwater management practices.
18
-------
4. RESULTS AND DISCUSSION
4.1. CHANGES IN CSO EVENT FREQUENCY
The extent to which CSO mitigation may be under-designed if based on historical
precipitation was estimated by determining changes in the frequency of the historical benchmark
event under future climate conditions. It is assumed that municipalities will design mitigation
measures (e.g., a deep storage tunnel or stormwater management practices to reduce runoff) to
meet the four-event-per-year Presumption approach threshold, and if historical precipitation is
used as the design standard, the effectiveness of those mitigation measures could be reduced in
the future. The metric presented in this section is the estimated percent change in CSO event
frequency relative to the four events per year allowed under the LTCP Presumption approach.
For example, a 50% increase in CSO event frequency would equate to two additional CSO
events per year, on average, or a total of six CSO events per year if the mitigation measures were
designed using historical precipitation data.
4.1.1. Great Lakes Region
In the Great Lakes Region, the regional average annual CSO frequency during the future
period from 2060-2099 was estimated to increase between 13% (based on the CCCM, 1-day
averaging period) and 70% (based on the HADCM2, 4-day averaging period) relative to the
assumed historical condition of four events per year. In other words, the average number of CSO
events per year would increase to 4.5 using the lowest projected average change, and 7.1 using
the highest average projected change (see Table 3).
Table 3. Regional average percent change in CSO frequency in
the Great Lakes region, 2060-2099. The percent change is
expressed relative to the four events/year standard, e.g., a 50%
change equals two additional CSO events/year.
Moving Average
1-Day
4-Day
CCCM
13.4%
18.8%
HADCM2
49.4%
70.0%
Figure 2 shows the distribution of the percentage change in CSO frequency (number of
communities) in the Great Lakes Region based on a 1-day moving average of daily precipitation.
The HADCM2 projects an increase in daily precipitation totals for all locations, while the
19
-------
I/)
0)
Jp
'E
3
E
o
O
8
0)
.a
E
3
J?>'
.&
q>
Change Interval (percent)
.,<£>
Q>
Figure 2. Percent change in frequency of CSO events in the Great
Lakes region relative to historical values based on Canadian (CCCM)
and Hadley (HADCM2) Model climate projections for the future period
from 2060-2099 (1-day).
CCCM projects decreases at 10 of the communities (5%). Accordingly, based on the CCCM,
there are 10 communities that are projected to experience a decrease in CSO frequency of 10 to
0.1%. At the other end of the spectrum, the HADCM2 predicts an 80 to 90% increase in CSO
frequency for 10 communities. For these 10 CSS communities, this would mean an additional
32 CSO events per year, on average. Figure 3 shows a similar plot based on the 4-day averaging
period. These results generally indicate wider distributions and greater impacts (though for the
CCCM, there is an increase in the number of communities with fewer CSO events). Figure 4
shows the cumulative distributions for the HADCM2 and CCCM for both averaging periods.
4.1.2. New England Region
In the New England Region, the regional average annual CSO frequency during the
future period from 2025-2050 was estimated to change from -24% and 14% relative to the
assumed historical condition of four events per year (Table 4). In other words, the average
number of CSO events per year would decrease to 3.0 using the lowest projected average change
and increase to 4.6 using the highest average projected change.
Figure 5 shows the distribution of the percentage change in CSO frequency (number of
communities) in the New England Region based on a 1-day moving average of daily
20
-------
E
J
O
o
0)
E
•^
K^> K> ^U ^5
Change Interval (percent)
Figure 3. Percent change in frequency of CSO events in the Great
Lakes region relative to historical values based on Canadian (CCCM)
and Hadley (HADCM2) Model climate projections for the future
period from 2060-2099 (4-day).
50 100
Percent Change
150
200
——Canadian 1-day
Hadley 1-day
Canadian 4-day
Hadley 4-day
Figure 4. Cumulative distribution of percent change in CSO frequency
in the Great Lakes region relative to historical values based on future
climate projections for the period 2060-2099.
21
-------
Table 4. Regional average percent change in CSO frequency in
the New England region, 2025-2050. The percent change is
expressed relative to the four events/year standard, e.g., a 50%
change equals two, additional CSO events/year.
