PA
United States
Environmental Protection
Agency
finking Water
Third Report to Congress
rv ,^\ Printed on recycled paper
-------
Drinking Water
Infrastructure Needs Survey
and Assessment
Third Report to Congress
I
ft
iAiii
June 2005
U.S. Environmental Protection Agency
Office of Water
Office of Ground Water and Drinking Water
Drinking Water Protection Division
Washington, DC 20460
-------
This document is available for sale to the public through:
. Educational Resource information Center
Clearinghouse for Science, Mathematics, and
Environmental Education ^ERIC/CSMEE)
1929 Kenny Road
Coiumbus>OH 43210-1015
1-800-276-0462
-or-
Nationa! Technical Information Service
5285 Port Royal Road
Springfield,VA22161
1-80Q-533-NTIS or 1-703-487-4650
This document is also available efectronicaiiy on the
internet at www.epa.gov/safewater/needs.htmi
-------
EXECUTIVE SUMMARY 1
Methods for the Assessment 2
Methods Used to Assess State Needs 2
Methods Used to Assess American Indian and Alaska Native Village Water System Needs 3
Models for Assigning Costs to Projects Without Costs 4
Total National Need 4
Total Need Compared to Previous Needs Assessments 5
Total Need: System Size and Type 6
Total Need: Current and Future 6
Total Need: Project Type , 7
The Regulatory Need 9
Security Needs 10
Needs for Small Water Systems 11
Needs of American Indian and Alaska Native Village Water Systems.... 11
Challenges for Future Assessments 11
Conclusions 12
SCOPE OF THE ASSESSMENT AND SURVEY METHODS 13
Scope of the Assessment 13
Assessment Methods 16
Conducting the State Survey for Large and Medium Systems 16
Improvements for the 2003 Needs Assessment of Medium and Large Systems 16
Method for Estimating the Small System, Not-for-Profit Noncommunity System, and American Indian and
Alaska Native Village Need 17
Documented Costs and Cost Models 18
Information Quality 19
FINDINGS 23
Total 20-Year National Need 23
Current and Future Needs 24
Total Need by Project Type ,. 26
The Regulatory Need 29
Existing Regulations 30
Security Needs 31
Community Water Systems Serving 10,000 and Fewer People 33
Not-for-Profit Noncommunity Water Systems 33
American Indian and Alaska Native Village Water System Need 34
American Indian Water System Needs 35
Alaska Native Village Water System Needs 35
Total Need Compared to Previous Needs Assessments 36
APPENDIX A—PAYING FOR AND FINANCING INFRASTRUCTURE IMPROVEMENTS 39
APPENDIX B—METHODS: SAMPLING AND COST MODELING 43
APPENDIX C—SUMMARY OF QUALITY ASSURANCE PROCEDURES 51
APPENDIX D—SUMMARY OF FINDINGS 57
APPENDIX E—SUMMARY OF FINDINGS FOR SYSTEMS SERVING 10,000 AND FEWER PEOPLE 67
APPENDIX F—GLOSSARY 69
-------
These photographs show the construction and completion of a tank in Bartle.sville. Oklahoma.
-------
Exhibit ES-1:1999 versus 2003 Current and Future Need 6
Exhibit ES-2: Total 20-Year Need 7
Exhibit ES-3: Total 20-Year Need by Project Type , 8
Exhibit ES-4:20-Year Regulatory and Non-Regulatory Need 9
Exhibit ES-5:20-Year Regulatory Need 10
Exhibit ES-6: Total 20-Year American Indian and Alaska Native Village Water System Need by Project Type .... 11
Exhibit 1: Overview of Needs by System Size and Type 24
Exhibit 2: Overview of Needs by State 25
Exhibit 3: Total Need by Project Type 27
Exhibit 4:20-Year Total Regulatory and Non-Regulatory Need 29
Exhibit 5:20-Year Regulatory Need 30
Exhibits: Security Needs by Project Type 32
Exhibit 7: Total 20-Year Need for Not-for-Profit Noncommunity Water Systems by Project Type 33
Exhibit 8: Total American Indian and Alaska Native Village Water System Need by Project Type 34
Exhibit 9: Total 20-Year Need by Project Type for American Indian Water Systems 35
Exhibit 10: Total 20-Year Need by Project Type for Alaska Native Village Water Systems 35
Exhibit B-1: Size Category Definitions 44
Exhibit B-2: Community Water System Sample Sizes 46
Exhibit B-3: Conventional Treatment Plant Project Cost Curve 48
Exhibit D-1: Total Need for Water Systems in the States by Project Type 58
Exhibit D-2: Current Need for Water Systems in the States by Project Type 59
Exhibit D-3: Total Need for Water Systems in the States by System Size 60
Exhibit D-4: Current Regulatory Need for Water Systems in the States 61
Exhibit D-5: Total Existing Regulatory Need for Water Systems in the States 62
Exhibit D-6: Total Need for American Indian and Alaska Native Village Systems by EPA Region 63
Exhibit D-7: Total Need by Project Type for American Indian and Alaska Native Village Water Systems 64
Exhibit D-8: Total Existing Regulatory Need for American Indian and Alaska Native Village Water Systems 65
Exhibit D-9: Total Proposed and Recently Promulgated Regulatory Need 66
Exhibit E-1: Total Need for Systems Serving 10,000 and Fewer People 68
-------
s
Fort Peck-Dry Prairie Regional Water Svstetn. This raw water intake pump station will drcnv art
average daily flow of 5.5 million gallons per day (MOD) from the Missouri River to serve 22
communities in northeastern Montana. These communities are currently served hv individual
wells with poor water quality.
-------
ACKNOWLEDGMENTS
Many dedicated individuals contributed to the 2003 Drinking Water Infrastructure Needs Survey and
Assessment. We would like to thank the state and EPA Needs Assessment Coordinators and their supporting
staff and contractors for active participation and continuing interest in the project. Not listed are the operators
and managers of the approximately 4,000 systems who spent their valuable time completing the
questionnaires we sent to them. We thank them for their assistance.
MarkSpinale-U.S. EPA Region 1
Theodore Dunn, Raymond Jarema-Connecticut
Shannon Gallant, Roger Grouse-Maine
Michael Maynard, Jack Hamm-Massachusetts
Richard Skarinka-New Hampshire
Gary Chobanian, Romeo Mendes-Rhode Island
Rodney Pingree-Vermont
Raymond Kvalheim-U.S. EPA Region 2
Roger Tsao-New Jersey
Steve Marshall-New York
Eva Hernandez, Olga Rivera-Puerto Rico
Christine Lottes-Virgin Islands
Don Niehus-U.S. EPA Region 3
George Rizzo-U.S. EPA Region 3, District of
Columbia
Ed Hallock-Delaware
Daniel Piasecki, Saeid Kasraei, Chris Carski-
Maryland
Renee Bartholomew, DanaAunkst-Pennsylvania
Thomas B. Gray-Virginia
Patrick Taylor-West Virginia
Walter Hunter, Dale Froneberger-U.S. EPA Region 4
Eric Holt-Alabama
Craig Diltz, Tim Banks-Florida
Aileen Trotter-Georgia
Carol Moore-Kentucky
Sheila Williams, Keith Allen-Mississippi
Sid Harrell-North Carolina
Tom McDonough-South Carolina
Khaldoun Kailani-Tennessee
William Tansey-U.S. EPA Region 5
Rick Cobb, Eric Portz-lllinois
Shelley Love, Cortney Stover-Indiana
Richard Benzie-Michigan
Brian Noma-Minnesota
Stacy Barna-Ohio
James Witthuhn-Wisconsin
Tom Poeton-U.S. EPA Region 6 and New Mexico
Raymond Thompson-Arkansas
Julie Comeaux-Louisiana
Robert K. Mullins-Oklahoma
Fawn McDonald-Texas
Kelly Beard-Tittone-U.S. EPA Region 7
Roy Ney-lowa
William Carr-Kansas
Stephen Jones-Missouri
Larry Steele-Nebraska
Minnie Moore Adams, Adrienne Rivera-U.S. EPA
Region 8 and Wyoming
John Payne-Colorado
Marc Golz, Gary Wiens-Montana
Chuck Abel-North Dakota
Jim Anderson-South Dakota
Tim Pine-Utah
Jose Caratini, Sara Jacobs-U.S. EPA Region 9
Susan Cox, Barry Pollock-U.S. EPA Region 9, Pacific
Islands
Moncef Tihami-Arizona
Uyen Trinh-Le, Robin Hook-California
Angel Marquez-Guam
Bill Wong, Denise Dang-Hawaii
Adele Basham-Nevada
Richard Green-U.S. EPA Region 10
Sherri Trask, James Weise-Alaska
Nancy Bowser-Idaho
Mike Grimm-Oregon
Peter Beaton-Washington
EPA Office of Water
David Travers, Lisa Almodovar and Robert Barles-
Needs Assessment Coordinators
Chuck Job-Drinking Water Infrastructure Branch
Chief
The Cadmus Group, Inc.-Prime Contractor
-------
In 2003, the U.S. Environmental Protection Agency conducted Us third assessment1 of the
nation's public water system infrastructure needs. The total national need for drinking water
investment is $276.8 billion over the next 20 years. The 2003 Needs Assessment documents
the continued need to install, upgrade, and replace the infrastructure on which the public relies
for safe drinking water.
The U.S. Environmental Protection Agency's (EPA's or
"the Agency's") third national assessment of public
water system infrastructure needs shows total
investment needs of $276.8 billion over the next 20
years. This document, the Third Report to Congress,
conveys the results of the 2003 Drinking Water
Infrastructure Needs Survey and Assessment and
covers the 20-year period from January 1. 2003,
through December 31, 2022.
The national total comprises the infrastructure
investment needs of the nation's approximately
53,000 community water systems2 and 21,400 not-
for-profit noncommunity water systems3 found in all
50 states, Puerto Rico, the Virgin Islands, the Pacific
island territories, and the District of Columbia.
American Indian and Alaska native village water
systems are also included in the total need. Among
the needs reported in the 2003 Needs Assessment
are projects to protect public health, to preserve the
physical integrity of water systems, to convey treated
water to homes and commercial and industrial
establishments, and to ensure continued compliance
with specific Safe Drinking Water Act (SDWA or "the
Act") regulations.
Public water systems continually install, upgrade, and
replace the infrastructure on which the public depends
for safe drinking water. Projects reported in the 2003
Needs Assessment range from replacement of
short sections of deteriorated water mains to
construction of large-scale, state-of-the-art
treatment plants that produce drinking water from
sea water. Many projects were identified as current
needs; many more projects will arise over the next
20 years as existing infrastructure reaches the end
of its useful life.
The cost of infrastructure investment is borne
primarily by water system customers in the form of
water rates. However, general revenues from
federal, state, and local governments may
supplement revenues from users. For major capital
improvements, long-term financing is often critical; it
allows communities to spread out the cost of
improvements over the expected life of a project,
thereby allocating the costs to those customers who
Sections 1452(h) and 1452(i)(4) of the Safe Drinking
Water Act direct EPA to conduct an assessment of
drinking water infrastructure needs every 4 years.
The results are used to allocate Drinking Water
State Revolving Fund monies to the states and
tribes. In partnership with tfte states, EPA
undertakes a survey of drinking water utilities as a
basis for the Agency's assessment. EPA conducted
prior surveys in 1995 arid 1999.
1 EPA's previous assessments of infrastructure need in 1995 and 1999 were called "Needs Surveys" because the assessment relied primarily on
survey methods. In 2003, EPA relied in part on surveys but a!so on analysis of previous survey data. Accordingly, the term "assessment" is more
appropriate. Hereinafter, these studies will be referred to as "Needs Assessments.'
2 A community water system is a public water system that serves at least 15 connections used by year-round residents or that reguiarly serves
at least 25 residents year-round. Cities, towns, and even small communities such as retirement homes are examples of community water
systems.
s A noncommunity water system is a public water system that is not a community water system and that serves a nonresidential population of at
least 25 individuals or 15 service connections daily for at least 60 days of the year. Schools and churches are examples of noncommunity water
systems.
-------
Page 2
2003 Drinking Water Infrastructure Needs Survey and Assessment
benefit from the improvements. Despite the
importance of these projects for protecting public
health, some utilities may encounter difficulties in
obtaining affordable financing for such improvements.
The Drinking Water State Revolving Fund (DWSRF)
was established by Congress in the 1996 SDWA
Amendments to help public water systems obtain
financing for improvements necessary to protect
public health and comply with drinking water
regulations. Between FY1997 and FY2004, Congress
appropriated $6.96 billion for the DWSRF. The
DWSRF is one of many local, state, and federal
programs that water systems can use to supplement
Between FY1997 and FY20Q4, Congress
appropriated more than $6.96 billion for the
DWSRF program. Through June 30, 2004, states
had received $5.7 billion in capitalization grants,
which, when combined with state match, bond
proceeds, loan repayments, and other funds,
made for a total of $9.7 billion in funds available for
loans. As of that date, states fcad made close to
\ 3,700 loans totaling $8.0 billion, leaving $1J billion,
which had not yet been allocated to Joans- The
total assistance provided represented 166 percent
of the awarded federal grants or 83 percent of the
total funds available.
user fees and help finance large-scale capital
investments. Appendix A provides a more detailed
discussion of financing for water system
improvements in the context of sustainable
infrastructure.
As mandated by the SDWA, EPA uses the results of
the most recent infrastructure needs assessment to
allocate DWSRF funds to the states based on their
share of the total national need, with each state
receiving at least 1 percent of the available DWSRF
funds. For example, the 1999 Needs Assessment
found 22 states and the District of Columbia each had
less than 1 percent of the total national need (in
aggregate, 11.3 percent of the total national need).
However, from 2002 to 2005, each of these states
were eligible for 1 percent of the annual DWSRF
allotments (or, in aggregate, 23 percent of the total
DWSRF allotment). The discrepancy may be due, in
part, to a number of these states participating in the
needs assessments effort to a lesser degree than the
other states.
Eligible projects are funded according to each state's
priority system, consistent with public health criteria
specified in the SDWA. EPA also uses the
assessment results to allocate the tribal set-aside (up
to 1.5 percent of the DWSRF annual appropriation) for
American Indian and Alaska native village water
systems.
Methods for the Assessment
The approach for the 2003 Needs Assessment was
developed by EPA in consultation with a workgroup
consisting of representatives of the states and EPA
Regions. The state/EPA workgroup refined the
methods used for medium and large water systems in
1995 and 1999 based on lessons learned from these
assessments and options made available from
technological advancements in the Internet. To
account for the needs of small community water
systems, EPA adjusted the 1999 Needs Assessment
findings to January 2003 dollars and reallocated the
needs to states based on the current inventory of
small systems. The needs for not-for-profit
noncommunity water systems, American Indian water
systems, and Alaska native village water systems
were based on the 1999 Needs Assessment findings
adjusted to January 2003 dollars.
Methods Used to Assess State Needs
Medium and Large Systems. EPA used
questionnaires to collect data on infrastructure needs
from medium and large water systems (see
Appendix B for a discussion of different system size
categories). EPA sent questionnaires to all of the
nation's 1,041 large water systems (those that serve
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page 3
over 50,000 people) and aii 301 of the medium
systems that serve between 40,001 and 50,000
people. This census included 1,342 systems.
Questionnaires were also sent to a random sample of
2,553 of the 7,337 systems serving 3,301 to 40,000
people. Approximately 96 percent of all questionnaires
were completed and returned.
Questionnaires for most systems were returned by
systems to their state contacts, who reviewed the
information for completeness and then added projects
or improved documentation of projects as needed. In
some cases, states completed the questionnaires for
the systems. States then forwarded their amended
questionnaires to EPA for review and tabulation. EPA
reviewed al! 128,600 projects submitted to ensure that
each met strict documentation requirements and
were allowable DWSRF projects. This individual
project review resulted in removal o? 23,600 projects
due to ineligibility or inadequate documentation. States
were given the opportunity, through an interactive Web
site, to provide additional information on projects for
EPA consideration.
Smaii Systems. Smal! systems serving populations
of 3,300 or fewer have often lacked the staff and
planning documents needed to respond to the
questionnaire. Therefore, for the 1999 Needs
Assessment, EPA conducted site visits to identify and
document their infrastructure needs. Site visits were
conducted at 599 of the approximately 45,000 small
community water systems and at 100 of the
approximately 21,400 not-for-profit noncommunity
systems.
Because these data were collected on site by EPA
using consistent and comprehensive system
interview tools, there was a high level o? confidence in
the findings. In addition, the small system need from
the 1995 Needs Assessment, also collected using
EPA site visits, was comparable to the findings in the
1999 Needs Assessment, indicating that the need
was properly identified and did not decrease over
time.
This man in a native village in AiaskajUls several
containers frotn (his watering point to §nt drinking
water for -hivfe/m ify:
For these reasons, EPA used the 1999 data to
estimate small system need. The Agency determined
an average cost per system for each of several strata
(based on population and source type) from the 1999
data. The Agency then adjusted this cost to 2003
dollars and reallocated the small system need to each
state based on the number of small systems active at
the time of the 2003 Needs Assessment.
Methods Used to Assess American Indian and
Alaska Native Village Water System Needs
For many of the same reasons that apply to other
sma!! systems, the 1999 questionnaires for small
American Indian systems were completed during on-
site visits with information provided by EPA and the
Indian Health Service (IMS). AH 19 American Indian
systems serving more than 3,300 people completed a
questionnaire and were provided technical support
upon request, EPA estimated Alaska native village
water system needs by census, using key personnel
and data resources made available by representatives
of the Alaska Native Health Consortia. the IMS, and
Village Safe Water. Because of the high level of
confidence in the 1999 findings, EPA adjusted the
need from the 1999 Needs Assessment for American
Indian and Alaska native village systems from 1999
dollars to 2003 dollars, and used that estimate for this
2003 Needs Assessment.
-------
Page 4
2003 Drinking Water Infrastructure Needs Survey and Assessment
Examples of Cost Modeling
When modeling the cost of construction of a
complete conventional water treatment plant,
items included are:
«* Coagutatiofi, fioceuiation, sedimentation.
filtration, waste handling, and the
building;
» Ail raw and finished water pumps;
» The finished water clearwell and
disinfection;
9 Tanks;
* Process control system and buiteling;
ami
* Engineering design and contingencies,
When modeling the cost of replacement of
distribution mains, components included are;
* Pipe cost trenching, becfding, backfill,
hydrant^ valves, road repair, easements,
and service leads from the main to the
curb stop; and
» Engineering design and contingencies.
When modeling the cost of construction of a
storage tank, items included are:
» Tank;
» Ali appurtenances including piping, water
level controls, and valves; and
* Engineering design and contingencies.
Models for Assigning Costs to Projects
Without Costs
During the 1999 Needs Assessment, EPA invested
considerable effort in obtaining project cost
information from data submitted by systems. With this
cost information, models were developed for nearly all
types of projects included in the assessment. For
2003. most of those project costs were not expected
to change beyond typical adjustments for inflation.
except for automated meter reading devices for
domestic water meters and the cost of pipe
installation and rehabilitation. The workgroup
determined that efforts for 2003 should focus on other
areas of the assessment, and that most of the 1999
cost models could be adjusted to 2003 dollars. The
"cost modeling" text box discusses the components
of three types of cost models. Appendix 8 provides
more detail on the cost models used for the
assessment.
EPA did develop new cost models for automated
meter reading projects and for transmission and
distribution pipe installation and rehabilitation using
2003 project data. The new pipe models were
developed using the same method as those used for
the 1999 Needs Assessment. The 2003 meter model
reflects the expected increase in cost to
accommodate new, more efficient technology.
Total National Need
The 2003 Needs Assessment found that the nation's
water systems need to invest $276.8 billion over the
next 20 years in order to continue to provide clean and
safe drinking water to their consumers. The need
includes installation of new infrastructure as wel! as
rehabilitation or replacement of deteriorated or
undersized infrastructure. It also includes the need to
address aging infrastructure that is adequate now but
will require replacement or significant rehabilitation
over the next 20 years.
Most of the needs are not related to violations of any
SDWA regulations. Instead, they are ongoing
investments that systems need to make to continue to
deliver water to their customers, as well as to remain
in compliance with regulations.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Pages
Total Need Compared to Previous
Needs Assessments
The 1995 and 1999 Needs Assessments estimated
the total national need at $167.4 and $165.5 billion
respectively." The findings of this assessment
estimate a need of $276.8 billion, exceeding the
previous assessments' national need by more than 60
percent.
The methods used to collect and evaluate needs in
EPA's 2003 Needs Assessment remained largely
unchanged from those used in 1995 and 1999, except
for an emphasis on capturing previously
underreported needs for infrastructure rehabilitation
and replacement.5 EPA recognized the necessity to
more accurately capture these infrastructure needs.
