United States
Environmental Protection
Office of Water (4601M)
Office of Ground Water and Drinking Water
Total Coliform Rule Issue Paper
Distribution System Inventory, Integrity and Water
January 2007

U.S. Environmental Protection Agency
Office of Ground Water and Drinking Water
Standards and Risk Management Division
1200 Pennsylvania Ave., NW
Washington DC 20004
American Water Works Association
Background and Disclaimer
The USEPA is revising the Total Coliform Rule (TCR) and is considering new possible
distribution system requirements as part of these revisions. As part of this process, the
USEPA is publishing a series of issue papers to present available information on topics
relevant to possible TCR revisions. This paper was developed as part of that effort.
The objectives of the issue papers are to review the available data, information and
research regarding the potential public health risks associated with the distribution
system issues, and where relevant identify areas in which additional research may be
warranted. The issue papers will serve as background material for EPA, expert and
stakeholder discussions. The papers only present available information and do not
represent Agency policy. Some of the papers were prepared by parties outside of EPA;
EPA does not endorse those papers, but is providing them for information and review.
Additional Information
The paper is available at the TCR web site at:
Questions or comments regarding this paper may be directed to TCR@epa.gov.

Table of Contents
Table of Contents	iii
Abstract	v
1.0 Introduction	1
2.0 Buried Infrastructure Challenges Facing the Water Industry	2
3.0 AWWA and AwwaRF Studies and Surveys	3
3.1	AWWA and AwwaRF Studies	3
3.2	AWWA's Water://Stats Database	7
4.0 EPA Surveys and Other Federal Government Analyses	13
4.1	EPA Community Water System Survey (EPA, 2002a)	14
4.2	EPA Drinking Water Infrastructure Needs Survey (EPA, 2005)	15
4.3	EPA White Papers on Distribution Systems	16
5.0 Condition Assessment of Water Distribution Pipes	17
6.0 Conclusions	20
7.0 References	21

Distribution System Infrastructure Inventory
and Integrity
This white paper reports on the availability of data about distribution system infrastructure,
and the ability to answer selected questions using these data. The paper does not address water
quality, policy needs, or potential research projects. Water distribution systems comprise
complex networks of infrastructure components. Currently, available data provide more
information on distribution systems than existed a decade ago. At the national level, data with
which to describe distribution systems is good, but the information has not in all cases been
verified. The data reported are mainly from recent AwwaRF reports, AWWA's Water Industry
Data Base and Water://Stats surveys, and EPA's Community Water System Survey (CWSS)
and Needs Survey. Data on the extent of water mains, finished water storage, hydrants, some
types of valves, and customer service lines are generally good. Very little data are available on
other components of distribution systems or on premise plumbing. The practice of condition
assessment is intended to support asset management programs rather than general conclusions
about the overall condition of the nation's water distribution system infrastructure.
Implementation of asset management systems that require condition assessment varies from
utility to utility - some utilities have complex data systems, while many utilities rely on paper
files, maps, and the experience of the utility staff.

Distribution System Infrastructure Inventory
and Integrity
1.0 Introduction
The purposes of this white paper are to report on the availability of data about distribution
system infrastructure and to summarize answers to selected questions that can be supported by
the data. In simple terms the purpose of the water distribution system infrastructure is to
supply water to all customers at sufficient pressure and volume to provide for their needs as
well as for fire suppression (water quantity aspects), while also protecting the quality of the
water as prescribed by various standards (water quality aspect). It is important that distribution
systems deliver water reliably and protect the quality of the water that is delivered (National
Research Council, 2006). These water distribution systems involve complex networks of
infrastructure components consisting of pipes, joints, valves, and other appurtenances. In
addition, water travels through service lines and premise plumbing systems before arriving at
the customer's tap.
Throughout the paper, infrastructure issues are discussed using terminology that is not in all
cases standardized. When terms are introduced, working definitions are presented, and
acronyms are explained when they appear.
The term "inventory" refers to the identification, location, and description of distribution
system components such as pipe segments, valves, and other parts. The term "condition" refers
to appraisal of the current physical integrity of a component compared to its original designed
condition. In this instance "physical integrity" of a component is a measure or estimate of flaws,
defects, or decay that could reduce its service life (time from installation to replacement), as
compared to original physical condition.
While distribution systems may affect drinking water quality and while water quality may
affect health, this paper does not address these possible effects. It also does not make
recommendations about policies or needed research. The paper is principally focused on
reporting about sources and extent of data that is available and how it bears on the following
•	How much and what types of pipe and fittings are in service today?
•	How much and what types of pipe are being installed and renewed today?
•	How many and what types of storage tanks exist?
•	How many and what types of fire hydrants and valves are in service today?
•	How is the condition of distribution systems assessed? What is the knowledge
base about the condition of distribution systems?
•	What other appurtenances can be assessed?

The knowledge base about distribution system infrastructure has improved greatly since the
1986 Amendments to the Safe Drinking Water Act (SDWA). Prior to that date, neither
individual utility studies nor national surveys were very extensive in their reporting of
infrastructure data. The emergence of electronic database and Geographic Information System
(GIS) technology, along with recent waves of activity in vulnerability assessment and "asset
management" have led to more interest in conducting infrastructure inventories. An inventory
of a distribution system comprises identification, location, and description of components such
as pipe segments, valves, and other appurtenances.
Although extensive distribution system infrastructure data were not published prior to about
1986, utility surveys by the American Water Works Association (A WW A) actually began much
earlier. Additionally, prior to about 1960 the literature contains a number of short papers about
problems and remedies with cast iron water mains. During the 1970s more information and
basic data were collected, but the data available increased more rapidly after 1986. The data
available has been collected using different means. The primary categories of data are AWWA
and EPA surveys and AwwaRF case studies of one or more utilities. Other studies have been
published, but they rely on data from these primary sources. A list of references is available in
Grigg (2004), which provides a synthesis of the information available on water distribution
system infrastructure.
2.0 Buried Infrastructure Challenges Facing the Water Industry
The buried infrastructure challenges facing the water industry were summarized in one of a
series of papers that were prepared for EPA to provide information about potential distribution
system requirements being evaluated under the 6-year review of the Total Coliform Rule (TCR)
(American Water Works Service Co., Inc., May 2002).
The paper outlines how most distribution pipes installed from the late 1800s to the late 1960s in
the United States were of cast iron. It describes how casting technologies changed from pit
casting to centrifugal casting, which made a thinner pipe wall and lighter pipe possible. The
paper also outlines how cement mortar pipe lining improved resistance to internal corrosion,
how jointing changed from lead to a plasticized sulfur cement compound called "leadite," and
how "leadite" joints failed more often than the older lead joints. Further improvements in
jointing occurred with the introduction of rubber gaskets. The next major advancement was the
development of ductile iron pipe, which has a different internal metallic structure due to the
metal's graphite content. Then, polyvinyl chloride (PVC) and high-density polyethylene
(HDPE) pipe technologies were developed, that are not subject to the corrosion processes that
affect iron pipe. The paper does not discuss reinforced concrete pipe or prestressed concrete
cylinder pipe (PCCP). Some PCCP has experienced catastrophic failures due to production
processes that led to failure of reinforcing bands. The paper does not discuss the use of asbestos
cement (AC) pipe, which was significantly used in the 1950s and 1960s but was discontinued
due to concerns over asbestos. However, the paper does contain a diagram, which is
reproduced here as Figure 1, which shows the eras when asbestos cement and other types of
pipe were predominant.