Moving Average
CCCM
HADCM2
1-Day
4-Day
-10.4%
-24.5%
8.8%
14.4%
110
Change Interval (percent)
Figure 5. Percent change in frequency of CSO events in the New
England region relative to historical values based on Canadian
(CCCM) and Hadley (HADCM2) Model climate projections for the
future period 2025-2050 (1-day averaging period).
precipitation totals. The HADCM2 model projects an increase in daily precipitation totals for
the majority of locations, while the CCCM model projects decreases for the majority of the
communities. Based on the CCCM model, there are 10 communities that are projected to
experience an increase in CSO frequency of more than 10%, 76 communities that will have less
than a +10% change, and 49 communities that are projected to experience a decrease in CSO
frequency of more than -10%. Alternatively, the HADCM2 model predicts 48 communities to
have an increase in CSO frequency of more than 10%, 64 communities with less than a +10%
change, and 23 communities with decreases in CSO frequency exceeding -10%. Figure 6 shows
similar plot based on the 4-day averaging data. Figure 7 shows the cumulative distributions for
the HADCM2 and CCCM models for both averaging periods.
22
-------
E
J
O
8
ii—
o
0)
E
100
90
80
70
60
40
D Canandian • Hadley
•^
•^ -^
,°> ^°> ^°> ^°> ^°> ^°>
^V
Change Interval (percent)
Figure 6. Percent change in frequency of CSO events in the New
England region relative to historical values based on Canadian
(CCCM) and Hadley (HADCM2) Model climate projections for the
future period 2025-2050 (4-day averaging period).
75 -50 -25 0 25 50 75 100 125 150 175
Canadian 1-day
Hadley 1-day — —Canadian 4-day
Hadley 4-day
Figure 7. Cumulative distribution of percent change in CSO event
frequency in the New England region relative to historical values based
on future climate projections for the period 2025-2050.
23
-------
The inconsistency in direction of projected precipitation changes in the New England
Region complicates interpretation of the results. For example, with the CCCM model, if control
measures in the New England Region are designed based on historical precipitation
characteristics, climate-related changes could result in a 10-25% decrease in CSO events
depending on the averaging period considered, i.e., an increase in the effectiveness of mitigation
measures. A similar analysis based on the Hadley model suggests a 9-14% increase in
overflows. The relatively near-term period 2025-2050 was selected as the focus of this analysis
to be more relevant to current decision making. It is possible, however, that these near-term
projections do not capture longer term trends that may not become detectable until farther into
the future. Despite the inconclusive results in the New England Region in the near term,
resource limitations did not permit analysis of additional future time periods (e.g., 2060-2099) in
the New England Region.
4.2. POTENTIAL MITIGATION REQUIREMENTS
A second objective of this assessment is to estimate the potential required changes in
system capacity to meet the four-event-per-year mitigation target under future climate
conditions. This information may be useful to municipalities interested in implementing CSO
mitigation measures (e.g., a storage basin) that are robust to future changes in climate. Two
metrics are presented in this section to address this general question: (1) the estimated percent
change in the future magnitude of the benchmark daily event that must be captured to meet the
four-events-per-year Presumption standard relative to the historical value, and (2) the estimated
percent changes in stormwater runoff volume resulting from changes in daily precipitation that
must be accommodated by the system. Each metric provides an indication of how CSO
mitigation design parameters could be modified to account for future climate change, albeit the
actual mitigation measures required for any specific CSS will vary considerably depending on
the specific attributes of the system.
It should also be noted that estimates of potential increases in system capacity assume
that CSS infrastructure upgrades are the predominant approach used in CSO mitigation.
Stormwater management to reduce the runoff entering CSSs is not explicitly considered,
although the scaling factors used to estimate stormwater runoff capture a range of potential
changes in hydrologic response.
4.2.1. Potential Changes in Benchmark Daily Precipitation
Changes in the magnitude of the benchmark daily precipitation event that must be
captured to meet the four-events-per-year Presumption approach standard are indicative of the
potential changes in system capacity required to meet this mitigation target under future climate
24
-------
conditions. The relationship between different attributes of precipitation and the volume of
timing of stormwater runoff is complex and highly variable. By assuming a simple relationship
between daily precipitation totals and stormwater runoff, however, the changes in CSS capacity
that would be needed to adapt to future climate change can be approximated. For example, if a
1:1 correspondence is assumed between daily precipitation and stormwater runoff generated, a
10% increase in the benchmark daily event would imply that the design of the system would
need to be sized for a roughly 10% increase in runoff volume to account for climate change.
4.2.1.1. Great Lakes Region
In the Great Lakes Region, the regional, average daily total precipitation corresponding to
a recurrence interval of 4 events per year during the future period from 2060-2099 is projected to
increase from approximately 5 to 16% relative to historical values (see Table 5).