This objective is consistent with EPA's initiative for
"sustainable infrastructure," (See Appendix A) which
emphasizes improved management of assets,
including collection of better data on infrastructure
condition, and long-term planning for rehabilitation and
replacement. For the 2003 Needs Assessment, it is
likely that a more systematic approach to asset
identification and evaluation led some systems and
states to consider and report a larger number of
replacement and rehabilitation projects. EPA has
some anecdotal evidence that states began to
investigate the backlog of projects that had been
deferred in the past.
Systems' and states' efforts to correct underreporting
appear to have been successful. States reported
many more projects (covering all types of need) in
While the 2003 Needs Assessment estimate
represents a substantial increase in Reed from the
previous assessments, it is still witnm the range
identified in other reports.
»156B/J1678
EPA 11)85 /
ss9*5tnents
$4203
WIN Estimate
CEO Estimate
$3038 (point est.)
EPA's "Clean Water and Drinking Water
Infrastructure Gap Analysis" estimated
drinking water systems* 20-year capital
needs within a range of $170 to $493 billion,
with a point estimate of $303 billion.6
The Congressional Budget Office (CBO)
report "Future Investment in Drinking Water
and Wastewater Infrastructure," estimates
annual water system needs of $12,2 to
$21.2 billion, which would extrapolate to a
20-year total need in the range of $245 to
$424 billion/
The Water Infrastructure Network's (WiN's)
"Clean and Safe Water for the 21st Century.
A Renewed National Commitment to Water
and Wastewater Infrastructure/ estimates
water system needs of $21 billion annually,
which extrapolates to $420 billion over 20
years.8
* The 1895 and 1999 total needs have been converted to January 2003 dollars for comparison purposes. The 1995 need in 1995 dollars was
$138.4 billion. Tne 1999 need in 1999 dollars was $150.9 billion.
* In the 1999 Needs Assessment, EPA noted the problem of underreporting. Quality assurance reviews of data from 1995 confirmed this. For a
comparison of the 1999 EPA Needs Assessment with other estimates, see Congressional Budget Office, op. cii., Chapter 2.
6 U.S. Environmental Protection Agency, "Clean Water and Drinking Water Infrastructure Gap Analysis," (September 2002), p. 5. Needs were
assumed to be in 1999 dollars based on the date of the report and planning period used. Needs have been adjusted to 2003 dollars for
comparison purposes.
7 Congressional Budget Office. "Future Investment in Drinking Water and Wastewater infrastructure." {November 2002), p. ix. Needs were
reported in 2001 dollars and have been adjusted to 2003 dollars for comparison purposes.
E Water Infrastructure Network, "Clean and Safe Water for the 21 st Century - A Renewed National Commitment to Water and Wastewater
Infrastructure," (undated), p. 3-1. Needs were assumed to be in 19S9 dollars based on the planning period and data used. Needs have been
adjusted to 2003 dollars for comparison purposes.
-------
Page 6
2003 Drinking Water Infrastructure Needs Survey and Assessment
2003 than in the previous assessments. In the 1999
Needs Assessment, there were 61,400 projects for all
large and medium systems. In the 2Q03 Needs
Assessment, there were 128,600 projects for ail large
and medium systems. Equally important, the largest
increase in 2003 (both in doiiars and in percentage)
compared to previous assessments came in future
needs. Current needs increased by about 50 percent,
but future needs rose by over 100 percent as shown
in Exhibit ES-1.
$200.0
$150.0
$100,0
S50.0
SO.O
Current
For this comparison, the 1999 Needs Assessment results have been
adjusted to January 2003 cioiiars-
This increase suggests the 2003 Needs Assessment
was more complete in capturing the longer term
needs to address aging infrastructure that is currently
adequate, but will require replacement or significant
rehabilitation over the next 20 years. While EPA
cannot confirm that systems reported all of their 20-
year needs, the increase in both the number of
projects and the total need indicates much of the
underreporting was eliminated.
The Agency's objective to better capture the true 20-
year need did not outweigh the primary imperative to
maintain the credibility of the assessment and
determine the need of individual states. EPA made a
considerable effort to ensure that the 2003 Needs
Assessment retained the stringent documentation and
eligibility requirements of both of the previous
assessments. In addition, the 2003 Needs
Assessment incorporated further quality assurance
measures to prevent over-reporting of needs.
Total Need: System Size and Type
As shown in Exhibit ES-2, the nation's 1,041
largest community water systems (those serving
populations more than 50.000 people) account
for $ 122.9 billion, or 44 percent, of the total
national need. Medium and small community
water systems also have substantial needs of
$103.0 billion and $34.2 billion, respectively.
These figures include the needs for small,
medium, and large systems in the Pacific island
territories and Virgin Islands, which are $509.1
million and $172.6 million, respectively. Not-for-
profit noncomrnuniiy water systems have
infrastructure needs of $3.4 billion. American
Indian water systems need $1.3 billion in
infrastructure improvements, while Alaska native
village systems need $1.2 billion.9
Total Need: Current and Future
The 2003 Needs Assessment differentiates "current
needs" from "future needs;" the definitions of these
two types of needs, as well as examples, are
described below. About 60 percent of the total needs,
$165.0 billion, are identified as current needs. In
Appendix D, Summary of Findings, Exhibits D-2 and
D-7 present a breakdown of current needs by project
type. Although current needs have increased in dollars
from previous assessments, they are a smaller
percentage of the total need in 2003 (60 percent,
compared with 68 percent in 1999). As discussed
! These estimates slightly exceed the total $2.4 billion American Indian and Alaska native village system need because of rounding.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page?
above, this is evidence of successful efforts to more
accurately capture "future needs."
Current Needs. Current needs are projects that a
system considers a high priority for near-term
implementation to enable a water system to continue
to deliver safe drinking water. For instance, a system
may have had numerous leaks and breaks in a
section of main that should be replaced before a
major main break occurs and inhibits the delivery of
safe drinking water.
A system with current needs is not necessarily in
violation of any health-based drinking water standard
or in the midst of responding to
an emergency. For example, a
surface water treatment plant
may currently produce safe
drinking water, but the plant's
filters may require replacement
because of their declining
effectiveness. By replacing the
filters the plant would be able to
continue providing safe water
and avoid emergency situations.
accommodate future growth (e.g. extension of service
lines to new housing developments). However, for
both current and future needs, the 2003 Needs
Assessment did include DWSRF-eligible projects that
had reasonable accommodation for expansion of
capacity that is consistent with the design life of the
infrastructure (e.g., replacing deteriorated 6-inch pipe
with new, and larger capacity, 12-inch pipe).
Total Need: Project Type
Every project in the 2003 Needs Assessment belongs
to one of five categories of need: transmission and
distribution, treatment, source, storage, or "other."
Future Needs. Future needs
are projects that water systems
do not currently need, but would
expect to address in the next 20
years as part of routine
rehabilitation or replacement of
infrastructure because of
predictable events, e.g., reaching
the end of a facility's service-life.
Approximately 40 percent of the
total need, $111.8 billion, is
reported as future needs.
Growth-Related Needs. To be
consistent with the eligibility
requirements for the DWSRF,
the 2003 Needs Assessment did
not include projects that would be
undertaken solely to
System Size and Type
Large Community Water Systems
(serving over 50,000 people)1
Medium Community Water Systems
(serving 3,301 to 50,000 people)1
Small Community Water Systems
(serving 3,300 and fewer people)1' 2
Costs Associated with the Recently Promulgated Arsenic Rule3
Not-for-profit Noncommunity Water Systems"
American Indian and Alaska Native Village Water Systems4'5
Subtotal National Need
Costs Associated with Proposed and Recently Promulgated
Regulations (Taken from EPA Economic Analyses)
Total National Need
BS8888838S88888888^a»p88888a
Need
SI 22.9
$103.0
S34.2
$0.9
$3.4
$2.4
$266.3
$9.9
S276.8
Note: Numbers may not lota! due to rounding.
' Ooes not include the costs associated with the recently promulgated Arsanic Fiuie and proposed or
recently promulgated SDWA regulations; these costs are included on a separate line in tiiis table.
2 1999 Needs Assessment findings adjusted to January 2003 dollars and reallocated based on 2003
inventory of smail systams.
3 Does not include costs for American Indian and Alaska native village water systems to comply with
the recently promulgated Arsenic Rule; these costs ars incorporated in lha estimate for American
hdian and Alaska native village water systems.
4 199S Needs Assessment findings adjusted to January 2003 dollars.
5 [nciudes cost for compliance with the recently promulgated Arsenic Rule.
- - - - - - - -----
-------
Page 8
2003 Drinking Water Infrastructure Needs Survey and Assessment
Exhibit ES-3 illustrates the total 20-year need by
category based on project type.
• Transmission and Distribution. With $183.6
billion needed over the next 20 years,
transmission and distribution projects
constitute the largest category of need,
accounting for almost two-thirds of the total
need. Little of this category of need is related
to any federal mandate. Instead, utilities need
to install and maintain distribution systems to
provide potable water to their customers while
preventing contamination of that water prior to
delivery. Although treatment plants or elevated
storage tanks are usually the most visible
components of a water system, most of a
system's infrastructure is underground in the
form of transmission and distribution mains.
Failure of transmission and distribution mains
can interrupt the delivery of water leading to a
loss of pressure, possibly allowing a backflow
of contaminated water into the system. Broken
transmission lines also can disrupt the
treatment process. The transmission and
distribution category also comprised the
largest proportion of the total need in the 1995
and 1999 Needs Assessments. Its increased
share of the total in 2003 reflects EPA's
emphasis on fully capturing previously
Exhibit ES-3; Total 20-Year Need by
• ••-. .. • • ........ . : :•, •. ,,•• ... : - •--• •
(iii biiUbp of Jariuaty 2
Transmission .;^8§|
jin«t DistrifcvSfe>n
I-S8S6
Note- Numbers may not iotal due to rounding.
underreported rehabilitation and replacement
needs, most of which were in this category.
The underreporting in the 1995 and 1999
Needs Assessments was due in part to the
limitations of planning documents. The
transmission and distribution category
includes the installation and rehabilitation of
raw and finished water transmission mains
and distribution mains and replacement of lead
service lines, flushing hydrants, valves,
meters, and backflow prevention devices.
• Treatment. Treatment projects represent the
second largest category of need, $53.2 billion,
nearly one-fifth of total need, over the next 20
years. This category consists of projects
needed to reduce contaminants through
treatment processes such as filtration,
disinfection, corrosion control, and aeration.
The installation, upgrade, or rehabilitation of
treatment infrastructure also enables removal
of contaminants that can cause chronic health
effects or taste, odor, and other aesthetic
problems.
• Storage. The total 20-year need for storage
projects is $24.8 billion. This category includes
projects to construct new or rehabilitate
existing finished water storage tanks.
Construction of new tanks is necessary if the
system cannot provide adequate flows and
pressure during peak demand periods. Many
projects in this category involve rehabilitating
existing tanks to prevent structural failures or
sanitary defects that can allow microbiological
contamination.
« Source. The source category includes
projects that are necessary to obtain safe
supplies of surface water or ground water. The
infrastructure needs in this category include
the installation and rehabilitation of drilled wells
and surface water intakes. The total 20-year
needs for source water projects are $12.8
billion.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page 9
• Other. Other needs account for an estimated
$2.3 billion. This category captures needs that
cannot be assigned to one of the prior
categories. Examples include emergency
power generators not associated with a
specific system component, computer and
automation equipment, and projects for
system security,
The Regulatory Need
The SDWA requires that public water systems meet
national standards to protect consumers from the
harmful effects of contaminated drinking water.
Although ail of the infrastructure projects included in
the 2003 Needs Assessment promote the SDWA's
public health objectives, most are driven by the need
to provide an essential service to the utility's
customers. However, some of the projects are
directly attributable to specific SDWA regulations.
This report refers to these needs collectively as the
"regulatory need," The total regulatory need is divided
into two broad categories: the need associated with
existing SDWA regulations, and the need associated
with recently promulgated and proposed regulations.
The second category accounts for new or proposed
regulations that may impact systems in the near
future, even though systems have not yet
determined the extent to which they will need
capita! investment to achieve compliance. As
shown in Exhibit ES-4, the total regulatory need
is $45.1 billion, or only 16 percent of the total
national need.
While most of the total need is not driven by
compiiance with a particular regulation, properly
maintaining a system's infrastructure is not only
economical in the long run, but also is protective
of public health. These nonregulatory costs
include routine installation, upgrade, and
replacement of basic infrastructure and are
borne by the system regardless of regulations.
Existing SDWA Regulations. The estimated needs
directly associated with existing SDWA regulations
(including the recently promulgated Arsenic Rule that
will be effective in January 2006) are $35.2 billion. The
total capital cost of compliance with the recently
promulgated Arsenic Rule (from the Economic
Analysis for the final rule) was included in this category
because state-specific occurrence data were
available, allowing EPA to allocate costs to states.
Exhibit ES-5 displays the regulatory need by existing
regulation and differentiates between current and
future needs.
Microbial Contaminants. Projects that address
microbiologicai contamination comprise 86 percent, or
$30.2 billion, of the total existing regulatory need.
Under the SDWA, the Surface Water Treatment Rule
(SWTR), the Interim Enhanced Surface Water
Treatment Rule (IESWTR), and the Total Coliform
Rule (TCR) are designed to remove or inactivate
mtcrobiai contaminants in drinking water. Microbial
contaminants, such as Giardia and E. colt, can cause
acute gastrointestinal illness and, in extreme cases,
death. The installation of a treatment plant to filter a
surface water source or the replacement of an aging
disinfection system are examples of needs in this
category.
Exhibit ES-4; 20-Year R$gMlat9|y and
Non-Regulatory Need
(Iii5|te
Need
"
20-Year
Maecf
$231,
Note. Numbers may nol total due to rounding.
-------
Page 10
2003 Drinking Water Infrsstruclure Needs Survey and Assessment
Regulations
Existing SDWA Regulations
Interim Enhanced Surface Water Treatment
Rule and Surface Water Treatment Rule1
Total Coliform Rule1
Nitrate/Nitrite Standard1
Costs Associated with the Recently
Promulgated Arsenic Rule
Lead and Copper Rule
Total Trihalomethanes Standard
Other Regulations2
Subtotal National Need
Costs Associated with Proposed and
Recently Promulgated Regulations (Taken
from EPA Economic Analyses)3
Total National Need
Chemical Contaminants.
Projects designed to protect
the public health from
chemical contaminants
comprise $5.0 biliion. or 14
percent, of the total existing
regulatory need. This
category includes projects
necessary for compliance
with the existing Nitrate/Nitrite
Standard, the Lead and
Copper Rule, the Total
Trihalomethanes Standard,
and the recently promulgated
Arsenic Rule, as weli as
other regulations that set
maximum allowable limits for
organic and inorganic
contaminants. Examples of
projects in this category
include aeration facilities to
remove volatile organic
compounds or projects to
add corrosion control to
reduce the leaching of lead
from pipes.
Proposed or Recently
Promulgated Regulations.
The total need associated
with proposed and recently
promulgated regulations is
$9.9 billion. Of this total, $3.2 billion is for the
regulation of acute contaminants under the Long Term
I and/or the Proposed Long Term 2 Enhanced Surface
Water Treatment Rules (LT1ESWTR and/or
LT2ESWTR), the Proposed Ground Water Rule, and
the Filter Backwash Recycling Rule. The remaining
$6.7 billion is for chronic contaminants regulated
under the Stage 1 and/or the Proposed Stage 2
Disinfectants/Disinfection Byproducts Rules (Stage 1
and Stage 2 DBPR), the proposed Radon Rule, and
the recently promulgated Radionuclides Rule. The
2003 Needs Assessment obtained the costs for this
category from the Economic Analysis published for
each rule; they are not estimates from respondents to
Current Need
$16,463.1
$1,283.5
$404.1
S 1,633.5
$ 123.5
§1,075.2
$20,982.9
$20,982.9
Future Need
$11,063.0
81,349.1
S97.2
$962.1
$371.9
$75.2
$255.4
$14,174.0
$9,927.4
$24,104.4
Total Need
$27,526.2
$2,632.6
$501.4
$962.1
$2,005.4
$198.7
$1,330.6
$35,156.9
$9,927.4
$45,084.3
Note: Numbers may not total due to rounding.
1 Regulations for contaminants that cause acute health effects.
2 Includes regulated Volatile Organic Chemicals (VOCs), Synthetic Organic Chemicals (SOCs), horganic
Chemicals (DCs), and Radicnuciides.
3 hcludes regulations for contaminants that cause acute and/or chronic health effects. In the Economic Analyses,
the compliance costs with some regulations are given as a range. In calculating the $9.9 biliion need, the 2003
Needs Assessment used tPA's lead option, unless one was not available, in which case ttw 2003 Needs
Assessment used the higher estimate, "(hese estimates include oniy the capital costs (i.e., excludes operation
and maintenance costs). Costs for the recently promulgated Arsenic Rule are not included in this row.
the 2003 Needs Assessment questionnaire. These
costs are added to the total national need for this
assessment, but do not affect individual states' total
need or allocation because the Economic Analysis
relies on regional data only.
Security Needs
Water systems have long included protections against
vandalism and natural disasters as part of their water
system improvement programs. However, systems
have only recently begun to address more robust
security needs to identify and protect the system from
terrorist-type activities. Because the 2003 Needs
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page 11
Assessment was concurrent with this expanded
security evaluation and planning process, many
systems may not have adequately captured these
specific needs for the 2003 Needs Assessment.
Systems with completed vulnerability assessments
and corrective action plans often did not have
documented costs for those improvements. These
were not the types of costs that EPA was prepared to
model. It is anticipated that these needs will be more
completely reported in future assessments. The total
security need estimated from the 2003 Needs
Assessment is $1.0 billion.
Needs for Small Water Systems
Approximately 45,000 of the nation's 53,000
community water systems serve 3,300 or fewer
people. Small water systems' 20-year infrastructure
need is estimated to be $34.2 billion. The total is
based on findings from the 1999 Needs Assessment,
adjusted to 2003 dollars and applied to the 2003
inventory of small systems. Small water systems face
many unique challenges in providing safe drinking
water to consumers. The substantial capital
investments required to rehabilitate, upgrade, or install
infrastructure, without the economies of scale
available to larger systems, represent one challenge.
Although the total small system need is modest
compared to the need of
larger systems, the costs
borne on a per-household
basis by small systems are
significantly higher than those
of larger systems.
Needs of American
Indian and Alaska
Native Village Water
Systems
The total need for American
Indian and Alaska native
village systems is $2.4 billion
over 20 years. The totai is
also based on findings from
the 1999 Needs Assessment, adjusted to 2003
dollars, and the portion of the total capital cost of
compliance with the recently promulgated Arsenic
Rule attributed to these systems. Exhibit ES-6
presents the total need by project type for these
systems. The total 20-year need for American Indian
systems is $1.3 billion, and for Alaska native village
systems is $1.2 billion.
Challenges for Future Assessments
All assessments that include surveys impose a data
collection burden on respondents. EPA has
considered options to reduce respondent burden in
each of the assessments (1995, 1999, and 2003).
These efforts must be renewed in planning for the
next assessment. EPA will pay particular attention to
the number of projects to be considered in a 20-year
planning effort, the comprehensiveness of the data
collection goal, and documentation requirements for
each project. All of these factors create a burden for
participating water systems, state agencies, and EPA.
While the data obtained through the survey and
assessments are extremely valuable for many
applications, the approach used to collect the data is
regularly reviewed by EPA to determine more efficient
and effective ways to capture the fuli need.
Ilihiit E$
Alaska Native Village ^^'.i^j^iHisil^J' fcy Proje
'"••- -vKv:-:. ••••••-•• : ' • • • •••• '':''
Treatment
$0,5
Note: Numbers may rot total due to rounding.
-------
Page 12
2003 Drinking Water Infrastructure Needs Survey and Assessment
EPA is addressing two additional issues for future
assessments: engineering assumptions of life cycles
for future rehabilitation and replacement projects, and
encouraging greater response rates from systems in
states receiving the minimum 1-percent DWSRF
capitalization grants.
With respect to life cycle assumptions, more explicit
nationally applicable guidelines would facilitate
consistency from the outset of the assessment. This
would streamline quality control efforts and eliminate
the need to identify assumptions used in projections
of infrastructure replacement and rehabilitation needs
(and reject projects where assumptions are
inconsistent with industry practice). Regarding
response rates, states that are near or beiow 1
percent of the total national need have little incentive
to promote responses from systems in their
jurisdictions. This can lead to underestimates of the
needs in these states.
In the estimation of total national needs, these two
issues may partially offset each other. (Inconsistent
engineering assumptions may drive needs up, but low
response rates in states receiving minimum
capitalization grants may drive needs down.) Yet,
these issues can affect the relative distribution of
needs among states receiving more than 1 percent of
the DWSRF appropriation. Without more complete
participation in states receiving minimum
capitalization grants, questions may be raised about
the appropriateness of the current statutory approach.