Figure 1. Timeline of Pipe Technology in the U.S. in the 20th Century
(American Waterworks Service Co., Inc., 2002)
'•1 \ • • in \i
!; ,
—	\ 1 i )^1U\ I
y\ 11 niok ) [ |


i i i
—1— r-~







1 .\S%M
U '(


¦ T*

t M T'OP
i j-vJ

...I pp

4 jv hvn
S i'..'!
f. fii.Ml!
X mi:



tV,. ;n it


1 IB


{ ,l>"
< .m ii


(•" |K "U
i «'in.




' •
Pi [ uiv-v


M lit 11 (1

1 1
J tone. fltCPS

nmm* com, 
Table 1. List of AWWA and AwwaRF studies on distribution system infrastructure
Study author
O'Day et al. (1986)
Surveyed six utilities (New York, Denver, Philadelphia, Louisville, East
Bay, and Kenosha). Report contains often-cited information about failure
mechanics, condition assessment, management methods and other
Deb et al. (1990)
Studied seven utilities and surveyed 35 utilities about renewal practices.
Most appurtenances were replaced rather than rehabilitated. Valves and
hydrants were both replaced and rehabilitated. Shows that renewal is
practiced, especially in large utilities, but other than cleaning and lining,
new technologies were not in widespread use.
Kirmeyer et al.
Surveyed twenty utilities in 1992. Also summarized AwwaRF's three
expert workshops on distribution systems, held in 1990, 1991, and 1992.
This research needs section of this report is often cited for its presentation
of distribution system statistics.
Stratus Consulting
AWWA commissioned an independent assessment of distribution system
needs that led to a 20-year estimate of $325 billion.
Deb et al. (1998)
This report about a prioritization model also reviewed distribution system
Deb et al. (2000)
Surveyed 37 utilities about O&M, including European utilities. Leak
detection was the least common maintenance activity, among tasks such
as hydrant flushing and testing. 81% had corrosion control procedures
and 60% had procedures for main breaks. 70% had maintenance history
databases. While statistics show that utilities engage in the activities, they
do not reveal the extent to which they implement them.
Cromwell et al.
Surveyed 20 utilities and reported needs of more than $250 billion over
the next 30 years to replace pipes and infrastructure. This does not
include more than $12 billion per year that utilities spend on infrastructure
repairs or Safe Drinking Water Act compliance.
Cromwell et al.
(2001 b)
Benchmarking and process comparisons of asset management practices
between 15 North American and 2 Australian utilities.
Deb et al. (2002)
Found from WATERASTATS that in 1995 there were 23-breaks/100
miles/year. Break rates in Europe are higher, on the order of 50-
breaks/100 miles/year. The data show scatter in break rates, especially
for small utilities.
Grigg (2004)
Collected and synthesized data from approximately 50 utilities in three
workshops, site visits, and surveys
While these are examples of leading practices, the effectiveness and extent to which they are
used varies widely. Two utilities may report in a survey that they use a computer-based
maintenance management system, but one may have a system that is highly integrated with
their asset management strategy and yielding significant benefits, while the other may have just
purchased a stand-alone work order management system off the shelf, which limits the benefits
obtainable in asset management unless it is used in an integrated fashion. So far, the extent to
which utilities are benefiting from these technologies is mostly contained in case studies, such
as those in the reports given in Table 1.

While more recent survey data is available, the Kirmeyer et al. (1994) report for AwwaRF offers
a comprehensive view and useful statistics because a main activity of the study was to process
and analyze the available AWWA and utility data about distribution system infrastructure,
whereas data from more recent AWWA surveys requires further analysis to determine trends
and conclusions. The study's authors conducted their own surveys, used the AWWA Water
Industry Data Base (WIDB) survey data, and visited utilities. Tables 2, 3, and 4 are based on the
Kirmeyer et al. (1994) study, and present in a capsule form key data about inventories and
management practices. The more recent surveys confirm the basic data in these tables.
The data in Table 2 was compiled using a number of approaches. First, data was compiled from
results of AwwaRF projects available at the time, including three expert workshops on
distribution systems. Twenty utilities were studied in depth. The expertise of a project advisory
committee that included experts who had completed past projects was tapped. And, a special
workshop during AWWA's 1992 Distribution Systems Symposium was conducted to collect
data and opinions from additional utilities. The data in Table 2 is described by Kirmeyer et al.
(1994) as based on the above project information, AWWA's Water Industry Data Base
(predecessor to AWWA's Water://Stats database), and reports of the Water Industry Technical
Action Fund (WITAF).
Table 2. Statistics of U.S. distribution systems (Kirmeyer et al., 1994)
Distribution system elements
Project findings
Estimated length of distribution piping
880,000 miles
Estimated replacement value of piping
$348 billion
Condition of piping more than 30 years old
28% excellent, 43% good, 26% fair, 3% poor (these
composite figures are based on surveys of 20 utilities and
their reports of condition for water quality, structural
performance, and hydraulic performance)
Estimated number of pipe breaks
237,600 breaks/year (27-breaks/100 mi/yr)
(note: this is a different data set than Deb et al., 2002,
which reported 23-breaks/100 mi/yr)
Primary types of existing piping
48% cast iron, 19.2% ductile iron, 15.1% AC*
Estimated new piping
13,200 miles/year (DIP*47.7%. PVC* 38.7%), CPP* 12.5%)
(cost $2.8 billion per year)
Estimated pipe replacement
4,400 miles per year
Value of replacement
$1,742 billion per year
Lead service lines
2.3 to 5.1 million
Cost to replace lead service lines
$10-14 billion
Fire hydrants
5.85 million
Percent of O&M* budgets to T&D*
Total O&M budget to T&D
$4.5 billion per year
Inadvertent system losses (defined as
losses other than "authorized" losses, e.g.,
leaks, inaccurate meters, etc.)