Table 5. Regional average percent change in the benchmark
daily total precipitation in the Great Lakes region during the
future period 2060-2099. Changes are expressed relative to
historical values.
Moving Average
1-Day
4-Day
CCCM
4.8%
5.1%
HADCM2
16.2%
14.9%
CSS communities in certain parts of the Great Lakes Region, however, are projected to
experience future reductions in the benchmark daily event. Figure 8 shows the distributions of
projected changes in the four-event benchmark daily total precipitation in the Great Lakes
Region based on a 1-day moving average of daily precipitation. Projections based on the CCCM
model suggest four CSS communities will experience a decrease ranging from -5 to -0.1 percent
in the benchmark event. At the other end of the spectrum, projections based on the HADCM2
model suggest a relatively large 25% to 29.9% increase in the benchmark daily event for two
communities. The range of values is attributed to spatial variability in climate change
projections for different locations within each region (i.e., the VEMAP grid locations to which
CSS individual communities are mapped). Figure 9 shows the distributions of projected changes
in the four-event benchmark daily total precipitation based on the 4-day averaging period.
Figure 10 shows the cumulative distributions of projected changes in benchmark daily total
precipitation from the HADCM2 and CCCM models for the 1- and 4-day averaging periods.
25
-------
~<2>
jy
j?
°>
Change Interval (percent)
Figure 8. Percent change in the four-event benchmark daily total
precipitation in the Great Lakes region relative to historical values based
on Canadian (CCCM) and Hadley (HADCM2) Model climate projections
for the future period 2060-2099 (1-day averaging period).
120
110
.2100
~.Q>
~<*
Change Interval (percent)
Figure 9. Percent change in the four-event benchmark daily total
precipitation in the Great Lakes region relative to historical values
based on Canadian (CCCM) and Hadley (HADCM2) Model climate
projections for the future period 2060-2099 (4-day averaging period).
26
-------
-5
10 15 20
Percent Change
25
30
35
Canadian 1-day
Hadley 1-day — —Canadian 4-day
Hadley 4-day
Figure 10. Cumulative distribution of percent change in four-event
benchmark daily total precipitation in the Great Lakes region relative
to historical values based on future climate projections for the period
2060-2099.
4.2.1.2. New England Region
In the New England Region, the regional average daily total precipitation corresponding
to a recurrence interval of four events per year during the future period from 2025-2050 is
projected to change from -6% to 3% relative to historical values (Table 6). The regional-average
change based on the different AOGCM models disagree in sign, with the CCCM model
predominantly suggesting decreases in average intensity, and the HADCM2 model
predominantly suggesting increases. This inconsistency in the direction of change complicates
interpretation of results in the context of necessary changes in system capacity to adapt to
climate change. Nonetheless, results do provide information on the potential magnitude of
change relative to historical values in this region.
Table 6. Regional average percent change in the benchmark daily
total precipitation in the New England region during the future period
2025-2050. Changes are expressed relative to historical values.
Moving Average
1-Day
4-Day
CCCM
-3.2%
-5.7%
HADCM2
2.8%
3.1%
27
-------
Figure 11 shows the distributions of projected changes in the benchmark daily
precipitation event in the New England Region based on a 1-day moving average of daily
precipitation. Projections based on the CCCM model suggest three communities may experience
an increase in benchmark daily event exceeding 10%, whereas projections based on the
HADCM2 model suggest 21 communities may experience increases in the benchmark daily
event exceeding 10%. Figure 12 shows the distributions of projected changes in the four-event
benchmark event intensity for the 4-day averaging period. In each case, the projected changes in
event magnitude based on the CCCM model are shifted slightly to the left, indicating a decrease
relative to projections based on the HADCM2 Model. Figure 13 shows the cumulative
distributions of projected changes in benchmark daily precipitation from the HADCM2 and
CCCM models for the 1- and 4-day averaging periods.
120
\N ^
V^ V'N >K
Change Interval (percent)
Figure 11. Percent change in the four-event benchmark daily total
precipitation in the New England region relative to historical values
based on Canadian (CCCM) and Hadley (HADCM2) Model climate
projections for the future period 2025-2050 (1-day averaging period).
28
-------
100
A*y ^
Change Interval (percent)
Figure 12. Percent change in the four-event benchmark daily total
precipitation in the New England region relative to historical values
based on Canadian (CCCM) and Hadley (HADCM2) model climate
projections for the future period 2025-2050 (4-day averaging period).