EPA realizes these issues should be discussed with
stakeholders before data collection begins on the next
assessment. Stakeholders on this issue include
states, their Governors and Legislators, the water
supply industry and its associations, and researchers,
particularly those who have specialized in empirical
research on the useful life of pipe. As the
Congressional Budget Office noted in 2002, methods
of estimation and assumptions about requirements for
rehabilitation and replacement typically drive national
estimates of infrastructure needs.10 The Agency
recognizes that reaching agreement on the approach
to this issue in future assessments will improve the
credibility of the estimates that are submitted to
Congress.
Finally, EPA recognizes that assessment methods
result in uncertainty in the estimated needs. The
sampling plan for medium and large systems was
designed to produce estimates of the tola! need for
each state with 95 percent confidence intervals that
are ±10 percent. However, sampling error is only one
source of uncertainly. The assessment also involves
statistical cost models and economic analyses of
regulations. Each of these creates additional
uncertainty. While the 2003 Needs Assessment does
not include a comprehensive quantitative analysis of
uncertainty, EPA plans to continue efforts to more
accurately characterize these in future assessments.
Conclusions
The 2003 Drinking Water Infrastructure Needs Survey
and Assessment, the third such national effort by
EPA, estimates that the nation's public water systems
need to invest $276.8 billion over the next 20 years to
ensure the continued provision of safe drinking water
to consumers.
The findings of the previous assessments, conducted
by EPA in 1995 and 1999, indicated that the need was
most likely underreported because of limitations of
water system planning documents. EPA believes that
changes made to the assessment to address
underreporting resulted in a more complete
assessment of the 20-year need.
The need to rehabilitate and replace infrastructure is
expected to increase as systems age, particularly if
funding constraints limit the systems' ability to meet
these needs. The needs summarized in this report
highlight the challenges facing water systems as they
cope with aging infrastructure in the 21s1 century.
10 Congressional Budget Office, op. cit., pp. 13-17.
-------
SCOPE OF THE ASSESSMENT AND SURVEY
The 2003 Drinking Water Infrastructure Needs Survey and Assessment represents the
collective efforts of the states, EPA, and thousands of water systems—all of which participated
in identifying and documenting infrastructure needs. This chapter provides an overview of the
methods used by these participants to assess drinking water needs. It also describes the
refinements made to the methods used in the 1995 and 1999 Needs Assessments to improve
the accuracy of the results, and the extent of reliance on the 1999 Needs Assessment in
determining the need for small, American Indian, Alaska native village, and not-for-profit
noncommunity water systems.
Scope of the Assessment
Goai and Purpose. EPA's goal for the 2003 Drinking
Water Infrastructure Needs Survey and Assessment
was to document the 20-year national infrastructure
needs for the approximately 53,000 community and
21,400 not-for-profit noncommunity public water
systems eligible to receive DWSRF assistance.
Needs were assessed for the 20-year period
beginning January 1, 2003. and ending December 31,
2022. A total of approximately 4,000 medium- and
large-population public water systems completed the
2003 Drinking Water Infrastructure Needs Survey and
Assessment questionnaire. Medium and large
systems' infrastructure needs projected over the next
20 years (excluding costs to comply with the recently
promulgated Arsenic rule) constituted 82 percent of
the total need.
* States. The 1996 Safe Drinking Water Act
(SDWA) Amendments direct EPA to assess
the needs of water systems, and to use the
results of the assessment to allocate DWSRF
funds. To this end, the Agency designed an
assessment that would provide accurate
estimates of need for each of the states. The
DWSRF funds are allocated based on each
state's share of the total national need
(although, under SDWA, each state receives a
minimum allotment of 1 percent).
The survey of medium and large systems
was designed to provide a nigh level of
precision for each state's estimate of need.
For most of the survey, a precision target of
95 percent ±-10 percent was established.
Territories. The results of the assessment
are also used to allocate the 0.33 percent of
the DWSRF appropriation designated for the
Pacific island territories. Therefore, the
workgroup designed the assessment to
generate separate estimates of need for
Guam, American Samoa, the Commonwealth
of Northern Mariana Islands, and the U.S.
Virgin Islands. Needs for the Virgin Islands
were determined by adjusting 1999 needs to
2003 dollars. The assessment results dictate
what percentage of the 0.33 percent will go to
each territory.
American Indian Communities and Alaska
Native Villages. For this assessment, the
need determined from the 1999 Needs
Assessment was adjusted and used to
determine the 2003 need. The results are
used to help determine how to allocate funds
that are available through the DWSRF to
American Indian and Alaska native village
water systems.
-------
Page 14
2003 Drinking Water Infrastructure Needs Survey and Assessment
This corroded, valveless filter is badly in need of replacement.
Eligible Needs. Since the purpose of the
assessment is to allocate DVVSRF funds, EPA
included only projects that met the eligibility criteria
established under the DWSRF program.'1 In general,
projects eligible for DWSRF funding facilitate
compliance with the SDWA's National Primary
Drinking Water Regulations or otherwise significantly
further the health protection objectives of the Act.
Categories of Need by Project Type. Each project
was assigned to one of five categories of need based
on the project type: source, transmission and
distribution, treatment, storage, or "other." This
classification shows where the nation's water
systems need to make capital investments.
* The source water category includes projects
necessary to obtain adequate quantity and
quality of surface water and ground water
supplies. Examples include wells, surface
water intakes, and spring collectors.
• The transmission and distribution category1
includes the needs associated with installing
or rehabilitating raw and finished water
transmission pipes, distribution water mains,
pumping stations, flushing hydrants, valves,
water meters, and backflow prevention
devices.
* The treatment category includes projects
needed to deal with microbial pathogens and
chemical contaminants present in the water
supply.
• The storage category includes projects to
construct new or rehabilitate existing finished-
water tanks.
* The "other" category is reserved for needs that
cannot be assigned to one of the four major
categories. Examples include emergency
power generators not assigned to specific
types of projects, computer and automation
projects, and projects to address security,
Current and Future Needs. For the 2003 Needs
Assessment, EPA distinguished between current and
future needs for the 20-year period from January 1,
2003, through December 31, 2022. Current needs are
projects that systems consider a high priority for near-
term implementation that will enabie a water system
to continue to deliver safe drinking water. An example
of a current need is replacement of a section of
distribution line that is susceptible to breaks or leaks.
Future needs are projects that are not necessary at
the time of the assessment but that water systems
expect to undertake within the next 20 years. These
include routine rehabilitation and replacement
projects. For example, a system may anticipate that it
will need to rehabilitate a storage tank in
approximately 10 years, or that it needs to replace a
certain length of distribution pipe every year over the
20-year period to phase out old pipe. These future
needs were underreported in previous assessments,
in part due to limitations of the planning documents.
11 EPA's assessment excluded DWSRF-eiigible needs which do not involve the installation, replacement, or rehabilitation of infrastructure; 'or
example, refinancing loans, conducting studies, and acquiring other water systems.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page 15
To mitigate underreporting for this assessment, EPA
made changes in the format of the questionnaire and
trained state coordinators on needs assessment
tools. The new questionnaire asked systems to
review their entire inventory of infrastructure assets
and consider what projects might be necessary to
manage those assets through the end of
2022. The questionnaire also provided
examples of appropriate projects and related
documentation. The Agency encouraged
states to help systems review their inventories
and identify realistic estimates of system
needs. Many states visited or called each
system within their jurisdictions to facilitate
completion of the questionnaires. States used
in-house inventories (where available) to
ensure that ail major infrastructure was
considered. Some states used their own
analyses o? infrastructure condition to identify
needs.
Reasons for Need. The questionnaire also
asked systems to identify and code the
reason, or reasons, each project was needed.
Options included:
Projects for consolidation with and/or
connection to an existing public water system.
Projects for extending service to existing
homes without adequate water quantity or
quality.
0.75 million gallon ground level storage tank in Ksnnan,
California wax constructed to compensate for the reduced capacity of
three wells that am being constructed to replace three larger
contaminated wells.
* Projects for existing infrastructure that
is, or will be, old or deteriorated by the end of
the 2003 Needs Assessment period.
• Projects to correct a deficiency in source
water quantity caused by current user
demand.
* Projects to correct a deficiency in storage
capacity caused by current user demand.
Not surprisingly, a majority of the systems and states
listed "replacement or rehabilitation of old or
deteriorated infrastructure" as the primary reason for
need. Sixty-seven percent of projects listed "old and
deteriorated infrastructure" as the only reason for
need, and 77 percent listed this as at least one of the
reasons for need if more than one reason was
provided.
» Projects to correct existing pressure problems Security Needs. Projects intended wholly or in part
not related to fire flow. to address security needs were separated into the
following categories:
» Projects to obtain or maintain compliance with
an existing regulation. • Projects to prevent or detect an intrusion or
security violation.
• Projects to obtain or maintain compliance with
a secondary standard. • Major security projects.
-------
Page 16
2003 Drinking Water Infrastructure Needs Survey and Assessment
» Communication needs for security.
* Projects for redundancy or to respond to a
security breach.
* Projects to address safety issues.
Assessment Methods
The 2003 Needs Assessment consisted of two
components: a new survey of needs for large and
medium systems; and an estimate of needs for
systems serving 3,300 or fewer persons, not-for-profit
noncommuntty water systems, American Indian
systems, and Alaska native village systems. These
two components are discussed below.
A workgroup of state and EPA representatives
developed the methods for the 2003 Drinking Water
Infrastructure Needs Survey and Assessment. The
workgroup decided to adopt the genera! approach of
the 1995 and 1999 Needs Assessments. However,
the workgroup refined the questionnaire to prompt
more complete assessment of needs. These
refinements were based on lessons learned from the
1999 Needs Assessment regarding effective interview
methods for capturing needs that are not included in
relatively short-term water system planning
documents. The workgroup also revised some
documentation policies to reduce the burden on
systems without compromising the validity of the data.
Communications options made available by changes
to the Internet aiso allowed more efficient information
exchange on specific projects between EPA and
states.
Conducting the State Survey for Large
and Medium Systems
EPA and the states developed a questionnaire used to
collect infrastructure needs from large and medium
community water systems. The questionnaires were
provided to all of the nation's water systems serving
over 40.000 people and from a random sample of
systems serving 3.301 to 40,000 people. Each
system received a package containing the
questionnaire, instructions, an example of a
completed questionnaire, and a list of frequently
asked questions.
Systems returned the questionnaires and
accompanying documentation to their state contacts.
The states reviewed each questionnaire to ensure
that systems identified all of their needs and that the
projects fulfilled the eligibility and documentation
criteria. If these criteria were not met, the states had
the option of contacting the system to obtain more
information. EPA conducted a final review of each
project and entered the information into a database.
Web-based communications allowed the states to
review the data, including any changes made by EPA.
Using the project Web site, states couid identify
projects not meeting the established criteria and
submit additional documentation of the project need or
the cost to support a project.
improvements for the 2003 Needs
Assessment of Medium and
Systems
Compared with the previous two assessments where
EPA had a substantial role in data coliection, the 2003
Needs Assessment placed the responsibility for
collecting data primarily on the states. To assist
states, EPA held 2-day training sessions at eight
regional locations. These training sessions were
designed to educate state coordinators, staff, and
their contractors on the approach, available needs
assessment tools, and documentation criteria. EPA
also worked directly with each state in reviewing
responses for the first five questionnaires to maintain
consistency.
As an improvement over the 1999 questionnaire, the
workgroup modified the design of the 2003
questionnaire to prompt systems to more thoroughly
consider their entire infrastructure inventory and
projects that might be needed over the next 20 years.
The 2003 questionnaire asked the system about the
length and diameter, or number and size, of major
pieces of existing infrastructure. The questionnaire
-------
200x3 Drinking Water infrastructure Needs Survey and Assessment
Page!?
included tables to record the gross infrastructure
inventory and asked foliow-up questions to prompt the
system to consider the 20-year need for rehabilitation
or replacement of the infrastructure and whether it
was adequate to meet the needs of existing
consumers. The questionnaire provided examples of
projects and acceptable documentation, and
simplified the data collection format into three
category-related tables—transmission and
distribution; source, treatment, storage, pumping, and
other; and backfiow prevention devices/assemblies,
flushing hydrants, service lines, valves, and water
meters.
The workgroup reconsidered some policies that had
been adopted for the 1999 Needs Assessment.
Specifically, the workgroup decided that it was not
necessary to require systems to identify ownership of
backfiow prevention devices or non-lead service lines.
if the projects were identified as needs, the workgroup
assumed that they were likely the responsibility of the
public water system.
Advances in Internet technology prompted the
development of an interactive Web site that allowed
states and EPA to track survey progress and
O S) ..',. S3
Thf user-friendly We.h nite allowed KM andstatsx to communicate the status
of the survey and projects submitted.
communicate questionnaire and project status
updates. States were able to identify projects that
required additional documentation and to respond to
most issues via the Web site.
Another policy change was related to the eligibility of
domestic water meter projects. In the 1999 Needs
Assessment, systems were limited to metering
currently unmetered systems or replacing meters that
were currently malfunctioning. In 2003, recognizing
the vaiue of metering to water audits, conservation
programs, and asset management, the workgroup
allowed metering of unmetered systems and a single
replacement of each existing meter over the 20-year
assessment period. Under the new policy, the meter
projects for large and medium systems accounted for
$12.1 billion in need. This amount is included in the
total transmission and distribution category of need.
For the 1999 Needs Assessment, if a project was
categorized as a regulatory need, systems were
required to include as part of their documentation a
laboratory report showing an actual or imminent
violation of a maximum contaminant level (MCL) or
treatment technique requirement. For the 2003 Needs
Assessment, the workgroup decided that an actual
laboratory slip was not needed as part of
the documentation.
Method for Estimating the
Small System, Not-for-Profit
Nonepmmunity System, and
American Indian and Alaska
Native Village Need
Small Systems and Not-for-Profit
Noncommunity Systems. Small
systems serving 3,300 or fewer people
and not-for-profit noncommunity systems
generally lack the personnel and planning
documents necessary to complete the
questionnaire. Therefore, for the 1999
Needs Assessment, EPA conducted site
visits to determine the infrastructure
needs of these systems.
-------
Page 18
2003 Drinking Water Infrastructure Needs Survey and Assessment
EPA believes that the needs captured from the site
visits in 1999 represented a fair and complete
assessment of these systems' 20-year needs.
Findings from 1999 were very similar to the findings in
1995, indicating that system's needs did not change
significantly over a 4-year period. Because there was
a high level of confidence in the data obtained from
the site visits, EPA decided that it could estimate 2003
needs by adjusting the 1999 needs to 2003 dollars.
The total national small system need was then
reallocated to each state based on the number of
systems that existed in each stratum in 2003.
The 1999 not-for-profit noncommunity needs were
likewise adjusted to 2003 dollars and assigned to
each state's need.
American Indian and Alaska Native Village Needs.
During the 1999 Needs Assessment EPA helped the
American Indian and Alaska native village water
systems complete their questionnaires.
* American Indian Systems. In 1999. all 19
medium-sized American Indian systems
completed a questionnaire with technical
support from EPA. The Agency conducted site
visits at 78 randomly selected small systems
to represent the 781 small American Indian
systems.
* Alaska Native Village Systems, In 1999,
questionnaires were mailed to the two
medium-sized systems. For the 172 small
systems, representatives from the Alaska
Native Village Health Consortia, the IMS, and
the Village Safe Water completed the
questionnaires, with assistance from EPA.
Because of the high level of confidence in the findings
from 1999, EPA did not survey these systems again in
2003. Instead, EPA adjusted the data from 1999 to
2003 dollars to estimate the 2003 needs for these
systems.
Documented Costs and Cost Models
If systems had documented cost estimates for a
given project, EPA converted these costs to January
1. 2003 dollars and applied the cost to the system's
total need. If no costs were available, the
questionnaire requested information about the project
so that EPA could model a cost for the project. For
example, if a system identified a need to replace a
section of leaking pipe, but lacked cost
documentation, the system supplied the length and
diameter of pipe to be replaced. Based on this
information, EPA modeled the cost for this project.
The number of projects submitted without cost
documentation increased in 2003 compared with the
previous assessments. Of the 105,000 accepted
projects, 82 percent were submitted without costs.
This increase resulted in a heavy reliance on cost
modeling.
Acceptable Documentation
For Need and/or Cost Documentation:
» Capital Improvement Plan or Master Plan
* Facilities Pian or Preliminary Engineering
Report
* Grant or Loan Application Form
« Engineer's Estimate
For Need Documentation Only:
* Intended Use Plan/State Priority List
« Indian Health Service Sanitary Deficiency
System Report
« Comprehensive Performance Evaluation
(CPE) Results
» Sanitary Survey
« Monitoring Results
« Other Need Document
For Cost Documentation Only:
9 Cost of Previous Comparable Construction
« Other Cost Document {such as
manufacturer's catalog costs)
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page 19
In addition to developing requirements for
documenting needs, the workgroup set rigorous
documentation criteria for assessing the legitimacy
and scope of project costs. If systems submitted
project costs, there had to be documentation showing
that the cost had undergone an adequate degree of
professional review. These would have included
Capital improvement or Master Plans developed for
the system by professional engineers, tabulations of
bids received for a project developed by contracting
firms, or costs of previously completed projects of
comparable scope. Documentation had to be detailed
enough that EPA could review all component costs
included in the estimate. This enabled EPA to model
portions of the project that had been omitted from a
cost estimate, or to delete DWSRF-ineligible portions
of the submitted cost (such as interest payments).
In general, EPA used the models developed from the
1999 Needs Assessment data and adjusted the 1999
data to 2003 dollars for the 2003 Needs Assessment.
For the 1999 Needs Assessment, 59 models were
developed to assign costs to infrastructure needs—
from replacing broken valves to building new
treatment plants. Most of the cost models were
derived from projects that listed both cost estimates
and modeling parameters. For some types of need,
the 1999 Needs Assessment data proved inadequate
for a statistically significant model. Therefore, for 19 of
the models, EPA obtained cost data from additional
sources—engineering firms and state DWSRF
programs—to supplement data submitted by
respondents.
For the 2003 Needs Assessment, EPA derived new
models for transmission and distribution piping and
meters. A new meter model was needed to
accommodate improvements in standard technology.
Since the 1999 Needs Assessment, the standard
technology for domestic water meters changed from
predominantly manual-read meters to radio-read
meters. This new technology had a higher cost, so a
new model was appropriate.
EPA also updated the cost models for transmission
and distribution pipe based on cost information
received from the 2003 Needs Assessment. These
models had not been updated since the 1995 Needs
Assessment. Because the transmission and
distribution category represents the largest
percentage of need, developing up-to-date models
was a high priority.
Information Quality
The findings of the 2003 Needs Assessment are
reinforced by adherence to EPA's Guidelines for
Information Quality,12 which implement the Data
Quality Act for the Agency. Appendix C of this report
contains more detail on information quality.
Quality Assurance. The most fundamental
requirement for information quality is the Agency's
Quality System. EPA implements the system on a
project basis through the development of a quality
assurance project plan (QAPP), the cornerstone of
which is the definition of data quality objectives
(DQOs). The Agency uses the results of this
assessment to allocate DWSRF capitalization grants
to states. Allocations are made on the basis of
proportional state need for water systems eligible for
DWSRF funding. Therefore, this project (like those
that preceded it in 1995 and 1999) sought to
maximize the accuracy of the state-level estimates of
infrastructure needs. Decisions about precision levels
were also established by a state/EPA workgroup that
met regularly during the 2003 Needs Assessment.
Many water sy&ems are impr&vtttg the
efficiency and accuracy ofrvater usage.
data collection by replacing old and
outdated water meters with new radio-
read meters. Hand-held radio meter units
communicate -with the meter iramtniiter
fron, a remote location, such as a vehicle*
This dramatically reduces lahor hours
rie&fed to collect water tisage data.
12 U.S. Environmental Protection Agency, "Guidelines for Ensuring and Maximizing the Quality. Objectivity, and Integrity of Information Disseminated
by the Environmental Protection Agency," EPA/260R-02-008 (October 2002).
-------
Page 20
2003 Drinking Water Infrastructure Needs Survey and Assessment
Accuracy was maximized through the following
steps. First, since this was a sample survey, the
workgroup established targets for precision of
estimates (acceptable sampling error). These
decisions shaped sample design. Second, EPA used
quality assurance (QA) procedures from the QAPP to
ensure that "eligible infrastructure" was clearly
defined and that documentation standards were
rigorously enforced. For a project to be included in the
2003 Needs Assessment, systems and states had to
submit documentation describing the purpose and
scope of the project for each need. The
documentation was reviewed by EPA to determine if
each project submitted for the 2003 Needs
Assessment met the eligibility criteria for DWSRF
funding and allowability criteria set for the 2003 Needs
Assessment. The workgroup established the
documentation requirements so that uniform criteria
were applied to all questionnaires. These
requirements not only lend credibility to the findings,
but also address the issue of fairness when the
results are used by EPA to apportion DWSRF funds.