Cost of water losses
$2.8 billion per year
* Key to table: AC = asbestos-cement; DIP = ductile iron pipe; PVC = polyvinyl chloride; CPP =
concrete pressure pipe; O&M = operations and maintenance; T&D = transmission and distribution.

As shown above, Kirmeyer et al. (1994) estimated that some 2.3 to 5.1 million lead service lines
are still in service. Replacement of the utility portion was estimated to cost $3.4 to $5.1 billion.
Complete replacement of the utility and residential portions of existing lead service lines was
estimated to cost $10 to $14.1 billion.
Table 3 was prepared from a survey question posed by Kirmeyer et al. (1994) that asked utilities
to list the five most common causes of main breaks. The data did not distinguish between types
of materials. Table 4 is based on survey responses to the question: "What are your criteria for
deciding whether a particular section of pipe is to be replaced?" Note that the top criterion for
pipe replacement is "number of leaks or breaks." This might be construed to imply a reactive
approach, but some would argue that leaks and breaks are, in fact, the most cost-effective and
integrated measures of pipe condition that are available to support a predictive approach to
replacement (Cromwell, 2001b; Hughes, 2002).
Table 3. Causes of main breaks (Kirmeyer et al., 1994)
Causes of main breaks
Percent of utilities reporting
Weak joints
Earth movement or settling
Internal corrosion
Corrosive soils
Construction or utility digging
Stray DC current
Seasonal changes in water temperature
Heavy traffic load
Tidal influences
Changes in system pressure
Water hammer
Air entrapment
Table 4. Criteria for pipe replacement (Kirmeyer et al., 1994)
Criteria for pipe replacement
Percent reporting
Number of leaks or breaks
Age of pipe
Low flow
Condition or type of material
Size changes required
Water quality
Soil condition
Street construction work
Elimination of dead ends
Amount of damage by leaks/breaks

A survey by CH2M Hill was reported at AWWA's 1985 Distribution System Symposium and
described by (O'Day et al., 1986). This limited survey is significant in that it demonstrates the
variability in distribution system management actions taken among small, medium, and large
utilities. Percentages of reported actions are shown in Table 5. The data show the expected
results that small utilities participated in infrastructure management activities to a lesser extent
than larger utilities.
Table 5. Percentage of surveyed utilities practicing management activities shown
(O'Day et al., 1986)

Management activity
< 5 mgd
5-50 mgd
> 50 mgd
Leak detection surveys
Method to determine replacement need
Computer model of system hydraulics
Reports of main breaks
Steps to remove scale and tuberculation
Revenue to finance renewal program
Budget and planning for replacement
3.2 AWWA's Water://Stats Database
AWWA's Water://Stats database (AWWA, 2004) is the most current survey of the drinking
water industry produced by AWWA. AWWA's water industry surveys began before 1900, and
the information they provide can serve as a historical reference and current source of
information about water distribution infrastructure.
A compilation of the surveys since 1945 was provided by AWWA (Keeley, 2003). It shows that
surveys were conducted every five years from 1945 through 1970, then surveys were conducted
more frequently. The number of utilities responding varies from a low of 211 (1985 survey) to a
high of 1,397 (1981 survey), with the average between 1945 and 1985 being 770 utilities (Seidel,
1985). Grigg (1988) reviewed data in the 1984 survey. Prior to about 1980, the surveys focused
on water production and rates. Survey data and associated reports show that management
attitudes have changed, requiring the collection of different data. For example, discussion at the
1985 AWWA annual conference stressed the need for capital management programs to sustain
infrastructure (O'Day et al., 1986). However, the attitude among 33 of the large utilities
surveyed was that O&M expenditures were adequate to maintain reliability even though 23 of
them had reported some deferred maintenance.
The modern survey effort was launched as a joint AWWA/AwwaRF project in 1989/1990 as the
Water Industry Data Base (WIDB). It was intended to develop detailed profiles of individual
utilities that could also be aggregated to profile the large system segment of the industry
(Cromwell et al., 1990). The initial survey was sent only to the 3,000 water systems that serve
more than 10,000 people. Some 1097 responses were obtained, representing only 2% of
community water systems, but about half of the total population served by community water
systems (112 million). Respondents reported a total of 436,000 miles of distribution pipe, broken

down as: 50% cast iron, 20% ductile iron, 15% AC pipe, and 15% other materials. Pipe
replacement rates were 0.6%/yr versus 1.6%/yr for pipe expansion. The survey also
documented the presence of 11,000 storage facilities, broken down as: 60% steel tanks, 15%
concrete tanks, and 20% below-ground clearwells and reservoirs.
The AWWA/ AwwaRF Water Industry Data Base effort was renamed WATER://STATS in
1996, and since then, surveys focus on specific subjects, such as finance (1999 survey of 672
utilities), distribution systems (2002 survey of 337 utilities) and rates (2004 survey of more than
250 utilities).
The 2002 Water://STATS Distribution Survey (AWWA, 2004) includes a set of questions that
focus on the distribution system rather than on general utility profiles. It was sent to 3,000 water
utilities and the response rate was 11%. Data were collected between June 2002 and April 2003.
The survey covers pipe materials, valves, fire hydrants, finished water storage facilities,
corrosion control, pumping capacity, metering, customer service lines, water auditing, leakage
management and infrastructure needs. Water audit and leakage management data is in a format
developed in 2000 by the International Water Association.
The 337 utilities that were surveyed served 59,389,902 in population, and had 14,339,261
customer service lines, and 146,435 wholesale connections. Total length of pipe was 202,000
miles for the population served, and if increased by ratio to current total population (2004 US
population of 292.5 million), it reaches 980,000 miles, a figure that is roughly comparable to the
1992 estimate of 880,000 miles (Table 2, above) drawn from the prior WIDB survey (Kirmeyer et
al., 1994). It is noted, however, that both WIDB and WATER://STATS results indicate that the
length of pipe per capita varies with system size, so using overall averages to extrapolate is only
a broad approximation.
Data available are summarized in the next section. They include utility information, types of
services provided, pipe material, finished water storage, water conveyance, valves, fire
hydrants and flushing, customer metering, customer service lines, customer service lines
responsibilities, corrosion control, water supply auditing, leakage management, and
More analysis is needed to separate wholesale and retail customers before ratios such as miles
of line and valves per capita can be compared meaningfully. Also, the data must be processed to
homogenize values and to facilitate comparison. Data in the 2002 survey do not show expansion
and replacement by pipe type.
The data from the 2002 survey might suggest that both pipe expansion and replacement have
slowed since the WIDB-based estimates by Kirmeyer et al. (1994). But interpretation of these
broad extrapolations should not be stretched that far. In contrast, the more detailed inventory of
distribution system components, profiled in the 2002 Water://Stats Survey is useful in
providing deeper insights into more parameters.