-20
-10 0 10
Percent Change
20
30
40
• Canadian 1-day o Hadley 1-day — —Canadian 4-day • Hadley 4-day
Figure 13. Cumulative distribution of percent change in four-event
benchmark daily total precipitation in the New England region relative
to historical values based on future climate projections for the
period 2025-2050.
29
-------
4.2.2. Potential Changes in System Capacity
Future changes in precipitation are an important indicator of the potential impacts of
climate change on CSOs, particularly in the context of precipitation design criteria upon which
CSO mitigation strategies are based. In addition to precipitation, however, CSO events are
influenced by specific watershed and sewershed attributes affecting the generation and routing of
stormwater entering the system, and attributes of the CSS including the conveyance, storage
capacity, and wastewater treatment plant capacity. Detailed modeling of individual systems is
required to properly account for these factors, but was beyond the scope of this study. Rather,
coarse, order-of-magnitude estimates of the potential changes in stormwater runoff needing to be
accommodated by CSSs in each region were made by applying a high- and low-range scaling
factor to projected future changes in precipitation.
Chiew et al. (1995) suggest that the percent change in runoff is about twice the percent
change in precipitation in wet and temperate areas. The United Nations Food and Agriculture
Organization (FAO) suggests that the percent change in runoff increases at half the rate of the
increase in precipitation (Critchley and Siegert, 1991). In the United Kingdom, an assessment
based on climate change projections for the year 2080 suggests that increases in rainfall of 40%
would approximately double the frequency and volume of flooding (Wilkinson and Balmforth,
2004). The range of these estimates reflects changes in the hydrologic response associated with
changes in climate in different regions and watersheds. Future changes in watershed (and
sewershed) landuse, infrastructure, and urban stormwater management practices will also
influence the hydrologic response to changes in climate.
To capture a broad range of potential changes in hydrologic response, the estimates of the
FAO and Chiew et al. (1995) were used as lower and upper bounds on the change in runoff per
unit change in precipitation. Table 7 shows the estimated regional average changes in runoff
resulting from application of FAO and Chiew et al. scaling factors to projected changes in
benchmark event intensity in the Great Lakes Region. Table 8 shows the same estimates for the
New England Region.
30
-------
Table 7. Estimated regional average percent change in runoff volume for the
Great Lakes region for the future period 2060-2099. Changes are expressed
relative to historical values.
CCCM Model
HADCM2 Model
Averaging
Period
% Change
Precipitation
Change
Runoff
(0.5x
scaling)
Change
Runoff
(2x
scaling)
% Change
Precipitation
Change
Runoff
(0.5x
scaling)
Change
Runoff
(2x
scaling)
1-Day
4-Day
4.8
5.1
2.4
2.6
9.6
10.2
16.2
14.9
8.1
7.5
32.4
29.8
Table 8. Estimated regional average percent change in runoff volume for the
New England region for the future period 2025-2050. Changes are expressed
relative to historical values.
CCCM Model
Averaging
Period
1-Day
4-Day
% Change
Precipitation
-3.2
-5.7
Change
Runoff
(0.5x
scaling)
-1.6
-2.9
Change
Runoff
(2x
scaling)
-6.4
-11.4
HADCM2 Model
% Change
Precipitation
2.8
3.1
Change
Runoff
(0.5x
scaling)
1.4
1.6
Change
Runoff
(2x
scaling)
5.6
6.2
4.3. LIMITATIONS AND FUTURE RESEARCH
This screening-level report assesses the potential order of magnitude of climate change
impacts on CSO mitigation in the New England and Great Lakes Regions. The purpose is to
assess whether the potential implications of climate change on CSOs in these regions warrant
further consideration and study in these regions. Accordingly, a methodology was used that
involved a number of simplifying assumptions and other limitations. This section outlines some
key limitations of the approach that must be considered when interpreting results and presents
several suggestions for future research.
31
-------
4.3.1. Limitations
Presumption Approach-The CSO Control Policy Presumption approach standard of no
more than four events per year is assumed to be the main objective of CSS managers, and
increasing system storage is assumed to be the primary mitigation measure used to meet this
standard. In practice, there are many site-specific issues that make development of long-term
control plans very complicated, and not all planners choose to increase storage as their primary
mitigation measure. Moreover, the CSO Control Policy also allows use of a Demonstration
approach based on levels of effluent treatment and water quality standards of the receiving
waterbody. Thus, the analysis takes a relatively simple approach to a very complex and
heterogeneous system; the approach lacks the technical robustness that would result from the use
of sewer system models such as the Storm Water Management Model (SWMM). Modeling and
analysis that captures the effects of changes in rainfall, snowfall, snowmelt, and rain-on-snow
runoff events and system design would be particularly helpful.