Of the 128,600 projects submitted to the survey, EPA
deleted 18 percent that failed to meet the
documentation criteria, or appeared to be ineligible for
DWSRF funding. Some projects were adjusted to
correct a variety of measurement problems: overlaps
between two projects (raising the issue of double-
counting), inconsistency with project documentation,
and use of overly aggressive infrastructure iife cycles
by states where system planning documents were not
used or available.
To adjust for the use of aggressive infrastructure life
cycles, EPA made technical adjustments to individual
projects based on engineering literature and
benchmarks of engineering practices. The Agency
tailored adjustments to the unique assumptions
implemented by each state and then negotiated with
state officials. EPA's genera! direction of these
adjustments was to place a cap on the state's
assumptions about the rate of rehabilitation and
replacement of pipe, unless there was project-specific
documentation of a need provided by the water
system.
Other subjects discussed in the QAPP were: training
and certification of staff working on data collection and
evaluation; standards for questionnaire design and
survey implementation; procedures for manual
editing, coding, and data entry; automated data
validation; database quality assurance; tabulation
quality assurance; and QA for report preparation.
Transparency and Reproducibility. EPA's
Guidelines on Information Quality explain that
influential information (such as this report) "should be
subject to a higher degree of quality (for example,
transparency about data and methods). Such
Quality Assurance
The 2003 Meeds Assessment followed the Agency's Guidelines for Ensuring and Maximizing Information
Quality (2002). EPA's goal for these guidelines is to ensure the quality, objectivity, utility, and integrity of
information disseminated by the Agency. These guidelines are particularly important for projects such as the
2003 Meeds Assessment, which influences public policy decisions.
The 2003 Needs Assessment workgroup implemented the guidelines through quality assurance and
reprodiicibility of its results. Also, given the influential nature of the report, EPA ensured a high degree of
transparency regarding data, assumptions, analytic methods, and statistical procedures. :
For more information on quality assurance, see Appendix C. For more information on data, assumptions,
analytic methods, a«d statistical procedures, see Appendix B.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment Page 21
transparency facilitates reproducibility of this
information, and reproducibility should meet
commonly accepted standards."
The 2003 Needs Assessment (like those in 1995 and
1999) maintained high standards of transparency. For
example, ail decisions about the study approach,
analytical methods, cost models, and statistical
methods, were presented to the workgroup for their
review. All data collected by this study were made
available on-line to state experts for their review and
comment.
Appendix B contains information on the statistical
methods and cost modeling procedures that were
used in the preparation of this report. Given this
information, and access to the database, any qualified
third party could reproduce the results of this
assessment.
-------
FINDINGS
The 2003 Drinking Water Needs Survey and Assessment estimated the capital investment
needs of the nation's approximately 53,000 community water systems and 21,400 not-for-profit
noncommunity water systems. Appendix D provides greater detail of the need by state.
Total 20-Year National Need
The 2003 Needs Assessment indicates that
community water systems and not-for-profit
noncommunity water systems need $276.8 billion over
the next 20 years to install, upgrade, and replace
infrastructure. For the 2003 Needs Assessment,
states were required to present documentation that
described the purpose and scope of each project. In
general, infrastructure projects were acceptable if they
were needed to protect public health or to maintain the
delivery of potable water to homes. Such projects
varied greatly in scale, complexity, and cost—from
rehabilitating a small storage tank to constructing a
high-capacity water treatment plant for a large
metropolitan area. EPA excluded projects solely for
future growth, fire flow, and general operation and
maintenance needs.13 However, EPA included projects
to rehabilitate or replace significant components of
deteriorated infrastructure because they were not
considered operation and maintenance.
The estimate of total national need represents all
community water systems and not-for-profit
noncommunity water systems in the states, Puerto
Rico, the Virgin Islands and the Pacific island
territories. District of Coiumbia, American Indian
communities, and Alaska native villages.
Exhibit 1 shows the total national need by system size
and type, and by current and future need. The nation's
1,041 largest community water systems (serving more
than 50,000 people) account for $122.9 billion, or 44
percent of the total need. Medium and small
community water systems have needs of $103.0
billion and $34.2 billion, respectively. These figures
include the needs for small, medium, and large
systems in the Pacific island territories and Virgin
Islands, which are $509.1 million and $172.6 million,
respectively. Not-for-profit noncommunity water
systems have $3.4 billion in estimated needs. The
American Indian and Alaska native village system
needs total $2.5 billion: American Indian water
systems need $1.3 billion in infrastructure
improvements, and Alaska native villages need $1.2
billion.
Because public water systems are not expected to
have accurate estimates of their capita! needs for
proposed or recently promulgated regulations, EPA
used capital costs from Economic Analysis
documents for the rules to estimate those needs.
Proposed or recently promulgated regulations
account for $9.9 billion of the total national need, in
addition, the need for compliance with the recently
promulgated Arsenic Rule is $1.0 billion. This
includes the cost of compliance for water systems in
the states ($947.4 million) as well as water systems
serving American Indian communities and Alaska
native villages ($14.7 million).
Most of the infrastructure needs in the assessment
represent projects that systems would address as
preventive measures to ensure the continued
provision o? safe drinking water rather than as
corrective actions to address an existing violation of a
drinking water standard. EPA recognized that the
majority of the total national need stems from the
inherent costs of producing and delivering water—
which involves an ongoing need to install, upgrade,
and replace the basic water system infrastructure.
13 Projects soiely for operation and maintenance, darns, reservoirs, future growth, and fire "ow are generally ineligible for DWSRF assistance.
-------
Page 24
2003 Drinking Water Infrastructure Needs Survey and Assessment
System Size and Type
Large Community Water Systems
(serving over 50,000 people)2
Medium Community Water Systems
(serving 3,301 to 50,000 people)2
Small Community Water Systems
(serving 3,300 and fewer people)2'3
Costs Associated with the recently
promulgated Arsenic Rule4
Not-for-profit Noncornmunity Water Systems5
American Indian and Alaska Native Village
Water Systems5'6
Subtotal National Need
Costs Associated with Proposed and
Recently Promulgated Regulations
(Taken from EPA Economic Analyses)
Total National Need
T™™*™
Current Need
$80.7
$56.4
$24.4
$1.2
$2.3
$165.0
$165.0
I
Future Need
$42.1
$46.6
$9.8
$0.9
$2.2
$0.2
$101.8
$9.9
$111.8
mswwsmmm
Total Need
$122.9
$103.0
$34.2
$0.9
$3.4
$2.4
$266.8
$9.9
$276.8
Number of
Systems1
1,041
7,638
43,039
21,400
974
Note: Numbers may not totai due to rounding.
' Number of largo, medium, and small systems in determined from the 2003 Needs Assessment sample frame. Number of not-for-
profit, American Indian, and Alaska naiive village systems is determined from the 1999 (Meeds Assessment sample frame. The
numbers in the 2003 Needs Assessment may differ from the Safe Drinking Water Information System (SDWIS) due to changes in
system inventories and the way 1hs 2003 Needs Assessment classifies some systems (I.e., systems that serve Alaska native villages
are classified in SDWIS as small systems, but are classified in the 2003 Needs Assessment as Alaska native village water systems).
2 Does not include the costs associated with the recently promulgated Arsenic Rule and proposed or recently promulgated SDWA
regulations; Uiese costs are included on a separate line in this table.
3 1 939 Needs Assessment findings adjusted to January 2003 dollars and reallocated based on 2003 inventory of small systems.
' Does not include exists for American Indian and Alaska native village water systems to comply with the recently promulgated Arsenic
Rule; those costs are incorporated in the estimate for American Indian and Alaska native village water systems.
5 1989 Needs Assessment findings adjusted to January 2003 dollars.
5 Includes cost for compliance with the recently promulgated Arsenic Ruie.
Exhibit 2 provides an overview of the needs by state.
Appendix D provides a more complete breakdown of
needs for each state.
Current and Future Needs
Of the totaf national need, $165.0 billion are for
current needs. Although most systems have current
needs, this does not preclude their delivery of safe
drinking water to their customers. Rather, many
current needs are preventive projects to avoid water
quality problems. For example, a system may
conclude that some of its 50-year-old pipe is
deteriorated. Although the system is in compliance
with all regulations, the condition of the pipe makes
compliance with the Total Coliform Rule difficult, and
occasional breaks may cause interruptions in service.
-------
2003 Drinkino Water Infrastructure Needs Surrey and Assessment
Page 25
American Samoa*
Guam*
Northern Mariana Is'
i:
Ii|QP;:::;il:llllIIIJ.:::r;
Includes need for the recently promulgated Arsenic Regulation. Does not include needs for American Indian and Alaska native
village water systems.
"The needs for American Samoa. Guam, the Northern Mariana Islands, and the Virgin Islands are less than $1 billion each.
-------
Page 26
2003 Drinking Water Infrastructure Needs Survey and Assessment
The size of current need reflects the age and
deteriorated condition of the nation's infrastructure.
Many water systems were constructed 50 to 100
years ago. Some systems have adopted a reactive
approach to capital investment that involves replacing
or upgrading infrastructure only as it fails. For
example, a system may fix leaks in the distribution
system, rather than invest in rehabiiitation or
replacement. A more pro-active approach of planned
rehabilitation or replacement should prove iess costly
over the long run and reduce the likelihood o?
emerging risks to public health.
Future needs account for $111.8 billion of the total
need. Future needs are projects that are not currently
necessary. Nevertheless, systems will need to
undertake these projects during the 20-year period of
the assessment to ensure the continued provision of
safe drinking water. Future needs address
components of a water system that operate
adequately now, but will exceed their design life or
performance capabilities within the next 20 years. For
example, a recently constructed storage tank
operates adequately now, but based on historic
trends, the system knows that the tank will require
some major rehabilitation within the next 20 years.
Total Need by Project Type
Infrastructure needs of water systems can be
grouped into four major categories based on project
type-source, transmission and distribution, treatment,
or storage—each of which fulfills an important
function in delivering safe drinking water to the public.
Most needs were assigned to one of these categories.
An additional "other" category is comprised of projects
that do not fit into one of the four categories.
Examples are system-wide security or computer
controls. Exhibit 3 shows the total national need by
water system size and type and by project type.
Transmission and Distribution Needs.
Transmission and distribution projects represent the
largest category of need (two-thirds of the total need).
$183.6 billion over the next 20 years. Of this total,
$120,0 billion is identified as current needs. Although
the least visible component of a public water system,
the buried pipes of a transmission and distribution
network generally account for most of a system's
capital value, it is not uncommon for even medium-
sized systems to have several hundred miles of pipe.
Little of this $183.6 billion is related to any federal
mandate. Projects are typically driven by the utilities'
need to install and maintain distribution systems to
provide potable water to their customers while
preventing contamination
of that water prior to
delivery.
industry benchmarks
indicate that although most
systems address iess than
1 percent of their existing
pipe per year, an aggressive
program would; provide for
replacement or rehabilitation
of as much as 1 to 2 percent
of a system's total pipe per
year.
Transmission and
distribution projects
include replacing aging
and deteriorated water
mains, refurbishing pipes
to remove build-up on
pipe walls, looping dead-
end mains to avoid
stagnant water, installing
water mains in areas where homes do not have a
safe and adequate supply, and installing pumping
stations to maintain adequate pressure. This category
also includes projects to address the replacement of
appurtenances, such as valves that are essential for
controlling flows and isolating problem areas during
fipenurstmg is an effective way to upgrade deteriorated
pipe. A pneumatic bursting head is attached to new pipe
and (hreaded tkmugh the old pipe, As if-passes through
the old pipe, the bursting head destroys the old pipe,
compacting it into- the. surrounding sail making room
far the new pipe aftha same or even larger diameter.
Pi.pebursling is a preferred method of pipa upgrade since
it is $eifii-tn&ich}&>x and minimises' disruption to xireets.
homes, and businesses in the area.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page 27
repairs, hydrants to flush the distribution system to
maintain water quality, and meters to record flow.
Replacing or refurbishing transmission and
distribution mains is critical to providing safe drinking
water. Failures in transmission and distribution fines
can interrupt the delivery of water and possibly allow
backsiphonage of contaminated water. Deteriorated
distribution mains can pose acute health risks by
providing an environment in which bacteria will grow.
The rate at which pipe requires replacement or
rehabilitation varies greatly by the age of the pipe, soil
characteristics, weather conditions, construction
methods, and pipe material. Systems that have
neglected to rehabilitate or replace mains may have
more aged infrastructure, and therefore a higher level
of need.
In addition, some pipe materials have not stood the
test of time. Galvanized pipe is particularly
susceptible to corrosion in certain soils. Unlined cast
System Size and Type
Large Community Waler Systems
(serving over 50,000 people)1
Medium Community Water Systems
(serving 3,301 to 50,000 people)1
Small Community Water Systems
(serving 3,300 and fewer people)1'2
Costs Associated with the Recently
Promulgated Arsenic Rule3
Not-for-profit Noncommunity Water
Systems4
American Indian and Alaska Native
Village Water Systems" r>
Subtotal National Need
Costs Associated with Proposed and
Recently Promulgated Regulations
(Taken from EPA Economic Analyses)
Total National Need
Distribution and
Transmission
$89,779.9
$73,454.4
$18,624.3
$425.3
$1,347.3
$183,631.1
$183,631.1
Treatment
$20,091.3
$14,906.2
$6,164.1
$947.4
$670.2
$462.2
$43,241.4
$9,927.4
$53,168.8
Storage
$6,994.5
$9,473.3
$6,263.8
$1,620.3
$490.3
$24,842.2
$24,842.2
Source
$4,715.8
$4,392.8
$2,871.0
$681 .0
$135.1
$12,795.6
$12,795.6
Other
$1 ,270.2
$790.9
$248.3
$0.8
$13.6
$2,323.7
$2,323.7
Total Need
$122,851.7
$103,017.4
$34,171.5
$947.4
$3,397.5
$2,448.5
$266,834.1
$9,927.4
$276,761.5
Note: Numbers may not total due to rounding.
' Does not include the costs associated with the recently promulgated Arsenic Rule and proposed or recently promulgated SDWA regulation; these costs
are included on a separate line in this table.
* 1 993 Needs Assessment findings adjusted to January 2003 dollars and reallocated based on 2003 inventory of smail systems.
3 Docs not include costs for American Indian and Alaska native village water systems to comply with sha recently promulgated Arsenic Rule; these costs
are incorporated in the estimate for American Indian and Alaska native village water systems.
' 1939 Needs Assessment findings adjusted to January 2003 dollars.
'-• Inclsxtes cost for compliance with tho recently promulgated Arsenic Rute.
-------
Page 28
2003 Drinking Water Infrastructure Needs Survey and Assessment
iron pipe and ductile iron pipe are susceptible to
internal corrosion. Furthermore, health concerns
associated with asbestos make asbestos cement
pipe undesirable. Many water suppliers are
systematically removing these types of mains and
replacing them with ductile iron or poiyvinyl chloride
(PVC).
Treatment Needs. The total 20-year need for
treatment is $53.2 billion, of which $23.7 billion are
current needs. This category includes the installation
or rehabilitation of infrastructure to reduce
contamination through, for example, filtration,
The workgroup developed 47 dHferent treatment
codes for the 200$ Needs Assessment to identify
the specific type of treatment being employed by a
system. This ensured that the cost of the project
was modeled appropriately if costs were not
provided,
disinfection, corrosion control, and aeration. Since the
majority of the capital costs for proposed and recently
promulgated regulations are related to treatment,
these costs also are included in this category.
Treatment facilities vary significantly in scale
depending on the quality of source water and type of
contamination. Treatment systems range from a
simple chlorinator for disinfection to a complete
conventional treatment system with coagulation,
fiocculation, sedimentation, filtration, disinfection,
laboratory facilities, waste handling, and computer
automated monitoring and control devices.
Treatment technologies primarily address two general
types of contaminants: those with acute health effects
and those with chronic health effects.
An acute health effect usually occurs within hours or
days of short-term exposure to a contaminant. Acute
illnesses are associated mostly with microbial
contaminants, although some chemical
contaminants, such as copper and nitrate, also can
cause acute health effects. Gastrointestinal illness
resulting from the ingestion of microbial pathogens is
the most common acute health effect.
Chronic health effects develop typically after long-term
exposure to low concentrations of chemical
contaminants. Examples of these effects include
cancer and birth defects. The largest need associated
with contaminants that pose chronic health effects is
treatment for lead. Research has shown that
exposure to lead may impair the mental development
of children and cause other chronic health effects
such as high blood pressure.
The treatment category also includes projects to
remove contaminants that adversely affect the taste,
odor, and color of drinking water. Treatment for these
"secondary contaminants" often involves softening the
water to reduce magnesium and calcium levels or
applying chemical sequestrants for iron and/or
manganese contamination. Although not a public
health concern, the aesthetic problems caused by
secondary contaminants may prompt some
consumers to seek more palatable, but less safe or
more expensive, sources of water.
Storage Needs, The total 20-year need for storage
projects is $24.8 billion, $12.9 billion of which are
current needs. This category includes projects to
construct or rehabilitate finished water storage tanks.
A 'water system with sufficient storage can provide an
adequate supply of treated water to the public even
during periods of peak demand. The system can
sustain the minimum pressure required to prevent the
intrusion of contaminants into the distribution network.
Moreover, many states require that systems have the
storage capacity to provide a 1 - to 2-day supply o?
water in the event of an emergency, such as a water
source being temporarily unusable.
Source Needs. The total 20-year need for source
water infrastructure is $12.8 billion. Of this total, $6.7
billion are current needs. The source category
includes needs for constructing or rehabilitating
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page 29
surface water intake structures, raw water pumping
facilities, drilled wells, and spring collectors,
Drinking water is obtained from either ground water or
surface water sources. Wells are typically considered
ground water sources; rivers, fakes, other open
bodies of water, and wells under direct influence of
surface water are considered surface water sources.
Whether drinking water originates from ground or
surface water sources, its raw water quality is an
important component in protecting public health. A
high quality water supply can minimize the possibility
of microbiai or chemical contamination and may not.
require expensive treatment facilities. Many source
water needs involve construction of new surface
water intake structures or drilling new well fields to
obtain improved raw water quality.
A water source should also provide enough water
under all operating conditions to enable the water
system to maintain minimum pressures, even at peak
flows. Low water pressure may result in the intrusion
of contaminants into the distribution system through
backsiphonage. The 2003 Needs Assessment
includes projects to expand the capacity of intake
structures and acid new wells to address supply
deficiencies.
Other Needs. Needs not included in the previous
categories are labeled "other" needs. These needs
account for $2.3 billion of the total 20-year need.
Examples of "other" projects include system-wide
telemetry or Supervisory Control and Data Acquisition
(SCADA), and system-wide security measures.
The Regulatory Need
As shown in Exhibit 4,16 percent of the total national
need, or $45.1 billion, is for compliance with current,
new. and proposed SDWA regulations. Although al! of
the projects in the 2003 Needs Assessment are
needed to attain or maintain compliance with the
SDWA regulations and goals, most are driven by the
need to provide an essential service—potable water—
to the utility's customers. However, some of the
Exhibit 4, 20yeaf Total RegtilettQify
;and No^pigulalory Nf ed
;{fl£.:$ifc^
Ragoiatoty
£48- *
•- 26-Year &em-Regufstory
Note: Numbers may not total due to rounding.
projects are directly attributable to specific regulations
under SDWA. These projects are collectively referred
to as the "regulatory need." Most of the regulatory
need involves the upgrade, replacement, or
installation of treatment technologies.
The total regulatory need is divided into two broad
categories: existing SDWA regulations ($35.2 billion),
and recently promulgated or proposed regulations
($9.9 billion). Exhibit 5 displays the regulatory need by
type of existing regulation. For reporting purposes, the
recently promulgated Arsenic Rule is included in the
existing regulations section because the total need
has been distributed amongst the states.
Proposed or Recently Promulgated Regulatory
Needs. The total need to comply with proposed or
recently promulgated regulations is $9.9 billion. Of the
total, $3.2 billion is to address microbia! contaminants
that have acute health effects. The total costs of these
regulations are included in the 2003 Needs
Assessment as future regulatory needs.
The regulations included in this category are the
Stage 1 and Stage 2 Disinfectants/Disinfection
Byproducts Rules (Stage 1 and Stage 2 DBPR), the
Radon Rule, the Ground Water Rule, the Filter
Backwash Recycling Rule (FBRR), the Long Term 1
-------
Page 30
2003 Drinking Water Infrastructure Needs Survey and Assessment
Regulations
Existing SDWA Regulations
Interim Enhanced Surface Water Treatment Rule and Surface
Water Treatment Rule1
Totai Coliform Rule1
Nitrate/Nitrite Standard1
Costs Associated with the Recently Promulgated Arsenic Rule
Lead and Copper Ruie
Total Trihalomethanes Standard
Other Regulations2
Subtotal National Need
Costs Associated with Proposed and Recently Promulgated
Regulations (Taken from EPA Economic Analyses)3
Iota! National Need
Total Need
S27.5
$2.6
$0.5
$1.0
$2.0
S0.2
$1,3
$35.2
$9.9
$45.1
Note: Numbers may noi total due to rounding.