Table 6. Comparison of population-based extrapolations to national totals from AWWA 1989/90
WIDB survey results and AWWA 2004: Water://Stats 2002 survey
(1097 utilities)
(Kirmeyer et al., 1994)
Water://Stats 2002
(337 Utilities)
(AWWA, 2003)
Miles of pipe
Expansion per year, miles
Replacement per year, miles
Shown below are:
•	A matrix showing availability/quality of the data available for each subject;
•	An assessment of the capability to disaggregate national figures into regional and/ or
system size categories; and,
•	An analysis of whether trends can be observed for regions or system sizes
Table 7. Matrix of data availability in AWWA 2004: Water://Stats 2002 survey
Quality of data
Pipe data
Data on miles of pipe by type is very detailed. Data on
expansion and renewal does not specify pipe
materials. Data on pipe condition and on failure
mechanics is only anecdotal.
Finished water storage tanks
Inventory data is very detailed by type of tank.
Condition data is not available. Data is available on
Joints and gaskets (not included)
Data not available.
Inventory data and data on maintenance and
exercising is detailed.
Customer service lines
Inventory data on customer service lines is provided.
Data is available on type of line, but not on condition
of lines.
Distribution system meters
No data is available.
Customer meters
Inventory data is provided, but no data on condition or
reliability is available.
Valves (gate, butterfly, PRV)
Data on number of gate, butterfly, and pressure
reducing valves is available, but condition data is not
Data not available
Backflow preventers
Data not available in Water://Stats.
Other system appurtenances (e.g.,
blowoffs, air release valves)
Data not available

Table 8 presents the results from the 2002 Water://Stats data. The raw data could be
disaggregated by size of pipe and size of system, But such assessment was beyond the scope of
this paper to address. Note that the data on percent of total miles of pipe is based on averages of
reported data by utilities of this statistic, and is not computed from the data on miles in place as
reported in Table 8. This method of computing the averages will produce small differences in
the right hand column of Table 8, but is not significant in terms of estimating the national
inventory of pipe.
Table 8. Pipe Inventory of AWWA 2004: Water://Stats 2002 Survey

% of total miles of
Pipe material
Miles in place
(WaterStats, 2002)
pipe *
(as reported by
Ductile iron, CML
Asbestos cement
Cast iron, unlined
Cast iron, CML
Ductile iron, unlined
Other 1
Concrete pressure
Other 2
Other 3
Other 4
* (Percentages do not add to 100 because of data inconsistencies).
Responses for the "other" categories were: galvanized iron, H DPE, wrought iron, black iron,
copper, steel cylinder pipe, plastic, cement-stove, fiberglass (Permastrand), concrete lined steel
cylinder, steel, arch concrete masonry, polybutylene, and unknown. Utilities listed different
materials for "Other (1, 2, 3, 4)," and these cannot be correlated with pipe material type, such as
HDPE, black iron, etc.
Finished water storage tank data from the 2002 Water://Stats survey show a total of 4,929
storage tanks among the surveyed utilities. The types of tanks are shown in the Table 9.
Capacities are also reported, but quality of the data in the survey tables should be assessed
before totals can be reported. Some utilities may have reported capacity in gallons, rather than
millions of gallons, and an analysis of the data should be carried out before total capacities are
reported. Hydrant data from the 2002 survey are reported in Table 10. Data on repairs and
inspections are on an annual basis.

In Table 11, data on customer service lines are shown. The total number of lines surveyed was
14,120,646, which serve a population of over 59 million customers. Extrapolation of the reported
value of 3.3% for lead pipe suggests that there are some 2.3 million lead service lines in use, and
this estimate compares well to the estimate by (Kirmeyer et al., 1994) of 2.3 to 5.1 million lead
service lines still in service.
Table 9. Storage facilities (AWWA 2004: Water://Stats 2002 survey)
Storage tank type
Number in service
Welded ground storage
Welded elevated
Reinforced Concrete
Welded standpipes
Bolted ground storage
Bolted standpipes
Table 10. Hydrant data (AWWA 2004: Water://Stats 2002 survey)
Profile data
Hydrants in system, dry barrel
Hydrants in system, wet barrel
Total hydrants
Hydrant repairs, dry barrel
Hydrant repairs, wet barrel
Hydrant inspection, dry barrel
Hydrant inspection, wet barrel
Miles of pipe flushed annually
Hydrants flushed

Table 11. Customer service lines (AWWA 2004: Water://Stats 2002 survey)
Service line type
Percent *
Copper pipe
Lead pipe
Polyvinyl chloride
Cast iron
Asbestos cement
Other 1
Other 2
* Percentages do not sum to 100 due to
inconsistencies in reporting of data.
Reported in "other" categories are: ductile iron, plastic, brass, CAI, DUC, wrought iron,
Tubelog, cement lined wrought iron, KITEC (aluminum/PE composite), Tuballoy, and HDPE.
Data summaries do not clearly distinguish which are "Other 1" and "Other 2."
Data in Table 12 represents the data available on the valve types and size in service as reported
by the surveyed utilities.
Table 12. Valve data (AWWA 2004: Water://Stats 2002 survey)
Valve type
Number in service
Gate valves, 12 in and smaller
Gate valves, larger than 12 in
Butterfly valves, 12 in and smaller
Butterfly valves, larger than 12 in
PR valves, 12 in and smaller
PR valves, larger than 12 in
Other equipment and appurtenances (general data is not available on these elements of the
distribution system):
•	Pumps
•	Backflow preventers
•	Other system appurtenances (e.g., blowoffs, air release valves)
•	Joints and gaskets
•	Distribution system meters
•	Customer meters