Benchmarking-This analysis is based on a benchmarking comparison of historical and
future daily precipitation totals based on weather observations and AOGCM projections. A
simple ranking technique was used in this study to identify benchmark events rather than more
sophisticated statistical techniques. The use of statistical techniques for identifying benchmark
events may yield slightly different results. Moreover, the historical data sets used to establish
storm intensity, duration, and frequency may go back further in time than the 25-year period
used for the New England Region or the 40-year period used for the Great Lakes Region. To the
extent that there are underlying trends in precipitation3 toward increasing intensity of storm
events, use of a short (more recent) time period from the historical data set as the baseline may
bias estimates of future change relative baseline values based on other historical periods.
General Circulation Models-Use of the four-event-per-year standard from the
Presumption Approach as a standard benchmark required a focus on values at the high end of the
precipitation distribution-the 100th largest out of 9,130 values (25 years) for the New England
Region and the 160th largest out of 14,600 values (40 years) for the Great Lakes Region. The
validity of these values is a function both of the validity of the AOGCM projections and the
validity of the downscaling approach used to manipulate those projections. The VEMAP
precipitation data used in this study is based on AOGCM projections (CCCM and HADCM2
models) at a monthly time step. Monthly average values were converted to daily projections
based on the stochastic weather generator, WGEN. The initial AOGCM outputs and the use of a
weather generator to generate daily data incorporate many limitations and a considerable degree
3 For example, the observed 20th century values for annual precipitation were up to 25% greater than pre-20th century
records for the eastern coastline of the New England Region, where many of the CSS communities are located (U.S.
Department of State, 2002).
32
-------
of uncertainty, as described earlier in this report and in the VEMAP documentation. All of those
limitations and uncertainties apply directly to these results. Moreover, most models are more
valid in their estimation of central tendencies than in the tails of the output distribution, and to
the extent that we are concerned here with the high-end tail, the results should be used with
caution.
Averaging Period-The VEMAP data used in this study was at a daily time step.
Although the analysis uses 1- and 4-day averaging periods to bracket the effects of short, intense
storms versus longer or multiple precipitation events in sequence, it is possible that by choosing
shorter or longer periods the results could change. For example, the approach does not
distinguish between intense precipitation "pop-up" events (e.g., a thunderstorm yielding 2 inches
of rain in 1 hour) versus a steady rain that accumulates 2 inches over a full 24-hour period.
These short, but intense, events can quickly overwhelm a CSS and result in an overflow or
surcharged sewer condition. More sophisticated ways for addressing sub-daily event
characteristics would improve the results of this study. Higher temporal resolution evaluation of
hydrologic effects (e.g., an hourly time-step) will require additional precipitation data sets and/or
methods for disaggregating daily precipitation to finer time scales. In either case, the use of
hydrologic models such as SWMM would likely be required.
Snowmelt-Analysis of total daily precipitation data from the VEMAP data set does not
distinguish between rain and snow. All precipitation is treated as rainfall. Snowfall is less likely
than rainfall to result in elevated peak flows in a CSS at the time of the precipitation event, but
snowmelt at a later date can have significant impacts on sewer flows, especially if the snowmelt
occurs during a rain event. Moreover, rain on snow events can result in significant runoff, with
virtually all of the rainfall converted to runoff, supplemented by melting snow. This is a
significant limitation for our analyses of the New England and Great Lakes Regions, given the
generally cold temperatures throughout these regions during the winter months.
Future Time Period of Concern-This analysis is strongly influenced by variability in
climate change projections in time and space. In the case of the long-term (2060-2099)
projections for the Great Lakes Region, the CCCM and HADCM2 AOGCMs each predict an
increase in CSO event frequency (with respect to a historical benchmark) and the intensity of the
four-events-per-year benchmark storm event. In the New England Region, however, more near-
term (2025-2050) projections disagree in the direction of change; the HADCM2 model projected
increases and the CCCM model projected decreases in the frequency of CSO events and intensity
of the four-events-per-year benchmark event. The methodology is, thus, not able to determine if
the differences in AOGCM behavior are due to differences in the two time periods or to
differences in the two geographic regions. This is a limitation of this study that requires
additional investigation.
33
-------
This inconsistency does, however, highlight an important consideration for assessing the
future implications of climate change. On one hand, it is desirable to focus on a more distant
future when changes in climate, if present, are more likely to be detectable. One the other hand,
a focus on the near-term provides information more useful and relevant to current decision
making. Finding the right balance is important to the overall value of an assessment activity.