' Ragufcitions for contaminants that cause acute health effects.
'l Includes regulated Volatile Organic Chemicals (VCCs), Synthetic Organic Chemicals (SOCs).
Inorganic Chemicals (OCs); and Radionuclidss.
a Includes regulations tor contaminants that cause acute and/or chrortc health effects. In the
Economic Analyses, the compliance costs with some regulations are given as a range. In
calculating the $9.9 biliion need, the 2003 Needs Assessment used EPA's lead option, untess
one was not available, in wrtich case The 2003 Needs Assessment used tha highest estimate.
These estimates include only the capital costs (i.e., excludes operation and maintenance costs).
Costs for the recently promuigaied Arsenic Rule are noi included in this category.
and Long Term 2 Enhanced Surface Water Treatment
Rules (LT1ESWTR and LT2ESWTR), and the
Radionuclides Rule. Capital cost estimates for each
of these rules are provided in Exhibit D-9. EPA derived
the estimates from the Economic Analysis (EA) that
the Agency published when proposing each
regulation, or from the final EA (if the regulation has
been promulgated).
In general, water systems can readily identify the
infrastructure needs required for compliance with
existing regulations, but most systems have not yet
determined the infrastructure needed to comply with
future or recently promulgated
regulations. Therefore, relying on
systems to report the costs of future or
recently promulgated regulations
would significantly understate the true
need. Because of this, EPA relied on
EAs to estimate these compliance
costs.
However, since the EAs rely on
regional data, they are noi good
predictors of state-specific needs.
Therefore, the costs associated with
the proposed or recently promulgated
regulations other than the new arsenic
standard are allocated at a national
level, not apportioned to each state.14
Existing Regulations
Microbial Contaminants. The
Surface Water Treatment Rule
(SWTR), the Interim Enhanced
Surface Water Treatment Rule
(IESWTR) and the Total Coliform Rule
(TCR) are SDWA regulations that
address microbial contamination.
Projects directly attributable to these
regulations account for $30.2 billion, or
86 percent of the total existing
regulatory need. (Note: Numbers may
not total due to rounding.)
The SWTR and the IESWTR account for almost all of
the microbial contaminant-related need and most of
the total regulatory need. This reflects the fact that the
majority of the nation's large municipal systems use
surface water sources. Under these regulations, all
systems using surface water sources must provide
treatment to minimize microbial contamination. In
most cases, this means installing filtration plants to
remove and inactivate microbial pathogens, such as
the bacterium E. coli, the virus Hepatitis A, and the
protozoan Giardia lamblia. Projects associated with
" See the section in Appendix B, "Estimating Costs for Proposed and Recently Promulgated Regulations,0 for a more detailed discussion.
-------
2003 Drinking Water infrastructure Needs Survey and Assessment
Page 31
«»;§: mmmm®m®
Many water sources in the tiati&n have arsenic levels above the future
regulator;.' limit of 0.01 Q milligrams par liter (tng/L). To meet the more.
'stringent arsenic regulation, the pilot plant shown is examining
athwptive and specially media for optima! arsenic removal.
chemicals. Examples of projects include
aerating water to remove volatile organic
compounds, such as tetrachloroethyiene,
and applying corrosion inhibitors to reduce
the leaching of lead from pipes in home
plumbing. This category includes regulation
of more than 80 inorganic or organic
chemicals for which infrastructure projects
may be needed.
Most chemical contaminants are associated
with chronic health effects such as cancer,
reproductive difficulties, and liver or kidney
problems. However, nitrate levels above the
health-based standard can cause an acute
illness, known as "blue baby syndrome," a
condition in which infants are deprived of
oxygen in the bloodstream. Also, excessive
copper levels can induce acute
gastrointestinal illness.
these regulations also include rehabilitating and
upgrading existing treatment facilities. Disinfection for
compliance with the IESWTR and the SVVTR would
also protect the system from TCR violations.
Chemicai Contaminants. Existing SDWA regulations
to minimize chemical contamination accounts for $5.0
billion of the total regulatory need. This estimate
includes projects attributable to the Nitrate/Nitrite
Standard, the recently promulgated Arsenic Rule, the
Lead and Copper Rule, the Total Trihaiomethanes
Standard, and other regulations that set MCLs or
treatment techniques for organic and inorganic
Security Needs
Since the September 11lh tragedy, there has been a
concentrated national focus on our vulnerabilities, and
water systems are no exception. The Public Health
Security and Bioterrorism Preparedness and
Response Act of 2002 requires any community water
system that serves more than 3,300 people to prepare
a Vulnerability Assessment. Systems serving at least
50,000 people should have completed the vulnerability
assessments during the data collection period of the
2003 Needs Assessment.
Current and Future Regulatory Needs
Of the $45.1 billion total regulatory need, $21.0 billion is the current need for attaining and maintaining
compliance with existing regulations. Most water systems with current regulatory needs are currently not in
violation of any health-based standards, Rather, these systems identified needs that would enable them to
continue to maintain compliance with existing regulations. Water systems also identified projects for future
regulatory needs, such as projects that are largely due to the routine rehabilitation or replacement of
infrastructure. For example, most conventional filtration plants require the refurbishment of pumps, filters,
chemical feed units, and other components within a 20-year period. Ail of the: costs associated with the
proposed or recently promulgated regulations are included as future regulatory needs.
-------
Page 32
2003 Drinking Water Infrastructure Needs Survey and Assessment
Although water systems have begun to identify their
security needs, responses indicate that in 2003 many
did not yet have a complete grasp of these needs or
their costs. States reported that systems had difficulty
in determining security needs because many were in
the process of developing their vulnerability
assessments, but had not yet completed their
comprehensive review.
Despite these limitations, the 2003 Needs
Assessment represents some information on security
needs from medium and large water systems, and
EPA did receive a substantial response from larger
metropolitan utilities regarding their security needs. Of
the total national need, $1.0 biiiion was identified as
having security as at least one reason for the need.
The 2003 Needs Assessment put security projects
into five categories. Exhibit 6 provides the breakdown
of the total security need into these categories:
« System-wide or major security projects
• For security reasons, systems could not
reveal detail
* Projects to prevent or detect an intrusion or
security violation
- Fencing, lighting, cameras
• Projects for redundancy or to respond to a
security breach
•• Generators, parallel pipelines, redundant
tanks
» Communication needs for security
- System Control and Data Acquisition
(SCADA) or telemetry
» Projects to address public health and safety
issues
- Changing from gas to liquid chlorine
In addition, many other projects included a component
of security as part of a large project. An example is
inclusion of a security fence in the description of a
project for construction of a new treated water storage
Exhibit 6: Security N6eds by Proj|
{Percentage of Si bWion Security Neetl
x-x-x^x-x-x-: :•:-: :•:•:•:•• :-:-•:-: •:•:. T*T - - - . .-.-.-:. •. •. .-..••. • • •.-.•.•.. .-.•: •;•:-:-;:. :•;•:•:•:-:•;•:- •:•*• . .•:•: .•:-: x-x-:-x-xr-
18 CcxYwutticaUon needs for security
aii Proj<fof redtHKiaDuy ui w> respond w>» ssc
^ PfnjRrts til j.»RVi«:i at rinfatS asi »:lsi?c«r
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page 33
With inweased security awareness, (he total national security need
could risa significantly in the fittum. Fencing, security cameras and
closed circuit television, and well housing are all common first
generation security needs. In the future, these items will likely be
included in the. cos! of building new facilities.
Community Water Systems Serving
10,000 and Fewer People
The 2003 Needs Assessment estimates for systems
serving fewer than 10,000 people represent $72 billion
or 28 percent of the total national need for community
water systems regulated by the states. In
approximately one-third of the states, these systems'
needs comprise over 50 percent of the state's total
need. Exhibit E-1 presents the 20-year needs for
systems serving 10,000 people and fewer by state.
The SDWA requires that states use at least i5
percent of their DWSRF funding for financial
assistance to water systems serving populations of
10,000 or less. Through FY2003, states had
allocated 40 percent of their assistance to those
systems.
Systems serving 10,000 people and fewer face
considerable economic challenges in delivering safe
drinking water to their consumers. The substantial
capital investments required to rehabilitate, upgrade,
or install infrastructure represent one such challenge,
Smail systems lack the economies of scale that allow
larger systems to spread the costs of capita!
improvements among their many consumers. For
example, the installation of a new 1.0 million gallons
per day (MOD) conventional treatment plant designed
to serve a community of 5,000 people may cost
approximately $2.5 million or $500 per person,
whereas a 20 MOD plant serving 150,000 people may
cost $30 million but will cost $200 per person. The
cost per household is substantially higher for the
smaller community. Moreover, larger systems are
usually able to purchase material in quantities that
result in significant savings.
Not-for-Profit Noncommumty Water
Systems
EPA adjusted the 1999 Needs Assessment results to
January 2003 dollars to determine the estimate of
need tor not-for-profit noncornmunity water systems.
These systems need to invest $3.4 billion in
infrastructure improvements over the next 20 years.
Of this total, $1,2 billion is identified as current needs
to ensure the continued protection of public health.
Exhibit 7 presents the not-for-profit noncornmunity
need by project type. In comparison to community
(in millions of January 003|lpiar$)
-------
Page 34
2003 Drinking Water Infrastructure Needs Survey and Assessment
water systems, noncommunity water systems
typically have limited distribution networks; therefore,
a higher percentage of their needs are storage needs.
The needs of not-for-profit noncommunity systems
comprise a small proportion of the tola! national need.
This reflects the limited infrastructure required for a
noncommunity system. In spite of their modest
contribution to the total national need, noncommunity
water systems are important. Approximately half of
the nontransient1s noncommunity systems are
schools and daycare centers serving water to
sensitive populations. For this reason, the Agency
believes that investing in the infrastructure of these
systems is an important contribution to public heaith.
American Indian and Alaska Native
Village Water System Need
Because of the effort made in the 1999 Needs
Assessment, and the high confidence level in the data
from that effort, EPA did not resurvey the American
Indian and Alaska native viilage water systems for the
2003 Needs Assessment. Instead, the need
established in 1999 was adjusted to 2003 dollars and
used as an estimate for the 2003 need.
According to the 2003 Needs Assessment, the
American Indian and Alaska native village water
systems need to invest an estimated $2.4 billion in
capital improvements over the next 20 years. Of this
total, $2.3 biilion is identified as current needs to
ensure the continued provision of safe drinking water.
Exhibit 8 presents the total need by project type for
American Indian and Alaska native viiiage systems.
The 2003 Needs Assessment indicates that of the
estimated $2.4 billion total needs, American Indian
water systems need to invest $1.3 billion and
Alaska native village water systems need to invest
$1,2 billion in capital improvements. (Note:
Numbers do not total due to rounding.) EPA
estimates that American Indian and Alaska Native
water systems will need to invest $14.7 million to
comply with the recently promulgated Arsenic Rule.
t TbtaJ American .Indian and Alaska Native Village Water System Need
(in millions of January 2003 dollars)
American Indian Water System Need
Alaska Native Village Water System Need
ycict
Note-. Numbers may no', total due to rounding.
n There are two types of noncommunity water systems: those that serve transient populations (s g.. restaurants, roadside rest areas) anc! those
that serve the same populations more than 6 months of the year (e.g., schools, factories, and office buildings). The second type are called
"ncntransieni" noncommuniiy sysioms.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page 35
For American Indian systems, the widely
dispersed and remote location of many
communities and the limited availability
of water resources are among the
logistical challenges that account for
high per-household needs. Alaska native
village water systems face higher costs
because of their remote arctic locations
and the unique design and construction
standards required in permafrost
conditions.
American Indian Water System
Needs
The total 20-year need for American
Indian systems is $1.3 billion. Of this
total, approximately $1.1 billion is
identified as current needs to provide
safe drinking water. Exhibit 9 presents
the total need by project type for
American Indian systems. Exhibit D-6 presents the
American Indian need by EPA Region.
Alaska Native Village Water
System Needs
The total 20-year need for Alaska native
village systems is $1.2 billion. Of this
total, approximately $1.1 billion is
identified as current needs to ensure the
continued provision of safe drinking
water. Exhibit 10 shows the total Alaska
native village need by project type. The
Alaska native village need contributes a
disproportionately large share to the
total national need on a per-household
basis. The need for Alaska native
villages differs from other community
water systems in that costs for storage
in Alaska native villages exceed those
for treatment needs.
Categories of Need
Distribution and
Transmission
Treatment
Storage
Source
Other
Total Need
Current Need
$758.5
$172.4
$116.6
$71.1
$12.8
$1,131.4
Future Need
$56.8
$2-1.3
$34.1
$16.7
$0.0
$131.9
Total Need
$815.4
$196.7
$150.7
$87.8
$12.8
$1,263.3
Note: Nuf-bers may not tola! due to rounding.
American Indian water system needs were adjusted from 1i599 Needs Assessment findings to January
2003 dollars.
Ooes not include me costs associated wild the recently promulgated Arsenic Rule and proposed or recently
promulgated SDWA regulators.
Categories of Need
Distribution and
Transmission
Treatment
Storage-
Source
Other
Total Need
^•'IIIIMM'IIMIIMttlMIBa
Current Need
S523.5
S232.5
$320.9
$38.0
$0.8
81,120.6
mm?mmmii88i88®£
Future Need
S3.5
$18.4
$18.8
S9-3
$0.0
$49.9
Total Need
$531.9
$250.9
$339.6
$47.3
$0.8
$1,170.5
Note: Numbers may not total due to rounding.
Alaska native village mate' system reeds were adjusted from 1999 Needs Assessrnerrt findings to January
2C03 dollars.
Does rot include the costs associated with the recently promulgated Arsenic Rule- and proposed or recently
promulgated SDWA regulators.
-------
Page36
2003 Drinking Water Infrastructure Needs Survey and Assessment
Totai Need Compared to Previous
Needs Assessments
The total need of $276.8 billion established by the
Agency from the 2003 Needs Assessment
substantially exceeds the 1995 Needs Assessment
estimate of $167.4 billion and the 1999 Needs
Assessment estimate of $165.5 billion.1fi
The 2003 Needs Assessment workgroup identified
several factors that came into play in capturing what
is believed to be a more accurate representation of
total national need for this assessment, as follows:
® First, this was the third Drinking Water
Infrastructure Needs Survey and Assessment.
Most states had a much better understanding
of how the assessment was conducted than in
1995 as well as in 1999.
« For each assessment, questionnaires were
sent to all of the largest water systems;
therefore, many of the utilities were also
familiar with the process.
« For the 2003 Needs Assessment, the
questionnaire included several pages
prompting systems to more closely examine
the current condition of the entire system
inventory, and to better consider their
replacement and rehabilitation needs for aging
infrastructure.
* Criteria for replacement of domestic water
meters was modified and the requirement for
documentation of system ownership of
backflow prevention devices and service lines
was removed.
• The interactive Web-based database enabled
states to more easily and clearly submit
additional information to EPA.
• EPA conducted extensive state training for the
2003 Needs Assessment at several regional
locations to help states understand the
questionnaire itself and the process to be
followed, and to underscore the importance of
cooperating with the 2003 Needs Assessment
and accurately representing total water
system needs.
Needs Assessment Tools
* The Needs Assessment Guide
* The toil free helpline
* The Needs Survey Web site
* Direct access to contractor support
One comparison between the 1999 and the 2003
Needs Assessments is the number of projects
submitted. In 2003, the projects submitted for large
and medium systems alone totaled 128,600. In 1999.
the total projects received for medium and large
systems was 61,400. This underscores the effort by
states and systems to provide information on
infrastructure needs.
Other differences include the following:
• The 1995 Needs Assessment included the
$6.3 billion capital need associated with dams
and untreated water reservoirs.17 After EPA
completed the first Needs Assessment, these
needs were determined to be ineligible for
DWSRF assistance and were consequently
excluded from the 1999 and 2003 Needs
Assessments.
18 The 1995 and 1999 total needs have bean converted to January 2003 dollars for comparison purposes. The 1995 need in 1995 dollars was
$138.4 billion. The 1999 need in 1999 dollars was $150.9 billion.
" Costs adjusted to 2003 dollars for comparison purposes. Costs ware originally $5.2 billion in 1995 dollars.
-------
2003 Drinking Water Infrastructure Needs Surrey and Assessment
Page37
Unlike the 1995 Needs Assessment, the 1999
and 2003 Needs Assessments each included
$3.4 billion1" in needs for not-for-profit
noncommurtity water systems that are eligible
for DWSRF funding.
* The varying estimates of costs associated
with the proposed and recently promulgated
regulations also contribute to the difference
between the assessments.
Despite these variations, the fundamental methods
used to collect and evaluate needs in 2003 remained
largely unchanged from the 1995 and 1999 Needs
Assessments. Most importantly, the 2003 Needs
Assessment retained the stringent documentation and
eligibility requirements of the previous assessments.
This photograph shoves the installation of a new raw
water line to the water treatment plant in Bartlesyillc.
Oklahoma. This DWSRF-fundeil project was
constructed to correct a deficiency inflow to ihii water
treatment plant.
1S Costs adjusted to 2003 dollars for comparison purposes. Needs for not-for-profit nonccmmunity water systems were $3.1 billion in 1999
dollars.
-------
APPENDIX A—PAYING FOR AND FINANCING
Paying for infrastructure Improvements
The 2003 Drinking Water Infrastructure Needs Survey and Assessment shows that the nation's public drinking
water systems need to invest $276.8 billion over the next 20 years to continue providing water that is safe to
drink. Investments of $165 billion are required to meet current needs. Given the size of the estimated needs,
how will utilities pay for these infrastructure improvements?
Although much of a water system's needs are met through consumer's rates, this funding does not always
cover the fuil cost of major capital investments. For this reason, local, state and federal programs have been
developed to help fill the gap.
The 1996 Safe Drinking Water Act (SDWA) Amendments created the Drinking Water State Revolving Fund
(DWSRF). The purpose of this program is to provide low-cost loans to drinking water systems. Federal
assistance to systems regulated by states comes in the form of "capitalization grants" to the states. This
"capital" is used by the states to start the revolving loan funds. As loans are paid off, money becomes available
for re-lending. Congress has appropriated more than §6.96 billion for the DWSRF from FY1997 through
FY2004.
In addition to EPA, other federal agencies have low-interest loan or grant programs. The largest of these
programs is provided by the Department of Agriculture through its Rural Utilities Service (RUS), which received
appropriations of $1.3 billion in FY2003 for both water and wastewater projects. The second-largest, program is
provided by the Department of Housing and Urban Development (HUD) through its Community Development
Block Grants; total disbursements for both water and wastewater in FY2003 amounted to $479 million. Finally,
the Economic Development Administration (EDA) in the Department of Commerce provides funds for physical
infrastructure, including water and wastewater systems.
Many states also provide loans and grants to water utilities from monies that their own legislatures have
appropriated. Some of these are coordinated with DWSRF capitalization grants. State funds (through matching
appropriations, leveraged bonds, principal loan repayments, or interest) account for 42 percent of the funds
available through the DWSRF. Other loans and grants may be coordinated with other available federal
assistance (including RUS, HUD, and EDA).
As the Congressional Budget Office (CBO) noted, "Ultimately, society as a whole pays 100 percent of the
costs of water services, whether through ratepayers' bills or through federal, state, and local taxes."19
19 Congressional Budget Office, op. til., page ix.
-------
Page 40 2003 Drinking Water Infrastructure Needs Survey and Assessment
Reducing the Cost of infrastructure
In 2002, EPA issued a report identifying that over the next 20 years a significant funding gap could emerge
between clean water and drinking water infrastructure investment needs and current levels of spending. The
following year, a national meeting was held entitled, "Closing the Gap: Innovative Responses for Sustainable
Water Infrastructure." where participants recognized that current spending and operational practices would
need to change in order to avoid the emergence of a funding gap that wouid hamper efforts to provide future
safe drinking water. The participants further recognized that federal funding is and will remain limited; initiatives
to adequately address the potential emerging gap will need to be based on improved management and water
conservation as methods for reducing the cost of infrastructure.
The concept of "sustainable infrastructure," announced at the January 2003 meeting, consists of "four pillars":
« Full Cost Pricing of Water. There are strong economic arguments for shifting more of the cost of
water from taxes to rates, and they are closely linked with smart water use. If consumers pay the full
cost of water, and if this results in higher rates, then the rate will send an appropriate "price signal" to
consumers and encourage conservation. The CBO recently estimated that future infrastructure
investment needs could be paid by ratepayers, and that this investment would increase water bills from
0.5 percent of income to 0.9 percent of income, on average,20 If these rate increases create problems
for low-income or fixed-income households, a wide variety of mechanisms are available to mitigate the
impacts, such as rate reductions or local subsidies to these households in the form of "life-line" water
rates.