The surveyed population of 59,389,902 is about 20% of the national population in 2004. Table 13
summarizes data and extrapolates it to the year 2004 population, simply on the basis of
population. The tenuous nature of such extrapolation should be respected.
Table 13. Summary of population extrapolations from AWWA 2004: Water://Stats 2002 survey
Dist. system infrastructure
Surveyed population
served of 59.4 million
Extrapolated to 2004 U.S.
population of 292.5 million
Pipe miles
Storage tanks
Total hydrants
Total service lines
Total valves
Expansion, miles
Replacement, miles
4.0 EPA Surveys and Other Federal Government Analyses
Studies sponsored by the Environmental Protection Agency (EPA) include surveys and case-
based research. Surveys were produced to support needs studies for infrastructure funding, and
case-based research studies include studies by the Cincinnati and Edison Laboratories, and
comprehensive reports such as Smith et al. (2000).
EPA's Community Water Systems Survey is based on an extensive, stratified sample of systems
(EPA, 2002a). It includes estimates of miles of pipe in place (by diameter), miles replaced, and
replacement costs. The survey also includes information on storage facilities (by type),
connections, customers, and cross connection and backflow controls. It provides information on
pipe age, but not about materials. It does not include information on appurtenances.
EPA's Drinking Water Infrastructure Needs Survey collects data on funding needs that include
replacement of distribution infrastructure, but it does not inventory the actual infrastructure in
place or its condition (EPA, 2001a). EPA has also conducted a study on modeling the costs of
infrastructure (2001c) in support of the Needs Survey.
EPA's 2002 Clean Water and Drinking Water Infrastructure Funding Gap Analysis (EPA, 2002b)
makes national projections of pipe replacement investment needs derived from estimated pipe
age profiles in 20 cities developed by AWWA (Cromwell, 2001a). It concludes that most of the
funding need for pipe replacement lies beyond the 20-year horizon of the study, with needs
ramping up continually through a projected peak in 2040.
The General Accounting Office and Congressional Budget Office also conduct studies of
distribution system issues from time to time, but they normally do not conduct original surveys
(GAO, 1980, 2001; CBO, 2002). These studies rely on data available from other sources and on
limited surveys to develop policy recommendations with budgetary implications for the federal

government. They have broadly concurred that investment needs for replacement investment
are growing and will present a large need.
4.1 EPA Community Water System Survey (EPA, 2002a)
The Community Water System Survey (CWSS) is a broad profile survey of the industry that has
been replicated by EPA in 1976,1982,1986,1995, and 2000. The most recent replications
introduced a carefully designed stratified sampling design intended to represent the diversity
of systems types and sizes. The 1,806 systems included in the 2000 survey sample represent a
census of systems serving more tan 100,000 population. The response rates ranged from 56 to
63% for system serving more than 3,300 persons. EPA performed field visits to boost response
rates to the 85 to 99% range in systems size categories serving fewer than 3,300 persons. EPA
also applied a QA protocol to review of the 1,246 survey responses.
The 2000 CWSS results show that, overall, 47% of all capital expenditures is devoted to
distribution and transmission infrastructure, a proportion that is fairly consistent across system
size categories. The overall proportion of capital expenditures for storage facilities is 12%, which
tends to be higher - up to 20% - of total outlays in small systems.
Data on pipe age shows 78% is less than 40 years old; 18% is 40 to 80 years old; and only 4% is
more than 80 years old. As shown in Table 14, the age profile of pipe assets documented in the
2000 CWSS is markedly different by system size, with large systems being generally older than
small systems. This is consistent with the fact that roughly half of all small systems are
suburban systems that are necessarily younger than the urban areas they adjoin. Overall
replacement rates are less than 1% per year, varying from 0% to almost 2%, from small to large
Table 14. EPA 2000 Community Water System Survey Data on Pipe Assets
size (pop
Percent of pipe per system
by age class (yrs)
Average miles of pipe per system
by diameter (inches)
> 80
> 10
< 100
500,000 +


The 2000 CWSS provides data for each system size category on the total miles of pipe in place
by diameter. As shown in Table 14, mains greater than 10-inches in diameter exist mainly in
larger systems. Systems serving fewer than 3,300 persons typically have less than 1-mile of such
pipe. In addition, the CWSS has data on the number of service connections in each system size
category, enabling estimates of the total number of service lines in place nationally, by
extrapolation. This combination of factors should also enable a good basis for developing miles
of pipe/connection relationships by system size that could be used for extrapolation to estimate
national totals for pipe assets by diameter and age. Since the CWSS also documents the number
of storage facilities and their capacities, extrapolation to national totals for storage tanks should
also be possible.
The 2000 CWSS also contains details about the presence of cross connection and backflow
controls. It clearly documents a lesser degree of penetration of such practices in small systems.
4.2 EPA Drinking Water Infrastructure Needs Survey (EPA, 2005)
The EPA Needs Survey has been replicated in 1995,1999, and 2003 to serve as the basis for
reports to Congress documenting the extent of investment requirements in support of the State
Revolving Loan Fund program. The survey is conducted with the assistance of state
governments who have a stake in assuring a high response rate in order to substantiate the need
for their share of SRF funds. The 2003 survey was conducted as a census for 1,342 systems
serving more than 40,000 persons and as a random sampling of 2,553 systems serving between
3,300 and 40,000 people. For systems serving fewer than 3,300, the 2003 needs estimates were
developed by extrapolation from the 1999 results that were based on a sample of 599 systems
for which needs were documented by field visits conducted by EPA.
The analytical objective of the Needs Survey is to document projected capital investment needs
over a 20-year horizon based on site-specific information provided by respondents to document
planned investment projects. The data is subjected to QA protocols at both the state and EPA
levels. Because specific projects are less formulated when they are farther off in time, the earlier
versions of the Needs Survey were suspected to have understated the total needs by missing
some longer term needs. The 2003 survey was implemented with extra measures to enhance the
articulation of long-term needs. The result was an estimated total 20-year need of $277 billion,
60% more than the previous estimates of $167 billion. EPA concluded the increase is attributable
to longer-term projects. The system size breakdown of the $277 billion total is as follows: $123
billion for large systems (> 50,000 people); $103 billion for medium size systems; and $34 billion
for small systems (< 3,300 people). The order of magnitude of the 2003 total needs estimate is
consistent with other major national estimates of investment needs.
Of the total estimated need of $277 billion, $184 billion is estimated to be required for
transmission and distribution projects and $25 billion is identified for storage projects. Table 15
presents the breakdown of these projected needs by system size.