The work in the New England and Great Lakes Regions described in this report was conducted
as two separate projects. Work in the New England Region was done subsequent to that in the
Great Lakes Region, and the more near-term focus was selected to be more relevant to current
decision making. Unfortunately, despite the inconclusive near-term results in the New England
Region, resource limitations did not allow analysis of additional future time periods (e.g.,
2060-2099) in this region.
4.3.2. Future Research
This study is only a first step towards understanding a very complex issue, the
implications of which will vary significantly depending on location and system. The following
are suggested areas of follow-up research to this study.
Utilize More Recent Climate Mode/s-Newer AOGCM runs have become available since
the VEMAP Phase II data set was created in the mid 1990s. There are also additional
approaches for obtaining daily, time-step projections through updated weather generators or
AOGCMs that generate daily precipitation data. More up-to-date AOGCM results would reflect
advances in the state of the science, and regional models (e.g., the UK Hadley Centre's PRECIS
model) can provide better resolution on a regional basis.
Case Studies-A smaller scale case study approach looking in detail at a few communities
would provide more accurate data on system responses on an hourly basis. These case studies
would likely involve the use of detailed sewershed runoff models such as SWMM. These
models typically utilize continuous precipitation data, so a method for applying AOGCM outputs
to modify historical continuous precipitation data would need to be created. This would provide
a more robust analysis as event intensity on an hourly (or shorter time-step) basis for predicting
CSO events would provide a more accurate basis than the daily precipitation data utilized in this
screening-level analysis. The use of hydrologic models for more detailed case study assessments
would also allow for better handling of the complex hydrologic responses to climate change
including the effects of changes in snow, snowmelt, and temperature.
Determine Best Practices for Characterizing Design Storms-One opportunity to provide
useful guidance would be to establish a straightforward approach for modifying the Intensity-
Duration-Frequency (IDF) curves and design storms commonly utilized for water resource
engineering design and planning. To develop guidance, it would be useful to review (1) the
34
-------
current practices associated with creating IDF curves (e.g., how far back in the historical record,
what statistical techniques are used) and (2) the extent to which recent trends in increasing storm
intensity are already embedded in the DDF curves. One possible approach for modifying the way
that design storms are calculated would be to utilize research examining trends in precipitation
intensity over the last century. For example, research by Groisman et al. (2005) evaluated
historical precipitation data for the United States and determined that there were statistically
significant trends which indicated an increase in the intensity of the heavy (upper 5% of
precipitation) events. Specifically, their research has found a 4.6% increase in event intensity
per decade for the largest 5% of precipitation events; a 7.2% increase in event intensity per
decade for the largest 1% of precipitation events; and a 14.1% increase in event intensity per
decade for the largest 0.1% of precipitation events. Relationships like these could be used, along
with assumptions on the design lifetime of CSSs, to provide adjustment factors for characterizing
future storm intensity.
Effectiveness of Best Management Practices in CSO mitigation- The threat of increased
future precipitation variability could provide additional motivation for rigorous inflow and
infiltration (I&I) mitigation programs that maximize the capacity of the existing CSS. By
eliminating flow in sewers due to groundwater infiltration and runoff from gutter downspouts
and sump pumps, additional capacity can be made available to reduce the frequency and severity
of CSOs. Many municipalities already have aggressive I&I programs in place, and the potential
for increased precipitation intensity in the future under climate change makes this mitigation
option even more important. Moreover, a variety of stormwater best management practices
(BMPs) and low-impact design strategies can be implemented to reduce runoff peaks entering
CSSs including rain gardens, rain barrels, green roofs, and other practices. Assessment of the
design, implementation, and effectiveness of I&I programs for adapting to climate change is an
important area requiring additional study.
35
-------
5. CONCLUSIONS
This screening-level analysis assesses the potential order of magnitude of climate change
impacts on CSO mitigation in the New England and Great Lakes Regions. The purpose is to
determine whether the potential implications of climate change on CSOs in these regions warrant
further consideration and study, and secondly, to evaluate the need for decision support tools and
information enabling CSS managers to better incorporate consideration of climate change into
their decision making processes. As such, this assessment and report is only a first step towards
understanding a complex issue, the implications of which will vary significantly in different
locations and for different systems.