« Better Management. There are proven management methods to reduce the cost of providing safe
drinking water and improving performance. One of these is asset management. This is a data-driven
approach to prioritizing investments in infrastructure so that they meet customer expectations. Armed
with detailed information on the age, condition, and performance of infrastructure, systems would be
able to replace infrastructure as needed to meet performance standards. This would optimize
investment. Savings from asset management approaches are estimated to be 10 percent of the capital
investment. Ten percent of the estimated infrastructure needs in this assessment ($276.8 billion) would
be $27.7 billion over 20 years, or $1.38 billion per year—more than the current federal contribution in
capitalization grants through the DWSRF. A related concept is environmental management systems
(EMS). These are comprehensive assessments of the utility's operations for continual improvement in
operations, resulting in better performance and lower cost.
* Efficient Water Use. Much of the needed investment reported in EPA's Needs Assessment consists of
installing new distribution pipe, treatment, or storage to meet the needs of the existing U.S. population.
These projects are sized to accommodate reasonably anticipated growth. Decreasing water use,
however, might reduce the projected increase in design capacity, thereby reducing investment needs.
EPA estimates that there could be a 20 percent reduction in water use if simple conservation methods
were introduced. This may translate to smaller capacity plants, which in turn would have reduced
capital and operating costs.
20 Ibid., page xvi.
-------
2003 Drinking Water infrastructure Needs Survey and Assessment
Page41
* Watershed Approach. There is great potential for cost savings in what EPA has broadly described as
the "watershed approach" to management. This term refers to policies that include broad stakeholder
involvement, hydroiogicaily defined geographic boundaries, and coordinated management across ail
policies that affect water. Specific practices may include incentives for pollutant reduction, purchase of
easements to minimize or eliminate poilutant sources, and conversion of land uses where such
approaches are cost effective.
No single initiative will answer the question of how to pay for the infrastructure needs identified in this
assessment. Yet, each has great potential, and none has been fully exploited. Taken together, and used in a
coordinated fashion with the significant levels of financial assistance available at the federal and state levels,
they provide an outline of how to pay for these infrastructure needs.
Regional "water systems serving rural areas require, long lengths of mains per
household se.rved. The. Part Pevk-f)rv Prairie. Regional Water System in nortfrwust
Montana serves a population
-------
Survey Design
EPA's 2003 Needs Assessment relied on a survey to determine the needs for medium and large water
systems. The survey is based on a random sample of water systems. This section provides an overview of
the survey design. A detailed description o? the design is in "2003 Drinking Water Infrastructure Needs Survey,
EPA iCR #2085.01."
Sample Frame
The first step of the sample design is to develop the sample frame. The sample frame is a list of all members
(sampling units) of a population from which a random sample of members will be drawn for the survey. The
sample frame is the basis for the development of a sampling plan to select a random sample. To ensure that
the survey accounted for all community water systems in the nation, the universe of water systems (from
which the samples were drawn) was obtained from the federal Safe Drinking Water Information System
(SDWIS-FED). SDWIS-FED is EPA's centralized database for information on public water systems. It includes
the inventory of all public water systems in the states and territories from which the states verify information
regarding population served, water sources, and other important variables for their systems. For the 2003
Needs Assessment's sample frame database, systems were categorized by source water and population
served. Some systems sell water to other water systems; for purposes of the survey, the population of the
purchasing systems is included in the seller's population.
EPA sent the sample frame, with the population served and the water sources, to the states for their review
and updated it based on the states' comments. The 2003 Needs Assessment excluded systems serving
populations of 3,300 or fewer, so these systems were dropped from the list. A sample of systems was then
selected from this updated sample frame.
Sample Design
EPA drew separate samples for each of the 50 states, the District of Columbia, and each of the trust territories.
The sampling design for the survey was stratified random sampling within each state. In stratified samples, the
population is divided into nonoverlapping subpopulations called strata and a simple random sample is taken in
each stratum. Stratification may increase the precision of the estimates when the population is divided into
subpopulations with similar characteristics within each stratum. Some water systems, as a group, will have
different needs than other groups of water systems. For example, large water systems generally require much
greater investment than do small systems, and systems that utilize surface water require more treatment (and
therefore incur more costs) than systems that utilize ground water. In this assessment, water systems were
stratified by source water type and system size based on the population served in each system.
-------
Page44
2003 Drinking Water Infrastructure Needs Survey and Assessment
Water Source. Systems were classified as either surface water or ground water systems. Systems
that use surface water, even if they also use ground water sources, were classified as a surface water
system. AN other systems were classified as ground water systems. Systems that rely exclusively on
purchasing treated water have very few treatment needs; therefore, their needs are more similar to
ground water systems than systems using and treating surface water sources. For this reason,
systems that solely purchase water were included in the ground water strata.
System Size. Systems were further stratified by the size of the population served. The size categories
varied by state and water source. In some cases, systems were divided into four size categories: 3,301
to 10,000,10,001 to 40,000, 40,001 to 50,000, and more than 50,000. In other cases, they were divided
into five categories: 3.301 to 10,000,10,001 to 25,000, 25,001 to 40,000, 40,001 to 50,000, and more
than 50,000. Five size categories were used if it resulted in smaller sample sizes than four size
categories. (Note that the population of purchasing systems was included when systems were
assigned to size categories, as described above.) Exhibit B-1 shows the size categories used by
different EPA drinking water programs.
Programs
2003 Needs
Assessment
Public Water System
Supervision Program
Drinking Water State
Revolving Fund
Size Categories
Extra Small
N/A
<500
N/A
Small
< 3,300
501 - 3,300
< 10,000
Medium
3,301 - 50,000
3,301 - 10,000
N/A
Large
> 50,000
10,001 -100,000
N/A
Very Large
N/A
>1 00,000
N/A
For systems serving populations of 3,301 to 40,000, EPA selected a random sample of systems from each
stratum. The target precision for the estimate of the need for each state determined the number of systems
selected in each stratum, as described below. The survey sample included 2,553 community water systems
serving populations of 3.301 to 40,000 out of the national inventory of 7,337 systems.
Systems serving more than 40,000 people were sampled with certainty. There is a relatively small number of
these systems in many states, but they serve a large share of the population and account for a large share of
the need. The survey included all of the nation's 1,342 systems serving populations of more than 40,000. At the
direction of the workgroup, it was assumed that systems serving more than 40,000 that do not respond to the
survey (approximately 4 percent) have no need and do not contribute to the needs of their state.
States were given the option of sampling with certainly the full set of systems serving populations of 3,301 to
40,000, rather than using a random sample of these systems. One state chose this method.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment Page 45
Sample Size Determination
The 2003 Needs Assessment workgroup determined the sample size for each state to achieve the target
precision of 95 percent ±10 percent for each state's estimate of need. The sample size for each state was
determined to achieve the target precision set for each state's estimate of need. The sample size was
selected so that the state's need would be estimated within 10 percent of the amount of the true need with 95
percent confidence. For example, if the survey estimates indicate a need of $2.0 billion, then there is a 95
percent probability that the interval of $1.8 to $2.2 billion includes the true need. Data from the 1999 Needs
Assessment were used to estimate the average need and standard deviation of the need for each state, by
stratum. These estimates were then used to calculate the sample size required for each state to meet the
precision target. Systems serving populations of 3,301 to 40,000 were oversampled to account for system
nonresponse. EPA assumed the response rate would be 90 percent, based on data from the 1999 Needs
Assessment. Once the sample size was selected for each state, the number of samples for each stratum
was allocated in a way that minimizes the sampling error of the estimate. See Exhibit B-2 for the sample sizes
for each state.
Weighting the Systems
EPA weighted the systems serving populations of 3,301 to 40,000 to account for variable probabilities of
selection and differential response rates. Weighting the data allows inferences to be made about all systems,
not just those included in the sample, but also those not included in the sample or those that did not respond to
the survey. For instance, in a given stratum in a given state, one system may be given a base weight of 10.
This means that only 1 in 10 systems in this stratum is included in the survey, and the needs of this system
represent its own and those of nine other systems.
The base weights and nonresponse adjustments reflect the probability of selection for each system and
adjustments for system level nonresponses, respectively. Systems serving more than 40,000 people received
a weight of one because they were selected with certainty.
Data Collection
The 3,895 medium and large systems in the survey were mailed a questionnaire package. Systems were
asked to identify capital projects needed to protect public health for current customers and for households
without access to safe drinking water. The questionnaire prompted systems to provide:
• A description of the infrastructure need
® Documentation explaining why the project is needed
« An indication of whether the project is a current or future need
• An indication of whether the project involves installing new or rehabilitating existing infrastructure
* An indication of whether the project is triggered by a Safe Drinking Water Act (SDWA) regulation
* A documented cost estimate, if available
• Design capacities of projects without costs for cost modeling
-------
Page46
2003 Drinking Water Infrastructure Needs Survey and Assessment
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Puerto Rico
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
American Samoa
Guam
Nordiem Mariana is.
Virgin islands
Total Number of Systems in Inventory
Paputaiityn Served
3,301-4r>seoQ ^40,000
293
14
77
145
432
84
42
17
0
295
175
28
36
402
181
115
72
225
192
31
39
205
245
139
185
152
28
41
26
33
189
46
300
211
25
281
139
87
259
99
19
131
40
235
697
77
30
110
140
95
147
22
0
3
4
2
45
2
18
16
224
19
17
6
1
92
39
3
4
58
19
13
10
11
17
1
13
39
47
21
6
14
3
3
6
2
35
6
62
29
4
34
11
15
74
23
8
14
2
51
99
17
2
31
25
8
16
4
1
1
1
0
Total
33&
1&
3&
1S1
S5&
133
23
1
387
214
31
40
4SO
200
128
S2
23&
2G&
32
52
244
292
160
191
16&
31
44
32
3&
224
52
362
31 &
150
192
335
122
27
145
42
286
796
94
32
135
193
2&
1
4
5
2
Number of Systems Selected in Sample
Population Served
3,301-40,000 *40,OCK}
134
10
26
74
32
2!
1!
6
0
25
26
23
17
71
97
31
16
129
102
17
5
46
33
57
99
44
7
31
7
9
40
12
300
76
19
91
54
34
55
19
8
66
25
140
72
11
15
110
36
58
80
17
0
3
4
2
45
2
18
16
224
19
17
6
•i
92
39
3
4
58
19
13
10
11
17
1
13
39
47
21
6
14
3
3
6
2
35
6
62
29
4
34
11
15
74
23
8
14
Z
51
99
17
2
31
25
8
16
4
1
1
1
0
Total
12
44
90
2S8
4&
12
1
Tf7
8&
28
2?
129
tie
44
28
140
IS
IS
as
m
78
1QS
58
10
34
13
n
78
18
3S2
105
23
125
65
49
123
42
16
30
27
171
23
17
m
66
96
2?
1
$
2
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment Page 47
Systems returned the completed questionnaires to the states for review, along with the supporting need and
cost documentation. The states reviewed each questionnaire to ensure that systems thoroughly identified their
needs and that ai! projects were documented and described correctly. The states had the option of providing
supplemental information if documentation of need or cost was inadequate. In many instances, the states
contacted the systems to obtain additional information. The states then forwarded the questionnaires to EPA
for final review. EPA reviewed each project for Drinking Water Infrastructure Needs Survey and Assessment
eligibility criteria, conformance to workgroup policies, adequacy of documentation of need, and documentation
of reported costs. EPA accepted or edited project information accordingly and coded each deficiency or
change made to each project. Once EPA's review was completed, the questionnaires were entered into a
database. This database was made available on the Internet to provide states with a final opportunity to review
their systems' data and provide additional information regarding the changes or deficiencies recorded by EPA.
EPA's review process in 2003 has evolved from the procedures used in 1995 and 1999. Although some states
were involved in data collection for the 1995 Needs Assessment, EPA assumed primary responsibility for
reviewing needs and. whenever necessary, contacting systems to obtain further documentation. The greater
involvement of the states—with their familiarity with the systems—accounts in part for the larger number of
projects received for the 1999 Needs Assessment. In 2003, the states were given more extensive training by
EPA and more responsibility for the review of the surveys. For this assessment, the number of projects as well
as total need increased significantly. This increase is believed to be a much more complete and accurate
representation of the nation's total water system capital needs.
Estimating Needs of Water Systems
Estimating Needs for Large and Medium Community Water Systems
Each system that responded to the survey provided information regarding each of its investment needs. The
sample included data on 128,600 infrastructure projects. Some of the medium and large drinking water
systems provided capital improvement plans or engineering reports to document the costs of their
infrastructure projects. However, approximately 82 percent of the projects lacked cost estimates. EPA used
models to assign costs to these projects. For the most part, EPA developed the cost models from the 1999
Needs Assessment and adjusted the costs to 2003 dollars to estimate current costs. EPA developed two new
models for the cost of installing and rehabilitating pipe and a third model for installation of domestic meters.
New models were needed for pipe installation and rehabilitation because the models had not been updated
since the 1995 Needs Assessment. A new model was needed for service meters because generally accepted
technology had changed from manual-read to radio-read meters. All costs provided by systems or modeled by
EPA were converted to January 2003 dollars.
Exhibit B-3 provides an example of a cost curve used to apply costs to a. new conventional treatment plant
project. A cost model would have been used if a system knew that it needed to rehabilitate a conventional
filtration treatment plant that no longer met performance standards but did not have documentation of cost. If
the system provided the design capacity of the plant on the questionnaire, EPA would have applied the specific
cost model for rehabilitating this type of plant.
-------
Page48
2003 Drinking Water Infrastructure Needs Survey and Assessment
1,000,000,000 3
100,000,000
10,000,000
8 1,000,000
100,000
10,000
1,000
0.01
0.1 1 10 100
Millions of Gallons per Day
1000
The investment need for each system in the sample was estimated based on the reported and modeled costs
of each project in the sample. The total need for medium and iarge systems was then estimated for each state
by applying the sample weights to the total need for each system. The need for each system was multiplied by
the sample weights; this product was then summed across all systems to produce the total need for medium
and large systems in each state.
Estimating Needs for Small Community Water Systems
EPA estimated small system need based on the findings of the 1999 Needs Assessment. The 1999 Needs
Assessment collected data on a national sample of small systems. These needs were adjusted to January
2003 dollars using a factor of 1.097 and apportioned among the states based on the inventory of small
systems. EPA believes that the 1999 data are credible because they were collected through EPA site visits by
water system specialists who had extensive experience working with small systems, and who received
training in the project eligibility and documentation criteria established for the survey.
Estimating Needs for American Indian, Alaska Native, and Not-for-Profit Noncommunity Water
Systems
EPA estimated needs for American Indian, Alaska native village, and not-for-profit noncommunity water
systems based on the findings of the 1999 Needs Assessment. In 1999, EPA conducted site visits or provided
assistance in completing the questionnaire to all American Indian systems, Alaska native village systems, and
to a sample of approximately 100 not-for-profit noncommunity water systems. Data collection and cost
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment Page 49
modeling were completed using the same methods applied to small community water systems. The needs
calculated from the 1999 data were adjusted to January 2003 dollars for the 2003 Needs Assessment effort,
Estimating Costs for Proposed and Recently Promulgated Regulations
A portion of the needs collected in the 2003 Needs Assessment are attributable directly to SDWA regulations.
Systems were able to identify projects needed for compliance with existing regulations. However, most
systems had not yet identified the infrastructure needed to comply with proposed and recently promulgated
regulations. Consequently, the need for complying with these regulations was based on the Economic Analysis
(EA) that EPA presents when proposing or finalizing each regulation. The 2003 Needs Assessment did not
include the costs of regulations that were proposed after August 2003.
The costs associated with most future and recently promulgated regulations are included in the total national
need only, not allocated at the state level. In general, the use of EAs to allocate these costs to each state is
problematic, given that the cost of a regulation is not necessarily a direct function of the number of systems in
each size and source category. The cost of compliance with a new regulation will vary significantly from state
to state if the contaminant occurs mostly in specific regions of the country. Allocating costs based solely on the
inventory of systems would fail to capture this variation.
However, the recently promulgated Arsenic Rule is somewhat different in that many states did have
occurrence data for the number of systems with arsenic over 10 parts per billion. Therefore, the total national
cost of complying with the recently promulgated Arsenic Rule was taken from the EA and allocated to each
state based on these occurrence data.
-------
information Quality
The 2003 Needs Assessment foliowed the Agency's Guidelines for Ensuring and Maximizing Information
Quality.21 EPA's goal is to ensure the quality, objectivity, utility, and integrity of information disseminated by the
Agency. The Agency developed the guidance document to incorporate the government-wide guidelines issued
by the Office of Management and Budget (OMB) pursuant to section 515 of the Treasury and General
Government Appropriations Act of 2001 .^ Information quality is particularly important when the Agency
disseminates the results of research, and where those research results iead to policy decisions. Because the
results of the 2003 Needs Assessment will be used to allocate Drinking Water State Revolving Fund (DWSRF)
capitalization grants, data quality is critical.
Quality Systems: The cornerstone for maximizing information quality is the Agency's Quality System. All EPA
Offices, and all contractors working for EPA, have Quality Management Plans (QMPs) that outline detailed
procedures for quality assurance and quality control. The specific procedures required for each project are
documented in a quality assurance project plan (QAPP). The plan outlines all of the steps that the project team
will follow to ensure that quality is built into the project from the start. Since the 2003 Needs Assessment was
similar to the Needs Assessments of 1999 and 1995. the QAPP included all lessons learned from the previous
research.
The most important task at the start of each project is the definition of data quality objectives (DQOs). These
define the policy decisions that will result from the research and the precision targets for data collection. The
DQOs for this project were established for the 1995 Needs Assessment and, with some slight modifications,
these DQOs remained the same for the 2003 Needs Assessment. The primary DQO for the 2003 Needs
Assessment was to maximize precision of the estimates of state needs. The specific precision requirement for
each state was that the maximum half-width of the 95 percent confidence interval estimate of the total need
was to be no more than ± 10 percent of the total need for each state. Since the 2003 Needs Assessment relied
on a survey of a random sample of systems, the precision target for the survey was defined in terms of
acceptable sampling error. For more information on the sample design and the quality assurance procedures
for the sample frame, see Appendix B.
A distinctive feature of the Needs Assessments is that important questions are decided by a state/EPA
workgroup that meets regularly throughout the project. At the start of the 2003 Needs Assessment, the
workgroup met to review the lessons learned from the last assessment. The workgroup reaffirmed the DQOs;
and made suggestions for improvements in data collection for 2003, One problem identified in 1995 and 1999
was an apparent underreporting of needs. This was addressed explicitly in the 2003 Needs Assessment
approach. The workgroup recommended:
21 U.S. Environmental Protection Agency, Guidelines for Ensuring and Maximizing the Quality, Objectivity, and integrity of Information Disseminated
by the Environmental Protection Agency, EPA/260r- 02-008 (October 2002).
« Federal Register, Vol. 67, No. 36, February 22, 2002. pp. 8452-6460.
-------
Page 52 2003 Drinking Water Infrastructure Needs Survey and Assessment
• Changes in the questionnaire that wouid force systems to think more comprehensively about their
needs, including those not covered in existing capital improvement plans (CIPs), and
• Revised and enhanced training for state coordinators for their increased role in 2003 (assisting
systems, reviewing questionnaires, and tracking questionnaires through the review process).
In addition, the 2003 Needs Assessment workgroup used quality assurance techniques that had worked well
in previous surveys. Recognizing that one of the largest potential sources of error in a sample survey is
nonresponse, EPA took steps to ensure high response rates. These steps included the following:
• Questionnaires were shipped with a prepaid return envelope via Federal Express,23 enabling EPA to
track the shipments.
* EPA made available a toll-free telephone helpline (operated by an EPA contractor) to answer questions
about the questionnaire.
* The contractor electronically tracked all questionnaires and provided lists of nonrespondents to the
states for follow-up.
» EPA contacted utility organizations and specific systems, as requested by states, to encourage
participation.
The result of these efforts was a 96 percent response rate to a mail survey.
EPA designed several procedures to ensure quality control of the data collected by the questionnaires.
* The first step was intensive training of all professionals who were involved in the review of
questionnaires. It was critical that all personnel (EPA, state, and contractor) have a shared
understanding of the objectives of the data collection. All personnel also needed detailed training in the
completion of the questionnaires and use of the project coding.
• Systems sent their completed questionnaires to their state coordinators. This gave the states an
opportunity to review the questionnaires, request additional information, and make corrections based
on their knowledge of the systems.
• The states then sent the questionnaires to EPA's contractor, The Cadmus Group, Inc., who also had
provided technical support to the Needs Assessments in 1995 and 1999. Cadmus professional staff
reviewed the questionnaires to ensure that they met agreed-upon survey policies and quality
standards. One critical objective of this review was to eliminate all unallowable or undocumented
needs. Another was to ensure data were coded correctly and were consistent with each project's
documented purpose, scope and cost. Changes made at this stage of review were coded so that
states could see the rationale for these changes.
" In response to an inquiry in 1995: EPA calculated the costs and benefits of using Federal Express versus the U.S. Post Office. Given the rate
structure that EPA had negotiated with Federal Express, EPA demonstrated that this was cheaper and more effective.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment Page 53
® After this initial contractor review, the questionnaires were forwarded to a senior Cadmus professional.