Table 15. EPA needs survey data on distribution, transmission and storage needs
System size
(pop served)
Source &
Distribution &
Storage needs
Total needs
($ 2003)
Large systems
Small systems
(< 3,300)
All community
4.3 EPA White Papers on Distribution Systems
As mentioned earlier, a white paper that was prepared for EPA offered a summary of
distribution system infrastructure issues facing the nation (American Water Works Service Co.,
Inc., 2002). Eight other white papers were prepared to address issues about distribution systems
and are listed in this section because they may contain information that will help the reader
understand infrastructure-related distribution system issues.
The paper on infrastructure covers the problems of aging and corrosion (American Water
Works Service Company, 2002). It discusses general issues, such as the current condition of
buried infrastructure, capital needs, technical issues of buried infrastructure, and assessment
Several of the papers discuss how infrastructure failures can open paths to contamination. One
paper covers how installation and repair of water mains might introduce possible routes to
contamination (AWWA and Economic and Engineering Services Inc., 2002a), and another
covers the potential health implications of failures at covered storage facilities (AWWA and
Economic and Engineering Services Inc., 2002b). A paper on intrusion explains the possible
roles of pressure transients, or specialized backflow situations, in contaminating water mains
(LeChavallier, Gullick, and Karim, 2002). Another paper, on cross-connection control, explains
backflow and cross-connection risks (EPA Office of Ground Water and Drinking Water, 2002).
A paper on permeation and leaching is about external threats to pipes, or how chemicals can
penetrate plastic pipes (AWWA and Economic and Engineering Services Inc., 2002c).
Three papers focus on water quality issues. One explains decay of water quality over time in
distribution systems (AWWA and Economic and Engineering Services Inc., 2002d). Another
paper discusses microbes associated with biofilms, diseases, pathogen routes to the DS, and
management indicators (EPA, 2002). A third paper (AWWA and Economic and Engineering
Services Inc., 2002e) explains nitrification and especially the associated health risks. Nitrification

is explained as a microbial process that mainly involves oxidation of ammonia to nitrite and
5.0 Condition Assessment of Water Distribution Pipes
Effective assessment of pipe condition is required to plan renewal programs for water
distribution systems. This section of the paper describes utility condition assessment practices
as observed in an industry survey, workshops, case studies and various publications. It also
summarizes issues faced by utilities in implementing condition assessment.
In March 2002, WERF convened a workshop to define research priorities in asset management
(WERF, 2002). The top-ranked research need arose from the lack of standardized guidelines for
conducting condition assessments and using such data to understand asset condition and
performance. AwwaRF and WERF have a joint project ongoing to address fill this gap
(Urquhart, 2004).
The context for understanding the objective of condition assessment is anchored in principles
that have been established in the global best practice of asset management, as documented by
utility practitioners in Australia and New Zealand in the International Infrastructure
Management Manual (NAMS, 2006). The objective of asset condition and performance
assessment is not to manage asset condition, but to manage failure risk (Urquhart, 2005). The
purpose of condition data is to make an assessment of the remaining life of the asset so that
rehabilitation or replacement investment can be planned and implemented before failures occur
that would cost the utility more than it would to have avoided such failures through asset
management. This risk management context provides the basis on which the cost of condition
assessment is justified. In the standard practice of asset management, there is an important risk
prioritization step in which a differentiation can be made between "reactive" assets and
"proactive" assets (Urquhart, 2004). It is worth noting that the best practice in applying this risk
management protocol takes full account of the environmental and social costs of failures in
applying the risk management discipline in a triple bottom line sense.
Reactive assets are those for which the consequences of failure are quite low. Main breaks on
small lines on residential streets would be an example. Given the cost of condition assessment
on small lines and the difficulty of predicting specific failures on a hundred miles or more of
individual small lines, it is much more cost-effective to simply fix lines when they break. Some
Australian utilities have in fact focused their entire small mains asset management program on
that objective by focusing on responsiveness to failure as the best risk management approach —
reducing the cost per break repair and the time-out-of-water (Cromwell, 2001b). Condition data
is useful in planning rehabilitation and replacement investments for small mains, but these
reactive assets are approached as a population of assets, using statistical analysis of pooled data
on such parameter as break trends segmented by pipe material and soil type in order to assess
overall replacement needs. This type of statistical analysis of aggregate performance data, such
as breaks, does not cost as much as some other forms of actual condition assessment of specific
lines and is therefore more suitable to lower priority risks.

Typical data used for such analysis includes pipe age, pipe material, pipe diameter, soil
conditions, number of breaks, any rehabilitation that might have been conducted on the pipe,
pressure of operation, and complaints of taste, odor, color, or low pressure associated with the
delivered water. However, the lack of standardized procedures and common terminology for
recording data on leaks and breaks has challenged adoption of such programs. For instance,
some utilities do not differentiate by pipe failure type, yet the mode of failure can provide
insight on the condition of the pipe. It is often assumed that the mode of failure is corrosion
failure. While this is an important type of failure, there are a variety of factors that contribute to
failure (pipe break). As another example, many northern climate pipe failures occur in the fall
and spring, as soil temperatures are either decreasing with the advent of winter, or increasing
with spring. These pipe failures tend to be circumferential failures typically due to soil
movement, and have little or nothing to do with corrosion. Other types of failures, when
properly identified and analyzed, can also yield useful condition information
New factors to consider within this risk management framework have recently come to light
and will have to be incorporated. First, there is growing evidence that pipes very often leak
before they break. Further, it may be the case that some breaks would not even occur if the leaks
had been fixed when they first began, preventing them from potentially undermining pipe
foundations and producing stresses where there were none (Hughes, 2002). This suggests that
leak detection could become a much more significant tool for proactive efforts, even applied to
small mains.
A second major area of concern involves not just the physical condition of pipes (especially
small mains), but also their water quality performance. Performance failure is just as important
as physical failure. With tightening regulatory requirements on water quality at the tap and the
potential disruption of the chemical and biological equilibrium in old pipes when subjected to
different waters produced by advanced treatment processes, the useful life of some pipes may
be shortened if stable performance in terms of delivered water quality cannot be recovered.
Thus, water quality monitoring must be regarded as an essential component of asset condition
assessment. In addition to these concerns, there is also growing concern that repeated main
breaks in small lines may be an important source of contamination threats. This could have the
effect of either increasing the cost of main repairs or decreasing the number of failures that
should be endured prior to replacement. The effect is the same - shorter pipe life.
"Proactive" assets are those for which the consequences of failure are quite high, making it
worthwhile to be proactive about managing failure risk. An example would be the loss of a
large main lying under Main Street, causing significant damage and disruption in addition to
substantial service outage. Because so much is at stake, the cost entailed in conducting and
evaluating condition assessment data is more than justified for such "proactive" assets that are
also called "critical assets" in the parlance of vulnerability assessments recently conducted in
the US.
At this time the accepted method of recording results of condition assessments for 'proactive"
assets (or, critical assets) is a five-point scoring system, such as the following (Morrison, 2005):