In the Great Lakes Region, projected long-term (2060-2099) changes in precipitation
suggest that if CSO mitigation efforts are designed based on historical precipitation values, many
systems could experience increases in the frequency of CSO events beyond their design capacity,
resulting in increases in overflow volume discharged to receiving waters. In the New England
Region, projected near-term (2025-2050) changes in precipitation are inconsistent, with
projections based on the CCCM and HADCM2 AOGCM models disagreeing on the direction of
change. This difference in direction complicates interpretation of the results and highlights
uncertainties associated with the AOGCM climate projections. The near-term results for the
New England Region are best considered inconclusive, however, neither confirming nor refuting
the likelihood of future impacts due to climate change. This analysis did not consider long-term
(2060-2099) projections of change in the New England Region—a limitation that precludes any
direct comparison of impacts in the two study regions. Future study is required to address this
issue in the New England Region.
It must be noted that the methodology used in this study involves a number of
simplifying assumptions and other limitations. Key limitations include the inherent uncertainty
in AOGCM projections, use of a weather generator to downscale to daily data, lack of
consideration of snow and snowmelt, and the use of simple scaling factors to estimate
stormwater runoff rather than detailed modeling of individual CSSs. Results are thus not
intended, nor the methods appropriate, to provide site-specific information on the potential
impacts or mitigation requirements for any individual system or facility.
Investments in water infrastructure tend to be long-term, capital-intensive, and, in many
cases, irreversible in the short- to medium-term. It is thus prudent to consider that today's
decisions could influence the ability of treatment facilities to accommodate changes in climate
for decades into the future. As indicated in this report, knowledge of future climate change at the
local scales required by CSS managers is still subject to uncertainty, particularly with respect to
changes in precipitation. Throughout much of the United States, however, there is empirical
36
-------
evidence for a trend toward increasing precipitation intensity during the last century (Karl and
Knight, 1998). Irrespective of the uncertainties in the AOGCM projections, consideration of
these changes may be warranted. Moreover, uncertainty regarding future climate change should
be considered in context. CSS managers already include other factors with similar uncertainty
and long-term effects in their design such as total impervious area, sewered area, population
served, and per capita water demand.
Faced with the prospect of future climate change, opportunities may exist where current
CSO mitigation efforts can be upgraded at little additional cost to provide an added margin of
safety to account both for near-term extreme events and the potential future effects of climate
change. No-regrets opportunities may also exist where actions taken today to address current,
other water quality concerns can provide additional benefits in the context of adapting to climate
change.
Finally, it is important to recognize that each CSS and CSS community has a unique set
of attributes, existing challenges, constraints, and other factors that must be considered in
determining what reasonable and appropriate actions should be taken to manage any increase in
risk associated with climate change. The focus of this report, CSOs, is not meant to imply that
CSOs are the single or even the greatest source of water quality impairment in these areas. Other
sources of impairment including non-point loading from agriculture, urban development, and
other sources also occur in the study regions. Moreover, CSS managers are also faced with a
range of regulatory and other challenges not related to climate change. Accordingly, responding
to climate change will require a holistic approach that considers climate change in the context of
other impacts on CSSs and regional water quality to determine what reasonable and appropriate
actions can be taken.
Although limited in scope, this screening-level analysis provides a first step towards a
better understanding of climate change impacts on CSOs in the New England and Great Lakes
Regions. Results suggest that certain systems may be vulnerable to future climate change, and
that there is a need for more detailed, site-specific analyses including the development of
decision support tools and information. Regardless of whether or not CSS managers choose to
include climate change in their long-term planning, it is preferable that the decision be
intentional and not due to lack of awareness of the problem.
37
-------
REFERENCES
Burian, SJ; Nix, SJ; Pitt, RE; et al. (2000) Urban wastewater management in the United States: past, present, and
future. J Urban Tech 7(3):33-62. Available online at
http://www.sewerhistory.org/articles/whregion/urban_wwm_mgmt/urban_wwm_mgmt.pdf.
Chiew, F; Whetton, P; McMahon, A; et al. (1995) Simulation of the impacts of climate change on runoff and soil
moisture in Australian catchments. J Hydrology 167(1):121-147.
Daly, C; Neilson, RP; Phillips, DL. (1994) A statistical-topographic model for mapping climatological precipitation
over mountainous terrain. J Appl Meteorol 33(2):140-158.
Critchley, W; Siegert, K. (1991) Water harvesting: a manual for the design and construction of water harvesting
schemes for plant production. FAO (Food and Agriculture Organization) of the United Nations, Rome, Italy: FAO
Corporate Document Repository; AGL/MISC/17/91. Available online at
http://www.fao.org/docrep/U3160E/U3160EOO.htm.