The purpose of this level of review was to ensure quality control of all contractor work.
• Cadmus personnel also did in-depth reviews of the first five questionnaires submitted by each state,
including a telephone conference call with state personnel. This enabled Cadmus to talk through any
problems with the questionnaires so that corrections could be made in state review processes. These
interactions with states were enhanced by regular telephone conference calls with the state/EPA
workgroup where state concerns could be resolved in a collaborative process.
After the review, the contractor entered the information from the questionnaires into a data system. The data
entry process used an automated program keyed to the questionnaire, which precluded any invalid entry for
each question. To further assure quality, the program included 100 percent verification using double key entry.
During all of these steps, the contractor used an electronic tracking system to track the progress of the
questionnaires. Each time the questionnaire changed hands, from the time it was mailed out through each
review step and data entry, the contractor knew exactly who had the document. This information also was
shared with the states though a dedicated Web site. The states knew the status of each questionnaire. They
could see changes that had been made during the review process, and they had an opportunity to modify
project information through the Web site, by fax, or by mail. All modifications made by reviewers were coded to
create a record that explained all changes.
Quality control of the database consisted of several steps. The first step was automated computer edits
looking for out-of-range values for any variable. The second step was automated logic edits. Some of these
tests looked for extreme values for specific variables (e.g., a small system that reported it needed to replace
10,000 miles of distribution pipe, probably meant to report 10,000 feet of pipe). Other automated tests focused
on relationships between variables. For example, if a system purchased all of its water, it would be unlikely to
have a major treatment plant.
Variables that failed any of these tests were identified in a report, and a data supervisor was able to examine
the original questionnaire to determine whether the anomaly occurred in the original data. If the anomaly was
in the questionnaire, then questions could be posed to the state.
As in past assessments, EPA clearly defined the concept of "eligible infrastructure" in the questionnaires and
training. EPA also used quality control procedures to rigorously enforce that definition when reviewing project
documentation. For a project to be included in the 2003 Needs Assessment, documentation describing its
purpose and scope had to accompany each need. The documentation was reviewed by EPA to determine
whether the projects submitted for the 2003 Needs Assessment met the eligibility criteria for DWSRF funding
and allowability criteria set for the 2003 Needs Assessment. The state/EPA workgroup established the
documentation requirements so that uniform criteria were applied to all questionnaires. These requirements
not only lent credibility to the findings, they also addressed the issue of fairness in using the results to apportion
DWSRF funds.
Of the 128,600 projects submitted to the survey, EPA deleted 18 percent that failed to meet the documentation
criteria or appeared to be unallowable based on workgroup criteria or ineligibility for DWSRF funding. EPA
-------
Page 54 2003 Drinking Water Infrastructure Needs Survey and Assessment
adjusted the projects to correct a variety of measurement problems: overlaps between two projects (raising
the issue of double-counting), inconsistency with project documentation, and use of overly aggressive
infrastructure life cycles by states where system planning documents were not used or available.
To adjust for the use of aggressive infrastructure life cycles in estimating need, EPA made technical
adjustments to individual projects based on engineering literature and benchmarks of engineering practices.
The adjustments were tailored to the unique assumptions implemented by each state and were negotiated
with state officials. The general direction of these adjustments was to place a cap on the state's assumptions
about the rate of rehabilitation and replacement of pipe, unless there was project-specific documentation of a
need provided by the water system.
internal and External Review: A June 7.1994 EPA policy makes it clear that peer review should be part of
the design of any research project. In fact, the policy states that "peer review at the planning stages can often
be extremely beneficial." The 2003 Needs Assessment is the third in a series of assessments that EPA
performed (every 4 years) since 1995. Peer review has been part of the planning process from the very
beginning. Continued external review, provided by the state/EPA workgroup, was essential in ensuring that the
research met its intended quality objectives.
EPA sought external review of its approach to the first Needs Assessment, in 1995. Since the 2003 Needs
Assessment was a major data-collection project that required substantial efforts from water systems and
states, EPA distributed its study approach to industry and professional associations. These organizations
provided helpful criticism of the approach, which led to changes in the study design. Because the results o? the
assessments have a direct impact on states (through the allocation of DWSRF capitalization grants), the
Agency has consulted regularly with a workgroup composed of federal and state personnel. Since 1995, state
personnel have provided technical reviews of each study's approach, data collection methods, and analysis.
EPA developed the statistical design for the assessment in 1995, in consultation with the workgroup. The
workgroup specifically reviewed the critical decision on the proposed level of statistical precision. States,
especially, were consulted about their preferences for the level of precision of the state-specific estimates that
EPA would use to determine the allocation of DWSRF capitalization grants among them. Upon receipt of the
precision targets, EPA developed a statistical design, using the Neyman allocation formula, which would most
efficiently achieve those objectives. The entire design was subjected to internal review by statisticians in the
Office of Water (OW) and the Office of Regulatory Management and Information (now the Office of Information
Analysis and Access in the Office of Environmental Information). The statistical design was further reviewed by
specialists in the OMB during the evaluation of the Information Collection Request (ICR) for the study. The
statistical design has remained basically the same since 1995; any changes to the design have been noted in
the ICRs, which are reviewed by statisticians in EPA and at OMB.
Closely related to the technical approach and statistical design is the QAPP. Consistent with Agency policy on
quality assurance, the QAPP is reviewed by an independent, quality specialist in OW before work can proceed.
The quality assurance process is also the subject of audit by EPA Quality Staff, thereby providing additional
internal peer review by experts in quality assurance methods. The QAPP is updated completely at the start of
each assessment cycle, and it is amended as necessary during the assessment period.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment Page 55
The results of the 1995 and 1999 Needs Assessments have been critically reviewed by external parties,
especially those in the water utility industry24 and government accounting offices.25'28 These reviews have been
instrumental in the changes to the study design. For example, past Assessments had been criticized for
underestimating infrastructure needs, so EPA changed the research design for the 2003 Needs Assessment to
correct that problem. Such external reviews, coupled with the ongoing input from the external members of the
state/EPA workgroup, provide a continuous source of ideas for improving the quality of each assessment,
Transparency and Reproducibility: EPA's Guidelines on information Quality explain that influential
information (such as this report) "should be subject to a higher degree of quality (for example, transparency
about data and methods). Such transparency facilitates reproducibility of this information, and reproducibility
should meet commonly accepted standards." The Information Quality guidelines emphasize the importance of
transparency in information that is disseminated to the public and used to make policy decisions. EPA believes
that transparency is also useful during the research process, especially when state officials are important
stakeholders.
The 2003 Needs Assessment (like its predecessors) maintained high standards of transparency. Since 1995,
the Needs Assessments have been guided by a state/EPA workgroup that meets regularly throughout the study
period. These face-to-face meetings are supplemented by telephone conference calls. Every important
decision about the assessments—from the technical approach, questionnaires, data collection methods, and
statistical design, to the cost models, analysis of data, and, after submission to Congress, the Report to
Congress-—is discussed by the workgroup. Decisions generally are made by consensus. A centra! concern of
the workgroup is fairness to all stakeholders,
At the beginning of each new assessment, EPA summarized the lessons learned from the previous
assessment. Ail lessons are rigorousiy analyzed, including follow-up research to establish a solid record of
evidence. These lessons provide a basis for making changes in the technical approach or assessment design.
The most important lesson learned in 1999 was an apparent underreporting of needs due in part to limitations
of system planning documents. This lesson was addressed by the workgroup and resulted in major changes in
the assessment.
EPA has developed and enhanced the methods by which states can review the projects submitted and action
taken to ensure its quality during the 2003 Needs Assessment itself. The Agency improved this process for the
2003 Needs Assessment and facilitated states' ability to review information from their systems and to provide
comments on those data. The objective was greater transparency in the assessment process.
One area of weakness in the first two Reports to Congress was the lack of sufficient details on the
methodology. The details provided in those reports were similar to the information found in the Reports to
Congress that had been prepared by the Clean Watersheds Needs Surveys (CWNSs) for two decades. With
the issuance of EPA guidelines on information quality, however, it is appropriate for EPA to change the Agency's
''* American Water Works Association Water Industry Technical Action Fund. Dawn of the Replacement Era: Reinvesting in Drinking Water
Infrastructure (Denver, CO: May 2001).
ffl Congressional Budget Office: Future Investments in Drinking Water and Wastewater irtfrastnjciure (Washington, DC: November 2002). This
report is particularly useful because it provide a comparative analysis o? the methodologies of ali studies of infrastructure needs.
26 United States General Accounting Office: Key Aspects of EPA's Revolving Fund Program Need to be Strengthened, GAO-02-135 (January 2002).
-------
Page 56 2003 Drinking Water Infrastructure Needs Survey and Assessment
approach and provide greater detail. The ultimate goal is to provide enough information so that a reader who
had access to the Needs Assessment database could reproduce the results of the 2003 Needs Assessment.
To ensure that level of transparency, EPA has provided additional detail on the research methods in this
Report to Congress. EPA also has referenced, and will make available via the EPA Web site, the technical
approach document. That document provides detailed background information on every important research
design decision, as well as full details on the statistical methods used to draw a representative sample, and
the methods used to create sample weights for data analysis.
-------
Needs for Water Systems in the States2728
(community water systems and not-for-profit noncommunity water systems)
Exhibit D-"f—Total Need for Water Systems in the States by Project Type
Exhibit D-2—Current Need for Water Systems in the States by Project Type
Exhibit D-3—Total Need for Water Systems in the States by System Size
Exhibit D-4—Current Regulatory Need for Water Systems in the States
Exhibit D-5—Total Existing Regulatory Need for Water Systems in the States
Needs for American Indian and Alaska Native Village Water Systems
Exhibit D-6—Total Need for American Indian and Alaska Native Village Systems by EPA Region
Exhibit D-7-—Total Need by Project Type for American Indian and Alaska Native Village Water Systems
Exhibit D-8—Total Existing Regulatory Need for American Indian and Alaska Native Village Water Systems
Needs Attributable to Future Drinking Water Regulations23
Exhibit D-9—Total Proposed and Recently Promulgated Regulatory Need
Note: Numbers in Exhibit D-1 through D-9 may not total due to rounding.
27 Exhibits 0-1 through D-5 do not include naeds for American Indian or Alaska native village water systems. These needs are reported separately
In Exhibits D-6 through D-8.
«" Exhibits D-1, D-3, and D-5 through D-8 include costs associated with the recently promulgated Arsenic Rule but do not include costs associated
with other proposed or recently promulgated SOW A regulations.
" Exhibit D-9 includes costs associated with proposed or recently promulgated SDVVA regulations for waters systems in the states, American
Indian communities, and Alaska native villages.
-------
Page58
2003 Drinking Water Infrastructure Needs Survey and Assessment
state
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District o! Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
iridiana
iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
Nf:w Mexico
New York
North Carolina
North Dakota
Ohio
Oklahorria
Oregon
Pennsylvania
Puerto Rico
Rhode island
South Carolina
South Dakota
'fennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Subtotal
American Samoa
Guam
North Mariana is.
Virgin Isip.nds
Subtotal
Total
Transmisston
and Distribution
917.6
444.2
7,262.9
2,296.3
18,052.7
3,472.8
336.2
143.2
132.5
10,387.3
6,911.1
630.5
430.7
3,353.3
2,503.6
2,602.5
1 ,303.9
2,162.0
2,923.6
547.3
2,562.3
6,611.0
793/<1
3,362.3
914.5
4,625.5
469.0
737.3
£64.0
321.2
5,081.1
498.9
10,664.8
7,502.5
282.8
7,084.6
3,714.3
2,519.6
7,838.9
1.593.3
290.1
970.3
704.4
2,131.3
18,423.0
481.2
229.4
1,986.7
4,382.3
478.8
3,948.4
193.4
181,920.0
12.1
204.8
69.8
77.1
363.8
132,283.6
Ttaatmenf
415.2
63.2
1,114.2
727.S
4,830.1
99S.3
175.5
36.9
0.0
1,073.3
43.7
12S.9
2,463.0
741.4
373.4
233.8
313.0
576.7
i10.8
800.2
377.1
1,935.5
1,179.7
291.6
686.9
152.7
371.0
152.9
109.3
703.5
261.8
2,403.1
1,333.9
130.7
1,330.5
653.6
559.9
1 ,550.9
471.9
71.3
108.6
151,4
313.0
5,531.7
37.0
77.7
403.4
785.3
155.7
1,054.7
45.7
42,650,9
5.3
8.1
78.1
36.8
128.3
42,779.2
Storetja
302.9
126.4
463.5
346.3
3.005.5
452.2
96.3
39.3
15.5
S83.4
573.5
S4.5
111.8
1:170.3
477.2
328.4
256.4
254.8
317.0
i20.6
453.2
622.1
834.6
566.0
270.3
463.9
115.8
125.8
134.6
114.7
736.2
112.7
1,166.6
950.3
77.1
827.0
267.2
342.7
1.030.1
154.5
28.0
105.6
92.9
242.5
1,941.9
S2.G
60.3
324.0
1,077.3
159.8
575.0
42.7
24,223.6
11.1
27.7
3S.9
53.7
128.3
24,351.9
tt»uet
48.3
45.1
216.8
156.1
1,704.3
370.8
40.1
20.3
0.0
936.5
318.5
34.7
52.1
1,284.7
284.4
170.7
115.0
53.4
242.2
47.2
115.4
318.2
371 .5
274.6
160.1
171.7
48.2
107.8
53.5
47.5
322.7
46.2
449.1
478.6
60.5
371.0
162.3
230.6
457.5
45.6
9.3
50.9
37.8
63.6
1,033.5
34.4
24.1
133.5
382.6
48.S
337.7
15.1
12,604.6
2.7
32.2
9.2
11.6
55.7
12,660.5
OthSf
4.9
2.6
42.S
12.5
278.8
31.5
40
1.1
1.3
137.9
141.2
4.2
5.6
225.5
25.2
28.9
16.8
20.6
47.2
S.4
31.7
126.2
132.0
77.3
7.9
10.3
3.6
12.2
7.0
2.9
72.1
2.7
124.0
153.9
5.7
71.1
6.8
14.8
52.9
13.5
3.4
10.2
3.3
20.0
139.6
1.7
3.3
17.5
44.2
8.1
22.3
1.3
2,296.7
1.2
6.3
1.2
13.5
2,310.2
Totaf
631,5
5323.5
653.1
240.S
149,4
1S,4
63$r.S
264,38&.&
'Does not irwkide needs associated with proposed or rece:iUy promulgated regulaiiore, except for the receiitly
promulgated Arsenic Hula.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page 59
Stats
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Coliimbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
towa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
Nsw Mexico
Now York
North Carolina
Norm Dakota
Ohio
Oklahoma
Oregon
Peniisylvania
Puerto Rico
Rhode Island
South CaroBna
South DaKota
Tc-messee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Subtotal
American Samoa
Guam
North Mariana Is.
Virgin isiands
Subtotal
Total
Tratismisstor*
and Distribution .
304.7
257.3
6,346.6
1 ,539.0
11,819.4
1 ,672.8
242.1
112.1
69-5
9,495.0
6,331.4
266.4
318.1
4,653.1
1 .588.6
1 ,935.5
847.1
1 ,379.1
2,267.1
381.6
2,217.6
4,737.4
5,447.1
1 ,759 0
730.6
2,166.2
405.8
441.4
284.3
201.1
2,641.7
375.4
8.078.0
2.987.0
201.2
2,934.5
1 ,524,1
2,242.3
6,297.0
1 .003.3
234.6
573.1
220.7
1,014.6
9,974.2
382.1
157.3
997.0
3.198.7
336.2
1 ,708.3
96.8
118,414.9
11.4
204.6
56.2
70.7
3425
118,757.8
rceatift&nt
48.3
21.6
624.8
291.6
3,252.8
513.3
94.7
18.6
0.0
1,809.7
625.0
33.S
50.2
908.1
343.0
136.5
88.6
155.2
272.0
49.1
662.8
290.0
1 ,087.1
619.0
139.1
171.8
39.8
262.1
17.3
35.5
442.3
31.5
2,066.4
636.2
74.2
824.1
1287
499.6
1,186.9
294.6
25.0
39.8
60.5
111.3
2,981.6
31.7
34,8
174.2
2855
86.3
529.3
15.0
23,240.9
4.1
7.7
64.8
24.1
100.7
23,341.5
stows*
44.7
48.3
272.7
148.8
1,960.6
241.3
34.2
16.1
8.1
720.0
38 1 .9
56.1
44.4
513.1
225.7
158.9
151.5
154.6
182.1
67.6
353.8
365.3
341.9
216.4
130.9
142.8
50.1
47.0
54.2
41.5
408.5
34.2
687.0
309.7
37.8
337.2
86.0
660.4
574.5
97.9
10.7
47.8
38.9
112,3
656.8
64.5
24.8
130.5
579.1
66.5
215.0
17.3
12,372.3
10.9
27.S
12.2
33.7
84.4
12,457.3
Souths
13.0
19.8
118.8
77.5
1,134.0
315.5
20.5
10.8
0.0
599.0
2S6.4
29.8
30.2
236.0
140.1
70.4
71.3
25.1
147.7
25.4
71.4
118.7
154.2
101.8
96.0
68.1
25.7
49.2
18.C
24.1
222.1
20.1
248.0
186.8
39.7
177.6
51.3
142.0
345.1
31.5
4.6
29.3
20.7
27.4
3762
14.9
14.1
58.4
21 1 .3
28.4
198.7
8.7
6,538.0
2.6
32.1
7.5
10.5
52.7
6,590.7
OthSf
3.1
1.4
41.0
10.1
183.7
29.6
2.5
0.9
1.3
117.1
134.4
2.5
4.9
139.0
22.6
19.1
10.6
14.8
42.4
3.9
29.8
50.7
125.9
45.1
55
0.9
3.2
5.9
2.0
2.4
42.4
2.2
115.2
109.1
4.6
51.3
5.3
14.0
41.9
9.4
2.2
7.1
2.7
10.2
51.3
1.4
22
13.4
22.2
5.4
17.4
1.1
1,595.9
1.1
6.3
2.9
1.2
11.4
1,607.4
Total
413.8
348.4
7.404.B
2,068.8
™»JM5M.
2,772.4
394.2
1585
78.9
12,7*1,8
7,733.1
408,2
44?>?
$,4493
2,328.0
2,328.4
1,1735
1,728.8
2,911.4
527,5
3,335.4
3,562,1
7,168,2
2,741.3
1.1C2.2
2,555.8
SS4J
SOS.8
me
304>5
3,757.0
463.6
12,194.6
4,228,8
357.4
4,324,8
1,785-7
3,558.3
8,545.3
1,436.7
277.1
637.1
3414
\3SJ6A
14^040.1
484,?
233.5
1,374.6
4,238.7
522,7
2,668.6
138.8
162,162.5
30,0
2?8,5
143.6
140.1
8S2.2
162,754.?
-------
Page 60
2003 Drinking Water Infrastructure Needs Survey and Assessment
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Deis ware
District of Columbia
Florida
Georgia
Hawaii
baho
Ilinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
'stessachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Puerto Rico
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Subtotal
American Samoa
Guam
North Mariana is.
Virgin Islands
Subtotal
Total
Large CWSs
615,2
163.6
5,556.5
778.7
19,326.6
2.664.7
1S5.0
72.1
149.4
7,903.1
4,825.6
477.7
83.4
6,095.0
1,054.1
716.3
475.1
656.9
1.143.6
76.3
2,947.0
2,808.7
5,994.0
1.453.9
65.2
1,027.1
121.1
484.1
522.8
22.4
2.887.6
369.8
10.130.4
4,632.5
35.6
4,189.1
1 ,060.7
1 .409.0
5,733.7
1 .094.5
234.9
451.7
11.1
530.0
15,212.5
154.3
2.2
1 ,203.5
2,299.6
43.6
1 ,895.3
16.2
122,555.0
0.0
221.6
75,0
0.0
296.7
122,851.7
Medium CWSs
782.4
254.4
2,988.3
2.137.3
5,823.3
2,022.8
121.5
7.4
0.0
6,011.5
3,411.1
213.0
169.8
5,835.5
2.157.4
1.953.9
716.1
1,3788
2,175.8
429.2
640.4
5.459.5
3.840.8
3.018.2
664.5
3.88S.O
245.2
472.4
171.4
121.5
3,486.3
159.8
2,517.6
4.997.2
343.8
4,186.2
2,857.8
2.122.5
3.495.3
707.!