1.	Little or no deterioration, performance more than adequate.
2.	Minor deterioration, performance adequate.
3.	Mildly deteriorated, short-term performance just adequate, however will require
renewal or replacement soon.
4.	Severely deteriorated and in need of repair, renewal or replacement.
5.	In danger of immediate failure, requires emergency repair or replacement.
Non-destructive inspection is commonly done in wastewater lines using closed-circuit
television (CCTV) cameras. This technology has provided valuable information to wastewater
managers and the five point scoring system has actually been put to use for wastewater mains
using such data. However, the application of CCTV is much less valuable in the potable water
environment where failure modes are different, access to these pressurized systems is much
more limited, and pipes are usually smaller. The unique nature of pressurized water
distribution system presents a significant technical challenge for universal scoring protocols.
This requires adaptable tools and training to address the myriad pipe sizes, materials, and ages,
as well as fittings with tight bends and other constrictions. Advanced applications will be
required for the future, which may include real-time assessment, smart pigs, automated pipe
data registration and other technologies.
Most non-destructive inspection technologies require some type of hardware access to the
interior of the pipe, and for the pipe to be dewatered for effective inspection. The requirement
for access to the interior of the pipe can result in high first-time inspection costs because water
systems may need to modify their system. For instance, a major US city recently spent
approximately $700,000 for inspection of one mile of 36-inch diameter pipe. Most of this cost
was associated with gaining access to the interior of the pipe. Approximately 17% of the cost
was the actual inspection. The dewatering requirement, traffic control, inspection manhole
installations, and pavement restoration in an urban environment can result in high indirect
costs, and also severely limit the timeframes when inspection is feasible. The overall cost of
nondestructive inspection at this time limits the economic application of these technologies to
larger diameter pipe (typically 24-inches or greater in the US). The failure of large pipes in this
size range is a rare event, but typically creates significant direct damage and service outage
issues and thus makes the associated cost justifiable — if the utility recognizes the inherent risk
management context of asset management.
One of the greatest success stories of non-destructive testing in the potable water sector has
resulted from a number of catastrophic failures of large diameter PCCP due to failure of metal
reinforcing bands. Ingenious use of magnetic, acoustic and fiber optic technologies have
produced a substantial toolkit for utilities facing this risk (Johnson and Shenkiryk, 2006),
demonstrating that perhaps technology can rise to the challenge in this area.
Additional research has been called for during the past decade (Kleiner, 2005) to include
nondestructive test methods for determining the condition of existing pipe, improved leak
detection equipment and methods to measure losses, studies of causes for pipe, joint, lining,

and coating deterioration, including corrosion, and development of better in-situ methods to
test condition. Taking into account the evolution of research and trends in the water supply
industry, it seems likely that in the future utilities will more actively manage distribution
systems with more monitors, safeguards, and protective systems. For instance, recent work on
water accounting has developed a defined set of terms and a considerably increased
understanding of water leakage. Partly due to these developments work is now ongoing to
improve real-time pressure management, which is directly related to leakage, and leak
detection hardware. While this is just one example, it is representative of a general trend.
Both technologies and methods are evolving to improve condition assessment capabilities. This
conclusion is based on a recent synthesis of AwwaRF reports on distribution system
infrastructure. AwwaRF has commissioned a number of recent studies on distribution system
infrastructure, and experts recommended more and continued research on failure mechanisms
with different types of pipe, causes for pipe, joint, lining, and coating deterioration, and
continuing integration of results, along with more focused and practical guidance for utilities in
this complex arena. In general, the goal is to develop more accurate, user-friendly test methods
to determine condition of pipe, to expand understanding of causes for deterioration, leaks, and
breaks, and to prevent problems and predict length of life under various conditions.
6.0 Conclusions
The paper summarized the available data on the inventory and condition of the nation's water
distribution infrastructure. Taken together, the available data and companion studies provide
much more information on distribution systems than existed even as recently as a decade ago.
EPA has conducted a number of studies about distribution systems, including surveys and
research investigations, and more detailed data is available in AWWA's Water://Stats
program. AWWA and AwwaRF have also conducted a number of separate studies, both relying
on existing data and on surveys and/or case studies.
At the national level, the database of inventory information on distribution systems is fairly
good. However, the national database is built from utility-level information that has not in all
cases been verified. While some data on age of constructed facilities is available, data on
condition of systems is weak.
The paper reported on data contained in the literature and distribution system data from
AWWA's 2002 Water://Stats survey, which is the latest available. The matrix of data
availability shows that data on pipes, finished water storage, hydrants, some types of valves,
and customer service lines is generally good. Very little data is available on other components
of distribution systems.
National data from EPA surveys and the AWWA Water://Stats surveys can be disaggregated
to provide regional and/or system size categorical data. Trends can be analyzed for regions or
system sizes, and by comparisons with previous surveys, time trends can be evaluated.

From recent research, the water industry has a good understanding of how condition
assessment is practiced by utilities. Condition assessment is not done consistently by utilities,
and system condition is not well known by most utilities. Gross indicators such as "poor" or
"good" condition are normally reported, rather than remaining life and more definite
indicators. The art and science of condition assessment need further improvements. While tools
for condition assessment hold promise, more development and training are necessary to
advance the state of knowledge.
7.0 References
American Water Works Service Co., Inc. 2002. Deteriorating Buried Infrastructure
Management Challenges and Strategies. Issue Paper for EPA.
http:/ /www.epa.gov/safewater/tcr/tcr.html#distribution. Accessed February 13, 2004.
AWWA and Economic and Engineering Services Inc. 2002a. New or Repaired Water
Mains. Distribution System Issue Paper.
http://www.epa.gov/safewater/tcr/tcr.html#distribution. Accessed February 17, 2004.
AWWA and Economic and Engineering Services Inc. 2002b. Finished Water Storage
Facilities. Distribution System Issue Paper.
http://www.epa.gov/safewater/tcr/tcr.html#distribution. Accessed February 17, 2004.
AWWA and Economic and Engineering Services Inc. 2002c. Permeation and Leaching.
Distribution System Issue Paper.
http://www.epa.eov/safewater/tcr/tcr.html#distribution. Accessed February 17, 2004.
AWWA and Economic and Engineering Services Inc. 2002d. Decay in Water Quality
Over Time. Distribution System Issue Paper.
http://www.epa.gov/safewater/tcr/tcr.html#distribution. Accessed February 17, 2004.
AWWA and Economic and Engineering Services Inc. 2002e. Nitrification. Distribution
System Issue Paper, http://www.epa.gov/safewater/tcr/tcr.html#distribution.
Accessed February 17, 2004.
AWWA. 2004. Water://Stats 2002 Distribution Survey. Denver, CO: AWWA.
Congressional Budget Office (CBO). 2002. Future Investment in Drinking Water and
Wastewater Infrastructure.
Cromwell, J., E. Speranza, and H. Reynolds. 2001a Reinvesting in Drinking Water
Infrastructure: Dawn of the Replacement Era. Denver, CO: AWWA.
Cromwell, J., G. Nestel, R. Albani, L. Paralez, A. Deb, and F. Grablutz. 2001b. Financial
and Economic Optimization of Water Main Replacement Programs. Denver, CO:
AWWA Research Foundation.
Cromwell, J.E. Ill, R.G. Lee, and E. Kawczynski. 1990. Development of the AWWA
Water Industry Data Base: A "Facts" Machine for the Future. Cincinnati, OH: AWWA
Annual Convention.