Gleick, PH. (2000) Water: the potential consequences of climate variability and change for the water resources of
the United States. The report of the Water Sector Assessment Team of the National Assessment of the Potential
Consequences of Climate Variability and Change. Oakland, CA: Pacific Institute for Studies on Development,
Economics, and Security; 151 pp. Available online at http://www.gcrio.org/NationalAssessment/water/water.pdf.
GLRA (Great Lakes Regional Assessment Group). (2000) Preparing for a changing climate: the potential
consequences of climate variability and change - Great Lakes. In: Sousounis, PJ; Bisanz; JM; eds. A summary by
the Great Lakes Regional Assessment Group for the U.S. Global Change Research Program. Ann Arbor, MI:
University of Michigan. Available online at http://www.gcrio.org/NationalAssessment/greatlakes/greatlakes.pdf.
Groisman, PY; Knight, RW; Easterling, DR; et al. (2005) Trends in intense precipitation in the climate record. J
Climate, 18:1326-1350.
IPCC (Intergovernmental Panel on Climate Change). (2001) Climate change 2001: the scientific basis. In:
Houghton, JT; Ding, Y; Griggs, DJ; et al, eds. Contribution of Working Group I to the Third Edition of the Third
Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University
Press. Available online at http://www.grida.no/climate/ipcc_tar/wgl/index.htm.
IPCC (Intergovernmental Panel on Climate Change). (2007) Climate change 2007: the physical science basis. In:
Solomon, S; Qin, D; Manning M; et al., eds. Contribution of Working Group I to the Fourth Assessment Report of
the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press. Available online at
http://www.ipcc.ch/ipccreports/ar4-wgl.htm.
Karl, TR; Knight, RW. (1998) Secular trends of precipitation amount, frequency, and intensity in the United States.
Bull Am Meteorolog Soc 79(2) :231-241.
NAST (National Assessment Synthesis Team). (2000) Climate change impacts on the United States: the potential
consequences of climate variability and change. National Assessment Synthesis Team, US Global Change Research
Program. Cambridge, UK: Cambridge University Press. Available online at
http://globalchange.gov/pubs/nast_2000.html.
Richardson, CW; Wright, DA. (1984) WGEN: A model for generating daily weather variables. Washington, DC:
United States Department of Agriculture, Agriculture Research Service, ARS-8. 83 pp. Available online at
http://soilphysics.okstate.edu/software/cmls/WGEN.pdf.
38
-------
Rosenbloom, NA; Kittel, TGF; Kaufman, C; et al. (2003) A user's guide to the VEMAP Phase II Database.
VEMAP (The Vegetation/Ecosystem Modeling and Analysis Projects). Available online at
http://www.cgd.ucar.edu/vemap/users_guideV2.html Updated August 2003.
Schladweiler, JC. (2005) Tracking down the roots of our sanitary sewers: design choices and philosophies. Arizona
Water Pollution Control Association. Sewerhistory.org. Available online at
http://www.sewerhistory.org/chronos/design_choices.htm.
U.S. Department of State. (2002) U.S. Climate action report 2002. Third national communication of the United
States of America under the United Nations framework convention on climate change. U.S. Global Change
Research, Washington, DC. Available online at http://www.gcrio.org/CAR2002/.
U.S. EPA (Environmental Protection Agency). (1994) Combined sewer overflow (CSO) control policy. Federal
Register 59(75): 18688-18698. Available online at http://www.epa.gov/npdes/pubs/owm011 l.pdf.
U.S. EPA (Environmental Protection Agency). (2004) Report to Congress: impacts and control of CSOs and SSOs.
Office of Wastewater Management, Washington, DC; EPA/833/R-04/001. Available online at
http://cfpub.epa.gov/npdes/cso/cpolicy _report2004.cfm.
U.S. EPA (Environmental Protection Agency). (2006) Water permitting 101. Office of Wastewater Management,
Washington, DC. Available online at http://www.epa.gov/npdes/pubs/101pape.pdf.
Wilkinson, B; Balmforth D. (2004) Effects of climate change on sewer system performance. The United Kingdom
Water Industry Research and MWH UK, Ltd. Available online at
http://www.wapug.org.uk/pastjapers/Dunblane_2004/D2004wilkinson.pdf.
39
-------
United States
Environmental Protection
Agency
National Center for Environmental Assessment
Office of Research and Development
Washington, DC 20460
Official Business
Penalty for Private Use
$300
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERM IT NO. G-35
------- |