116.7
498.6
722.2
1.820.4
9,896.8
300.5
107.9
872.3
2,764.2
209.1
2,834.7
122.3
102,812.6
13.2
50.2
96.9
446
204.8
103,0174
Small CWSs'
288.1
1870
4671
566.6
2,016.7
627.0
328.5
157.9
0.0
1,018.2
766.6
115.4
408.9
1.450.9
662.7
792.G
729.7
272.2
757.5
237.7
292.7
248.7
1.012.2
743.3
906.0
1.005.5
373.0
375.5
172.8
369.9
370.9
35S-8
2,003.0
1.035.8
209.8
1,054.0
854.0
674.2
1 1.520.7
471.0
360
280.2
243.5
396.0
2,964.2
231.4
274.6
709.1
1,404.8
5663
776.6
144.7
33,965.1
18.7
72
25.1
135.4
186.4
34.171.5
KPNCWS*'
32
51.0
155
6.1
34.6
1.1
22.9
2.7
0.0
106.6
11.5
0.8
31.5
92.0
147.3
15.4
2.9
0.9
12.5
28.8
82.0
27.5
394.4
224.1
8.0
32.7
42.3
13.4
11.9
51.7
170.0
12.8
105.4
308.8
4.5
235.7
18.6
46.4
235.3
!.0
13.5
13.5
4.3
24.1
39.8
10.8
0.1
76.6
36.9
40.2
403.8
10.2
3,3975
0.0
0.0
0.0
0.0
oa
3,397.5
Racerttty
Promulgated
Arsenic Bute ''
0.0
15.4
92.5
0.0
118.2
7.9
15.0
0.8
0.0
1.2
0.8
5.5
33.3
23.4
0.4
25.7
7.1
0.0
17.4
S.9
1.2
10.3
63.7
21.0
0.8
4.0
6.7
3.7
33.3
30.1
0.8^""
21.0
56.2
5.9
13.1
19.0
13.1
15.4
5.1
5.1
1.6
1.6
8.7
0.0
56.2
9.9
9.9
3.6
106.1
0.8
27.7
4.8
945.B
0.4
0.0
0.8
0.4
1.6
947.4
total
tjsaaS
601,5
s,ims
3*538.7
27,871.5
5,323.8
653,1
240JS
149.A
15,040.7
&G1 ?.§
812,5
727.0
13^96,8
4,031.&
3333.3
1,3308
iJjSC&S
4,1053
831.8
3,863.2
8*554,7
11, -311.1
5?4S05
tJ544,S
5,958.2
789,3
1,354.0
912,1
S95.6
•"""•"igSTSIf
3222
14,812.5
10,980,2
606.8
»,884,,1
4,804.2
4,2$74S
10.99(5.3
2,2?&8
40£6
1,245.6
989.8
2,778.4
2&,169.6
706,9
334.8
2J?65.0
S,S7t,7
801,9
5,938,1
298.2
263,69^1
32Ji
279x0
i«r,a
100*4
689^
284.385J5
: 1939 Drinking Water Infrastructure Needs Survey and Assessment findings were used to calculate the need for systems serving 3,300 and
fewer people (smalls) and not-for-profit noncommum'ty water systems (NPNCWSs). 1999 Needs Assessment results were adjusted to January
2003 dollars.
? Data did not alloy/ allocation of costs by system size for the recently promulgated Arsenic Rule.
-------
>003 Drinking Water Infrastructure Needs Survey and Assessment
Page 61
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
Nortii Carolina
Nortii Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Puerto Rico
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Subtotal
American Samoa
Guam
Nortii Mariana Is.
Virgin islands
Subtotal
Total
JS
15.9
162
353.7
210.9
3,0232
436.5
72.5
7.2
0.0
64.9
483.0
14.5
367
423.6
179.6
31.6
39.8
145.5
137.3
32.9
504.0
253.3
885.7
54.6
20.5
972
22.8
226.8
7.1
14.8
295.1
102
1,937.0
505.3
38.3
7122
91.4
5212
8482
3382
16*
25.'!
27.7
76.9
2,421.0
15.5
27.9
140.3
975
66.1
ilS.O
9.5
16,200.0
0.4
7.5
40.7
402
885
16,288.9
fee
0.0
6.8
0.4
02
261.2
0.9
0.2
0.0
0.0
4.4
0.1
0.0
0.3
S5.0
1,5
21.3
0.0
22.3
12
33.3
1.4
64.5
547.4
4.0
02
0.8
0.4
0.1
0.1
05
3.5
0.1
4.6
105.0
3.1
44.0
0.2
0.9
3.4
0.0
0.1
2.8
0.0
02
15.7
0.1
0.1
15.6
42.3
0.4
11.0
O.i
1,282.4
0.0
0.0
0.0
0.0
0.0
1,282.4
Nttrsts/
Nitrite
0.0
0.1
02
0.1
72.7
0.2
0.3
0.1
0.0
0.5
0.5
0.0
0.3
6.1
02
1735
11.7
0.0
02
0.1
0.1
0.1
0.4
1.8
0.!
0.4
02
51 .0
0.1
0.3
0.1
0.2
5.7
2.4
0.1
39.1
0.2
0.3
93
0.1
0.0
O.i
0.1
0.1
0.9
0.1
0.1
0.4
0.9
0.1
21.9
O.i
404.1
0.0
0.0
0.0
0.0
0.0
404.1
Copper Bute
3.8
0.6
9.1
3.9
14.5
6.7
1.0
0.4
1.4
146.7
53.3
0.7
1.3
161.0
30.6
8.1
16.6
40.8
3.3
2.4
275
88.8
255.0
40.4
4.3
13.6
1.4
3.7
0.7
15
912
1.3
81.3
17.4
0.8
158.0
13.6
15.5
65.1
4.5
32.9
3.0
0.6
3.4
26.7
0.8
2.3
4.5
15.1
5.6
124.7
1.1
1,632.6
0.0
0.0
0.0
0.7
0.7
1,633.3
rm«*
0.0
0.0
1.0
0.0
16.4
0.0
0.0
0.0
0.0
40.0
2.5
13.3
0.0
0.0
0.0
0.0
0.0
2.7
22
0.4
0.5
0.0
9.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
102
0.0
0.0
20.9
0.0
0.0
0.3
0.0
0.0
0.0
0.0
1.5
1.4
0.0
0.0
0.4
0.0
0.0
0.0
0.0
123.S
0.0
0.0
0.0
0.0
0.0
1235
Other
4fi
22
115
3.0
233.5
0.0
0.4
0.4
0.0
11.1
15
0.4
05
1723
23
27.1
2.0
OS
2.6
2.1
2.7
46.1
24,i
393
3.7
2.7
05
16.4
0.4
05
27.1
0.7
15.3
25.9
OS
5.5
1.9
25
1S5
05
O.i
OJ5
05
\S
342
0.6
03
20.6
23
1.8
299.7
02
1,075.1
0.0
00
0.0
OX!
0.1
1,0752
JSSt
244
29.8
979,9
219.2
$,621.4
445,1
74.4
8.2
1,4
267*
840,9
28.9
33.1
836>1
2145
261.9
70,1
212.$
146.3
71 .&
536.4
452.3
1,722^
140,1
28.9
114^8
254
298,0
8.5
17.9
4ir,i
12.6
2,104.1
655.$
49,1
979.7
tOTjt
840,4
941.8
34JU
49.9
3S.1
29.9
84.0
2,500.0
17.1
30^
tai.a
1582
74.0
&73,&
11,0
20,7*7.8
0.4
7&
40.7
405
8S.6
25,807*4
-------
Page62
2003 Drinking Water Infrastructure Needs Survey and Assessment
Stats
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Rorida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New YorK
North Carolina
North Dakota
Ohio
Qklarioma
Oregon
PeaTsyivania
Puerto Rico
Rhode bland
South Carolina
South Dakota
Tftnnessss
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Subtotal
American Samoa
Guam
North Mariana k.
Virgin islands
Subtotal
Total
SWTR/
IE&WTR
3)3.7
19.3
673.7
472.6
3,911.1
865.0
87.1
7.2
0.0
101.2
723.2
15.5
41.7
1,357.2
221.4
68.4
60-2
260.6
285.9
59.8
524.7
721.8
1;222.3
133.7
34.3
338.9
91.7
240.3
93.0
18.6
346.3
186.8
2,057.8
1,617.9
107.3
966.2
537.4
597.8
1,037.6
500.9
41.7
67.7
53.2
224.8
4,501.7
48.9
42.9
283.0
221.1
103.0
308.5
23.0
27,250.0
0.5
7.6
40.7
48.0
96.3
27,346.8
TCB
0.1
7.2
1.8
0.5
399.0
1.6
i.O
0.3
0.0
18.6
1.6
0.1
!.2
285.0
2.0
23.8
0.7
32.5
1.9
44.7
1.9
68.0
992.2
4.9
0.6
2.1
1.2
0.7
0.4
1.3
3.9
0.8
8.5
107.4
5.8
465.1
3.4
1.9
5.4
0.2
0.9
6.1
0.4
0.6
18.7
0.4
0.5
23.7
57.1
0.8
22.4
0.3
2,631.5
0.0
0.0
0.0
0.0
0.0
2,631.5
Nitrate?
HiWte
0.0
0.1
0.2
O.I
119.3
0.2
0.3
0.1
0.0
0.5
0.5
0.0
0.3
7.8
0.2
173.5
24.8
0.0
0.2
0.1
0.5
O.I
0.4
1.8
0.1
0.4
0.2
72.0
0.1
0.3
0.1
0.2
13.4
2.4
0.1
3S.1
0.2
0.3
9.3
0.1
0.0
0.1
0.1
0.1
O.S
0.1
0.1
0.4
0.9
0.1
28.7
0.1
501.4
0.0
0.0
0.0
0.0
0.0
501.4
$&4
210,*
2,267.0
1,691.$
130.9
1,752,1
569.9
S34,3
1<1<«,t
5113
77.1
80,6
77.1
232.3
4,654,9
6Q.3
56.2
348.7
4Q4.S
121*
840.1
315
34^624
0.9
7.6
415
49.1
33.1
34,961,7
-------
2003 Drinkinq Water infrastructure Needs Surrey and Assessment
Page 63
Exhibit D-6: Total Need for American fndiMlfKi Alaska Native VjHage
(20-year need in millions of January 2003 dollars)
EPA Region ;;
Region 1
Region 2
Region 3a
Region 4
Region 5
RegionG
Region 7
Region 8
Region 93
Region 10"
Alaska Native Systems
American Indian and Alaska Native Need to Comply
with the Recently Promulgated Arsenic Rule
Totai
Total Meed1
4.3
6.6
0.0
19.5
172.5
166.7
15.7
146.3
602.0
129.8
1,170.5
14.7
2,448,5
1 1S99 Drinking Water Infrastructure Needs Survey and Assessment findings converted to January
2003 dollars, includes costs associated with the recently promulgated Arsenic Rule.
2 There are no American Indian wator systems in EPA Region 3.
3 Navajc water systems are located in EPA Regions 6. 3. and 9, but for purposes of this report, all
Navajo needs are shown in EPA Region 9.
" Needs for Alaska natjva village water systems are not included in the EPA Region 10 total.
If-
-------
Page 64
2003 Drinking Water Infrastructure Needs Survey and Assessment
Project Typfc
Transmission and Distribution
Treatment2
Storage
Source
Other
Total
Cwrrer&M«e«t$
1,287.0
404,9
437,4
109.1
13.6
2,252.0
Future Neette
60.3
57.3
52.9
26.0
0.0
196.5
Total Nee?!1
1,347.3
462.2
490.3
135,1
13.6
2,448.5
' 1399 Drinking Water Infrastructure Needs Survey and Assessment findings convened to January 2003 dollars, deludes costs
associated with the recently promulgated Arsenic Rule.
' Treatment category includes needs for the recently promulgated Arsenic Rule.
-------
2003 Drinking Water Infrastructure Needs Survey and Assessment
Page 65
Regula&m
Regulations for Contaminants
with Acute Health Effects
Regulations for Contaminants
with Chronic Health Effects
Total
Current Need*
175.3
0.2
175.5
Future Nee«te
5.1
14.7
19.8
Total Need
180.4
14.9
195,3
1999 Drinking Water infrastructure Needs Survey and Assessment findings converted to January 2003 dollars. Includes costs
associated with she recently promulgated Arsenic Rule.
-------
Page 66
2003 Drinking Water Infrastructure Needs Survey and Assessment
Regulation
Stage 1 Disinfectants/Disinfection Byproducts Rule
Long-Term 1 Enhanced Surface Water Treatment
Rule
Filter Backwash Recycling Rule
Ground Water Rule
Stage 2 Disinfectants/Disinfection Byproducts Rule
Long-Term 2 Enhanced Surface Water Treatment
Rule
Radon Rule
Radionuclides Rule1
Total2
RaftgeofCOSfS
Low
Estimate
936.8
1 :290.9
144.8
167.2
High
Estimate
1,150.2
1 ,685.7
5,794.2
883.3
Estimate included
in the 2003 Needs
A««^sswem
2,582.7
193.1
157.8
1,150.2
491.7
1,685.7
2,782.8
883.3
9,927.4
1 The high and low estimates represent the two approaches presented in the November 2000 "Economic Analysis of the
Radionuclides National Primary Drinking Water Regulations." The total capital costs were determined by averaging the total capital
costs for compliance with the maximum corrtaminarrt level (MCL.) set at 20 mlcrograms per liter (ug/L) and 40 uj/L. for each of the
two approaches. The final rule set the MCL at 30 ug/L
2 In calculating tr» $9.9 billion need associated with proposed and recently promulgated regulations, fc'PA used the lead option,
unless one was not available in which case EPA used the more conservative estimate. These estimates Include only the capital
costs (i.e., excludes operation and maintenance costs). Costs for the recently promulgated Arsenic Rule are not included in this
table.
-------
Needs for Water Systems in the States23 (community water systems)
Exhibit E-1-—Total Need for Systems Serving 10,000 and Fewer People
Note: Numbers in Exhibit E-1 may not total due to rounding.
20 Exhibit E-1 does not include costs associated with proposed or recently promulgated SDW.A regulations, including the recently promulgated
Arsenic Rule.
-------
Page68
2003 Drinking Water Infrastructure Needs Survey and Assessment
^^^^g^onlg
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Comecilc'Jt
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kfintuci4 615.0 "KM*
1J357.5 9,011.9 13-1%
1,825.0 3,632.6 45.0%
3.S38.S 27,668.6 14,3%
TM1.5 5,314.5 S7.i%
356.8 615.1 58,0%
157.9 ?37.3 66.5%
0.0 149.4 0-0%
£575.2 14,932.9 17,2%
2,318.1 9,0053 24.6%.
1614 8061 $£S%
491,4 662.? 74,2%.
3,308,5 13,381.4 24,7%
1,800,« 3,8842 41,2%
W&1.S 3,462.8 36.4%
1f»93 1,9209 52,0%
958,« ?,807 9 33,S%
1,715,1 4,076.9 4£1%
466.3 793.1 5&8%.
373,3 3,880.0 9,6%
1,888,3 8,516.9 19,8%
SXnS 10,847.0 20,9%
1,,881.1 5,2154 441%.
1,378,1 1,6357 «4,3%
3jOa9.0 5,921.6 51,2%
440.8 740.3 58.5%.
835,7 1,332.0 47,7%
316-g 867.0 ^$.7%
383,2 513.8 74,6%
1^95.0 6,744.8 19,2%
440.0 8384 49.3%
«*76.2 14,6509 1&6%
^8184 10,6654 26.4%.
353.1 589.3 S3.S%
5334.8 9,429.4 26.9%
S^>39.3 4,772.5 42,7%
1^156,5 4,205 7 32.3%
X,808<1 10,749.8 3&1%
79S.2 2,272.7 3&0%
45-1 387.6 11.6%
411,8 1,2305 33,3%
4^7.5 976.9 47,9%
1,457.0 2,746.4 5&1%
6,335.6 28,073.5 22.5%
$104 6862 ?•«%
346,8 334.7 80,1%
1,068.5 2,784.9 3&4%
2.S88.S 6,468.7 40.0%
695.8 820.9 &4.B%
1,7134 5,5066 ' 31,1%
1885 2832 70.3%
71,208,8 259,352.7 27.8%
18,7 31 9 «50,T%
26.4 2V9.0 9.5%
•MZJ) 197.0 «1.9%
1385 1800 75.7%
303,4 687.9 44.1%
71,512,.0 260,040.6 27.5%
19S9 Drinking Water Infrastructure- Needs Survey and Assessment findings were used to cafcuiats the need for systems serving 3,300 and
fewer people. 1993 fteeda Assessment results were adjusted to January 2C03 dollars.
-------
Acute health effects: health effects resulting from exposure to a contaminant that causes severe symptoms to
occur quickly—often within a matter of hours or days. Examples include gastrointestinal illness and "blue baby
syndrome."
Capital improvement plan (CIP): a document produced by a local government, utility, or water system that
thoroughly outlines, for a specified period of time, all needed capital projects, the reason for each project, and
their costs.
Chronic health effects: health effects resulting from long-term exposure to low concentrations of certain
contaminants. Cancer is one such health effect.
Coliform bacteria: a group of bacteria whose presence in a water sample indicates the water may contain
disease-causing organisms.
Community water system: a public water system that serves at least 15 connections used by year-round
residents or that regularly serves at least 25 residents year-round. Examples include cities, towns, and
communities such as retirement homes.
Current infrastructure needs: new facilities or deficiencies in existing facilities identified by the state or system
for which water systems would begin construction as soon as possible to avoid a threat to public health.
Engineer's report: a document produced by a professional engineer that outlines the need and cost for a
specific infrastructure project.
Existing regulations: drinking water regulations promulgated under the authority of the Safe Drinking Water Act
by EPA; existing regulations can be found in the Code of Federal Regulations (CFR) at 40 CFR 141.
Finished water: water that is considered safe and suitable for delivery to customers.
Future infrastructure needs: infrastructure deficiencies that a system expects to address in the next. 20 years
because of predictable deterioration of facilities. Future infrastructure needs do not include current
infrastructure needs. Examples are storage facility and treatment plant replacement where the facility currently
performs adequately but will reach the end of its useful life in the next 20 years. Needs solely to accommodate
future growth are not included in the Needs Assessment.
Ground water: any water obtained from a source beneath the surface of the ground, which has not been
classified as ground water under the direct influence of surface water.
Growth: needs planned solely to accommodate projected future growth are not included in the 2003 Needs
Assessment. Eligible projects, however, can be designed for growth expected during the design life of the
project. For example, the 2003 Needs Assessment would allow a treatment plant identified as a current need
and expected to treat water for 20 years. Such a plant could be designed for the population anticipated to be
served at the end of the 20-year period.
-------
Page 70 2003 Drinking Water Infrastructure Needs Survey and Assessment
Infrastructure needs: the capital costs associated with ensuring the continued protection of public health
through rehabilitating or building facilities needed for continued provision of safe drinking water. Categories of
need include source development and rehabilitation, treatment, storage, and transmission and distribution.
Operation and maintenance needs are not considered infrastructure needs and are not included in this
document.
Large water system: in this document, this phrase refers to a community water system serving more than
50,000 people.
Medium water system: in this document, this phrase refers to a community water system serving from 3,301 to
50,000 people.
Microbiological contamination: the occurrence of protozoan, bacteriological, or viral contaminants in a water
supply.
Noncommunity water system: a public water system that is not a community water system and that serves a
nonresidential population of at least 25 individuals or 15 service connections daily for at least 60 days of the
year. Examples of not-for-profit noncommunity water systems include schools and churches.
Potable water: water that is fit to drink.
Public water system: a system that provides water to the public for human consumption through pipes, other
constructed conveyances, if such system has at least 15 service connections or regularly serves an average
of at least 25 individuals daily at. least 60 days out of the year.
Regulatory need: a capital expenditure required for compliance with regulations.
Safe Drinking Water Act (SDWA): a law passed by Congress in 1974 and amended in 1986 and 1996 to
ensure that public water systems provide safe drinking water to consumers. (42 U.S.C.A. §300f to 300J-26)
Small water system: in this document, this phrase refers to a community water system serving 3,300 people or
fewer.
Source rehabilitation and development: a category of need that includes the costs involved in developing or
improving sources of water for public water systems.
State: in this document, this term refers to ail 50 States of the United States, Puerto Rico, the District of
Columbia, American Samoa, Guam, the Northern Mariana Islands, and the Virgin islands.
Storage: a category of need that addresses finished water storage needs faced by public water systems.
Supervisory Control and Data Acquisition (SCADA): an advanced control system that collects ail system
information for an operator and allows him/her, through user-friendly interfaces, to view all aspects of the
system from one place.
-------
2003 Drinking Water infrastructure Needs Survey and Assessment Page 71
Surface water: all water that is open to the atmosphere and subject to surface run-off, including streams,
rivers, and lakes.
Transmission and distribution: a category of need that includes replacement or rehabilitation of transmission or
distribution lines that carry drinking water from the source to the treatment plant or from the treatment plant to
the consumer.
Treatment: a category of need that includes conditioning water or removing microbiological and chemical
contaminants. Filtration of surface water sources, pH adjustment, softening, and disinfection are examples of
treatment.
Watering point: a central source from which people without piped water can draw drinking water for transport
to their homes.
------- |