Deb, A.K., Y.J. Hasit, H.M. Schoser, and J.K. Snyder. 2002a. Decision Support System for
Distribution System Piping Renewal. Denver, CO: AWWA Research Foundation:
Distribution Systems.
Deb, A.K., F.M. Crablutz, Y.J. Hasit, J.K. Snyder, G.V. Loganathan, and N. Agbenowski.
2002b. Prioritizing Water Main Replacement and Rehabilitation. Denver, CO: AWWA
Research Foundation.
Deb, A.K., K.A. Momberger, Y.J. Hasit, and F.M. Grablutz. 2000. Guidance for
Management of Distribution System Operation and Maintenance. Denver, CO: AWWA
Research Foundation: Distribution Systems.
Deb, A.K., Y.J. Hasit, F.M. Grablutz, and R.K. Herz. 1998. Quantifying Future
Rehabilitation and Replacement Needs of Water Mains. Denver, CO: AWWA Research
Foundation: Distribution Systems.
Deb, A.K., Y.J. Hasit, and F.M. Grablutz. 1995. Distribution System Performance
Evaluation. Denver, CO: AWWA Research Foundation: Distribution Systems.
Deb, A.K., Y.K. Hasit, and F.M. Grablutz. 1995. Distribution System Performance
Evaluation. Denver, CO: AWWA Research Foundation: Distribution Systems.
Deb, A.K., Y.J. Hasit, J.A. Williams Jr., and R. Jacob. 2001. New Techniques for Precisely
Locating Buried Infrastructure. Denver, CO: Awwa Research Foundation.
Environmental Protection Agency. 2004. Issue Papers.
http:/ /www.epa.gov/safewater/tcr/tcr.html#distribution. Accessed January 26, 2004.
Environmental Protection Agency. 2002a. Community Water Systems Survey - 2000. 815-R-
02-005A,B. December.
http://www.epa.gov/safewater/consumer/cwss 2000 volume i.pdf. Accessed
January 27, 2004.
Environmental Protection Agency. 2002b. Health Risks from Microbial Growth and
Biofilms in Drinking Water Distribution Systems Distribution System Issue Paper.
http://www.epa.gov/safewater/tcr/tcr.html#distribution. Accessed February 17, 2004.
(Authorship of paper is not noted).
Environmental Protection Agency. 2002c. The Clean Water and Drinking Water
Infrastructure Gap Analysis. 816-R-02-020. September.
Environmental Protection Agency (Office of Ground Water and Drinking Water). 2002d.
Potential Contamination Due to Cross-Connections and Backflow and the Associated
Health Risks. Distribution System Issue Paper.
http://www.epa.gov/safewater/tcr/tcr.html#distribution. Accessed February 17, 2004.

Environmental Protection Agency. 2005. Drinking Water Infrastructure Needs Survey.
Third Report to Congress. EPA 816-R-05-001. Washington, DC: U.S. Environmental
Protection Agency.
Environmental Protection Agency. 2006. 2003 Drinking Water Infrastructure Needs Survey:
Modeling the Cost of Infrastructure. EPA 816-R-06-007. Washington, DC: U.S.
Environmental Protection Agency.
Government Accountability Office (GAO). 2001. Report on Water Infrastructure:
Information on Federal and State Financial Assistance.
Government Accountability Office (GAO). 1980. Comptroller General of the United
States. November 24,1980. Additional Federal Aid for Urban Water Distribution
Systems Should Wait Until Needs Are Clearly Established. CED-81-17.
Grigg, N.S. 2004. Assessment and Renewal of Water Distribution Systems. AwwaRF
Report Number 91025. Denver, CO.
Grigg, N.S. 1988. Water Utility Operating Data, 1984. American Water Works
Hughes, D. et al. 2002. Assessing the Future: Water Utility Infrastructure Management.
Denver, CO: AWWA.
Johnson, B. and M. Shenkiryk. 2006. Condition Based Asset Management for Critical
Water Pipelines. Underground Infrastructure Management, May/June 2006.
Keeley, K. 2003. AWWA Survey of Operating Data, 1945-1985. Unpublished document.
Denver, CO: AWWA.
Kirmeyer, G.J., W. Richards, and C.D. Smith. 1994. An Assessment of Water Distribution
Systems and Associated Research Needs. Denver, CO: AWWA Research Foundation:
Distribution Systems.
Kleiner, Y. et al. 2005. Risk Management of Large Diameter Water Transmission Mains.
Denver, CO: AwwaRF.
LeChavallier, M.W., R.W. Gullick, and M. Karim. 2002. The Potential for Health Risks
from Intrusion of Contaminants into the Distribution System from Pressure Transients.
Distribution System White Paper.
http:/ /www.epa.gov/safewater/tcr/tcr.html#distribution. Accessed February 17, 2004.
Morrison, R. 2004. Condition Assessment - Back to Basics. Underground Infrastructure
Management. November/December.
NAMS (National Asset Management Steering Group). 2006. International Infrastructure
Management Manual. New Zealand.
National Research Council. 2006. Drinking Water Distribution Systems: Assessing and
Reducing Risks. Committee on Public Water Supply Distribution Systems. Washington,

O'Day, D.K., R. Weiss, S. Chiavari, and D. Blair. 1986. Water Main Evaluation for
Rehabilitation/Replacement. Denver, CO: AWWA Research Foundation and
Environmental Protection Agency.
Seidel, H.F. 1985. Water Utility Operating Data: An Analysis. Journal of American Water
Works Association Vol. 77, No. 5, p 34-41, May.
Smith, L.A., K.A. Fields, A.S.C. Chen, and A.N. Tafuri. 2000. Options for Leak and Break
Detection and Repair of Drinking Water Systems. Columbus, OH: Battelle Press.
Stratus Consulting. 1998. Infrastructure Needs for the Public Water Supply Sector.
Report for American Water Works Association. Boulder, CO. December.
Urquhart. A. 2005. Incorporating Condition Assessment into a Comprehensive Asset
Management Program. Underground Infrastructure Management. November/
Urquhart, A. 2004. Protocols for Assessing Condition and Performance of Water and
Wastewater Assets. Denver, CO: AwwaRF.
WERF. 2002. Research Priorities for Successful Asset Management: A Workshop.
Alexandria, VA.